Prevention Of Malaria In Pregnancy.pdf

ANNIKA LAUNIALA

Prevention of Malaria in Pregnancy in Malawi Encounters and non-encounters between global policies, national programmes and local realities

ACADEMIC DISSERTATION To be presented, with the permission of the Faculty of Medicine of the University of Tampere, for public discussion in the Small Auditorium of Building B, Medical School of the University of Tampere, Medisiinarinkatu 3, Tampere, on December 10th, 2010, at 12 o’clock.

UNIVERSITY OF TAMPERE

ACADEMIC DISSERTATION University of Tampere, Tampere School of Public Health Doctoral Programs in Public Health (DPPH) Finland

Supervised by Professor Juhani Lehto University of Tampere Finland Professor Marja-Liisa Honkasalo University of Linköping Sweden

Reviewed by Associate Professor Hanne Overgaard Mogensen University of Copenhagen Denmark Professor Tuula Vaskilampi University of Eastern Finland Finland

Distribution Bookshop TAJU P.O. Box 617 33014 University of Tampere Finland

Tel. +358 40 190 9800 Fax +358 3 3551 7685 [email protected] www.uta.fi/taju http://granum.uta.fi

Cover design by Mikko Reinikka

Acta Universitatis Tamperensis 1560 ISBN 978-951-44-8255-7 (print) ISSN-L 1455-1616 ISSN 1455-1616

Tampereen Yliopistopaino Oy – Juvenes Print Tampere 2010

Acta Electronica Universitatis Tamperensis 1009 ISBN 978-951-44-8256-4 (pdf ) ISSN 1456-954X http://acta.uta.fi

SUMMARY The main methods used are focus group discussions (FGDs) and in-depth interviews (IDI) with pregnant women, women of reproductive age, female elders and men covering multiple themes ranging from pregnancy and delivery issues to malaria in pregnancy and its prevention. As part of the interviews, different kinds of ranking and classification exercises were conducted regarding pharmaceuticals, obstetric complications and illnesses. A total of 16 FGDs and 64 IDIs were conducted mainly in the communities. At the Lungwena ANC, observations were carried out in order to understand what kind of health education and services pregnant women receive and how malaria prevention activities are implemented in practice. In addition, the study included data from a knowledge, attitudes and practices (KAP) survey, totaling 248 interviews and from the three year work experience with national level programme planning and management of community based malaria prevention programmes in Malawi.

Malaria in pregnancy is a multidimensional public health problem that has severe adverse consequences for the pregnant woman and her unborn baby. Pregnant women are at risk of delivering babies with low birth weight (below 2500g), and malaria increases the risk of maternal anaemia that may lead to maternal death. In sub-Saharan Africa alone there are approximately 25-30 million pregnant women at risk of the disease. To date, medical anthropology research on malaria in pregnancy has been scarce. Little is known about the factors that affect pregnant women’s and their extended family members’ motivation and agency to seek treatment and preventive services from the antenatal clinic (ANC) which serves as a platform for implementation of the current malaria prevention strategies, namely intermittent preventive treatment and insecticide treated nets. Yet it is acknowledged that prevention programmes depend largely on pregnant women’s participation and adherence with services. This applied medical anthropology research was planned to provide comprehensive understanding of factors affecting prevention of malaria in pregnancy from the perspective of pregnant women and their extended family members, as well as to provide insights into the interwoven factors at local, national and global level that affect planning and implementation of tailored and context specific malaria prevention programmes. The research plan is based on a three year experience (1998-2001) of working in malaria prevention in Malawi.

The main findings from this study show that prevention of malaria in pregnancy needs to be extended beyond the pregnant women. Men as husbands, and female elders, have specific roles and responsibilities to maintain the well-being of the pregnant women in the community. The emphasis of prevention programmes aiming to replace cultural misconceptions with appropriate public health information is also problematic because it often fails to recognise that local understanding of malaria is formulated in encounters among community members, health workers and the like, and these syncretic explanatory models of malaria are diverse among community members. There is no one kind of local understanding of malaria in pregnancy. In addition, the findings explain why malaria in pregnancy is not perceived as a major danger among the Yao women suggesting that the local risk concept differs from the public health risk definition. Examining malaria in the context of pregnancy and perceived

This study was conducted among the Yao in Lungwena and its surrounding villages in rural Malawi. The main fieldwork was carried out in 2002 and a complementing focused data collection in 2006. The study employed an ethnographic approach and multiple methods in order to comprehensively illustrate what happens when the local reality of a pregnant Yao woman encounter the globally agreed priorities and policies transformed into national malaria prevention programmes implemented in a rural antenatal clinic setting. i

vulnerabilities shows that pregnant women are exposed to multiple dangers ranging from witchcraft, extra-marital relationships, multiple illnesses to everyday worries in a resource poor setting. Malaria is only one of the many dangers emerging in the local community context. To conclude, this study clearly demonstrates that an ethnographic approach and multimethod design provide important knowledge and novel insights into understanding malaria in pregnancy that cannot be found utilising a simplistic KAP survey alone. Malaria in pregnancy is a complex problem, and its prevention at the ANC is affected by multiple interwoven factors ranging from socio-cultural factors to economic and structural factors, many of which are beyond the agency of the pregnant women, men and female elders. The current emphasis on health education fails to address the everyday constraints and needs of pregnant women. Programmes should place more emphasis on pregnancy and efforts to achieve positive pregnancy outcome than prevention of a single disease during pregnancy. Based on the findings several recommendations for improving prevention programmes are made.

ii

TIIVISTELMÄ lähikylissä Malawin maaseudulla. Varsinainen kenttätyö toteutettiin 2002 ja täydentävä, fokusoitu aineistonkeruu 2006. Tässä tutkimuksessa on käytetty etnografista lähestymistapaa ja useita erilaisia menetelmiä, jotta olisi mahdollista kuvata kokonaisvaltaisesti raskaana olevan maaseudun naisen arkitodellisuutta sekä globaalin malariapolitiikan ja kansallisten malarianehkäisyhankkeiden kohtaamista äitiysterveyshuollossa. Pääasiallisina tutkimusmenetelminä on käytetty ryhmäkeskusteluja ja syvähaastatteluja eri-ikäisillä naisilla ja miehillä. Käsitellyt teemat ovat vaihdelleet raskauteen ja synnytykseen liittyvistä kysymyksistä malarian hoitoon ja ehkäisyyn raskauden aikana. Lisäksi haastatteluissa on tehty erilaisia sairauksien, lääkkeiden ja raskauskomplikaatioiden luokitteluja. Tutkimuksessa on tehty yhteensä 16 ryhmäkeskustelua ja 64 syvähaastattelua. Lungwenan äitiysklinikalla tehtiin myös systemaattista havainnointia naisille annettavan terveyskasvatuksen sisällöstä ja toteuttamistavoista sekä itse terveystarkastuksista ja raskauden aikaisen malarian ehkäisypolitiikan toteuttamisesta. Lisäksi tutkimuksessa kerättiin 248 ’tieto, asenne ja käytäntö’ kyselylomaketta (KAP survey) ja aineistona hyödynnetään myös kolmen vuoden käytännön työkokemusta malarian ehkäisyhankkeista ja niiden toteuttamisesta kansallisella tasolla.

Raskauden aikainen malaria on monisyinen kansanterveydellinen ongelma, jolla on vakavia seurauksia sekä raskaana olevalle naiselle että hänen syntymättömälle vauvalleen. Malarian seurauksena raskaana oleva nainen voi synnyttää pienipainoisen vauvan (alle 2500g) ja hänellä voi esiintyä anemiaa, joka pahimmillaan voi johtaa äitiyskuolemaan. Yksistään Saharan eteläpuoleisessa Afrikassa on 25-30 miljoona naista, joilla on riski sairastua raskauden aikaiseen malariaan. Raskauden aikaisen malarian ehkäisy, joka koostuu jaksottaisesta ehkäisevästä lääkehoidosta sekä hyttysmyrkyllä käsiteltyjen hyttysverkkojen alla nukkumisesta, toteutetaan osana äitiysterveydenhuoltoa. Raskaana olevan naisen osallistuminen ja myötämielisyys tiedetään keskeiseksi tekijöiksi ennaltaehkäisevien ohjelmien onnistumisessa. Lääketieteellisen antropologian piirissä ei ole juurikaan tutkittu raskauden aikaista malariaa. Niinpä tekijöistä, jotka vaikuttavat raskaana olevan naisen ja hänen perheensä (extended family) motivaatioon ja mahdollisuuksiin hakeutua äitiysterveydenhuollon palveluihin ja saada raskauden aikaista malariaehkäisyä, tiedetään melko vähän. Tämä soveltavan lääketieteellisen antropologian piiriin kuuluva tutkimus on suunniteltu lisäämään tietoa erityisesti raskauden aikaisen malarian ehkäisyyn liittyvistä sosio-kulttuurisista tekijöistä, ensisijaisesti raskaana olevan naisen ja hänen perheensä näkökulmasta. Lisäksi tutkimuksen tarkoituksena on lisätä ymmärrystä moninaisista paikallisella, kansallisella ja globaalilla tasolla toisiinsa linkittyneistä tekijöistä, jotka keskeisesti vaikuttavat malarianehkäisyhankkeiden suunnitteluun ja toteuttamiseen. Tutkimuksen suunnittelussa hyödynnettiin kolmen vuoden käytännön kokemusta (1998-2001) malarianehkäisyhankkeiden suunnittelussa ja toteuttamisessa Malawissa.

Tutkimuksen keskeiset tulokset osoittavat, että raskauden aikaisen malarian ehkäisy tulee kohdistaa myös miehiin, ennen kaikkea aviomiehiin, ja yhteisön vanhoihin naisiin (female elders). Heillä on keskeinen rooli ja vastuu raskaana olevan naisen hyvinvoinnin ylläpitämisessä. Tutkimuksen tulokset tuovat myös esille, että paikallinen malarian selitysmalli (explanatory model) muodostuu yhteisön eri jäsenten ja terveydenhuoltohenkilökunnan kohtaamisissa, jolloin nämä mallit ovat synkreettisiä (eri tietolähteistä pitkällä aikavälillä kumuloitunutta tietoa) ja erilaisia eri yhteisön jäsenten keskuudessa. Malarianehkäisyhankkeiden pyrkimys poistaa kult-

Tutkimus on toteutettu Yao-heimon keskuudessa Lungwenassa ja sitä ympäröivissä iii

tuurisia väärinkäsityksiä on ongelmallinen, koska hankkeiden toteuttajilta puuttuu syvällinen ymmärrys paikallisen tiedon muodostumisesta ja sen moninaisuudesta: ei ole olemassa yhtä paikallista ja kulttuurista malarian selitysmallia, jonka voisi korvata länsimaisen lääketieteen selitysmallilla malarian aiheuttajista, tarttumistiestä, hoidosta ja ehkäisystä. Tulosten avulla on myös ymmärrettävissä, miksi raskauden aikaista malariaa ei koeta merkittävänä vaarana raskaana olevien naisten keskuudessa. Tulokset viittaavat siihen, että paikallinen riskikäsitys eroaa kansanterveystieteen ja epidemiologian riskikäsityksestä. Malarian tutkiminen raskauden ja koettujen paikallisten uhkien kontekstissa osoittaa, kuinka raskaana olevat naiset altistuvat moninaisille vaaroille noituudesta avioliiton ulkopuolisiin suhteisiin, moniin sairauksiin ja jokapäiväisiin huoliin äärimmäisen niukoissa ja köyhissä olosuhteissa. Malaria on vain yksi monista paikallisia ihmisiä huolestuttavista sairauksista ja huolista joita alueella esiintyy. Yhteenvetona voidaan sanoa, että tämä tutkimus osoittaa selkeästi, kuinka etnografisella lähestymistavalla ja erilaisten menetelmien yhdistelmällä on mahdollista tuottaa tärkeää uutta tietoa ja uusia näkökulmia raskaudenaikaisesta malariasta, mitä ei olisi mahdollista saavuttaa käyttämällä esimerkiksi yksistään ’tieto, asenne ja käytäntö’ kyselytutkimusta. Raskaudenaikainen malaria on monisyinen ongelma jonka ehkäisyyn äitiysklinikalla vaikuttavat sosio-kulttuurisiset-, taloudelliset-ja rakenteelliset tekijät. Suurin osa näistä tekijöistä on niin raskaana olevien naisten kuin miesten ja kylän vanhimpien naisten vaikutusvallan ulkopuolella. Malarianehkäisyhankkeiden tämänhetkinen painopiste terveyskasvatuksessa ei riittävästi huomioi raskaana olevien naisten jokapäiväisiä haasteita ja tarpeita. Malariahankkeiden painopiste tulisi kohdistaa enemmän itse raskauteen ja yrityksiin saavuttaa positiivinen lopputulos raskaudelle ja vähemmän yksittäisen taudin ehkäisyyn raskauden aikana. Tutkimuksen löydösten pohjalta on tehty useita suosituksia malarianehkäisyhankkeiden parantamiseksi. iv

ACKNOWLEDGEMENTS my first article. Special thanks to Teija Kulmala for providing me an opportunity to return back to Malawi for a short period. She has also taught me valuable lessons about academic authorship and research ethics.

The seeds of this research were planted nearly ten years ago in Malawi while I was still working as a UNICEF project officer responsible for malaria prevention. After three inspiring and challenging years I still felt that I needed to continue working in the field of malaria prevention. However, this time as an anthropologist that could bridge the apparent gap between malaria programs and the local people. If I had known what a long and rough journey ahead of me I might have reconsidered and decide otherwise. Luckily I didn’t know. I have many people to thank for supporting me during the hard times, as well as sharing with me the happy moments of this journey. The following people deserve special mention:

Associate Professor Hanne Mogensen and Professor Tuula Vaskilampi for agreeing to serve as official reviewers of the dissertation and for providing constructive comments. In Malawi several people deserve special thanks. Dr. Juan Ortiz (UNICEF) for always having time for meeting me and sharing the latest information and progress concerning malaria prevention programs. Juan’s enthusiasm for finding ways to reduce the burden of malaria has encouraged me to carry out this research. Mr. Ketema Bizuneh (UNICEF) for responding to my many queries and concerning national level malaria prevention issues and requests for unpublished research and progress reports. My appreciation also goes to Dr. Allan Macheso (Ministry of Health and Population) who supported this research initiative and provided valuable guidance on the research permit application process.

My supervisors Professor Juhani Lehto and Professor Marja-Liisa Honkasalo. Juhani for accepting me to the School of Public Health to study under his supervision. I am particularly indebted to Juhani for always finding money to support my attendance in international conferences. He has always been supportive and encouraging, and he has allowed me to find my own path in this research. Marja-Liisa for supporting and teaching me to become an independent researcher. She has taught me to trust myself as a researcher, to take responsibility and to make my own decisions.

My hardworking and dedicated research assistants - Shaibu, Innocent, Stephano, Misonzie, Ben, Rashid, Gertrud, Zacharia and Eunice - without whom I could not been able to carry out this research and collect the data. Shaibu deserves a special thank for responding to all my queries over these years and performing all the tasks I have asked him to do. I am also very grateful to him for allowing me to be involved in his family life and sharing many of the hardships in the everyday life.

Dr. Mary Hamel (CDC Atlanta) for providing encouraging and constructive guidance in the early stages of this work. In particular thank you for suggesting that, I focus my research on pregnant women and malaria. I am also very thankful to Dr. Rick Steketee (CDC Atlanta) for commenting on my research proposal and for emphasizing the importance of translating anthropological research results into practical program recommendations.

The women and men in Lungwena who warm-heartedly welcomed me, another msungu, to come and ask questions that probably made little sense for them in the contexts of everyday life and hardships.

My follow up group members Professor Matti Hakama and Dr. Teija Kulmala. I am particularly indebted to Matti Hakama for his guidance regarding peer review comments on

v

will of all the senior colleagues of the network who have read and commented on my manuscripts.

My Irish friends Cathy and Brian who took care of me during my visits in Lilongwe. There was never a dull moment in your company! Outi Määttänen-Burke and her husband John for also warmly welcoming me to stay in their home.

Professor Jussi Kauhanen (School of Public Health and Clinical Nutrition, University of Eastern Finland) for providing me an academic surrounding where to continue my research. Thank you also for all the colleagues and particularly to Irma Nykänen and Maarit Sinikangas. Irma - thank you for support and sharing you office with me. Maarit - thank you for helping me to forget my research every now and then by providing all kinds of interesting projects to work on.

My fellow researchers and good friends - Heli and Johanna - for your significant support during these years. You have been my valuable unofficial research team with whom I have been able to discuss and reflect on my research questions and data already in Malawi. Johanna - thank you for the inspiring anthropological discussion on the veranda of your house and letting me to stay in your small but cozy house during my fieldwork. Heli - thank you for your hospitality and for providing important opportunities also for social life during the weekends.

Warm thank you to my current colleagues in Crisis Management Centre Finland for support and encouragement, and in particularly to Ari Kerkkänen and Petteri Taitto for allowing me to finalize my research work by taking a leave of absence of several months during this year.

My research colleagues from the School of Public Health – Leena Tervonen- Goncalves, Pia Solin, Valerie Flax and Arundathi Char for your friendship and valuable peer support particularly when I was still living and working in Tampere. Pia and Leena for making me feel welcome in the School of Public Health and Leena also for continuing to be my crucial contact point at the School of Public Health. Valerie for giving me hope that it is possible to finish this work. I also appreciate your advices on the final stages of this work. Arundathi for the many inspiring Skype conversations between Finland and India.

All my friends and in particular Paula, Adrienn, Mikko and Stepani for advices, support and all the moments when it has been possible to talk ‘nonsense’ and simply to enjoy life. My parents Ulla-Marja and Kari for your love, support, patience and encouragement during all these years. You have also provided love, care and attention to our children. Without your unselfish support this thesis would have never been done. My husband Petri and our children Alli, Aarni and Ahti for your love and patience. Petri – no words can express my gratitude for your support and tolerance concerning this ‘madness’ during all these years and at the same time kept me in my senses! Alli, Aarni and Ahti – thank you for reminding me everyday what is important in life.

Professor Marja Jylhä and Dr. Kirsi LummeSandt for their support through the Doctoral Programmes in Public Health. Thank you for your understanding and flexibility when I moved with my family to Kuopio. MA Paula Nieminen for linguistic editing of the thesis and answering my many last minute queries promptly.

Doctoral Programmes in Public Health, Nordic Africa Institute and Koneen Säätiö have provided personal funding that has allowed full time work on this research.

Dr. Holly Williams and Dr. Caroline Jones for providing valuable support through the PSSMC network. Thank you also for the good

vi

CONTENTS SUMMARY .........................................................................................................................................i TIIVISTELMÄ...................................................................................................................................iii ACKNOWLEDGEMENTS.................................................................................................................v TABLE OF CONTENTS...................................................................................................................vii LIST OF ORIGINAL ARTICLES......................................................................................................ix LIST OF FIGURES AND TABLES....................................................................................................x ABBREVIATIONS............................................................................................................................ xi 1 INTRODUCTION...................................................................................................................... 1 2 MALARIA IN PREGNANCY: A LITERATURE REVIEW ................................................. 3 2.1. Global perspective: International health, global strategies and policies .............................. 3 2.1.1. Overview and consequences of malaria infection during pregnancy ............................... 4 2.1.2. Global strategies related to malaria in pregnancy ........................................................... 7 2.1.3. Recognition of importance of socio-cultural understanding of malaria ......................... 11 2.2. National perspective: Malawi .............................................................................................. 11 2.2.1. Burden of malaria in pregnancy in Malawi................................................................... 11 2.2.2. Prevention of malaria in pregnancy.............................................................................. 12 2.2.3. Malaria research in Malawi.......................................................................................... 14 2.3. Social scientific perspectives ................................................................................................ 16 2.3.1. Contribution of social scientific research on malaria so far........................................... 17 2.3.2. Malaria in pregnancy: an understudied topic in medical anthropology ......................... 19 2.4. Summarizing what is known and the ‘research gaps’ ........................................................ 24 3 OVERALL PURPOSE AND OBJECTIVES OF THE STUDY ............................................ 26 4 STUDY SETTING AND METHODOLOGY ......................................................................... 27 4.1. Entering the field and the research site ............................................................................... 27 4.2. Positioning and establishing relationships in the field ........................................................ 31 4.3. Local reality.......................................................................................................................... 32 4.3.1. Socio-cultural context .................................................................................................. 32 4.3.2. Everyday life and hardships in Lungwena .................................................................... 33 4.3.3. Medical system and pluralism...................................................................................... 34 4.4. Methodology ......................................................................................................................... 35 4.4.1. Ethnographic approach and multi-method study design (IV) ........................................ 35 4.4.2. Important role of local research assistants (IV)............................................................. 35 4.2.3. Data collection procedures........................................................................................... 37 4.4.4. Data analysis ............................................................................................................... 46 4.4.5. Limitations of the study ............................................................................................... 46

vii

5 RESULTS................................................................................................................................. 47 5.1. Gender and malaria in pregnancy (II, III) .......................................................................... 47 5.1.1. Pregnant women and prevention of malaria at the ANC ............................................... 48 5.1.2. Role of men ................................................................................................................. 48 5.1.3. Role of female elders ................................................................................................... 50 5.2. Local understanding of malaria in pregnancy (I, II, III).................................................... 52 5.2.1. Diversity of knowledge and perceptions .................................................................52 5.2.2. Syncretic explanatory model of malaria in pregnancy .................................................. 57 5.2.3. Health education on malaria......................................................................................... 59 5.3. Malaria in the context of multiple vulnerabilities in pregnancy (III)................................. 61 5.3.1. Explaining why malaria in pregnancy is not perceived as dangerous among the Yao.... 61 5.3.2. Recognition of economic and structural factors............................................................ 63 6 SCIENTIFIC CONCLUSIONS BASED ON EMPIRICAL FINDINGS ............................... 65 7 RECOMMENDATIONS FOR MALARIA PREVENTION PROGRAMMES...................68 8 DICUSSION ............................................................................................................................. 71 8.1. Challenges concerning utilization of qualitative research results ...................................... 71 8.1.1. Understanding bureaucratic aspects of international health organizations (IV) ............. 72 8.1.2 Evidence in programme planning and policy-making .................................................... 76 8.2. Applied medical anthropology and the way forward.......................................................... 79 9 RECOMMENDATIONS FOR FUTURE RESEARCH ......................................................... 83 10 REFERENCES....................................................................................................................... 85 ANNEXES: Annex 1: Main findings according to each publication.................................................................. 99 Annex 2: An example of a theme guideline for FGDs and IDIs .................................................. 105 Annex 3: KAP survey questionnaire ........................................................................................... 107

ORIGINAL PUBLICATIONS

viii

LIST OF ORIGINAL ARTICLES The dissertation is based on the following original articles, referred to in the text by their roman numerals I-IV.

I.

Launiala Annika and Teija Kulmala 2006. The importance of understanding the local context: Women’s perceptions and knowledge concerning malaria in pregnancy in rural Malawi. Acta Tropica 98: 111-117.

II.

Launiala Annika and Marja-Liisa Honkasalo 2007. Ethnographic study on factors affecting compliance to intermittent preventive treatment of malaria during pregnancy among the Yao women in rural Malawi. Transactions of the Royal Society for Tropical Medicine and Hygiene 101(10):980-9.

III.

Launiala Annika and Marja-Liisa Honkasalo 2010. Malaria, danger and risk perceptions among the Yao in rural Malawi. Medical Anthropology Quarterly 24(3): 399-420.

IV.

Launiala Annika 2009. How much can a KAP survey tell us about people’s knowledge, attitudes and practices? Some observations from a medical anthropology research on malaria in pregnancy in Malawi. Anthropology Matters Journal Vol 11(1):1-11.

ix

LIST OF FIGURES AND TABLES FIGURES Figure 1. Global distribution of malaria transmission risk…………………………………………...3 Figure 2. Timeline of the development of the malaria armamentarium………………………...........4 Figure 3. Effects of malaria in pregnancy on maternal, newborn, infant and child health…………..6 Figure 4. Adverse consequences of malaria in pregnancy in stable transmission areas......................7 Figure 5. IPT dosing schedule in the framework of FANC …...…………………………………….9 Figure 6. Country categorization by malaria control status and burden…………………………….10 Figure 7. Map of Malawi……………………………………………………………………………28 Figure 8. Map of Mangochi District………………………………………………………...............30 Figure 9. Map of Lungwena health centre catchement area………………………………...............45 Figure 10. Interwoven factors that affect prevention of malaria in pregnancy in Lungwena............67

TABLES Table 1. Summary of field visits, data collection methods, target groups and main themes……………………………………………………………………………………………….41 Table 2. Summary of local understanding of malaria in pregnancy………………………………...55

x

ABBREVIATIONS ACT ANC CDC CQ DDT DFID DOT EBM EBPM FANC FGD GAD GMAP GOM HQ HRAP IDI IPT IRS ITN IUGR KAP LBW LLIN LTHCP MDG MiP MIPESA MoHP NMCP NGO PSI PSSMC RDT SP STI TBA TDR TUMCHP UN UNICEF UNDP RBM VHC WHO WID

Artemisin-based combination therapy Antenatal clinic Centers for Disease Control and Prevention Chloroquine Dichloro-diphenyl-trichloroethane United Kingdom Department for International Development Direct observed therapy Evidence based medicine Evidence based policy making Focused antenatal care Focus group discussion Gender and development Global Malaria Action Plan Government of Malawi Headquarters Human Rights Approach in Programming In-depth interview Intermittent preventive treatment Indoor residual spraying Insecticide treated net Intrauterine growth retardation Knowledge, attitudes and practices Low birth weight Long-lasting insecticidal nets Lungwena Training Health Centre Project Millennium Development Goal Malaria in Pregnancy Malaria in Pregnancy Eastern and Southern Africa Coalition Ministry of Health and Population National Malaria Control Program Non-governmental organization Population Services International Partnership for Social Sciences in Malaria Control Rapid diagnostic test Sulfadoxine-pyrimethamine Sexually transmitted illnesses Traditional birth attendant Special Programme for Research and Training in Tropical Diseases Tampere University Mother and Child Health Project United Nations United Nations Children’s Fund United Nations Development Programme Roll Back Malaria Village Health Committee World Health Organization Women in Development

xi

1 INTRODUCTION between local reality, malaria prevention programmes and global malaria policies despite the national level programme efforts: particularly little attention was paid to examining the local socio-cultural context prior to programme planning and implementation. Thus I planned this applied medical anthropology study in order to better understand the socio-cultural factors that affect access to and utilization of preventive services of malaria in pregnancy in the community and health facility context from the perspective of pregnant women. Later I also explored these issues from the perspective of other actors involved in maintaining the wellbeing of a pregnant woman, namely husbands and female elders. In the thesis I draw upon the data I collected during fieldwork and upon my subjective experience working for UNICEF Malawi.

In the present study, I explore malaria in pregnancy among the Yao men and women in rural Malawi. Malaria in pregnancy is an enormous public health problem in the whole of sub-Saharan Africa. It has severe consequences for a pregnant woman and her unborn infant. An infant born to a mother with malaria is more likely to have low birth weight (LBW), which is the single greatest risk factor for death during the first months of life. The risk of maternal death also increases considerably as a pregnant woman suffering from malaria is likely to develop severe maternal anaemia. (Dellicour et al. 2010; WHO 2004.) Malaria is also a disease of the poor living in rural areas. The majority of more than a million malaria deaths per year occur among people living with less than a dollar per day, i.e., among “the bottom billion” (Hsiang et al. 2009). Malaria mostly affects pregnant women living in rural areas where access to health care and effective malaria treatment are greatly hampered by the poor state of the formal health care system. Today malaria control and prevention is a top global priority. The main strategies to reduce the burden of malaria in pregnancy are promotion of intermittent preventive treatment (IPT) and use of insecticide treated nets (ITNs) through the antenatal clinic platform. (GMAP 2008.) Yet, progress in reducing the burden of malaria in pregnancy has been slower than expected (Hill and Kazembe 2006).

I follow the hermeneutic research tradition, which aims to understand and interpret human practices and cultures, as well as to investigate meanings. I have grounded this research in theoretical frameworks in medical anthropology (Janzen 1976; Kleinman 1980) in order to gain a comprehensive understanding of malaria in pregnancy among the Yao in Lungwena and its surrounding villages in rural Malawi. My background as an anthropologist and involvement in development aid in Finland and in Malawi have considerably shaped my way of thinking and perceiving the world and influenced the way I position myself in the field of medical anthropology1. I have adopted an applied position that is common among medical anthropologists working in the field of international health (see

My motivation for this research is largely based on my work experience as a malaria project officer for UNICEF Malawi (19982001), where I was responsible for community-based malaria prevention projects, carried out in collaboration with the national malaria control programme (NMCP) and other international organizations. During this time, I became aware of how difficult it is to plan and implement meaningful programmes tailored to fit the local socio-cultural context that will lead to the expected impact on local people’s lives. I observed a clear discrepancy

1

I find Richard Parker’s (2002: 39-40) account of the factors that condition his interpretation and writing on “Administering the epidemic: HIV/AIDS policy, models of development and international health” illuminating. He describes how his involvement in international health working in the global programme on AIDS, being an anthropologist by training and being a gay man who has lived and worked with HIV and AIDS in the north and south, has shaped his views and affects the way he explains paradoxes related to the HIV/AIDS pandemic.

1

contribution of social scientific research to malaria and examine in more detail the few studies available on malaria in pregnancy, and based on the literature review I summarize the main research gaps in malaria in pregnancy. In Chapter 4, I elaborate on the study setting and methodology, which are very important in anthropological research. Chapter 5 consists of presentation of the results in relation to the current global strategies and on-going international discussions in malaria prevention outlined in the literature review. I also discuss gender aspects in the community and show how also men and female elders are crucial actors in taking care of the well-being of a pregnant woman. I further discuss the current emphasis on health education as an important tool to change behavior and to replace cultural misconceptions with appropriate information. I also show the importance of understanding the socio-cultural context in order to comprehend local people’s perceptions of malaria in pregnancy and how this may affect acceptance and utilization of preventive services. In Chapter 6, I present the broad scientific conclusions of the research, while the findings of each individual article are summarized in Annex 1. In Chapter 7, I present recommendations to improve malaria prevention programmes and particularly suggestions as to what type of information is needed for planning meaningful context sensitive programmes. Chapter 8 is a general discussion of the national and global context factors that hinder planning culture sensitive programmes and utilization of anthropological and qualitative research evidence. I attempt to link my subjective experience working in malaria prevention in Malawi to analysis of the factors affecting international health programme planning and policies. I also discuss the way forward in applied medical anthropology, and in the Chapter 9, I give some recommendations for future research in malaria in pregnancy.

Manderson 1998). This experience has also formulated my approach to a pragmatic direction, and an important goal has been to conduct research that would provide results that are translatable into practical contributions to be utilized in malaria programme planning. Because of my pragmatic approach, I decided not to write my thesis in the form of a monograph as is commonly done in anthropology, but instead to follow the public health and medical tradition of writing a thesis based on four publications in peer reviewed journals. I felt it was important to get the research results published in order to contribute to the ongoing international discussions on malaria prevention. I also write more as a public health person than an academic anthropologist. This is due to my background working in development aid, but also because I have made an effort to translate the results of this study understandable for a wide audience that goes beyond medical anthropologists, reaching professionals who work in the field of malaria prevention and infectious diseases because they are the ones at the ground level making the decisions and implementing the policies regarding malaria. Therefore, I have also published in tropical medicine journals (articles I, II) in addition to the anthropological journals (articles III, IV). I have organized the structure of the thesis as follows. I first take the reader through a literature review that encompasses a global perspective that includes latest biomedical evidence on malaria in pregnancy and the current global malaria in pregnancy initiatives, strategies and action plans. Secondly, I elaborate on the national context of Malawi, how evidence and global programmes have trickled down to national level programmes and policies, as well as on what kind of research on malaria has been conducted in Malawi. I find it important that the reader gets a good insight into the global and national contexts because I have developed the research objectives taking into account these two contexts. This perspective is also lacking in the individual articles because of the tight word limits of the journals. Thirdly, I review the 2

2 MALARIA IN PREGNANCY: A LITERATURE REVIEW

which at least 20% were among children under five years. (World Malaria Report 2009; see Figure 1. Global distribution of malaria transmission risk.)

2.1. Global perspective: International health, global strategies and policies Malaria is among the five top killer diseases worldwide, being the second in Africa, after HIV/AIDS. The global health burden of malaria is enormous. It is estimated that each year worldwide there are 300-500 million cases of acute illness due to malaria and approximately 1.1-2.7 million deaths due to malaria, mostly among children under five years of age. Half of the world's population is at risk of malaria, the majority of them living in endemic malaria areas in sub-Saharan Africa where 90% of all cases occur. In 2008 an estimated 243 million cases led to estimated 863 000 deaths in the region out of

Malaria control and eradication has a long history. From 1945 onwards, there were successful efforts to reduce malaria with dichloro-diphenyl-trichloroethane, popularly known as DDT. In 1955, the Global Malaria Eradication campaign was launched by the eighth World Health Assembly for all malarious countries, excluding countries in subSaharan Africa and Madagascar. The eradication campaign included indoor residual spraying (IRS), primarily with DDT, as a vector control tool, which was combined

Malaria endemicity Very high High Moderate Low No malaria

Figure 1. Global distribution of malaria transmission risk Source: World Malaria Report 2005 3

malaria-endemic countries etc. (Alilio et al. 2004; Keusch et al. 2010; Wernsdorfer et al. 2009.) In 1992 a Global Malaria Control Strategy was adopted as a response to the increased malaria burden and in 1998, the RBM Partnership was established to coordinate global efforts in combating malaria. Today long-lasting insecticidal nets (LLINs) and artemisinin-based combination therapy create new opportunities for large-scale malaria control. (GMAP 2008; see Figure 2. Timeline of the development of the malaria armamentarium.)

with case management. As a result of this eradication campaign, 37 of the 143 countries where malaria was endemic in 1950 were freed of malaria by 1978, 27 countries being in Europe or the Americas. In some countries, eradication efforts failed and by 1973 strategies shifted to long-term integrated control programmes and the Global Malaria Eradication campaign was abandoned. Little attention was paid to malaria over the subsequent years. Despite the end of the official WHO campaign, a number of countries have successfully eliminated malaria since that period, including Tunisia (1979), Maldives (1984), and the United Arab Emirates (2007). In the 1980s, malaria, mortality and morbidity began to increase again. This was attributed to the increase in parasite and vector resistance due to the current anti-malarial drugs and insecticides, the weakening of traditional malaria control programmes, rapid decentralization and integration into deteriorating primary health services, and the development of humanitarian crisis situations in many

1880 Plasmodium Falciparum discovered

1881 Plasmodium staining method developed

1887 Role of mosquitoes in malaria transmission discovered

1911 Mathematical model for malaria transmission developed

1932 Mepacrine for routine treatment introduced

2.1.1. Overview and consequences of malaria infection during pregnancy Malaria in pregnancy is an enormous public health problem particularly in malariaendemic areas in Africa. In sub-Saharan Africa alone, there are about 25 to 30 million pregnant women at risk of malaria infection

1959 Resistance to chloroquine in P. falciparum emerges 1945 IRS introduced

1985 ITNs introduced

1946 Chloroquine, amodiaquine, proguanil, and pyrimethamine introduced

1939 Insecticidal property of DDT discovered

1985 Mefloquine introduced

1972 Artemsinin discovered

Figure 2. Timeline of the development of the malaria armamentarium Source: Wernsdorfer et al. 2009: 96

4

2001 ACTs developed

1995 RDTs introduced

due to Plasmodium falciparum2 that causes multiple adverse consequences for the pregnant women, the foetus and the newborn (Dellicour et al. 2010; WHO 2004; see Figure 3. Effects of malaria in pregnancy on maternal, newborn, infant and child health). Although malaria during pregnancy puts all pregnant women at risk, clinical evidence shows that primigravidae are particularly more susceptible to severe forms of P. falciparum infection than multigravidae (Menendez 1995). Nevertheless, recent evidence also suggests that consequences for infants of multigravidae women may be greater than previously expected. (Desai et al. 2007). The burden of malaria infection during pregnancy is ever bigger with HIV coinfection (ter Kuile et al. 2004).

In stable transmission areas, the effect of malaria infection during pregnancy is associated with maternal anaemia. It has been estimated that approximately 26% of maternal anaemia cases are due to malaria during pregnancy. Malaria infection during pregnancy is also known to directly and indirectly cause maternal mortality, but to what extent, is less known. (Desai et al. 2007.) One estimate is that severe anaemia resulting from P. falciparum infection is responsible for approximately 10,000 maternal deaths every year in Africa (Guyatt and Snow 2001). The effects of malaria during pregnancy are also devastating for the newborn infant. According to estimates, approximately 20% of the low birth weight (LBW defined as birth weight below 2500g) deliveries are a consequence of malaria during pregnancy. LBW, in turn, is associated with poor infant development and survival, and contributes to around 100,000 infant deaths: 11,4% of neonatal deaths and 5,7% of all infant deaths or 17 deaths per 1000 live births according to another estimate. (Cottrell et al. 2005; Desai et al. 2007; Menendez et al. 2000; WHO 2004.) In addition, malaria during pregnancy is estimated to be responsible for 70% of intrauterine growth retardation cases (IUGR) and 36% of preterm deliveries. Placental malaria also increases the risk of stillbirth particularly during the third trimester of pregnancy (Desai et al. 2007). The main adverse consequences of malaria during pregnancy in stable transmission areas are summarized in Figure 4. Adverse consequences of malaria during pregnancy in stable transmission areas.

In sub-Saharan Africa, where the burden of malaria during pregnancy is the greatest, malaria transmission is predominantly stable 3. Characteristic of malaria infection during pregnancy in stable transmission areas is that it is often asymptomatic due to the preexisting immunity that has been acquired through frequent exposure to P. falciparum malaria infections since childhood. Absence of clinical symptoms such as fever makes it difficult to recognize the disease which therefore often remains untreated. Findings from clinical studies suggest that placental malaria is common and approximately one in four pregnant women has evidence of malaria infection at the time of delivery. (Desai et al. 2007.)

2

Malaria is an infectious disease caused by Plasmodium parasites, spread by a vector, the Anopheles-mosquito. P. vivax, P. ovale, P. malariae and P. falciparum are the four Plasmodium species capable of infecting humans. According to recent evidence (White 2008), a fifth Plasmodium species, knowlesi, has been discovered in South East Asia, namely in Malaysia. P. knowlesi is considered less severe than P. falciparum but more severe than the three other known species (ibid.). 3 Malaria transmission areas are divided into stable (moderate or high transmission) and unstable (or low) transmission areas. In stable transmission areas, malaria is endemic.

5

Recurrent/new plasmodium infections?

Maternal anaemia Cerebral malaria Severe malaria

Maternal outcomes

Maternal mortality?

1st trimester

2nd trimester

Conception

60 days

3rd trimester

28 weeks Abortion

Birth

28 days

1 year 5 years

Adult

Next generation

Stillbirth Preterm delivery/

Neonatal mortality

Low birthweight

Congenital malaria

Child outcomes

IUGR/low birthweight

Infant mortaility Infant anaemia Infant parasitaemia? Other infectious morbidity? Undernutrition?

Poor developmental/ behavioural outcomes? Increased risk of metabolic diseases Short stature (contributing to low birthweight in next generation?)

<5-year mortality

Figure 3. Effects of malaria in pregnancy on maternal, newborn, infant and child health4 Source: Desai et al. 2007: 95 4

According to Desai and colleagues (2007:95), the effect of malaria in pregnancy in the first trimester is unclear, possibly resulting in early abortions. However, exposure to antimalarials during the 1st trimester may also increase the risk of abortion. Little is also known about possible effects in the first year of life and throughout adulthood.

6

Acquired immunity – high

In absence of HIV infection, 1st and 2nd pregnancies are at higher risk

Asymptomatic infection high

Anaemia Placental sequestration Altered placental integrity

Less nutrient transport Maternal morbidity

Low birth weight

Higher infant mortality

Figure 4. Adverse consequences of malaria during pregnancy in stable transmission areas Source: WHO 2004

2.1.2. Global strategies related to malaria in pregnancy

WHO strategic approach to prevention and control of malaria in pregnancy

The burden of malaria in pregnancy has been an invisible public health problem for a long time because of its asymptomatic nature and it has therefore received relatively little attention among international organizations. Only recently the vulnerability of pregnant women and severe consequences of the disease for pregnant women and their infants have been recognized and efforts have been made to develop effective prevention and control strategies to mitigate the impact of the disease. During the past ten years, there have been several global initiatives and summits to accelerate the control of malaria and efforts to build up partnership with maternal health programmes and malaria control in pregnancy through focused antenatal care (FANC) and the WHO Making Pregnancy Safer Programme. (World Malaria Report 2008; Yartey 2006.)

The first approach adopted by WHO was to recommend a full antimalarial treatment during the first ANC visit followed by weekly chemoprophylaxis for all pregnant women in malaria endemic areas. Chloroquine (CQ) was the most commonly used drug for chemoprophylaxis. This strategy, however, had limitations that included widespread resistance of P. falciparum to CQ, pregnant women’s poor compliance with a weekly regimen throughout pregnancy, adverse effects and the bitter taste associated with CQ. Collaboration with malaria control and reproductive health programmes was also weak. (WHO 2004; Yartey 2006.) The second and current approach recommended by WHO in areas of stable P. falciparum transmission is a three-pronged approach that includes IPT, ITNs or LLINs and prompt case management agreed upon in the Abudja Declaration. IPT consists of pro7

in most of the African countries. Therefore, the ANC is also perceived as an excellent platform for accelerating the malaria in pregnancy strategy of IPT and ITNs (see Fig 5. IPT dosing schedule in the framework of FANC). This, however, requires strong collaboration between national malaria control programmes (NMCP) and safe motherhood programmes at all levels of the health care delivery system. The role of NMCP is to provide technical assistance and supervisory support to safe motherhood programmes that implement malaria in pregnancy strategies through the ANC services. (WHO 2004 Yartey 2006.)

viding all pregnant women with at least two preventive treatment doses of an effective, preferably single-dose antimalarial drug during routine ANC visits. The first dose should be given after quickening. Currently sulphadoxine-pyrimethamine (SP) is the most effecttive drug for IPT despite the increasing resistance of SP to P. falciparum. ITNs are particularly targeted at pregnant women and children under five as studies have demonstrated that wide-scale use of ITNs can reduce mortality and morbidity of children under five (D’Allessandro et al. 1996; Neville et al. 1996). There is also evidence that protection by ITNs can reduce the risk of LBW babies (ter Kuile et al. 2004). Efforts are made to scale up production and use of LLINs in response to the low re-treatment rates of ITNs. In addition, access to correct, affordable and appropriate antimalarial treatment within 24 hours of the onset of symptoms is supported and promoted, as is provision of iron supplements for aneamia. (WHO 2004; Yartey 2006.)

The Roll Back Malaria Initiative and Abudja Declaration In 1998, WHO, UNICEF, UNDP and the World Bank formed a Roll Back Malaria (RBM) partnership in order to provide a coordinated global response to malaria control aiming to halve the malaria burden by 2010. In 2000, 44 heads of state from malaria endemic areas in Africa met in Abudja and committed to the realization of the RBM goal. According to the commitment of the African states, known as the Abudja Declaration, it was agreed that 60% of pregnant women in malaria endemic areas should have access to effective treatment and prevention of malaria with IPT and ITNs by 2005. Later the target was increased to reach 80% of pregnant women by 2010. The core of the RBM initiative is a committed and co-coordinated partnership, which has expanded rapidly since its establishment, consisting now of over 500 partners ranging from malaria endemic countries to bilateral and multilateral development partners, the private sector, nongovernmental organizations (NGOs), and community-based organizations, foundations, and research and academic organizations. (RBM 2005-2015.)

Focused antenatal care and collaboration with safe motherhood programmes WHO also recommends a four-visit schedule for FANC. The first visit and enrolment should ideally take place during the first trimester (before 16 weeks). The second visit should take place in the sixth or seventh month (24-28 weeks), third visit in the eight month (32 weeks) and fourth visit in the ninth month (36 weeks). More visits can be scheduled if necessary. The FANC focus is on quality over quantity. All pregnant women are perceived to be at risk, and the main goal of the FANC is to assist women to maintain normal pregnancies through identification of pre-existing health conditions, early detection of complications, health promotion and disease control, birth preparedness and complication readiness. These basic services are adapted to and focused on each woman’s individual situation and needs. Approximately 70% of pregnant women attend an ANC at least once during pregnancy 8

WHO-recommended visits

2

3

IPT1

IPT2

1

4

IPT3 (optional)*

Rate of fetal growth**

Conception

Quickening

Birth Pregnancy

* In areas where HIV prevalence among pregnant women >10%, a third dose should be administrated at the last scheduled visit ** The rate of fetal growth is low within the first trimester of pregnancy, the rate increases rapidly in the second trimester and then declines in the last month of pregnancy

Figure 5. IPT dosing schedule in the framework of FANC Source: WHO 2004:11 control and elimination efforts (GMAP 2008; see Figure 6. Country categorization by malaria control status and burden.)

Again, in 2008, the world leaders and the global malaria community gathered on occasion of the 2008 Millennium Development Goal (MDG) Malaria Summit in New York to endorse an ambitious Global Malaria Action Plan (GMAP 2008). The current overall strategy is to reduce malaria morbidity and mortality by reaching universal coverage and strengthening health systems. The three components of the global malaria strategy are outlined in the GMAP as follows: 1) to control malaria to reduce the current burden and sustain control as long as necessary; this entails scaling-up for impact of preventive and therapeutic interventions, and sustaining control over time, 2) to eliminate malaria over time country by country and 3) to research new tools and approaches to support global

The GMAP (2008) emphasizes scaling up of IPT in pregnancy through FANC, paying attention to delivery of interventions for the prevention and control of malaria during pregnancy through IPT, LLINs, effective case management and other interventions such as iron supplementation against anaemia for ensuring healthy pregnancy outcomes as well as maternal and child survival. However, it recognizes several challenges that hamper scaling up of malaria in pregnancy (MiP) interventions, namely availability of MiP interventions in ANC clinics and regular and timely attendance throughout pregnancy. 9

interventions aiming to reduce the burden of malaria in pregnancy will require rigorous efforts to minimize missed opportunities and to ensure that all women attending ANC receive the necessary interventions for malaria prevention and control.

Despite the high ANC attendance in many countries, IPT coverage remains low, and substantially lower than ANC coverage, clearly demonstrating the extent of missed opportunities in the delivery of IPT and LLINs through antenatal care services. According to the GMAP (2008: 61), scaling up

Figure 6. Country categorization by malaria control status and burden Source: GMAP 2008 reduce extreme poverty. During the summit, eight time-bound targets, known as the Millennium Development Goals, were formulated and agreed to be achieved by 2015. MDG 4 and 5 focus on maternal and child health aiming to reduce the maternal mortality ratio by three quarters and under-five mortality by two-thirds of the rates in 2000. MDG 6 focuses on malaria and other infectious diseases (HIV, TB) aiming at halting/reversing the incidence of malaria by 2015. Core strategies to achieve these targets are promotion of skilled care at birth, use of effective malaria prevention and treatment interventions. Prevention of malaria in pregnancy is considered central to achievement of the MDGs 4 and 5 in malaria endemic areas. (Yartey 2006.)

The main obstacles to scaling up are considered to include persistent stock out of drugs for IPT, inadequately trained personnel, poor supervision and tools for effective monitoring and evaluation of programme effectiveness. Community engagement is considered crucial and therefore, communication and behaviour change methodologies are considered essential to ensure that the communities develop an appreciation of the need for skilled care during pregnancy and that women receive all the interventions they need in a timely manner with community support. (GMAP 2008.) Millennium Development Goals In 2000, world leaders from 189 countries came together in New York to adopt the United Nations Millennium Declaration commitment to a new global partnership to

10

the plan, strong tailored messages will be created to improve recognition of malaria symptoms and risk groups, as well as correct use of individual-level interventions, to address specific regional and community needs and to involve local leaders and participation of the population. These plans do not, however, separately recognize pregnant women and prevention of malaria in pregnancy. (GMAP 2008.)

2.1.3. Recognition of importance of sociocultural understanding of malaria During the past ten years, international health actors have recognized the importance of planning programmes based on understanding of socio-cultural aspects of disease prevention and local needs in order to ensure sustainable programmes. Therefore, particularly WHO has called for analysis of social and cultural determinants of women’s access to the ANC; use of the preventive services as the ANC forms the main platform for reduction of the burden of malaria among pregnant women, newborns and infants. WHO strongly advocates that African countries that have adopted the regional strategic framework for prevention and control of MiP should improve programme implementation through research on the multiple social and cultural factors that affect women’s perceptions, motivation and agency to access the ANC services. (WHO 2004:20-21.)

2.2. National perspective: Malawi 2.2.1. Burden of malaria in pregnancy in Malawi Malaria is endemic in Malawi, and each year approximately half a million women become pregnant and are at risk of malaria by P. falciparum, which causes over 85% of malaria infections in the country. Direct fatal consequences of malaria in pregnancy are rare because malaria transmission is stable in the country. Yet malaria in pregnancy has severe consequences. It causes severe maternal anaemia, the overall prevalence of which among pregnant women has been estimated to be as high as 66% (Malawi RBM 2004). Maternal anaemia is a major cause of maternal mortality which is also extremely high: 807 deaths per 100,000 live births in Malawi (UNICEF 2008). Malaria in pregnancy also leads to LBW, and studies show that approximately 20% of babies born are below 2500 grams. It has also been found that at the time of delivery, up to 40% of primigravidae and secundigravidae have placental malaria. (MDHS 2004; Rogerson et al. 2000.)

The RBM Plan 2005-2015 advocates understanding of the gender aspects of malaria prevention and household decision-making dynamics influencing the ability to access care. According to the RBM plan, there are three key issues that require addressing when planning, implementing and evaluating malaria prevention programmes and strategies. These are: 1) socio-cultural and behavioural factors that influence women’s and men’s approach to malaria prevention and treatment; 2) women’s and men’s access to different resources to pay for healthcare and women’s ability to pay on individual and household livelihoods; and 3) gender and decisionmaking levels. The Global Malaria Action Plan (GMAP 2008) emphasizes that service delivery is not just about delivering products but also about appropriate use. Communication and behavior change methodologies are considered central tools to ensure the appropriate use of interventions. Participation of service providers and intervention users in programme planning is considered important in order to improve health-seeking practices and care-providing behaviours. According to

The national estimate of prevalence of malaria in pregnant women is 19%. Yet the severity of malaria in pregnancy depends on the mother’s gravidity, age and intensity of malaria transmission that is particularly influenced by season and location. There are significant differences between districts in the country. Malaria transmission in the lakeshore area is high because of the presence of stagnant water pools. Thus pregnant women, particularly primigravidae and secundigra11

venues under direct observed therapy (DOT). 2) Any pregnant women with febrile illness for which other causes (urinary tract infections, respiratory infection, etc) cannot be found should receive a treatment dose of three tablets of SP to be taken under DOT. Women should be counseled that it is safe to take SP on an empty stomach. Women with known allergy to sulfa drugs (typically a skin rash) should be treated with quinine which is safe in pregnancy. 3) Pregnant women should be encouraged to sleep under ITNs to avoid mosquito bites.

vidae, living in Mangochi district are more likely to be infected by malaria than pregnant women living in Blantyre district, as a study conducted in the late 1980s showed. (Rogerson 2000.) The economic burden of malaria is enormous. In the Malaria Strategic Plan 2005-2010, malaria is defined as “a disease of poverty - it affects the poorest and keeps them poor.” At the beginning of the 1990s it was estimated that an average Malawian family spends up to 20-30% of their annual income on malaria treatment (Ettling et al. 1994). Malawian adults are not able to work an average of 25 days per year due to malaria, which directly contributed to loss of income. In addition, the cost of drugs to treat malaria easily burdens the poor families and depletes their scarce resources. According to estimates, a lowincome family spends more than one-quarter (28%) of their yearly income to treat malaria illness. (Malawi RBM Consultative Mission 2004.)

In Malawi approximately 93% of pregnant women attend the ANC at least once during their pregnancy. At the district level, prevention of MiP is integrated with the safe motherhood programme. In 2002, Malawi adopted the WHO recommended FANC package, according to which the recommended amount of visits is four for a normal pregnancy, scheduled at specific times during the pregnancy.

2.2.2. Prevention of malaria in pregnancy Commitment to the RBM Initiative and its targets

Malaria Policies and MiP programme implementation

Malawi has also signed the Abudja Declaration and committed to achieve the goals of the RBM initiative that is to reduce malaria morbidity and mortality in Malawi by half by the year 2010 and to reduce this burden by a further 50% by 2015. The specific objectives to be achieved by 2010 (malaria policy) are as follows: At least 60% of pregnant women have access to effective treatment within 24 hours of the onset of symptoms (up from 20% in 2001). At least 60% of women aged 15-49 sleep under ITNs which are accessible and affordable (up from 13% in 2001). At least 60% of all pregnant women receive presumptive intermittent treatment for malaria in pregnancy (up from 30% in 2001).

In 1993, Malawi adopted the WHO threeprong approach as one of the first countries in the sub-Saharan region and carried out a policy change regarding malaria in pregnancy. In the same year a decision was also made to change the first-line drug for treatment of uncomplicated malaria from CQ to SP as the first African country. In 2002, a national malaria policy was developed outlining the following policy on presumptive intermittent treatment5 in pregnancy: 1) All pregnant women should receive at least two treatment doses of SP at least one month apart at the ANC and at TBA 5

I apply throughout my thesis the term ‘intermittent preventive treatment’ instead of presumptive intermittent treatment because IPT is commonly used by all international organizations such as WHO, and also by Ministry of Health and Population (MoHP) in the various malaria documents.

12

Progress so far

The current target is that 90% of pregnant women have received two doses of IPT-SP and sleep under an ITN, as well as have access to appropriate and prompt treatment. (Malawi Malaria Strategic Plan 2005-2010.)

Estimates concerning the current coverage of MiP targets vary depending on the source. Concerning IPT-SP, a nationwide survey estimated that 38% of pregnant women received only one dose of SP, while 41.9% received two doses and 17.9% three or more doses. In addition, 42.9 % of households were reported to own an ITN and 34.1% of pregnant women were said to utilize and sleep under an ITN. (Kadzanzira and Munthali 2004.) According to a UNICEF funded survey, 67% of household owned an ITN and 47% of pregnant women slept under an ITN (UNICEF 2008). The recent Malawi MDG report (2008) only reports progress concerning ITN coverage. According to the report, there has been an increase from 42% to 58% regarding the number of households with at least one net. In addition, there has been an increase in the proportion of population with access to treatment from 14% in 2004 to 20% in 2006.

Major partners involved In the beginning the MiP programme was implemented vertically, and there was no collaboration between reproductive health and malaria control. Today collaboration between these two programmes has been reinforced through establishment of a MiP technical group, but occasional verticalism has occurred at the national level. At the district level, the MiP programme is integrated in the safe motherhood programme. According to an assessment of country experiences in the adoption and implementation of MIP policies in 2006, IPT was given DOT in 80% of the ANCs. (MIPESA 2006.) Major partners involved in MiP are the RBM partners including WHO, UNICEF, UNDP, World Bank, DFID, USAID/CDC and the Japan International Cooperation Agency. In addition, there are research institutions, Malawi College of Medicine as well as NGOs such as Population Services International (PSI), Africare, Médecins Sans Frontieres, World Vision International, Plan Malawi, Canadian Physician for Aid and Relief etc. The RBM partnership is considered to be dynamic, although there has been constraints regarding communication and information sharing6. The National Malaria Control Program (NMCP) also suffers from lack of staff, as does the whole health sector in the country, despite the commitment of the Government of Malawi (GoM) and RBM partners. (Malawi RBM 2004.)

Challenges identified Regarding implementation of the MiP programme and particularly IPT-SP at ANC, several challenges have been reported. These include staff attrition, late ANC attendance, lack of adequate cups and clean water, stock out of SP, perceptions concerning SP and its bitterness, limited understanding of purpose of IPT-SP and low uptake of the second dose of IPT (MIPESA 2006). The ITN scaling up has made progress, yet the human resources crisis, shortages of ITNs, poor access to ITNs, use of ITNs for other purposes than prevention of malaria and poor water sanitation are recognized as major weaknesses affecting ITN programme implementation (Malawi MDG Report 2008). Communication strategy aiming at behaviour change

6

My own experience (2000-2001) of the RBM partnership concurs with the observation of the partnership as dynamic. The RBM taskforce met regularly to discuss the implementation of RBM activities at the premises of MoHP with all the main UN partners, DFID and PSI were present. Inadequate communication and information sharing is a persistent problem in any development sector involving multiple donor agencies and national partners.

The Ministry of Health and Population has developed a malaria communication strategy as part of the Malaria Strategy Plan 20052010. The ultimate goal of this strategy is to 13

ensuring relevancy of research for programme planning, have hampered effective and frequent dissemination of research results among the international and national partners involved in malaria prevention. In addition to the NMCP, malaria research is conducted by several institutions including the Malawi College of Medicine in Blantyre, the Welcome Trust, the Centers for Disease Control and Prevention (CDC), the Blantyre Integrated Malaria Initiative, the Blantyre malaria project, the Centre for Social Research in Zomba and the various international organisations in Malawi. (Malawi RBM 2004.)

change people’s behaviour to better comply with and adopt the current malaria treatment and prevention strategies in order to achieve the set targets. The underlying starting point in the strategy is that Malawians just accept malaria and perceive it as something they just have to live and often die with. The emphasis is on raising awareness on the mosquitomalaria link as well as the malaria and poverty link. The strategy outlines that, considering the general population, there is a lack of awareness that malaria is a serious problem, there are misconceptions about how malaria is transmitted, misconceptions about antimalarials, misconceptions about risk (seasonality), beliefs and misconceptions that ITNs do not work etc. Regarding IPT, health workers and pregnant women share the misconception that antimalarials cannot be taken on an empty stomach and beliefs that SP can cause a miscarriage. Tools for achieving behaviour change are health talks, the radio, jingles and video shows. (NMCP 2005.)

Since the 1990s there has been a growing interest in investigating various clinical aspects of MiP; its adverse consequences and efficacy of drugs used in prevention of MiP in Malawi. A search in the web-based MiP library7 using ‘Malawi’ as a search term gives 171 references, with the majority being references of clinical studies. Findings from some of the studies have also influenced the malaria policy in Malawi. For example, research conducted by CDC in collaboration with MoHP on prevention of malaria in pregnancy led to a policy change as the study showed that IPT-SP was much more effective than CQ chemoprophylaxis (Schultz et al. 1994). Several subsequent studies confirmed that IPT-SP reduces the adverse consequences of malaria in pregnancy such as LBW and maternal anaemia (Steketee et al. 1996; Verhoeff et al. 1998).

2.2.3. Malaria research in Malawi The Government of Malawi promotes evidence-based policy-making. In the malaria policy (2002), operational research is recognized as a key intervention that will provide policymakers, programme managers and stakeholders with up-to-date information that can be used for improving and amending programmes. Regarding prevention of malaria in pregnancy, studies regarding factors influencing ANC visits are considered important because of the tendency to delay enrolment to the 2nd trimester. Also the Malaria in Pregnancy Eastern and Southern Africa Coalition calls for operational research “into technical, socio-economic and anthropological dimensions of MiP to ascertain what works and what doesn’t work.” (MIPESA 2006: 34.)

7

The Malaria in Pregnancy (MiP) Library is a comprehensive bibliographic database of published and unpublished literature relating to malaria in pregnancy that is updated regularly (accessed 1 Feb. 2010 http://www.updatesoftware.com/Publications/Malaria/). It also includes a trial registry of planned and ongoing trials. The MiP Library is a product of the Malaria in Pregnancy Consortium that is a five-year programme of research to evaluate new and improved existing interventions for the prevention and treatment of malaria in pregnancy.

A large amount of malaria research data exist in Malawi that is co-ordinated by the Research Unit of the MoHP. Nevertheless, inadequate resources of the Research Unit, as well as the lack of a central repository for research publications and of mechanisms for 14

Southern region), phungu and wimbwambwa (among Tumbuka and Tonga in the Northern region). In addition to malaria, malungo was used to refer to fever caused by other diseases making it difficult to be sure when malungo was used to mean malaria and when fever. Findings regarding local illness aetiology and treatment-seeking practices concerning children under five were similar to those of Bisika (1996) and Munthali (2005). This study also explored perceptions and practices regarding prevention of malaria, including traditional prevention methods and modern methods such as ITNs that is a key programme activity by UNICEF.

So far, some social scientific research has been conducted concerning socio-cultural aspects of malaria, particularly concerning children under five years of age (e.g. Bisika 1996; Munthali 2005). A rapid ethnographic study in Southern Malawi explored local illness beliefs and practices, particularly concerning malaria and its case management among children (Bisika 1996). This study revealed that the local terminology used to describe childhood illnesses is diverse. The local word most commonly used to refer to malaria in children is malungo, yet there are children who are labeled otherwise although they most likely suffer from malaria. Concerning convulsions, it was discovered that mothers explain convulsions as a separate indigenous illness kambanga, rarely associating it with severe malaria. Findings from a medical anthropology study among the Tumbuka in Northern Malawi showed that local illness perceptions, distance from care, unavailability of drugs at the health facility and perceptions of the ‘right’ treatment for febrile illnesses all contributed to treatmentseeking delays regarding malaria in underfive children (Munthali 2005). In addition, there are a number of unpublished qualitative studies on perceptions and utilisation of ITNs as part of the social marketing programme of PSI. Studies conducted in urban Blantyre district showed that respondents’ knowledge of the mosquito-malaria link was good, yet inconsistent net use was found in relation to seasons with fewer mosquitoes (Matinga 2000).

To date, there have been some surveys assessing coverage and utilization of IPT-SP and ITNs by pregnant women as well as surveys on malaria knowledge, attitude and practices among childbearing women (Holtz et al. 2004; Kadzanzira and Munthali 2004; Schultz et al. 1994). Schultz and colleagues (1994) found that the majority of recently pregnant women perceived malaria as a problem; 47% associated malaria with miscarriage, 15% with a small and weak baby, and 12% with stillbirth and 4% maternal anaemia. Kadzanzira and Munthali (2004) reported that 50.8% of heads of households considered both pregnant women and children under five to be most at risk for malaria and only 6.1% considered pregnant women to be most at risk. A recent study in Southern Malawi evaluating the impact of a community-based delivery of SP to increase the IPT-SP coverage discovered that while the communitybased delivery significantly increased the proportion of women receiving SP as per the policy, it also had unintended and unwanted consequences as the number of women attending ANC fell below 90%. (Msyamboza et al. 2009).

In 2001, an ethnographic malaria research commissioned by UNICEF8 was carried out in five districts: Nkhata Bay, Mzimba, Kasungu, Mangochi and Mwanza. This study showed that although the English word ‘malaria’ was widely used in all district, other local terms were also used such as malungo (among Chewa and Yao in the Central and

Qualitative and ethnographic research particularly on malaria in pregnancy has been scarce. At the end of the 1980s, a multimethodological research was conducted on compliance with malaria chemoprophylaxis in pregnancy. The research included an ethno-

8

During this time I was working for UNICEF Malawi and was responsible for the malaria prevention activities. I wrote the terms of reference for the study, and two Malawian medical anthropologists were commissioned to carry out the study.

15

graphic study of malaria in pregnancy, which contributed to explaining the local understanding of malaria in pregnancy and use of chemoprophylaxis in pregnancy before the policy change from CQ to SP. (Helitzer-Allen 1989.) Based on the findings, this study recommended that programme planners should learn about the factors affecting compliance before undertaking an intervention design and planning health education messages. In addition to the complex local malaria understanding and terminology (see Chapter 2.3.2.), it was found that CQ, the drug used as chemoprophylaxis, was feared due to its bitter taste and belief that it can cause an miscarriage (Helitzer-Allen et al. 1994). Further, a cost-effectiveness analysis of alternative interventions was conducted (Helitzer-Allen et al. 1993), the role of ethnographic research in malaria control was elaborated on (Helitzer-Allen et al. 1993) and differences between qualitative and quantitative data sets discussed (Helitzer-Allen and Kendall 1992).

2.3. Social scientific perspectives To date, virtually no medical anthropology research has been carried out on malaria in pregnancy. According to Williams and Jones (2004: 516-517), little is known about pregnant woman’s treatment-seeking practices during pregnancy, and willingness and ability to engage in preventive activities. The influence of household structure and concepts of vulnerabilities, or protection of the mother and the unborn baby have neither received attention among social scientists. Ribera and colleagues (2007) have, however, noticed that there are encouraging signs that interest among social scientists is growing. 9 There is a large amount of social scientific literature on malaria focusing on children under five years. Research has also been carried out regarding women from the perspective of mothers and caretakers of children. (Heggenhougen et al. 2003; McCombie 1996; Ribera et al. 2007; Williams and Jones 2004). Due to the scarce amount of literature specifically focusing on malaria and pregnant women, I have broadened the scope of this literature review to cover social scientific research on malaria in general in order to have a better overview of current status of social scientific malaria research.

Regarding the use of SP and compliance with IPT-SP, a few qualitative studies have been conducted in the urban areas of Blantyre District. Findings from these studies have suggested that the main factors affecting compliance are related to health staff’s knowledge of timing of SP, inadequate stock of SP and women’s unawareness of purpose of SP given at the ANC etc. The majority of women showed no problems in taking SP at the ANC, although there were some beliefs related to SP such as that one should not take SP without food and that SP leads to LBW, miscarriage and dizziness. (Ashwood-Smith et al. 2002; Masache 2002; Ngoma 1999.) The UNICEF study (2001) also touched upon the issue of IPT-SP revealing that SP was perceived too powerful for a pregnant woman, and complaints of nausea and discomfort because of the size of the tablets were voiced by some women.

9

For example, the Malaria in Pregnancy Consortium’s research programme also has a research team on anthropological studies on malaria in pregnancy. This research team is coordinated by a medical anthropologist, Robert Pool. The focus of these studies is on local perceptions, experiences and behaviours relating to pregnancy, illness, and in particular malaria in pregnancy. In addition, acceptability of new interventions and patterns of use of existing interventions and other health resources by pregnant women are explored as well as broader social and cultural factors that may affect interventions for malaria in pregnancy at facility, district, and national level (for more details see http://www.mipconsortium.org/).

16

used for all types of fever in addition to fever caused by malaria. Also studies in Tanzania (Winch et al. 1996; Kamat 2006) have demonstrated that the local word homa covers several types of fever, homa ya malaria only being one of them. A study conducted in rural Ghana (Agyepong 1992) showed that the symptom complex of asra closely corresponds to the clinical definition of malaria, however covering also fevers caused by other diseases common in the study area, yet difficult to distinguish from fever caused by malaria (see also Ahorlu et al. 1997). A study in Kenya showed that on the one hand, among the Mijikenda, homa is used to refer to many febrile manifestations while on the other hand, among the Luo, malaria is perceived as a distinct disease entity (Mwenesi et al. 1995). A rapid ethnographic study in Zambia (Williams et al. 1999) reported that community members recognized more than a hundred illnesses that cause fever in children, out of which malungo, cilunguzi or malaria were used to describe a syndrome associated with mosquitoes.

2.3.1. Contribution of social scientific research on malaria so far Local understanding of malaria and its causalities A starting point for investigating local understanding of malaria has been to explain local notions and illness taxonomy related to malaria. Several studies (Agyepong 1992; Kengeya-Kayondo et al. 1994; Mwenesi et al. 1995; Winch et al. 1996) have revealed that the local word and illness taxonomy used for malaria is broad and ambiguous, showing gaps and overlaps concerning the biomedical and lay conceptualisation of malaria. A common finding has been that malaria is conceptualised as a broad symptom complex with symptoms varying between general body weakness, chills, joint pains, lack of appetite, headache and fever etc. In the absence of a vernacular word for malaria, it is common that in the lay conceptualisation a local term commonly translated as fever into English is also used to refer to malaria. There are, however, exceptions, such as soumaya in Burkina Faso which is literally translated as ‘a state of being cold’ (Dugas et al. 2009; Okrah et al. 2002; Samuelsen et al. 2004). A study in Uganda found that the local word for malaria (omusujja) is also used synonymously with feeling unwell (Kengeya-Kayondo et al. 1994) similar to findings in Tanzania (Hausmann-Muela et al. 1998) where the local notion (homa) has a broader meaning of general malaise or diffuse body pains.

Explaining malaria causalities Investigation of local illness causalities and particularly people’s knowledge of the mosquito-malaria link has been central for studies aiming to understand local perceptions concerning malaria. The majority of studies demonstrate that normal malaria 10 is perceived as a mild illness, and attributed to natural causes.11 A few studies have shown

10

Further, studies show that in many countries there are different types of feverish illnesses overlapping with the symptoms and aetiology of the biomedical definition of malaria. For example, Kengeya-Kayondo and colleagues (1994) reported that community members were aware of several types of fever (omusujja), some being related to specific disease entities, others to specific symptoms and yet others to specific aetiologies such as omusujja caused by mosquitoes. Thus in everyday conversation the word omusujja is

Many studies show that people make a distinction between normal and severe malaria. Malaria is defined ‘normal’ when the illness manifests itself as mild feverish illness. Malaria becomes severe when severe symptoms are present such as high and persistent fever combined with vomiting, coughing and severe headache. (Williams and Jones 2004.) 11 In medical anthropology illness causalities have been commonly investigated utilising Foster’s (1976) dichotomy of naturalistic and personalistic factors. Less utilized has been Young’s (1976) theoretical framework to investigate causalities through so-called internalized and externalised factors. Anthropologists have also debated whether most causes are interpreted as resulting from natural or supernatural causes. Pool

17

are perceived to be caused by witchcraft and the linkage with cerebral and severe malaria is not well known. Illnesses with convulsions are commonly classified as a specific indigenous children’s illness. (Baume et al. 2000; Hausmann-Muela et al. 1998; Kamat 2006, 2008; Okrah et al. 2002.)

that in some communities people have not been very aware of the mosquito-malaria link and transmission mode (Agyepong 1992; Mwenesi et al. 1995). However, a large number of studies have found that in many communities people know well that normal malaria is caused by mosquitoes, although they may be unclear about the actual transmission mode. In these communities it is also common that in addition to mosquitoes, malaria is attributed to several other natural causes such as working hard, drinking dirty water, overeating, over-consumption of sweet foods, wind, rain, excessive exposure to sun etc.12 (Ahorlu et al. 2007; Hausmann-Muela et al. 1998; Nuwaha 2002; Ohrah et al. 2002; Samuelsen et al. 2004; Winch et al. 1996). Nevertheless, a study conducted in Tanzania (Hausmann-Muela et al. 2002) showed that normal malaria can be associated with supernatural causes such as spirits and witchcraft (see also Ahorlu et al. 2007 and HelitzerAllen et al. 1993). Hausmann-Muela and her colleagues (2002) described how the community explained that witches can cause “fake” malaria by imitating the symptoms of malaria, or by linking the two aetiologies syncretically13 with each other, showing how indigenous illness causalities and biomedical explanatory models interact and produce hybrid ideas (Nichter 2008: 44).

Treatment-seeking practices There is a large amount of research on the treatment-seeking practices particularly among mothers and other caregivers of children under five from different parts of subSaharan Africa. Findings from these studies show that mild fever in children (corresponding to uncomplicated malaria) is often managed at home with tepid sponging, traditional herbs and/or pharmaceuticals bought from drug vendors (Ahorlu et al. 1997; Baume et al. 2000; Kengeya-Kayondo et al. 1994). When the symptoms of illness change and become severe, including high fever, vomiting and diarrhoea, treatmentseeking practices often change from home management to seeking care outside the home. Treatment options often vary from government health facilities to private clinics. Distance to the nearest health care facility is as an important factor. (Agyepong and Manderson 1994; Tarimo et al. 2000.) By contrast, treatment for symptoms such as convulsions and unconsciousness (often interpreted as an indigenous children’s illness) is often sought from different types of traditional healers (Ahorlu et al. 1997; Hausmann-Muela et al. 1998). A general conclusion has been that beliefs associated with convulsions and supernatural causations is a main cause for delaying prompt biomedical care that could be life saving. However, a recent study in Tanzania by Kamat (2008) showed that although there is a shared cultural knowledge regarding degedege (a life threatening indigenous folk illness, cerebral malaria from the biomedical perspective), there are great variations in the treatmentseeking practices of parents, calling attention to the micropolitics of therapy management. There is also a vast amount of studies which

Although studies show that normal malaria is perceived as a mild illness with little consequences for everyday life, people are aware that malaria can be severe and have drastic consequences particularly for children under five. Many studies describe how convulsions (1994) argued that most illnesses are due to witchcraft while Green (1999) argued the opposite. 12 This is called multiple causality, in other words, an illness is believed to be caused by any one of the several causes or sometimes by a combination of several causal factors (Nichter 2008:42). 13 Syncretic interpretation of illnesses causalities refers to cumulative interpretation; people add and combine explanations when trying to make sense out of the illness (Nichter 2008: 44, 62). See also Pool and Geissler (2007: 43-45) about criticism of medical systems (Kleinman 1980) and discussion about medical pluralism and syncretism.

18

and January 2010 I conducted a systematic literature search. As a result I selected 17 studies (some published by the same author) for a closer review (see also Ribera et al. 2007 for an overview). I utilized the data bases of the PSSMC as well as the Malaria in Pregnancy Library of Malaria in Pregnancy Consortium that had 4120 bibliographic references at the end of 2009, the majority, however, being references of clinical and quantitative survey studies. I also searched for publications using PubMed, Medline, AnthroSource and JSTOR journal collection, and browsed through particular journals published in Africa. There is a possibility that I have missed some medical anthropology studies of malaria in pregnancy which are published as monographs, or are unpublished Ph.D dissertations. I used broad selection criteria in order to reach as many as possible of the very few social scientific publications. My criteria included 1) studies focusing on malaria in pregnancy or pregnant women and malaria; 2) studies conducted in sub-Saharan Africa,14 3) studies using an ethnographic approach, qualitative methods and/or a mixture of qualitative and quantitative methods. I excluded surveys utilizing only quantitative methods and/or studies assessing coverage of the MiP interventions. I have summarised below the findings from the reviewed publications according to the thematic areas emerging from the review as follows: 1) perceptions concerning malaria in pregnancy, 2) illness classification and perceived causalities, 3) biomedical knowledge of adverse consequences 4) perceptions, knowledge and use of medication during pregnancy, 5) compliance with intermittent preventive treatment, and 6) timing and utilisation of the antenatal services.

have described the sources of care, patterns of treatment-seeking for uncomplicated malaria, type and use of traditional healers, perceptions and utilisation of pharmaceuticals from drug vendors and shopkeepers etc. (see Heggenhougen et al. 2003; McCombie 1996; Williams and Jones 2004 for an overview of social scientific research on malaria.) Many sub-Saharan African countries have recently made changes in their drug policy from SP and CQ to artemisin based combination therapies (ACTs) in order to overcome the drug resistance of P.falciparum. This has already received attention among medical anthropologists. A recent study in Cote d’Ivoire (Granado et al. 2009) showed, contrary to some previous findings (Mwenesi 1994; Nichter 1989), that pharmaceuticals bought from private-sector vendors involved no interaction between the customer and the vendor, and hence, did not shape the explanatory model (Kleinman 1980) or drug choices made by the customer, suggesting implications for the implementation of the new drug policy on ACTs. Kamat (2009) explored in Tanzania mothers’ perceptions of antimalarials such as SP and discussed implications for the new ACT drug policy, calling for caution when implementing a ‘one drug fits all’ approach to malaria control. 2.3.2. Malaria in pregnancy: an understudied topic in medical anthropology As a starting point for planning this study focusing on malaria in pregnancy (spring 2002) I conducted a literature search that to my surprise resulted only in a handful of publications. I also asked the members of the Partnership for Social Sciences in Malaria Control (PSSMC) network if anyone knew about relevant publications or even unpublished studies of MiP. Some members confirmed that there is not much social scientific research on MiP available, yet there was willingness and intention to commence such studies in Tanzania, Mozambique and Zambia by these members. Again in December 2009

Perceptions concerning malaria in pregnancy Community perceptions concerning malaria in pregnancy, particularly whether pregnant 14

There are very few studies on MiP or pregnant women outside of sub-Saharan Africa, none of them based on ethnographic fieldwork (see also a review made by Purohit and Mahapatra 2009).

19

Illness classification and perceived causalities of malaria in pregnancy

women are perceived to be vulnerable to malaria and whether malaria in pregnancy is perceived as dangerous to pregnant women or not, are important from a programme perspective because they influence motivation to use preventive measures (Manderson 1998; Nichter 2008: 47-50). So far, not much is known about local perceptions, and the information available is vague, containing little details. A number of studies have touched upon the issue and found that malaria (referring to normal fever) is perceived as a common illness for pregnant women (HelitzerAllen et al. 1993; Ndyomugyenui et al. 1998; see also Winch et al. 1996). Helitzer-Allen and her colleagues (1993) found in rural Malawi that having malungo during pregnancy was perceived as normal, particularly just before and after delivery. In addition, malaria attacks during pregnancy were not considered to need any treatment; the disease would cure on its own. A study in rural Ghana (Ahorlu et al. 2007) found that normal malaria is considered a frequent illness during pregnancy that is easily treatable with herbal and/or biomedical medicines. A study in Uganda (Mbonye et al. 2006a; Mbonye et al. 2006b) showed the opposite; malaria was perceived to be a leading cause of morbidity and mortality in pregnancy, except among primigravidae and adolescents. Yet women perceived a fever (omusujja gwa nakawere) with symptoms such as mild fever, breast pain and lower abnormal pain, to be common and normal among newly delivered women. The researchers considered it alarming that women consider fever after delivery normal because it can be a sign of pelvic infection which is one of the common causes of maternal mortality. 15

Local notions and illness taxonomy related to malaria in pregnancy have received some attention in research. A study in Uganda (Mbonye et al. 2006a) concluded that malaria in the communities was known as omussujja (fever) and omusujja caused by mosquitoes, although it was used to a much lesser extent. The study also found that respondents ranging from pregnant women, opinion leaders, midwives, retired female teachers to TBAs etc. made a distinction between mild malaria (olusujjasujja) presenting with multiple mild symptoms and severe malaria (omusujja omungi) presenting with severe symptoms such as high fever, rigors, shivering etc. A study in rural Ghana (Ahorlu et al. 2007) illustrated how three different types of fever were distinguished, namely malaria fever, jaundice fever and yellow fever. Fever associated with malaria was further defined as common fever that is easy to treat and male fever that is difficult to cure and may result in mad behaviour. Findings from an ethnographic study in Malawi (Helitzer-Allen et al. 1993) are comprehensive and detailed concerning the local understanding of malungo, a local word used to refer to malaria by the professional sector (Kleinman 1980) and lay people in their everyday conversation. Helitzer-Allen et al. (1993) concluded that on the one hand, malungo (fever) is a term glossing several ‘malungo diseases’ each with distinct symptoms and aetiology and, on the other hand, malungo is used to describe the symptoms of these ‘diseases’. Hence malungo can be used to refer to malungo caused by mosquitoes (malungo wa udzuzu) or any other type of malungo, even malungo caused by evil spirits (malungo wa majini). Studies have particularly focused on exploring community knowledge of the mosquitomalaria link, and in many cases a comprehensive analysis of broader causalities and understanding of local illness aetiology is lacking (Foster 1976). Findings show that in some areas such as rural Nigeria (Okonofua et al.

15

Kengeya-Kayondo et al. (1994) and Nuwaha (2002) reported that malaria in pregnancy is perceived as a dangerous disease for pregnant women in the communities although the focus of the study was not on malaria in pregnancy.

20

have been explored to some extent. Findings from rural Ghana (Ahorlu et al. 2007) suggested that among the community members there was some recognition of the linkage between malaria and anaemia (shortage of blood). Studies in Uganda (Mbonye et al. 2006a; 2006b) showed that the majority of respondents associated severe malaria and miscarriages, but not less severe malaria and stillbirths, anaemia and LBW. Findings from rural Sudan (Adams et al. 2008) showed that both pregnant women and midwives had limited understanding of the consequences, maternal anaemia being the best known adverse consequence. In Gambia (Brabin et al. 2009) both women and men knew that malaria during pregnancy can have direct consequences such as anaemia and pre-term birth or indirect consequences such as the foetus falling ill due to poor nutritional status of the pregnant woman.

1992), the level of knowledge between adolescent pregnant girls, men and traditional healers varied, and particularly men and traditional healers did not attribute malaria to mosquitoes. Instead, staying in the sun too long, bad water and indiscriminate sex were perceived as main causes of malaria. By contrast, studies in Sudan (Adam et al. 2008) and in Ethiopia (Belay and Deressa 2008) found that the level of knowledge concerning mosquitoes as a causal agent was good. However, besides mosquitoes, it was also common that other causal attributes were identified such as flies, stagnant water, eating stalk of maize/sorghum, exposure to cold air, drinking dirty water and working in the sun (Adam et al. 2008; Belay and Deressa 2008). Findings from a study in rural Ghana are similar illustrating that in the study communities multiple causalities are associated with febrile malaria illness among which mosquito bite and heat from sun/fire were most prominent. Also evil forces were recognized as one of the causes of febrile malaria illness. (Ahorlu et al. 2007.)

Perceptions, knowledge and use of medicine during pregnancy Very few malaria studies have explored perceptions and use of herbs during pregnancy. Findings suggest that particularly bitter tasting herbs during pregnancy are often prohibited due to their abortifacient characteristics, as was found in Malawi (Helitzer-Allen et al. 1993) and in Uganda (Ndyomugyenui et al. 1998). One study in Gambia, however, reported that a certain herb could be taken even during the first trimester while another herb could not be taken at all. A study in Uganda (Mbonye et al. 2006a) stated that no known fear was associated with the use of traditional herbs during pregnancy. According to a study in rural Ghana (Ahorlu et al. 2007), pregnant women are not allowed to take either herbal medication or pharmaceuticals without the advice of ‘experts’.

Helitzer-Allen and her colleagues (1993) have thoroughly investigated the local understanding of malaria aetiology. According to the findings, the illness complex malungo comprises a total of seven subcategories with distinct aetiologies ranging from natural causes such as mosquitoes, rain, wind, hard work, other airborne methods, dirty water and food, kulipuka (a parent suffering from mauka will get malungo) to supernatural causes, namely majini (spirits and witchcraft). It is also possible to suffer from a combination of naturally and supernaturally caused malungo, and the causality could be re-interpreted from natural into supernatural due to the complex local transmission aetiology. Thus in the latter case the interpretation of causalities is syncretic.

Hardly any of the malaria studies have reported general perceptions and knowledge of the use of pharmaceuticals during pregnancy. A study in Gambia (Brabin et al. 2009) revealed that men’s knowledge mainly concentrated on describing the colour of medication

Biomedical knowledge of adverse consequences of malaria in pregnancy Community perceptions of adverse consequences concerning malaria in pregnancy 21

gained of perceptions and use of SP during pregnancy. A study in Uganda (Mbonye et al. 2006a) showed that SP is perceived as a very strong drug and therefore, fears of miscarriage and foetal abnormalities are associated with its use during pregnancy. A major barrier to the use of SP was its cost: it is an expensive drug compared with CQ, and it is recommended to be taken with extra fluid such as juice, which further increases the cost. However, most people considered SP an effective drug to treat malaria in pregnancy if CQ treatment fails.

whereas not much was known about the names and purpose of medication. Misconceptions also existed, and some men thought that paracetamol may cause abortion. The women did not know much about the drugs they were given at the ANC either, but despite the unclarity, they accepted the drugs because they were given by a nurse or a doctor, and SP given at ANC was perceived as a safe drug. Findings from a recent study in Ghana (Smith et al. 2010) are similar as it was found that drugs received at the ANC were described by some women according to names and by others according to colour and size. Women’s knowledge of antimalarials also varied; some made no distinction between different drugs, others did not know the purpose of drugs, yet others thought drugs were for treatment or prevention etc. Nevertheless, receipt of drugs was reported to be among the main motivating reasons to attend the ANC. Helitzer-Allen and her colleagues (1993, 1997), who conducted a more detailed investigation into community perceptions of medication, showed that medications were described by their characteristics; taste, smell, perceived efficacy, size, form and colour. Sweet drugs were generally perceived as good and bitter tasting as bad. Medication use during pregnancy was more restricted than at other times, and all bitter tasting medications, medication perceived as strong, capsules and bitter herbs were considered prohibited because they can potentially cause a miscarriage.

Understanding and explaining compliance with intermittent preventive treatment IPT with SP is recommended for prevention of malaria in pregnancy in malaria endemic areas in sub-Saharan Africa. The pressure to monitor achievement of the RBM goal for IPT-SP has particularly promoted the assessment of IPT-SP coverage. Findings indicate that in many countries coverage is low and insufficient (see Hill and Kazembe 2006 for an overview). Nevertheless, factors contributing to low coverage are not well known because significantly less attention has been paid to exploring the multiple factors influencing compliance from a broader perspective that allows understanding the diversity of factors from the perspective of health systems, pregnant women and the community. A small number of studies have investigated understanding the purpose of IPT and factors influencing compliance with chemoprophylaxis. A study in rural Ethiopia found that there was an apparent lack of knowledge of reasons for using chemoprophylaxis during pregnancy. Findings suggested that factors influencing low compliance were lack of knowledge about availability of CQ and fear of CQ causing a miscarriage, heavy household duties and insufficient time for attending to health care needs and long distances to access services. (Ghebreyesus et al. 1996.) Helizer-Allen and Kendall (1992) showed that the utility of CQ as a preventive measure was

The main attention has been devoted to exploring perceptions concerning drugs used for IPT such as CQ and SP. In the 1990s, CQ was widely used in Africa for prophylaxis and for IPT, as well as in first-line treatment for clinical malaria. Some studies have reported perceptions concerning CQ, namely that it is perceived as a bitter drug and feared because of its potential to cause a miscarriage. Studies have reported pregnant women’s unwillingness and fear to take CQ (Ghebreyesus et al. 1996; Helitzer-Allen et al. 1993; Ndyomugyenui et al. 1998; see also Mbonye et al. 2006a). Some knowledge has been 22

weeks), and due to late ANC enrolment, there is a danger that a pregnant women does not receive the second dose of SP in time. Some of the malaria studies have explored factors influencing the timing of the first ANC visit and utilization of the services. In Uganda (Mbonye et al. 2006a; Ndyomugyenyi et al. 1998), where use of ANC services is low and irregular, it has been found that factors affecting use of ANC services varied from inadequate services, no need to visit ANC when healthy, husband’s reluctance to waste money on transport to ANC when wife is healthy, ignorance and high tendency to use herbs during pregnancy, lack of drugs at ANC, routine ANC check-up not perceived as beneficial particularly if no drugs received, nurses’ negative attitude, long distance to having no money for drugs. In Tanzania national and district level malaria officers attributed late enrolment to individual and community-based socio-cultural factors as well as to health system based factors such as quality of services and structural factors such as poverty and distance from services (Mubyazi et al. 2008).

perceived to be low. The majority of respondents considered malungo unpreventable and therefore CQ should not be taken as a prophylactic, particularly because it was also considered a dangerous drug for pregnant women. Studies of IPT-SP suggest that knowledge of IPT-SP and its benefits during pregnancy is low (Adams et al. 2008; Brabin et al. 2009; Mbonye et al. 2006a) and use of SP during pregnancy is mainly associated with treatment of malaria (Mbonye et al. 2006a). Furthermore, pregnant women themselves have little awareness of the timing of SP doses in pregnancy, as findings from Uganda (Mbonye et al. 2006a) and Gambia (Brabin et al. 2009) show. Problems concerning health workers’ knowledge and skills have also been observed in Tanzania (Anders et al. 2008; Mubyazi et al. 2008). Some studies have revealed health system based factors that contribute to low compliance, including insufficient drug stocks (Anders et al. 2008; Mubyazi et al. 2008), health workers’ inadequate training in IPT guidelines (Anders et al. 2008), understaffing, inadequate skills and poor motivation (Mubyazi et al. 2008) and lack of clean water to carry out DOT on IPT (Mubyazi et al. 2008). In addition, late enrolment in ANC was seen as a major causal factor for delayed IPT-SP in Tanzania (Mubyazi et al. 2008). Another study in Tanzania (Anders et al. 2008) showed that timely uptake of IPT depended on health workers’ performance rather than pregnant woman’s individual characteristics or timing of the first ANC visit.

Studies have placed little emphasis on exploring pregnant women’s motivation to attend the ANC. In Uganda it was found that women attended the ANC mainly to learn the position of the baby, to receive tetanus vaccination, treatment when sick and to receive the ANC card (Ndyomugyenyi et al.1998). A study in Ghana showed that pregnant women’s motivation to attend the ANC was strongly influenced by their own health and that of their baby. In addition, women had a strong trust in the health staff and accepted drugs and blood test, not always understanding the biomedical purpose (Smith et al. 2010).

Timing and utilization of the ANC services ANC forms an important platform for implementation of MiP interventions. Findings from Uganda (Ndyomugyenui et al. 1998), Kenya (Hamel et al. 2001) and Malawi (Holtz et al. 2004) show that ANC enrolment tends to be delayed to the second trimester of pregnancy. Yet timing of the first ANC visit is crucial as the first dose of SP as per the IPT policy should be given at four months (16-20 23

Today many of the women attend an ANC that has adopted the WHO recommended focused antenatal care package and delivers MiP prevention interventions as part of the package.

Contributions of pregnancy and maternal health ethnographies to malaria research It is important to recognize that malaria in pregnancy is interwoven with and inseparable from pregnancy and maternal health issues. Therefore, anthropological research on pregnancy and maternal health provides important insights into and lessons learned from understanding different aspects of pregnancy and motherhood; factors influencing use of antenatal services, treatment-seeking practices during pregnancy and the role of community in maternal health among other things (e.g. Allen 2002; Chapman 2006; MacCormack 1982). Marchia Inhorn’s (2006) review article “Defining women’s health” gives an excellent overview of the many inspiring ethnographies touching upon maternal health issues that are also very central for research on malaria in pregnancy.

There is a large amount of social scientific research that focuses on malaria in children under five years and women as caregivers. So far, the main focus has been to examine the public health knowledge gaps among the study population. Contributions from studies on malaria in pregnancy are still rather meagre, but some general points can be summarised as follows: Perceptions of malaria in pregnancy vary from dangerous to common and not that worrisome illness in pregnancy. Also findings from some studies suggest that fever during pregnancy is perceived as normal, particularly before and after delivery.

2.4. Summarizing what is known and the ‘research gaps’

Local notions and illness taxonomies regarding malaria in pregnancy have revealed differentiation of mild and severe malaria/fever and that the local term used for malaria encompasses several feverish illnesses.

Public health and biomedical evidence show that more than 25 million pregnant women live in a malaria endemic area in sub-Saharan Africa. Each year millions of these women suffer from malaria during pregnancy, with those being pregnant for the first or second time and/or those suffering from HIV/AIDS being most vulnerable to the disease. Due to the asymptomatic nature of malaria in pregnancy, many of the pregnant women may not even be aware of being sick or know that suffering from asymptomatic malaria might have adverse consequences for them as well as for the pregnancy outcome. Many of these women become anaemic, some of them more severely than others. As a consequence of anaemia combined with poor nutritional status they may give birth to a baby with LBW. Birth weight below 2500g increases the morbidity of the baby and the likelihood to die before the age of five. In the worse case scenario severe anaemia during pregnancy can lead to maternal death. The majority of these 25 million pregnant women attend an antenatal clinic at least once during their pregnancy.

Studies have concentrated on investigating the malaria-mosquito link, showing differences in knowledge regarding mosquitoes as a vector, and reporting other causal explanations such as flies, stagnant water, sun and overworking. Studies of perception and use of medication have mainly concentrated on antimalarials such as CQ and SP, and on examining factors that may affect a pregnant woman’s motivation to comply with IPT-SP. Studies report unwillingness to take CQ and SP because of their bitter taste that is associated with the fear of miscarriage and foetal abnormalities. Studies have also reported limited knowledge regarding benefits of IPT-SP and that there are service related factors 24

that hamper implementation of IPT-SP at the ANC.

perceptions and knowledge of the community.

Regarding timing and utilization of ANC, it has been found that pregnant women tend to delay enrolment, which affects appropriate timing of the first SP dose. Reasons associated with delayed enrolment range from lack of drugs, poor services to cultural beliefs.

So far, the focus on disease has been dominant in malaria in pregnancy studies and as a consequence, no attention has been paid to examining malaria in the context of pregnancy and in relation to other potential dangers to a positive pregnancy outcome.

All in all, not much is known about malaria in pregnancy as to its complexity and burden for the communities in resource poor setting. Among medical anthropologists, malaria in pregnancy has remained an understudied topic. To date, many of the medical anthropologists have concentrated their efforts on studying malaria in children, and treatment-seeking practices. There are several research gaps that require detailed examination and interpretation of results based on understanding of the socio-cultural context. The main gaps can be summarised as follows:

Little is known about pregnant women’s and community members’ knowledge of adverse consequences of malaria in pregnancy such as miscarriage and LBW, or the association with maternal anaemia. Studies have concentrated on exploring what pregnant women do not know, while attention should be paid to exploring what they know, including local indigenous knowledge. Little is known about pregnant women’s treatment-seeking practices when suffering from malaria, or about their perceptions and practices regarding the use of medications (pharmaceuticals and traditional medicine) during pregnancy.

Virtually no attention has been paid to gender aspects of malaria in pregnancy. Little is know about a pregnant woman’s agency to maintain the well-being of herself and the unborn baby and agency to access treatment and preventive services when needed. Household and extended family members involved in maintaining the well-being of a pregnant woman have not received any attention. Almost nothing is known about household members’ roles and responsibilities in relation to pregnant woman and malaria control.

The present study was planned to fill these research gaps and to gather contextual information that could be used for programme planning purposes and for planning future studies in Malawi and elsewhere in malaria endemic areas.

Little is known about pregnant women’s and community members’ understanding of malaria in pregnancy. The majority of studies fail to go beyond simplistic knowledge, attitude and practice studies to gain deeper insights into the perceptions and explanations behind them, to understand why malaria in pregnancy is or is not perceived as a dangerous illness, and to examine what factors contribute to the

25

3 OVERALL PURPOSE AND OBJECTIVES OF THE STUDY 2. To explain the local understanding of malaria in pregnancy and to elucidate what type of knowledge concerning malaria in pregnancy exists in the community and how this knowledge is generated (I,II,III);

The overall purpose of this study is two-fold: first, to gain socio-cultural understanding of pregnancy related issues, practices and perceived vulnerabilities from the perspective of Yao women and men, and second, to explore encounters and non-encounters between local reality and national and global level factors, and how these affect acceptance and utilisation of malaria in pregnancy interventions among pregnant women, as well as planning meaningful and context-specific malaria prevention projects.

3. To examine malaria in the context of perceived pregnancy related vulnerabilities, to gain insights into factors that weigh into the local understanding of dangers in pregnancy and to explain how malaria in pregnancy is related to them (III);

The specific objectives of the study are as follows:

4. To discuss global and national level factors that affect linking research, practice and policy, and utilization of social scientific research evidence in programme planning and policymaking.

1. To examine household and extended family member participation in taking care of the well-being of a pregnant woman; their perceived roles and responsibilities affecting pregnant women’s access to care and preventive services at the antenatal clinic, as well as the perceived constraints that hinder them to perform their roles (II, III and unpublished data);

26

4 STUDY SETTING AND METHODOLOGY 4.1. Entering the field and the research site leagues, I employed James and Shemu directly. I never had any cause to regret my decision because these two young men were bright, extremely trustworthy and hardworking. During the time they worked for me they became fathers, and suddenly my responsibilities as an employer grew as I also became responsible for the well-being of their families. Victor was an older man. According to his story, he was originally trained as a school teacher but his certificates had gone missing and he could no longer work as a teacher. Therefore, he came to seek work with us and we felt pity for him and decided to give him a job as a guard. He was a very humble and modest man, always trying his best. James, Shaibu and Victor became an excellent team, and during the three years I never needed to intervene in their work; they organised and agreed among themselves the work shifts, and I always had a guard on duty. Grace was the matron in the house and at the top of the hierarchy. She made sure that things went smoothly, and Shemu, James and Victor obeyed her when we were not around. These four people were my closest link to the local reality of Malawians struggling with the everyday hardships of an urban city. Sometimes I feel that my financial and social support for them and their families, and my husband’s medical services 16 were the most tangible accomplishments during the three years, and had a direct impact on the wellbeing of several Malawian people.

I first arrived in Malawi in August 1998 to work in the UNICEF Malawi office as a junior project officer. I settled down in Lilongwe, the capital city, where all the multiand bilateral international organizations, donors and government offices are located (see Figure 7. Map of Malawi). I occupied a UN staff house together with my husband. The house was big; three bedrooms and a living room, and a huge beautiful garden surrounded with high fences (to keep out uninvited visitors as I was told). For an anthropologist who was used to basic living conditions in the field, this was a huge upgrade. Becoming a UN staff member meant that suddenly I owed a car and was an employer for four local people that took care of my house while I was busy working in the office. I was not really given any choice to live differently from the rest of the UN staff; the security regulations were strict, and once the word went around that a new expatriate had arrived in the country, there was a queue of people outside my house offering their services to work as a guard, housekeeper, cook, gardener, driver and so on. I ended up employing four people permanently. I employed Grace as my housekeeper. She was in her 50s and very experienced as she had worked for several expatriates during the past 10 years. She moved to live in the small staff house in the compound. She was a modest woman, very trustworthy and hardworking. She taught me how ‘everyday household business’ is run and made my life very easy. Then I employed Shemu, James and Victor to work as guards. Shemu and James were young men in their mid-twenties. Originally they worked as guards for a private security company with which I first had a contract through UN. In a few months I became irritated with the way the security company treated its employees and against the UN regulations and warning of my col-

16

At the time my husband was still a medical student, but for Malawians it did not matter. For them, my husband was a medical doctor and therefore trusted and respected, and people came to ask for his advice and prescriptions for all kinds of health and social matters.

27

Lilongwe

Figure 7. Map of Malawi adequate understanding of the underlying socio-cultural factors, often labeled as misconceptions and beliefs forming barriers to development and improvement. This field period laid an important basis and motivation for the forthcoming research and fieldwork. It led me towards applied medical anthropology at the intersection of international health and development aid. In Chapter 8 I will draw on this experience when discussing the global and national level malaria prevention context, knowing very well that this is a subjective account based on participant observation during daily work in an international health organization. As Foster (2009: 683) points out, information gathered in this way is ‘interpreted’ rather than ‘analyzed’.

A large part of my work in UNICEF concentrated on health issues and particularly malaria prevention programmes. I was also responsible for activities regarding community-based primary education projects during the first two years. My main task was to plan, manage and supervise the community-based ITN projects in collaboration with the NMCP and district malaria officers. I also conducted regular monitoring visits to the rural sites in the five UNICEF supported districts. My visits in some of the most remote places were often received with surprised but delighted comments on me being the first white person visiting the project and once even the first white person ever seen by an old woman in rural Nkhata Bay. Despite the visible poverty, the villagers welcomed me warmheartedly offering me local meals of cooked nsima (maize porridge) with relish and even sometimes served together with chicken despite my appeal to save it for the family. During these visits, I learned about the everyday hardship of rural Malawians and at the same time, I became painfully aware of how little impact the wellintentioned development projects had on local people’s lives. The monitoring visits revealed that there was a real gap between policies, programmes and local reality. Nor was there

In the beginning my plan was to conduct research in two different sites to be able to carry out data comparisons. I was thinking that Kasungu in the central region and Nkhata Bay in the Northern region would be suitable places; different enough as the first was inland occupied by Chewa farmers and the second a lake town full of fishermen mainly belonging to the Tonga and Tumbuka tribes. These two districts were also familiar to me through my many monitoring visits to the UNICEF supported ‘ITN communities.’ This 28

(TUMCHP) and invested in building research facilities next to the health centre to support their maternal and child health clinical trails. Within the TUMCHP, there have been Finnish and Malawian students aiming for a Ph.D. degree coming and going for more than ten years. The local people have become very used to the msungu (white man) during the past twenty years18.

plan turned out to be unrealistic simply because of lack of funding. At the time I started my fieldwork I did not yet have any funding and decided to carry out the research in a place which would be the easiest logistically and the cheapest in terms of accommodation and daily living costs. The only place I could think of was Lungwena, situated on the Eastern shore of Lake Malawi, approximately 300 kilometers from the capital city and about 30 kilometres from the nearest town (see Figure 8. Map of Mangochi District).

I came to Lungwena for the first time as an anthropologist at the beginning of September 2002. It was an easy field to enter because of the long engagement of Finns in different capacities; as NGO workers and as researchers. I lived in the health centre compound in one of the staff houses occupied by a Finnish NGO worker, Johanna. She was both an anthropologist and trained midwife, and discussions with her at the beginning of the fieldwork (and throughout the process) were an excellent way of getting to understand the life and people in Lungwena. The living standards in Lungwena were poor compared with those in Lilongwe, but as a place to conduct research, the standards were great. The staff house had two small rooms, a small kitchen with a gas stove and a cold water shower. Thanks to the Finnish NGO project and TUMCHP, there was a generator that provided electricity for four hours (6-10 p.m.) in the evening and solar panels that provided electricity during the daytime. I was able to use the computer, to write up my field notes every evening, and keep in touch with family, friends and supervisors with a wireless mobile phone internet connection.

I had visited Lungwena a couple of times while working in Lilongwe, although the visits were not work related and I never went there as a UNICEF project officer. I visited Lungwena because of a Finnish couple (Leenu and Kytö Salin), who were implementing a NGO project called Lungwena Training Health Centre Project (LTHCP).17 I knew them already from Uganda where they had implemented a similar project by the same NGO. When the LTHCP project started in the early 1990s, the Lungwena health centre buildings existed but there was no health staff. The LTHCP was established to improve the health of the communities, to use the Lungwena training health centre as an experimental training site for provision of comprehensive primary health care services, to develop a model for training cadres of health providers among rural areas, carrying out research on health problems and searching for their solutions. The project ended in 2002 and was handed over to the Malawians (LTHCP internal evaluation report 2000). At the end of the 1990s another Finnish NGO implemented a reproductive health project in the area with one Finnish project worker who spent six months per year in Malawi for a period of three years. In the mid 90s, the Medical Faculty of University of Tampere started a research project, the Tampere University Mother and Child Health Project

18

The long period of Finnish involvement is coming to an end as most of the research activities and facilities of the Medical School of University of Tampere have been moved next to the district hospital in Mangochi town to accommodate the needs to carry out a large international research project. The local research staff are now commuting daily between Lungwena and Mangochi.

17

LTHCP was funded by a Finnish NGO, Mannerheim Child Welfare League, and the Finnish Ministry for Foreign Affairs. From the early 1990s to year 2000 there were Finnish expatriates managing the project in collaboration with the Malawian project team.

29

Research area Lungwena catchment area

Figure 8. Map of Mangochi District

30

luckily, Johanna’s house was not occupied by anyone, and we were able to move there. The house had deteriorated considerably in three years, and the solar panels had been stolen from the roof. There was dust, dirt and mouse faeces everywhere, but for a couple of weeks it did not really matter. My trustworthy research assistant, Shaibu, had organized everything prior to my arrival. He had informed the chiefs in the villages about the continuation of my research and done preliminary arrangements for FGDs and interviews. He had contacted two more research assistants, Innocent and Stephano, who had previously worked with Johanna, to assist me during the busy two weeks.

Hence, although living in a remote area, I was very much connected to the rest of the world unlike most of the people living in Lungwena. I bought my food from the small trading centre nearby which sold basic food items such as sugar, maize flower, tomatoes, onions, bread and fish, and when visiting Mangochi I bough other ‘luxury’ items such as coffee, pasta and tinned food. My daily routine was to wake up at sunrise, no later than 6 a.m., to be at the health centre by 8 a.m., around the time the last pregnant women were arriving or to cycle to the villages to conduct interviews, and to return before dark, work on my data and e-mails and go to bed at 10 p.m. when the generator was turned off. Occasionally I had an opportunity to spend a night or two in Mangochi, staying in the house of another Finnish researcher, Heli, and her two sons. She conducted nutrition research in Lungwena. These short periods with Heli and her family served as important mental relaxation that helped me live and carry out fieldwork in sandy and hot Lungwena. I returned to Finland at the end of December.

4.2. Positioning and establishing relationships in the field In anthropological research one cannot leave out discussion about the position one takes in the field (intentionally or unintentionally), how local people perceive one and how this may affect the quality of the data one collects (Bernard 2002; Fetterman 1989). I arrived in Lungwena alone without my husband. I was a white childless female, in many ways the exact opposition of the women I interviewed. In the beginning I was concerned with how the local community would perceive and allow me to enter the field; whether my gender and unmarried childless status would affect building rapport with the women. In contrast to Susan Dwyer-Shick’s (1992: 258259) experiences in Turkey, where she had to bend the truth about her daughter’s care taking back home in order to gain trust and build rapport with the Turkish women, I soon learned that most of the women in Lungwena did not seem much bothered about my gender status. I suspect that this was due to the fact that there had been so many white childless females conducting research in Lungwena that there was nothing odd about me.

I returned to Lungwena in April 2006 for a very short period of two weeks. I arrived together with a Finnish medical doctor and her boyfriend. She was starting her Ph.D. research project, and her supervisor had asked me to accompany her to the field in order to supervise and assist her to start up the qualitative part of her fieldwork that consisted of focus group discussions (FGDs). Although being pregnant (28 weeks), I did not hesitate to take up the offer as it allowed me to conduct focused data collection on some of the questions left unanswered from my first fieldwork. I also felt that I needed to connect with the field again in order to keep my motivation up and to remind myself that this research is important19. Returning to Lungwena was not as smooth as before. Due to a misunderstanding, there were no rooms available in the Lungwena health centre hostel, but

At the beginning of each interview, I introduced myself as a student from Finland, and having worked previously in Lilongwe aiming to learn from the women and men in

19

Conducting research alone with no research team with whom to share and reflect upon ideas, as well as experiences far away from the field, makes it very hard to keep one’s motivation up year after year.

31

rare during the first fieldwork. (See more on building rapport in Chapter 4.4.2. Important role of research assistants).

Lungwena issues related to pregnancy and its outcome. I tried, however, to emphasize that I was not working for the Lungwena health centre nor was I part on the on-going medical trials. I thought that being associated with the medical trials and Lungwena health centre might make people suspicious and reduce their willingness to openly discuss issues related to ANC care. My understanding is that despite my efforts, the local people were not able to make much difference between the projects and white people coming and going. It is difficult to say whether this worked against me or had no impact. I also tried to create a trustful atmosphere by emphasizing that I handled all information confidentially. I conducted all interviews anonymously without recording the names of the women and men. I felt that building rapport with the women was in most cases rather easy, particularly with women who had several children. This was probably because the research themes focused on their pregnancies and motherhood, themes they were happy to talk about and obviously, were experts in their own field. Only the young and first time pregnant girls had to be convinced to talk because they felt they knew nothing and that they could not have any knowledge that would interest me.

4.3. Local reality 4.3.1. Socio-cultural context Marriage and gender roles among the Yao The population living in Lungwena and the surrounding areas are mainly Yao by ethnic origin. The majority of them are Muslims practicing Islam adapted to the local communities. In Malawi there are several ethnic groups that fall into either the matrilineal or partilineal system. The central and southern regions are predominantly matrilineal, whereas the northern region is patrilineal. The Yao are matrilineal and thus descended through the female lineage, meaning that children belong to the mother’s matrilineage and acquire their rights to land, as well as assistance and support to maintain their wellbeing through the mother and matrilineal kin. Mitchell (1962:30) points out that obtaining assistance from the father’s side is therefore perceived as an act of grace as all the obligations lie on the side of the mother’s relatives. Thus the nature of a Yao marriage is economic and political. Both monogamous and polygamous marriages are found in the area. (Paz Soldan et al. 2007; Thorold 1993.)

When I arrived in Lungwena for the second time, I noticed a difference in local people’s attitudes and interest towards me. Now I was pregnant myself and my belly was big and visible. It was very easy to start discussions about pregnancy related issues and vulnerabilities for two reasons. Firstly, now I was able to connect with the women much better having a personal experience of pregnancy, delivery and motherhood. This time the questions I asked were much more meaningful also for myself than during the first fieldwork. I was also much better able to feel and understand the everyday hardships of being a pregnant woman in Lungwena. Secondly, my visible pregnancy allowed building rapport even with the inexperienced young girls. They also asked questions and advice concerning their pregnancy, which was

According to Mitchell (1962:29), marriage among the Yao is a formal mechanism through which social relationships that include specific rights and duties are defined between a husband and a wife, but also between descent groups of the husband and wife as well as between the children within the marriage and all these people. The Yao marriage that goes beyond the married couple extending the relationships to the two descent groups has implications for the patterns of expected behaviour towards each other. Marriage signifies a set of rights and duties between the spouses that are mainly related to children and sexual obligations. In practice this means that a husband has exclusive 32

system are associated with improved level of education, temporary migration to work in the mines in South Africa, due to the extreme poverty situation and the AIDS crisis in Malawi (Chimbiri 2007).

sexual access to his wife and at the same time it is his responsibility to give his wife children. A husband is also supposed to support his wife financially, ensure shelter, food and basic necessities, as well as to protect her from harm. Whereas a wife’s responsibility, in addition to bearing children, is to cook food, draw water, collect fire wood, pound the maize, work in the farm, care for the children, care for the husband when ill etc. (Mitchell 1962: 31; Mvula and Kakhongwa 1997: 27; see also Mitchell 1956.)

4.3.2. Everyday life and hardships in Lungwena I would not describe Lungwena and the surrounding villages as a typical place in Malawi. It has been rather isolated due to its geographical location between Lake Malawi and the bordering hills of Mozambique and due to poor infrastructure until 2001-2003 when a tar road was constructed between Lungwena and Mangochi, significantly improving the access to the nearest villages and Mangochi town. Before the tar road was built, access to the district hospital was virtually cut off for months during the rainy season because then the road would transform into a muddy path, sometimes impassable even to 4by4 vehicles. The number of local minibuses and private cars has increased considerably since the building of the tar road. Before, a day could pass without seeing a local minibus on the road.

Social structure and power relationships Traditionally in a matrilineal system, the head of the household is always a man: a husband, maternal uncle, brother or grandfather (Eriksen 2001: 102-103). Also in Yao society men are viewed as leaders and to have authority, but not directly over their wife and children but over their sisters and their children. There is a special relationship between the sister and most often with the oldest brother in the family, called as ‘owner of the women’ (asyene mbumba), according to Mitchell (1962: 33), or the maternal uncle. This means that the maternal uncle owns and controls the inherited property and family resources such as land. He is also responsible for the welfare of his sisters, extended to all their children that belong to his matrilineal descent group.

Most households are living under the ultrapoverty line with less than 25 U.S. cents per day and experiencing periodical hunger. Only a minority of families are able to buy basic food items such as bread, sugar and cooking oil. Fishing and farming, mainly maize, are common occupations, but basically all fish and maize is consumed by the households themselves. (Benson 2002.) Due to the poverty situation and lack of cash earning opportunities in the area, only a few are able to pay for local transportation. Thus access to services is hard as most households have very little cash available, if at all. A Malawian woman spends on average 12 hours each day working regardless of the season, which is twice as much as the work time spent by men. Women have an important role in producing, storing, processing and preparing food for the family needs, whereas men are more often

According to tradition, the husband is supposed to move in with his wife’s relatives who complicate his possibility to carry out his duties concerning her sister and their children (Mitchell 1962). According to Kishindo (1995: 27), this also means that the localized matrilineal descent groups lose their male members to other descent groups in near or distant villages, and the departing male members are replaced by men from other descent groups who father their children through marriage. According to Chimbiri (2007), there are signs that the matrilocal residence system is slowly changing, and nowadays a woman can also move in with the husband’s relatives. These changes in the traditional residence 33

other providers such as private hospitals and clinics, commercial companies, the Army and the Police, accounting for the remaining 2%. In addition, there is a large folk sector consisting particularly of traditional healers, namely herbalists that belong to the Herbalist Association of Malawi, having approximately 75,000 members. Private shops also sell basic antimalarial drugs. (GoM 2005.)

engaged in growing commercial crops. This is attributed to the fact that women have less access to credit, land and inputs than men. (Mvula and Kakhongwa 1997: 27.) The burden of malaria in pregnancy is high in Lungwena and its surroundings. A community-based cohort study found that at ANC enrolment 42% of pregnant women had maternal anaemia and 27% had peripheral malaria parasitaemia. The study also revealed that 22% of the deliveries were preterm (before 38 week of gestation) associated with primiparity and peripheral malaria paracitemia of the mother. (Kulmala 2000: 49.)

Medical pluralism is characteristic of Lungwena and its surroundings. The government-supported Lungwena health centre and three outreach clinics provide normal preventive and curative health services that include treatment for the common illnesses, antenatal and delivery care, under-five clinics and vaccinations. The services are free of charge. The estimated population in the Lungwena catchments area is approximately 27 400, living in 27 villages (Lungwena H/C data Jan 2010). The health centre staff include two medical assistants (compared with only one in 2002 and 2006), four midwifes (compared with three in 2002 and 2006), 18 health surveillance assistants, one environmental officer, two ward attendants, one ground worker and two security guards. Antenatal attendance is high with about 95% of the pregnant women attending the ANC at least once during their pregnancy. The health centre dispensary is mostly well stocked with basic drugs, and uncomplicated pregnancies and deliveries are handled at the clinic. In 2009, 659 deliveries were registered in the clinic. However, severely ill children and women with pregnancy and delivery complications are referred to either St. Martin’s Mission Hospital in Malindi that has 100 beds (15 km from Lungwena on the Mangochi to Makanjira road) or Mangochi District hospital (about 35 km away). In addition, there are 13 traditional healers (asing’anga) that are herbalists, six spiritual healers (chisako), five religious healers (shehe), a few shopkeepers and traditional

4.3.3. Medical system and pluralism The Yao classify and interpret illnesses in the framework of natural and personalistic causalities, as do many other ethnic groups in Malawi and in sub-Saharan Africa (Foster 1976; Green 1999; Young 1976). According to my data, the majority of illnesses in Lungwena and nearby areas are interpreted as resulting from natural agents, and treatment is primarily sought from official health care providers and during pregnancy from the ANC. Some of the natural causes are also interpreted as acts of God (mulungu). There are also indigenous illnesses that are perceived to be caused by supernatural agents such as witches (afiti) or evil spirits (majini) that often require negotiation within the therapy management group regarding treatment-seeking (Janzen 1978), and treatment provided by traditional healers (asing’anga). Morris (1985: 17) has argued that “there are no witchcraft diseases in Malawi, only certain kinds of misfortunes which may in specific contexts be attributed to witches (afiti).” In Malawi, the Ministry of Health is the main service provider in the professional sector, accounting for 60% of the health services. The remaining 40% of the health services are provided by the Christian Health Association of Malawi that accounts for 37% of the services, and by local government (1%) and 34

footsteps of Malinowski, strolling leisurely through the study village observing the family practices as a daily routine and writing ethnography, full of rich descriptive data. Unfortunately, in my case this was not possible for two main reasons. Firstly, I had no funding for the fieldwork except my own savings after three years of working for UN and secondly, for personal family reasons I was not able to stay again for one to two years in Malawi. Being realistic and knowing the time limitations for conducting fieldwork I decided to plan an applied medical anthropology study applying an ethnographic approach. Applied medical anthropology here refers to research that takes place in the intersection of public health and medical anthropology focusing on a public health problem in the theoretical frameworks of sickness and healing (Janzen 1976, Kleinman 1980). The ethnographic approach refers to fieldwork that is conducted in a limited time period, utilizing the same methods as are used in ethnographic research (participant observation, in-depth and key informant interviews, informal discussions etc.), but spending less time on ‘strolling around and participating in the daily routines of the local people’, although this is an important part of anthropological research (Fetterman 1989). Conducting applied research also allowed utilization of multiple methods not traditionally used in ethnographic research. Among other things, I utilized both FGDs and individual in-depth interviews as well as a quantitative knowledge, attitude and practice (KAP) survey. I used data sources collected using these different methods to cross-validate the collected information and to check the reliability of responses (Helitzer-Allen and Kendall, 1992; Agyepong et al., 1995; Bhattacharyya, 1997).

birth attendants (asamba) who provide services to the local communities. 20 4.4. Methodology 4.4.1. Ethnographic approach and multimethod study design (IV) There is no one definition of ethnography but it is generally agreed that it refers to “a research process in which the anthropologist closely observes, records, and engages in the daily life of an other culture - an experience labeled as the fieldwork method - and then writes accounts of this culture, emphasizing descriptive detail” (Marcus and Fisher 1986: 18). Hence, ethnography refers to both a process (methodology) and a product, a written account in the form of a monograph containing thick, descriptive and rich data in a particular socio-cultural setting (Savage 2007). The core of ethnography is a long fieldwork that is carried out at least for a year, but often longer. An anthropologist conducting ethnographic research is expected to learn the local language in the process. Characteristic of the fieldwork is that the researcher is in no harry, s/he will spend several months to establish a position in the study community, to build rapport and to identify key individuals who have an important role in guiding the anthropologist during the fieldwork. S/he will participate in community life as much as possible, trying to become an insider with whom all the local traditions and wisdom is shared with. Bronislaw Malinowski is considered the founder of this ethnographic method, which he describes in his book, Argonauts of the Western Pacific, originally published in 1922. Being trained as a cultural anthropologist, conducting fieldwork was a natural thing, and something I was anxious to start. As probably many anthropologists, I would have wanted to conduct ethnographic research following the

4.4.2. Important role of local research assistants (IV) Although I lived a total of three and half years in Malawi I never learned to speak fluent Chichewa, spoken in Lilongwe and through-

20 Information regarding the number of traditional

and religious healers was collected by one of my research assistants in January 2010.

35

interpreter. Eunice felt it was strange that I wanted to go to the villages and do the interviews myself when she thought she was perfectly capable of interviewing around any possible topic I wanted. According to her, previous Finnish researchers had not conducted the interviews by themselves. Instead, they had sent her to do the interviews with instructions, topic guides and questionnaires. This explained why she had a difficult time remaining in her role as an interpreter translating my questions.

out the central region, nor Chiyao that is spoken in the Southern region and in Lungwena. While I was working in Lilongwe I took Chichewa language lessons every now and then, but it was hard to stay motivated by the language training when English was the official working language and everybody living in the capital and towns were able to speak English, even those with little education. I used my simple and basic Chichewa mainly during my field visits to show respect to local people who were both happily surprised and amused by my efforts. My ability to understand Chichewa conversation was much better than speaking the language. Choosing Lungwena as my research site meant that I was completely dependent on finding research assistants who would act as my interpreter because the local language spoken in the area was Chiyao that has nothing in common with Chichewa. Choosing and training research assistants became a very important task because the quality of the data collected depended now also on my ability to train the research assistants to understand the broad study aims and methodology used for collecting data as well as to act as an interpreter and to build rapport with the respondents and communities.

My next interpreter was Rashid, a young man in his twenties from Chapola village. He was polite, learned fast and did not question my way of doing research. His English was good, and he was able to read and write as he had completed secondary school. At first I was worried about how the interviews would go with a male interpreter and whether the women would feel free to talk about pregnancy related issues. To my surprise, a male interpreter was not that big an issue for women particularly when they were experienced and older than Rashid. Interviews with young girls were often difficult, for several reasons. For example, girls often said that they know nothing and I should ask the more experienced women in the villages. Building rapport was not always possible with young girls who were shy and probably scared of me. They were not convinced that I would not tell the nurses about their answers regarding the antenatal clinic. Rashid turned out to be a good and trustworthy research assistant with whom I found it easy to work with. We had occasional problems regarding translations of specific words regarding pregnancy and delivery, as he was not always sure of their public health meaning.

Upon my arrival to Lungwena and after introducing my plans to carry out fieldwork for several months in the surrounding communities, I was immediately offered assistance by one of the research assistants affiliated with the clinical trials of the Medical School of University of Tampere. Eunice was a very talkative and confident woman from Taliya village. Unlike most Malawians, she offered her opinions and advice on whom I should choose as my research assistants and how I should organize my fieldwork. In the beginning I liked her active role very much because it was quite different from the more common, rather passive way of doing things and waiting to have clear orders. I soon learned that Eunice liked doing things on her own. Her proactive style started to work against her as a research assistant and after a while I decided to look for another

For the KAP survey, I had to employ more local research assistants, and Eunice was eager to find me suitable candidates. She introduced me to Gertrud, Ben, Zacharia and Shaibu who I ended up employing. Having four different research assistants showed that there can be clear differences between the skills and motivation of the research 36

researchers from the North would have a difficult time gathering any data, be it qualitative FGDs or survey data because of language, cultural factors and difficult circumstances. In most cases, local research assistants remain uncredited for the work they do and are invisible. Often the research process and data collection is not described transparently although the use of local research assistants, who have no formal training on any kind of scientific research, have a positive or negative impact on the quality of the data. If one would conduct research in the North and use research assistants who have in most cases only primary education and occasionally high school education to collect the data for a Ph.D. research, the academic community would certainly react and question the quality of the data and reliability of the results. Local research assistants in the South are trusted with enormous responsibilities, which often go unnoticed by the research community in the North, unfamiliar with practices regarding research in rural settings such as Lungwena.

assistants; some are able to build rapport with the survey respondent and use probes for the open-ended questions while others can hardly read the questions and correctly fill in the answers. Training the research assistants and pre-testing the questionnaires thus became very important and even then, I encountered multiple challenges while conducting the survey data collection (described in article IV). When returning to the field I employed Innocent and Stephano, two young men in their mid-twenties to assist with the research together with Shaibu. Since Innocent and Stephano had previously been trained by and worked with Johanna, they were knowledgeable of how anthropologists conduct their research and what is expected from research assistants. Innocent and Stephano were hardworking and very reliable assistants who were eager to learn more skills and carried out their work carefully. They took on a tremendous workload in a short time period. Out of all my research assistants, I became closest to Shaibu and his family. Shaibu was also in his twenties living in the nearby Chapola village, a husband and father of one daughter. He was very eager to learn new things and develop his skills as a research assistant. He first translated all my interviews from Chiyao into English, and conducted KAP survey interviews. He was very good in building rapport with the pregnant women and did his work carefully. When I returned to Finland Shaibu remained my contact point to Lungwena, and whenever I needed clarifycations or translations I called Shaibu or sent him an e-mail. Through him I was able to keep a tiny, but important connection to the field and occasionally heard about the everyday hardships in Lungwena; the hunger periods, delayed rains, increases in maize prices, increasing poverty and family sorrows.

4.2.3. Data collection procedures At the beginning of fieldwork, I first spent time to get to know and train my research assistants. I trained my research assistants in the study objectives, data collection methodology and their role as my assistants. I also spent time to get to know the previous experience of my research assistants and their learned habits that caused some problems during the fieldwork (IV). In addition, we discussed issues such as how to build rapport during the FGD, how to moderate and take notes, how to probe and ensure that the discussion goes on and no one of the group members dominates or intimidates others. We also discussed and rehearsed being a translator; my role as the one asking questions and the translator’s role to translate as accurately as possible. Training and supervising the research assistants was an ongoing process, with some needing more guidance than others. We collected data both in the surrounding communities and in the Lungwena health centre compound. The

Local research assistants have a significant role in data collection, yet they role is often underreported and left unrecognized in research publications as well as in Ph.D. dissertations. It is, however, a fact that most 37

Individual interviews with women of reproductive age (n=34), pregnant women (n=8) and men (n=22). I conducted all interviews with women of reproductive age in the villages where the women lived (in 2002). Each interview lasted about one hour to one and a half hours. I asked the questions and my research assistant translated them. I also kept notes while I interviewed. One woman, who was also pregnant, refused to participate in the study. I interviewed the pregnant women at one of the rooms in the health centre because of my own pregnancy that did not allow walking or cycling long distances (in 2006). Each interview lasted 30 minutes to an hour. My research assistant acted as a translator. No one refused to participate. I interviewed the men at the health centre compound following the same procedure as above (in 2006). No one refused to participate. At the beginning of all in-depth interviews my research assistant briefly introduced the purpose of the interview according to my instructions. After the introduction I asked them to give informed consent which was tape recorded and then conducted the interview according to the themes I had chosen. The interviews never followed the same pattern because my questions and probes depended on the answers. During the interviews I tried to probe for longer narratives on different themes such as delivery and illnesses experiences. At the end of the interview I always asked if the participant had anything s/he wanted to ask me. Some asked for medical advice, others were curious if we have malaria in Finland, and yet others asked advice for marital problems and husbands’ unwanted behaviour. I answered the questions that did not concern medical advice, and concerning medical advice I suggested the person seek advice from a medical person. During my second visit, I noticed that participants were curious about my pregnancy and asked about pregnancy related issues in Finland. They also wanted to know how far my pregnancy was and

different methods and target groups are described in more detail below (see also Table 1. Summary of field visits, data collection methods, target groups and main themes). Focus groups with women of reproductive age (n=9), men (n= 4), female elders (n=2) and pregnant women (n=1). I utilized FGDs at the beginning of the data collection because FGDs are a good tool to generate data on people’s experiences, opinions and concerns regarding certain themes such as reproductive health issues and malaria in pregnancy as in this present study. Group interaction allows discussions and commenting on each other’s experiences making it possible to have an idea of the main concerns regarding the research issues quite quickly. (Kitzinger and Barbour 2001.) All FGDs, except the one with pregnant women, were conducted in the villages (see sampling and selecting of villages and participants later). The FGD with pregnant women was conducted in the Lungwena health centre compound because after an ANC visit it was possible to get pregnant women together for a FGD. Each FGD lasted about 1-1.5 hours. One research assistant moderated the discussion and another one took notes. I was present in onethird of the discussions to observe the group dynamics and atmosphere, and to provide guidance to my assistants. I preferred to carry out the FGDs at the beginning of the research because it allowed me to gain insights into local perceptions regarding the main themes I had selected for the study. The main themes covered: 1) how the communities define malaria as an illness, and local taxonomy and etiology, 2) perceptions regarding malaria in pregnancy, 3) treatment and prevention of malaria in pregnancy, 4) channels of communication in the community, 5) therapy management and treatment-seeking practices in pregnancy, 6) pregnancy and delivery complications and 7) the roles and responsibilities of family members. There were slight variations regarding the themes discussed depending on the target group. 38

pregnancy. All interviews were carried out in 2002 following similar procedures as described above regarding in-depth interviews. All interviews took place in the village and lasted approximately one to one and a half hours.

how I was feeling. These in-depth interviews form the core of the data. Illness and complications classification exercise. During the FGDs and IDIs I asked the participants to classify what illnesses they consider dangerous for pregnant women in the area and to describe symptoms, kinds of causes, treatment options, prevention methods, possible consequences for pregnancy and commonness. Lastly, I asked the participant to say which illness they considered the most dangerous and why. I asked about pregnancy complications in a similar way. If malaria did not emerge as one of the illnesses, I probed for it. The purpose of this exercise was to find out if the Yao have shared criteria for defining when an illness or a pregnancy complication is perceived as dangerous for a pregnant woman, and what factors might weigh into determining the dangerousness of an illness for a pregnant woman.

Structured cross-sectional survey of MiPrelated knowledge, attitudes and practices. I developed a structured cross-sectional survey questionnaire based on the WHO survey protocol for assessment of social, economic and cultural aspects of malaria (Agyepong et al. 1994). I modified the questionnaire to better fit the local context based on the preliminary findings from the FGDs and in-depth interviews. I used the survey to gain confirmation on the local malaria classification system and etiology. I covered many of the same themes as in FGDs and IDIs for triangulation purposes (see Annex 3, KAP survey questionnaire). The research assistants completed a total of 248 survey questionnaires with women of reproductive age. The questionnaire was translated twice, first from English into Chiyao and then back from Chiyao into English and finally back into Chiyao by different translators. Use and applicability of a KAP survey are discussed in detail in article IV.

Drug sorting exercises. I carried out an exercise of sorting out the main drugs available from local shops and out-patient clinic and ANC. I asked each woman to identify the drugs and tell me if they had ever used them and for what purpose. The aim was to gather information on the knowledge and perceptions related to using drugs in pregnancy. The drugs shown were SP, iron tablets and Panadol® from the ANC, Novidar® (SP) and tetracycline capsules from the shops at Lungwena trading centre and Vitamin A capsules from the Lungwena health centre out-patient clinic.

Notes from the interviews. In addition to tape recording the interviews, I kept notes during the interviews because it allowed me to keep track of the major issues emerging from them, as well as to modify my themes and to follow interesting leads. It turned out that tape-recording of two interviews failed, but with the help of my notes I was able to capture the main points from those interviews.

Key informant interviews with traditional birth attendants (TBA, n=4), traditional advisors (anankungwi, n=2). I interviewed TBAs about their role in managing pregnancy related problems as well as home deliveries. I also explored themes related to malaria in pregnancy. I interviewed traditional advisors to find out more about their role in the initiation ceremony for first time pregnant girls (litiwo), about their perceptions and knowledge regarding pregnancy related problems and malaria in

Observations at the ANC. I and my research assistants observed ANC sessions to document the antenatal clinic procedures, the content and style of individual counseling and communication related to drug use and malaria prevention. Observations were carried out during ten ANC days on randomly chosen dates. All health education sessions were tape39

recorded and translated into English by my research assistants. The purpose of the observations was to gain insights into the quality of ANC services, distribution of knowledge about pregnancy and malaria related issues by the nurses and use of antimalarial drugs, such as SP. I also wanted to better understand women’s perceptions of the services described during the interviews and to verify women’s explanations about what happens at the ANC. Data from the ANC register. I collected data from the ANC register from February to mid-December concerning the first visits to ANC. This data show the number of women attending the ANC each month, from which villages, trimester of pregnancy and if the women received SP.

40

Table 1. Summary of field visits, data collection methods, target groups and main themes Field visit periods

Methods

August 1998October 200121

Participant observation as a malaria project officer Key informant interviews Informal discussions KAP surveys UNICEF Ethnographic malaria study FGDs, including drug sorting exercise

SeptemberDecember 2002 Lungwena

IDIs, including drug sorting exercise Key informant interviews

Target groups

FGD (n) not recorded

Village Health Committees International Aid Community National Malaria Control Program Roll Back Malaria Taskforce District malaria officers Women of reproductive age Men

Interviews (n) not recorded

7 1

Factors influencing community participation ITN implementation Central-level barriers affecting programme planning and implementation

Knowledge of pregnancy Perceptions of ANC services; Perceptions of drugs and restrictions in pregnancy (food, drugs, practices) Knowledge and perceptions of different illnesses and treatment-seeking patterns Decision makers in household; Delivery practices and complications Knowledge of malaria treatment and prevention

Women of reproductive age

34

TBAs Traditional Advisors

4 2

21

Themes explored

Same themes as above Knowledge, perceptions and advice regarding pregnancy and delivery Initiation ceremony litiwo and pregnancy related cultural taboos and advice

I lived and worked in Lilongwe. As part of my work I carried out multiple supervision and monitoring visits to ITN project districts: Mzimba, Nkhata Bay, Kasungu, Mangochi and Mwanza. Since this was a field period that laid the ground for the actual fieldwork I did not keep record of all the people I interviewed during my supervision and monitoring visits, or of the number of people who participated in the various studies.

41

Field visit periods

April 2006

Methods

Target groups

Observations at the ANC

Pregnant women and the nurse in duty (about 180 PWs observed)

ANC registry data KAP survey

Jan-Oct 2002 (10 months) Women of reproductive age (interviewed in communities) Pregnant women (at ANC) Men Pregnant women Women of reproductive age Female elders

FGDs, including

IDIs Observations at the ANC

FGD (n)

Interviews (n)

Themes explored Observations were unstructured and notes were kept from the arrival of first woman to the departure of last woman Observations concentrated on health education sessions that were also tape-recorded (content and communication dynamics) Nurses were observed carrying out the antenatal examination and the IPT-SP policy

200

Themes as above

48 3 1 2 2

Men Pregnant women Pregnant women and the nurse in duty (43 PWs observed)

22 8

Communication concerning pregnancy, Knowledge and perceptions of pregnancy risks, worries regarding pregnancy, Perceptions of dangerous illnesses, Role of husband and other relatives in management of pregnancy and complications, Motivation for and use of ANC services, and management of first pregnancy Complications and delivery practices, Knowledge and perceptions of drugs, knowledge of western and indigenous, traditional illnesses (especially malaria and fever) Treatment-seeking practices and prevention of malaria. As above Same as 2002 (see above)

In total 64 in-depth interview participants, 125 FGD participants (all groups 8 participant, except PWs 5), 223 PWs observed at the ANC and 248 KAP survey participants

42

own pregnancy and the limited time available, I decided that the best option is to ask the respondents come to me instead of me going to the villages.

Sampling and selecting respondents for FGDs and IDIs I conducted sampling and selection of respondents in collaboration with the community management committee consisting of chiefs representing the Lungwena H/C catchement area. The catchment area is divided into four parts (known locally as gulu) and each gulu is managed by one of the chiefs. The chiefs suggested that I should select at least one village from each gulu; eventually two villages from each gulu were chosen. I used convenient sampling and chose the villages nearest to the Lungwena H/C (max. distance 5 km from the H/C) because I used a bicycle to get to the villages. A very long distance would have made the data collection difficult because of the extreme weather conditions (+40 C in the shade). I would have spent a lot of time just travelling between Lungwena and the villages, and because of the heat I anticipated that it would have been very exhausting for me. I chose the and IDIs: Gulu 1: Gulu 2: Gulu 3: Gulu 4:

Sampling and selecting respondents for FGDs and IDIs I used a different sampling method to select the villages for the KAP survey. I divided the Lungwena H/C catchment area into 3 geographical areas as follows (see Figure 9. Map of Lungwena health centre catchment area): 1) Mountain side: Kwilasya, Masasyasya, Talia, Rashidi, Magongwe, M’dala Makumba, Liwale, Mbanda (8 villages) 2) Lakeside south of Lungwena towards Malindi/Mangochi: Kapinjiri, Chapola, Biti Kalanje, Mlani Chapola, Mtaka, Milombwa, Matenganya (7 villages) 3) Lakeside north of Lungwena towards Makajira: Mpundi, Ngombe, Chilonga, Tumbwe, Fowo, Mtumbula, Mwanjati, Mbale, Mdoka (9 villages)

following villages for the FGDs Moto and Biti Kalanje Kwilasya and Talia Mpundi and Kapinjiri Mtumbala and Fowo

I then drew a random sample of two villages from each geographical area that were M’dala Makumba, Rashidi, Matenganya, Chapola, Mwanjati and Fowo22. Lungwena H/C was chosen as the 7th venue for the KAP. After this the next step was to randomly select the first household in each village from the H/C register. Random selection was done as follows: depending of the number of households in the village, I asked a person to pick a number, e.g., between 1 and 300. For example, if the number was 50, I looked for

The community management committee was informed about my return to Lungwena (in 2006), and I asked permission to collect some additional data for my research. I was allowed to draw a random sample of seven villages from the Lungwena H/C catchment area. These were Kwilasya, Rashidi, M´dala Makumba, Chapola, Ngombe and Fowo. In addition, the FGD of primigravidae was conducted in the compound of Lungwena H/C because I considered the ANC the best option to reach enough primigravidae at one time in order to conduct a FGD. The men for the indepth interviews were conveniently sampled from the nearby Lungwena trading centre by my research assistants. Pregnant women for the IDIs were sampled conveniently among the women attending the ANC. Due to my

22

The probability proportionate to size (PPS) was calculated as follows: Cluster 1: M’dala Makumba 290 H/Hs (total 1669 H/Hs) 0.17 * 200 = 35 (34.7) and Rashidi 98 H/Hs 0.06 * 200 = 12 (11.7) Cluster 2: Chapola 292 H/Hs 0.17 * 200 = 35 (34.7) and Matenganya 138 H/Hs 0.08 * 200 = 16 (16.53) Cluster 3: Fowo 367 H/Hs 0.22 * 200 = 44 (43.9) and Mwanjati 484 households 0.29 * 200= 58 (57.9) households

43

the no. 50 household in the register and that house was the starting point for the survey. In all villages every 8th household was interviewed. At the ANC 48 questionnaires were collected during 4 days. My plan was to interview every third woman. This did not work out because the women kept changing their places. Some were examined while others were interviewed, making it difficult to select every third woman. I later realized that this could have been avoided with a number system. The main purpose of doing the interviews at the clinic was to see if there are any major differences in the answers because of the interview setting.

44

Figure 9. Map of Lungwena health centre catchment area 45

4.4.4. Data analysis follow-up of some of the pregnant women. Participant observation of the daily lives of pregnant women and their treatment-seeking would have provided more reliable information on the actual treatment-seeking practices in pregnancy. Secondly, the importance of structural and economic factors as barriers to care and preventive services emerged from the data analysis. A more detailed investigation of household assets, control of funds and negotiation on use of funds in the extended family would have been important because it would have allowed a more comprehensive understanding of the barriers to care. Thirdly, interviews with health workers at the Lungwena health centre would have been important for gaining a comprehensive understanding of problems regarding IPT policy implementation, and nurses’ perceptions and attitudes regarding pregnant women. Fourthly, this study took way too long to finish (eight years), during which time there have most probably been changes in the national level malaria prevention activities and practices which I am not able to capture, and some of the results and recommendations may not be applicable to the current situation.

Data analysis is an on-going process that already starts during the fieldwork. I kept notes during the in-depth interviews and used those notes to carry out a preliminary analysis of the main themes emerging from the data. When I returned to Finland, I typed all the handwritten translated FGDs and interviews into Word documents. Then I read all the interviews and manually marked the main themes and categories emerging from the discussions. I then compiled a matrix where I summarised the themes and main findings in order to have a broad overview of the main findings. I also looked for similarities, differences, variations and contradictions between different data sets. I used different kinds of taxonomy maps to analyze the local meanings of malaria and its types (Bernard 2002). At a later stage I also used Atlas.ti software for Windows version 5.0 (Scientific Software Development GmbH) for sorting out the qualitative data after code creation and to identify interviews for close manual analysis. I analysed the KAP survey data using SPSS analysis software. 4.4.5. Limitations of the study I observed some clear limitations concerning the study. First, there are some methodological and data analysis limitations. I included a KAP survey into the study plan because, on the one hand, I wanted to develop my skills in conducting quantitative research and, on the other hand, to clarify my doubts and concerns regarding the KAP survey methodology. The KAP survey data have been underutilised in the research because of my limited understanding of quantitative analysis. I have only analysed frequencies and some of that data were published in the first two articles (I, II). After gaining more confidence as a researcher I also became rather critical about the method itself (IV), and this criticism is not reflected in the first two articles (I, II). Another methodological limitation was that I did not conduct close 46

5 RESULTS responsibility for caring for those who fall ill from malaria, yet in most endemic countries these same women don’t have access to information, decision-making power and financial resources that would allow for effective disease prevention and treatment at community level” (Kvinnoforum: A guide to gender and malaria resources, p. 8).

5.1. Gender and malaria in pregnancy (II, III) The Beijing Platform for Action in 1995 signified a shift in gender approach from the narrow Women in Development (WID)23 to a more comprehensive Gender and Development (GAD) approach. This meant that international aid communities as well as governments around the world agreed to promote, ensure and integrate the global strategy of ‘gender mainstreaming’ into policy and programme planning. The concept ‘gender’ is used to describe socially constructed roles of men and women that are learned when growing up from childhood to adulthood, and that vary in different contexts and can change in time. Whereas ‘gender mainstreaming’ is "-a strategy for making women's as well as men's concerns and experiences an integral dimension of the design, implementation, monitoring and evaluation of policies and programmes in all political, economic and societal spheres so that women and men benefit equally and inequality is not perpetuated. The ultimate goal is to achieve gender equality” (Economic and Social Council 1997). The importance of gender and its effect on malaria prevention and decision-making dynamics influencing access to care have been recognized by RBM that has advocated a more broad analysis of gender aspects in malaria prevention (RBM 2005-2015). A few years ago also the Global Gender and Malaria Network urged to look at malaria from a gender perspective as: “Women have the greatest

To date, the gender perspective and its translation into practices have been largely missing in malaria prevention, despite the emphasis on tailoring malaria control to local situations by WHO already in the early 1990s. The global strategy on gender mainstreaming and growing evidence on the effects of gender dynamics on health related practices and well-being of women have had little impact on programme planning (Hoggenhaugen et al. 2003; RBM 2003; Tanner and Vlassoff 1998; Williams and Jones 2004). Researchers have not paid much attention to gender either, although it has been identified as an important area lacking research, as stated by Williams and Jones (2004) and more recently by Ribera and his colleagues (2007), who flagged the importance of understanding influence of decision-making and intra-household hierarchies on access to services regarding ma-laria in pregnancy. Findings drawn form a handful of malaria studies show that gender norms and values affect men’s and women’s exposure to mosquitoes. In many sub-Saharan African countries such as Malawi, women get up before dawn to carry out household chores and are exposed to anopheles mosquitoes that are known to be active from very late night to the early morning hours (Vlassoff and Manderson 1998; Chavasse 2002). Evidence clearly shows that pregnant women and particularly those of first and second parity as well as HIV positive pregnant women, are particularly vulnerable to malaria infection due to their decreased immunity (Duffy and Fried 2005; Steketee et al. 2001). Vulnerability to malaria

23

The Women in Development term and approach emerged at the beginning of the 1970s. It was first adopted by the Unites States Agency for International Development. Central to this approach was to recognize that women are key contributors for economic development, yet they had been completely ignored (Moser 1993:2). According to Moser (1993: 3), WID has been criticised because women were focused on in isolation, by promoting measures (access to credit and employment) to better integrate women into the development process, ignoring the real problem, women’s subordinate status to men.

47

women’s willingness to reveal their pregnancy. Witchcraft is perceived to be a common phenomenon, and pregnant women are perceived particularly vulnerable to witchcraft related problems discouraging them to reveal their pregnancy at an early stage. Findings also suggest that enrolment and attendance are delayed (or the opposite promoted) depending on the wellbeing of the pregnant woman. Long distances may also discourage some women from timely attendance. Walking long distances is often uncomfortable during pregnancy, and indirect costs such as transport and food may negatively affect women’s motivation to attend the ANC, as has been found in other studies (Mbonye et al. 2006a; Ndyuomugenui et al. 1998) As in Uganda (Mbonye et al. 2006a) and in Ghana (Smith et al. 2010), the pregnant women in Lungwena and its surrounding villages expressed their motivation to attend the ANC in order to be examined by nurses as it is considered the best way to know that the pregnancy is progressing well. ANC enrolment for receiving the ANC card was also considered important because it assured nurses’ atten-dance in case of an obstetric complication. (II.)

also varies according to different age groups, and particularly adolescent girls are vulnerable to malaria. A study in Malawi revealed that non-pregnant and pregnant adolescent girls had significantly higher parasite rates than women over 19 years of age (Brabin and Brabin 2005). Some studies have also shown that adolescent pregnant girls are less likely to attend an ANC and seek timely care for malaria (Mbonye et al. 2006a; Okonofua et al. 1992). A study in Kenya showed that women caring for their children or needing access to treatment themselves had to ask their husband for permission to access services (Molyneux et al. 2002). Women may also have difficulties accessing care because they lack economic support from their husbands or if they disagree with the recommendations of the therapy management group in the household which often includes their husband and family elders (Tolhurst and Nyonator 2006). 5.1.1. Pregnant women and prevention of malaria at the ANC As in many sub-Saharan countries, also in Malawi the MiP programme implementation is carried out in the antenatal setting in collaboration with the safe motherhood programme and in the framework of FANC. Nevertheless, despite the high ANC attendance, achievement of the RBM target on IPT has faced problems caused by late enrolment at the ANC, often delayed until the second trimester. Other problems have been inappropriate timing of SP-1 and low coverage of SP2. (Hamel et al. 2001; Holtz et al. 2004; Ndyuomugenui et al. 1998.) Findings from Lungwena concerning ANC attendance and timing of enrolment suggest that women wait until they are able to confirm their pregnancy. Both women and men perceived miscarriages as common and also recognized a condition called kupumula, which means that the woman’s body is just resting for one to three months (III). Therefore, confirmation of pregnancy takes time and foetal movement (around 18-20 weeks at the earliest) is the best way to be sure about the pregnancy. There are also cultural factors that affect some

5.1.2. Role of men Findings from this study show that men perceive their role as important in maintaining the well-being of a pregnant woman.24 It is the role of a husband to take care of the needs of a pregnant woman; to provide nutritious food to ensure the baby’s growth, and to provide clothes and necessities such as soap. “To do everything to ensure that his seeds are growing well”, as described by a young husband, father of one child. Although child bearing issues are traditionally considered women’s responsibility (Kishindo 1995), there are some implications that perceptions are changing. Some men saw it as the husband’s responsibility to ensure that his pregnant wife attends the ANC. They were also keen to know what happens at the clinic. 24

Most of the data presented here are unpublished and part of a manuscript under preparation.

48

chores: “You can’t assist your wife when she is pregnant. After the delivery she’ll talk awkwardly to other people saying ‘he is my brother, he does much of my work when I’m pregnant’. There is also a danger that other men will laugh at a husband who does “women’s work”.

There was even a desire to be involved with the ANC activities. The importance of involving husbands with the ANC activities was reasoned with the husband’s power in the family: he is the one who makes decisions, and therefore it would be better to allow him access to the ANC. Many of the men, however, preferred to keep their traditional role and leave pregnancy issues to women. Men should only be involved when there are problems requiring the husband’s attention.

Theobald and her colleagues (2006:301) state that one of the many challenges in ensuring gender equity in disease control programmes is “the need to recognise that neither ‘women’ nor ‘men’ constitute homogenous groups”. Among the matrilineal Yao, heterogeneity between women’s and men’s groups is visible. Men as husbands or men as maternal uncles have a different role in maintaining a pregnant woman’s well-being that is particularly visible in decision-making dynamics (Mitchell 1962). Perceptions and experiences regarding roles, responsibilities and power to decide are not homogenous among different groups either. For example, the role and responsibility of the maternal uncle was debatable among the Yao men. According to many, the husband is the owner of the woman within the marriage because he is the one who has built the house and controls the family’s resources. Yet men are painfully aware of the power relationships within a matrilineal society strengthened by the tradition of husbands moving to live in their wife’s villages and relatives as described by one Yao husband: “Wherever one goes in this world and marries, that’s not your home. The place where you have gone to marry does not belong to you, but belongs to owners of your wife and where she comes from and your wife belongs to her mother and her uncle. So the place where he is staying belongs to the wife’s uncle, and her biological mother.” There are signs of change in the living arrangements as 20% (n=248) of the women who participated in the KAP survey said that they live in their husband’s village.

Husbands have an important economic role that influences women’s access to care as noted in studies carried out in Kenya (Molyneux et al. 2002) and in Ghana (Tolhurst and Nyanator 2006). Studies in Malawi have shown that marriage is an economic necessity for women (Mitchell 1962; Schantz 2005). In Lungwena men had an important role being in charge of the household financial assets. Men themselves recognize this role and particularly the young husbands voice their worry and concern to be able to fulfill this duty. Women also recognized men’s financial responsibilities and duties. A pregnant woman, age 26, described her husband’s role as follows: “Whatever I want he does provide and even if I’m down with malungo [malaria], he does provide money to visit the hospital. - - When I’m pregnant, he does everything wisely, he even borrows money”. Some women also expressed their concerns and worries as well as powerlessness: “When I am pregnant, what if you find problems with what he [husband] is suppose to help you with and you get sick all the time, yet there is no money to go to the hospital, what should you do?” The perceptions of some of the female elders were rather harsh: “The husband just stands aside. He just brings relish, that’s all. Some men even beat their wives.” According to traditional gender roles, women carry out the household duties although some men recognized pregnant women’s need for rest. Perceptions of appropriate gender roles and expectations for appropriate behaviour are rooted deep in the culture and tradition. Many men considered it impossible to assist women with household

Most men shared the view that a husband is in charge of the family matters and has the right to make decisions concerning the pregnant woman when the situation is not severe. As 49

5.1.3. Role of female elders

studies on treatment-seeking practices of children under five years (Ahorlu et al. 1997; Hausmann-Muela et al. 1998; Munthali 2005) have shown, the seriousness of the situation is likely to affect involvement of extended family members in negotiatiating what action to take (Janzen 1978). The results from this study suggest that in situations that are interpreted serious and where the well-being of the pregnant woman and unborn baby is in danger, the authority and responsibility extends to a maternal uncle who in Yao society is responsible for ensuring the welfare of her sister’s children (Mitchell 1962). Findings from a safe motherhood study (Matinga et al. 2000: 18) revealed that among the Yao the maternal uncle and the pregnant woman’s mother featured prominently during delivery undermining the husband’s role in decision-making.

Female elders,25 such as traditional advisors and grandmothers, have been left without any recognition in malaria prevention programmes (Vlasloff and Tanner 1998). Findings from this study clearly show that female elders among the Yao have an important advisory role in pregnancy and delivery issues as one woman explains; “Female elders are the ones who can best explain problems in delivery because they are the ones who encounter these problems and solve them.” “The elders are the ones who help us here in the village. If they fail to assist us, then we rush to the hospital”. (see also Matinga 2000; Mitchell 1956). Traditional advisors (anankungwi), who are highly respected, elderly and experienced members of the community, perform the initiation ceremony (litiwo) to first time pregnant girls (mwali ndembo). The general purpose of litiwo is to provide advice on pregnancy; how to manage it and to prepare her for delivery. Another important purpose is to give moral education on appropriate social and sexual behaviour. (III, Mitchell 1956: 134-5; Phiri I 1997: 35-6; Zulu 1996: 32.)

Men’s and women’s perceptions of who has the power over the pregnant woman differed among the respondents. In general, the women agreed that the husband as head of the household is responsible for the pregnant woman because he is the one who made her pregnant. The husband deals with the everyday minor issues but the maternal uncle has the ultimate responsibility and authority as head of the clan. As for falling sick while pregnant, women say that they do not need to ask for anybody’s advice, they just go straight to the ANC to get medication when they perceive the illness to be mild. Sometimes they consult the husband, or the head of the household, who then consults the rest of the family. According to the findings from the KAP survey, 41% of the women said that they consult their husband when they fall ill, 36% consult their mother or grandmother and 7% their maternal uncle. According to the women, the maternal uncle is involved when the illness is perceived to be serious and dangerous. It is also possible that the maternal uncle can be asked for financial assistance for medication and transport costs. According to the women, men do not have much say unless money is required for transport or medication.

Female elders are perceived as custodians of particularly indigenous knowledge and herbal medications: “According to our culture, our parents advise us that we should take traditional medicine, but with hospital medicines, they never help” explained one woman during a FGD, followed by another woman: “That’s true. After visiting the hospital nothing helps, but when you use traditional medicine you get cured. So which treatment should you then follow?” In the case of traditional, local illnesses, women often first consult the husband who then most often seeks advice from the village elders. Nevertheless, not all appreciate the role of elders, as said by a pregnant woman: “There are some elders who don’t recommend 25

Part of the data presented here is unpublished and part of a manuscript under work, or has been presented in the Medical Anthropology at the Intersections Conference in Yale, 2009.

50

stubbornly relying only on traditional ‘harmful’ practices.

treatment from the hospital, but then you agree with your husband to go there for medication. The problem is that the elders believe in traditional issues. They are not used to hospital treatment.” Despite the different views concerning female elders and their cultural wisdom, they have decisionmaking power and they are important actors in quest for therapy as the following male response in a FGD suggests; ”We organize transport for her if the elders say that she needs to be referred to [mission hospital] Malindi”. Studies on maternal health in Malawi (Kerr et al. 2008; Seljeskog et al. 2006; Tolhurst et al. 2008) have shown that grandmothers have a powerful and multifaceted role in extended family influencing decision-making related to maternal and child health. Advice on motherhood issues from female elders, such as grandmothers and mother-in-laws, are trusted and listened to (Kerr et al. 2008).

Policies and programmes on prevention of malaria in pregnancy recognize the vulnerability of pregnant women due to malaria infection and can be considered gender sensitive in this respect. While I agree that it is very important to target pregnant women, I emphasize the fact that gender is not just about pregnant women. I support the emerging agreement that gender mainstreaming in infectious disease control, including malaria prevention programmes, requires extending the target groups beyond women to include other family members. 26 The current narrow focus on pregnant women and the antenatal setting ignores the community and extended family context where decisions to seek care are made and prevention practices are accepted or rejected (Nichter 2008). A study in rural Ghana showed the complex intra-household bargaining dynamics over children’s health care, clarifying the importance of including fathers and household elders as target groups in malaria control (Tolhurst et al. 2008). Some lessons can also be drawn from the field of maternal health that has advocated in the post-Cairo era the important role of men and communities in ensuring the continuum of maternal care between households and health facilities in order to reduce the

The grandmothers themselves see themselves as advisors to the pregnant women in pregnancy and delivery issues, just as their own grandmothers used to do. Although they have noticed that some young girls nowadays fail to heed their advice: “Nowadays these young girls consider your counseling as outdated and they don’t accept it.” Some lessons regarding the exclusion of female elders can be drawn outside of malaria research. In Senegal Aubel and colleagues (2004) attributed the lack of attention to grandmothers in maternal and child health (MCH) policies and programmes to two main factors. Firstly, the MCH focus has been on mothers ignoring the role of family and/or household and secondly, there are several negative stereotypes related to the role of grandmothers, discrediting their experience (see also Kerr et al. 2008). Studies in Senegal (Aubel et al. 2004) and Malawi (Kerr et al. 2008), clearly demonstrate that grandmothers are able to learn and integrate new information into their practices and positively influence those of fertile women, proving wrong the stereotype of grandmothers being incapable of learning new things and

26

Drawing on my work experience in UNICEF Malawi, the community based ITN projects lacked a gender perspective as they mainly target men. The community based ITN programme was implemented through the village health committees (VHCs) that were found in every village throughout Malawi. VHCs are voluntarily formed committees whose members are elected among the village members. UNICEF strongly encouraged that every committee should have at least one female member and many times there was one. Although in my opinion the woman was rather a token member than a fully recognized committee member. In most cases the chairperson of the committee was male and, for example, during my regular monitoring visits to the villages around Malawi I always met only male committee members unless I specifically also asked to meet the female member who, however, was in most cases busy taking care of the household duties or working in the farm and maize field.

51

staggering maternal mortality rates across the sub-Saharan Africa (WHO 2005).

globally and nationally (Heggenhougen et al. 2003: 37; Nichter 2008:7). 29

5.2. Local understanding of malaria in pregnancy (I, II, III)

In medical anthropology the focus has been on exploring what the local population knows about a certain public health problem; what local people think about it, what causes they attribute to it and more importantly, how they cope with the problem. Thus the approach in medical anthropology is different from public health that emphasizes knowing what the local population does not know, valuing public health knowledge over local knowledge and understanding of the problem. Yet knowing what local people know is crucial when new public health ideas and practices are promoted. (Hahn and Inhorn 2009: 6; Nichter 7.) As in many studies focusing on malaria in under-five children, none of the studies of malaria in pregnancy has elaborated on the diversity of perceptions related to it. In fact, investigations into the differences in the way illness is explained and responded to in the same society by different age groups and genders are largely missing from social scientific research (Nichter 2008: 9).

Malaria prevention programmes are still many times planned without adequate understanding of the local socio-cultural context 27 (see also Gardner and Lewis 1996; Nichter 2008). Programme planners and health professionals often perceive culture as a barrier (Sillitoe 2002) and knowledge held by local populations as wrong, containing misconceptions and not having adequate understanding of the public health and biomedical model of malaria in pregnancy. Misconceptions are assumed to be a significant contributing factor to low and inadequate compliance and coverage of the interventions (Lambert and McKevitt 2002). I use the word assume because these conclusions are rarely supported by research evidence that goes beyond simplistic knowledge, attitude and practice studies and an attempt to explore and explain local understanding from the lay perspective within a broader socio-cultural, historical, political and economic context that is central to medical anthropology research (Lambert and McKevitt 2002; Ribera et al. 2007). 28 The majority of studies have concentrated on knowing the “gaps in biomedical knowledge” and “cultural misconceptions” held by local populations because a major concern in malaria pre-vention is whether the local population and particularly pregnant women perceive malaria in pregnancy to be such a severe disease that encourages them to prevent it by utilising the prevention interventions, namely IPT and ITNs, promoted by RBM partners

5.2.1. Diversity of knowledge and perceptions Regarding the local understanding of malaria in pregnancy, I realized that the meaning of malungo was complex and ambiguous. Malungo is a Chichewa term that is commonly used to refer to malaria by professionals in the health care settings and its literal translation is joint pain, but it can also be translated as body pains or fever. My findings were in line with those of Helitzer-Allen and her colleagues (1993) who concluded that malungo is a term glossing several ‘malungo diseases’, each with distinct symptoms and aetiology. In other words, malungo is used to refer to many types of malungo illnesses, not

27

This argument is supported by my personal experience as a project officer responsible for planning malaria prevention programs discussed in more detail in Chapter 8. 28 Publications by Deborah Helitzer-Allen in the late 1980s and early 1990s are the only examples of medical anthropology research on MiP that I have come across.

29

See, for example, publications of Susanne Hausmann-Muela and Vinay Kamat of medical anthropology studies on malaria moving beyond perceptions to explain the logic of local understanding and multiple factors influencing treatment seeking practices.

52

and respond to my questions. There was a clear difference whether I spoke with a woman who had been pregnant many times or with a young girl pregnant for the first time. When returning to the field in 2006 I wanted to explore again the local understanding of malaria in pregnancy as well as other illnesses considered dangerous for pregnant women in the area, but this time from a broader perspective. I included men and female elders as a specific target group whose perceptions and knowledge I wanted to explore in more detail as these two groups emerged from women’s interviews as important actors providing support for pregnant women; men providing economic support and female elders social support and advice. I also revisited the local understanding of malaria in pregnancy through interviews with pregnant women during their ANC visits. As an end result from this process I was able to confirm my assumption that the Yao have a complex and diverse understanding of malaria in pregnancy that overlaps with the public health and biomedical understanding of the disease. (I, III.)

just malaria. The women recognized more than ten malungo types, but none of the types were associated with evil spirits (majini), showing the only main difference from the earlier findings. (I; Helitzer-Allen et al. 1993.) Studies of local illness classifications are common in medical anthropology research, and important lessons have been learned, contributing to our understanding of the interface between ethnomedicine and public health (see for an overview Nichter 2008: 6778). In malaria research, contributions have particularly been related to understanding the ambiguity of local terms used for malaria as well as relationships with folk illnesses caused by supernatural agents that manifest themselves through convulsions and cerebral malaria (Hausmann-Muela et al 1998; Kamat 2006; Winch et al 1996; see Williams and Jones 2004 for overview). Based on the findings of this study I made an important observation: malungo illness categories were vague and not shared by all members of the community. Although there was a certain consensus on the meaning of malungo, it could change in conversation (I,III). As Pool (1994: 117-118) has argued, local illness categories are produced and reproduced in encounters with local people. Furthermore, illness categories are dynamic, changing in time (Nichter 2008:76). Therefore, there is a need to be cautious when using the local terms in health education (Helitzer-Allen et al. 1993; Winch et al. 1996: 1057). Also generalization of illness categories to be applicable to other ethnic groups nationwide is not wise although most likely there are similarities (see UNICEF 2001).

Among all groups and respondents there was a big difference in the symptoms listed: on average each respondent reported one to three symptoms. Knowledge of causal agents of malaria in pregnancy varied. Many said that malaria is caused by mosquitoes, others mentioned several other natural causes while some said that they do not know what causes malungo (I,III). A similar finding among all groups was that treatment of malaria in pregnancy was considered best to be handled at the clinic. Nurses and medical staff were considered the most knowledgeable to treat malaria in pregnancy and any pregnancy related concern. Home treatment with pharmaceuticals and/or herbs was not recommended because of fear associated to bitter medication that was known to cause miscarriage. SP was perceived rather tasteless, although some women associated it with bitter taste. Nevertheless, it was well tolerated and perceived as a suitable pharmaceutical to be used during pregnancy, particularly because it was administrated at the ANC and the nurses were

I also found that there was no clear uniformity concerning perceptions and knowledge among women and men (III). Nichter (2008: 4) has pointed out that people of the same cultural background learn similar ways of talking and thinking, but their knowledge ‘of, about and how’ varies. I also realized that age, experience (multiple pregnancies), exposure to information through informal (community-based) and formal (ANC) channels affected women’s ability to narrate 53

considered trustworthy. (II.) Approximately half of the KAP survey respondents said that malaria in pregnancy can be prevented while the rest said that it cannot be prevented or they did not know. Also in other groups opinions varied from ‘possible to prevent’ to the opposite. ITNs were the best known method to prevent malaria. Knowledge and purpose of IPT-SP was mostly not known (II, III). Out of the 48 women, one pregnant woman interviewed immediately after her ANC visit said that malaria in pregnancy can be prevented with SP while 13 women mentioned ITNs (see article II, table 1). SP administered at ANC was interpreted as suggestive of malungo. Regarding prevalence of malungo, responses varied between common and uncommon. Some pregnant women said that malungo is not common because “most women take preventive measures” whereas female elders did not know if malungo is common. They just reckoned that “after the establishment of hospitals women are going for malungo treatment.” (II,III.) I have compiled all the knowledge presented by each group into Table 2. Summary of local understanding of malaria in pregnancy. Table 2 shows the large amount of knowledge available in the community. It is, however, important to keep in mind that there were great individual differences between the respondents as well as the groups.

54

Table 2. Summary of local understanding of malaria in pregnancy Malungo

Health education at ANC

Pregnant women

Women of reproductive age Mild fever Shivering Vomiting Diarrhoea Loss of appetite Don’t know

Symptoms (I)

Mild fever – high fever Body pains Shivering Feeling cold Feeling unwell Headache Vomiting Don’t know

Etiology (I,III)

People say it’s mosquitoes Don’t know

Treatment (I)

Panadol SP at ANC Mapilisi at ANC Advised not to buy medication from shops SP Visiting ANC and getting some pills to be taken at home ITNs When going to the ANC nurses give some pills No prevention

Prevention (II)

“We give you medicine for malungo” “Malungo has reached a critical stage, so we give you medicine called Fansidar” “You take Fansidar here at the clinic. --- As a result we are going to defeat malungo” “You get Fansidar for malungo and you should buy at least one mosquito net” “Government wants each and every pregnant women to have at least one net”

55

Female elders

Men

Headache Back pain Don’t know

Mild fever - high Fever Body pains Shivering Feeling cold Headache Joint pains Don’t know

Mosquitoes (not all mosquitoes) Rubbish / unclean compound Hard working Bathing in cold water Don’t know Panadol, Aspirin “because malaria in pregnancy not very strong illness” Mapilisi at clinic

Mosquitoes Don’t know

Mosquitoes Hard working Bathing in the lake Poor diet Unsafe water Cold weather Don’t know Panadol, Aspirin Mapilisi at clinic In the home setting danger of wrong drug

ITNs Cleaning the compound to keep mosquitoes and dirt away “Nurses say that keeping your home clean prevents diseases” Covering water sources No prevention Starting ANC early Taking Fansidar at the ANC

ITNs Treatment before delivery ”Maybe going to hospital?”

Panadol SP Mapilisi at ANC

Attending ANC Go to ANC for pills No prevention Good hygiene Eating well cooked food God’s will Burning leaves to create smoke that chases mosquitoes away

Malungo

Health education at ANC

Pregnant women

Consequences (I,II)

“If you suffer from malungo you are likely to become anaemic and you miscarry” “Most pregnant women are dying here because of this malungo, and the government is concerned. This is why the price [ITN] has been reduced from MK140 to MK50.”

No consequences because “you are just given Fansidar and then you fall sick every month, which is not serious” No consequences, you just become too weak to work. Miscarriage if one doesn’t take enough medication Miscarriage if severe malungo Woman becomes weak and if unlucky, may die Dehydration If severe, malungo can lead to miscarriage Not common “Not common because most women take preventive measures” Common

Prevalence (III)

Perceived severity (I,II,III)

Not dangerous because one can just get SP “Malungo is dangerous but we don’t take it seriously”

56

Women of reproductive age Miscarriage Maternal death Causes lack of body fluids and blood (a malungo type) Baby in the womb may become weak and thin

Not common Common

Not serious, malungo is just an illness Not serious, malungo is the first visible sign of pregnancy Can be dangerous because of consequences for the pregnant woman and baby

Female elders

Men

Miscarriage Can lead to death without treatment Causes woman to be useless in delivery Premature birth

No consequences Harmless illness Feeling uncomfortable Miscarriage Can cause maternal death if not treated in time Can cause maternal death if cerebral malaria Women can become unconscious Baby born weak

Not common “Don’t know if common but after the establishment of hospitals pregnant women go for malungo treatment” Common

Not common Common

Can be very severe if the pregnant woman is reluctant to attend the ANC

logics, resulting in syncretic models. Nichter (2008: 44, 62) defines syncretism as the production of hybrid ideas formulated through interaction between indigenous illness causalities and biomedical explanatory models.

5.2.2. Syncretic explanatory model of malaria in pregnancy There is a danger that the reader will interpret Table 2. on local understanding of malaria in pregnancy that the knowledge of a rural Yao, a pregnant woman, a woman of reproductive age, a female elder or a man, is full of misconceptions, inaccurate information and beliefs that need to be corrected with “appropriate health education messages.” In other words, the local understanding of malaria in pregnancy is perceived to be containing “wrong knowledge” that needs to be replaced with correct biomedical and public health knowledge. I argue that this type of interpretation and action would be too shallow and would not lead to the expected result: improved knowledge, changed practices and achieved targets. Before meaningful, tailored and culture sensitive health education messages can be developed we need to know the underlying factors contributing to building up an explanatory model of an illness (Kleinman 1980).

The first step is to recognize that the biomedical and public health knowledge among health professionals also varies and affects the content and emphasis of health education messages at different times, as also pointed out by Hausmann-Muela and her colleagues (2002). Views between professionals may also differ causing disagreements on what preventive measures should be promoted as a national strategy. During one of the consultative stakeholder meetings organised by UNICEF as part of the malaria policy consultations in 2001, a Resident Director of an International NGO made a comment that spreading health education messages that recommend communities to clean up their compounds, to drain puddles and to cut down the vegetation around houses in order to prevent malaria is useless and a waste of time. He explained that Anopheles mosquitoes breed only in clean, sunlit water and small natural breeding sites such as animal footprints and small areas of flooded grass. Water sources nearby houses are often polluted with faeces, rotting vegetation and garbage and therefore not suitable breeding places of Anopheles mosquitoes. According to the Resident Director, mobilising communities to clean up their communities was therefore not justified for malaria control. His statement generated a small storm in the conference room particularly among the government officials from MoHP that held the opposite view. In Malawi cleaning compounds and cutting down vegetation and grass were considered an important part of reducing the burden of malaria following the strategies of malaria eradication campaigns between the 1960s and 1980s (see Alilio et al. 2004; Brown 1997 for historical overview). This incident showed that it can be hard to accept new ideas and knowledge, and to incorporate them into practices when they go

An important part of the puzzle is to understand that the explanatory models of malaria in pregnancy are syncretic, formulated through encounters with health education messages on malaria distributed through different kinds of information channels. The local understanding of malaria in pregnancy is a syncretic model containing a mixture of cultural knowledge learned through experience, advice from elders in the community, as well as biomedical knowledge received through various kinds of formal channels such as the ANC. Hausmann-Muela and her colleagues (2002) introduced the concept of medical syncretism into the local understanding of malaria among rural population in Tanzania. They define syncretism as “blending of biomedical with indigenous concepts” (ibid: 404). In their study they showed how malaria-related biomedical knowledge transmitted in health education messages coexisted, interacted and merged with local pre-existing ideas and 57

conducted in the Lungwena area (Shame et al. unpublished report) revealed that most people in the study area had problems understanding Chichewa although they claimed that they did understand the language. As a consequence, a lot of information remained understood. Montgomery and colleagues (2006) described a similar type of delivery of health education on malaria in Tanzania where health educators ‘preach’ and ask women to repeat the answers creating a hierarchal power structure that undermines the agency of the mothers.

against one’s own views (see also Yoder 1997). The way messages are conveyed and in what circumstances also matters. The nurses who implement the MiP interventions also have their own understanding and knowledge that differ from the knowledge of epidemiologists, medical doctors and national and international professionals. Nurses transmit health education messages based on their explanatory model that is a mixture of public health and indigenous knowledge, experience and assumptions regarding community understanding of malaria in pregnancy. The results from the observations conducted at the ANC show that the content of health education messages on MiP was vague, unspecific and not containing any information on the preventive purpose of IPT-SP. The nurses used words such as “we give you medicine for malungo”, “malungo has reached a critical stage”, “we give you Fansidar – to defeat malungo” (see Table 2). IPT-SP was administered under DOT with no instructions or communication. On the wall of the ANC ward there was a poster saying: “Amai apakati imwani SP pa mwezi wa 4 ndi 7 kuti mupewe malungo”, meaning a pregnant woman should take SP between 4 and 7 months of pregnancy to prevent malaria. The nurses also spent approximately 15 to 20 minutes to sell subsidised ITNs to each pregnant woman attending the clinic as part of the joint Donor and Government effort to scale up ITNs nationwide through the ANC platform. The nurses were very motivated to sell the nets as they received a commission of 5MK for each net sold. It was, however, observed that nurses gave no instructions on how to use the net or how and when to re-treat the net. All their efforts were devoted to selling the nets. The nurses used Chichewa as their language of communication, which is not the mother-tongue of the pregnant women and the language used in the communities. This increases the likelihood of interpreting the messages differently than intended in the public health perspective. (II.) A sociolinguistic study of communicative practices

A closer look at the findings shows how pregnant women have merged and integrated MiP-related messages transmitted by nurses at the ANC as well as interpreted the practices among the nurses on implementing the IPTSP policy and ANC procedures as part of their explanatory model of malaria in pregnancy (see Table 2 presented earlier). Findings suggest that women interpret IPT-SP as a suggestive of malungo rather than a preventive measure: “Leaving home you think you are fine. You are surprised that the nurses discover that you are suffering from malungo, because they give you these pills, Fansidar.” The abdominal examination by nurses reinforces the interpretation that ‘pills’ are given for treatment because SP is administered immediately after examination without instructions or explanations: “We take Fansidar only after being examined and the nurses have found out that we are ill”. Thus it may seem for the pregnant women that all of them are suffering from malungo, strengthening the perception that malaria in pregnancy cannot be prevented. Nor did the pregnant women seem that worried about the consequences of malaria in pregnancy, “Malungo in pregnancy has no consequences because you are just given SP at the ANC, and even if you don’t take Fansidar you just feel sick every now and again.” It is very possible that many of the pregnant women perceive malungo as a mild illness that is easily handled with SP administered at ANC. After all, nurses detect that a pregnant women is suffering from malungo during a routine 58

and mild malungo. In the old days there were none.” They had also noticed that after the establishment of the Lungwena health centre at the beginning of the 1990s pregnant women have been going there for ‘malungo treatment’ (III.) Malawi adopted the IPT-SP policy in 1993 when ANCs started to implement it as part of the routine ANC services. In all groups ITNs were recognized as a preventive measure among those who said that malaria in pregnancy can be prevented. I attribute this largely to the fact that since 2001, there have been enormous efforts to scale up ITN coverage. PSI, the main agency implementing the ITN programme, has been very successful in promoting the nets through social marketing and utilising various kinds of communication channels such as the radio and local newspapers. Also in Lungwena, pregnant women were really keen to buy the nets which they saw as desirable household items. Thus women of reproductive age, female elders and men regularly saw pregnant women returning from the Lungwena ANC with their nets. Men also heard radio messages on “chitezezo nets” while spending time at the Lungwena trading centre.

ANC checkup, while in most cases she does not feel sick or have any symptoms. Also concerning the prevalence of malungo, the opinions among pregnant women varied from common to uncommon. Again it seems that the introduction of ‘preventive measures’ for reducing the burden of malaria has affected perceptions concerning prevalence and the need to worry about malungo during pregnancy. (II, III.) To sum up: the analysis suggests that the messages transmitted by nurses and their practices regarding IPT-SP implementation have had unintended consequences. Instead of creating awareness of dangers of malaria in pregnancy, they seem to have done the opposite, namely to reduce pregnant women’s fears regarding the illness. A closer look at the explanatory models of fertile women, female elders as well as men show how they have also merged and incorporated biomedical and public health knowledge creating a syncretic model of malaria in pregnancy. For example, both the fertile women and the men had a broad understanding of aetiology and preventive measures; a mixture of mosquitoes, poor hygiene, unsafe water, cleanliness etc. This suggests that the community mobilisation messages on clearing the bushes and keeping the compound clean from the global malaria eradication era have also trickled down to the local communities during the past 30-40 years also in Malawi, and been merged with the local understanding of causalities and preventive measures. Helitzer-Allen and her colleagues (1993) suggested that three malungo types, malungo caused by mosquitoes, malungo caused by food contaminated by flies and malungo due to dirty water, had been created through exposure to health workers’ and agriculture extension workers’ health education messages. The female elders had the least knowledge of malaria. Their knowledge concentrated on the indigenous illnesses in the area rather than on malaria which they considered a rather new illness in the area; “To tell you the truth, we know malungo these days. There are two types of malungo; serious

5.2.3. Health education on malaria Health education on malaria aiming at behaviour change is one of the key interventions in prevention of malaria (GMAP 2008). A common recommendation among social scientific studies is that there is a need to develop health education messages to improve the local populations’ knowledge of malaria regarding public health facts leading to behaviour change (e.g. Gikandi et al. 2008; Nganda et al. 2004). This recommendation is often based on the assumption that poor and low compliance is caused by ignorance on the part of the pregnant women and the community. It is taken for granted that knowledge of importance of IPT or mosquitoes as a transmitting vector leads to behaviour change. Hausmann-Muela and her colleagues (2002) criticised this common approach in malaria research where results on the local under59

nity in supporting the behaviour change of adolescent girls has been crucial (Hogg et al. 2005.)30

standing of illness classification are interpreted through the biomedical-traditional dichotomy. Good (1994) has criticised the hierarchical and superior position of scientific medical knowledge over folk beliefs and indigenous knowledge; scientific knowledge forms the basis for the conceptual framework of the Health Belief Model that is used to health education and promotion in malaria and other infectious diseases (see also Yoder 1997).

Some lessons learned can also be drawn particularly from health education on sexually transmitted illnesses such as HIV/AIDS. In Malawi many projects have used drama groups and plays on specific topics tailored to fit the cultural context utilizing the local language. Local drama groups are well received by the communities as they are popular and entertaining events bringing together the whole community. Hanne Mogensen (1995:99-100), who has conducted a medical anthropology study of AIDS in rural Zambia, advocates participatory theatre in which concepts of the participating audience are used as a starting point for the story to be acted by the drama group. According to her, health education should be based on local people’s own perceptions of and narratives on obstacles to behaviour change rather than exposing local communities to our ‘Northern’ understanding and knowledge (Mogensen 1995).

Observations at the ANC show that health education messages are delivered in a teachercentred and dominative manner. The pregnant women have a passive listening role, occasionally being asked to repeat what the nurse has said. (II.) According to Montgomery and colleagues (2006), teacher-centred education where patients are asked to repeate aloud the message in their own words allows medical staff to maintain greater control over women and as an adverse consequence create a selffeeling of being uneducated and lose faith in one’s own knowledge. They further point out the paradox in health workers’ discourse and perceptions of mothers: they are at the same time knowledgeable of their children’s affairs, as well as uneducated and ignorant rural women with little capacity to understand health education messages (ibid). In Malawi, according to the conventional teaching approach, first grade primary school students are already taught to respect the teacher who stands in front of the class and asks the students to repeat his/her words (personal observation during monitoring visits to UNICEF supported primary education projects 19981999). According to Stambach (1994), the east African schooling system strengthens the teacher’s authoritative position: being a teacher at a primary school or a nurse in a rural health centre means having authority marked with a superior position of the ‘educated’. An Adolescent Girls Literacy Project (AGLIT) in Southern Malawi showed that participatory teaching and pedagogical methods that are functional led to good results and increased the self-esteem of the adolescent girls. Also involvement of the wide commu-

There is clear evidence that increasing and improving knowledge does not necessarily lead to changes in perceptions and practices because behaviour and practices are far more than just knowledge and beliefs. Social, economic, structural and political factors influence people’s willingness and agency to change their behaviour, to adapt or to reject new treatment and prevention practices promoted by international health programmes. (Espino et al. 1997; Helitzer-Allen et al.1993; Nichter 2008: 6-10.) Therefore, “tailored messages”, advocated by RBM partners, can only be planned and implemented when several issues are taken into account (see Chapter 7 Recommendations for malaria prevention programmes).

30

I visited the AGLIT project in Chikwawa district in 2000. The teaching pedagogy differed from the conventional primary education pedagogy, concentrating on a functional curriculum that takes account of community needs.

60

harmless illness. Drawing on the findings it is possible to suggest that there are four factors that weigh into the perception of risk among the Yao, i.e., how dangerous an illness is perceived (III). 31

5.3. Malaria in the context of multiple vulnerabilities in pregnancy In Malawi, despite FANC and collaboration with the safe motherhood programme, prevention of malaria is managed as a single disease approach with little association to other diseases and conditions problematic in pregnancy (MIPESA 2006). A few social scientific studies (Kengeya-Kayondo et al. 1994; Mbonye et al. 2006a; Nuwaha 2002) have reported that malaria is perceived as dangerous for pregnant women. Nonetheless, these studies have not explored what factors contribute to the local risk perception of malaria, nor studied malaria in relation to other illnesses and maternal health problems. An underlying assumption in studies that have discovered that malaria is a mild illness, is that the perception of nonseverity is due to the local conceptualisation of malaria as a broad symptom complex, often referred to as fever in the local language (Agyepong 1992; Ahorlu et al. 1997; Winch et al. 1996) and feeling unwell or general malaise (HausmannMuela et al. 1998; Kengeya-Kayondo et al. 1994) without further investigation into the factors weighing into local risk perceptions.

31

The findings presented and conclusions made here regarding local risk peceptions are a result of a long process of more than six years. Before entering the field in 2002, I had already made attempts to shift my approach towards a more horizontal approach (Nichter 2008: 80) by taking pregnancy as the starting point for this research. After returning to Finland I carried out data analysis and wrote a long draft manuscript titled: “Why is malaria in pregnancy not perceived as a dangerous disease among the pregnant Yao women?” I soon realised that the manuscript contained too many interlinked topics with no clear focus and theoretical discussions. Although there were good points, many that I have utilized in my published articles, it was obviously too messy and would have required a huge job to get it published in a medical anthropology journal. Being based at the school of public health, I also lacked an anthropological peer group that would have allowed important reflections regarding my data analysis, and eventually I got discouraged. At the same time I was also expecting my first child, and I felt pressure to publish my first article. At this point of the research I was still very much feeling and thinking as a UNICEF project officer, more than an anthropologist, and I considered writing a public health type of article much easier than an anthropological one. Therefore, although I realised that the local understanding of MiP had to do with the many other dangerous illnesses in Lungwena, as well as pregnancy and delivery complications that emerged from the women’s narratives, I ended up writing an article in the framework of the traditional single disease approach (I). Later on when I had an opportunity to return to the field, I came back to question why malaria in pregnancy was not perceived as dangerous, and carried out analysis of my data in order to identify data gaps and to focus research themes. I decided to focus on exploring pregnancy related vulnerabilities to see how MiP was related to them. Being a mother and pregnant, expecting my second child, were also important factors when planning the focused data collection because they allowed me to understand motherhood and pregnancy much better in the context of everyday life in rural Malawi. During the fieldwork, I thought about the perceived dangers and risks for my pregnancy in Finland and for a pregnant woman in a rural Yao village, realising that we were worlds apart and used a different risk discourse. Back in Finland when I had already started analysing the data and drafting the article (III), I received the news of Onni’s death, which triggered me further to question the international health approach regarding pregnant women at risk of malaria.

5.3.1. Explaining why malaria in pregnancy is not perceived as dangerous among the Yao Focusing on issues that crosscut diseases is important because they allow better understanding of local perceptions of vulnerabilities and risk, and consequently allow us to draw lessons for health related practice, as Nichter (2008: 80-81) has pointed out. Examining malaria in pregnancy in relation to other illnesses perceived as dangerous for pregnant women in Lungwena and its surrounding villages, as well as examing it in the context of pregnancy, provided valuable insights into local risk perceptions. This made it possible to establish women’s exposure to multiple vulnerabilities during pregnancy, namely regarding witchcraft and extra-marital affairs, and to find explanations on why women described malaria in pregnancy as a rather 61

The type of risk is also meaningful: does the illness expose the pregnant woman to medical risks that she has agency to control and/or social risks that a pregnant woman has limited agency to control? Malaria exposes her to medical risks while STIs and particularly HIV/AIDS, because of extra-marital relationships, also expose her to social risks (risk to valued social relationship, Nichter 2008: 58) that can outweigh medical risks, as Bujra (2000: 74) and Nichter (2008: 58-59) have shown in their studies. Acceptance of medical risks of STIs can also be attributed to the perception that STIs such as malaria are easily treated at the Lungwena ANC. Hence judging risks in contexts where there are multiple diseases, health concerns and dangers, and in relation to other risks, is highly valid (Nichter 2008: 60). As Gramiccia (1981: 386-387) argued already almost 30 years ago that in the context of everyday hardships such as poverty and hunger coupled with diseases and conditions, prioritising elimination of malaria makes little sense. Hence, it is important to recognize that malaria is not a stand alone issue in people’s lives and to understand how it weighs in relation to other everyday concerns (Jones and Williams 2004; Kamat 2008).

These factors are 1) perceived adverse consequences in pregnancy, 2) easiness to treat and cure an illness, 3) transmission mode and agency to control, and 4) type of risk (social/medical). Both men and women considered sexually transmitted illnesses (STIs, referring here to gonorrhea, chancroid and syphilis), HIV/ AIDS and malaria the most dangerous illnesses in the area because they can cause either a miscarriage or at worst a maternal death, even though adverse consequences of malaria were perceived to be preventable with mapilisi at an ANC (referring to IPT) and attending the ANC.32 STIs and malaria were considered possible and easy to treat at the ANC, whereas HIV/AIDS was perceived neither as treatable nor curable making it a really worrisome and dangerous illness. As many malaria studies (Agyepong 1992; Mwenesi et al. 1995; Nuwaha 2002; Samuelsen et al. 2004; see also Jones and Williams 2004) have suggested, normal malaria is perceived as mild because mosquitoes and other causal attributes are natural. Also STIs and HIV/AIDS are caused by natural agents, but with a significant difference from malaria regarding a pregnant woman’s agency to control. A pregnant woman has limited possibilities to control STIs and HIV/AIDS because she would have to be able to control her husband’s extramarital relationships that are perceived as common. Neither condom use nor divorce is perceived as a viable prevention option (Schatz 2005). A pregnant woman can prevent malaria by applying the currently available, free of charge public health measures: IPT administered at ANC, presuming that she is able to access the services. The issue of sleeping under an ITN is more complex, and pregnant women’s agency to utilise an ITN may depend on factors beyond their control.33

Malaria prevention actors at global and national level find it crucial for successful MiP interventions that pregnant women’s ANC attendance, following the FANC schedule, is ensured. While this is important, it is also important to realise that ignoring local risk perceptions may have adverse consequences for the achievement of MiP targets and programme sustainability. In Malawi, utilization of ITNs is still low, which purchase a net, sleeping patterns in a household, perceptions on ITNs etc. (Rashed et al. 1999; UNICEF 2001). In Lungwena most pregnant women seemed to have money to buy a net during a routine ANC visit. There were rumours circulating that some pregnant women bought nets for the fishermen. This may be very true as it was observed pregnant women purchasing nets on behalf of non-pregnant people such as the Malawian research staff affiliated to the health centre. Use and sleeping patterns were outside of the scope of this study.

32

I see this perception as an unintended consequence of implementation practices of IPT-SP policy combined with nurses’ health education on malaria in pregnancy as discussed earlier (see Chapter 5.2.2) 33 Studies on access and use of ITNs have shown that hindering factors are related to availability funds to

62

consequence with IPT-SP. Nor does the approach recognize the influence of these factors on men’s ability to perform their duty to take care of the well-being of a pregnant woman. Missing from the global discussion on malaria as a disease of poverty and inequality is a discussion about the social burden of the disease in the wider socio-cultural, economic and political contexts of everyday suffering (Jones and Williams 2004). Several anthropologists have called attention to examining malaria in the wider framework of perceived vulnerabilities affected by multiple socio-cultural, economic, political and structural factors (Kamat 2008; Sommerfeld et al. 2002; Williams and Jones 2004; see also Heggenhougen et al. 2003.) Brown (1997: 132-134) has argued that the major reason for reappearance of malaria is directly associated with increased poverty all over the world. Farmer and his colleagues (2006: 1686) use the concept ‘structural violence’ to show the complexity of social arrangements that place individuals and populations in harm: “the arrangements are structural because they are embedded in the political and economic organization of our world; they are violent because they cause injury to people (typically, not those responsible for perpetuating such inequalities)”.

is most likely due to the local risk perceptions of men and women. In addition, the IPT-SP intervention is sustainable as long as the pregnant women attend the ANC and receive SP under DOT. If women’s ANC attendance is interrupted for one reason or another, this will presumable have an impact on the MiP and safe motherhood interventions at the ANC. A recent study in Malawi showed that introduction of community-based IPT-SP delivery negatively affected the ANC attendance reducing it below 90% although it increased the IPT coverage (Msyamboza et al. 2009). Among the Yao, community-based delivery of drugs such as SP would not necessarily be a feasible option because of the mistrust regarding administration of drugs by non-professional people due to fear of miscarriage. (II.) 5.3.2. Recognition of economic and structural factors It is globally recognized that malaria is a disease of poverty and inequality: it is a disease of poor people and it keeps people poor. Costs of poverty due to malaria are often presented in quantifiable terms of lost Gross Domestic Product, which alone in Africa is estimated to be 12 billion USD each year (Gomes 1993; Sachs and Malaney 2002; Stratton et al. 2008). The economic burden of malaria on households is also acknowledged in terms of lost productivity and household expenditure on malaria treatment (Asenso-Okyere and Dzator 1997). The economic burden of malaria is used to advocate the importance of prevention of malaria. Poor and inadequate compliance is often attributed to pregnant women’s ignorance or deviant behaviour, ignoring economic and structural factors beyond their control. More than ten years ago Farmer (1997) pointed out that we should not exaggerate people’s agency and power to take action that will improve their health and wellbeing. The current MiP approach does not pay attention to the multiple structural and economic factors that affect pregnant women’s access to services and ability to comply, for example, with the FANC schedule and as a

I did not directly study structural and economic factors affecting pregnant women’s ability to access MiP interventions in Lungwena. Yet these factors; poor services, lack of skilled personnel and equipment, long distances, poor infrastructure, scarcity of household financial assets and circumstances that can in a worst case scenario lead to tragic outcomes, emerged from the interviews with both women and men. Baby Onni was only four days old when he died because of malaria (article III). In his case, the tragedy started when the mother developed symptoms of fever around 32 weeks of pregnancy. The husband advised her to seek medical care at the Lungwena health centre, where services are free and which is less than two kilometres from Chapola village where Onni’s parents live. There she received SP, which is the first63

pay for a bicycle taxi, return fee MK10034. She will spend an average of 5 to 6 hours on the visit, leaving home around 7 a.m., spending approximately 2 hours in the clinic, and returning home around 11 a.m. During her visit she will spend MK50 for a lunch snack if she has money. If she is referred to St. Martins’ mission hospital she will also have to find money for medical bills that are around MK600 for treatment of malaria. When walking or travelling with a bicycle taxi is not possible, a pregnant woman needs to find funds to hire a vehicle that can cost up to MK3000. Indirect costs related to ANC visits can be a factor delaying attendance particularly when a woman is feeling well. With a 100 kwacha a pregnant woman can buy 1 kg of maize flour or depending of the season 520 tomatoes or a quarter litre of cooking oil etc.

line treatment for malaria in pregnancy, and was sent back to home. Her condition failed to improve and after three days of observation, Onni’s parents made a decision to seek care again because now they interpreted the situation to be serious. The mother returned to the Lungwena health centre, and this time she was referred to St. Martin’s mission hospital, 16 kilometres from Lungwena. Onni’s father had to have money for transport and the services. At St. Martins, Onni was born prematurely, about 8 weeks earlier than expected, nevertheless weighing nearly 3 kg. Two days after the delivery, Onni’s mother and Onni were discharged. During the first night home Onni developed fever and refused to suckle at the breast. The next morning Onni’s parents took him to Lungwena, and he was again referred to St. Martins. Again Onni’s parents had to find money for transport, services and treatment. At St. Martins Onni was diagnosed with malaria and both Onni and his mother received treatment. At this point Onni already had a high fever and the following night he died. Onni’s story shows that his parents did their best to ensure Onni’s survival, but they had very little chances because of the circumstances and structural factors. In four days there were three visits to Lungwena, two referrals to Malindi, an obvious lack of communication between Lungwena and St. Martins, lack of advice and support for the parents and unfortunate timing. Sub-Saharan Africa, where approximately 90% of malaria cases occur, there are plenty of babies that have had similarly tragic fates. According to estimates, every 30 seconds a child dies due to malaria, accounting for 3000 child deaths each day (WHO 2009).

Pregnancy and delivery are a cause of worry because of the uncertainties related to the outcome. Particularly men expressed their worry when the time of delivery approaches. They worried that the delivery might not progress as it should. The woman may need emergency obstetric care and the nearest service is available at St. Martins Mission hospital. A caesarean section costs MK10,000 and the total costs can amount to MK16000 including the operation, medical attendance, medication, food and transport35. Funds such as these are rarely available for anyone living in the area. Mobilization of household assets is a critical factor influencing people’s access to health care (Obrist et al. 2007:1858). The burden of malaria in pregnancy can only be reduced if we tackle the problem comprehensively by applying “integrated designs of interventions that are placed within the broader social, cultural, political, and economic context” (Williams and Jones 2004: 160).

The majority of families in Lungwena live in extreme poverty having little if any cash available to be used on indirect costs that include transport, travel time associated with ANC attendance or sudden treatment-seeking needs. For example, a pregnant woman from Moto village lives seven kilometres away from the Lungwena health centre. If she needs to seek treatment for malaria she will either have to walk seven kilometres, or get funds to

34

Excahange rate 1 Usd = MKW 147, 1 Euro = MKW 214 (January 2010) 35 Information of prices related to transport, medication and food items were collected in January 2010 by my research assistant.

64

6 SCIENTIFIC CONCLUSIONS BASED ON EMPIRICAL FINDINGS

Second, I studied the local understanding of malaria in pregnancy aiming for an idea of what kind of knowledge existed in the community and how this knowledge has been generated. There is a diversity of perceptions and knowledge regarding malaria in pregnancy among different groups in the community, and there is no one kind of local understanding that is shared by all. Illness categories are vague and constantly negotiated in encounters between people. Explanatory models of malaria are syncretic models containing a mixture of biomedical knowledge and local indigenous knowledge generated through formal public health and informal community-based channels. The explanatory models of different groups differ depending on their main sources of information. Pregnant women’s models are influenced by health education given by nurses at the ANC. Observations of types of messages and communication methods at the ANC show that there is a danger that the IPT-SP policy and its implementation can have unintended consequence: it may lessen the perceived dangerousness of malaria in pregnancy among pregnant women. Access to public health information on malaria in pregnancy varies between different groups: female elders, who have an important advisory role, do not have direct access to public health information, whereas men rely on sporadic health education messages through the radio and hearsay from their pregnant wives. Yet these groups participate in maintaining the wellbeing of a pregnant woman and are important decision-makers when treatment options are discussed.

The scientific conclusions drawn here are related to the broad purpose of this study that was to gain contextual understanding of perceptions, knowledge and practices regarding pregnancy among the Yao and to examine how these affect utilization of MiP interventions at the ANC and planning meaningful programmes. I have compiled the findings of each individual article (I-IV) into Annex 1. I first examined household members’ participation, including their roles and responsibilities, in taking care of the well-being of a pregnant woman. Yao men as husbands have an important role in ensuring the wellbeing of the pregnant wife. This role is directly linked to their role as household heads. Husbands are both decision-makers in family matters and participate in treatment-seeking for mild illnesses. They are specifically responsible for financial matters and are expected to have financial assets to cover costs concerning pregnancy related problems and illnesses. Responsibilities between the husband and the maternal uncle are debatable, but in general, the maternal uncle is perceived to have the ultimate responsibility as head of the clan, and his assistance is called for in serious matters. Female elders, that is, traditional advisors and grandmothers, have an important advisory role in pregnancy related matters and motherhood issues because they are widely respected and trusted in the communities. Female elders are perceived as guardians of indigenous knowledge and they enforce and maintain local traditions. Female elders may also participate in decision-making concerning pregnancy related treatment-seeking practices, particularly at the time of delivery. Both husbands and female elders have limited agency to perform their duties due to inadequate access to information on malaria in pregnancy, as well as due to the prevailing structural and economic factors common in resource poor settings such as Lungwena.

Third, I wanted to understand why malaria in pregnancy was not perceived as a dangerous illness, by judging malaria in the context of pregnancy, multiple dangers and illnesses. Yao women’s vulnerability during pregnancy is increased by extra-marital affairs and witchcraft related factors which women have limited, if any agency to control. In addition, there are multiple illnesses, particularly sexually transmitted illnesses that are perceived as 65

stant balancing act between different factors and different actors. Malaria is only one of the many concerns in everyday life. In this context, current efforts of preventive programmes aiming to improve pregnant women’s public health knowledge of malaria may gain rather little attention and impact because they don’t address directly the main concerns and factors that affect pregnant women’s practices for seeking treatment and prevention in Lungwena.

dangerous for pregnant women, malaria thus being just one of the many illnesses. The risk perceptions among the Yao are different from those of public health explaining why malaria in pregnancy is not perceived as dangerous. The findings suggest that there are four interwoven factors that weigh into the perception of risk and explain how dangerous an illness is perceived, namely 1) perceived adverse consequences in pregnancy caused by the illness, 2) easiness to treat and cure the illness, 3) transmission mode and agency to control the illness, and 4) the type of risk caused by the illness is meaningful (social/medical risk). Malaria in pregnancy may have adverse consequences such as miscarriage and at worst lead to maternal death, but this is rare because these consequences are easy to prevent by mapilisi (IPT) from the ANC. Malaria in pregnancy is also perceived to be caused by natural agents and therefore easy to treat and cure. It is also preventable by ITNs as most pregnant women seem to have money to purchase a net. Malaria in pregnancy exposed women to medical risks only. In addition, both women and men expressed worries related to everyday life in a resource poor setting and structural and economic factors (many of which are beyond the control of the pregnant women and the men) affecting access to health care. I have summarised the conclusions drawn above and the findings of each individual article (see Annex 1) to Figure 10. Interwoven factors that affect prevention of malaria in pregnancy in rural Malawi (see next page). The figure shows the complexity of the everyday context in which a pregnant woman lives in rural Lungwena; multiple factors (witchcraft, extra-marital affairs, multiple illnesses, everyday worries, structural and economic factors) that affect a pregnant woman’s access to preventive services, her agency to carry her pregnancy to term in the ANC and community context, and involvement of household and extended family members in maintaining her well-being and assisting her to access health services (roles, responsibilities, agency). A pregnant woman’s life is a con66

Figure 10. Interwoven factors that affect prevention of malaria in pregnancy in rural Lungwena. PROGRAMME IMPLEMEN-

COMMUNITY AND EVERYDAY CONTEXT: Perceived vulnerabilities in pregnancy:

TATION CONTEXT:

ANC & MiP interventions nurses’ compliance with ANC guidelines SP-IPT DOT, but no instructions periodic shortages of SP problems regarding timing of SP-1 Health education vague, no clear reference to prevention of MIP conducted not using the mother tongue PWs

FACTORS INFLUENCING ANC ATTENDANCE: examination to ensure access to services if complications during delivery

Witchcraft – common phenomenon; causes illnesses and delivery complications (miscarriages, maternal death, malowe) Extra-marital relationships – STIs, HIV/AIDS (exposure to social risks) Magnitude of illnesses – malungo only one out of 27 illnesses (exposure to medical risks) Multiple everyday worries - fears of delivery complications, extra-marital relationships and STIs, ill willing relatives, falling ill, inability to walk and perform household duties, bewitchment, economic constraints at the time of delivery (hierarchy of risks) Gender roles, rights & responsibilities

PREGNANT WOMEN (PW)

age experience EM of MiP access to information limited agency

FACTORS INFLUENCING ENROLMENT: confirming pregnancy (movements) cultural issues – revealing pregnancy experience and need distance (+ economic factors)

MEN – HUSBAND, OTHER MALE FAMILY MEMBERS

economic support advice on ANC attendance no direct access to ANC/public health information EM of MiP differs from PW’s model limited agency – financial constraints

FEMALE ELDERS - TRADITIONAL ADVISORS, GRANDMOTHERS advisory role, initiation ceremony approriate behaviour medical and nutritional prohibitions – bitter herbs – miscarriage least knowledge of malungo (MiP EM) no direct access to ANC/public health information

Support and therapy management group (Decision-making dynamics)

Local understanding of illnesses (treatment & prevention practices) Local understanding of ‘risks’ (perceptions and practice)

Structural and economic factors

67

7 RECOMMENDATIONS FOR MALARIA PREVENTION PROGRAMMES Comprehensive gender analysis and HRAP tools will allow identification of all stakeholders and planning of appropriate activities that allow their participation and take account of their capacity gaps.

A large part of my motivation to conduct the present research stems from my personal work experience and recognition that development aid programmes are often planned with inadequate understanding of the socio-cultural context. Thus the pragmatic aim of this research was to gather particularly socio-cultural data that could be used for planning meaningful and context specific programmes aiming to improve ultimately maternal health in Malawi. Based on the results of this study, I make broad recommendations that can be used to improve health programmes, including MiP interventions in Malawi and elsewhere in resource poor settings in the region. These are as follows: 1.

Target groups should be extended beyond pregnant women to include household and extended family members. This study shows that men (husbands and maternal uncles) and female elders (traditional advisors and grandmothers) have an important role in maintaining the well-being of a pregnant woman. They participate in decisionmaking concerning treatment-seeking and preventive practices, and provide advice to the pregnant woman. Groups in the communities are not homogeneous and may differ considerably in programme settings within a country. Therefore, in order to plan meaningful activities reaching all target groups, programmes need to know who in the household and extended family are involved and how, what are their specific roles and responsibilities in maintaining the well-being of a pregnant woman, and what are their capacity gaps to fulfill their duties. Answers to these questions can be gathered by conducting a gender analysis and utilizing the human rights approach in programming (HRAP) tools to plan a tailored context specific malaria prevention programme. 68

2.

Programmes should target economic and structural factors that hinder pregnant women’s access to and utilization of ANC services. Successful implementation of maternal health activities, including MiP, requires appropriately timed ANC attendance. In resource poor settings there are economic factors such as lack of funds for transport, services, medication, everyday well-being such as enough nutritious food, as well as structural factors that manifest themselves in the form of poor services, lack of skilled health care personnel, lack of equipment and medication, long distances to access care etc. These multiple factors hinder pregnant woman’s access and motivation to enroll on and reattend the ANC. Programmes should put much more effort to understanding what kind of economic and structural factors exist in the programme setting, and plan activities to overcome these barriers. The current focus on improving knowledge of malaria should be altered to primarily focus on the economic and structural factors beyond the pregnant woman’s and household members’ agency, and only secondarily on knowledge, which is known to have less impact on treatment-seeking practices (Espino et al. 1997; Nichter 2008: 6-10).

3.

Health education on malaria should be tailored to reach all target groups and delivered by applying participatory and learning-centred pedagogical methods. Improving knowledge of pregnant women and the community is most likely to continue as one of the

3.3.Programmes need to identify different kinds of informal communication channels through which different target groups best receive the information. In Malawi health education on malaria is promoted through media such as the radio and newspapers reaching a minority of the rural population, and mainly men who spend their time in the rural trading centres while the women work in the maize fields and are occupied with household chores. In rural areas one can also see posters at the health centres and at some schools reaching only the few who can read, again missing the majority of rural women, and particularly the female elders. In settings such as Lungwena, songs are a popular means of distributing advice and could be utilised as an effective tool for information sharing in the community.

main activities in prevention of malaria in pregnancy. In its current form it is not that useful, and several changes need to be done in order to increase its impact on treatment and prevention-seeking practices. Specific recommendations to improve health education are as follows: 3.1.Programmes should pay attention to the type of health education messages given. Messages should be clear and simple, giving accurate information in order to avoid misinterpretation that may lead to unintended consequences. Health education should also contain clear instructions on what to do when certain symptoms appear at home. Also pregnant women should be explained that malaria in pregnancy is asymptomatic, explaining why preventive services such as IPT-SP and ITNs are important. 3.2.Programmes first need to define target groups to be involved (see recommendation 1), and to examine what the different target groups know. It is important to provide appropriate information also for husbands, who participate in decision-making and quest for therapy and for female elders, who have a significant advisory role in the community. Programmes need to alter their emphasis on addressing ‘misconceptions and gaps in public health and biomedical knowledge’ of malaria to building on the existing knowledge of target groups and recognize the diversity of knowledge among different target groups. Discrediting local knowledge and concentrating on what people do not know leads to poor understanding of local knowledge and logic behind practices.

3.4.Programmes need to change the current pedagogical teacher-centred approach to a more participatory and learning by doing approach. In the North, the current discussions on the best ways to promote adult learning range from ‘learning by doing’, problembased learning to different kinds of hermeneutic models where the starting point is that the student already has various types of prior knowledge and experience that s/he uses to reflect upon new information, its acceptance and absorption (Reigeluth 1999.) These pedagogical approaches have shown good impact on adult learning and would be worth a try also in Malawi and elsewhere in the region (Hogg et al. 2005). Programmes need to develop the teaching skills of nurses and medical staff, who most probably are 69

not aware that their way of delivering health education messages is not very effective nor always understandable. 3.5. Programmes need to utilize the local vocabulary and language as well as to generate shared meanings and concepts to be used in health education messages in the programme setting. The knowledge and understanding of local illnesses and concepts are diverse and require development of shared meanings before they can be distributed in health educa-tion messages. Participatory theatre provides tools for defining the vocabulary, concepts and shared meanings (Mogensen 1995: 99100).

70

8 DICUSSION 8.1. Challenges concerning utilization of qualitative research results

annual report 2009.) In addition, global initiatives such as RBM have been calling for social-scientific research on socio-cultural factors and household dynamics, among other things. Although the role of social scientific research in malaria control has increased considerably since the 1990s and suggestions to improve interventions have also been made, very little of this evidence seems to be translated into practice (Napolitano and Jones 2006; Williams and Jones 2004). According to Williams and Jones (2004: 515), “It is unclear to what extent this is due to failure to communicate the results of the research, to the lack of importance that is often placed on this type of work, or to the fact that many of the recommendations either conflict with current policies or lie outside the sole control of malaria control programs.” They identify four factors contributing to the limited utilization of social scientific research that are as follows: 1) the lack of trained social scientific researchers in malaria control, 2) the lack of awareness among Ministries of Health and NMCPs of the variety of disciplines within social science, 3) the lack of awareness among social scientists regarding the problems malaria prevention programmes are faced with, and difficulties to interpret and use social scientific information and 4) the socalled ‘magic bullet’ type expectations on social scientific research to fix all problems (Williams and Jones 2004). While I agree on the factors Williams and Jones suggest as reasons for failure to apply social scientific findings in programme planning, I also suggest that we examine in more detail the bureaucratic aspects of multilateral international agencies36 which I consider central in discus-

In November 2004, health ministers around the world came together to discuss health research in Mexico City. A key message from this Ministerial Summit on Health Research was that there is a need to strengthen the interaction between researchers, policymakers and the potential users of the research. There are on-going debates on how best to bridge the gap between what is known and what gets done in practice, i.e., the ‘know-do’ gap (Pablos-Mendez et al. 2005.) More than 20 years ago Ramalingaswami (1986) pointed out that we have all the knowledge and means to improve health, but we are lacking a shared understanding and tools for achieving this. Illhealth and premature deaths in resource poor settings such as Lungwena will continue to persist despite the large amount of social scientific research findings available and the existence of various kinds of cost-effective interventions such as IPT and ITNs that are effective in preventing malaria. Unless we are able to understand the complex and interlinked factors that hinder utlization of social scientific research evidence we have little chances to find ways to bridge the gap between research, practice and policy. The importance of social scientific research in communicable disease control has been recognized for 30 years and a Special Programme for Research & Training in Tropical Diseases (TDR) was established by UNICEF, UNDP, the World Bank and WHO in 1979. The broad focus of this programme has been to support social, economic and behavioural research that identifies constraints in, and opportunities for, control and prevention of such diseases as malaria in resource poor settings by elucidating social, cultural, economic, health system and policy related factors. The practical aim has been to propose strategies and solutions to remove the barriers identified through research. (TDR

36

There are many types of international health organizations (see an example of classification by Foster 2009: 685). From now on, international health organization will only refer to multilateral organizations such as the different United Nations agencies (WHO, UNICEF; UNDP etc.), unless I make a specific remark otherwise.

71

programme priority areas in international health.

sion about hindering factors in translation of social scientific research results into practice.

Externalized problem - Communities perceived as a problem

8.1.1. Understanding bureaucratic aspects of international health organizations

The first aspect is the commonly held perception and attitude that communities and traditional practices are the main problem. In other words, programmes are planned to change community practices. This perception and attitude is strongly influences by the early days of technical aid and dominance of Western science, when many Western countries, including Finland, sent technical assistance, namely experts in agriculture, health, forestry, animal production and education to developing countries because it was believed that Western know-how was the key to improving the situation in developing third world countries (Foster 2009: 686-687). This perception and the attitude that ‘we know better than the less educated communities’ is still alive despite the buzz words or mobilizing metaphors generated by policy discourse (Mosse 2004: 663)37 such as ‘community participation and dialogue’, ‘partnership´ and ‘local ownership’. When communities are perceived as a problem, there is no need to understand and examine what the communities actually know and what their practices are (Nichter 2008: 7). The social scientific literature review shows that the majority of studies have mainly examined the gaps in ‘public health’ knowledge, with a few notable exceptions (e.g. Hausmann-Muela et al. 2002; Helitzer-Allen et al. 1994; Kamat 2008). As Foster (2009: 688) points out, there is little if any reflection on the programme planning and implementation practices of the organization itself. Accepting and recognizing that there is something wrong in the way programme planning and policies are made would mean that one should make changes in the organi-

International health organizations are often criticized for their “one solution fits all” approach, perceiving culture as a barrier, failure to achieve initiatives such as ‘Health for All by 2000’, shifting priorities and reinventing the wheel etc. (see Castro and Singer 2004; Whiteford and Manderson 2002). While there is some truth in the criticism, I find it important that we understand the everyday reality and constraints of international health organizations in the dynamic and multi-layered context in which international development is implemented and negotiated (Crewe and Young 2002:5), and more importantly, how development aid and multilateral international organizations work, rather than just passing judgment on its failure, as Mosse (2004) points out. Criticism will not provide a conducive environment or tools for change towards better programmes, strategies and informed policies. So far, there have not been many descriptions of bureaucratic aspects of international health organizations (Foster 1987; 2009; see also Ramalingaswami 1986), although a number of medical anthropologists have explored these issues in their studies (Justice 1999; Nichter 2008, Parker 2002; Pfeiffer 2004; Yoder 1997). Working three years for UNICEF Malawi provided me with experience of what it is to work as an anthropologist in international health and helped me gain understanding of some of the factors that make it challenging to plan meaningful and culture sensitive programmes based on social scientific evidence. Drawing upon my subjective experience and Foster’s (1987; 2009) comprehensive analysis of bureaucratic aspects of international organizations, I will next discuss some of the aspect that in my opinion directly or indirectly affect the applicability and utilization of social scientific research in malaria control and in any one of the

37

According to Mosse (2004: 663), these vague and conceptually imprecise mobilizing metaphors are needed to conceal different interests, to allow compromise, to build consensus and coalitions, and to distribute agency among different development aid actors at global, national and local level.

72

tions in resource-poor countries such as Malawi38. Many staff members do not have a Ph.D. nor experience of research. In particular, there is a lack of social scientific research capacity both among international and national staff members as well as among local counterparts, as noted by several studies (Theobald and Nhlema-Simwaka 2008¸ Williams and Jones 2004). I was the only staff member in UNICEF Malawi country office who had training in anthropology. There where others who were more familiar with survey methodology and those with no knowledge. KAP surveys were very popular methods to collect data for programme planning because they were easy to conduct, relatively cost-effective and fast, producing ‘hard’ numbers and ‘generalisable’ results (IV).39

zation itself and its practices. We have ample evidence of how difficult it is to change practices from the many ‘failed community-based projects’. Organizational changes are likely to be even more difficult because of the power relations and hierarchies in an international organization. According to Crewe and Young (2002: v), “staff within bureaucracies tend to resist fundamental challenges to the status quo.” As long as the attitude and perception of communities as a problem persists, it is most likely that research agendas focus more on assessing progress and superficial knowledge, attitudes and practices than on exploring local practices, gender aspects and household dynamics through ethnographic approach. Lack of capacity and experience in research (IV)

The office had a research committee that assessed the tenders of local Malawian researchers and decided which institution was commissioned to carry out research. The problem with this in-house committee was that it consisted of staff members who did not have the needed background and experience to assess the substance of the research plans; they mainly commented on the budgets and timelines, and often chose the cheapest tender. This naturally affected the type of research conducted; surveys were most cost-effective

The second aspect is related to the skills and capacity available in the international organizations. Organizations are made up of people with different backgrounds, education, gender, age, working skills and interests that affect the way of performing daily tasks and achieving the programme targets. Many international staff members are ‘generalists’ who have learned their knowledge of diversity of programme priorities ranging from child protection to nutrition, early childhood development, safe motherhood, primary education, water and sanitation etc., through their work in international development. Many have no educational background in these fields and are therefore dependent on the strategies and guidelines issued by the headquaters (HQ) and regional officers, and advice from experts and consultants in these fields. Staff members working in the health sector are often an exception as many of them have a background in public health and epidemiology. Yet, in reality the expertise of individual staff members in a health organization is not very deep nor are they able to follow academic debates and have access to latest publications, often due to the everyday challenges of getting the work done and the difficulties accessing social scientific and anthropological publica-

38

Access to information has improved considerably through internet and open access publications since the time I worked in Malawi. However, a large bulk of social scientific publications remain out of reach of international and national staff members and government counterparts because they are published either as monographs or specific journals mainly targeting a small group of academic people. Trostle (2002: 291) points out that “we [anthropologists]are distanced from the local even in a publication championing the local.” 39 The strong belief that a KAP survey is an appropriate and adequate method to collect data among littleeducated people in resource poor settings triggered me to try out the method; to gain knowledge and first-hand experience about its strengths and weaknesses. Having also experience about resistance to accepting ethnographic research findings, I thought that triangulation of qualitative and quantitative methods might be useful and increase the credibility of ethnographic findings among programme planners and policy-makers (IV).

73

context. Therefore, many stay in the ‘global context’ and become distanced from the local reality. In the long run it is easier to be disconnected than connected. Staff turnover is often high, considerably affecting the institutional memory of the organization, hampering learning from past experience. Foster (2009: 692) attributes the failure to learn from past experience to either be an inherently bureaucratic structure or to staff memebers’ reluctance to give credit to others. Whatever the reason, national staff members and counterparts have to deal with constant arrivals and departures of expatriates, complicating continuity of work. I left Malawi just after the ethnographic malaria study was finalized and before the annual planning meetings took place. My fear is that when I left Malawi the results were left unused without incorporation into programme planning because in the office and among the local counterparts there was no capacity to translate ethnographic results into practice. The way the results were written in the report did not give any tools for project officers to translate them into practice. Recommendations such as “planning Information Education and Communication campaigns utilizing local terminology” are of no use unless we also understand the correct communication channels, gender and household dynamics and are able design culture sensitive IEC materials. I also speculate that there might have been another issue that affected the lack of interest to utilize the results, namely the reluctance to believe that the results were true. Before my departure from the country, we organized a meeting to disseminate the findings to project officers in NMCP and in the MoHP as well as to the district malaria officers. A few of the medical officers from the MoHP reacted negatively to some of the results concerning local community practices, such as taking a convulsing child to a traditional healer, arguing that the results were not true. We learned a valuable lesson: one has to think through how to communicate results meaningfully and constructively to professionals with a biomedical and public health background. According to Yoder (1996: 139), medical doctors and nurses

and the least time consuming. During the three years, I managed to pull through one medical anthropology research on malaria with the support of my supervisor, a broadminded medical doctor. A considerable amount of effort was needed to convince both the research committee and the UNICEF representative on the meaningfulness of a medical anthropology study because they were not familiar with the anthropological research tradition and its possible contributions (see Napolitano and Jones 2006) and also because there was a clear sense of research fatigue in the office. In my opinion, too many studies had been conducted and too little of the results had been translated into practice due to lack of research capacity. I agree with Lairumbi and colleagues (2008) that to convince professionals who have experience only with methodologies yielding quantitative results or have no research experience, it is very important to communicate the relevance of anthropological and qualitative research in a justifiable manner, to open up the process and present results understandably. Theobald and Nhlema-Simwaka (2008) also argue that the importance of skills and rigour in qualitative research practice needs to be stressed and the analysis process should be demystified. Otherwise, there is a danger that results from anthropological and social scientific studies will just be shared as amusing anecdotes of local beliefs during cocktail parties with no effect on programme planning, as an international malaria epidemiologist once told me. According to his experience, anthropological research on malaria had proven to be useless because it is rarely oriented to asking “how do we change our program”; it just describes the issue. Staff rotation and turnover The third aspect that disrupts work and slows down the progress is rotation of permanent international staff members in every two to four years from one country to another. It is hard to be motivated and excited after each rotation, and to make an effort to get to understand the national and even more so the local 74

the community-based projects. Now, however, less attention was paid to the actual use of nets and IEC messages distributed by nurses during their sales speeches (II). My work performance was assessed according to my success in spending the funds by the end of the year and showing progress in terms of the number of distributed ITNs. Sustainability and culture-sensitiveness of the projects were not priorities, nor did I have time to occupy myself with these concerns. I also spent a lot of my time trying to track down my counterparts and receipts of funded project activities as accountability for funds was taken very seriously by UNICEF. Giantism takes attention away from importance of culture and context, and consequently from the need to conduct proper ethnographic research.

trained in biomedicine show the greatest resistance to recognition of local knowledge, yet they are the professionals implementing the health care programmes. Justice’s (1999: 336) experience from Nepal shows that ‘cultural evidence’ is perceived as unwelcome because its full consideration may threaten the existence of a whole programme and therefore it has to be rejected. Giantism - no room for small context specific programmes The fourth aspect concerns funding, budgets, pressure to spend funds and to secure more funds. Crewe and Young (2002: 9) use the term giantism to describe the pressure to attract more funding and bigger budgets as they are equivalent to the greater status of organizations. Giantism affects programme designs and the need to scale up programmes nationwide. Tailoring context specific programmes that tackle specific problems are often unattractive because they do not absorb enough funds although their importance may be recognized. Small projects are easily left for local NGOs that are perceived to deal with the so-called grassroot problems. Community participation has been one of the core values of UNICEF, and in malaria prevention community participation was promoted through community-based ITN projects ran by village health committees. My task was to supervise and support NMCP and district malaria officers to implement these ITNs projects, first in three districts and later in five districts (out of 27 districts). The progress was slow, and many problems were detected relating to the capacity of the village health committees to manage these projects. At the same time the RBM Initiative put enormous pressure to scale up the ITN projects. Frequent messages from the UNICEF HQ and the regional office in Nairobi called for substantial progress in order to meet the Abudja targets. New partnerships were formed and strategies adopted. UNICEF began to support PSI by procuring ITNs while PSI distributed the nets through the ANC platform. Progress in terms of distributed nets was impressive compared with

Long-term programme planning and topdown priorities The last aspect is related to the inflexible long-term planning processes that hamper utilization of social scientific research evidence after the programme inception. The UNICEF country programme followed a fiveyear programme cycle that was evaluated in the middle of the programme. The mid-term evaluation was carried out mainly using survey methods and key informant interviews. No significant evidence emerged from the evaluation that would have required any major change in the remaining 2.5-year programme cycle. Each year there were annual planning meetings of each section in collaboration with local stakeholders. During the three years, I participated annually in formulating the project’s action plan for the education and health sector as well as in the overall programme plan of operation. Planning was mainly done as deskwork based on our own expertise, and not on scientific evidence (referring here particularly to qualitative research to voice the needs and concerns of communities). An exception was a one-week long workshop on planning a five-year malaria programme based on HRAP in collaboration with central and district level stakeholders. During the planning process, we also 75

having experience of participating in developing a national malaria policy in Malawi, I am interested in discussing EBPM in relation to my own subjective experience. I find it important to first discuss who decides when, what type of evidence is used to what purpose, and secondly to discuss perceptions of what counts for ‘evidence’ in policy and programme planning.

conducted key informant interviews and focus group discussions among community members in Kasungu District, which was one of the UNICEF supported districts. Although engaging community members in the planning process was a step forward, it was obviously very superficial. It is also a fact that programme priorities, policies and strategies are defined at the HQ level, which means that country offices have little say over the priorities. According to Justice (1999: 336), the prevailing health policies that “inundate local realities as they sweep downward from policy-making circles to planners” are part of the bureaucratic culture of international health organizations. Policymakers and programme planners at HQ level are disconnected from the local reality.

Who decides and what type of evidence The demand for ‘evidence’ became central in biomedical health research in order to inform daily medical practice in the late 1980s. The work of Archie Cochrane, a Scottish epidemiologist and other epidemiologists that followed his work developing the first tools for ‘best evidence based practice’ in the early 1990s form the roots of EBM, referring to “clinical practice that draws on the ‘best evidence’ available to inform treatment decisions” (Lambert 2009: 16.) In 1993, a global network of centres and individuals contributing to improving decision-making by applying systematic reviews concerning the effects of healthcare interventions called the Cochrane Collaboration, was established (Cochrane Collaboration 2010.) In the mid1990s, following the rapid arise of EBM, pressure and demand for EBPM also awakened and policymakers, government officials and programme planners were urged to apply scientific facts in developing policies instead of the common practice of developing policies based on political ideologies. (Behague et al. 2009.) In the medical field, evidence is closely linked to improving the effectiveness of biomedical interventions. This type of evidence, ‘evidence of effectiveness’ (Lambert 2009: 16), is often gathered applying randomized controlled trials that allow collection of strict, standardized and quantifiable data (Bahague et al. 2009).40

8.1.2 Evidence in programme planning and policy-making Bureaucratic aspects of international organizations alone fail to explain why so much of the social scientific research results remain unused in malaria prevention programme planning and policy-making. My opinion is that interlinked to this discussion is the current popular trend particularly among medicine and policy-making that everything has to be ‘evidence-based’. This enthusiasm has even been described as the evidence-based movement of the elites in academic medicine (Mykhalovskiy and Weir 2004; Pope 2003). While professionals with a medical background have critically discussed different aspects of evidence-based medicine (EBM), social scientists have been slow to respond to the fast rise of EBM and evidence-based policy-making (EBPM; Mykhalovskiy and Weir 2004). In medical anthropology discussions on notions of evidence (Hastrup 2004; Lambert 2006, Lambert 2009) and ‘evidencebased medical anthropology’ (Ecks 2008) have only recently started. It is not my intention to engage in the critical debates on EBM and EBPM because I am fully aware that I am not knowledgeable enough about them and they do not fall under my expertise. However,

40

Among the medical people there are also debates about what counts for evidence, and criticism towards randomized controlled trials. Hence the discussion here may seem too simplistic, unintentionally drawing a dichotomy between quantitative and qualitative evidence.

76

tions for social scientific research as its value on programme planning has already been recognized.

Evaluation criteria for evidence are often based on the authority of the methods, raising a question of how other forms of evidence gathered by e.g. qualitative and anthropological methodology can be integrated within evidence based programme planning and policymaking (Lambert 2009). Behague and collagues (2009) also argue that the transposition of EBM to non-clinical domains has caused a number of problems particularly in global policy-making where the primary imperative for policy-making is based on the need to provide generalized research results that have been collected utilizing study designs and methods replicable from one country to another. The demand for evidence-based malaria control and prevention is seen in the large number of randomized control trials on drug efficacy and epidemiological surveys on IPTSP and ITNs carried out through the subSaharan Africa. Due to the increasing evidence on resistance of P. falciparum to SP, the malaria treatment policy has now been changed or is about to be changed in many of region’s countries. At the end of 2007, Malawi launched a new drug policy promoting artemisin combination therapy (ACT) as the new first-line treatment following the WHO recommendations (Malenga et al. 2009). Evidence used to change the national malaria treatment policy was primary based on that derived from epidemiological and randomized control studies as was also done in 1993 when malaria drug policy was changed from CQ to SP (Malenga et al. 2009; Williams et al. 2004). Anthropological studies on socio-cultural factors affecting drug use practices have been largely missing as evidence for policy-making or left unused (Kamat 2009). Helitzer-Allen (1989) conducted medical anthropology research in Malawi in the late 1980s on perceptions and use of antimalarial drugs in pregnancy, yet these results were largely left unused. I suspect, based on my own experience, that the demand by international organizations such as WHO for certain types of ‘evidence’ affects national level perceptions and practices on ‘appropriate evidence’ for policy-making. In the long-run this might have positive implica-

Evidence-based policy-making suggests that policies are formulated on the basis of evidence, but policy formulation can also be a way to sustain the global initiatives and priorities of international organizations. Mosse (2004: 663) has suggested that in development aid; “policy is an end rather than a cause; result, often a fragile one, of social processes.” In other words, development policies are not driven by policy, but by the need to maintain relationships between a wide range of stakeholders with diverse and sometimes even conflicting interests. Mosse (2004: 646) argues that policy ideas are produced socially, and a policy serves as a unifying element providing the needed support to implement different kinds of programmes and strategies. Policies also require constant translation into practice and it is the ‘skilled brokers’ (referring to development aid workers of all kinds) that make sense out of the policy translating it into the institutional languages of its stakeholder supporters (Mosse 2004:647). I agree with Mosse that policies are socially developed, but I also think that they are top-down driven, particularly by multilateral international organizations, from global to national levels. Since the establishment of RBM partnership and signing of the Abudja Declaration, there were discussions on a need to have a national malaria policy among the various UN agencies and CDC in Malawi. The need for a policy was also discussed with NMCP, MoHP and other RBM partners to get a sense of national level interest and support for a policy. In 2000, a new head of health sector arrived in UNICEF Malawi. He was very keen to see malaria prevention going to scale. In 2001 UNICEF decided to take a lead and to provide funds for the development of a national malaria policy. As I was the responsible officer for malaria prevention in the office, national malaria policy development became my primary task for the year 2001. My task was to 77

ANC without any instruction and appropriate health education seems to have lessened pregnant women’s perceptions of dangerousness of malaria (see also Castro and Singer 2004).

provide all the necessary support for NMCP and MoHP in the policy process. As Mosse (2004) says, there are diverse, sometimes even contradicting interests that have to be negotiated and therefore a consultant, who was of African origin but outside of Malawi specialized in policy development and knowledgeable on malaria prevention issues, was commissioned to carry out the difficult task. During several months he interviewed key persons in government and in the international donor community. There were also a series of consultative stakeholder meetings. I left the country before the policy was finalized in early 2002, but my experience from this process was that the need for policy was recognized at the national level, but the pressure for a policy came from the global level, prompted by the RBM Initiative. It was a social process based on prior evidence concerning ITNs and IPT. The national policy followed the recommendations of the RBM initiative and to my knowledge, no new evidence was gathered at this point. Social scientific researchers were not part of the consultative process most likely because there simply was no recognized social scientific group advocating their perspective, nor the knowledge and recognition of perspectives of local people; the ones who were expected to comply with the policy. The majority of stakeholders had a background in biomedicine, public health, epidemiology, environmental health etc. I, who was an anthropologist, was far too inexperienced to understand the whole process of policy development to think of a way to promote and incorporate the perspective of the local beneficiers to the policy. Nor would I have been able to have much impact alone. Formulation of a topdown driven policy is sometimes the only solution to reach consensus among stakeholders at national level and to gain commitment to reduce the burden of a disease such as malaria. Yet a policy is no guarantee of success. Implementation of the policy at the ground and local levels is a real challenge, and sometimes a well-intentioned policy may have unintended consequences as this present study shows: administration of IPT-SP at the

Perceptions of social scientific research as ‘soft’ and ‘lacking generalisability’ (IV) Gilson and McIntyre (2008) point out the important fact that the perceived credibility of researchers shape the extent to which policymakers are willing to learn from research findings. Many of us have observed that among professionals with a medical background, social scientific research is often perceived as soft, descriptive, merely anecdotal, lacking generalisability and, in other words, not credible (Ecks 2008; Napolitano and Jones 2006; Theobald and Nhlema-Simwaka 2008). This means that when the credibility of disciplines such as anthropology that primarily relies on a wide range of qualitative research methods and long fieldwork periods is questioned, we remain rather powerless in terms of evidence-based policy-making, and results from our studies are easily marginalized and poorly utilized. The literature review showed that although the amount of social scientific studies is slowly increasing to respond to the call for studies of socio-cultural aspects of malaria and MiP, there are only a few studies that go beyond simplistic KAP studies (Ribera et al. 2007; Nichter 2008: 6-7). In most cases qualitative methods are used superficially to describe local and cultural beliefs without deeper analysis of the cultural logic and understanding of the socio-cultural context. Many times the only qualitative method used is a FGD combined with a survey. Studies have encompassed such issues as cultural beliefs affecting motivation and delay to enroll ANC timely, beliefs concerning SP and factors affecting achievement of IPT-SP RBM targets. Yet rarely have these studies explored explanations behind perceptions and ‘cultural beliefs’, elaborated on the meanings and tried to understand why these beliefs exist and how 78

translate anthropological research into a language understood by non-anthropologists (Theobald and Nhlema-Simwaka 2008).

they are created. Investigation of gender and malaria in the context of perceived vulnerabilities is largely missing (see Literature review 2.3.). One important reason for the failure to move beyond KAP studies is that professionals having a medical background do not really understand why a KAP survey is perceived by anthropologists (e.g. HausmannMuela et al. 2003; Pelto and Pelto 1997) as a rather poor tool to gather evidence on sociocultural issues and practices, and what possible advantage an anthropological study could provide compared with a KAP survey, as found by Napolitano and Jones (2006). For professionals having a medical background, a KAP survey is already a ‘qualitative’ method whose results can be nicely analysed by computer software such as SPSS and presented in a compact ‘quantifiable’ form. KAP surveys are advocated through journals with high impact factors such as the Lancet by wellintentioned professionals with a medical background who have recognized the importance of socio-cultural factors. For example, a recent paper by Crawley and colleagues (2007) discussed the factors that influence successful translation of global MiP strategies into national policy and programme implementation. Among the major problems identified was limited evidence base to support policy change. They also called for systematic examination of the cultural and operational constraints to the delivery and uptake of IPTSP, recommending KAP studies to gain understanding of the constraints regarding implementation of the IPT policy. Cultural beliefs and practices were seen as potential factors negatively affecting programme implementation that could be overcome by developing innovative communication strategies for improving coverage. This is a powerful message advocated by researchers that are perceived credible by policymakers and programme planners in international health organisations and NMCPs. Medical anthropologists either need to accept that KAP surveys will be used despite their many weaknesses, including poor cross-cultural validity (IV, Stone and Cambell 1984; Manderson and Aaby 1992), or they will have to find ways to

8.2. Applied medical anthropology and the way forward There is no doubt that applied and critical medical anthropology can have a significant contribution, providing meaningful insights into the understanding of public health problems such as malaria (e.g.Winch 1999, Kamat 2009), acute respiratory infections (e.g. Nichter 1993), childhood pneumonia (e.g. Mull 1999) and the like,41 as well as providing important critical observations of public health policy issues and their implementation, among other things (e.g. Janes 2009, see the edited collection of Castro and Singer 2004). Through my research I have been able to show what added value applied medical anthropology research brings to understanding of a public health problem such as malaria in pregnancy. I provide insights into the complex and multifaceted gender dynamics at household level affecting pregnant women’s agency to make decisions to seek care and prevention for malaria in pregnancy (Chapter 5.1.). I demonstrate the diversity and syncretism of local perceptions and knowledge of malaria in pregnancy affecting the design of health education messages (Chapter 5.2). In the context of pregnancy related vulnerabilities, I illustrate important differences between the local risk perceptions and the public health at-risk approach that may affect acceptance and rejection of preventive methods such as IPTSP and ITNs (Chapter 5.3.). Hence the core question I pose here is not what applied medical anthropology can contribute. Instead I ask what happens if we fail to become actively engaged in developing theoretically and practically orientated disciplinary responses to the growing international 41

There is a large amount of excellent contributions to medical anthropology in public health. Those interested in finding out more should read the following books: Whiteford and Manderson (2002), Hahn (1999), Hahn and Inhorn (2009), Nichter (2008).

79

medical professionals for qualitative methods is that they apply the methods without the important theoretical and contextual understanding needed for interpretation of the results. At worst, results from these studies are turned into findings such as ‘blaming the caretakers for neglecting their child’s wellbeing because of certain cultural beliefs’. As Lambert and McKevitt (2002: 210) point out, “Qualitative research is in danger to be reduced into a limited set of methods that require little theoretical expertise, no discipline based qualifications, and little training.”

pressure to provide social scientific knowledge, including anthropological knowledge, on shedding light on the complex and interlinked socio-cultural, economic, political and structural factors that have an adverse effect on the well-being of women, men and children in resource poor settings. Over the last 20 years there has been a growing emphasis on interdisciplinary42 research due to acknowledgement of the complex nature of health problems. Today, many research funding organizations at the global level such as UNESCO, WHO TDR, European Union (7th Framework Research Programme), and at the national level, e.g. the Academy of Finland (see Bruun et al. 2005), consider interdisciplinary research a prerequisite for funding of research programmes on international health issues. Interdisciplinary research has become a popular trend and a politically correct buzz word. Professionals having a medical background have been quick to respond to this call. More and more professionals who do not have training in anthropological methods and theory have started to conduct ‘qualitative research’ applying FGDs and other rapid appraisal techniques (Lambert and McKevitt 2002; Manderson and Aaby 1992; Manderson 1998). FGDs have been particularly advocated and appreciated by non-anthropological researchers because they are rapid techniques guided by relatively strict rules and procedures, yet yielding qualitative descriptive data about cultural beliefs and knowledge (Khan and Manderson 1992). Drawing on my experience in Malawi I also argue that the ease of using FGDs is further increased by the fact that local research assistants can be trained to moderate FGDs based on topic guides, whereas IDIs and ethnographic interviews require skills and training in the methodology and theory acquired through extensive training. What is deeply worrisome with this increasing enthusiasm of

Since the late 1970s there has been an on-going discussion about the role and contribution of medical anthropology in public health and control of infectious diseases (Bibeau 1997; Foster 1982; Hemmings 2005; Manderson 1998; Rylko-Bauer et al. 2006; Yoder 1997; see also Hahn and Inhorn 2009; Inhorn and Brown 1997; Nichter 2008). There have been concrete efforts to provide solutions to improve collaboration and interdisciplinary between anthropologists and medical professionals. Trostle (2005) discussed the potential collaboration of anthropology and epidemiology, and most recently Ecks (2008) made three propositions for an ‘evidence-based medical anthropology’. Some anthropologists have been making attempts to develop rapid methodologies to respond to the needs of public health communities. For example, a malaria manual was developed in 1995 (Agyepong et al. 1995). There have also been critical voices among medical anthropologists. Good (1994) criticized the engagement of medical anthropology in public health due to asymmetrical power relationships between medical knowledge and local folk beliefs that have led to the development of health belief models aiming at replacing the false local beliefs with accurate public health knowledge’.43 Browner (1999) has warned against the growing tendency for anthropolo-

42

43

Interdisciplinary research refers to an approach that aims to building a holistic and shared view of a complex problem integrating separate disciplinary data, methodologies, conceptual and theoretical frameworks (Bruun et al. 2005: 28).

Drawing on my experience in Malawi, I agree with Good (1994) and Manderson (1998) that power and authority are enjoyed by organizations responsible for policies, resources and programs privileging their knowledge over their target groups.

80

otherwise medical anthropology is in danger of being marginalized in the field of international health. An imperative fact is also that the reduction of the burden of malaria in pregnancy and any other public health problem is beyond the scope of any single discipline. The success of current control measures used to reduce malaria infection and mortality is dependent on understanding malaria in the context of perceived vulnerabilities affected by cultural, social, political and economic factors (Hahn and Inhorn 2009; Lambert and McKevitt 2002; Manderson 1998; Nichter 2008.) Interaction with other disciplines should be seen as an opportunity to bring new insights and theoretical developments into the field of applied medical anthropology.

gists to alter their research plans to meet the needs of biomedical professionals, referring to ‘medicalization of medical anthropology’. According to Van der Geest (1995), applied anthropologists engaged in public health enjoy no prestige among colleagues based in academic institutions. Napolitano and Jones (2006) describe how some of the anthropologists described their research work in international health as not real anthropology. Having personal experience of conducting applied medical anthropology research I can recognize the tension and criticism towards applied research and self-undervaluation of such research. My concern, however, lies more on the fact that anthropological research methods can be used by anyone, even by those with no training. Paul Farmer said in his keynote address at the Medical Anthropology at the Intersections Conference at Yale University in 200944 that if medical anthropologists do not engage in studying public health problems utilising the theoretical and methodological contributions of medical anthropology and strive to translate their results into practical solutions, other health related disciplines will. This may very well lead to use of ‘qualitative methods’ of ethnographic fieldwork without contextual understanding, resulting in false interpretation of the results that at worst may do more harm than good (see also Manderson 1998). If we fail to take action we will end up watching other disciplines overtake the research work that we as anthropologists are so much better capable of doing than our counterparts from other non-qualitative disciplines. The pressure for interdisciplinary research leaves no other choice than to collaborate and seek partnerships outside of our discipline,

It is noteworthy to realise that matters of international relations are closely tied and interlinked with global health. Decisions on development policies in Finland and in the European Union affect funding and priorities in global health in the South. According to Pfeiffer and Nichter (2008:412), the only way for effective health action is to get political. We need to start a loud campaign to advocate the added value of applied medical anthropology research and demand equal interdisciplinary partnership. Interdisciplinary partnership that can be considered equal would mean that anthropologists would be heard already at the research planning stage as experts and team leaders, and not just as data collectors or ‘CVs’ in the research proposals to attract funding. In addition, adequate funding should be budgeted and allocated to carrying out proper applied medical anthropological research. There is a need to look for innovative ways to carry out fieldwork; to apply multiple methods and in several time phases in order to be able to better respond to today’s time pressure to develop rapid responses and solutions yet without compromising the research outcome. University faculties providing training in anthropology also need to generate understanding of fieldwork and multimethod research that goes beyond the traditional ethnographic fieldwork. It is

44

The conference took place 24-27 September, 2009, and its aim was to celebrate past achievements in medical anthropology and to foster an exchange of ideas that will inspire path-breaking work in the next 50 years. It brought together more than a thousand medical anthropologists, including faculty, students, and practitioners from all over the world, approximately half of the participants coming outside of United States. All the keynote addresses are available at http://www.yale.edu/macmillan/smaconference/speaker s.html (accessed 13 March, 2010).

81

also important to make funding agencies at the national and global levels aware of the dangers of unskilled use of qualitative methods and the importance of rigour and transparent research process. This means that anthropologists need to engage in critical discussions and publish more in non-anthropological journals in order to reach decisionmakers and programme planners in international health. I believe that the criticism expressed by Good (1994) and van der Geest (1995) is more about fear than reality today when we collaborate with medical professionals or work in international organizations. We, however, need courage to contribute and when needed, challenge our medical and international health counterparts in a language that they understand and make partnerships with those who are in a position to make decisions regarding strategies, programmes and policies. Lastly, it is important to recognise and keep in mind that all of us working in the field of international health have a common goal: to improve the health of children, women and men of different ages in resource poor settings, and we can only achieve this goal if we work in partnership because of the complexity of the problems in question.

82

9 RECOMMENDATIONS FOR FUTURE RESEARCH IPT-SP to prevent a miscarriage will miss the point. In addition, it would be worthwhile to explore women’s understanding of anaemia and perceptions regarding low birth weight babies, as these two are the main adverse consequences of malaria in pregnancy and yet, most of the women did not associate them with malaria in pregnancy.

The findings from this research, presented in this thesis, call for novel approaches to studying malaria in pregnancy in resource poor settings. First of all, this research clearly demonstrates that malaria in pregnancy is not just about pregnant women suffering from malaria, but more about pregnancy itself and pregnant women’s effort to have a positive pregnancy outcome. Therefore, it would be important to explore malaria in pregnancy in the context of pregnancy and everyday life, at the expense of the traditional focus on disease. Focusing on pregnancy and management of wellbeing of pregnant women during pregnancy will allow a better and more holistic understanding of the multiple factors that affect treatment-seeking practices and utilisation of preventive services promoted through the ANC platform. The multiple factors include identification of those who are involved in quest for therapy and in what kind of capacity, how decisions are negotiated within a household and the extended family, and recognition of factors that are beyond the agency of the pregnant women and other family members involved.

Thirdly, this study shows that a major effort in malaria prevention is to improve people’s knowledge, considered replete with gaps in biomedical knowledge and cultural misconceptions, believing that this will lead to an expected behaviour change. The culture of blame has been a common justification for health education at the expense of acknowledging economic and structural factors that play an important role in access to services. Future research should explore pregnant women’s economic situation within a household; decision-making, controlling, mobilizing and earning funds to be used for accessing care and preventive services. Similarly, studies should explore men’s, namely the husband’s role and responsibilities to provide funds for his pregnant wife’s care and his capacity to earn and mobilize funds when needed. Little is also known about the roles in and responsibilities of other relatives for mobilizing assets. Structural factors such as long distances and poor quality services require attention, and how these these factors affect women’s motivation to attend formal services should also be examined. More evidence is needed on the interlinkages of economic and structural factors and household costs of attending ANC services and accessing treatment for malaria in pregnancy.

Secondly, it would also be very useful to investigate in more detail maternal health concepts, their local meanings and impact on treatment-seeking practices and utilization of preventive services. For example, the concept of miscarriage and management of mild symptoms in the early stage of pregnancy (first trimester) require attention. This study found that there is a concept called kupumula meaning that the woman’s body is resting for 1-3 months after which menstruation starts again, and another concept for miscarriage which is more similar to the biomedically defined miscarriage. It would be worth exploring what kinds of causes and symptoms are associated with kupumula and miscarriage, and how they are treated and prevented. When the conceptual understanding of miscarriage differs from the biomedical definition, health education by nurses on taking

Fourthly, studies should examine the diversity of knowledge, perceptions and practices among women and men of different ages regarding maternal health issues and ANC services in order for programmes to better target project activities and health education. Gender focused research should also identify the most 83

vulnerable groups among pregnant women and particularly explore the primigravidae, who have no experience and have least access to information. Also the advisory role of female elders and decision-making power within a household and extended family require further investigation. Lastly, so far attention has been paid to exploring community knowledge, attitudes and practices in order to improve compliance with treatment regimes and preventive services ignoring the knowledge, attitudes and practices of programme planners and health care professionals working in malaria prevention. It would be very important to investigate national level project planning and decisionmaking in order to identify the critical barriers that hinder tailoring needs-based contextspecific programmes that reach the appropriate target groups through suitable project activities. Too little is known about the attitudes of professionals regarding local people and their practices, and how these attitudes affect programme planning. Shifting the focus from the community context to the global and national level context could eventually contribute to organizational changes and lead to better programmes and achievement of development targets.

84

10 REFERENCES Lessons from the past 100 years in Africa. American Journal of Tropical Medicine and Hygiene 71 (Suppl 2): 268–278.

Adam I, Omer E, Salih A, Khamis A and Malik E (2008): Perceptions of the causes of malaria and of its complications, treatment and prevention among midwives and pregnant women of Eastern Sudan. Journal of Public Health 16 (2): 129-132.

Allen DR (2002): Managing motherhood, managing risk. Fertility and danger in West Central Africa. Ann Arbor: University of Michigan Press.

Agyepong I (1992): Malaria: ethnomedical perceptions and practice in an Adangbe farming community and implications for control. Social Science and Medicine 5: 131–137.

Anders K, Marchant T, Chambo P, Mapunda P and Reyburn H (2008): Timing of intermittent preventive treatment for malaria during pregnancy and the implications of current policy on early uptake in northeast Tanzania. Malaria Journal 7(79). doi:10.1186/1475-2875-7-79

Agyepong I and Manderson L (1994): The diagnosis and management of fever at household level in the Greater Accra Region, Ghana. Acta Tropica 58: 317–330.

Asenso-Okyere WK and Dzator JA (1997): Household cost of seeking malaria care: A retrospective study of two districts in Ghana. Social Science and Medicine 5: 659–667.

Agyepong I, Arynee B, Dzikunu H and Manderson L (1995): The malaria manual: Guidelines for rapid assessment of social, economic and cultural aspects of malaria, Social and economic research, UNDP/ World Bank/WHO, Special Programme for Research and Training in Tropical Diseases (TDR). WHO Geneva. Accessed on 14 Sept 2010 at: http://opentraining.unesco-ci.org/cgibin/page.cgi?g=Detailed%2F1205.html;d=1

Ashwood-Smith H, Coombes Y, Kaimila, N, Bokosi M and Lungu K (2002): Availability and use of sulphadoxinepyrimethamine (SP) in pregnancy in Blantyre District. A safe motherhood and Blantyre Integrated Malaria Initiative (BIMI) joint survey. Malawi Medical Journal 14: 8—11.

Ahorlu CK, Dunyo SK, Afari EA, Koram KA and Nkrumah FK (1997): Malaria-related beliefs and behaviour in southern Ghana: Implications for treatment, prevention and control. Tropical Medicine and International Health 2: 488-499.

Aubel J, Toure I and Diagne M (2004): Senegalese grandmothers promote improved maternal and child nutrition practices: The guardians of tradition are not averse to change. Social Science and Medicine 59: 945-959.

Ahorlu CK, Kwadwo AK and Mitchell GW (2007): Children, pregnant women and the culture of malaria in two rural communities of Ghana. Anthropology and Medicine 14(2): 167-181.

Baume C, Helitzer-Allen D and Kachur P (2000): Patterns of care for childhood malaria in Zambia. Social Science and Medicine 51: 1491–1503.

Alilio MS, Bygbjerg IC and Breman JG (2004): Are multilateral malaria research and control programs the most successful?

Behague D, Tawiah C, Rosato M, Some T and Morrison J (2009): Evidence-based policy-making: The implications of 85

Brabin L, Stokes E, Dumbaya I and Owens S (2009): Rural Gambian women's reliance on health workers to deliver sulphadoxine – pyrimethamine as recommended intermittent preventive treatment for malaria in pregnancy. Malaria Journal 8:25 doi:10.1186/1475-2875-8-25.

globally-applicable research for contextspecific problem-solving in developing countries. Social Science and Medicine 69: 1539–1546 Belay M and Deressa W (2008): Use of insecticide treated nets by pregnant women and associated factors in pre-dominantly rural population in northern Ethiopia. Tropical Medicine and International Health 13(10): 1303-1313.

Brown PJ (1997): Culture and the global resurgence of malaria. In: The anthropology of infectious diseases, pp. 119-144. Eds. MC Inhorn MC and PJ Brown, the Netherlands: Gordon and Breach Publishers.

Benson T (2002): Malawi: An atlas of social statistics. National Statistic Office, Zomba, Malawi and International Food Policy Research Institute, Washington, USA.

Browner CH (1999): On the medicalization of medical anthropology. Medical Anthropology Quarterly 13: 135-40.

Bernard HR (2002): Research methods in anthropology. Qualitative and quantitative approaches, third ed. Walnut Creek, CA: Altamira Press.

Bruun H, Hukkinen J, Huutoniemi K and Thompson Klein J (2005): Promoting interdisciplinary research. The case of Academy of Finland. Edita Oy. Helsinki: Finland.

Bhattacharyya K (1997): Key informants, pile sorts, or surveys? Comparing behavioural research methods for the study of acute respiratory infections in West Bengal. In: The anthropology of infectious diseases, pp. 211-238. Eds. MC Inhorn MC and PJ Brown, the Netherlands: Gordon and Breach Publishers.

Bujra J (2000): Risk and trust: Unsafe sex, gender and AIDS in Tanzania. In: Risk revisted, pp. 59-84. Ed. P Caplan, London: Pluto. Castro A and Singer M (2004): Unhealthy health policy: A critical anthropological examination. Walnut Creek: Altamira Press.

Bibeau G (1997): At work in the fields of public health: The abuse of the rationality. Medical Anthropology Quaterly 11(2): 246255.

Chapman R (2006): Chikotsa, secrets, silence, and hiding: Social risk and reproductive vulnerability in Central Mozambique. Medical Anthropology Quaterly 20(4): 487-515.

Bisika TJ (1996): Malaria case management in children in Malawi: The case of Namasalima. An unpublished report of a rapid ethnographic study. Centre for Social Research, University of Malawi.

Chavasse D (2002): Know your enemy. Some facts about the natural history of Malawi’s Anopheles mosquitoes and implications for malaria control. Malawi Medical Journal 14(1): 7-8.

Brabin L and Brabin BJ (2005): HIV, malaria and beyond: Reducing the disease burden of female adolescents. Malaria Journal 2005, 4:2 doi:10.1186/1475-2875-42

Chimbiri A (2007): The condom is an intruder in marriage: Evidence from rural Malawi. Social Science and Medicine. 64:1102-1115. 86

Cochrane Collaboration (2010):Accessed on 9 April 2010 at http://www.cochrane.org/

Microbiology and Immunology 295: 169– 200

Cottrell G, Jean-Yves M, Drissa B and Michel C (2005): Is malarial placental infection related to peripheral infection at any time of pregnancy? American Journal of Tropical Mediciene and Hygiene 73: 1112-1118.

Dugas M, Dubé E and Bibeau G (2009): Translating malaria as sumaya: Justified convention or inappropriateness? Anthropology and Medicine 16(3): 307-318. Dwyer-Shick S (1992): Some consequences of fieldworker’s gender for cross-cultural research. In: The naked anthropologist. tales from around the world, pp. 253-263. Ed. PR DeVita, Belmont, California: Wadsworth Publishing Company.

Crawley J, Hill J, Yartey J, Robalo M, Serufilira A, Ba-Nguz A, Roman E, Palmer A, Asamoa K and Steketee R (2007): From evidence to action? Challenges to policy change and programme delivery for malaria in pregnancy. Lancet Infectious Diseases 7: 145–155.

Ecks S (2008): Three propositions for an evidence-based medical anthropology. Journal of the Royal Anthropological Insitute S77-S92.

Crewe E and Young J (2002): Bridging research and policy: Context, evidence and links. Working paper 173. Overseas Development Institute. Accessed on 15 Feb 2010 at: http://www.odi.org.uk/resources/download/ 151.pdf.

Eriksen TH (2001): Small Places, Large Issues. An Introduction to Social and Cultural Anthropology. Second Edition. London: Pluto Press. Espino F, Manderson L, Acuin C, Domingo, F and Ventura E (1997): Perceptions of malaria in the Philippines: Transmission and prevention of disease. Acta Tropica 63: 221–239.

D'Alessandro U, Langerock P, Bennett S, Francis N, Cham K, Greenwood BM (1996): The impact of a national impregnated bed net programme on the outcome of pregnancy in primigravidae in the Gambia. Transactions of the Royal Society of Tropical Medicine and Hygiene 90 (5): 487-492.

Ettling M, McFarland DA, Schultz LJ, Chitsulo L (1994): Economic impact of malaria in Malawian households. Tropical Medicine and Parasitology 45(1):74-9.

Dellicour S, Tatem AJ, Guerra CA, Snow RW and ter Kuile FO (2010): Quantifying the number of pregnancies at risk of malaria in 2007: A demographic study. PLoS Med 7(1):e1000221. doi:10.1371/journal.pmed.1000221

Farmer PE (1997): Social scientists and the new tuberculosis. Social Science and Medicine 44: 347—358. Farmer PE, Nizeye B, Stulac S, Keshavjee S (2006): Structural violence and clinical medicine. PLoS Med 3(10): e449. doi:10.1371/journal.pmed.0030449

Desai M, ter Kuile F, Nosten F, McGready R, Asamoa K, Brabin B. and Newman R (2007): Epidemiology and burden of malaria in pregnancy. The Lancet Infectious Diseases (7) 2: 93-104.

Fetterman D (1989): Ethnography. Step by step. Applied Social Research Methods Series Volume 17. USA: Sage Publications.

Duffy PE and Fried M (2005): Malaria in the pregnant woman. Current Topics in 87

Foster G (1976): Disease etiologies in nonwestern medical systems. American Anthropologist 78: 773–782.

Good B (1994): Medicine, rationality, and experience: An anthropological perspective. New York: Cambridge University Press.

Foster G (1982): Medical anthropology and international health: Retrospect and prospect. Human Organization 41: 189-197.

Gramiccia G (1981): Health education in malaria control - why has it failed? World Health Forum. 2: 385-393.

Foster G (2009): Bureautic aspects of international health programs. In: The anthropology of infectious diseases, pp. 681-701. Eds. MC Inhorn MC and PJ Brown, the Netherlands: Gordon and Breach Publishers.

Granado S, Obrist Tanner M (2009): Treating malaria in Anthropology and 331.

B, Manderson L and The moment of sale: Abidjan, Côte d’Ivoire. Medicine 16(3): 319-

Green E (1999): Indigenous Theories of Contagious Disease. California: Altamira Press.

Gardner K and Lewis D (1996): Anthropology, development and the postmodern challange. USA: Pluto Press.

Guyatt HL and Snow RW (2001): The epidemiology and burden of Plasmodium falciparum-related anemia among pregnant women in sub-Saharan Africa. American Journal of Tropical Medicine and Hygiene 64: 36–44.

Ghebreyesus TA, Alemayehu T, Bosman A, Witten KH, Teklehaimanot A (1996): Community participation in malaria control in Tigray region, Ethiopia. Acta Tropica 61: 145–156.

Hahn RA and Inhorn MC (2009): Anthropology and public health. Bridging differences in culture and society. Second edition. New York: Oxford University Press

Gikandi PW, Noor AM, Gitonga CW, Ajanga AA and Snow RW (2008): Access and barriers to measures targeted to prevent malaria in pregnancy in rural Kenya. Tropical Medicine and International Health 13 (2): 208–217.

Hahn RA and Inhorn MC (2009): Introduction, pp.1-31. In: Anthropology and public health. Bridging differences in culture and society. Eds. RA Hahn and MC Inhorn, New York: Oxford University Press.

Gilson L and McIntyre D (2008): The interference between research and policy: experience from South Africa. Social Science and Medicine 67: 748-759.

Hamel MJ, Odhacha A, Roberts JM and Deming MS (2001): Malaria control in Bungoma district, Kenya: A survey of home treatment of children with fever, bednet use and attendance at antenatal clinics. Bulletin of World Health Organization 79, 1014– 1023.

GMAP (2008): Roll Back Malaria Partnership. Accessed on 17 Apr 2010 at: http://www.rollbackmalaria.org/gmap/gmap .pdf GoM (2005): Malaria strategic plan 20052010. Scaling up malaria control interventions. Draft 1. Malawi: Lilongwe.

Hastrup K (2004): Getting it right. Knowledge and evidence in anthropology. Anthropological Theory 4(4): 455-472.

Gomes M (1993): “Economic and demographic research on malaria: A review of the evidence”. Social Science and Medicine 37: 1093-1108. 88

prophylaxis during pregnancy in Malawi. Acta Tropica 58: 225–266.

Hausmann-Muela S, Muela Ribera J and Tanner M (1998): Fake malaria and hidden parasites-the ambiguity of malaria. Anthropology and Medicine 5(1): 43–61.

Hemmings CP (2005): Rethinking medical anthropology: How anthropology is failing medicine. Anthropology and Medicine 12(2): 91-103.

Hausmann-Muela S, Muela Ribera J, Mushi AK and Tanner M (2002): Medical syncretism with reference to malaria in a Tanzanian community. Social Science and Medicine 55:403-413.

Hill J and Kazembe P (2006): Reaching the Abudja target for intermittent preventive treatment of malaria in pregnancy in African women: a review of progress and operational challenges. Tropical Medicine and International Health 11: 409—418.

Hausmann-Muela S, Muela Ribera J and Nyamongo I (2003): Health-seeking behaviour and the health system's response. DCPP Working Paper no. 14. Accessed on 2 Feb 2010 at: http://www.dcp2.org/file/29/wp14.pdf.

Hogg A, Makwiza B, Mlanga S, Broadhead R and Brabin L (2005): “Finding a curriculum that works under trees: literacy and health education for rural adolescent girls in Malawi”. Development in Practice 15 (5): 655 – 667.

Heggenhougen K, Hackentha V and Vivek P (2003): The behavioural and social aspects of malaria and its control. An introduction and annotated bibliography. TDR/STR/SEB/VOL/03.1. Accessed on Nov 2009 at: http://apps.who.int/tdr/svc/publications/tdrresearch-publications/social-aspectsmalaria-control

Holtz TH, Kachur SP, Roberts JM, Marum LH, Mkandala C, Chizani N, Macheso A and Parise ME (2004): Use of antenatal care services and intermittent preventive treatment for malaria among pregnant women in Blantyre District, Malawi. Tropical Medicine and International Health. 9(1):77-82.

Helitzer-Allen D (1989): Examination of the Factors Influencing utilization of the antenatal malaria chemoprophylaxis program Malawi, Central Africa. Ph.D. Thesis. Johns Hopkins University. Baltimore, Maryland.

Hsiang MS, Panosian C and Dorsey G (2009): Understanding malaria. In: Shrinking the malaria map: A Prospectus on malaria elimination, pp. 81-94. Eds. RGA Feachem, AA Phillips and GA Targett. Accessed on 16 April 2010 at: http://www.malariaeliminationgroup.org/pu blications/shrinking-malaria-mapprospectus-malaria-elimination

Helitzer-Allen D and Kendall C (1992): Explaining differences between qualitative and quantitative data. A study of chemoprophylaxis during pregnancy. Health Education Quarterly 19(1), 41—54.

Inhorn MC (2006): Defining women’s health: A dozen messages from more than 150 ethnographies. Medical Anthropology Quaterly 20(3): 345-378.

Helitzer-Allen D, Kendall C and Wirima JJ (1993): The role of ethnographic research in malaria control: an example of Malawi. Research in the Sociology of Health Care 10: 269–286.

Janes CR (2009): An ethnographic evaluation of post-Alma Ata health system reforms in Mongolia: Lessons addressing health inequities in poor communities. In:

Helitzer-Allen D, Macheso A, Wirima JJ and Kendall C (1994): Testing strategies to increase use of chloroquine chemo89

Kerr RB, Dakishoni L, Shumba L, Msachi R and Chirwa M (2008): ‘We grandmothers know plenty’: Breastfeeding, complementary feeding and the multifaceted role of grandmothers in Malawi. Social Science and Medicine 66: 1095-1105.

Anthropology and public health, pp 652680. Eds. RA Hahn and MC Inhorn, New York: Oxford University Press. Janzen J (1978): The quest for therapy: Medical pluralism in Lower Zaire. Berkeley: University of California Press.

Keusch GT, Kilama WL, Moon S, Szlezák NA and Michaud CM (2010): The global health system: Linking knowledge with action - learning from malaria. PLoS Med 7(1): e1000179. doi:10.1371/journal.pmed.1000179

Justice J (1999): Neglect of cultural knowledge in health planning: Nepal’s assistant nurse-midwife program. In: Anthropology in Public Health. Bridging Differences in Culture and Society, pp. 327344. Ed. RA Hahn, New York: Oxford University Press.

Khan M and Manderson L (1992): Focus groups in tropical diseases research. Health Policy and Planning 7(1): 56-66.

Kadzanzira JM and Munthali A (2004): The coverage and utilization of insecticide treated nets and malaria prevention and treatment practices at the community level in Malawi. Lilongwe: Government of Malawi, National Malaria Control Programme and Ministry of Health and Population.

Kishindo P (1995): Family planning and the Malawian male. Nordic Journal of African Studies 4(1): 26-34. Kitzinger J and Barbour RS (2001): Introduction: the challenge and promise of focus groups. In: Developing focus group research. Politic, theory and practice, pp. 120. Eds. RS Barbour and J Kitzinger,. London: Sage Publications.

Kamat VR (2006): "I though it was only ordinary fever!" Cultural knowledge and the micropolitics of therapy seeking for childhood febrile illness in Tanzania. Social Science and Medicine 62:2945–2959.

Kleinman A (1980): Patients and healers in the context of culture. Berkeley: University of California Press.

Kamat VR (2008): Dying under the bird's shadow: Representations of degedege and child survival among the Zaramo of Tanzania. Medical Anthropology Quarterly 22(1):67-93.

Kulmala T (2000): Maternal health and pregnancy outcomes in rural Malawi. Acta Universitatis Tamperensis, University of Tampere.

Kamat VR (2009): Cultural interpretations of the efficacy and side effects of antimalarials in Tanzania. Anthropology and Medicine 16:293–305.

Kvinnoforum: A guide to gender and malaria resources. Accessed on 3 Feb 2010 at: http://www.rollbackmalaria.org/docs/advoc acy/gm_guide-en.pdf

Kengeya-Kayondo JF, Seeley JA, KajuraBajenja E, Kabunga E, Mubiru E, Sembajja F and Mulder DW (1994): Recognition, treatment seeking behaviour and perception of cause of malaria among the rural women in Uganda. Acta Tropica. 58: 267–273.

Lairumbi GM, Molyneux S, Snow R, Marsh K, Peshu N and English M (2008): Promoting the social value of research in Kenya: Examining the practical aspects of collaborative partnership using ethical framework. Social Science and Medicine 67: 734-747. 90

Campbell C, Molyneux M, Bronzan R, Dodoli W, Ali D and Kabuluzi S (2009): Developing national treatment policy for falciparum malaria in Africa: Malawi experience. Transactions of Royal Society of Tropical Medicine and Hygiene 103 (Suppl 1):S15-8.

Lambert H and McKevitt C (2002): Anthropology in health research: From qualitative methods to multidisciplinarity. British Medical Journal 325: 210–3. Lambert H (2006): Accounting for EBM: Notions of evidence in medicine. Social Science and Medicine 62: 2633-2645.

Malinowski B (1961): Argonauts of the Western Pasific. New York: E.P. Dutton (first published 1922).

Lambert H (2009): Evidentiary truths? The evidence of anthropology through the anthropology of medical evidence. Anthropology Today 25(1): 16–20.

Manderson L (1998): Applying medical anthropology in the control of infectious disease. Tropical Medicine and International Health 3(12): 1020-1025.

LTHCP internal evaluation report (2000): Unpublished report.

Manderson L and Aaby P (1992): An epidemic in the field? Rapid assessment procedures and health research. Social Science and Medicine 35(7): 839-850.

MacCormack CP (1982): Ethnography of fertility and birth. London: Academic Press Inc. Malaria policy (2002): National Malaria Control Programme, Community Health Sciences Unit, Division of Preventive Health Services, Ministry of Health and Population. Lilongwe: Government of Malawi.

Marcus GF and Fisher MJ (1986): Anthropology as cultural critique. An experiemental moment in the human sciences. Chicago: The University of Chicago Press.

Malawi malaria strategic plan 2005-2010. Ministry of Health and Population. Lilongwe: Malawi.

Masache E (2002): An assessment of the knowledge, beliefs and practices of pregnant women towards malaria infection in pregnancy and its prevention in rural Blantyre. Collage of Medicine, Blantyre, Malawi. Unpublished report.

Malawi millennium development goals report (2008): Ministry of Economic Planning and Development. Lilongwe: Malawi. Accessed on 5 Feb 2010 at: http://www.undp.org.mw/index.php?option =com_content&view=article&id=74:malawi -releases-the-2008-mdgsreport&catid=1:latest-news&Itemid=96

Matinga PU (2000): Reports on focus group discussions conducted in Blantyre District. PSI Malawi. Unpublished report. Matinga PU, Ashwood-Smith H and Bokosi M (2000): Qualitative needs assessment report for Chikwawa, Mangochi and Thyolo Districts. Safe motherhood project. Blantyre: Malawi.

Malawi RBM consultative mission (2004): Essential actions to support the attainment of the Abudja targets- 19-23 April 2004. Unpublished report. Accessed on 10 Jan 2010 at: http://www.rollbackmalaria.org/partnership/ country/docs/SAfrica/reaping_malawi.pdf

Mbonye AK, Neema S, Magnussen P. (2006a): Malaria in pregnancy, risk perceptions and care seeking practices among adolescents in Mukono district

Malenga G, Wirima J, Kazembe P, Nyasulu Y, Mbvundula M, Nyirenda C, Sungani F, 91

Molyneux CS, Murira G, Masha J and Snow RW (2002): Intra-household relations and treatment decision-making for childhood illness: A Kenyan case study. Journal of Biosocial Science (2002), 34(1):109-131.

Uganda. International Journal of Adolescent Medicine and Health 18(4): 561-573. Mbonye AK, Neema S, Magnussen P. (2006b): Preventing malaria in pregnancy: A study of perceptions and policy implications in Mukono district, Uganda. Health Policy and Planning 21(1):17-26.

Montgomery CM, Mwengee W, Kong'ong'o M, Pool R (2006): 'To help them is to educate them': power and pedagogy in the prevention and treatment of malaria in Tanzania. Tropical Medicine and International Health. 11(11):1661-9.

McCombie S (1996): Treatment seeking for malaria: A review of recent research. Social Science and Medicine 43(6): 933-945. MDHS (2004): Preliminary Report. National Statistic Office. Zomba: Malawi.

Morris B (1985): Chewa conceptions of disease – symptoms and etiologies. The Society of Malawi Journal 38:14-38.

Menendez C (1995): Malaria during pregnancy: A priority area of malaria research and control. Parasitol Today. 11(5):178-83.

Moser CON (1993): Gender planning and development. Theory, practice and training. London: Routledge.

Menendez C, Ordi J, Ismail MR, Ventura PJ, Aponte JJ, Kahigwa E, Font F, Alonso PL (2000): The impact of placental malaria on gestational age and birth weight. Journal of Infectious Diseases. 181(5): 17401745.

Mosse D (2004): Is good policy unimplementable? Reflections on the ethnography of aid policy and practice. Development and Change 35(4): 639-671. Msyamboza KP, Savage EJ, Kazembe PN, Gies S, Kalanda G, D'Alessandro U and Brabin BJ (2009): Community-based distribution of sulfadoxine-pyrimethamine for intermittent preventive treatment of malaria during pregnancy improved coverage but reduced antenatal attendance in southern Malawi. Tropical Medicine and International Health 14 (2): 183 – 189.

MIPESA (2006): Assessment of MIPESA country experiences in the adoption and implementation of malaria in pregnancy policies. WHO, USAID and ACCESS. Accessed on 18 April 2010 at: http://www.rollbackmalaria.org/partnership/ wg/wg_pregnancy/docs/MIPESA_Report20 06.pdf Mitchell JC (1956): The Yao village. A study in the social structure of a Malawian people. UK: Manchester University Press.

Mubyazi GM, Bygbjerg IC, Magnussen P, Olsen Ø, Byskov J, Hansen KS and Bloch P (2008): Prospects, achievements, challenges and opportunities for scaling-up malaria chemoprevention in pregnancy in Tanzania: the perspective of national level officers Malaria Journal 7:135. doi:10.1186/14752875-7-135

Mitchell JC (1962): Marriage, matriliny and social structure among the Yao of Southern Nyasaland. International Journal of Comparative Sociology 3: 29-42. Mogensen H (1995): AIDS is a kind of Kahungo that kills. The challenge of using local narratives when exploring AIDS among the Tonga of southern Zambia. Oslo: Scandinavia University Press.

Mull DS (1999): Anthropological perspectives on childhood pneumonia in Pakistan. In: Anthropology in public health. Bridging differences in culture and society, pp. 8492

pyrimethamine use in pregnancy in Blantyre city. Collage of Medicine, Blantyre, Malawi. Unpublished report.

114. Ed. RA Hahn, New York: Oxford University Press. Munthali A (2005): Managing malaria in under-five children in a rural Malawian village. Nordic Journal of African Studies 14(2): 127–146. Mvula P and Kakhongwa P (1997): Beyond inequalities: Women in Malawi. Harare: SARDC.

Neville CG, Some ES, Mung'ala VO, Mutemi W, New L, Marsh K, Lengeler C and Snow RW (1996): Insecticide-treated bednets reduce severe morbidity from malaria among children on the Kenyan coast. Tropical Medicine and International Health 1: 139–146.

Mwenesi H (1994): The role of drug delivery systems in health care: the case of self medication. African Journal of Health Sciences 1 (1): 42-48.

Nichter M (1989): Anthropology and international health: South Asian case studies. Dordrecht, the Netherlands: Kluwer Academic Publishers.

Mwenesi H, Harpham T, and Snow RW (1995): Child malaria treatment practices among mothers in Kenya. Social Science and Medicine 40: 1271–1277.

Nichter M (1993): Social-science lessons from diarrhea research and their application to ARI. Human Organization 52 (1): 53-67. Nichter M (2008): Global health. Why cultural perceptions, social representations, and biopolitics matter. Tuscon: the University of Arizona Press.

Mykhalovskiy E and Weir L (2004): The problem of evidence-based medicine: Directions for social science. Social Science and Medicine 59: 1059-1069.

NMCP (2005): Malaria Communications Strategy 2005-2010. Draft, June 17, 2005. National Malaria Control Program, Lilongwe: Malawi

Napolitano DA and Jones COH (2006): Who needs “pukka anthropologists”? A study of the perceptions of the use of anthropology in tropical public health research. Tropical Medicine and International Health 11(8): 1264-1275.

Nuwaha F (2002): People’s perceptions of malaria in Mbarara, Uganda. Tropical Medicine and International Health 7: 462– 470.

Ndyomugyenyi R, Neema S and Magnussen P (1998): The use of formal and informal services for antenatal care and malaria treatment in rural Uganda. Health Policy and Planning 13(1): 94–102.

Obrist B, Iteba N, Lengeler C, Makemba A, Mshana C, Nathan R, Alba S, Dillipn A, Hetzel MW, Mayumana I, Schulze A and Mshinda H (2007): Access to health care in contexts of livelihood insecurity: A Framework for Analysis and Action. PLoS Medicine 4(10):e308. doi:10.1371/journal.pmed

Nganda RY, Drakeley C, Reyburn H and Marchant T (2004): Knowledge of malaria influences the use of insecticide treated nets but not intermittent presumptive treatment by pregnant women in Tanzania. Malaria Journal (3):42 doi:10.1186/14752875-3-42

Okonofua FE, Feyisetan BJ, DaviesAdetugbo and Sanusi YO (1992): Influence of socio-economic factors on the treatment and prevention of malaria in pregnant and non-pregnant adolescent girls in Nigeria.

Ngoma J (1999): Knowledge, attitudes and practices and beliefs about sulphadoxine93

Journal of Tropical Medicine and Hygiene 95: 309-315.

Chewa women in central Malawi. Malawi: Christian Literature Association in Malawi.

Okrah J, Traoré C, Palé A, Sommerfeld J and Müller O (2002): Community factors associated with malaria prevention by mosquito nets: An exploratory study in rural Burkina Faso. Tropical Medicine and International Health 7(3):240-8.

Pool R (1994): Dialogue and the interpretation of illness. Conversation in a Cameroon Village. USA: Berg Publishers. Pool R and Geissler W (2007): Medical anthropology. Understanding public health. New York, USA: Open University Press.

Pablos-Mendez A, Chunharas S, Lansang MA, Shademani R, Tugwell P (2005): Knowledge translation in global health. Bulletin of the World Health Organization (83) 10: 723.

Pope C (2003): Resisting evidence: The study of evidence based medicine as a contemporary social movement. Health 7(3): 267-282. Purohit B and Mahapatra A (2009): A review on high burden of malaria during pregnancy: Need of social science intervention. Ethno-Medicine 3(1): 33-38.

Parker R (2002): Administrating the epidemic: HIV/AIDS policy, models of development, and international Health. In: Global health policy, local reality. The fallacy of the level playing field, pp 39-55. Eds. LM Whiteford and L Manderson London: Lynne Rienner Publishers.

Ramalingaswami V (1986): The art of the possible. Social Science and Medicine 22(11): 1097-1103.

Paz Soldan VA, deGraft-Johnson JE, Bisika T and Tsui AO (2007): Social, economic and demographic determinants of sexual risk behaviors among men in rural Malawi: A district-level study. African Journal of Reproductive Health 11(2):33–46.

Rashed S, Johnson H, Dongier P, Moreau R, Lee C, Crepeau R, Lambert J, Jefremovas V, Schafferet C (1999): Determinants of the permethrin impregnated bednets (PIB) in the Republic of Benin: the role of women in the acquisition and utilisation of PIBs. Social Science and Medicine 49: 993–1005.

Pelto JP and Pelto GH (1997): Studying knowledge, culture, and behavior in applied medical anthropology. Medical Anthropology Quarterly 11(2): 147-163.

RBM 2005-2015. Global Strategic Plan. Accesses on 18 April 2010 at: http://www.rollbackmalaria.org/forumV/doc s/gsp_en.pdf

Pfeiffer J (2004): International NGOs in the Mozambique health sector: The “velvet glove” of privatization. In: Unhealthy health policy: A critical anthropological examination, pp- 43-62. Eds. A Castro and M Singer, Walnut Creek: Altamira Press.

Reigeluth CM (1999): Instructional-design theories and models: A new paradigm of instructional theory. USA: Lawrence Erlbaum Associates

Pfeiffer J and Nichter M (2008): What can critical medical anthropology contribute to global health. Medical Anthropology Quarterly 22(4): 410-415.

Ribera JM, Hausmann-Muela S, D'Alessandro U, Grietens KP (2007): Malaria in pregnancy: What can the social sciences contribute? PLoS Med 4(4): e92. doi:10.1371/journal.pmed.0040092

Phiri IA (1997): Women, presbyterianism and patriarchy: Religious experience of 94

practices in Lungwena Health and Agriculture Multidisciplinary Project Area. Unpublished report.

Rogerson SJ (2000): Malaria in pregnancy. Malawi Medical Journal 12(1): 6-11. Rogerson, SJ, van den Broek NR, Chaluluka E, Qongwane C, Mhango CG, and Molyneux ME (2000): Malaria and anemia in antenatal women in Blantyre, Malawi: A twelve-month survey. American Journal of Tropical Medicine and Hygiene 62(3): 335-340.

Sillitoe Paul (2002): Participant observation to participatory development: Making anthropology work. In: Participating in development. Approaches to indigenous knowledge, pp. 1-23. Eds. P Sillitoe, A Bicker and J Pottier, London: Routledge. Smith LA, Jones C, Adjei RO, Antwi GD, Afrah NA, Greenwood B, Chandramohan D, Tagbor H and Webster J (2010): Intermittent screening and treatment versus intermittent preventive treatment of malaria in pregnancy: user acceptability. Malaria Journal 14;9:18. doi:10.1186/1475-2875-918

Rylko-Bauer B, Singer M and Van Willigen J (2006): Reclaiming applied anthropology: Its past, present, and future. American Anthropologist 108(1): 178-190. Sachs J and Malaney P (2002): The economic and social burden of malaria. Nature 415: 680–685.

Sommerfeld J, Sanon M, Kouyaté BA and Sauerborn R (2002): Perceptions of risk, vulnerability, and disease Prevention in rural Burkina Faso: Implications for community-based health care and insurance. Human Organization 61(2):139–146.

Samuelsen H, Toe LP, Baldet T and Skovmand O (2004): Prevention of mosquito nuisance among urban populations in Burkino Faso. Social Science and Medicine 59: 2361–2371. Savage J (2000): Ethnography and health care. British Medical Journal 321:14001402.

Stambach A (1994): “Here in Africa, we teach; students listen”: Lessons about culture from Tanzania. Journal of Curriculum and Supervisions 9: 368-385.

Schatz E (2005): “Take your mat and go!”: Rural Malawian women’s strategies in the HIV/AIDS era. Culture, Health and Sexuality 7(5): 479–492.

Steketee RW, Wirima JJ, Slutsker L, Roberts JM, Khoromana CO, Heymann DL, Breman JG (1996): Malaria parasite infection during pregnancy and at delivery in mother, placenta, and newborn: efficacy of chloroquine and mefloquine in rural Malawi. American Journal of Tropical Medicine and Hygiene 55 (1 Suppl):24-32.

Schultz LJ, Steketee RW, Chitsulo L, Macheso A, Nyasulu Y, Ettling M (1994): Malaria and childbearing women in Malawi: Knowledge, attitudes and practices. Tropical Medicine and Parasitology 45(1):65-9.

Steketee RW, Nahlen BC, Parise ME and Menendez C (2001): The burden of malaria in pregnancy in malaria-endemic areas. American Journal of Tropical Medicine and Hygiene. 64 (Suppl. 1/2), 28–35.

Seljeskog L, Sundby J and Chimango J (2006): Factors influencing women's choice of place of delivery in rural Malawi. An explorative study. African Journal of Reproductive Health 10 (3): 66-75.

Stone L and Campbell JG (1984): The use and misuse of surveys in international

Shame E, Kanyerere T and Kamanga M. A sociolinguistic analysis of communicative 95

development: An experiment from Nepal. Human Organization 43(1): 27-34.

in Malawi. Social Science and Medicine. 67: 760-770.

Stratton L, O'Neill MS, Kruk ME, Bell ML (2008): The persistent problem of malaria: Addressing the fundamental causes of a global killer, Social Science and Medicine (67)5: 854-862

Theobald S, Tolhurst R, Squire SB (2006): Gender, equity: new approaches for effective management of communicable diseases. Transactions of the Royal Society for Tropical Medicine and Hygiene100 (4): 299-304.

Tanner M and Vlassoff C (1998): Treatment-seeking behaviour for malaria: A typology based on endemicity and gender. Social Sciene and Medicine. (4-5):523-32.

Thorold A (1993): Metamorphoses of the Yao muslims. In: Muslim identity and social change in sub-Saharan Africa, pp. 7990. Ed. L Brenner, Bloomington: Indiana University Press.

Tarimo DS, Lwihula GK, Minjas JN and Bygbjerg IC (2000): Mothers’ perceptions and knowledge on childhood malaria in the holoendemic Kibaha district, Tanzania: Implications for malaria control and the IMCI strategy. Tropical Medicine and International Health 5(3), 179–184.

Tolhurst R, Amekudzi YP, Nyonator FK, Bertel Squire S, Theobald S (2008): "He will ask why the child gets sick so often": The gendered dynamics of intra-household bargaining over healthcare for children with fever in the Volta Region of Ghana. Social Science and Medicine. 66(5):1106-17.

TDR annual report (2009): Innovation for health Research that makes a difference. Accessed 10 Jan 2010 at: http://apps.who.int/tdr/svc/publications/abo ut-tdr/annual-reports/tdr-report09 Doi: 10.2471/TDR.10.978-924-1599702

Tolhurst R and Nyonator FK (2006): Looking within the household: Gender roles and responses to malaria in Ghana. Transactions of Royal Society of Tropical Medicine and Hygiene 100(4):299-304

Ter Kuile FO, Parise ME, Verhoeff FH, Udhayakumar V, Newman RD, van Eijk AM, Rogerson SJ and Steketee RW (2004): The burden of co-infection with human immunodeficiency virus type 1 and malaria in pregnant women in sub-saharan Africa. American Journal of Tropical Medicine and Hygiene 71(2 Suppl):41-54.

Trostle J (2002): International health research: The rules of the Game. In: Global health policy, local reality. The fallacy of the level playing field, pp 291-313. Eds. LM Whiteford and L Manderson London: Lynne Rienner Publishers. Trostle J (2005): Epidemiology and culture. USA: Cambridge University Press.

The Economic and Social Council (1997): Gender mainstreaming. Extract from (A/52/3, 18 September 1997). Accessed on 20 April 2010 at: http://www.un.org/womenwatch/daw/csw/G MS.PDF

UNICEF (2001): Ethnographic malaria research Report. Lilongwe: Malawi. UNICEF (2008): Annual Report. Lilongwe: Malawi.

Theobald S and Nhlema-Simwaka B (2008): The research, policy and practice interface: Reflections on using applied social research to promote equity in health

Van der Geest S (1995): Overcoming ethnocentrism: How social science and medicine relate and should relate to one

96

Winch PJ, Makemba AM, Kamazima SR, Lurie M., Lwihula GK, Premji Z, Minjas JN and Shiff CJ (1996): Local terminology for febrile illnesses in Bagamoyo district, Tanzania and its impact on the design of a community-based malaria control programme. Social Science and Medicine. 42: 1057–1067.

another. Social Science and Medicine 40(7): 869-872. Verhoeff FH, Brabin BJ, Chimsuku L, Kazembe P, Russell WB, Broadhead RL (1998): An evaluation of the effects of intermittent sulfadoxine-pyrimethamine treatment in pregnancy on parasite clearance and risk of low birthweight in rural Malawi. Annals of Tropical Medicine and Parasitology 92(2):141-50.

Whiteford LM and Manderson L (2002): Global health policy, local reality. The fallacy of the level playing field. London: Lynne Rienner Publishers

Vlassoff C and Manderson L. (1998): Incorporating gender in the anthropology of infectious diseases. Tropical Medicine and International Health. 3(12):1011-9. Review.

WHO (2004): A strategic framework for malaria prevention and control during pregnancy in the African Region, Brazzaville: WHO Regional Office for Africa, 2004. AFR/MAL/04/01.

Wernsdorfer W, Hay SI and Shanks GD (2009): In: Shrinking the malaria map: A prospectus on malaria elimination, pp. 95107. Eds. RGA Feachem, AA Phillips and GA Targett. Accessed on 16 April 2010 at: http://www.malariaeliminationgroup.org/pu blications/shrinking-malaria-mapprospectus-malaria-elimination

WHO (2005): The World health report: make every mother and child count. Geneva: Switzerland. Accessed on 10 Dec 2009 at: http://www.who.int/whr/2005/whr2005_en. pdf WHO (2009): Ten facts on malaria. Accessed on 10 Feb 2010 at: http://www.who.int/features/factfiles/malari a/en/index.html

White N (2008): Plasmodium knowlesi: the fifth human parasite. Clinical Infectious Diseases 46: 172-3. Williams HA, Kachur SP, Nalwamba NC, Hightower A, Simoonga C and Mphande PC (1999): A community perspective on the efficacy of malaria treatment options for children in Lundazi District, Zambia. Tropical Medicine and International Health 4 (10):1365-3156.

World Malaria Report (2005): WHO and UNICEF. Accessed on 11 Jan 2010 at: http://www.rollbackmalaria.org/wmr2005/p df/WMReport_lr.pdf. World Malaria Report (2008): WHO: Geneva. Accessed 20 April 2010 at: http://www.who.int/malaria/publications/ato z/9789241563697/en/index.html

Williams HA, Durheim D and Shretta R (2004): The process of changing national malaria treatment policy: lessons from country-level studies. Health Policy and Planning 19(6): 356-370.

World Malaria Report (2009): WHO: Geneva. Accessed 20 April 2010 at: http://www.who.int/malaria/world_malaria_ report_2009/en/index.html

Williams HA and Jones COH (2004): A critical review of behavioural issues related to malaria control in sub-Saharan Africa: what contributions have social scientists made? Social Science and Medicine 59: 501–523.

Yartey JE (2006): Malaria in pregnancy: access to effective interventions in Africa. International Journal of Gynecology and Obstetrics 94(3):364-73. 97

Yoder PS (1997): Negotiating Relevance: Belief, Knowledge, and Practice in International Health Projects. Medical Anthropology Quarterly 11 (2): 131-146. Young A (1976): Internalizing and externalizing medical belief systems: An Ethiopian example. Social Science and Medicine 10: 147-156. Zulu EM (1996): Social and cultural factors affecting reproductive behaviour in Malawi. A dissertation in Demography. University of Pennsylvania.

98

Annex 1: Main findings according to each publication Publication Main findings Local illness classification and naming Launiala A and Kulmala T (2006). malungo - ambiguous term meaning malaria, fever, body pain, feeling unwell and joint pains The importance of used interchangeably for many types of feverish illnesses, body pains, and not just for malaria understanding the multiple malungo types defined according to symptoms and causes, but no specific type related to pregnancy local context: a natural illness and recognition of a malungo type related to witchcraft although pregnant women perceived Women’s perceptions vulnerable to ill will and witchcraft and knowledge no single shared understanding – meaning of malungo produced in everyday encounters concerning malaria in Perceptions concerning malaria in pregnancy pregnancy in rural worrisome and common illness, but not specifically in relation to pregnancy Malawi. Acta Tropica fever can be interpreted as normal and a sign of pregnancy 98: 111-117. (I) Knowledge of malaria in pregnancy varies among the women transmission attributed to multiple natural causes, including mosquitoes knowledge of adverse consequences of malaria in pregnancy low. Miscarriage most often perceived as adverse consequence Recommendations: to focus on malaria in a wider context i.e. which issues are considered dangerous for pregnancy, why and how malaria is related to these issues in order to gain insight into multiple challenges pregnant women face and how they cope with them Launiala A and Honkasalo ML (2007): Ethnographic study on factors affecting compliance to intermittent preventive treatment of malaria during pregnancy among the Yao women in rural Malawi. Transactions of the Royal Society

Use of medication in pregnancy bitter tasting medicine (herbs and pharmaceuticals) perceived as dangerous as known to cause miscarriage knowledge of pharmaceuticals vague; described by colour, shape and expected purpose source of pharmaceuticals important during pregnancy; ANC trusted while local shops and traditional healers avoided due to wrong dosages and too powerful medication Knowledge and perceptions concerning IPT-SP SP administrated at ANC interpreted as suggestive of malungo purpose of IPT-SP mostly not known 60% (n=48) of pregnant women after administration of SP-IPT at ANC said that MIP cannot be prevented, 40% said that it can be prevented with ITNs (27%) and SP (1%). SP perceived as suitable for pregnant women; little perceived side effects 99

Publication Main findings for Tropical Medicine Knowledge and perceptions concerning ANC attendance and Hygiene knowledge of appropriate time to enrol to ANC varies from 3 to 6 months and above – average starting time 24 101(10):980-9. (II) weeks of pregnancy factors influencing timing of enrolment: 1) necessity to confirm pregnancy before enrolment (movements), 2) unwillingness to reveal pregnancy at early stage due to fear of ill will and witchcraft, 3) no need to attend ANC early when pregnancy progressing normally, 4) distance motivation to attend ANC: 1) to ensure that baby is growing well and 2) to ensure access to services in case of delivery complications Circumstances, practices and interaction at ANC ANC examinations conducted hastily with little verbal interaction SP administrated (together with iron supplements) DOT, but no explanations provided of the preventive purpose problems with appropriate timing of SP-1 IPT periodical shortages of SP communication one-sided and in Chichewa, a local language that is not the mother tongue of pregnant women in the area health education of MIP: vague, without clear reference to prevention of malaria Recommendations and conclusions: problems concerning compliance with IPT-SP should be addressed at the facility level and targeting the nurses research on compliance should be conducted from multiple perspectives including those of pregnant women and providers, as well as taking into account structural and economic factors to explore factors influencing ANC re-attendance to understand the multiple contexts shaping pregnant women’s knowledge and practices benefit of using an ethnographic approach Launiala A and Honkasalo ML (2010): Malaria, danger and risk among the Yao in rural Malawi. Medical Anthropology

Exposure to vulnerabilities in pregnancy: Witchcraft: a strong belief in witches and witchcraft a common phenomenon in the villages pregnancy exposes women to dangers of witchcraft and malevolent spirits that may cause illnesses and delivery complications (interpreted and managed applying the local illness explanatory models) witchcraft feared but not a constant worry, used as an explanation for unexpected and negative consequences (not all believe in witchcraft) 100

Publication Quarterly 24(3): 399320. (III)

Main findings Extra-marital relationships: common practice, also elsewhere in Malawi, particularly attributed to men several taboos related to menstruation, pregnancy, childbirth and sexual activity that have been traditionally used to control husband’s sexual behaviour. Old generation still reinforces these taboos through advice and initiation ceremonies such as litiwo for first time pregnant girls, whereas younger generation seem to have their doubts Illnesses perceived dangerous for pregnant women: a large amount of illnesses caused by natural and supernatural agents no uniformity regarding the most dangerous illness; STIs (gonorrhea, chancroid, syphilis), HIV/AIDS and malungo most often perceived as most dangerous knowledge regarding symptoms, causes, treatment, prevention (local understanding of each illness) varied among individual respondents and groups STIs men’s extra-marital relationship and danger of getting infected with STIs were associated knowledge of STIs were limited and infections were said to be discovered at the ANC where treatment was provided in the form of injection perceived adverse consequences: miscarriage, maternal death, being ill, stillbirth prevention of STIs was largely attributed to men’s behaviour: abstinence from extra-marital relationships. Pregnant women having little agency to control STIs HIV/AIDS has emerged as a dangerous illness that people talk about in terms of “stories they have heard from others” HIV/AIDS infection attributed to infidelity, particularly among men thinness and feeling sick on and off perceived as symptoms of AIDS diagnosis only through blood test at the clinic prevention hard: married women cannot ask husband to use condom or refuse to have sex with husband, and men seem to consider abstinence from extra-marital relationships hard despite the danger of HIV/AIDS perceived as dangerous because hard to prevent and no cure – eventually the woman dies Malungo perceptions of how common malungo in pregnancy is varied: considered uncommon by some pregnant women because “most women take preventive measures” whereas female elders had observed that “after establishment of hospitals women go for malungo treatment” etiology of malungo varied from mosquitoes to multiple other natural causes. normal or severe malungo was not associated with witchcraft. General opinion was that adults and pregnant women 101

Publication

Main findings do not suffer from severe malaria or convulsions (locally referred as kambanga caused by witchcraft that small children are more prone to) treatment (“pills”, Panadol, SP) handed out from clinic during pregnancy prevention methods varied from ITNs to balanced diet, cutting grass, getting medication in advance from the clinic to no prevention. perceptions and knowledge of consequences varied from no or mild consequences to miscarriage, maternal death. Particularly pregnant women expressed that there ate no consequences as long as one attends the ANC and receives pills (SP, Fansidar, Panadol) at the ANC. Also men considered ANC attendance as a way to prevent adverse consequences of malungo Multiple worries about everyday life women’s worries ranged from basic necessities to fears of delivery complications, husband’s sexual behaviour and STIs, ill willing relatives, falling ill, inability to walk and perform household duties, bewitchment, economic constraints at the time of delivery. men’s worries concentrated on the time of delivery. At early stage of pregnancy it is possible that the woman’s body is just resting (kupumula) and she is not pregnant. ANC was considered central to preventing and handling pregnancy related problems. men’s worries ranged from family misunderstanding and conflicts that may cause problems in pregnancy to uncertainty regarding the outcome of the pregnancy and particularly financial constraints (not having money for transport or operation if needed) that would hinder a husband from performing his duties. Conclusions: Malaria in pregnancy is not perceived as a dangerous and severe illness among the Yao because: pregnant women are exposed to co-existing, interlinked and overlapping everyday dangers ranging from witchcraft, extra-marital relationships, multiple illnesses to worries about uncertainties regarding delivery and financial constraints that are valued differently and prioritized based on their relevance at a given time. a wide range of dangerous illnesses for pregnant women exists in the area. Yet the perceived risk of each illness varies according to: 1) perceived adverse consequence; 2) the possibility and easiness to treat and cure the illness; 3) the transmission mode (natural or personalistic agent); and 4) type of risk (medical and/or social risk). social risks overweighing medical risks. STIs and HIV/AIDS can expose pregnant women to social risk and they may end up accepting the immediate medical risks of these diseases, whereas malaria does not expose to social risk. witchcraft is perceived as a common phenomenon that increases the vulnerability of pregnant women. It is perceived as a causal personalistic agent, potentially causing delivery complications. It is also used as an explanation for adverse pregnancy outcomes. 102

Publication

Launiala A (2009): KAP survey in a nonWestern setting: A critical discussion based on a medical anthropology study on malaria in pregnancy in rural Malawi. Anthropology Matters Journal. (IV)

Main findings structural factors (poor services, lack of skilled personnel and equipment, long distances, poor infrastructure) and economic factors (lack of cash earning possibilities, availability of cash and poverty) affect access to health care and are beyond the control of pregnant women in resource poor setting such as Lungwena. malaria in pregnancy should be studied in the context of pregnancy and the everyday life of the women as malaria is not a stand alone issue. Also men and extended family members, and especially grandmothers, should be included as target groups because they are intimately involved with management of maternal health at community level. the epidemiological and public health ‘at risk’ does not match the local perception of vulnerabilities and requires modification. Main aspects of a KAP survey: 1) Knowledge: narrow definition of knowledge as scientific facts and universal truths, aiming to identify knowledge gaps in biomedicine and public health excludes local indigenous knowledge and ‘beliefs’ perceiving them as erroneous information. difficult to ask questions that would capture the local knowledge 2) Attitudes: measuring attitudes problematic due to several factors: tendency to reply in a socially desirable manner rather than revealing one’s own attitude; influenced by interview setting, manipulative question formulation etc. high proportion of ‘agree’ answers that could be due to question formulation, tendency to avoid confrontation. 3) Practices: fails to explain why and when, i.e. the logic behind people’s behaviour difficult to formulate meaningful questions that take account the contextual factors affecting treatment-seeking practices. Challenges encountered in the field: translation of the questionnaire into local language – changing meanings of terms such as malungo (cultural reinterpretation) difficulty of obtaining information concerning sensitive topics such as complications during pregnancy (inadequate rapport between the interviewer and the respondents) difficulty of reaching beyond ‘yes’ and ‘no’ answers to obtain elaborative explanations (unintentional emphasis on quantity over quality, pressure of time, inadequate follow up of the quality of research assistants’ work, skills of individual research assistants) research assistants’ difficulties in unlearning old ways of collecting data courtesy bias (dislike of conflicts and rarely disagree to participate in a study, associated the study with the Lungwena 103

Publication

Main findings H/C and were worried giving negative answers e.g. concerning the ANC services, receiving goods or money in exchange for knowledge) Critical observations: without contextual understanding an inappropriate tool to gather data and interpret results assumption that there is a direct relationship between knowledge and action unnatural method in a rural non-Western setting lack of critical discussion about strengths, weaknesses and limitations of survey methodology use of multi-method designs allows overcoming of some of the survey method weaknesses Challenges and value of interdisciplinary: to find an appropriate way to communicate with professionals that have a ‘quantitative research training’ need for transparency regarding limitations of the study that should be communicated to programme planners and discussed openly in publications multi-disciplinary study teams may provide the best value

104

Annex 2: An example of a theme guideline for FGDs and IDIs THEME 1: COMMON ILLNESSES IN PREGNANCY

THEME 2: DELIVERY PRACTICES AND COMPLICATIONS

1) Could you tell me what are the common illnesses that pregnant women suffer from? Rainy season? Cold season? Hot season?

8) Practices and advising when first time pregnant? Who? What types?

2) What are the symptoms and treatment of these illnesses?

10) Complications regarding pregnancy and delivery? What kind? How and by whom managed?

9) Preparations for delivery? How is decision of delivery place made? Who involved? Probe for a delivery story.

3) Have you had any problems/ illnesses during your pregnancies? (Probe for symptoms, time of the pregnancy).

11) How do you know that your delivery is not going as expected? What are the danger signs? What can you do when you realise that the baby is refusing to come out? How long do you wait before you take any action? Who in the household decide what action should be taken? What kind of traditional practices are there to overcome problems in delivery?

4) What causes fever here? How is fever/different kinds of fevers treated? 5) Can you name any of the feverish illnesses common in this community? (Probe for malungo and different types of malungo) What does malungo mean? Difference between malungo and moto? What causes malungo/different malungos How are they treated? (probe 1st, 2nd … treatment choices and explanations) Waiting time before starting the treatment? How do you know which type of malungo you are suffering from How is severe malungo (malungo with convulsions) called?

12) Prevention of complication in pregnancy? 13) Perceptions and knowledge on maternal deaths? THEME 3: SUPPORT GROUP IN PREGNANCY 14) Who in your family is the head of household? Who has he got the authority to decide what to do when problems arise in the family? Role of other family members?

6) Prevention of malungo? kinds of methods?

15) Advising in pregnancy? What type of advices do you get concerning you pregnancy and from whom? Prohibitions during pregnancy?

7) Dangerous illnesses for pregnant women in the area?

16) Daily work duties in pregnancy?

105

THEME 4: ANC ATTENDANCE AND PERCEPTIONS ON DRUGS 17) When you are pregnant do you go to ANC? Motivation? 18) What month of the pregnancy is the best time for the first visit? Why? 19) Tell me what happens at the ANC when you go there first time? 20) I have here some pills, have you ever seen these types and do you know what they are used for? 21) Do PW have any special restrictions in terms of pills, traditional medicine and food?

106

Annex 3: KAP survey questionnaire SECTION A: No. 1. 2.

BACKGROUND INFORMATION

Question Linalyamusi (Name of village) Mmusi mwacheni (Matrilineal 1= elo (yes) village) 2= lyayi (no), ligongo (specify)

3.

Asyene nlango wani (Head of household)

1= walume (husband) 2= njomba (maternal uncle) 3= wawojo (herself) 4= wakongwe wane (any other female) 5= wane (other) ligongo (specify)

4.

Ana mjonba wawo akutama wawo musi munomuno? (Does the matrilineal uncle live in the same village?) Akutama wandu walingwa nyumba (Number of household members) Walombele (marital status)

1= elo (yes) 2= lyayi ligongo (no, specify where and distance) 3= yine ligongo (other, specify) Achakulungwa (adults) Wanache (children) 6-18 Wanache (children) U-5 1= walombele (married) if yes, go to 7 2= wamasile (divorced) if yes, skip to 8 3= wamkwawo wawile (widow) if skip to 8 4= wangalombela (single) if yes, skip to 8 1= wangali mitale (no co-wives) 2= wakongwe wamo (1 co-wife) 3= achakongwe wawili (2 co-wives) 4= achakongwe watatu kapena kupunda (3 co-wives or more) 1= Yao 2= Chewa 3= Yine, ligongo (other specify) 1= chisilamu (islam) 2= yine, ligongo (other specify) 1= 15 mpaka-19 yaka 2= 20-24 3= 25-29 4= 30-34 5= 35-39 6= 40-44 7= nganguyimanyilila (don’t know) 1= elo, asale mwesi (specify month) 2= Iyayi 1= kamo (1 pregnancy) 2= kawili (2 pregnancies) 3= katatu (3 pregnancies)

5.

6.

7.

Achakongwe walingwa (Number of co-wives, excluding self)

8.

Wamtunduchi (Ethnic group)

9.

Dini chi (Religion)

10.

Yaka Yawo (Age)

11.

Akwete chitumbo (Pregnant at the time of interview) 12a. Aweleche kalingwa (Number of pregnancies)

107

Code

No.

Question

Code

4= kacheche (4 pregnancies) 5= kasano (5 pregnancies) 6= nsanu nachimo – nikupunda (6 or more pregnancies) 12b. Nambala ja wanache wakupagwa aliwumi (Number of children born alive) 1= wosope (all) 13. Wanache wawumi apali walingwa lelojino (Number of 2= yine, ligongo (other specify) children alive today) 14. Sukulu wasomile (level of 1= jwangasoma (none) education) 2= std. 1-2 3= std. 3-4 4= std. 5 kapena kupunda (std. 5 or more) 1= akusawalangaga (she can read) 15. Wakuwalangaga (Level of literacy) Show the respondent 2= wangawalanga (she can’t read) the Chichewa text and ask her to read it 16. Mbiya akusapatila kwapi 1= mukulima (farmimg) (Main source of income) 2= mukuloposya (fishing) 3= muganyu (piecework) 4= mubisimesi, ligongo (business specify) 5= yine, ligongo (other specify)

This question only for outpatient and outreach clinic respondent 1= pakulwala, ligongo 17a. Magongo chigakusya (Sick, specify) jawulilaga kuchipatala (Reason for visiting the clinic) 2= kapena mwana che akulwala, ligongo (child sick, specify) 3= kapena alongo, ligongo (relative sick, specify) 4= lpali yine, ligongo (other, specify) 17b. Imanyilo chi ni chikamuchisyo chi cha ulwele panyuma pa kulinjidwa? Alole mu bukuja wakulwala. What is the diagnosis and treatment of the disease after examination? This part only for ANC respondent 18a. Wajendele kalingwa ku sikelo ja nsingo? (Number of the ANC visit)

1= ulendo wandanda (mwezi chi) (1st visit, month) 2= ulendo wawili (mwezi chi) (2nd visit, month) 3= ulendo watatu (mwezi chi) (3rd visit, month) 4= ulendo wa ncheche kapena kupunda (mwezi chi) (4th visit, month)

108

18b

Alongosole ya ijuma ya nsigowo pa kulola mu kaundula ja sikelo ja misigo ali alinjidwe. Nambala ja kulinjidwa:_______________ Verify the weeks of pregnancy from the ANC register after the examination (at the end of the day)

1= ijuma 14 – mpakana 19 (1st trimester) 2= ijuma 20 – mpakana 28 (2nd trimester) 3= ijuma 29 – mpakana 40 (3rd trimester)

SECTION B: KNOWLEDGE Part B1: Knowledge on malaria Allow spontaneous answers. If you need to probe the different choices, tick next to the question/code indicating ‘probed’. All questions have to have an answer. If the respondent doesn’t answer – mark don’t know. The open-ended questions are written in English. No. 19a. 19b.

Question Ana malungo gakusa dandaulisyaga 1= elo (yes) akuno? (Does malungo worry you here) 2= Iyayi (no) Skip to 20 Naga ee ligongo chichi? (If yes, explain why)

22b.

1= ndawi ja wula (rainy) 2= ndawi ja lyuwa (hot) 3= ndawi ja mbepo (cold) 4= ndawi jili jose ja chaka (all year around) 1= achakulungwa (adults) Wani mumisi wakusakangala kulwala 2= wanache wasukulu (school malungo? (Who in the community is children) most likely to get malungo) 3= wanache (U-5) (U-5s) 4= achimasyeto wa misigo (pregnant women) 5= waliwose (everybody) 6= nganguyimanyilila (don’t know) 7= wane ligongo (other, specify) 1= elo (yes) Ana malungo gakusausya kwa achimasyeto wana misigo? (Is malaria a 2= Iyayi (no) If no, skip to 23 problem for pregnant women) Naga elo alongosole. (If yes, explain why)

No. 23.

Question Ana imanyilo ya malungo ni chichi?

20.

21.

22a.

Code

Ndawichi ja chaka jagakusasawusya nope malungo? (Which season of year is malungo a problem)

Code 1= kolesya chilu moto (fever) 109

No.

Question (What are the symptoms of malungo)

Code 2= mbepo pa chilu (feeling cold) 3= kutetemera (shivering) 4= kutapika (vomiting) 5= kupweteka ntwe (headache) 6= kukasanya mumalumbiko (joint pains) 7= yine ligongo (other, specify) 8= nganguyimanyilila (don’t know) 1= kukomoka (convulsions) 2=kukolesya nope chilu moto (high fever) 3= kulipuka (unconsciousness) 4= kupweteka ntwe (headache) 5= yine, ligongo (other, specify) 6= nganguyimanyilila (don’t know) 1= kulumidwa niwususu (mosquito bites) 2= kukamula masengo nope (working too hard) 3= usakwa (dirty surroundings) 4= kutama palyuwa (staying in sun) 5= majini (evil spirits) 6= nganguyimanyilila (don’t know) 7= yine, ligongo (other, specify) 1= panado/aspirin 2= fansida 3= ntela wa chiwandu (traditional medicine) 4= nganguyimanyilila (don’t know) 5= yine, ligongo (other, specify) 1= elo (yes) 2= Iyayi (no) Skip to 28a

24.

Ana imanyilo ya malungo gamakulungwa ni chichi? (What are the symptoms of severe malungo)

25.

Achimasyeto wana nsigo mpaka agapate uli malungo? (How can a pregnant woman get malungo)

26.

Wamasyeto wana nsigo naga alwasile malungo ntela chi wandanda wakusamwa? (What is the first choice for treating malungo when a woman is pregnant)

27a.

Ana malungo kwa chimasyeto wana misigo mpaka tugagosye? (Can malungo in pregnancy be prevented) Naga elo alongosole. (If yes, explain how)

27b.

28a.

Ana kwana malungo gamalekangane 1= elo (yes) lekangane? (Are there different kinds of 2= Iyayi (no) Skip to 30. 110

No. 28b.

Question malungo) Naga elo alembe mitundu josope ja malungo. (if yes, list all the types of malungo)

29.

Chiwuyi manye uli kuti ukulwala malungo gantunduchi? (How do know which malungo/fever you are suffering from)

30.

Ukusajembecheya ndawi jantiuli payakutendechegwa niyakupweteka yamalungo nli nganiwe kupochera chikamuchisyo? (How long do you wait between the onset of malungo symptoms and treatment)

Code

1= 1-3 awala (hours) 2= 3-6 awala (hours) 3= 6-12 awala (hours) 4= lisiku limo (24 hours) 5= masikugawiri (2 days) 6= yine (ligongo) (other, specify)

Part B2: Knowledge on fever No. 31a.

31b.

32a.

32b.

33.

34.

Question 1= elo (yes) Ana kwana ine ilwele akuno 2= Iyayi (no) Skip to 34. ikusakolesya chilu moto? (Are there other diseases here, which have fever as a main symptom) Naga kwana alembe mena ga yele ilwele yelekangane lekangane? (If yes, list all the names of these different feverish diseases)

Code

1= elo (yes) Kwele kolela kwachilu moto 2= Iyayi (no) Skip to no 33 kwetindane? (Do these fevers differ from one another) Naga elo alongosole kulekangana kwakwe. (If yes, explain how they differ)

Ana yele ilwele yakolesya chilu moto yi 1= elo (yes), 2= Iyayi (no) kusalekangana ni malungo? (Are these fevers different to malungo) 1= elo (yes) 2= Iyayi (no) Ana kolesya chilu moto chili chimanyilo chakuloka? (Is fever a normal sign in pregnancy)

Part B3: Knowledge on complications in pregnancy and delivery No. 35.

Question Chimanyisyo chachitumbo chikusawaga chichi? (What are the signs of pregnancy)

Code 1=kolesya moto pachilu (fever) 2=kutapika (vomiting) 3=ngalwala mwezi (stopping menstruation) 111

No.

Question

Code 4= kupela pachilu (being tired) 5= yine, ligongo (other specify) 6= nganguyimanyilila (don’t know)

No. 36.

37.

38.

Question Chiwuyi manye uli kuti chitumbo chana yisyausyo? (How do you know that the pregnancy has problem)

1= mwanache ngakwenda (baby is not moving) 2= kwasa miyasi/kapena damu (bleeding) 3= kulwala – lwala pafupi pafupi (on and off ill) 4= yine, ligongo (other specify) 5= nganguyimanyilila (don’t know) 1= kuchelewa kupagwa Chiwuyimanye uli kuti lisiku mwanache (long delivery) lyakuweleka yichisawusya? (How do 2= kwasa miyasi nope (bleeding you know that the delivery has a lot) problems) 3= yine, ligongo (other, specify) 4= nganguyimanyilila (don’t know) Ukuyimanyilila yampakana itendekasye 1= kwasa miyasi (bleeding) kusyausa chitumbo? (Do you know 2= yilwele, ligongo what can cause complications in (diseases, specify) pregnancy) 3= usawi/ malowe (witchcraft) 4= ulwele wakokwewe (prolonged labor) 5= yine, ligongo (other, specify) 6= nganguyimanyilila (don’t know)

SECTION C: No. 39a.

39b.

40a.

Code

ATTITUDES

Question Mpakana tuwe wakhulupirila malangizo 1= ngwiitichisya (agree) pampepe nintela wakupeleka adokotala 2= ngukaichila (not sure) 3= ngingwitichisya (disagree) ligongo welewo walijiganye. (You can Go to 39b trust the advices and medication given by the nurses, because they are educated) Ngakwitichisya ligongochichi. (If disagree, explain)

Wandu akusajawula kwa sing’anga naga ali walojire. (People go to

1= ngwiitichisya (agree) 2= ngukaichila (not sure) 112

Code

No.

40b.

41a.

41b.

42a.

42b.

43a.

43b.

44a.

44b.

45a.

45b.

46a.

Question sing’anga when they are bewitched / 3= ngingwitichisya (disagree) illness is caused by witchcraft) Go to 40b Ngakwitichisya ligongochichi. (If disagree, explain)

1= ngwiitichisya (agree) 2= ngukaichila (not sure) 3= ngingwitichisya (disagree) Go to 41b Ngakwitichisya ligongochichi. (If disagree, explain) Ikusakomboleka kukolesya chilu moto naga uli ukwete msigo. (It is normal to have fever when one is pregnant)

1= ngwiitichisya (agree) 2= ngukaichila (not sure) 3= ngingwitichisya (disagree)Go to 42b. Ngakwitichisya ligongochichi. (If disagree, explain) Achakongwe wamisigo akusalwalalwala malungo. (Malungo is very common disease in pregnancy)

1= ngwiitichisya (agree) Ikusasosekwa kupikanila malangizo 2= ngukaichila (not sure) gakusapeleka anankungwi naga uli 3= ngingwitichisya ukwete msigo. (The advices given by (disagree)Go to 43b. the anankungwi concerning pregnancy have to be obeyed) Ngakwitichisya ligongochichi. (If disagree, explain)

1= ngwiitichisya (agree) IF NO UNCLE, SKIP TO 45A. 2= ngukaichila (not sure) Nkusa wusya lunda kwamjomba wenu 3= ngingwitichisya (disagree) naga nli kulwalika (You ask advice from your maternal uncle when you are Go to 44b. severely sick) Ngakwitichisya ligongochichi. (If disagree, explain)

1= ngwiitichisya (agree) Itumbalala yosope akuno 2= ngukaichila (not sure) ikusatendegwa ligongo lya usawi. 3= ngingwitichisya (disagree) (Most of the maternal deaths here are Go to 45b. caused by witchcraft) Ngakwitichisya ligongochichi. (If disagree, explain)

Ndawi jambone kutanda sikelo nijanti naga nlintandite kupikana yagundagunda m’matumbo. (The correct time to visit ANC first time is when the child has started to move in the womb)

1= ngwiitichisya (agree) 2= ngukaichila (not sure) 3= ngingwitichisya (disagree) Go to 46b.

113

Code

No. 46b.

Question Ngakwitichisya ligongochichi. (If disagree, explain)

47a.

1= ngwiitichisya (agree) Chinkusajaulila ku sikelo nichanti madokotala akampochele chenene naga 2= ngukaichila (not sure) 3= ngingwitichisya mkwete yakusausya pakuweleka. (The (disagree)Go to 47b. main reason to visit ANC at least once is to make sure that if there are complication during delivery the nurses will accept you into the clinic) Ngakwitichisya ligongochichi. (If disagree, explain)

47b.

48a.

48b.

49a.

49b.

Code

1= ngwiitichisya (agree) Azamba akuno ngatukuwakulupilila 2= ngukaichila (not sure) ligongo ndawi syejinji akulepela 3= ngingwitichisya (disagree) kuwelekasya. (The TBA here are not Go to 48b. trusted anymore because they have failed so many times in assisting in the deliveries) Ngakwitichisya ligongochichi. (If disagree, explain)

1= ngwiitichisya (agree) 2= ngukaichila (not sure) 3= ngingwitichisya (disagree) Go to 49b. Ngakwitichisya ligongochichi. (If disagree, explain) Malungo gangakomboleka kugagosya naga uli nimsigo. (Malungo in pregnancy can’t be prevented)

SECTION D: PRACTICES Part D1: Use of ANC services No. 50.

51.

Question Adokotala akusansalila kuti mwesi wambone kutanda sikelo nlimkwete miyesi jilingwa? (Which month are you supposed to start ANC according to the nurses) Ligongo chichi nkusijawula kusikelo? (What is your main reason to go to ANC)

Code 1= 1-3 months 2= 4 months 3= 5 months 4= 6 months or later

1= kukumpima 2= Kulola naga mkwete yakusausya tijyale kuwelechela kuchipatala 3= yine (ligongo) Next question is only for ANC mothers after the examination.

114

No. 52.

Question Ntusalile mtela wimpochele kusikelo soni ni ulwele wake. (Can you explain me what pills you were given just now after examination and what for)

Code

Part D2: Delivery practices No. 53. 54.

55. 56.

57a.

57b.

58. 59a. 59b.

60.

Question Awelechele kalingwa ku nyumba? (How many babies have you delivered at home) If no home deliveries skip to 57a. Akusankamuchisyaga wani? (Who 1= mamawenu (mother) assisted in your home delivery) 2= akwegwe (mother-in-law) 3= angangawakongwe (grandmother) 4= azamba (TBA) 5= pangali (nobody) 6= yine (ligongo) (other, specify) Walijiganyisye yawuzamba? (Was this 1= elo (yes) person skilled/trained) 2= iyai (no) 1= kuweleka kwajangwiye Ligongochi nganimbelecheleje (delivery started suddenly) kuchipatala? (What is the main reason 2= kusowa talansipoti (lack of that you have not delivered at hospital) transport) 3= wansalile kuti welechele kunyumba (adviced to deliver at home) 4= yine, ligongo (other, specify) 1= elo (yes) Pakwete pemwasachile kuwelechela kumusi ninlinlepele nikuja kuchipatala? 2= iyayi (no) If no, skip to 57 (Have you ever started the delivery at home, but then due to problems you have gone to hospital) Naga elo alongosole. (if yes explain what was the reason)

Awelechele kuchipatala kalingwa? (How many babies have you delivered at hospital) Nkusagula kuwelechela kwapi? (Where 1= kumusi (home) do you prefer to deliver) 2= kuchipatala (hospital) Ligongochi nkusinsagulaga kuwelechela kumusi kapena kuchipatala. (Explain why you prefer delivering at home/hospital)

Yasawusyaga yichichi? (What have been the main problems in your deliveries)

1= jwakokwewe kupagwa mwanache (malowe) (baby refused to come out) 115

Code

No.

Question

Code 2= mwasigalile (placenta did not come out) 3= japali damu (bleeding a lot) 4= yine, ligongo (other, specify) 5= has never had problems

Part D3: Treatment seeking practices No. 61.

62.

63.

64.

Question Ulwelechi wawuli wakogoya kumundu jwakwete nsigo? (What is the most dangerous disease a pregnant woman can get here)

Code

1= malungo (malaria) 2= kuugula m`matumbo (diarrhoea) 3= cholera 4= majini (evil spirits) 5= kukamoka (convulsions) 6= yine, ligongo (other, specify) 7= ngingumanyilila (don’t know) 1= wankwenu (husband) Payele yilwele, nlinlwasile 2= njomba (uncle) akusankamuchisyaga wani? (When you are pregnant and you fall ill 3= mamawenu kapena with the disease (you mentioned above), angangawenu (mother/grandmother) from who do ask to advice/help you) 4= azamba (TBA) 5= yine, ligongo (other, specify) IF NO UNCLE, SKIP TO QUESTION. Chakachi chankamuchisye njombawenu nlilwasile nlope ntusalile yiyatendekwe. (Give an example when you last time asked your uncle to help when you severely ill)

Chakachi chimwapite kwasing’anga soni ligongochi? (When was the last time you went to a sing’anga and what was the reason)

116

Acta Tropica 98 (2006) 111–117

The importance of understanding the local context: Women’s perceptions and knowledge concerning malaria in pregnancy in rural Malawi Annika Launiala ∗ , Teija Kulmala School of Public Health, FIN-33014 University of Tampere, Finland Received 21 January 2005; received in revised form 26 October 2005; accepted 13 December 2005 Available online 2 May 2006

Abstract A current problem of malaria prevention programmes is that not enough attention is paid to understanding the local socio-cultural context prior to programme implementation. The aim of this study is to discover how Yao women in rural Malawi understand and explain malaria in pregnancy, how they perceive it and what type of knowledge they have on it. Women’s knowledge of the adverse effects of malaria in pregnancy is also investigated. At first phase a total of 34 in-depth interviews were conducted. At second phase a KAP survey (n = 248) was conducted for cross-validation of the qualitative information. The findings showed that there is neither a vernacular word for malaria nor malaria in pregnancy. Women used a local word, malungo, to refer to malaria. Malungo is an ambiguous disease term because of its multiple meanings which are used interchangeably to refer to many types of feverish illnesses of various causes, not only malaria. Most women did not perceive malungo during pregnancy as a serious illness. There were several other diseases from anaemia, STDs to cholera etc. that were perceived to be more dangerous than malungo. The local meaning of malungo also entailed an assumption that it is a common but fairly harmless illness. Women had limited knowledge of the adverse effects of malaria in pregnancy, the best-known adverse effect being miscarriage (28%, 52/189). A socio-cultural understanding of the implementation context is prerequisite for planning meaningful programmes for the pregnant women in rural Africa. © 2006 Elsevier B.V. All rights reserved. Keywords: Malaria; Prevention; Pregnancy; Malawi; Perceptions; Malungo

1. Introduction More than 30 million women become pregnant every year in malaria-endemic areas of Africa. All these women are exposed to the risk of malarial infection during pregnancy which may cause maternal anaemia, impaired fetal growth, miscarriage, stillbirth, prema-

∗

Corresponding author. Tel.: +358 50 3747069. E-mail address: [email protected] (A. Launiala).

0001-706X/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.actatropica.2005.12.008

ture birth and intrauterine growth restriction (Schulman, 1999; Menendez et al., 2000). Gestational malaria infection has been estimated to cause 75,000–200,000 infant deaths every year in stable malaria transmission areas (Steketee et al., 2001). Today, global initiatives, such as roll back malaria (RBM), have raised malaria in pregnancy as one of the priorities in malaria prevention programmes (AFRO, 2002; Williams and Jones, 2004). The RBM target regarding pregnant women is that 60% of them should have access to preventive measures such as insecticide

112

A. Launiala, T. Kulmala / Acta Tropica 98 (2006) 111–117

treated nets (ITNs) and intermittent preventive treatment (IPT) twice during pregnancy (The Africa Malaria Report, 2003). Social scientific literature focusing on malaria perceptions, treatment-seeking practices and prevention during pregnancy is scarce (Williams and Jones, 2004; see also McCombie, 1996). Major findings have stated that pregnant women perceive malaria as a common problem, yet the use of IPT during pregnancy is rather low (Helitzer-Allen et al., 1994; Schultz et al., 1994; Massele et al., 1997; Ndyomugyenyi et al., 1998; Ashwood-Smith et al., 2001). Women’s perceptions and knowledge about the threat of malaria in pregnancy have varied from acknowledged danger (Kengeya-Kayondo et al., 1994; Nuwaha, 2002; Comoro et al., 2003) to contradictory beliefs (Dulhunty et al., 2000) and no problem in pregnancy (Helitzer-Allen et al., 1993; Winch et al., 1996). It has been recognised that the success of worldwide malaria control depends on paying enough attention to the importance of socio-cultural factors in local treatment and prevention practices (Heggenhougen et al., 2003). Thus it is crucial to examine the local context before implementing preventive measures and to gain a thorough understanding of what the women consider dangerous for their pregnancies. It is also crucial to review the local malaria problem in relation to the local understanding and meaning of the disease. This article reports the results of descriptive social science research focusing on socio-cultural issues affecting the treatment and prevention of malaria in pregnancy among Yao women in rural Malawi. It aims to explain the local meaning of malaria, how women perceive and understand malaria in relation to pregnancy and what type of knowledge they have regarding malaria in pregnancy. The research is based on previous studies carried out in this area (Kulmala, 2000; UNICEF, 2001). 2. Subjects and methods The Ministry of Health and Population (MOHP) in Malawi granted ethical approval and a research permit, and data collection was carried out between September and December 2002 in rural Lungwena, Mangochi District on the eastern shore of Lake Malawi. The main ethnic group in the area is Yao, who are matrilineal and practice Islamic religion. The main occupations are farming and fishing (Kulmala et al., 2000). The catchment area consists of about 30,000 people, 6958 households and 25 villages. The area is divided into four traditional administrative divisions, called gulu in local language, and they are run by a chief of a gulu (Mitchell, 1956).

Data presented in this article was collected in two phases with triangulation of data sources to crossvalidate the collected information and to check the reliability of responses (Helitzer-Allen and Kendall, 1992; Agyepong et al., 1995; Bhattacharyya, 1997). Phase I was carried out between September and November, and phase II in December. Data collection methods used in phase I were in-depth interviews (IDIs) with women in reproductive age (verified by last born <1.5 years and still menstruating) and key informant interviews with four traditional advisers (anankungwi), two traditional birth attendants (TBA), one traditional healer and two men. Participants for IDIs were selected from eight villages (two villages from each gulu) by using convenience sampling (Agyepong et al., 1995; Bernard, 2002). Interviewees were selected in collaboration with gulu chiefs following specific selection criteria (reproductive age 15–44 years, not a relative of the chief, and women with single pregnancy or with many pregnancies). The principal author conducted the interviews with help of a research assistant who acted as a translator. All interviews were translated and transcribed by another assistant who was not present at the interview. Fifteen percent of the interviews were randomly checked for accuracy of translation. All interviews began with broad unstructured and non-directive questions about pregnancy, delivery and health of the pregnant women and gradually focused towards malaria in pregnancy, its treatment and prevention with more specific themes. The purpose was to gain a proper understanding of the socio-cultural context of managing pregnancy and malaria. Interviews ceased with the respondents when no new data emerged, i.e. saturation was reached (Bernard, 2002; Agyepong et al., 1995). A total of 34 women were interviewed. Phase II consisted of a cross-sectional KAP survey. The questionnaire was designed by the principle author based on the main findings from phase I. It contained 64 questions (12 open-ended) concerning the same issues as had been discussed or which had emerged from indepth interviews. The questionnaire was translated from English to Chiyao and back to English to ensure the accuracy of translation and it was pre-tested before actual use. The catchment area was divided into four geographical areas for logistical reasons, of which the Lungwena Trading Centre was one area. Interviews with women in reproductive age were organised at the antenatal clinic (ANC) of the Lungwena Health Centre and in two villages of the three other geographical areas each. In the six villages, altogether 200 interviews were aimed at, with samples proportional to the size of the village, which implied that a woman in every 8th household was inter-

A. Launiala, T. Kulmala / Acta Tropica 98 (2006) 111–117

viewed in each village. The first household was selected randomly from the Health Centre register and determined the starting point for the interviews. Thereafter, a woman in every 8th household along the path from the index house was systematically interviewed. If there was no success in the sampled household, a woman in the next household on the path was interviewed instead. At the ANC, every third woman was selected over 4 days. An exclusion criterion for survey respondents was participation in the IDIs. This sampling procedure resulted in 248 interviews, 200 in the villages and 48 at the Health Centre. In-depth interviews were analysed by using an ethnographic approach for identifying categories and concepts from the data as they emerged. Analysis was made by looking for similarities, differences, variations and contradictions. Local meaning of malaria and its types were analysed by developing disease taxonomies (Bernard, 2002). The Atlas.ti computer programme for Windows version 5.0 (Scientific Software Development GmbH) was used to sort the qualitative data after code creation. Survey data were described as frequency distributions in relation to the qualitative findings. No further statistical analysis was undertaken at this point because of the qualitative nature of the study. 3. Results 3.1. Background characteristics of respondents The background characteristics of the respondents are presented in Table 1. All women were in reproductive age, between 15 and 44 years. Those women who did not know their actual ages were asked specific questions to verify that they were in reproductive age (see Section 2). Most of the respondents were married, and the median number of previous pregnancies was four among IDI respondents and three among survey respondents. The majority of the women had no education and 23% of survey respondents were able to read. 3.2. Local meaning of malaria The women used a local word malungo to refer to malaria and to the fever caused by malaria. Moreover, it was used as a synonym for body pains and feeling unwell. Literally translated, malungo means joint pain. It was described as a term covering several feverish illnesses, varying in symptoms, severity and aetiology. The women identified two types of malungo, which were known to all of them, namely, malungo caused by mosquitoes and malungo caused by hard work. The

113

Table 1 Background characteristics of respondents in in-depth interviews (IDIs) and KAP survey in rural Malawi Background characteristics

IDIs number of participants (34)

KAP survey number of participants (248)

Age (years) 15–19 20–24 25–29 30–34 35–39 40–44 Age unknown Missing data

1 (3%) 8 (24%) 4 (12%) 2 (6%) 1 (3%) 1 (3%) 17 (50%)

18 (7%) 58 (24%) 46 (19%) 27 (11%) 13 (5%) 3 (1%) 82 (33%) 1

Marital status Married Co-habiting Missing data

30 (88%) 13 (38 %)

215 (87%) 125 (51%) 1

4 (1, 6)

3 (1, 6)

Parity Median (min, max) number of previous pregnancies Number of primigravid women Pregnant at the time of interview Education No education Primary education Secondary education Literate (tested)

3

16

8

70

21 (62%) 9 (26%) 4 (12%) n/a

160 (65%) 49 (20%) 39 (16%) 56 (23%)

women did not recognise any malungo type with specific references to pregnancy. They also described many other types of malungo according to their causes and symptoms as follows: “There is malungo caused by hard work, and malungo that causes body pains and stiff neck”. However, not all the women knew all the different types. The different types mentioned were malungo with body pains, malungo with vomiting and diarrhoea, malungo with joint pains and feeling cold, malungo with headache, malungo with convulsions, malungo with stiff neck, and malungo with shortage of blood. Further, many women made a distinction between normal malungo and severe malungo (malungo gamakulungwa). Sixtysix percent (160/241) of survey respondents said that there is only one type of malungo, which is the normal malungo. The connection between evil spirits (majini) and malaria was also investigated during the in-depth interviews. Most women knew that a majini (evil spirit) causes illnesses, but they argued that there is no such malungo type as malungo wa majini (malungo caused by evil spirits).

114

A. Launiala, T. Kulmala / Acta Tropica 98 (2006) 111–117

3.3. Perception and knowledge about malaria in pregnancy

Table 3 Known adverse effects of malungo in pregnancy among KAP survey respondents (n = 189) in rural Malawi

The women did not regard malungo in pregnancy as a dangerous disease unless intentionally probed during the in-depth interviews, and even then some women perceived a number of other diseases more dangerous than malungo, as was, for example, described by one woman: “It is dangerous if you suffer often from malungo, but STDs are also very dangerous because of the infection that can lead to miscarriage”. Diseases considered common and more dangerous than malungo were anaemia, diarrhoea, STDs, AIDS, cholera and kambanga (convulsions due to severe malaria) etc. Forty-eight percent of the survey respondents (118/247) nevertheless perceived malungo as the most dangerous disease for pregnant women. The women explained that malungo caused by hard work is common throughout the year, but malungo caused by mosquitoes is common in the rainy season due to the abundance of mosquitoes. “The most worrisome malungo is the one caused by mosquitoes. This malungo and diarrhoea we suffer a lot during rainy season. We are bitten by so many mosquitoes. There are also other malungo types like malungo caused by hard work, this you can get any time of the year”. Thirty-six percent of the survey respondents (85/239) considered malungo as a year-round problem. Twenty-four percent (58/241) perceived pregnant women to be the biggest risk group, but around 58% (140/241) perceived children under 5 years as the biggest risk group. Eighty-eight percent of the survey respondents (213/241) worried about malungo. The reasons for worry are described in Table 2, showing that pregnancy-related reasons are almost non-existent. The same respondents were asked if they considered malungo in pregnancy as

Known adverse effects of malaria in pregnancy

Table 2 Reasons for worrying about malungo given by KAP survey respondents (n = 213) in rural Malawi Reasons for worrying about malungo Causes death (children) Causes body pains Disability to work It is a dangerous disease It is a common disease Causes shortage of blood Causes convulsions in children It is a painful disease Causes miscarriage Because of the pregnancy Other

n 70 (33%) 25 (12%) 25 (12%) 22 (10%) 9 (4%) 8 (4%) 5 (2%) 11 (5%) 4 (2%) 1(0.5%) 33 (15%)

Causes miscarriage Causes maternal death Causes shortage of blood (anaemia) Causes weakness Because of the pregnancy Causes premature delivery It is common in pregnancy Causes body pains Disability to work Causes convulsions during labour It is a dangerous disease It is a painful disease Other

n 52 (28%) 23 (12%) 20 (11%) 14 (7%) 13 (7%) 11 (6%) 11 (6%) 9 (5%) 6 (3%) 2 (1%) 1(0.5%) 1(0.5%) 36 (19%)

a problem. Seventy-eight percent agreed (189/241) that it was a problem, but they were unable to give a detailed description of the adverse effects of malungo in pregnancy (Table 3). The aetiology of malungo was also studied. In the survey, 56% of the women (135/241) said that malungo is caused by mosquitoes. Thirty-six percent of women (87/241) said that they did not know what causes it. The in-depth interviews allowed the women to elaborate their ideas of the causal factors. “If it is the rainy season, malungo is caused by mosquitoes because rotten garbage causes problems. In the rainy season, mosquitoes bite you and they give you different diseases making you feel uncomfortable. You can also get malungo after working hard in the maize field. In the hot season, you can get malungo from overheating in the sun”. A general opinion in the communities was that malungo is a natural illness. The women described it as an illness that just happens. Some said that it is an illness from God. The women did not associate witchcraft with normal malungo although witches (afiti) pose a threat in everyday life, regularly causing all kinds of misfortunes and illnesses. For example, kambanga (convulsions due to severe malaria), a local children’s illness, was said to be caused by witchcraft. Furthermore, pregnant women were not considered more susceptible to mosquito bites nor did they report that the aetiology of malungo changed when a woman is pregnant. However, women perceived pregnancy as a sensitive condition and pregnant women more exposed to illnesses caused by witchcraft than the general population. One respondent described: “When you are pregnant you are advised not to talk loudly about it. Your enemies in the village can use witchcraft to cause harm to your unborn

A. Launiala, T. Kulmala / Acta Tropica 98 (2006) 111–117

baby. Because of the pregnancy you are vulnerable to ill will.” The women were asked to list symptoms of malungo. Ninety-five percent of the survey respondents (230/241) mentioned one to three symptoms of malungo. Those most frequently emerging were feeling cold, joint pain and fever. For severe malungo the women listed less symptoms than for malungo. Thirty-one percent of the survey respondents (75/241) knew of no symptoms for severe malungo, and those mentioned included convulsions, high fever, unconsciousness and headache. Women did not rank fever as the most important symptom of malungo. In fact, the in-depth interviews revealed that a mild fever during pregnancy was considered normal and some even interpreted fever as one sign of pregnancy (kuloka). “When someone is pregnant, fever becomes the first visible sign of the pregnancy. Some say that malungo is a sign of the beginning of pregnancy, but some say it is just a natural disease”. 4. Discussion As a disease term, malungo is ambiguous on account of its multiple meanings and definitions, which are used interchangeably to refer to many types of feverish illnesses, not just malaria. This finding conforms to other research findings in the region (Agyepong, 1992; Helitzer-Allen et al., 1993; Kengeya-Kayondo et al., 1994; Bisika, 1996; Winch et al., 1996; Launiala and Raijas-Walch, 1999; Espino et al., 1997; UNICEF, 2001). Further, more than 10 malungo types were discovered, which is comparable to earlier research among the Yao. Contrary to previous findings, however, the women in this study did not recognise malungo wa majini (malaria caused by evil spirits) as one malungo type (Helitzer-Allen et al., 1993). Despite the great number of varying malungo types presented, a specific malungo type related to pregnancy did not exist among the Yao women. Further, the meaning of malungo neither changed during pregnancy, nor was there a vernacular word used for malaria in pregnancy. Our results suggest that, overall, the local malungo categories are very vague, ambiguous and not shared by all members of the community. Categories are produced and reproduced in encounters with local people as argued by Pool in his studies in Cameroon (1994). There is a direct association between women’s knowledge about the causes and symptoms of malungo and the type of malungo; the more causes and symptoms a woman knows, the more types she is able to list. Thus, the local illness categories should not be automatically generalized to concern a particular ethnic group or other

115

ethnic groups in the same country, although similarities are found (see also Nichter, 1994). It is difficult to compare the results regarding perceptions and knowledge of malaria in pregnancy in the region because of the small number of pre-existing studies, which moreover differ in focus and methodology (Kengeya-Kayondo et al., 1994; Schultz et al., 1994; Comoro et al., 2003; Dulhunty et al., 2000; Winch et al., 1996; Helitzer-Allen et al., 1993). The findings from this study indicate that malungo is such a common disease as not to be considered a major problem, even during pregnancy. The women agreed that malungo is a worrisome illness, but not specifically in relation to pregnancy. Pregnancy-related concerns were expressed mainly when reference was made explicitly to malaria in pregnancy. Fever during pregnancy was likewise perceived normal by some women and a sign of pregnancy. As hard work in the maize field is likely to raise the body temperature, it might explain why some women considered fever normal in pregnancy. Our study results indicate that although women say that they worry about malaria and perceive it as a problem, it does not necessarily mean that they perceive malaria as dangerous and a threat to their pregnancy. Women’s knowledge regarding the aetiology, symptoms and adverse effects of malaria in pregnancy varied. About half of the respondents said that mosquitoes transmit malaria. Most women mentioned one to three symptoms of malaria, and zero to one symptoms of severe malaria. Also, one third of the women linked miscarriage to malaria, and a few linked malaria with low birth weight, stillbirth, anaemia, premature birth etc. Women’s low educational level in the study area, which is below women’s national average (31.1%) in rural areas, offers some explanation for women’s understanding of malaria (National Statistics Office (Malawi) and ORC Macro, 2001). In conclusion, the results from this study clearly demonstrate that the local meaning of malungo contains an assumption that malaria is a common and fairly harmless disease for pregnant women. Moreover, the women have little detailed knowledge on the adverse consequences of malaria in pregnancy. It is, therefore, a major challenge for the malaria prevention programmes to improve their knowledge about the local understanding of malaria. These programmes should also focus on malaria in a wider context, in order to gain a perspective of which issues women consider dangerous for pregnancy, and of why and how malaria is related to these issues. Focusing on malaria in relation to other pregnancy problems will give insight into the multiple challenges women face in the local context and how they cope with these challenges. For example, one possible

116

A. Launiala, T. Kulmala / Acta Tropica 98 (2006) 111–117

entry point for prevention programmes is to raise the issue of miscarriage and its causes, and to develop messages that stress the linkage between malaria and risk of miscarriage. It should be born in mind that these illiterate women do not comprehend complicated medical terminology yet they are a continuous target of all kinds of health education messages. Furthermore, knowledge and awareness do not necessarily entail changes in behaviour and practices (Farmer, 1997; Nations and Monte, 1997). Only if a proper understanding of the implementation context is achieved, can meaningful health programmes for pregnant women in rural Africa be planned. Acknowledgements Our sincere thanks are expressed to all the women in the Lungwena area who participated in this research. In addition, a warm thank you belongs to the Lungwena Health Centre staff and the hardworking research assistants: Rashid Osman, Eunice Willy, Misontzie Tweya, Shaibu Msosa, Ben Mwema, Gertrud Moses and Zacharia Abdul. Valuable institutional support was provided by the Finnish Family Federation and the Nordic Institute of African Studies (NAI) provided funding for the fieldwork. Doctoral Programs in Public Health (DPPH) of the University of Tampere deserve special thanks for providing funding that made it possible to write this article. And finally, our warmest thanks to Prof. Matti Hakama regarding his valuable advice in the field of epidemiology, and Heini Huhtala and AnnaMaija Kivisto for their statistical advice, all from the University of Tampere, Finland. References AFRO, 2002. Strategic framework for malaria control during pregnancy in the WHO Africa region, November 1. Agyepong, I., 1992. Malaria: ethnomedical perceptions and practice in an Adangbe farming community and implications for control. Soc. Sci. Med. 5, 131–137. Agyepong, I., Aryee, B., Dzikunu, H., Manderson, L., 1995. The malaria manual. Guidelines for the rapid assessment of social, economic and cultural aspects of malaria. Methods in tropical diseases, no. 2, UNDP/World Bank/WHO Special programme for Research and Training in Tropical Disease (TDR), TDR/SER/MSR/95.1. Ashwood-Smith, H., Coombes, Y., Kaimila, N., Bokosi, M., Lungu, K., 2001. Availability and use of sulphadoxine–pyrimethamine (SP) in pregnancy in Blantyre district. A safe motherhood and Blantyre Integrated Malaria Initiative (BIMI) Joint Survey. Malawi Med. J. 14, 8–11. Bernard, H.R., 2002. Research methods in anthropology. In: Qualitative and Quantitative Approaches, 3rd ed. Altamira Press, England. Bhattacharyya, K., 1997. Key informants, pile sorts, or surveys? Comparing behavioural research methods for the study of acute respiratory infections in West Bengal. In: Inhorn, M.C., Brown, P.J. (Eds.),

The Anthropology of Infectious Diseases. Theory and Practice on Medical Anthropology and International Health, vol. 4. Gordon and Breach Publishers, The Netherlands, pp. 211–238. Bisika, T.J., 1996. Malaria Case Management in Children in Malawi: the Case of Namasalima. A Report of a Rapid Ethnographic Study. Centre for Social Research, University of Malawi. Comoro, C., Nsimba, S.E.D., Warsame, M., Tomson, G., 2003. Local understanding, perceptions and reported practices of mothers/guardians and health workers on childhood malaria in a Tanzanian district—implications for malaria control. Acta Trop. 87, 305–313. Dulhunty, J.M., Yohannes, K., Kourleoutov, C., Manuopangai, V.T., Polyn, M.K., Parks, W.J., Bryan, J.H., 2000. Malaria control in central Malaita, Solomon Islands. 2. Local perceptions of the disease and practices for its treatment and prevention. Acta Trop. 75, 185–196. Espino, F., Manderson, L., Acuin, C., Domingo, F., Ventura, E., 1997. Perceptions of malaria in a low endemic area in the Philippines: transmission and prevention of disease. Acta Trop. 63, 221–239. Farmer, P., 1997. Ethnography, social analysis, and prevention of sexually transmitted HIV infection among poor women in Haiti. In: Inhorn, M.C., Brown, P.J. (Eds.), The Anthropology of Infectious Diseases. Theory and Practice on Medical Anthropology and International Health, vol. 4. Gordon and Breach Publishers, The Netherlands, pp. 413–438. Heggenhougen, K., Hackethal, V., Vivek, P., 2003. The behavioural and social aspects of malaria and its control: an introduction and annotated bibliography. Geneva, UNDP/World Bank/WHO Special Programme for Research and Training in Tropical Diseases (TDR), (TDR/STR/SEB/VOL/03.1). Helitzer-Allen, D.L., Kendall, C., 1992. Explaining differences between qualitative and quantitative data: a study of chemoprophylaxis during pregnancy. Health Educ. Quart. 19 (1), 41–54. Helitzer-Allen, D.L., Kendall, C., Wirima, J.J., 1993. The role of ethnographic research in malaria control: an example of Malawi. Res. Sociol. Health Care 10, 269–286. Helitzer-Allen, D.L., Macheso, A., Wirima, J., Kendall, C., 1994. Testing strategies to increase use of chloroquine chemoprophylaxis during pregnancy in Malawi. Acta Trop. 58, 255–266. Kengeya-Kayondo, J.F., Seeley, J.A., Kajura-Bajenja, E., Kabunga, E., Mubiru, E., Sembajja, F., Mulder, D.W., 1994. Recognition, treatment seeking behaviour and perception of cause of malaria among the rural women in Uganda. Acta Trop. 58, 267–273. Kulmala, T., 2000. Maternal Health and Pregnancy Outcomes in Rural Malawi. Acta Universitatis Tamperensis 870, University of Tampere. Kulmala, T., Vaahtera, M., Ndekha, M., Cullinan, T., Salin, M.-L., Koivisto, A.-M., Ashorn, P., 2000. Socio-economic support for good health in rural Malawi. E. Afr. Med. J. 77, 168–171. Launiala, A., Raijas-Walch, H., 1999. Biomedical and local concepts of malaria among the Baganda in Uganda. J. Finn. Anthropol. Soc. 4, 142–153. Massele, A.Y., Mpundu, M.N., Hamudu, N.A.S., 1997. Utilization of antimalarial drugs by pregnant women attending the antenatal clinic at Muhimbili Medical Centre, Dar es Salaam. E. Afr. Med. J. 74, 28–30. McCombie, S.C., 1996. Treatment seeking for malaria: a review of recent research. Soc. Sci. Med. 43, 945–993. Menendez, C., Ordi, J., Ismail, M.R., Ventura, P.J., Aponte, J.J., Kahigwa, E., Font, F., Alonso, P.L., 2000. The impact of placental malaria on gestational age and birth weight. J. Infect. Dis. 181, 1740–1745.

A. Launiala, T. Kulmala / Acta Tropica 98 (2006) 111–117 Mitchell, J.C., 1956. The Yao village. In: A Study in the Social Structure of a Malawian People. Manchester University Press, UK. National Statistics Office (Malawi) ORC Macro, 2001. Malawi Demographic and National Survey 2000. National Statistic Office and ORC Macro, Zomba, Malawi and Calverton, Maryland, USA. Nations, M.K., Monte, G.C., 1997. I’m not dog, No!; Cries of resistance against cholera control campaigns. In: Inhorn, M.C., Brown, P.J. (Eds.), The Anthropology of Infectious Diseases. Theory and Practice on Medical Anthropology and International Health, vol. 4. Gordon and Breach Publishers, The Netherlands, pp. 439– 481. Ndyomugyenyi, R., Neema, S., Magnussen, P., 1998. The use of formal and informal services for antenatal care and malaria treatment in rural Uganda. Research report. Health Policy Plann. 13, 94– 102. Nichter, M., 1994. Illness Semantics and International Health: the weak lungs–tuberculosis complex in the Philippines. Soc. Sci. Med. 38, 649–663. Nuwaha, F., 2002. People’s perceptions of malaria in Mbarara, Uganda. Trop. Med. Int. Health 7, 462–470. Pool, R., 1994. Dialogue and the Interpretation of Illness. Conversation in a Cameroon Village. Berg Publishers, USA.

117

Schulman, C.E., 1999. Intermittent sulphadoxine–pyrimethamine to prevent severe anemia secondary to malaria in pregnancy: a randomized placebo-controlled trial. Lancet 353, 632–636. Schultz, L.J., Steketee, R.W., Chitsulo, L., Macheso, A., Nyasulu, Y., Ettling, M., 1994. Malaria and childbearing women in Malawi: knowledge, attitudes and practices. Trop. Med. Parasitol. 45, 65–69. Steketee, R.W., Nahlen, B.C., Parise, M.E., Menendez, C., 2001. The burden of malaria in pregnancy in malaria-endemic areas. Am. J. Trop. Med. Hyg. 64 (Suppl. 1/2), 28–35. The Africa Malaria Report, 2003. World Health Organization and UNICEF, WHO/CDS/MAL/2003.1093. UNICEF, 2001. Ethnographic Malaria Study. Research Report. UNICEF, Malawi. Williams, H.A., Jones, C.O.H., 2004. A critical review of behavioural issues related to malaria control in sub-Saharan Africa: what contributions have social scientists made? Soc. Sci. Med. 59, 501–523. Winch, P.J., Makemba, A.M., Kamazima, S.R., Lurie, M., Lwihula, G.K., Premji, Z., Minjas, J.N., Shiff, C.J., 1996. Local terminology for febrile illnesses in Bagamoyo district, Tanzania and its impact on the design of a community-based malaria control programme. Soc. Sci. Med. 42, 1057–1067.

Transactions of the Royal Society of Tropical Medicine and Hygiene (2007) 101, 980—989

available at www.sciencedirect.com

journal homepage: www.elsevierhealth.com/journals/trst

Ethnographic study of factors influencing compliance to intermittent preventive treatment of malaria during pregnancy among Yao women in rural Malawi Annika Launiala a,∗, Marja-Liisa Honkasalo b a b

School of Public Health, FIN-33014 University of Tampere, Tampere, Finland Helsinki Collegium for Advanced Studies, P.O. Box 4, FIN-00014 University of Helsinki, Helsinki, Finland

Received 9 June 2006; received in revised form 12 April 2007; accepted 12 April 2007 Available online 20 July 2007

KEYWORDS Malaria; Pregnancy; Intermittent preventive treatment; Ethnography; Compliance; Malawi

Summary In Africa today one of the main strategies to reduce malaria infection during pregnancy is the promotion of intermittent preventive treatment (IPT). To date only a few studies have investigated the factors affecting compliance to IPT. This medical anthropology study aims to describe these factors from the perspective of pregnant women in rural Malawi. We examine women’s knowledge and perceptions about the use of medication in pregnancy and the timing and motivation concerning use of antenatal clinic (ANC) services. In addition, the circumstances and interaction at the ANC and the IPT implementation process are described. The data were collected by applying an ethnographic approach, including focus group discussions (n = 8), in-depth interviews (n = 34), drug identification exercises, participant observation and a ‘knowledge, attitudes and practices’ survey (n = 248). This study discovered several factors affecting IPT. These were: unclear messages about IPT with sulfadoxine—pyrimethamine (SP) from nurses; timing of SP-1; periodic shortages of SP; women’s limited understanding of IPTSP; tendency for late enrolment; and nurses’ underperformance. The results of this study show that understanding of the multiple contexts affecting malaria prevention is important, and that ethnographic research is useful for discovering and solving problems beyond the scope of many other research approaches. © 2007 Royal Society of Tropical Medicine and Hygiene. Published by Elsevier Ltd. All rights reserved.

1. Introduction In Africa today, as part of the Roll Back Malaria (RBM) Initiative, one of the main strategies to reduce malaria infection

∗

Corresponding author. Present address: Keh¨ akatu 52, 70600 Kuopio, Finland. Tel.: +358 50 3747069. E-mail address: annika.launiala@uta.fi (A. Launiala).

during pregnancy is the promotion of intermittent preventive treatment (IPT) with sulfadoxine—pyrimethamine (SP) during a routine antenatal clinic (ANC) visit. Although the need for understanding of the socio-cultural and economic factors affecting treatment and prevention of malaria in pregnancy has been recognised, the factors contributing towards compliance to IPT and good coverage rates have remained largely unidentified (Hill and Kazembe, 2006; Williams and Jones, 2004; WHO/UNICEF, 2005). To date,

0035-9203/$ — see front matter © 2007 Royal Society of Tropical Medicine and Hygiene. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.trstmh.2007.04.005

Prevention of malaria during pregnancy in rural Malawi there have been only some studies on factors influencing compliance with IPT (Ashwood-Smith et al., 2002; HelitzerAllen and Kendall, 1992; Massele et al., 1997; Schultz et al., 1994; Van Eijk et al., 2004). Results reported in this article are part of a broader medical anthropology study exploring the socio-cultural factors affecting treatment and prevention of malaria in pregnancy in rural Malawi (Launiala, unpublished doctoral study plan, 2001). The aim of this article is to describe the factors that influence compliance to IPT-SP in a rural Malawian health centre by providing a detailed and contextual account of local Yao women’s knowledge and perceptions of the use of medication in pregnancy, and timing and motivation concerning the use of ANC services. In addition, the circumstances and interaction at the ANC and the IPT implementation process are described. Compliance is a complex issue warranting exploration from diverse perspectives in order to be thoroughly understood (Farmer, 1997; Farmer et al., 2006; Humphery et al., 2001; Sumartojo, 1993; Trostle, 2005). Traditionally it has been investigated from the perspective of the doctors and health system, assuming that failure to comply is due to patients’ deviant behaviour, ignorance, misunderstanding, lack of motivation and responsibility (Conrad, 1985; Donovan and Blake, 1992; Homedes and Ugalde, 1993; Trostle, 1988). Thus, in order to provide new insights, we conducted ethnographic research, which permits the production of rich, detailed and contextual data for understanding IPT, particularly from the perspective of pregnant women in this case, but also the description of the implementation process and context. We applied a multiple method design, which included focus group discussions (FGDs), in-depth interviews (IDIs), drug identification exercises, participant observation and a ‘knowledge, attitudes and practices’ (KAP) survey. For understanding the women’s perspective, we conducted interviews and drug identification exercises in the community. We used the following research questions: What type of knowledge and perceptions do women have concerning medication during pregnancy? How do they perceive the use of SP during pregnancy and do they have knowledge of IPT-SP? What is the average time of starting ANC visits, and does this affect implementation of IPT-SP? What motivates women to attend the ANC — is it the IPT-SP? For describing the ANC context, we conducted participant observation in order to answer the following research questions: What actually happens at the ANC? Do women receive SP, as recommended by the policy guidelines? How do nurses and pregnant women communicate? Do nurses provide clear and appropriate health education on malaria prevention?

981 matrilineal, with monogamous and polygamous marriages (Launiala and Kulmala, 2006; Thorold, 1993). Health services are delivered by a government health centre, staffed by a clinical officer and three nurses. It incorporates the ANC and also serves as a training centre for the University of Malawi College of Medicine. Health care is supported by regular outreach clinics and a network of health surveillance assistants (HSAs) and traditional birth attendants (TBAs). The catchment area contains about 30 000 people. Approximately 95% of the women attend the ANC at least once during their pregnancy (Lungwena demographic and health survey, 2000, unpublished) (National Statistics Office (Malawi) and ORC Macro, 2001). From the early 1990s, there have been non-governmental organisation projects and clinical trials on child and maternity health. The area differs little from other rural areas in the country in terms of socio-economic details and health services (Benson et al., 2002; National Statistics Office (Malawi) and ORC Macro, 2001). Malawi has a national malaria policy that was developed in 2002 and includes an IPT policy. According to the policy, all pregnant women should receive two treatment doses of SP at least one month apart in weeks 20—34 of pregnancy at the antenatal clinic and at TBA venues under direct observed therapy (DOT). The latest national demographic and health survey estimated that 65.4% of pregnant women received one dose of SP and 28.9% received two doses during their last pregnancy in rural areas (National Statistics Office (Malawi) and ORC Macro, 2001). A more recent survey estimated an improvement in IPT coverage, stating that about 60% of pregnant women received two or more SP doses during their last pregnancy. There were, however, clear variations between the districts (Kadzandira and Munthali, 2004, unpublished report). Moreover, the insufficient description of the survey protocol suggests cautious interpretation of the results.

2.2. The study plan

2. Materials and methods

The overall aim of the study was two-fold: (1) to acquire adequate understanding of the research context through exploration of local knowledge and socio-cultural practices related to pregnancy and delivery from the Yao community viewpoint; and (2) to enhance understanding of the factors influencing treatment and prevention of malaria in pregnancy relative to the ongoing malaria prevention activities in Malawi (Launiala, Doctoral Programs in Public Health, University of Tampere). The study plan is based on previous research (UNICEF, unpublished) conducted in Malawi by the principal author (PA) and performed in collaboration with the National Malaria Control Programme, the District Health Officer and the malaria coordinator in Mangochi District.

2.1. The setting

2.3. Data collection

The study was conducted in rural Lungwena, Mangochi District on the eastern shore of Lake Malawi, Malawi. The main ethnic group in the area is the Yao, who speak Chiyao. The majority of the Yao practice Islam, which over time has been adjusted to accommodate the local customs. The Yao are

Ethnographic fieldwork was conducted between September and December 2002 in two phases, with triangulation of data sources to cross-validate the collected information and to check reliability of responses (Agyepong et al., 1995; Bhattacharyya, 1997; Helitzer-Allen and Kendall, 1992) (see

982

A. Launiala, M.-L. Honkasalo

Figure 1 Triangulation of data sources. ANC: antenatal clinic; SP: sulfadoxine—pyrimethamine; FGD: focus group discussion; IPT: intermittent preventive treatment; KAP: ‘knowledge, attitudes and practices’.

Figure 1). During the fieldwork the PA was the principal data collector and resident at the Lungwena Health Centre. In Phase I, the methods used were FGDs and IDIs, which included drug identification exercises with women of reproductive age and participant observation at the ANC. During interviews the women were shown common drugs administered at the health centre, and they were asked to identify them and say what they know about them. Interviews were conducted in eight villages by using convenience sampling. Informed consents were obtained verbally and tape-recorded before the interviews and participant observation. The interviews were translated and transcribed, and 15% were randomly checked for accuracy of translation. IDIs ceased when no new data emerged, resulting in a total of 34 interviews. Participant observation was conducted at the ANC during 10 conveniently chosen days. Health education sessions held by two nurses were tape-recorded during the observations and translated. In Phase II, a cross-sectional KAP survey with women of reproductive age was carried out. The questionnaire was designed around the main findings from Phase I. The questionnaire was translated twice to ensure accuracy of translation and pre-tested before actual use. Informed consents were obtained verbally and marked on the questionnaires. For logistical reasons, interviews were organised at the ANC and in six randomly chosen villages. A total of 200 survey interviews were aimed at the villages, with samples proportional to the village size. The first household in each village was selected randomly from the Health Centre register and served as the starting point. At the ANC, every third woman was selected over a 4 day period. An exclusion criterion for survey respondents was participation in the FGDs and IDIs in Phase I. This sampling procedure resulted in 248 survey interviews, 200 in the villages and 48 at the Health Centre (Launiala and Kulmala, 2006).

FGDs and IDIs were analysed using an ethnographic approach for identifying categories and concepts from the data as they emerged. Analysis was performed by looking for similarities, differences, variations and contradictions (Bernard, 2002). Survey data were described as frequency distributions in relation to the qualitative findings.

3. Results 3.1. Knowledge and perceptions about the use of medication during pregnancy During the IDIs and FGDs, we inquired whether there was any medication pregnant women should not take. The women were unanimous that all bitter-tasting drugs should be avoided, because they are known to cause miscarriage. Traditional medicine, such as herbs and roots, as well as chloroquine, quinine and penicillin were classified as prohibited medication due to bitterness. However, traditional medicine was sometimes used to prevent miscarriage, for example, as one woman explained, ‘‘Some people get traditional medicine when their pregnancy is not settling. This is called kutenganyika chitumbo (‘one is about to miscarry’)’’. Drugs in capsules were also considered dangerous, with a power to cause miscarriage. Sometimes capsules and bitter traditional medicine were also used to abort unwanted pregnancies. The following excerpt from an FGD illustrates women’s knowledge of drugs (M = moderator; R = respondent): M: Here are some of those pills you receive when you attend the ANC [the women are shown SP, panado (paracetamol), iron tablets, tetracycline capsules]. Please tell me what pills they are and what for.

Prevention of malaria during pregnancy in rural Malawi R1: This is fansidar [the woman picks up two fansidar (SP) and one panado pill]. The nurses instruct us to take them home, one in the morning and one in the evening. They are vitamins. R2: Fansidar is for prevention of malungo (malaria), for the mother and newborn baby. R3: Fansidar is only taken when you are ill. R4: These are capsules [the woman picks up a tetracycline capsule]. M: Can a pregnant woman take capsules? R1: Yes, you take if a doctor gives it to you. R2: No, you can’t and I am talking from experience. I miscarried when I took capsules. A msungu (white man) came from Lungwena health centre to the school and he gave me capsules. I was three months pregnant and I miscarried at home. I have never seen nurses give capsules at the ANC. R3: Fansidar we can take, but not capsules. Capsules are very strong and you can miscarry immediately, while with fansidar there is no harm. M: What are these pills for [the women are shown iron tablets]? R1: We get those pills for increasing our blood. R2: They are iron tablets for shortage of blood. M: What about these pills [the women are shown panado pills]? R1: It is fansidar. R2: No it is panado, there is a line. Fansidar has a cross [many women saying ‘ee’ to agree]. R3: We have learnt today that there are different types of white pills. We thought that they are all the same. We also explored whether the source of medication made a difference. The IDIs revealed that women preferred to receive their medication from the health centre, as one woman explained, ‘‘If the doctor or the nurse gives you the pills, you can take them, because they know about pregnant women and medicines. We trust them and are not afraid of taking pills at the health centre. But suppose that someone in the village, who is nobody, gives you pills and advice? You can’t take medical advice from a nobody’’. Non-pregnant women can acquire medication from traditional healers and small groceries in the villages and at Lungwena trading centre. In contrast, during pregnancy, medicines from traditional healers were feared because of their power and those

Table 1

983 from shopkeepers because of wrong dosages. Furthermore, a pregnant woman was advised against taking any unknown medicine outside of her household, because of its potential danger, as one mother spoke of her experience, ‘‘I was staying at my husband’s home. At four months pregnant, I aborted there. I was given not-well-known medicines. Because I was in foreign territory, I was just taking the medicines and so I miscarried at four months’’.

3.2. Use of SP during pregnancy and knowledge of IPT-SP According to the findings of the IDIs and FGDs, SP was well known. Although all women had heard about it, some had not taken SP while pregnant, because they had not suffered from malungo. Many interpreted the SP given at the ANC as suggestive of malungo, as described by a FGD participant, ‘‘Leaving home you think that you are fine. You are surprised that the nurses discover that you are suffering from malungo, because they give you these pills, fansidar’’. Women were also asked, ‘‘Can you take fansidar to prevent malungo?’’, to which answers were such as ‘‘Impossible! We take fansidar only after being examined and the nurses have found out that we are ill’’, ‘‘When you are fine, you can’t take pills to prevent a disease’’, ‘‘We use mosquito nets to prevent malungo’’. Survey respondents were also asked ‘‘Can malungo in pregnancy be prevented?’’ and those who replied ‘yes’, were further asked to explain how. Only a minority of the KAP survey respondents expressed a view that fansidar is for preventing malaria in pregnancy (see Table 1). Women reported very few side effects regarding SP, only slight bitterness, loss of appetite and dizziness. Regarding bitterness, one woman explained, ‘‘Fansidar is bitter, but the doctors recommend it to pregnant women. They know better why a pregnant woman can still take it, although it is bitter’’. In general, women found SP tasteless. Some women said that they use panado to dilute SP. With a few exceptions, women considered SP suitable during pregnancy.

3.3. Timing and motivation for attending the ANC During the IDIs most women said that they should start ANC between 12 and 24 weeks of pregnancy. Among the KAP sur-

Knowledge of malaria prevention in pregnancy from a ‘knowledge, attitudes and practices’ (KAP) survey in rural Malawi

Question

Can malaria in pregnancy be prevented? Please explain howa

Response

Yes ITNs SP Otherb Don’t know

Venue of response ANC (n = 48)

Villages (n = 190)

19 (40%) 13 (27%) 1 (2%) 5 (10%) 0

94 (50%) 56 (29%) 15 (8%) 22 (12%) 1

ANC: antenatal clinic; ITN: insecticide treated net; SP: sulfadoxine—pyrimethamine. a Those who responded ‘yes’ were asked to explain how malaria in pregnancy can be prevented. b Responses varied, from keeping the home clean, staying away from stagnant water, starting ANC early, cleaning surroundings, spraying Doom (insecticide), stopping having sex, taking panado pills to taking medical treatment, and so on.

984

A. Launiala, M.-L. Honkasalo

vey respondents, 22% (55/248) said that nurses have advised them to start the ANC during the first 3 months of the pregnancy, 39% (96/248) said 4 months, 20% (49/248) said 5 months and 14% (34/248) said from 6 months onwards. Investigation of the ANC register (2002) showed that the average time of the first visit was 24 weeks into pregnancy. At the IDIs and FGDs, several factors emerged influencing the timing. First, the necessity to confirm the pregnancy, and waiting until the fetal movements were felt was common. Many women had experienced miscarriage in early pregnancy and wished to be sure of this pregnancy before travelling all the way to the ANC. ‘‘Sometimes, it happens that after one or two months of the pregnancy, you start to menstruate again. Therefore, I wait until I feel the movement of the baby before I attend the ANC’’. Second, revealing pregnancy at an early stage was not advisable because of witchcraft and ill will in the community, e.g. a jealous co-wife might want to harm the pregnancy. Thirdly in a normal pregnancy, progressing without a problem and with the woman not feeling sick meant that the ANC starting time was often delayed. A vice versa situation was likewise possible, as one woman said, ‘‘I was sick, suffering from some disease and this prompted me to start the ANC early’’. The fourth important reason was distance. Women living far away preferred to start the visits late, with only a few months left before the delivery, as explained one mother, ‘‘I live far away and

Table 2

the pregnancy takes a long time, therefore I don’t hurry starting with the ANC. Even if I start late, like six months, I will end up going to the ANC several times’’. Most of the women had attended the ANC more than once. The reasons to attend the ANC included the desire to ensure that the baby was growing well. Moreover, a feeling of necessity was expressed as the women wanted to be sure of nurses’ attention in the event that complications emerge during delivery, as explained by an FGD participant when asked what happens to a woman who does not attend the ANC and gets complications: ‘‘She is chased away by the nurses. — First question they ask is ‘where is your ANC Card?’ If you don’t show it immediately, you are chased back to home or elsewhere. Then you start to dislike the ANC, why go there. ANC card is the pregnant woman’s passport book and whenever you go to ANC, first priority is the card for easy history follow-up’’. The KAP survey results were similar. The main reasons to attend the ANC, at least once, were to be examined (82%, 202/246) and to obtain the ANC card in case of complications (35%, 86/248).

3.4. Compliance with the ANC guidelines We conducted participant observation in order to assess to what extent nurses complied with the ANC guidelines,

An excerpt from the observation notes describing a typical morning at the antenatal clinic (ANC)

Time

Activity

8.15

22 women are waiting outside the clinic examination room. This is their first visit. Inside examination room, the cleaner of the health centre is measuring BP. Noticing me, she looks worried. I ask her where the nurse is. She says that the nurse is at the family planning unit. I go there to ask permission for observations. Nurse says that she will come soon to ANC. Observing the tablets on the table, I notice that SP and aspirin tablets look alike: white, round and same size. The only difference is that on one it reads SP/GoM and on the other 500/GoM. There is no way the women can differentiate these tablets unless they are able to read. Nurse arrives. She starts with health education asking questions to which women answer. They are talking in Chichewa. Occasionally, women laugh and say ‘ee’ — agreeing. Nurse gives health education on malungo (malaria) and fansidar (SP), explaining that women should take fansidar twice between 4 and 7 months. She tells that mosquito nets protect from malungo and that nets can be bought at the clinic for a special price now, much lower than in shops. Women say ‘mmm’, agreeing. The nurse gives a long sales speech about pregnancy and the importance to use mosquito net. She speaks in Chichewa and asks if women understand her, and if they have questions, they say ‘no’, and if they have money, they say ‘yes’. Then nurse tells me that all but one have money to buy a net. Cleaner starts health education. She speaks in Chiyao. She shows an ANC card and clothes. Women laugh. Nurse interrupts health education. She collects all ANC cards and starts to discuss with the cleaner. An HSA arrives to give injections. Cleaner continues the health education session. Cleaner stops health education and starts to measure BP. Women are very obedient and silent. Nurse is writing receipts for the nets. Women are waiting. Nurse calls women by their names. They come one by one to collect and pay a net. Women kneel down on the floor and hand over the money without speaking. Examinations start. Two women go into delivery room and lie down on bed. The nurse looks for signs of anaemia, measures fundal height, listens to fetal heart sounds, touches woman’s legs. Most of the time she says nothing to the woman she is examining. Two women have a question but her reply is inaudible. All other women are silent while being examined. After examination, cleaner gives them SP and iron tablets. The women never even look at the tablets. They just swallow SP with some water from the tap. No instructions are given. All examinations were carried out hastily. All women having been examined, given SP and iron tablets, the clinic closes.

8.30

8.40

9.03 9.15 9.25 9.35 9.50

10.30

BP: blood pressure; SP: sulfadoxine—pyrimethamine; GoM: Government of Malawi; HSA: health surveillance assistant.

Prevention of malaria during pregnancy in rural Malawi which stipulate the following tasks: (1) to inquire about recent episodes of illness; (2) to measure maternal weight and blood pressure; (3) to perform an abdominal examination, including manual external palpation of the fetus, listening to fetal heart sounds and measuring fundal height; and (4) to administrate, when necessary, iron supplements and IPT-SP. We discovered that recent episodes of illness were not elicited on a regular basis. Maternal weight was measured regularly, but blood pressure (BP) measurement depended on the number of women attending the clinic, whether someone was able to assist the nurse and if batteries were available. No attention was paid to the outcome of measurements. Abdominal examinations, including manual external palpation of the fetus, listening to fetal sounds and fundal height measurements were hurriedly performed for every woman. After examination, women were given SP under DOT, and iron tablets, but without any explanations or instructions. Table 2 gives a description of a typical morning at the ANC. In addition, we investigated the ANC register (2002) to find out at which week of pregnancy the first dose of SP (SP-1) was administered. Between February and May, SP was administered to all pregnant women during their first visit to the clinic, independent of the week of pregnancy, which, according to the register, had been between weeks 10 and 36. Between June and mid-July, the clinic was out of stock of SP. Between mid-July and the end of November, SP was administered at 18 weeks or more into pregnancy.

3.5. Communication between pregnant women and nurses Observations showed that verbal communication between nurses and women was one-sided. As the nurse talked, the women listened and spoke only when asked. In general, health education was given in Chichewa, mother tongue of the nurses, and only occasionally in Chiyao, which was the mother tongue of a majority of the women. Table 3 gives a description of the verbal communication between a nurse and a group of pregnant women.

985

3.6. Instructions/health education on malaria prevention Particular attention was paid to health education messages regarding malaria prevention, because the interviews in the communities revealed that women’s knowledge of malaria prevention was poor. The following health education excerpt gives an example of the messages given regarding IPT-SP at the ANC: Nurse: We also give you medicine for malungo. We all know malungo? Women: Yes. Nurse: This illness malungo has reached a critical stage, so we give you medicine called fansidar. Do we all know fansidar? Women: Yes Nurse: How many tablets of fansidar should we take? One woman: Three Nurse: What colour is fansidar? One woman: White Nurse: Have we all seen fansidar? Women: Yes Nurse: You take fansidar right here at the clinic. It does not matter if you have not eaten. We don’t let you take it home, because you may not take it or you share it with someone else. As a result we are going to defeat malungo. Have you ever heard about mosquito nets? Women: Yes. Nurse: We don’t have them now, but when we receive more, you have to get hold of money (MK50). If you suffer from malungo, you are likely to become anaemic, you can miscarry. Next time you come you will get a net. Please, my fellow ladies, we will have to try hard to obtain the money, because the government wants each and every pregnant woman to have at least one net. I have already said that malungo has reached a critical stage. If you suffer from malungo, you are likely to become anaemic,

Table 3 An excerpt from the observation notes describing verbal communication between a nurse and the pregnant women during a health education session Time

Activity

8.52

Nurse begins health education using Chichewa language. The topic is ‘importance of attending ANC’. The women are sitting in silence. Nurse finishes the talk and asks: ‘‘Do we now all know the reasons why we should visit ANC?’’. Women say ‘ee’ meaning ‘yes’. Nurse asks women to tell her these reasons. Women are silent. Finally, she gets three answers: (1) we get vaccine; (2) we get checked if the baby is growing well; and (3) we get blood tested for STI like chindoko (gonorrhea). Some women laugh at the answers. Nurse says ‘ee’ to each answer. Then she asks: ‘‘Anything else?’’ Women are silent. Then the nurse repeats the reasons she mentioned in the health talk: ‘‘(1) we check your blood to see if you have enough blood. Then you get 30 iron tablets so that blood suffices; (2) the baby is abnormal in the early months but later on it becomes normal; (3) when you are 7—9 months you should feel the baby moving; (4) we check you for swollen legs; (5) you get fansidar for malungo and you should buy at least one mosquito net; (6) we test your blood for STIs.’’

9.01

ANC: antenatal clinic; STI: sexually transmitted infection.

986

A. Launiala, M.-L. Honkasalo and if you have it often, you can miscarry. Do we all know about this? Women: Yes. Nurse: Most pregnant women are dying because of this malungo and the government is concerned. This is why the price has been reduced from MK140 to MK50. Is this clear? Women: Yes.

To test women’s understanding of health education on malaria prevention, 48 KAP survey respondents were interviewed immediately after a health education session for malaria prevention and IPT-SP under DOT at the ANC. Less than half of them (19/48) said that malaria in pregnancy was preventable and one said that it was preventable with SP (See Table 1).

4. Discussion The use of an ethnographic approach allowed us to discover new insights into the complexity of the multiple contexts shaping women’s knowledge and practices, as well as implementation of the IPT policy. We first asked: ‘‘What type of knowledge and perceptions do women have concerning medication during pregnancy?’’ The women had a clear idea of suitable medication, so that, based on taste and medicinal efficacy, medications with a bitter taste, such as herbs or capsules, were considered harmful and prohibited in pregnancy because of risk of miscarriage. The finding concurs with previous research (Etkin, 1992; Helitzer-Allen et al., 1997; Whyte et al., 2002). Women’s knowledge of the drugs given at the ANC was vague. Similar to the findings of Allen (2002) in Tanzania, very few women were able to identify drugs by name. Instead, drugs were identified by shape and colour and described, for example, in terms of their benefits, such as ‘pills preventing malaria’ and ‘pills increasing the blood’. The source of medication was considered important and the health centre was regarded as the most secure source. This can be explained by the women’s weak knowledge about drugs and the shared belief of danger of miscarriage due to ‘wrong’ medication. Writing down measurements at the ANC examination and subsequent giving of medication may also demonstrate the skills and authority of nurses over the uneducated Yao women. As Van der Geest et al. (1996) have suggested, filling a prescription can be considered a social act demonstrating the skills and authority of a medical person. Secondly, we asked: ‘‘How do women perceive the use of SP during pregnancy? Do they have knowledge of IPT-SP?’’. In general, SP was perceived as suitable medication for pregnant women. It was well tolerated, with few side effects, similar to the findings of another study elsewhere in Malawi (Ngoma, unpublished). Not many women complained about the slightly bitter taste and dizziness, a finding contrary to an earlier study in Malawi suggesting that nurses tend to believe that pregnant women refuse to take SP because of its bitter taste and dizziness as a side effect (Ashwood-Smith et al., 2002). A study conducted in Uganda showed that health workers had misconceptions, such as SP being too strong for pregnant women and capable of causing abortions (K2Research Ltd, unpublished). We also observed that nurses

tend to anticipate women to show disobedience unless DOT is practiced, but we did not conduct interviews to confirm this observation. Concerning IPT-SP, we found that women’s knowledge was limited, often linking SP to diagnosis of malaria. Women frequently offered that malaria in pregnancy was not preventable with SP, similar to the findings in Uganda where community members questioned the rationale of giving SP to pregnant women unless they were sick (K2-Research Ltd, unpublished). Women’s lack of knowledge of IPT-SP can be explained by the unclear health education messages at the ANC sessions given in the ‘wrong’ language, and the instructions omitted during SP administration. Thirdly, we asked: ‘‘What is the average time of starting ANC and does this affect implementation of IPT-SP?’’ and ‘‘What motivates women to attend the ANC, is it the IPT-SP?’’. The ANC register book showed week 24 of pregnancy as the average time for the first visit, suggesting rather late enrolment that may delay administration of SP-1, similar to studies in Kenya (Williams and Mungai, unpublished), elsewhere in Malawi (Ashwood-Smith et al., 2002) and in Tanzania (Allen, 2002). Our findings showed that many women knew that ANC visits should start before week 20, and the reasons for delaying enrolment varied from waiting until fetal movements were felt and unwillingness to reveal pregnancy at an early stage for fear of witchcraft and ill will to reluctance towards making multiple visits or complaints of long distances. Furthermore, women with many pregnancies and good experiences failed to see much need to attend the ANC. Women did not mention IPT-SP as a reason to attend the ANC. Instead a crucial motivation to attend the ANC at least once was to ensure emergency obstetric care and availability of treatment options if necessary, similar to the findings of Helitzer-Allen et al. (1997). Another motivating factor was the actual examination, to obtain the knowledge that the fetus was growing well. Whyte (2005) has argued that medical examinations are valued, because they indicate that care has been taken and the most modern technology offered. Examinations enhance confidence and hope for a good outcome. Therefore, it seems plausible that the ANC examinations provide hope for the women in their uncertainty surrounding pregnancy outcome. This could also explain motivation for re-attending the clinic, but further research is required to confirm this assumption and to explore other factors influencing re-attendance. For describing the ANC context we conducted participant observation. We first aimed to find out the answer to the question ‘‘What does actually happen at the ANC?’’. The observations showed a blunt picture of the situation. The nurses were rather dominating and performed their tasks vaguely and routinely, without paying attention to the measurements taken during the ANC examination. The pregnant women were passive and obedient recipients of care. The passiveness could be explained by Yao people’s respect for wisdom gained by age and education; here, nurses represent education and, occasionally, mature age. Culturally, the Yao women are not expected to be vocal before they have gained experience through multiple pregnancies and old age. Another explanatory factor of nurses’ dominance could be that they are not Yao but Chewa people. Traditionally, the Yao have been considered backward and ignorant by both the Chewa and the Tumbuka because of their lack

Prevention of malaria during pregnancy in rural Malawi of interest towards education. Moreover, having control over medicines and knowledge, the nurses may intimidate the pregnant women, particularly young women. As Whyte (2005) demonstrated, superiority in terms of medicinal and medical resources and skills often allows health care staff to assume an intimidating role. Secondly, we asked: ‘‘Do women receive SP as recommended by the IPT policy guidelines?’’. The data we collected answers this question only partly. The nurses carried out IPT under DOT. The women received SP-1, but there were problems in the appropriate timing. The ANC register showed that nurses gave SP regardless of the week of pregnancy, between weeks 10 and 36. This was supported by the observations. SP was given DOT to every woman examined, irrespective of the week of pregnancy. One reason for this might be that the nurse on duty, aware of our ongoing malaria treatment and prevention study, wanted to emphasise the giving of IPT-SP. We also discovered periodical shortages of SP, a difficulty also noted by Ashwood et al. (2002) in their study. For assessing the effectiveness of IPT policy, and administration and timing of SP-2, more research is needed. Thirdly, we asked: ‘‘How do nurses and pregnant women communicate?’’. In this study, verbal communication between nurses and pregnant women was almost nonexistent. At the health education sessions, women were sometimes asked more-or-less rhetorical questions, requiring just ‘ee’ (yes) for an answer. There were no instructions for administration of SP and iron tablets, which explains why women interviewed at the community could not tell what ‘pills’ they received at the ANC or for what purpose. However, despite the nearly complete lack of verbal communication, the women were keen on being examined and receiving the pills, suggesting that verbal communication was not considered important. Mogensen (2005) described a similar situation in Uganda and concluded that, although cognitive understanding is not important when patient and health workers interact, the ability to conduct oneself appropriately at the health unit is. The emphasis is on doing rather than talking. Lastly, we looked more closely at the health education messages: ‘‘Do nurses provide clear and appropriate health education on malaria prevention?’’. Several problems were noticed. Not only was there an excessive number of topics covered, but the messages were unclear and spoken in a tongue not native to the pregnant women. All messages about SP were vaguely discussed, without a clear reference to its use for malaria prevention, for example: ‘‘malungo has reached critical stage and that’s why women should take fansidar’’. Our study also had some limitations. First of all we focused on investigating compliance from the pregnant women’s perspective. However, during the data analysis we realised that we should have also included the providers’ perspective in the study plan. It would have been important to interview the nurses about their knowledge on the policy, as well as attitudes concerning the pregnant women and drug use, etc. In addition we should have investigated how economic and structural factors affect the women’s ability to comply in the area. The pregnant women are expected to walk to the ANC even from the furthest village, more than 10 km away, which they do unless complications or pains

987 render walking impossible. In the local Yao culture, men control the family spending, including money for treatment and transportation. Control of the economic factors is a particularly important issue in risk pregnancies and emergency obstetric care. A second limitation is that we focused on factors affecting women’s motivation and their gestational age at the first visit. It is, however, highly relevant to understand what motivates women to re-attend the ANC, and the factors influencing the timing of re-attendance. Thirdly, our data collection was limited to the timing and administration of SP-1 and to describing the IPT implementation process and context. However, quantitative data on SP-2 is required for assessing the effectiveness of the policy. Lastly, one may also raise a concern that due to the ethnographic approach, and particularly due to observations at one ANC, the results cannot be statistically generalised, and thus this is a limitation of the study. While this concern is valid for population-based studies, it is important to understand that we have used ethnography to voice the Yao women’s concerns and to make sense of their knowledge and perceptions, as well as to understand and describe the multiple contexts influencing IPT programme implementation. Observations at the ANC are crucial for providing understanding of circumstances and interaction related to IPT-SP implementation. Producing statistically generalisable results does not constitute an aim in ethnographic research (Savage, 2000; Trostle, 2005). This ethnographic research highlights important and relevant issues to be considered in future investigations regarding compliance with IPT. Special attention should be paid to investigation of factors influencing compliance from the providers’ perspective and the critical perspective that emphasises the effect of economic and structural factors on ability to comply (see Farmer, 1997; Farmer et al., 2006). In addition, factors affecting re-attendance at the ANC and its timing need to be studied. Anyone interested in compliance research must understand that compliance is such a multifactorial issue that for comprehensive understanding it has to be explored from multiple perspectives using multi-method study designs.

5. Conclusions The ethnographic approach and use of multiple methods contributed to a rich, detailed and contextualised description of women’s understanding of the use of medication and ANC services during pregnancy, as well as a description of the circumstances in which IPT-SP is implemented. We found out that the local health centre is considered a good (and only) source of secure medication during pregnancy, and that the medical staff is trusted and enjoys authority, hence forming an excellent basis for addressing factors affecting IPT-SP implementation. We discovered that the nurses gave unclear IPT-SP messages in a ‘wrong’ language and that their knowledge of the timing of SP-1 was inaccurate. Furthermore there were periodic shortages of SP. We learned that women themselves had little and vague knowledge of drugs and understanding concerning IPT-SP. Furthermore, women had a tendency to delay enrolment for various reasons and nurses underperformed their duties. To address these problems, we recommend training nurses about various aspects of the

988 policy (i.e. timing, appropriate delivery regarding messages about IPT and drugs, monitoring of drug supply), and regular upgrading of skills. These improvements are likely to lead to better work performance and quality of care. As a consequence, women’s understanding of IPT and motivation to use the ANC services regularly might also improve. We also advocate that any future research on compliance also includes the providers’ and a critical perspective, and that the ANC re-attendance issue is not overlooked. This study clearly shows that an understanding of multiple contexts is important for successful and sustainable programme implementation. Ethnographic research can be used to discover and solve problems beyond the scope of many other research approaches. Authors’ contributions: AL and M-LH designed the study plan; AL collected, analysed and interpreted the data; MLH supervised the data analysis. AL and M-LH drafted the manuscript and read and approved the final version. AL and M-LH are guarantors of the paper. Acknowledgements: Thanks are expressed to all the women in the Lungwena area who participated in this research; the Lungwena Health Centre staff and the hardworking research assistants: Rashid Osman, Eunice Willy, Misontzie Tweya, Shaibu Msosa, Ben Mwema, Gertrud Moses and Zacharia Abdul; the National Malaria Control Programme and Dr Kalanda for support concerning the study; the Finnish Family Federation for logistical support; Dr Jenny Hill (London School of Hygiene and Tropical Medicine) and Prof. Kari Launiala for reviewing and commenting on the manuscript; and Mrs Marja Vajaranta for language editing. Funding: The Nordic Institute of African Studies (NAI) provided funding for the fieldwork. Doctoral Programs in Public Health (DPPH) provided funding so that it was possible to write this article. Conflicts of interest: None declared. Ethical approval: National Health Sciences Research Committee of the Malawian Ministry of Health and Population. The College of Medicine Research Ethical Committee of the University of Malawi was informed about the research.

References Agyepong, I.A., Aryee, B., Dzikunu, H., Manderson, L., 1995. The malaria manual. Guidelines for the rapid assessment of social, economic and cultural aspects of malaria. Methods for Social Research in Tropical Diseases No. 2. UNDP/World Bank/WHO Special Programme for Research & Training in Tropical Diseases (TDR), TDR/SER/MSR/95.1. Allen, D.R., 2002. Managing Motherhood, Managing Risk: Fertility and Danger in West Central Tanzania. The University of Michigan Press, Ann Arbor, MI. Ashwood-Smith, H., Coombes, Y., Kaimila, N., Bokosi, M., Lungu, K., 2002. Availability and use of sulphadoxine-pyrimethamine (SP) in pregnancy in Blantyre District. A safe motherhood and Blantyre Integrated Malaria Initiative (BIMI) joint survey. Malawi Med. J. 14, 8—11.

A. Launiala, M.-L. Honkasalo Benson, T.D., Kaphuka, J., Kanyanda, S., Chinula, R., 2002. Malawi: An Atlas of Social Statistics. National Statistics Office, Zomba and International Food Policy Research Institute, Washington, DC. Bernard, H.R., 2002. Research Methods in Anthropology. Qualitative and Quantitative Approaches, third ed. Altamira Press, Walnut Creek, CA. Bhattacharyya, K., 1997. Key informants, pile sorts, or surveys? Comparing behavioral research methods for the study of acute respiratory infections in West Bengal. in: Inhorn, M.C., Brown, P.J. (Eds.), The Anthropology of Infectious Diseases. Theory and Practice on Medical Anthropology and International Health, vol. 4. Gordon and Breach Publishers, Amsterdam, pp. 211— 238. Conrad, P., 1985. The meaning of medications: another look at compliance. Soc. Sci. Med. 20, 29—37. Donovan, L.J., Blake, D.R., 1992. Patient non-compliance: deviance or reasoned decision-making? Soc. Sci. Med. 34, 507—513. Etkin, N.L., 1992. ‘‘Side effects’’: Cultural construction and reinterpretations of Western pharmaceuticals. Med. Anthropol. Q. 6, 99—113. Farmer, P.E., 1997. Social scientists and the new tuberculosis. Soc. Sci. Med. 44, 347—358. Farmer, P.E., Nizeye, B., Stulac, S., Keshavjee, S., 2006. Structural violence and clinical medicine. PLoS Med. e449, 1686— 1691. Helitzer-Allen, D.L., Kendall, C., 1992. Explaining differences between qualitative and quantitative data: a study of chemoprophylaxis during pregnancy. Health Educ. Q. 19, 41—54. Helitzer-Allen, D.L., Kendall, C., Wirima, J.J., 1997. Beliefs and practices during pregnancy in a Yao community in Malawi. in: Kalipeni, E., Thiuri, P. (Eds.), Issues and Perspectives on Health Care in Contemporary Sub-Saharan Africa. Studies in African Health and Medicine, vol. 8. The Edwin Mellen Press, Lewiston, New York, pp. 145—164. Hill, J., Kazembe, P., 2006. Reaching the Abudja target for intermittent preventive treatment of malaria in pregnancy in African women: a review of progress and operational challenges. Trop. Med. Int. Health 11, 409—418. Homedes, N., Ugalde, A., 1993. Patients compliance with medical treatments in the third world. What do we know? Health Policy Plan. 8, 291—314. Humphery, K., Weeramanthri, T., Fitz, J., 2001. Forgetting Compliance: Aboriginal Health and Medical Culture. Northern Territory University Press, Darwin, Northern Territory. Launiala, A., Kulmala, T., 2006. The importance of understanding the local context: women’s perceptions and knowledge concerning malaria in pregnancy in rural Malawi. Acta Trop. 98, 111— 117. Massele, A.Y., Mpundu, M.N., Hamudu, N.A.S., 1997. Utilisation of antimalarial drugs by pregnant women attending the antenatal clinic at Muhimbili medical centre, Dar es Salaam. East Afr. Med. J. 74, 28—30. Mogensen, H., 2005. Finding a path through the health unit: practical experience of Ugandan patients. Med. Anthropol. 24, 209—236. National Statistics Office (Malawi) and ORC Macro, 2001. Malawi Demographic and National Survey 2000. Zomba, Malawi and Calverton, MD, USA, National Statistics Office and ORC Macro. Savage, J., 2000. Ethnography and health care. BMJ 321, 1400—1402. Schultz, L.J., Steketee, R.W., Chitsulo, L., Macheso, A., Nyasulu, Y., Ettling, M., 1994. Malaria and childbearing women in Malawi: knowledge, attitudes and practices. Trop. Med. Parasitol. 45, 65—69. Sumartojo, E., 1993. When tuberculosis treatment fails. A social behavioral account of patient adherence. Am. Rev. Respir. Dis. 147, 1311—1320.

Prevention of malaria during pregnancy in rural Malawi Thorold, A., 1993. Metamorphoses of the Yao Muslims, in: Brenner, L. (Ed), Muslim Identity and Social Change in Sub-Saharan Africa. Indiana University Press, Bloomington, IN, pp. 79—90. Trostle, J.A., 1988. Medical compliance as an ideology. Soc. Sci. Med. 27, 1299—1308. Trostle, J.A., 2005. Epidemiology and culture. Cambridge University Press, Cambridge, UK. Van der Geest, S., Whyte, S.R., Hardon, A., 1996. The anthropology of pharmaceuticals: a biographical approach. Annu. Rev. Anthropol. 25, 153—178. Van Eijk, A.M., Ayisi, J.G., ter Kuile, F.O., Otieno, J.A., Misore, A.O., Odondi, J.O., Rosen, D.H., Kager, P.A., Steketee, R.W., Nahlen, B.L., 2004. Implementation of intermittent preventive treatment with sulphadoxine-pyrimethamine for control of malaria in pregnancy in Kisumu, western Kenya. Trop. Med. Int. Health 9, 630—637.

989 WHO/UNICEF, 2005. World Malaria Report 2005. WHO Roll Back Malaria Department and UNICEF. http://www.rbm.who.int/ wmr2005/ [accessed March 2007]. Whyte, S.R., 2005. Uncertain undertakings: practising health care in the subjunctive mood, in: Jenkins, R., Jessen, H., Steffen, V. (Eds), Managing Uncertainty. Ethnographic Studies of Illness, Risk and the Struggle for Control. Museum Tusculanum Press, University of Copenhagen, Copenhagen, pp. 245— 283. Whyte, S.R., Van der Geest, S., Hardon, A., 2002. Social Lives of Medicines. Cambridge University Press, UK. Williams, H.A., Jones, C.O.H., 2004. A critical review of behavioural issues related to malaria control in sub-Saharan Africa: what contributions have social scientists made? Soc. Sci. Med. 59, 501—523.

Annika Launiala School of Public Health University of Tampere, Finland Marja-Liisa Honkasalo Division of Health and Society University of Link¨oping, Sweden

Malaria, Danger, and Risk Perceptions among the Yao in Rural Malawi Findings from a study designed to discover how local understanding of malaria among Yao in Malawi relate to pregnancy risk definitions reveal that malaria in pregnancy is not perceived as a major risk. Using extended ethnographic field research and multiple methods, we argue a shift from narrow single-disease approaches to malaria during pregnancy is required and document women’s concerns about exposure to multiple vulnerabilities during pregnancy, including witchcraft, extramarital affairs, and multiple dangerous illnesses. Four dimensions are implicated in Yao perceptions of risk: perceived adverse consequences in pregnancy; ease of treatment and cure; transmission and agency to control; and type of risk (social–medical). We discuss implications and consider malaria program features needed to address the complexity of perceived vulnerabilities and living conditions in resource-poor settings. Keywords: [risk, malaria, sexually transmitted illnesses, pregnancy, structural factors, Malawi] In April 2007, one of my Malawian research assistants phoned with news of, I (principle author) thought, the expected birth of a son. He and his wife had honored me by inviting me to choose the baby’s name, and I eagerly awaited news. Sadly, he said that Onni (the name I chose, happiness in Finnish) had died, only four days after birth. We expected Onni in late May or early June. On April 5 my wife complained of fever, whereupon I advised her to seek medical attention at the Lungwena health center. There she was given SP (sulfadoxinepyrimethamine), but her condition did not improve. After three days we started to think that the situation was bad, and on April 9 she went back to the health center. From there she was referred to St. Martin’s mission hospital where she delivered after a hard struggle, just barely escaping surgery, but thank God she gave birth to a baby boy who weighed 2700g. MEDICAL ANTHROPOLOGY QUARTERLY, Vol. 24, Issue 3, pp. 399–420, ISSN 0745C 2010 by the American Anthropological Association. All rights 5194, online ISSN 1548-1387.  reserved. DOI: 10.1111/j.1548-1387.2010.01111.x

399

400

Medical Anthropology Quarterly

Two days later she and Onni were discharged from the hospital. The problems started at home. My wife tried to breastfeed Onni, and at first it seemed to go well. But on the following night Onni became feverish and refused to suckle at her breast. In the morning we went to the Lungwena health center from where we were again referred to St. Martin’s mission hospital. After some examinations the personnel found that Onni was suffering from malaria, and his temperature kept rising. Both my wife and Onni were given some treatment and admitted to the ward. I left for home around 4 pm and at that time Onni seemed to be improving. The following morning I was getting prepared to go to the hospital when they arrived with the corpse. They told me that Onni had died during the night, April 12. Onni’s untimely death propelled deeper reflections about my presumptions that malaria is perceived by everyone in Malawi as a major problem, one with dire consequences for pregnant women and young children. Despite, perhaps because of, my prior role as a UNICEF project officer responsible for malaria prevention activities, I had not questioned the disease focus. As the need is great to reduce the heavy burden of malaria in Malawi and sub-Saharan Africa where 25 million women become pregnant annually and are at risk of malaria, which significantly contributes to maternal and neonatal morbidity and mortality (Dellicour et al. 2010; Rogerson et al. 2007). Onni’s death made me wonder what factors influence women’s ability and motivations for malaria prevention during pregnancy, what kinds of local perceptions of pregnancy risks and vulnerabilities women have, how malaria during pregnancy is related to them, and more importantly how they match with the “at risk” approach used by international organizations and program planners. This article examines the multiple dimensions in local perceptions of risk and shows why the Yao people of rural Malawi do not perceive malaria in pregnancy as a major risk. We do this by moving beyond the conventional single-disease approach and explore malaria in the context of pregnancy. We pay attention to the perceived vulnerabilities—witchcraft, extramarital relationships, multiple illnesses, and worries in pregnant Yao women’s everyday lives—and examine how malaria is associated with these vulnerabilities. To reach a comprehensive understanding of risk perceptions, we investigate these issues from multiple perspectives among the Yao, including those of men, older women, pregnant women, and women of reproductive age as “it is the diversity in the experience and perceptions to risk that are important” (Nichter 2003:28). The following research questions guided our research: What kinds of vulnerabilities are pregnant women perceived to be exposed to in the communities? What local illnesses are perceived to be dangerous for pregnant women, why, and is malaria among these illnesses? What are the major worries among women and men related to pregnancy? In international health, epidemiological approaches to risk dominate. Risk is calculated as a probability in terms of the odds that something will occur (or will not occur) within a given population exposed to specific risk factors relative to a reference population and perceived risk (Frankenberg 1993:229; Gordis 2004). Frankenberg describes the risk approach in epidemiology as a way of justifying a medical intervention (1993:233). Anthropologists criticize the epidemiological

Malaria, Danger, and Risk Perceptions in Rural Malawi

401

risk definitions for its neglect of the sociocultural settings where risk is understood and negotiated and for failing to account for heterogeneity and variations in meanings of risk among groups (Frankenberg 1993; Lupton 1999:24; Trostle 2005:161). Mary Douglas’s pioneering work (1992) galvanized medical anthropologists to deeply engage the notion of risk, also referred to as hazard, chance, uncertainty, and danger (Lupton 1999:8–9; Nichter 2008). Advances in our understanding of risk have come as studies pursue several perspectives. Some examine risk through the notion of uncertainty and its manifestations in everyday life (Bledsoe 2002; Haram 2005; Honkasalo 2006; Whyte 1997; see also Caplan 2000). Others demonstrate how social threats and risks influence treatment-seeking and prevention practices and how they can outweigh physical and biological risks (Bujra 2000; Chapman 2006; Nichter 2003:29, 2008:58–59). According to Nichter (2003:29), it is important to notice when social risks (“risk to valued relationship” as defined by Nichter 2008:58) outweigh biologized and medicalized risk categories prioritized in public health settings. This enables us to understand why certain preventive methods are accepted and others not. It is also important to note that risks are valued and prioritized differently based on their relevance at a given time, rather than actively considered simultaneously all the time. Nichter (2008:60) refers to hierarchies of risks: “how risk is judged in context as well as in relation to other risks rather than in isolation” (see also Day 2000). Furthermore, some medical anthropologists have emphasized that a comprehensive understanding of risk perceptions and factors affecting exposure to risk also requires examination of structural forces—poverty, inequality, and livelihood insecurity—that all impinge on access to health care. In other words, risk is located in the living conditions in which it occurs and investigated in a larger framework of vulnerability and agencies of power (Farmer et al. 2006; Obrist et al. 2007; Sommerfeld et al. 2002). There is virtually no research exploring risk perceptions of and vulnerability to malaria in pregnancy. Instead, social scientists have focused more narrowly on malaria, its meaning and influence on illness experiences, treatment-seeking practices, and prevention among children under five (Ribera et al. 2007; Williams and Jones 2004). The few social scientific studies on pregnant women and malaria have examined treatment-seeking practices (Ahorlu et al. 2007), use of antenatal services (Launiala and Honkasalo 2007; Ndyomugyenyi et al. 1998), compliance with intermittent preventive treatment in pregnancy (Helitzer-Allen et al. 1994; Launiala and Honkasalo 2007; Mbonye et al. 2006; Mnyika et al. 1995), and perceptions of malaria in pregnancy (Launiala and Kulmala 2006; Winch et al. 1996). Yet, understanding experiences of vulnerability and perceived risks in a particular sociocultural and structural context are important as they influence how people respond to public health interventions such as prevention of malaria in pregnancy (Nichter 2008:11, 50).

Fieldwork Setting and Methods Lungwena is roughly 300 kilometers from the capital and 30 kilometers from the nearest town of Mangochi, on the eastern shore of Lake Malawi. The population of Lungwena and surrounding areas are mainly Yao by ethnic origin. The majority are

402

Medical Anthropology Quarterly

Muslims practicing Islam adapted to the local communities. The Yao are matrilineal, reckoning descent through the female lineage. They prefer cross-cousin marriage, and both monogamous and polygamous marriages exist (Mitchell 1956; Paz Soldan et al. 2007; Thorold 1993). The matrilineal head of household is generally a man: husband, maternal uncle, brother, or grandfather. The maternal uncle often owns and controls the inherited property and family resources such as land (Eriksen 2001:102–103). In Lungwena many men temporarily migrate to work in the mines in South Africa, and families struggle with extreme poverty and are affected by AIDS. The region is isolated because of poor infrastructure, harsh environment during the rainy season, and geography being located between Lake Malawi and the bordering hills of Mozambique. Until recently just a sand road linked Lungwena to Mangochi, impassable during the rainy season. Vehicles operate irregularly during the dry season. In 2001–03, a tar road was constructed greatly improving access to nearby villages and Mangochi town. Yet because of poverty and lack of cash-earning opportunities, few can afford local transportation. Most households live under the ultrapoverty line of less than $0.25 daily and experience periodic hunger. Only a few families can afford to buy basic food items such bread, sugar, and cooking oil. Fishing and farming, mainly maize, are common occupations, but basically all fish and maize are consumed by the households themselves (Benson 2002). Medical pluralism is characteristic of the area. A government-supported health center and three outreach clinics provide free medical treatment to roughly 30,000 people in 25 villages (Lungwena H/C data 2002). Also, traditional healers (asing’anga), Islamic healers, shopkeepers, and traditional birth attendants provide services to the communities. Antenatal care is available with about 95 percent of the pregnant women attending the antenatal clinic (ANC) at least once during their pregnancy. The clinic with its mostly well-stocked dispensary handles uncomplicated pregnancies and deliveries. However, severely ill adults, children, and women with pregnancy and delivery complications are referred to either St. Martin’s Mission Hospital in Malindi (16 km from Lungwena on Magochi to Makanjira road) or Mangochi District hospital (about 35 km from Lungwena). Patients must pay for transport and in Malindi also for the services. It costs families 750MK ($6.25) to 1500MK ($12.50) to transport, pay for services, and provide food for a child suffering acute malaria referred and admitted to St. Martin’s. Thus, with little or no cash available, few can access services in Malindi and Mangochi. The first author conducted extended fieldwork in 2002 and briefly in 2006. This ethnographic study used multiple methods, including focus-group discussions (FGDs), key informant and in-depth interviews (IDI), participant-observation, drugidentification and illness-ranking exercises, and a quantitative knowledge, attitude, and practice survey (Bernard 2002). In 2002 thirty-four IDIs were conducted with women of reproductive age, and key informant interviews with four traditional advisors (anankungwi), two traditional birth attendants, one traditional healer (sing’anga), and two men. Research assistants moderated eight FGDs and conducted 248 survey interviews. All IDIs and FGDs began with broad, nondirective questions about pregnancy, delivery, and health of the pregnant women and gradually focused toward malaria in pregnancy, its treatment, and prevention with more specific themes. The purpose was to gain a comprehensive understanding of the

Malaria, Danger, and Risk Perceptions in Rural Malawi

403

sociocultural context of managing pregnancy and malaria. (See Launiala and Kulmala 2006 and Launiala 2009 for details.) Return fieldwork extended investigation of themes from the first field visit and expanded data on men’s, pregnant women’s, and older women’s perspective. The themes explored included communication concerning pregnancy, knowledge, and perceptions of pregnancy risks, worries regarding pregnancy, perceptions of dangerous illnesses, role of husband and other relatives in management of pregnancy and complications, motivation for and use of ANC services, and management of the first pregnancy. Notably, along with others in sub-Saharan Africa, the Yao do not have a word for risk (Haram 2005; Sommerfeld et al. 2002). Therefore we used everyday Yao words such as problem (yakusausya), to worry (kudandaula), and dangerous (yakogoya). In 2006 three Yao trained as research assistants conducted data collection using thematic discussion guides and other study tasks. Discussion theme guidelines were translated from English into Chiyao and pretested. FGDs were moderated by two research assistants in seven villages, purposely chosen to be the same villages as during the fieldwork in 2002, and one FGD in the compound of Lungwena Health Center. Three FGDs (7 to 8 persons each) were held with men, two with older women and women of reproductive age, and one with young girls pregnant for the first or second time. The first author and a research assistant interpreter conducted 22 IDIs with men and eight with pregnant women to supplement the IDIs and FGDs with women of reproductive age conducted in 2002. The IDIs with men, selected from the nearby trading center, were conducted in the health center compound, and those with pregnant women, selected from women attending the antenatal clinic, in one of the Health Center offices. Participant-observation was also conducted at the ANC during two antenatal days, as were illness and pregnancy complication classification exercises.1 All the FGDs and IDIs were audiotape-recorded and translated from Chiyao into English for analysis.2

Malaria, Danger, and Risk Perceptions among the Yao Yao perceptions of multiple vulnerabilities during pregnancy are now described. We first examine how the women and men attribute witchcraft and extramarital relationships to increase the vulnerability of pregnant women. Next, we explore more closely the multiple illnesses perceived as dangerous for a pregnant woman, including malaria. Finally, prior to discussing factors influencing risk perceptions, we describe the worries women and men related to pregnancy and its outcome. Exposure to Vulnerabilities during Pregnancy During the first fieldwork the notion of vulnerability to witchcraft and extramarital relationships during pregnancy emerged from the women’s narratives in Lungwena. Women expressed vulnerabilities through different kinds of restrictions and prohibitions that impose limitations on a pregnant woman’s daily life. One set of restrictions related to proper behavior needed to reduce danger from witchcraft and malevolent spirits. Many women desired to keep their pregnancy a secret because they feared that it would arouse jealousy and ill will in the communities (Launiala

404

Medical Anthropology Quarterly

and Honkasalo 2007; see also Allen 2002; Chapman 2006). Another set of restrictions was related to food and medication during pregnancy. Pregnant women were often advised to avoid certain food items (mpingu) because of their potential harm to the unborn child.3 As for medication, women were advised not to take any bitter medicines because these are perceived to cause miscarriage (Launiala and Honkasalo 2007; see also Helitzer-Allen et al. 1997). Discussions with men and women in Lungwena showed that most of them shared a strong belief in witchcraft (ufiti) with the rest of the Malawians. Many studies in Malawi have shown that witches are believed to cause misfortunes and disasters, and witchcraft is used to explain occurrence of sudden deaths, severe illnesses, strange events, bad luck, and anything that interrupts the normal course of events (Englund 1996; Lwanda 2001; Mitchell 1956). Many people in Lungwena perceive pregnancy to be a vulnerable state exposing a woman to dangers of witchcraft and malevolent spirits, as illustrated by a husband, age 30: “Those who are pregnant are vulnerable to witchcraft because they are like an egg; if not handled properly they can break.” Because of susceptibility to witchcraft, pregnant women are regarded as vulnerable to illnesses and complications caused by witchcraft and evil spirits (see also Chapman 2006; Helitzer-Allen et al. 1997). It was commonly believed that quarrelling induces anxiety about the pregnancy outcome and also a danger of being cursed. Cursing (thembelero), in turn, can cause delivery complications and neonatal or maternal death. Many women, but also men, identified malowe (prolonged labor) as an example of complication attributed either to witchcraft, quarrelling, or female infidelity.4 Some women expressed fear of witchcraft during a home delivery because witches can easily hide in the crowd gathered outside the hut where the delivery takes place. They considered hospitals as secular and safe places. Many times women attributed negative pregnancy and delivery outcomes to witchcraft. A mother of three children explained: “My sister had some complications when she was delivering at home. We felt that the complications were caused by witchcraft. It took a long time for her to deliver, and when the baby was finally born, it was dead.” Chapman (2006) similarly reports that in Mozambique the most serious pregnancy and obstetric complications are attributed to personalistic reproductive threats of witchcraft and sorcery. Sometimes fear of witchcraft was expressed through a disappearing pregnancy, as a Yao man described: There are times when the wife seems to be pregnant and then the pregnancy disappears: no delivery, no miscarriage, no bleeding. Then the two of us sit down and think how we could solve the problem, and we agree that come what may, the wife will either die or not. Then I say: “Come on wife, sit on the bike and lets go to sing’anga,” and when we arrive at the sing’anga we ask for medicine to treat the wife only and nothing else, then we return home. We are afraid to inform our relatives because either I feel that maybe my mother is a witch and can be the one doing this, or maybe the wife’s mother is a witch, so we do the healing secretly. Another man had a similar opinion: “There are families that are prone to witchcraft. If a woman is pregnant, the pregnancy can disappear.” Feldman-Savelsberg (1994)

Malaria, Danger, and Risk Perceptions in Rural Malawi

405

described in her study in Cameroon how the Bangangt´e women are frightened of “vampires,” envious women who can eat or steal the fetus. Although many of the men and women said that they fear witchcraft in the village, witchcraft was often reserved to explain unexpected and negative consequences rather than a constant worry. A few said they did not believe in witchcraft at all; they thought that God decides the outcome of the pregnancy. Extramarital relationships during the pregnancy was a potential harm identified by both men and women of different ages. Men’s narratives suggested that extramarital relationships are common, as in the rest of the Malawi (Chimbiri 2007). A father of two children explained, “Frankly speaking the women try to control themselves, but mostly we men are the ones who bring the illnesses into the family because we never control ourselves. Of course there are some women who still go out with other men even if they are pregnant. A well-behaved woman cannot do that. She fears that if she does that [has sex with other men] she might experience a miscarriage.” Yet men expressed some efforts to avoid extramarital relationships during pregnancy. A father of seven, age 47, living with two wives explained, “The husband’s role changes when your wife is pregnant. You are more responsible and don’t indulge in promiscuity to avoid contracting diseases that can lead to a bad result concerning your wife’s pregnancy.” A variety of taboos on menstruation, pregnancy, childbirth, and sexual activity among these Malawians strengthen the intention to avoid extramarital sex. Among the Chewa these taboos are mdulo, and among Yao ndaka (Breugel 2001, 169–207; Lwanda 2001:153). According to ndaka, a pregnant woman is only supposed to sleep with her husband, and he should not have extramarital sexual relationships because this can harm the fetus or cause miscarriage. These taboos have traditionally been used to control the husband’s sexual behavior and maintain order in the family, is easily disturbed by adultery (Helitzer-Allen et al. 1997; Zulu 1996:186– 191). Nowadays, younger persons may have some doubts about these taboos, but the older generation still believes in the taboos and reinforces them through advice given in initiation ceremonies such as litiwo. The general purpose of litiwo has been to advise mwali ndembo (woman pregnant for the first time) about her pregnancy, how to manage it, and to prepare her for delivery. It is performed by anankungwi (traditional advisers), who are highly respected, elder, and experienced members of the community. Another important purpose has been moral education on appropriate social and sexual behavior (Mitchell 1956:134–135; Phiri 1997:35–36; Zulu 1996:32).5 In turn, cultural practices such as postpartum abstinence and polygamy are perceived to encourage men to have extramarital affairs (Bisika 2008). A Vast Number of Illnesses Are Dangerous for Pregnant Women The Yao classify and interpret illnesses in the framework of natural and personalistic causalities, as do many other ethnic groups in sub-Saharan Africa (Foster 1976; Green 1999; Young 1976). According to our data, many illnesses in Lungwena and nearby areas are interpreted as resulting from natural agents, and treatment is primarily sought from official health care providers and during pregnancy from the ANC. Some of the natural causes are also interpreted as acts of God (mulungu). There are also indigenous illnesses that are perceived to be caused by supernatural

406

Medical Anthropology Quarterly

agents such as witches (afiti) or evil spirits (majini) that often require negotiation within the therapy management group regarding treatment seeking (Janzen 1978) and treatment provided by traditional healers. Morris argues that “there are no witchcraft diseases in Malawi, only certain kinds of misfortunes which may in specific contexts be attributed to witches” (1985:17). Our findings suggest several illnesses that are considered more dangerous than malungo (Chichewa term for malaria). We explored in more detail what illnesses are perceived as dangerous for pregnant women and asked what factors are used to define the dangerousness of an illness. Analysis indicates that men and women considered 27 illnesses (and symptoms) dangerous for pregnant women.6 They mentioned most often gonorrhea (chizonono), chancroid (mabomu), syphilis (chindoko), HIV/AIDS (edzi), and malaria (malungo). There were altogether 64 accounts (53 percent) of STIs (gonorrhea, chancroid, syphilis),7 17 accounts of HIV/AIDS (14 percent), and 16 accounts (13 percent) of nonprobed malungo out of a total of 120 illness accounts. Notably, when asked which of the illnesses they consider the most dangerous for a pregnant woman, they ranked first HIV/AIDS, followed by STIs and malungo. A few also mentioned anemia, ndaka, kambanga (caused by epilepsy),8 cholera, and diarrhea, while some said that they do not know which illness is the most dangerous. A father of three children explained, “Each and every illness has its danger for a pregnant woman. The moment she is pregnant, she faces some problems, and if she also falls sick she will face two problems. The consequence can be endless pain, and she can lose the baby.” A pregnant woman, gestation week 32, listed gonorrhea, syphilis, chancroid, and malungo as the most common and dangerous illnesses in the area: “All of these illnesses are dangerous because they can cause serious things to the baby and you yourself become ill.” Next we examine STIs, HIV/AIDS, and malungo, ranked as the most dangerous by the majority of respondents, to better understand what factors contribute to their dangerousness. Sexually Transmitted Illnesses—Severe Consequences, Common, Easy to Treat, but Hard to Prevent. Men’s and women’s responses suggest that STIs attributed to extramarital relationships are perceived as common in Lungwena. Women in particular expressed worry about contracting an infection: “When I’m pregnant I am always afraid that my husband might go out with other women and contract STIs. These can cause problems to the unborn baby.” Women rarely knew symptoms associated with STIs, and many pregnant women said that STIs are only diagnosed at the clinic. Men identified some symptoms: for example pain on urination as a symptom of chizonono. Both men and women felt that gonorrhea, syphilis, and chancroid are easily treatable with an injection during a routine ANC visit. As for prevention, women mainly offered suggestions about the husband’s behavior, as a young pregnant woman described: “We do tell them [husbands] that I’ve missed my period so he should start counting the months of pregnancy. Then I tell him that now that I am pregnant, he should avoid sleeping around to prevent miscarriage.” Women’s responses suggest they can only hope that the husband takes their advice seriously because they have no agency to control his sexual desires during pregnancy. One woman’s solution was to advise her husband to take a second wife to avoid

Malaria, Danger, and Risk Perceptions in Rural Malawi

407

infidelity. None of the women mentioned the possibility of using a condom, as opposed to men who mentioned condom use, but only in relation to extramarital sex, not within the marriage. Divorce, although nowadays common in Malawi (Schatz 2005), is not a viable option for the women in Lungwena. Many men said that it is common for girls as young as 10–15 years to become pregnant in the hope of getting married. According to men, it is the parents who encourage their daughters to get pregnant and married at a young age because of the poverty in the area, as explained by one father: “These days girls are getting married between 12 and 14 years. Because of poverty, you can’t afford to feed, clothe, etc. all your children. So when someone comes and wants to marry your daughter, you can’t resist.” Men also recognized abstinence and fidelity as ways to prevent transmission of STIs. Many women considered STIs the most dangerous illnesses for pregnant women because of the severe consequences such as miscarriage, maternal death, death of a newborn, being ill, and danger of stillbirth. Many men were also knowledgeable about adverse consequences of STIs during pregnancy. HIV/AIDS: Severe Consequences, No Treatment, No Prevention, and No Cure. In 2002, HIV/AIDS was rarely mentioned as a dangerous illness. In women’s narratives there were only suspicions that some of the deaths in the villages were because of HIV/AIDS. Four years later, both women and men mentioned HIV/AIDS as one of the dangerous illnesses in Lungwena, indicating a significant change in awareness. Nevertheless, when asked is HIV/AIDS is common, women and men said that they just hear stories and suspect that someone has died of AIDS. Similarly, as elsewhere in Malawi, both men and women attributed HIV/AIDS infection to extramarital sexual relationships, casual sex outside of marriage, and infidelity (MANET 2003; Thorsen et al. 2008). They associated thinness and feeling sick on and off as typical symptoms of HIV/AIDS, although some said that one could not tell if a person is HIV+. They also considered HIV/AIDS very dangerous because it has no treatment and it cannot be cured; as a father of three bluntly stated, “The only way to find out if she is positive is to go to the clinic for a blood test. If she’s got it, there is no way you can treat her. That’s the end of her, she will die.” Women also felt that it was hard to prevent HIV/AIDS in marriage because it would mean abstaining from sex or divorcing one’s husband, as explained by one older woman in a FGD: “The only way to avoid getting HIV/AIDS is to divorce him [husband] once you have seen that he has love affairs with other women, outsiders. There is no way you can say no to your husband when you are living in the same household, sleeping in the same bed.” Many men also said that HIV/AIDS can be prevented by abstinence and faithfulness in marriage although it is not easy, as explained by one male respondent: “The men here in Kapinjiri are not faithful, they are going out with other women. Yet the best way to prevent a pregnant woman from contracting STIs is to be faithful to your partner and to avoid going out with extramarital sex partners.” Only one man said that condoms could be used to prevent HIV/AIDS, although many said that condoms could be used to prevent common STIs. Both men and women considered HIV/AIDS the most dangerous disease because there is no treatment or cure, and eventually the pregnant woman will die.

408

Medical Anthropology Quarterly

“Malungo Is Dangerous for a Pregnant Woman, but We Don’t Really Take It Seriously.” Yao meanings, etiology, symptoms, and treatment and prevention practices of malungo in pregnancy were examined from multiple perspectives. A diversity of views was observed. Concerning prevalence of malungo in pregnancy some pregnant women considered it common while others less common because “most women take preventive measures.” The young pregnant girls were not sure about their opinion, while the women of reproductive age perceived malungo as common. In the FGDs involving older women, one respondent explained, “We don’t know if malungo is common, but after the establishment of hospitals, women go to hospital for malungo treatment.” Men and women specified several symptoms of malungo ranging from fever, high fever, joint pain, headache, feeling cold, body pains, shivering, vomiting, diarrhea, to back pain; others did not know any symptoms. As for the etiology of malungo, again men and women reported multiple causes varying from mosquitoes to cold weather, bathing in the lake, hard work, unsafe water, poor diet, and ignorance. Pregnant women were more hesitant in their responses; some said they don’t know while others said “people say that it is mosquitoes.” Some women of reproductive age also explained that not all mosquitoes cause malungo. None of the men and women associated normal or severe malungo in pregnancy with witchcraft or evil spirits (majini). When listing all the different types of malungo, the women said that there is no such malungo type as malungo wa majini.9 During fieldwork in 2002, we discovered that the Yao recognize an indigenous children’s illness called kambanga that manifests through convulsions. Kambanga is most often interpreted to be caused by witchcraft, but others attribute the causes to cerebral malaria or epilepsy. Both women and men considered adults and pregnant women not to suffer from kambanga; only small children do. Several studies in sub-Saharan Africa have reported a similar illness considered life threatening in children under five that seems to fit a medical diagnosis of cerebral malaria marked by febrile convulsions. For example, there is degedege in Tanzania (Kamat 2008) and eyabwe in Uganda (Launiala and Raijas-Walch 1999). Both women and men felt that for treatment of malungo in pregnancy a pregnant woman should go to a clinic for mapilisi (pharmaR ceuticals). The pregnant women were more precise defining mapilisi as Fansidar  R (SP) or Panadol . Prevention methods in pregnancy varied from insecticide-treated nets (ITNs) to balanced diet, cutting down the grass, and getting medication (SP, mapilisi) in advance from the clinic (Launiala and Kulmala 2006; Launiala and Honkasalo 2007). The perceived consequences of malungo during pregnancy range widely from no consequences to uncomfortable feelings and unconsciousness, maternal death if cerebral malaria is not treated in time, and miscarriage, to very dangerous; if a pregnant woman is reluctant to go to the ANC the baby will be born weak and prematurely. Some also said that consequences depend on God’s will. According to the pregnant women, a severe malungo could result in miscarriage, and if the woman is unlucky she will die. Otherwise a pregnant woman just becomes dehydrated and too weak to work. A mother of five explained, “Malungo in pregnancy has no R at the ANC, and even if you consequences because you are just given Fansidar don’t take it you just feel sick every now and again.” The women of reproductive age said that malungo during pregnancy could cause miscarriage or maternal death if the woman does not receive treatment in time.

Malaria, Danger, and Risk Perceptions in Rural Malawi

409

The severity of danger from malungo was attributed to several factors. Men identified dangerousness with the fact that malungo attacks anyone suddenly. A pregnant woman suffering from severe malungo is likely to die. Women’s opinions varied as follows: “Malungo is dangerous for pregnant women, but we don’t really take it seriously” (pregnant woman); “Malungo is a dangerous illness for pregnant women because you can miscarry if you don’t take enough treatment. We don’t take it seriously, we just take aspirin. HIV/AIDS is more dangerous than malungo because you can treat malungo, while HIV/AIDS has no treatment” (mother of one); “Anemia can lead to maternal death and you cannot deliver the child. Malungo is easier because if you are dehydrated you can get a water drip and all [delivery] goes well” (woman, age 26). Young pregnant girls were not able to explain why malungo is dangerous, and among women of reproductive age there was no consensus, as the following FGD shows: Moderator: You have mentioned malungo, sores, influenza, and gonorrhea. Which one of these illnesses is the most dangerous and why? All R: All the illnesses are dangerous. M: We want to know which one of them is the most dangerous. R4: My friend will tell. R5: Gonorrhea because it causes miscarriage and a woman can deliver a dead baby. R4: No, the way I see it, the most dangerous illness is all of them because one dies. R6: I think that malungo is the most dangerous illness because one can fall sick around 6 a.m. and by 6 or 7 p.m. s/he will die if no proper medication is taken.

Multiple Worries about Everyday Life Realizing that we had indeed taken for granted that malaria is perceived as a major problem in pregnancy, we extended the study to focus broadly on worries about pregnancy itself. We sought to learn if malaria emerges as one among several everyday worries about pregnancy and if not, what kinds of worries pregnant women and men expressed in their narratives. Women and men expressed numerous worries. A young single girl, pregnant for the first time, worried greatly about her life situation, “I was in a boy–girlfriend relationship. When I got pregnant he ran away and went to live with his mother, who is a widow. I live with my mother and we don’t have soap, salt, sugar, or even paraffin. It worries me that I don’t have clothes. I also worry about food.” A woman, age 28, six pregnancies and mother of four, explained that she worries about the delivery: “My main worry is how the delivery will go. We do pray for a good delivery and not to undergo surgery.” A woman, age 27, mother of one, worried about her husband’s behavior: “I worry most that my husband goes out

410

Medical Anthropology Quarterly

with other ladies because he can contract diseases, like chindoko, chizonono, and chibomu, that can destroy me and the unborn baby.” Other worries ranged from illintentioned relatives who can bewitch a pregnant woman to fear of miscarriage, fear of surgery because of a narrow birth canal, worry about falling ill, and inability to walk and do household chores. Women also tended to worry about the things they had experienced and knew. Similarly, as Abel and Browner (1998) have shown that pregnant women make use of their own and other women’s experiential knowledge from previous pregnancies, a pregnant woman, age 26, described her experiences as follows: “When I was pregnant for the first time, I didn’t worry about anything. I thought all was well, not knowing that all deliveries can have complications, and mine is no exception. Now that I’m pregnant for the second time, I worry about the delivery and how it will be. I only pray to God to help me. We have heard that many women are being referred to Mangochi District Hospital because of complications such as having a leg, an arm, or sometimes the whole body out with only the head remaining inside.” Women of reproductive age had similar concerns. One woman said, “When I’m pregnant I’m always worried that someone can harm me because there are people who have a negative attitude and ill intentions toward you.” Another woman explained, “Bewitchment usually happens at the clan level. When I’m pregnant some other relatives from other families are jealous and take advantage of my pregnancy. So they bewitch you and on the day of delivery you die. But this is not very common.” Women also related worries to unfaithful husbands, extramarital relationships, STDs, uncertainties concerning the pregnancy outcome, and economic constraints during delivery, as one FGD respondent described: “You are not certain as to where you will deliver, at home or in a hospital. If in hospital, what means of transport do you use if you don’t have any money?” The results from young girls’ FGD pointed in the same direction. Girls mainly worried about the delivery process as they heard women encounter many problems and end up with an operation. Men expressed multiple worries about pregnancy and particularly the time of delivery. Many of the men explained that in the early stage of a pregnancy (first trimester) there is no need to worry because the wife may not even be pregnant; her body is just resting (kupumula). One husband reported that if there are signs of problems he just sends her wife to the ANC because the nurses there know best what to do. Trust in the ANC services and professionalism of nurses was expressed by many men, and also women (Launiala and Honkasalo 2007). Once it is sure that the wife is pregnant, several factors can threaten the pregnancy. Conflicts and misunderstandings among family members was a concern for many men: “When there are no misunderstandings on either side of the families, we don’t fear anything,” explained a young father of three. The time of delivery induced anxiety among many of the men because of the uncertainty related to it, “Some women are destined to die or experience stillbirth, you never know. Anything can happen also on the day of delivery” (father of two, age 30). Worries were particularly related to the fear of not being able to perform one’s duties as a husband and head of household: “You never know when she will deliver, what time and what will happen, or whether you have enough money for the transport” (a FGD respondent). A 27-year-old father of two described his distress as follows: “I’m always under pressure when she is about to deliver. I do ask myself: is she going to deliver well or

Malaria, Danger, and Risk Perceptions in Rural Malawi

411

not? So I do try to get some money in advance so that if anything happens I will be able to assist her accordingly. Maybe she can give birth through operation. The operation has nothing to do with a particular woman; any woman might have to undergo surgery whether she has given birth several times or not, it can happen any time.” The economic and structural factors constituted clear barriers for men to carry out their duties. As a father from Rashid village described: “In most cases she can go into labor suddenly. Most of us here are poor. Once the time has come, you can’t take that woman on a bicycle. Imagine the delivery starting at midnight and we are supposed to walk to Lungwena 6–7 kilometers. It is no laughing matter. The clinic is far from us, this is the problem we are facing. Sometimes it happens that the baby is born on the way to the clinic. All this is because of poverty: had we not been poor, we could have just taken her by car.”

Discussion To summarize, the Yao in Lungwena perceived coexisting, interlinked, and overlapping everyday dangers ranging from witchcraft, extramarital relationships, and multiple illnesses to worries about uncertainties regarding delivery and financial constraints. In the worst case scenario, these dangers and worries manifest as miscarriages, maternal deaths, disappearing pregnancies, failure to fulfill the role of husband, divorces, and loss of economic support. Hence, Nichter’s (2008:60) argument—that risks should be judged in contexts where there are multiple diseases, health concerns, and dangers and in relation to other risks—is highly valid as it allows us to see why malaria in pregnancy does not emerge as a major concern among women and men in Lungwena. The goal of a pregnant woman and her husband living in Lungwena is to do everything in their power to ensure a positive pregnancy outcome: a live child and a live mother. This means that all their efforts focus on prevention of reproductive threats caused either by natural or by personalistic agents that may have an adverse pregnancy outcome. Nevertheless, in a resource-poor setting such as Lungwena, where people have limited cash-earning opportunities, lack of money, periodic hunger, long distance to care, and so on, this means that the multiple threats and worries are valued differently and prioritized based on their relevance at a given time. This study identified a wide range of dangerous illnesses that all may potentially cause a miscarriage or even a maternal death, yet the perceived risk varies according to the illness and has an impact on the treatment and prevention strategies used. Based on these data we suggest that there are four main factors that contribute to the perception of risk in terms of how dangerous an illness is perceived to be. First is any illness with perceived adverse consequences (e.g., miscarriage, maternal death). STIs, HIV/AIDS, and malungo may have adverse consequences. As for malungo, however, women added that malungo has no consequences because of the mapilisi (referring to intermittent preventive treatment)10 received at the ANC or that there are consequences only if not treated in time or if the pregnant woman is reluctant to attend the ANC. These women share an assumption that a pregnant woman does not need to worry about malaria as long as she attends the ANC and receives the mapilisi. Second is the availability and ease of treatments and cures: the more difficult to treat and cure then the more dangerous the illness is perceived. STIs and malungo

412

Medical Anthropology Quarterly

are possible and easy to treat at the ANC, whereas HIV/AIDS has no treatment and no cure is making it a really worrisome illness. Third is the transmission mode (natural or personalistic agents): how a pregnant woman contracts an illness and whether she has agency to control the risk of getting infected. STIs, HIV/AIDS, and malungo are all caused by natural agents. The difference between these illnesses is that infection with STIs and HIV/AIDS involves other people through sexual intercourse, and a pregnant woman’s agency is limited when it comes to controlling the infection. Malungo infection is possible to prevent by applying the public health means, namely IPT administrated at the ANC and sleeping under an ITN. Fourth, the type of risk is meaningful: does the illness expose the pregnant woman to medical risks that she can control or social risks that a pregnant woman has limited ability to control? Malungo exposes her to medical risks, while STIs and particularly HIV/AIDS, because of extramarital relationships, also expose her to social risks (risk to valued relationship) that can outweigh medical risks. Analysis shows that extramarital relationships are a common practice among the men in Lungwena, putting pregnant women in danger of STIs and HIV/AIDS. Yet, married women have little power to influence their husband’s behavior. Divorce is not a possibility for these women living in extreme poverty. They are dependent on their husband’s support in terms of food, clothes, and medicine, as most of them have no economic resources to take care of these basic needs. Marriage, particularly during pregnancy, becomes an economic necessity (Chapman 2006; Chimbiri 2007; Schatz 2005). Condom use in marriage is not an option either as it may signal misconduct from the woman’s side, and a woman caught in adultery is likely to be divorced (Chimbiri 2007:1110; Shatz 2005:480; see also Bujra 2000:74). Although it may not be a conscious choice made by women in Lungwena, this ethnography depicts how a pregnant woman often ends up accepting immediate disease and medical risks of STIs and HIV/AIDS rather than placing herself at a social risk of being divorced by her husband, losing economic support, and even being cast out by her family and relatives because of HIV-related stigma (MANET 2003; Shatz 2005; Thorsen et al. 2008; see Nichter 2008:58–59 on social risk). Another explanation for accepting the medical risks of STIs can be that women and men consider STIs easily treatable at the Lungwena ANC. In the context of several dangerous illnesses that expose a pregnant woman to medical and social risk, we can understand why malaria in pregnancy is not such a big worry despite the medical risks it causes to a pregnant woman. A study undertaken in rural Malawi using psychological analysis of risk suggested that individuals’ risk perception of malaria bore no influence on whether they engaged in recommended behavior, and that social factors may affect compliance with malaria control and treatment procedures (Ager 1992). Sommerfeld et al. (2002) found in Burkina-Faso that getting AIDS was ranked highest in terms of severity but lowest in perceived personal vulnerability. In contrast, malaria (soymaya) was ranked lowest in perceived severity and highest in perceived vulnerability. They argue that people fear low-probability and highintensity risks (getting struck by lightning, getting AIDS, falling mentally ill) as much as high-probability risks (drought, lack of funds to buy medicine, see also Ager 1992 on assessment of risk perceptions). Strong concerns about witchcraft are interwoven into pregnant women’s lives. Although not a constant worry, both women and men considered witchcraft a

Malaria, Danger, and Risk Perceptions in Rural Malawi

413

common phenomenon in the communities, thus increasing the vulnerability of a pregnant woman. The presence of witchcraft becomes visible when the condition of a pregnant woman suddenly changes, putting her and the unborn baby in danger, and in the absence of a clear explanation for the situation. This is, for example, evident in the case of prolonged labor. When the labor takes too long, it arouses a suspicion of witchcraft and evil spirits, and the cultural interpretation is that the woman is suffering from malowe. Instead of taking the woman urgently to emergency obstetric care, the therapy management group tries to solve the situation by themselves or consult a sing’anga. Chapman (2006) described how the women in Mozambique are particularly worried about personalistic reproductive threats and prefer to utilize the popular treatment providers and folk treatments to deal with these threats (see also Allen 2002). Also when a pregnant woman is suffering from an illness and there is a sudden turn for the worse, or her suffering is persistent, suspicions about witchcraft and the involvement of evil spirits may arise. Hausmann-Muela and colleagues (1998) showed in Tanzania that when malaria cannot be diagnosed in an official biomedical health care facility or symptoms persist despite antimalarial treatment, the cultural model of malaria may be complemented with the logic of witchcraft. In Lungwena, where witchcraft is considered a common phenomenon and illness causalities are interpreted in a naturalistic and personalistic framework, it is certainly possible that the cultural illness models of STIs and HIV/AIDS can also be complemented with the logic of witchcraft if symptoms persist and there is no clear diagnosis. This, however, requires more investigation into STI-treatment–seeking practices. Nevertheless, women and men in Lungwena never complemented the cultural model of malaria in pregnancy with the logic of witchcraft or evil spirits, maybe because pregnant women rarely manifest convulsions and unconsciousness in areas with stable malaria transmission (WHO 2004:4).11 Household assets are a key limit to health care (Obrist et al. 2007:1858), and the majority of families in Lungwena lack these, unlike Onni’s father who did have an irregular income. Suddenly raising even 100MK ($0.80) for transport is often impossible. Men’s worries regarding possible delivery complications are real because in case of an obstetric emergency, the families are referred to Malindi or to Mangochi district hospital for a cesarean section, that is, referred to distant services not readily accessible because of costs. Onni’s story and women’s and men’s worries expressed show that it is very important that, in addition to the sociocultural factors that affect risk perceptions and treatment-seeking practices, we recognize the macrolevel factors: structural and economic factors that affect the outcome (see Launiala forthcoming 2010). Kamat (2008) has shown further how the political, economic, and structural factors that influence malaria-treatment–seeking practices among the Zaramo in Tanzania. Obrist and colleagues (2007:1587) highlight factors that affect access to health care in the broader context of livelihood insecurity because “unless additional efforts are made to enable poor people to gain access to these goods and services, and to more basic livelihood assets required to initiate treatment seeking, equitable access remains an empty formula of politicians and experts.” Farmer et al. (2006) use the concept of structural violence to highlight the social forces beyond the control of patients in resource-poor settings. Premature

414

Medical Anthropology Quarterly

deaths because of structural violence can only be reduced by efforts to improve equal access to health care. Malaria in pregnancy receives surprisingly little research attention despite its high priority on the public health agenda of WHO, UNICEF, and other Roll Back Malaria (RBM) partners (WHO 2008). The few studies on malaria in pregnancy have restricted their focus on treatment-seeking practices and prevention in a particular sociocultural context (Ahorlu et al. 2007; Mbonye et al. 2006; Ndyomugyenyi et al. 1998; Williams and Jones 2004). Ribera and colleagues (2007) call social scientists to go beyond simplified measures of knowledge, attitudes, and practices and incorporate sociocultural context, recognition of illness, perceived severity and susceptibility, perceived benefits, risks and capacity for action, and availability and accessibility. The models suggested by Ribera and colleagues (2007) are comprehensive in terms of paying attention to contextual factors, dynamic relationships, and community reaction concerning malaria in pregnancy. We argue that malaria in pregnancy should also be studied in the context of pregnancy itself: perceived vulnerabilities, exposure to multiple dangerous illnesses, and worries in everyday life. As Kamat (2008:86) has pointed out, “Malaria does not stand alone as an isolated issue in people’s lives, especially in resource-poor communities where people are confronted with other pressing concerns on a daily basis.” Thus, malaria needs to be explored in the context of pregnant women’s daily lives. It is also crucial to pay attention to the social relations of extended family members and decision-making practices in a household to have a better understanding of maternal-health therapy management. The diverse perceptions recorded in our study show that there is a need to extend the focus beyond pregnant women themselves. Although our study shows men have a major role in maintaining the well-being of a woman in resource-poor settings, men’s experiences and role in pregnancy and delivery remain relatively neglected in medical anthropology (Dudgeon and Inhorn 2004), as well as the role of other family members. To conclude, in the context of multiple risks, dangers, and worries, the epidemiological risk definition and the “at risk” language of international organizations is limited because it neglects local perceptions of vulnerabilities and perceived dangers. Chapman (2006:493) has suggested that the language of “reproductive threats” or “maternal vulnerabilities” is more appropriate and useful than the “at risk” discourse in settings such as Lungwena because it allows a wide scope for threats meaningful for the Yao women. Allen (2002:10–11) also showed in her study how two different systems of maternal health care in Tanzania, namely the biomedical and the nonbiomedical, address and respond differently to maternal risks. The Safe Motherhood strategy is responding to the “risk of motherhood,” while women are responding through local healing strategies to “risks to motherhood.” Our study also shows that both women and men in Lungwena were concerned about the dangers and threats to motherhood, and efforts were made to ensure that pregnancies would be carried to term. We suggest that the “at risk” approach used by international organizations should be modified to better respond to local risk perceptions. It is also important to recognize that malaria is not isolated from the whole of a pregnant woman’s life. Policies and programs to reduce malaria deaths and illness must address the complexity of perceived vulnerabilities and structural and economical

Malaria, Danger, and Risk Perceptions in Rural Malawi

415

conditions in resource-poor settings. This will enable more effective partnerships between malaria prevention and safe motherhood programs.

Notes Acknowledgments. Sincere thanks to the women and men in Lungwena, Lungwena Health Center staff, and research assistants: Shaibu, Innocent, and Stephano. Doctoral Programs in Public Health (DPPH) and the Academy of Finland project #205648 supported the study. Thanks to Jussi Kauhanen (School of Public Health and Clinical Nutrition, University of Kuopio) for institutional support, Pamela Feldman-Savelsberg (Department of Sociology and Anthropology, Carleton College) for manuscript comments, and anonymous reviewers for truly constructive comments. Special thanks to Marja-Liisa Honkasalo, my Ph.D. supervisor, for research guidance and contributions to this article. 1. We asked participants to classify illnesses considered dangerous to pregnant women, their severity and why, and to describe symptoms, causes, treatment, prevention, and consequences during pregnancy. Pregnancy complications were classified similarly. If malungo was not identified, it was probed. This method was designed to learn Yao’s shared criteria for defining when an illness is dangerous for a pregnant woman. 2. Ethical clearance and research permit was granted by the National Health Sciences Research Committee of the Ministry of Health and Population (MOHP). The College of Medicine Research Ethical Committee of the University of Malawi was also informed of the research. 3. Women identified these food items and harms to the baby: eggs (babies do not come out), maize (stomach ache after birth), pepper (blindness), tomatoes (skin sores), fish (allergies), soaked rice (baby is born dirty), sugar cane (skin peels off and pains), food from sadaka funeral feast (miscarriage), and food from any public ceremony (might be prepared by woman menstruating, or miscarried, or had recent sexual intercourse). Many who have reported that despite the beliefs, they may eat eggs, maize, or fish, if available. Given the poverty and annual hunger periods, adherence to food restrictions may be low. HelitzerAllen and colleagues (1997) report similar restrictions among the Yao on the other side of Lake Malawi. 4. Malowe was described as a curse caused when a pregnant woman quarrels with someone, which makes the baby to refuse to emerge until the “bad words” are reconciled. Hence, when the delivery is in process relatives ask people to come and confess quarrelling to take back “bad words” so the baby will emerge. Malowe may also be because of female infidelity, and the woman must confess and name the father. Prolonged labor suggests a need for paternity proof, conducted by the female relatives of the man whose wife is in labor. If needed, treatment is sought from a sing’anga; maternal services are often used as a last resort. 5. Litiwo is still known to be practiced at least in the Chikwawa, Thyolo, and Mangochi districts (Helitzer-Allen et al. 1997; Zulu 1996:32–35). In the study area (Mangochi district) not all primigravidae participated in litiwo. 6. Illnesses and symptoms elicited during interviews and FGDs were: gonorrehea, eclampsia, influenza, mwana mbepo, sores, asthma, chipata, sleeping sickness, back pain, malnutrition, mauka, likango, vomiting, blood pressure, epilepsy (kambanga), stomach pain, cholera, diarrhea, swollen legs, ndaka, coughing, anemia, syphilis, HIV/AIDS, severe malungo, malungo, and chancroid.

416

Medical Anthropology Quarterly

7. Following Green (1999:135) I distinguish sexually transmitted diseases (STD) from sexually transmitted illnesses (STI). However, I use STI to refer to both biomedically defined STDs and local indigenous sexual illnesses. 8. Kambanga is commonly viewed as a children’s illness manifested through convulsions, caused by evil spirits. Two men considered kambanga the most dangerous illness for a pregnant woman because it can cause unconsciousness, and not because of evil spirits but because of epilepsy. Sometimes epilepsy is called kambanga. 9. Most women knew that majini causes illnesses, but they argued that there is no such malungo type as malungo wa majini. Yet the majority of women also recognized that witches (afiti) regularly cause all kinds of misfortunes and illnesses in the area (see Launiala and Kulmala 2006). 10. Malawi national malaria policy since 2002 includes an intermittent preventive treatment (IPT) policy whereby all pregnant women receive two treatment doses of SP at least one month apart in weeks at the antenatal clinic and at TBA venues under direct observed therapy (DOT; see Launiala and Honkasalo 2007). 11. In areas where malaria transmission is stable (prevalence during pregnancy ranges from 10 percent to 65 percent), malaria in pregnancy is often an asymptomatic infection that leads to anemia (WHO 2004:4).

References Cited Abel, Emily K., and C. H. Browner 1998 Selective Compliance with Biomedical Authority and the Uses of Experiential Knowledge. In Pragmatic Women and Body Politics. Margaret Lock and Patricia A. Kaufert, eds. Pp. 310–326. Cambridge: Cambridge University Press. Ager, A. 1992 Perceptions of Risk for Malaria and Schistosomiasis in Rural Malawi. Tropical Medicine and Parasitological 43:234–238. Ahorlu, Collins K., Kwadwo A. Koram, and Mitchell G. Weiss 2007 Children, Pregnant Women and the Culture of Malaria in Two Rural Communities of Ghana. Anthropology and Medicine 14(2):167–181. Allen, Denise Roth 2002 Managing Motherhood, Managing Risk: Fertility and Danger in West Central Tanzania. Ann Arbor: University of Michigan Press. Benson, Todd D. 2002 Malawi: An Atlas of Social Statistics. Washington, DC: National Statistics Office, Zomba and International Food Policy Research Institute. Bernard, H. Russell 2002 Research Methods in Anthropology: Qualitative and Quantitative Approaches. 3rd edition. Walnut Creek, CA: AltaMira. Bisika, Thomas 2008 Cultural Factors that Affect Sexual and Reproductive Health in Malawi. Journal of Family Planning Reproductive Health Care 32(2):79–80. Bledsoe, Caroline 2002 Contingent Lives: Fertility, Time, and Aging in West Africa. Chicago: Chicago University Press. Breugel, J. W. M. 2001 Chewa Traditional Religion. Malawi: Christian Literature Association in Malawi. Bujra, Janet 2000 Risk and Trust: Unsafe Sex, Gender and AIDS in Tanzania. In Risk Revisited. Pat Caplan, ed. Pp, 59–84. London: Pluto.

Malaria, Danger, and Risk Perceptions in Rural Malawi

417

Caplan, Pat 2000. Risk Revisited. London: Pluto. Chapman, Rachel R. 2006 Chikotsa: Secrets, Silence, and Hiding: Social Risk and Reproductive Vulnerability in Central Mozambique. Medical Anthropology Quarterly 20(4):487– 515. Chimbiri, Agnes 2007 The Condom Is an “Intruder” in Marriage: Evidence from Rural Malawi. Social Science and Medicine 64:1102–1115. Day, Sophie 2000 The Politics of Risk among London Prostitutes. In Risk Revisited. Pat Caplan, ed. Pp. 29–58. London: Pluto. Dellicour, Stephanie, Andrew J. Tatem, Carlos A. Guerra, Robert W. Snow, and Feiko O. ter Kuile 2010 Quantifying the Number of Pregnancies at Risk of Malaria in 2007: A Demographic Study. PLoS Medicine. http://www.plosmedicine.org/article/info%3Adoi%2 F10.1371%2Fjournal.pmed.1000221;jsessionid=7994CB45A9DB05840FBC712A 859174A4.ambra02. Accessed June 4, 2010. Douglas, Mary 1992 Risk and Blame. Essays in Cultural Theory. London: Routledge. Dudgeon, Matthew R., and Marcia C. Inhorn 2004 Men’s Influences on Women’s Reproductive Health. Medical Anthropology Perspectives. Social Science and Medicine 59:1379–1395. Englund, Harri 1996 Witchcraft, Modernity and the Person: The Morality of Accumulation in Central Malawi. Critique of Anthropology 16(3):257–279. Eriksen, Thomas Hylland 2001 Small Places, Large Issues: An Introduction to Social and Cultural Anthropology. 2nd Edition. London: Pluto. Farmer, Paul, Bruce Nizeye, Sara Stulac, and Salmaan Keshavjee 2006 Structural Violence and Clinical Medicine. PLoS Medicine 3(10):e449. http://www.plosmedicine.org/article/info%3Adoi%2F10.1371%2Fjournal.pmed. 0030449. Accessed November 15, 2009. Feldman-Savelsberg, Pamela 1994 Plundered Kitchens and Empty Wombs: Fear of Infertility in the Cameroonian Grassfields. Social Science and Medicine 39:463–474. Foster G. 1976 Disease Etiologies in Non-Western Medical Systems. American Anthropologist 78(4):773–782. Frankenberg, Ronald 1993 Risk: Anthropological and Epidemiological Narratives of Prevention. In Knowledge, Power, and Practice: The Anthropology of Medicine and Everyday Life. Shirley Lindenbaum and Margaret Lock, eds. Pp. 219–244. Berkeley: University of California Press. Gordis, Leon 2004 Epidemiology. 3rd edition. Philadelphia: Elsevier Saunders. Green, Edward 1999 Indigenous Theories of Contagious Disease. Walnut Creek, CA: AltaMira. Haram, Liv 2005 AIDS and Risk: The Handling of Uncertainty in Northern Tanzania. Culture, Health and Sexuality 7(1):1–13.

418

Medical Anthropology Quarterly

Hausmann-Muela, Susanne, Joan Muela Ribera, and Marcel Tanner 1998 Fake Malaria and Hidden Parasites: The Ambiguity of Malaria. Anthropology and Medicine 5:43–61. Helitzer-Allen, Deborah L., Carl Kendall, and Jack J. Wirima 1997 Beliefs and Practices during Pregnancy in a Yao community in Malawi. In Issues and Perspectives on Health Care in Contemporary Sub-Saharan Africa. E. Kalipeni and P. Thiuri, eds. Pp. 145–164. Lewiston, NY: Edwin Mellen. Helitzer-Allen, Deborah L., Allan Macheso, Jack J. Wirima, and Carl Kendall 1994 Testing Strategies to Increase Use of Chloroquine Chemoprophylaxis during Pregnancy in Malawi. Acta Tropica 58:255–266. Honkasalo, Marja-Liisa 2006 Fragilities in Life and Death: Engaging in Uncertainty in Modern Society. Health, Risk and Society 8(1):27–41. Janzen, John 1978 The Quest for Therapy: Medical Pluralism in Lower Zaire. Berkeley: University of California Press. Kamat, Vinay 2008 Dying under the Bird’s Shadow: Narrative Representations of Degedege and Child Survival among the Zaramo of Tanzania. Medical Anthropology Quarterly 22(1):67–93. Launiala, Annika 2009 KAP Survey in a Non-Western Setting: A Critical Discussion Based on a Medical Anthropology Study on Malaria in Pregnancy in Rural Malawi. Anthropology Matters. http://www.anthropologymatters.com./ Accessed November 7, 2009. Launiala, Annika (forthcoming 2010) Prevention of Malaria in Pregnancy in Malawi: Encounters and Discounters of Global Policies, National Programs and Local Realities. Academic dissertation. Tampere: Tampere University Press. Launiala, Annika, and Marja-Liisa Honkasalo 2007 Ethnographic Study on Factors Affecting Compliance to Intermittent Preventive Treatment of Malaria during Pregnancy among the Yao Women in Rural Malawi. Transactions of the Royal Society for Tropical Medicine and Hygiene 101(10):980– 989. Launiala, Annika, and Teija Kulmala 2006 The Importance of Understanding the Local Context: Women’s Perceptions and Knowledge Concerning Malaria in Pregnancy in Rural Malawi. Acta Tropica 98:111–117. Launiala, Annika, and Helena Raijas-Walch 1999 Biomedical and Local Concepts of Malaria among the Baganda in Uganda. Journal of the Finnish Anthropological Society 24(4):142–153. Lupton, Deborah 1999 Risk. London: Routledge. Lwanda, John 2001 Tikutha: The Political Culture of the HIV/AIDS Epidemic in Malawi. In A Democracy of Chameleons: Politics and Chameleons in the New Malawi. Harri Englund, ed. Pp. 151–165. Stockholm: Elanders Gotab with the Nordic Africa Institute. MANET 2003 Voices for Equality and Dignity: Qualitative Research on Stigma and Discrimination Issues as They Affect PLWHA in Malawi. http://www.manetplus.com/ Resources/Downloadpage.htm. Accessed November 5, 2009.

Malaria, Danger, and Risk Perceptions in Rural Malawi

419

Mbonye, Anthony K., Stella Neema, and Pascal Magnussen 2006 Preventing Malaria in Pregnancy: A Study of Perceptions and Policy Implications in Mukono District, Uganda. Health Policy and Planning 21(1):17–26. Mitchell, J. Clyde 1956 The Yao Village: A Study in the Social Structure of a Malawian People. Manchester: Manchester University Press. Mnyika, K. S., J. Z. J. Killewo, D. K. Kabalimu 1995 Self-Medication with Antimalarial Drugs in Dar es Salaam, Tanzania. Tropical and Geographical Medicine 47:32–34. Morris, Brian 1985 Chewa Conceptions of Disease: Symptoms and Etiologies. Society of Malawi Journal 38:14–38. Ndyomugyenyi, R., Stella Neema, and Pascal Magnussen 1998 The Use of Formal and Informal Services for Antenatal Care and Malaria Treatment in Rural Uganda. Research Report. Health Policy and Planning 13:94– 102. Nichter, Mark 2003 Harm Reduction: A Core Concern in Medical Anthropology. In Risk, Culture and Health Inequality: Shifting Perceptions of Blame and Danger. Barbara Herr Harthorn and Laury Oaks, eds. Pp. 13–33. Westport, CT: Praeger. 2008 Global Health. Why Cultural Perceptions, Social Representation, and Biopolitics Matter. Tucson: University of Arizona Press. Obrist, Brigit, Nelly Iteba, Christian Lengeler, Ahmed Makemba, Christopher Mshana, Rose Nathan, Sandra Alba, Angel Dillip, Manuel W. Hetzel, Iddy Mayumana, Alexander Schulze, and Hassan Mshinda 2007 Access to Health Care in Contexts of Livelihood Insecurity: A Framework for Analysis and Action. PLoS Medicine 4(10):e308. http://www.plosmedicine.org/ article/info%3Adoi%2F10.1371%2Fjournal.pmed.0040308. Accessed 15 November, 2009. Paz Soldan, Valerie A., Joseph E. deGraft-Johnson, Thomas Bisika, and Amy O. Tsui 2007 Social, Economic and Demographic Determinants of Sexual Risk Behaviors among Men in rural Malawi: A District-Level Study. African Journal of Reproductive Health 11(2):33–46. Phiri, Isabel A. 1997 Women, Presbyterianism and Patriarchy: Religious Experience of Chewa Women in Central Malawi. Malawi: Christian Literature Association in Malawi. Ribera, Joan Muela, Susanne Hausmann-Muela, Umberto D’Alessandro, and Koen Peeters Grietens 2007 Malaria in Pregnancy: What Can the Social Sciences Contribute? PLoS Medicine 4(4):e92. http://www.plosmedicine.org/article/info%3Adoi%2F10.1371% 2Fjournal.pmed.0040092 (accessed November 20, 2009). Rogerson, Stephen J., Lars Hviid, Patrick E. Duffy, Rose F. G. Leke, and Dane W. Taylor 2007 Malaria in Pregnancy: Pathogenesis and Immunity. Lancet Infectious Disease 7:105–117. Schatz, Enid 2005 “Take Your Mat and Go!”: Rural Malawian Women’s Strategies in the HIV/AIDS Era. Culture, Health and Sexuality 7(5):479–492. Sommerfeld, Johannes, Mamadou Sanon, Bocar A. Kouyat´e, and Rainer Sauerborn 2002 Perceptions of Risk, Vulnerability, and Disease Prevention in Rural Burkina Faso: Implications for Community-Based Health Care and Insurance. Human Organization 61(2):139–146.

420

Medical Anthropology Quarterly

Thorold, Alan 1993 Metamorphoses of the Yao Muslims. In Muslim Identity and Social Change in Sub-Saharan Africa. L. Brenner, ed. Pp. 79–90. Bloomington: Indiana University Press. Thorsen, Viva, Johanne Sundby, and Francis Martison 2008 Potential Initiators of HIV-Related Stigmatization: Ethical and Programmatic Challenges for PMTCT Programs. Developing World Bioethics 8(1):43–50. Trostle, James A. 2005 Epidemiology and Culture. New York: Cambridge University Press. WHO 2004 A Strategic Framework for Malaria Prevention and Control during. Pregnancy in the African Region. WHO Regional Office for Africa, AFR/MAL/04/01. http:// www.cdc.gov/malaria/pdf/strategic_framework_mip_04.pdf. Accessed November 18, 2009. 2008 World Malaria Report. http://malaria.who.int/wmr2008/malaria2008.pdf, Accessed November 7, 2009. Whyte, Susan 1997 Questioning Misfortune. Cambridge: University of Cambridge Press. Williams, Holly A., and Caroline O. H. Jones 2004 A Critical Review of Behavioural Issues Related to Malaria Control in SubSaharan Africa: What Contributions Have Social Scientists Made? Social Science and Medicine 59:501–523. Winch, Peter J., A. M. Makemba, S. R. Kamazima, M. Lurie, G. K. Lwihula, Z. Premji, J. N. Minjas, and C. J. Shiff 1996 Local Terminology for Febrile Illnesses in Bagamoyo District, Tanzania, and Its Impact on the Design of a Community-Based Malaria Control Programme. Social Science and Medicine 42:1057–1067. Young, A. 1976 Internalizing and Externalizing Medical Belief Systems: An Ethiopian Example. Social Science and Medicine 10:147–156. Zulu, Eliya M. 1996 Social and Cultural Factors Affecting Reproductive Behavior in Malawi. Ph.D. dissertation, Population Studies Center. University of Pennsylvania.

Anthropology Matters Journal 2009, Vol 11 (1)

How much can a KAP survey tell us about people’s knowledge, attitudes and practices? Some observations from

medical

anthropology

research

on

malaria

in

pregnancy in Malawi By Annika Launiala (University of Tampere and University of Kuopio, Finland)

Knowledge, attitude, and practice (KAP) surveys are widely used to gather information for planning public health programmes in countries in the South. However, there is rarely any discussion about the usefulness of KAP surveys in providing appropriate data for project planning, and about the various challenges of conducting surveys in different settings. The aim of this article is two-fold: to discuss the appropriateness of KAP surveys in understanding and exploring health-related knowledge, attitudes, and practices, and to describe some of the major challenges encountered in planning and conducting a KAP survey in a specific setting. Practical examples are drawn from a medical anthropology study on socio-cultural factors affecting treatment and prevention of malaria in pregnancy in rural Malawi, southern Africa. The article presents issues that need to be critically assessed and taken into account when planning a KAP survey.

Background: KAP surveys There is increasing recognition within the international aid community that improving the health of poor people across the world depends upon adequate understanding of the socio-cultural and economic aspects of the context in which public health programmes are implemented. Such information has typically been gathered through various types of cross-sectional surveys, the most popular and widely used being the knowledge, attitude, and practice (KAP) survey, also called the knowledge, attitude, behaviour and practice (KABP) survey (Green 2001, Hausmann-Muela et al. 2003, Manderson and Aaby 1992, Nichter 2008:6-7). The KAP survey tradition was first born in the field of family planning and population studies in the 1950s. KAP surveys were designed to measure the extent to which an obvious hostility to the idea and organisation of family planning existed among different populations, and to provide information on the knowledge, attitudes, and practices in family planning that could be used for programme purposes around the world (Cleland 1973, Ratcliffe 1976). In the 1960s and 1970s, KAP surveys began to be utilised for understanding family planning perspectives in Africa (Schopper et al. 1993). Around the same time, the amount of studies on community perspectives and human behaviour grew rapidly in response to the needs of the primary health care approach adopted by international aid organisations. Hence KAP surveys established their place among the methodologies used to investigate health behaviour, and today 1

Anthropology Matters Journal

2009, Vol 11 (1)

http://www.anthropologymatters.com

they continue to be widely used to gain information on health-seeking practices (Hausmann-Muela et al. 2003, Manderson and Aaby 1992). The attractiveness of KAP surveys is attributable to characteristics such as an easy design, quantifiable data, ease of interpretation and concise presentation of results, generalisability of small sample results to a wider population, cross-cultural comparability, speed of implementation, and the ease with which one can train numerators (Bhattacharyya 1997, Stone and Campbell 1984). Nevertheless, over the years some researchers have criticised KAP surveys for taking for granted that the data provided offers accurate information about knowledge, attitudes, and practices that can be used for programme planning purposes (Cleland 1973, Nichter 1993, Pelto and Pelto 1997, Yoder 1997, see also Green 2001). A number of social scientists have also voiced their concern over the applicability of KAP surveys (Cleland 1973, Caldwell et al. 1994, Green 2001, Manderson and Aaby 1992, Nichter 1993, Ratcliffe 1976, Smith 1993). Yet in the international health community and among health programme planners, there is rarely any discussion about whether KAP surveys are an appropriate methodology to explore health-seeking practices that can be used for programme planning or not (Foster 1987). Lately there has not been much critical discussion among social scientists regarding this issue either. My experience with the use of KAP surveys for programme planning comes from Malawi, where I worked as a project officer for UNICEF from 1998 to 2001. During this time I was involved in several KAP surveys conducted by UNICEF in collaboration with their local partners. At that time KAP survey research was common practice in international health (see also Nichter 2008:6-7) Why KAP surveys? In the UNICEF Malawi office, there were several reasons. First of all, there were Malawians who had received training on survey and quantitative research (compared to only two medical anthropologists in the entire country according to my knowledge). Secondly, surveys were easy to conduct, rather cost-effectively, even nationwide. Thirdly, there was an assumption that the results could be generalised nationwide; and, moreover, the results, “hard numbers”, could be used to show progress to the funding agencies. During my three years in UNICEF I became rather doubtful about the usefulness of KAP survey data in programme planning because we rarely discussed the data quality and thus the usefulness of the results (see also Gill 1993). As a matter of fact, the findings were used only to a limited extent for programme purposes. This problem was recognized by many of us, both by national and international staff working in UNICEF as well as Malawian counterparts, but due to time constraints and lack of skills and mechanisms for translating the results into practice, research reports were frequently underutilised. When I started to develop a PhD research plan in 2002 for a medical anthropology study on malaria in pregnancy, I was interested in adding a KAP survey to the study design to gain first-hand practical experience with the method and to clarify my doubts and concerns about it. Thus, in addition to in-depth interviews, focus groups discussions and participant observation, I carried out a KAP survey with the assistance of four local research assistant at the antenatal clinic of the Lungwena Health Centre and in six villages of the health centre catchment area. In the villages 200 interviews were aimed at altogether, with samples proportional to the size of the village. At the antenatal clinic, every third woman was selected over four days. This

2

Annika Launiala

How much can a KAP survey tell us?

sampling procedure resulted in 248 interviews, 200 in the villages and 48 at the Health Centre (see more details Launiala and Kulmala 2006). The aim of this article is two-fold: to discuss the appropriateness of KAP surveys in understanding health-related knowledge, attitudes, and practices, and to describe some of the major challenges encountered in planning and conducting a KAP survey in this setting. This article is therefore divided into two main sections. The first section looks more closely at the main aspects of each element in a KAP survey, and the second concentrates on challenges encountered in the field. Throughout I will draw examples from the KAP survey I conducted as part of my PhD research on socio-cultural factors affecting treatment and prevention of malaria in pregnancy among the Yao in rural Malawi.

Main aspects of a KAP survey Whose knowledge counts?

In KAP surveys, the knowledge part is normally used only to assess the extent of community knowledge about public health concepts related to national and international public health programmes. Investigation of other types of knowledge, such as culture-specific knowledge of illness notions and explanatory models, or knowledge related to health systems, e.g. access, referral, and quality, is highly neglected (Hausmann-Muela et al. 2003). Lack of investigation of illness notions and explanatory models is probably due to the fact that community knowledge is the embodied knowledge of explanatory illness models, and treatment practices. It is contextualised, practice-based, and emergent in times of illness, and, therefore, very difficult to detect using KAP surveys as pointed out by Nichter (1993). The narrow focus on knowledge can further be explained by the definition of knowledge and the agreement on whose knowledge counts. Pelto and Pelto (1997) have pointed out that public health professionals usually share the view that knowledge and beliefs are contrasting terms. They have an implicit assumption that knowledge is based on scientific facts and universal truths (refers to “knowing” about biomedical information). In contrast, beliefs refer to traditional ideas, which are erroneous from the biomedical perspective, and which form obstacles to appropriate behaviour and treatment-seeking practices (see also Good 1994). This narrow definition of knowledge is also shared by international health communities. While they have recognized the role and engagement of communities in the management and prevention of diseases, such as malaria and acute respiratory infections (ARI), they still fail to recognize the value of the knowledge that the communities possess (Nichter 1993). There is, however, no specific reason why knowledge related to health systems are rarely investigated in KAP surveys. In anthropology, knowledge and beliefs are not contrasting terms (Pelto and Pelto 1997). In my study, I considered Yao women’s knowledge to include local indigenous knowledge and beliefs, and biomedical knowledge. For example, during the qualitative phase of my study I investigated the meaning of malungo (a local word used for malaria), and the results revealed that malungo is an ambiguous term with multiple meanings and definitions, which are used interchangeably to refer to many types of feverish illnesses, not just malaria. More than 10 different types of malungo were noted, but I observed that these local categories were vague, ambiguous and not shared by all members of the community. Instead categories were produced and

3

Anthropology Matters Journal

2009, Vol 11 (1)

http://www.anthropologymatters.com

reproduced in encounters with other community members (Launiala and Kulmala 2006). In the KAP survey I also tried to go beyond public health and biomedical knowledge by investigating types of malungo. I asked: “Are there different kinds of malungo?”, and 75% (n=248) of respondents said no, meaning that there is only one type of malungo. Of the remaining 25%, who said yes and were further asked to name the different types in an open-ended question, the majority said there were two types: normal malungo and/or severe malungo. This KAP survey result showed the difficulty (and pointlessness) of asking questions related to local notions of illnesses in the format of a KAP survey. Attitudes – can they be measured?

Measuring attitudes is the second part of a standard KAP survey questionnaire. However, many KAP studies do not present results regarding attitudes, probably because of the substantial risk of falsely generalising the opinions and attitudes of a particular group (Cleland 1973, Hausmann-Muela et al. 2003). In everyday English, the term attitude is usually used to refer to a person’s general feelings about an issue, object, or person (Petty and Cacioppo 1981). Furthermore, attitudes are interlinked with the person’s knowledge, beliefs, emotions, and values, and they are either positive or negative. Pelto and Pelto (1994) have also described causal attitudes or erroneous attitudes, which are considered derivatives of beliefs and/or knowledge. The act of measuring attitudes via a survey has been criticised for many reasons. When confronted with a survey question, people tend to give answers which they believe to be correct or in general acceptable and appreciated. Sensitive topics are particularly demanding. The survey interview context may influence the answer; whether the interview is conducted at a clinic or in a village, whether there are other people present, etc. The question formulation can be manipulative towards a favourable answer. Sometimes, the respondents may be uninformed about the issue and thus find it strange, but their attitudes are nonetheless measured. On occasion, the attitude scales (numbers/verbal) may fail to reflect the respondents’ answers (Cleland 1973, Hausmann-Muela et al. 2003, Pelto and Pelto 1994). I also included a section on attitudes in the KAP questionnaire that I used, following the typical statement formulation with three response alternatives (“agree”, “not sure”, “disagree”). I formulated the statements based on some key findings from the qualitative phase of my study, with the purpose of obtaining a clearer picture regarding whether these findings were shared by a larger proportion of Yao women, or if they were just solitary statements by individuals. The questionnaire contained 11 statements altogether. In addition to three response alternatives, I added an openended question to “disagree” responses in order to gain some understanding of why respondents disagreed. Moreover, I instructed the assistants to mark down when a respondent said that she did not know the answer. Analysis of the results raises some concerns about the possibility of measuring attitudes through a questionnaire. The high proportion of “agree” answers was eyecatching. In nine statements out of 11, less than 10% (24/248) disagreed, between 63% (157/248) and 99% (246/248) agreed, and between 2% (5/248) and 29% (72/248) were not sure. There were slightly more “agree” answers among the women who gave responses at the antenatal clinic than among the village respondents. There was only one statement to which there were more disagreeing than agreeing answers. This statement concerned the role of the maternal uncle: “You ask advice from your maternal uncle when you are severely sick.” To this, only 39% (97/248) agreed. 4

Annika Launiala

How much can a KAP survey tell us?

There may be several explanations for the high proportion of agreeing answers. One explanation could be that there is indeed strong agreement and cultural homogeneity among the Yao women. However, when taking into account the Yao women’s sociocultural background, which often includes little formal (Western-style) schooling and which emphasises the value of being non-confrontational, it is also possible that the question formulation can influence attitudes towards favourable, “agreeing” answers. For example, 99% (246/248) agreed to the statement which put forth the argument: “You can trust the advice and medication given by the nurses, because nurses are educated.” This statement was formulated on the basis of the qualitative findings, but the problem is that it would require a lot of courage from the women to disagree with this statement even if they thought otherwise. I concur with other studies (Cleland 1973, Hausmann-Muela et al. 2003) that researchers should be very cautious regarding the interpretation of results related to attitude measurement. It is important to take into consideration the underlying contextual factors that affect the reliability of the data. One way to improve the reliability of measuring attitudes is to transform some of the attitude statements into direct questions in the other sections and to assess whether there is any discrepancy between the results or not. What does a KAP survey tell us about practices?

A third and integral part of KAP surveys is the investigation of health-related practices. Questions normally concern the use of different treatment and prevention options and are hypothetical. KAP surveys have been criticised for providing only descriptive data which fails to explain why and when certain treatment prevention and practices are chosen. In other words, the surveys fail to explain the logic behind people’s behaviour (Hausmann-Muela et al. 2003, Nichter 1993, Pelto and Pelto 1994, Yoder 1997). Another concern is that KAP survey data is often used to plan activities aimed at changing behaviour, based on the false assumption that there is a direct relationship between knowledge and behaviour. Several studies have, however, shown that knowledge is only one factor influencing treatment-seeking practices, and in order to change behaviour, health programmes need to address multiple factors ranging from socio-cultural to environmental, economical, and structural factors, etc. (Balshem 1993, Farmer 1997, Launiala and Honkasalo 2007). I was aware of the limitation of KAP surveys when it came to explaining the logic behind treatment-seeking practices and the difficulty of formulating a structured question to elicit these practices, and thus I included very few questions about this subject in my KAP questionnaire. I had one question about the time elapsed between onset of symptoms and treatment, thinking that this was a relatively straightforward question. According to the results, 26% (63/240) received treatment between one to three hours, 25% (60/240) within 24 hours, 5% (13/240) within 2 days, but 41% (99/240) fell into the category “other”. Those who said “other” were asked to specify their answer. The most typical explanation was either “immediately upon attack”, or that they “took pills” (ranging from panado, aspirin, or fansidar to penicillin) when symptoms appeared. It seems that the respondents interpreted the meaning of the question differently. The time categories seemed not to make much sense. Some respondents wanted to emphasise that they take pills immediately when symptoms appear. The problem was, however, that these answers did not explain what pills were taken for what symptoms and why. The answer “taking medication immediately” is also questionable based on findings from the qualitative data and from other social scientific studies explaining treatment-seeking practices (for example Agyepong and

5

Anthropology Matters Journal

2009, Vol 11 (1)

http://www.anthropologymatters.com

Manderson 1994, Hausmann-Muela et al. 1998, Nyamongo 2002). The choice of treatment depends on the severity of the symptoms. It is common that people wait and see how the symptoms develop before deciding on the choice of treatment. Mild fever is commonly treated with panado and aspirin at home, and if the fever persists, one may visit a health centre, or if the fever develops to high fever causing convulsion, relatives may seek treatment from a traditional healer. During the past decades there has also been discussion concerning informant accuracy in reporting past events and how accurately the reporting reflects reality. According to Bernard et al. (1984:503), “on average, about half of what informants report is probably incorrect in some way”, causing major concern regarding the validity of the data. All this suggests that analysis of survey research should pay more attention to the interpretation and elaboration of results. Understanding and taking account of the research context should also be a prerequisite for survey research as it is for any kind of qualitative research.

Challenges encountered in the field and some explanations for them Despite the effort to avoid the weaknesses of survey research through careful planning, I encountered several practical challenges that require discussion (see Cleland 1973, Ratcliffe 1976, Stone and Campbell 1984; see also O’Barr et al. 1973 and Ross and Vaughan 1986). The first challenge was the translation of the questionnaire: each of my research assistants translated part of the questionnaire from English into Chiyao. Then I exchanged the translations among the assistants and asked them to translate the questionnaire back into English. This exercise showed how difficult translation is, because meanings change. For example, the Yao use the word malungo to refer to malaria, but the meaning of malungo is complex as it can be used to refer to any feverish disease, its meaning varying from body pains to fever and malaria (Launiala and Kulmala 2006). So, sometimes the research assistants translated malungo as fever, sometimes as body pains, and sometimes as malaria, complicating the formulation of the questions and the interpretation of the results. Another experience of how the meanings of the questions can change occurred after I had returned home from the field. During data analysis I needed to double-check the translation of two questions regarding fever in pregnancy. I sent the questions (in Chiyao) back to Malawi and asked my research assistants to translate them back into English. Surprisingly (or perhaps predictably), the meaning of both questions changed from the original, making it questionable to use the results based on these questions, which also caused doubts about the validity of the other results. I came across this problem of changing meanings already during the in-depth interviews that I conducted (with a research assistant who acted as an interpreter), but due to the nature of the interview method, I was able to better confirm the concepts and meanings used during the interviews. In surveys, the control of cultural reinterpretation of questions is more difficult, because of the lack of in-depth conversation and because a number of different research assistants are often used to collect the data. There are several explanations for these linguistic challenges. Chiyao, the language spoken in the area in which I was conducting research, is an exclusively oral language, containing concepts and words with no vernacular equivalents in English, and vice versa. And among the Yao, knowledge and information exchange often occurs through various social networks (Soldan 2004). Furthermore, the Yao have a 6

Annika Launiala

How much can a KAP survey tell us?

specific explanatory model for malaria that is embedded in local cultural understandings, which affects their perception, knowledge, conceptualisation, treatment-seeking practices, and so on (Helitzer-Allen and Kendall 1992, Launiala and Kulmala 2006), similarly to many other ethnic groups in sub-Saharan countries (Hausmann-Muela et al. 1998, Kengeya-Kayondo et al. 1994, Nyamongo 2002, Winch et al. 1996). Researchers with western scientific training often have little knowledge, understanding and sensitivity concerning the socio-cultural context in which they conduct their studies. This cultural gap between researchers with western scientific training and local respondents may not only cause misinterpretation and cultural reinterpretation of questions, but also throws up constraints to data analysis (Ratcliffe 1976, Stone and Campbell 1984). According to Stone and Campbell (1984), many surveys conducted in rural areas in the South can be faulted for failure to meet even the fundamental requirement of formulating questions in meaningful local categories that make sense to the respondents. The authors found yet a bigger problem in concepts which evoke special meanings and associations in respondents, who then give their answer based on the meaning (connotation) of the question rather than the formal content. For example, in their study in Nepal, a great number of “don’t know” answers were received, because many of the respondents had interpreted the question “Have you heard of abortion?” as asking about knowledge of abortion technique or about knowledge of who had had an abortion. Thus, even when a questionnaire is designed using local concepts and the questions are formulated on the basis of culture-specific data, it is challenging to control misunderstandings, changing meanings, and cultural reinterpretations. One way to try to address this challenge is to loosen the time schedule (if possible) and to spend time every day going through the survey responses together with the research assistants (thus including continuous quality check up and training). Another problem I encountered was the difficulty of obtaining information concerning sensitive topics, e.g. use of traditional healers and witchcraft. Although I had used cultural information to formulate the questions, it was inadequate to overcome the problem. For example, concerning causes of complications during pregnancy, 72% (178/248) of the respondents knew of no cause. Only 8% (20/248) mentioned witchcraft as the cause of complications, yet during the focus group discussions and in-depth interviews, women told several stories related to e.g. miscarriage, complications, and even maternal death caused by witchcraft. There are several explanations for this. Firstly, many Malawians still feel uncomfortable expressing their negative feelings and opinions openly, and discussing sensitive issues such as traditional healers and witchcraft. According to some of my Malawian colleagues in UNICEF, this was in large part due to the oppressive era of Kamuzu Banda (19641994). Under Banda’s rule, people lived in constant fear because there were spies everywhere, and people were known to disappear as a consequence of saying and doing the wrong things. During Banda’s time, the use of traditional healers was also strictly prohibited and sanctioned. Nevertheless, most Malawians have a strong belief in witchcraft (ufiti) as an active force. Its presence in everyday life becomes explicit in the so called ufiti discourse used to maintain or preserve a mystical construct, to stop a certain direction of discourse, and to serve as the ultimate explanation (Englund 1996, Lwanda 2002). My experience is that a questionnaire is a poor instrument to gather information on sensitive issues, because it does not allow for building rapport between the interviewer and the respondent.

7

Anthropology Matters Journal

2009, Vol 11 (1)

http://www.anthropologymatters.com

I was also interested in reaching beyond the “yes” and “no” answers, and therefore included open-ended questions for additional explanations in my KAP questionnaire. The results were, however, rather disappointing; they contained little information and many questions were left unanswered. On the other hand this weakness could be due to the fact that I unintentionally emphasised quantity over quality, as I did not put any limitation to the number of survey interviews conducted per day. So the faster the research assistants managed to complete the surveys, the sooner they received their salary to support their families. On the other hand, there is always the pressure of time in the field, too. We were rushing through the survey because of anticipation that the rainy season would start any day, making it difficult to reach the respondents in the villages. I too was busy conducting interviews and did not constantly follow up the work of the research assistants every day. The skills and enthusiasm of research assistants also vary and some are better at probing than others. All these factors may lead to many of the open-ended questions being left unanswered. I also encountered the problems caused by the previous training of the other research teams; unlearning previous ways of collecting data proved hard. There are a limited number of research assistants available at the study area, yet there have been many research teams over the years. This means that the same research assistants are employed in many studies, all having different aims and methods. The previous research teams conducting surveys had trained the assistants to probe the alternatives but I wanted the assistants not to probe unnecessarily, and to make a note when alternatives were probed. As a result some assistants experienced difficulties in learning to avoid unnecessary probing, and others forgot to indicate when they had probed the alternatives. As pointed out by Cleland (1973) already in the 1970s, probing or non-probing makes a difference to the results. My advice to proceed cautiously with probing the alternative answers led to a high number of “don’t know” responses. An alternative explanation, however, is that women’s knowledge most often concentrates on local, indigenous issues, and they had little or nothing to say when presented with questions emphasising public health and biomedical knowledge. Or the women may have been worried about giving wrong answers, or may have been afraid of answering, especially if their relatives and other community members were present, as was often the case in the villages. Lastly, there was also the problem of courtesy bias, meaning that respondents produced answers which they believed that the research assistants and health centre staff wanted to hear. Malawians are a polite people, and, disliking the idea of conflict, they rarely refuse to participate in a survey. This may cause a problem of courtesy bias, as reported in many studies criticising the use of surveys (Bhattacharyya 1997, Stone and Campbell 1984). The courtesy bias could have been further worsened by the fact that most respondents continuously assumed that this survey had something to do with the Lungwena health centre, which may have made them worry about what type of treatment they would receive if they were critical towards the services and care provided by the health centre. For example, answers to the survey questions related to the use of the antenatal clinic’s services seemed to be positive, yet during the in-depth interviews women voiced their criticism towards the antenatal clinic’s services. The problem of courtesy bias is further strengthened by the fact that local people are used to receiving money or goods in exchange for their knowledge. Interestingly, an unpublished report from a results dissemination meeting in the present study area shows that, given an opportunity, people are willing to express their concerns and even negative opinions about surveys (TUMCHP 2005). The report 8

Annika Launiala

How much can a KAP survey tell us?

revealed that the community in question was tired of participating in the many ongoing research activities. Many of the community members did not even differentiate between the different studies, and lacked understanding of the aims of these studies. Also, they expected handouts from the researchers after taking part in surveys, especially after answering very long questionnaires. Furthermore, they felt that some researchers exploited them by asking intimate questions about sexual issues and, at the same time, their private lives, as such questions are considered culturally inappropriate and against their moral code.

More critical discussion about the challenges encountered in the field is needed My experience in using a standard KAP survey questionnaire to collect data on knowledge, attitudes and practices concerning malaria in pregnancy in rural Malawi strengthens my opinion that as a method the KAP survey contains several weaknesses. Some of these weaknesses can be overcome through careful planning, pre-testing and training of research assistants. A bigger concern, however, is the appropriateness of a KAP survey to collect data, particularly on attitudes and practices, and the way the results are interpreted without contextual understanding. Ratcliffe (1976) argued already in the 1970s that uncritical reliance on a KAP survey’s ability to produce accurate data combined with limited comprehension of the socio-cultural context is likely to deliver a narrow understanding of the underlying factors, or even worse, a bogus interpretation of data. Another major problem is that many investigators use KAP surveys to explain health seeking behaviour assuming that there is a direct relationship between knowledge and action, as pointed out by Hausmann-Muela et al. (2003). I agree with other authors that a KAP survey is a poor method for obtaining information about sensitive issues, such as traditional treatment and prevention practices, and sexual behavior (Schopper et al. 1993, Smith 1993). At the most, it can be used to assess people’s knowledge about practices in general, but not about their actual day-to-day practices and the explanations behind them (Hausmann-Muela et al. 2003, Nichter 1993). I also agree with Radcliffe (1976) and Stone and Campbell (1984) who have argued that any kind of survey questionnaire is a rather unnatural method for collecting information in a rural setting in a nonWestern culture. The name “Knowledge, Attitudes, and Practice” (KAP) survey itself gives a misleading impression that we can easily use a KAP survey to collect data on health seeking practices and that this will be useable for programme planning. Professionals working in international health do not often have thorough methodological research training and thus they may take the use of certain methods for granted. What is lacking in today’s scientific discussion is an analytical discussion about the strengths and weaknesses in survey design and methodology, and the limitations of survey research in general. Often the data collection process is described superficially, following the standard procedures and leaving out the contextual description. Yet an open and transparent discussion is the only way to improve methods and to learn from mistakes. Presumably, despite the weaknesses of survey questionnaires, they will still be used for data collection in settings across the world. Therefore, I would argue that in addition to the open discussion of the limitations of the method, minimum prerequisites are to carefully consider what type of information can be collected with

9

Anthropology Matters Journal

2009, Vol 11 (1)

http://www.anthropologymatters.com

a questionnaire and to take into account the socio-cultural context in both the planning stage and when interpreting the results. Within public health programme research there has been an increasing trend towards multiple-method designs composed of a variety of qualitative and quantitative methods, in order to lessen the limitations of single method designs (Bhattacharyya 1997, Stone and Campbell 1984, see also Lambert and McKevitt 2002). The use of a multiple-method design allows contextualisation of knowledge and makes it possible to understand the logic behind treatment-seeking practices. One of the advantages of combining qualitative and quantitative methods is that it increases the validity of data if the study is appropriately designed. One should, however, keep in mind that any successful research outcome depends heavily on the skills of the researcher.

Conclusion: the challenges and value of interdisciplinarity Today’s health problems around the world cannot be solved by any discipline alone. The way forward is to enhance interdisciplinary cooperation between the social sciences and medicine. There are, however, challenges that need to be overcome before true interdisciplinarity can be achieved. One such challenge to anthropologists is to find a way to communicate with medical professionals and to be able to argue convincingly what anthropology can offer (Pool and Geissler 2007). When working with UNICEF in Malawi I often encountered resistance concerning my “anthropological” ideas and suggestions, presumably because most of my colleagues failed to understand the relevance of the study, and because I was unable to explain my ideas convincingly using the appropriate “public health language”. Napolitano and Jones (2006) have described similar problems and experiences among public health practitioners in the UK and in the Gambia, referring to their limited understanding of anthropology and its contributions, and the existence of ethnocentric fears. Some anthropologist might wonder why we should make an effort if medical researchers are perceived not to be taking any steps towards understanding anthropology. I guess it all depends on what drives us to do research. My motivation for trying to enhance collaboration between medicine and medical anthropology is based on the hope that in the end it will improve the well-being of rural Malawians. Conducting a KAP survey in a rural African setting – and in other types of settings as well – is problematic for a number of reasons. These problems and challenges should be openly discussed in scientific publications and communicated to programme planners. A KAP survey can be useful when the research plan is to obtain general information about public health knowledge regarding treatment and prevention practices, or about sociological variables, such as income, education, occupation, and social status. It is important, however, to know and understand what type of data can be generated by which method, and to choose appropriate methods in relation to the study objectives. If the objective is to study health-seeking knowledge, attitudes, and practices in context, there are suitable ethnographic methods available, including focus group discussions, in-depth interviews, participant observation, and various participatory methods. A combination of qualitative and quantitative methods may also prove effective, but I believe that the best value can be achieved only when the research team consists of experts from both qualitative and quantitative research traditions. Anthropologists working in international and public health should strive to find ways to enhance true interdisciplinarity.

10

Annika Launiala

How much can a KAP survey tell us?

References Agyepong, I. and L. Manderson. 1994. The diagnosis and management of fever at household level in the Greater Accra Region, Ghana. Acta Tropica 58, 317330. Balshem, M. 1993. Cancer in the community: Class and medical authority. Washington, DC: Smithson Inst. Press. Bernard, H. R., P. Killworth, D. Kronenfeld and L. Sailer. 1984. The problem of informant accuracy: The validity of retrospective data. Annual Review of Anthropology 13, 495-517. Bhattacharyya, K. 1997. Key informants, pile sorts, or surveys? Comparing behavioral research methods for the study of acute respiratory infections in West Bengal. In The anthropology of infectious diseases: Theory and practice on medical anthropology and international health (eds) M. C. Inhorn and P. J. Brown, 211-238. Amsterdam: Routledge Publishers. Caldwell, J.C., P. Caldwell, and P. Quiggen. 1994. The social context of AIDS in subSaharan Africa. New York: Population Council. Cleland, J. 1973. A critique of KAP studies and some suggestions for their improvement. Studies in Family Planning 4(2), 42-47. Englund, H., 1996. Witchcraft, modernity and the person: The morality of accumulation in Central Malawi. Critique of Anthropology 16(3), 257-279. Farmer, P.E. 1997. Social scientists and new tuberculosis. Social Science and Medicine. 44(3), 347-358. Foster, G. M. 1987. World Health Organization behavioural science research: Problems and prospects. Social Science & Medicine 24(9), 709-717. Gill, G. J. 1993. O.K., the data’s lousy, but it’s all we’ve got (being a critique of conventional methods). Gatekeeper Series no. 38. London: International Institute for Environment and Development (www.iied.org). Good, B. 1994. Medicine, rationality and experience: An anthropological perspective. Cambridge: Cambridge University Press. Green, C. E. 2001. Can qualitative research produce reliable quantitative findings? Field Methods 13(3), 3-19. Hausmann-Muela, S., R. J. Muela and M. Tanner. 1998. Fake malaria and hidden parasites – the ambiguity of malaria. Anthropology and Medicine 5(1), 43-61. Hausmann-Muela, S., R. J. Muela and I. Nyamongo. 2003. Health-seeking behaviour and the health system’s response. DCPP Working Paper no. 14. Give web address? Helitzer-Allen, DL. and C. Kendall. 1992. Explaining differences between qualitative and quantitative data: A study of chemoprophylaxis during pregnancy. Health Education Quaterly 19, 41-54. Kengeya-Kayondo, J. F., J. A. Seeley, E. Kajura-Bajenja, E. Kabunga, E. Mubiru, F. Sembajja and D. W. Mulder. 1994. Recognition: treatment seeking behaviour and perceptions of cause of malaria among the rural women in Uganda. Acta tropica 58, 255-266. 11

Anthropology Matters Journal

2009, Vol 11 (1)

http://www.anthropologymatters.com

Lambert, H. and C. McKevitt. 2002. Anthropology in health research: From qualitative methods to multidisciplinarity. British Medical Journal 325, 210213. Launiala, A. and M-L. Honkasalo. 2007. Ethnographic study of factors influencing compliance to intermittent preventive treatment of malaria during pregnancy among Yao women in rural Malawi. Transactions of the Royal Society of Tropical Medicine and Hygiene 101(10), 980-989. Launiala, A. and T. Kulmala. 2006. The importance of understanding the local context: Women’s perceptions and knowledge concerning malaria in pregnancy in rural Malawi. Acta Tropica 98, 111-117. Lwanda, J. 2002. Tikutha: the political culture of the HIV/AIDS epidemic in Malawi. In A democracy of chameleons: Politics and culture in the new Malawi (ed.) H. Englund, 151-165. Blantyre: Christian Literature Association in Malawi. Manderson, L. and P. Aaby. 1992. An epidemic in the field? Rapid assessment procedures and health research. Social Science & Medicine 35(7), 839-50. Napolitano, D. A. and C. O. H. Jones. 2006. Who needs “pukka anthropologists”? A study of the perceptions of the use of anthropology in tropical public health research. Tropical Medicine & International Health 11(8), 1264-1275. Nichter, M. 1993. Social science lessons from diarrhea research and their application to ARI. Human Organization 52(1), 53-67. ---------. 2008. Global health: Why cultural perceptions, social representations, and biopolitics matter. Tuscon: University of Arizona Press. Nyamongo, I. K. 2002. Health care switching behaviour of patients in a Kenyan rural community. Social Science &Medicine 54(3), 377-386. O’Barr, W., D. Spain and M. Tessler. 1973. Survey research in Africa: Its applications and limits. Evanston, IL: Northwestern University Press. Pelto, J. P., and G. H. Pelto. 1997. Studying knowledge, culture, and behavior in applied medical anthropology. Medical Anthropology Quarterly 11(2), 147163. Petty, R. E., and J. P. Cacioppo. 1981. Attitudes and persuasion—classic and contemporary approaches. Dubuque, IA: W. C. Brown Co. Publishers. Pool, R., and W. Geissler 2007. Medical anthropology: Understanding public health. Berkshire: Open University Press. Ratcliffe, J. W. 1976. Analyst biases in KAP surveys: A cross-cultural comparison. Studies in Family Planning 7(11), 322-330. Ross, D. A., and J. P. Vaughan. 1986. Health interview surveys in developing countries: A methodological review. Studies in Family Planning 17(2), 78-94. Schopper, D., S. Doussantousse and J. Orav. 1993. Sexual behaviors relevant to HIV transmission in a rural African population: How much can a KAP survey tell us? Social Science & Medicine 37(3), 401-412. Soldan, V. A. P. 2004. How family planning ideas are spread within social groups in rural Malawi. Studies in Family Planning 35(4), 275-290.

12

Annika Launiala

How much can a KAP survey tell us?

Smith, H. L. 1993. On the limited utility of KAP-style survey data in the practical epidemiology of AIDS, with reference to the AIDS epidemic in Chile. Health Transition Review 3(1), 1-15. Stone, L. and J. G. Campbell. 1984. The use and misuse of surveys in international development: An experiment from Nepal. Human Organization 43(1), 27-34. TUMCHP. 2005. Unpublished meeting report, distributed via email 13 July 2005. Department of International Health, University of Tampere Medical School, Finland. Winch, P.J., A. M. Makemba, S. R. Kamazima, M. Lurie, G. K. Lwihula, Z. Premji, J. N. Minjas and C. J. Shiff. 1996. Local terminology for febrile illnesses in Bagamoyo District, Tanzania and its impact on the design of a communitybased malaria control programme. Social Science and Medicine 42(7), 10571067. Yoder, P. S. 1997. Negotiating relevance: Beliefs, knowledge and practice in international health projects. Medical Anthropology Quarterly 11(2), 131-146.

About the author Annika Launiala holds an MA in Cultural Anthropology and is currently working as a project manager on a multidisciplinary project called “Multicultural aspects of Health” at the School of Public Health and Clinical Nutrition, University of Kuopio, Finland. She is also working on her PhD research concerning socio-cultural factors affecting treatment and prevention of malaria in pregnancy in rural Malawi. She can be contacted at [email protected] or [email protected]

13