Tunisia Implemented To Address Inequalities.pdf

  • Uploaded by: Hayet Ben Said
  • 0
  • 0
  • November 2019
  • PDF

This document was uploaded by user and they confirmed that they have the permission to share it. If you are author or own the copyright of this book, please report to us by using this DMCA report form. Report DMCA


Overview

Download & View Tunisia Implemented To Address Inequalities.pdf as PDF for free.

More details

  • Words: 25,549
  • Pages: 60
AfDB 2014 www.afdb.org

E c o n o m i c

B r i e f

What policies should be implemented to address inequalities in health care in Tunisia?

CONTENTS

Summary p.2 1 – General Introduction p.4

Key Messages

2 – Indicators of Health Status and Use of Health Care Services p.5

• Despite the progress achieved, health inequalities remain considerable and relatively little known in Tunisia. In light of the analysis conducted, there is significant elbow room for reducing these inequalities. In Tunisia, there are significant inequalities in care consumption between governorates for similar needs (those related to reproductive health, for example). There are also significant differences in the health status of the population of these governorates. The life expectancy of 74.5 years in 2009 does not exceed 70 years in Kasserine and Tataouine, but reaches 77 years in the governorates of Tunis and Sfax. The analysis indicates that:

3- Territorial Inequalities in

- The overall inequality in health spending declined from 2000 to 2010. The breakdown of the Gini index shows that this movement is almost entirely explained by the decrease in inequality in pharmaceuticals spending, which accounted for 42.2% of health spending in 2010. This trend can be attributed to a greater availability of pharmacies throughout the national territory. - The items where inequality has worsened and that had an inertia effect were long-term illnesses (17% of expenditure), hospital stay and medical surgery (8.6%) and radio and scans (8% of health spending). Such spending is related to the demographic and epidemiological transition. - Dental care is characterized by unusually high levels of inequality and lack of access for the disadvantaged classes.

Health Care Facilities p.9 4- Trend in inequalities in health spending in Tunisia between 2000 and 2010 p.27

• The main recommandations in this context are as follow:

5- General Conclusion p.41

- From the supply side: (i) In the public sector, it is necessary to revitalize primary health care by improving the operation (ii) It is also important to strengthen Level II which seems to be the weak link in the system. Better coverage of the territory in terms of Level II beds should necessarily go hand-in-hand with the provision of more specialized physicians for the poorest regions in light of the demographic and epidemiological transition. (iii) Efforts should be made to ensure that at each level the system performs its assigned tasks under the best possible conditions. These tasks should be clearly defined. Each hospital institution should have a scheme of work that allows for coherent strategic management. (iv) The specific incentives that were introduced to encourage physicians to settle in deserted areas should be evaluated. Public-public and possibly private-public partnerships should be instituted. Also, it is important to negotiate with corporations an institutional framework to better regulate the opening of private practices. (v) It is necessary to determine measures that should be implemented to enhance health care delivery at local or regional level, as part of an overall regional development policy. - On the demand side: (i) It is important to reduce financial barriers to health care access by better targeting the poor who benefit from free medical assistance. (ii) Pharmaceuticals are a significant drain on the budgets of the poorest households and it is necessary to reduce this weight by ensuring good governance of public pharmacies. (iii) There is a need to ensure a better collective coverage of longterm illness, hospital stay and medical surgery, x-rays and scans. Knowing the profile of households that incur these expenses will make it possible to better target them, if need be. (iv) Dental care continues to be characterized by extremely high inequalities in expenses. Improved coverage of the territory in terms of availability of dental practices and greater public awareness of the importance of dental health should curb one of the causes of the inequality. Similarly, a special processing of reimbursement for dental expenses by health insurance, apart from the recurrent expenses, should contribute to reducing inequalities in dental care access. - On the institutional side: (i) It is necessary to aim at reducing social and regional inequalities in health care. (ii) There is a need to produce and monitor indicators for assessing the progress of specific categories not only at the national level but also at the local level. It is important to conduct periodic surveys on the status of health, health care use, or the failure to seek health care for financial reasons.

Bibliography p.42 Annexes p.44

Zondo Sakala Vice President [email protected]

Jacob Kolster Director ORNA [email protected] +216 7110 2065

This paper was prepared by Salma Zouari, Ines Ayadi and Yassine Jmal, under the supervision of Vincent Castel (ORNA) and Sahar Rad (ORNA) et Laurence Lannes (OSHD). Overall guidance was received from Jacob Kolster (Director, ORNA). Ahmed Rekik and Chokri Arfa suggested improvements to the preliminary version of this research. Asma Baklouti, Mariem Ellouze, Rahim Kallel and Abdessalem Gouider each made an input.

A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

a

n

k

AfDB 2 0 1 4



Economic Brief w w w . a f d b . o r g

Summary

Therefore, there is clearly a need to develop a strategy for strengthening and revitalizing primary health in the country as well as enhancing Level II.

I

n Tunisia, there are significant inequalities in care consumption between

governorates for similar needs (those related to reproductive health,

1-2- Regarding human resource allocations, the inequality between

for example). There are also significant differences in the health status

governorates has decreased, except for physicians whether in the

of the population of these governorates. The life expectancy of 74.5

public or private sector. Although there has been a significant drop

years in 2009 does not exceed 70 years in Kasserine and Tataouine,

in the number of inhabitants per physician from 2002 to 2010,

but reaches 77 years in the governorates of Tunis and Sfax.

the gaps have widened between the better endowed governorates and the less endowed ones, while the variation coefficients have

Three hypotheses were then made:

l

l

increased.

Households, whatever the level of their resources and even when they

The availability of free medical practitioners is characterized by high

benefit from social security coverage, have unequal access to care

levels of inequality; the relationship between the most endowed and

because of inequities in the provision of health care services in their

the least endowed governorates is 14.3. This is followed by dental

immediate environment.

practices (ratio of 11.3) and hospital beds (10.7). The most evenly

Despite the importance of social coverage, households assume an

distributed resources are pharmacies and paramedical staff.

average of 41% of health spending in the form of out-of-pocket

l

expenditures. Therefore, households have unequal access to care

It would be advisable to review the criteria for opening positions of

arising from inequalities in income distribution and illustrated by

public health physician at regional level and the institutional framework

unequal health spending.

governing private practices. Like the practice of pharmacy, the practice

Due to the importance of out-of-pocket health care spending, the

of dentistry and medicine on a free basis should be better regulated.

regressive (or progressive) nature of care spending and its inelasticity

Similarly, public-private and especially public-public partnerships

compared to income, can give them a potentially catastrophic and

(such as agreements between academic physicians and regional

impoverishing character that makes unequal access to care even

hospitals) that may make disadvantaged areas more attractive as is

more acute.

being considered for specialists could be a solution. However, the implementation of such partnerships should be accompanied by

These hypotheses were tested on the basis of available statistical

measures to ensure their effectiveness for all stakeholders.

data. Health policy recommendations have been made. 1-3- Lastly, since the status of health care facilities in a governorate 1- On the assumption that the availability of care provision, whether

cannot be analysed by reference to a single determinant, all the

public or private, and good coverage of the national territory in health

components of the sector and the complementarity between different

infrastructure contribute to the decline in inequality in access to care,

providers should be taken into account simultaneously. For this purpose,

we analysed the trend of provision indicators by governorate and the

we have integrated the various determinants of facilities (by category

dispersion of these indicators through the use of cross-sectional data

and overall) in order to arrive at relatively homogeneous groups (called

of the 2010 health map and various longitudinal indicators published

clusters) and calculated for each governorate, a composite indicator

in the Statistical Yearbook of the National Institute of Statistics for the

of care provision that measures its position compared to other

period 1997-2010. Three aspects were analysed: infrastructure, the

governorates as well as the progress that may be achieved over time.

availability of beds and the provision of human resources. Among the three components of health care facilities, the geographic 1-1- With regard to infrastructure and bed availability, it turned out

distribution of medical human resources stands out as the most unequal,

that only the availability of PHCs declined over the last decade. Level

with a significant concentration on the coast. Despite an increase in the

II, which is the reference for Level I, would not be very effective because

density of physicians, regional disparities have widened. Qualitatively,

it lacks adequate technical equipment and specialized physicians. We

the inequalities are even more blatant and more than 2/3 of specialists

suspect that patients are referred to Level III which takes the place

are found on the coast as regards not only rare specialties but also the

of Level II, thus causing inefficiencies.

most common such as gynaecology and paediatrics.

2 A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

a

n

k

Economic Brief 2 0 1 4

Three governorates constantly fall within the most favoured cluster



AfDB

w w w . a f d b . o r g

l

the contribution of inequality in each SPY category or sub-category

l

the marginal effect - equalizer or non-equalizer – of the variation of a

whatever the aspect considered. They are Tunis, Sousse and Monastir.

to total health spending inequality;

Conversely, four governorates always fall within the most disadvantaged cluster: Jendouba, Kairouan, Kasserine and Sidi Bouzid. Between these

particular SPY on total health spending inequality.

two groups, the various governorates show more or less substantial deficits depending on the nature of the resources analysed. The scope

2-1- The overall inequality in health spending declined from 2000 to

of intervention required by each governorate may then be defined.

2010. The breakdown of the Gini index shows that this movement is almost entirely explained by the decrease in inequality in pharmaceuticals

Government intervention is necessary when there is a build-up of

spending, which accounted for 42.2% of health spending in 2010. This

inequalities. However, the choices related to the health sector and efforts

trend can be attributed to a greater availability of pharmacies throughout

to better allocate resources to priority areas can only be effective if they

the national territory.

form part of a comprehensive local development strategy in these areas. The reduction of economic, cultural and social differences between the

2-2- The items where inequality has worsened and that had an inertia

governorates can only facilitate and strengthen health reforms.

effect were long-term illnesses (17% of expenditure), hospital stay and medical surgery (8.6%) and radio and scans (8% of health spending).

2- Working from the assumption that inequalities in access to health

Such spending is related to the demographic and epidemiological

are linked to income inequalities, we assessed the inequality of out-of-

transition.

pocket household health expenditure and analysed the trends thereof and training through inequality indicators and their breakdown by

The reduction of the corresponding inequality requires specific

expenditure category. For this purpose, we referred to the individual

government policies that target the most vulnerable groups. The

data of the national surveys on household budget and consumption in

collective management of these expenditures still seems insufficient.

2000, 2005 and 2010. This data provides information on total health

Knowing the profile of households that incur these expenditures will

spending per person per year (SPY) and the various expenditure

help to better target them.

categories: routine medical care, special medical care, pharmaceuticals and medical devices or expenditure sub-categories (medical

2-3- Dental care is characterized by unusually high levels of inequality

consultations; dental care; radio, scanner and medical analysis; medical

and lack of access for the disadvantaged classes.

stay and surgery; special dental care; special radiology expenditure; childbirth; long-term diseases; drugs; other pharmaceuticals, etc.).

Improved coverage of the territory by dental practices and greater awareness of people about the importance of oral and dental health

This approach gave information about:

should curb one of the causes of this inequality. Similarly, a specific treatment for reimbursement made by health insurance that is non-

l

Overall inequality in health spending and its trends;

concurrent with current spending should contribute to the reduction of

l

inequality of SPY in each care spending item and sub-item;

inequalities in access to dental care.

3 A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

a

n

k

AfDB 2 0 1 4



Economic Brief w w w . a f d b . o r g

1. General Introduction

S

ince 1956, the foundations for a universal system of health care

is heavily subsidized or free for the beneficiaries of free health coverage

delivery have been established in Tunisia. For three decades,

who are estimated at 27% of the population.

the resulting benefits have been improved over time and a social security system has been put in place for employees. However, public

Despite the important size of community coverage of medical care by

expenditure on the health sector has slowed since the 1990s, and

insurance or by the State budget, private spending still remains very

private practices have gradually substituted public health services that

high and is increasing. In 2010, health care was financed by public

have experienced some decline in their quality and availability. Household

budgets to the tune of 23.8%, by health insurance resources (CNAM

health spending has risen sharply, sometimes leaving a heavy dent on

with 27.7% and private insurance with 7%) and out-of-pocket

the household budget. The most vulnerable segments are not spared

household spending that covered 41.2%.

(Arfa, ElGazzar, 2013). Thus, the health system faces many challenges, and it is important to: Since January 2011, there is heightened awareness of inequalities in the health status of the population and health care use. More attention

l

Reduce regional disparities in the provision of health care services;

is paid to issues of equity in access to care and there is more concern

l

reduce inequalities in household health spending;

about a more egalitarian distribution of health services throughout the

l

limit the amount of out-of-pocket household spending.

country. Therefore, it is necessary to better define the situation of health inequalities and its recent trend, and to identify policies that can help

Indeed, although health infrastructure covers almost the entire 1

country , there are inequalities in availability between the different

address the above challenges.

regions. The health system is predominantly public, with 87% of bed capacity in public hospitals and 13% in private clinics. On average,

We will begin by recalling some household health status indicators

Tunisia has 123 physicians per 100 000 inhabitants. However,

and the use of health services (Section I). We will then analyse the

physician density is much lower in the poorest regions where most

inequities in health care provision (Section II). Lastly, we will also

of the beneficiaries of free health care are found.

2

address the spending inequalities and their sources (Section III). In this respect, we will mainly use data from the 2010 Tunisia Health

With regard to the financing of the demand for care, the National Health

Map published by the Ministry of Health as well as data relating to

Insurance Fund (CNAM) covers about 68% of the total population. It

the health sector published by the National Institute of Statistics (INS)

covers both public and private health care services in the country. The

in the Statistical Yearbook of Tunisia between 1997 and 2010. We

majority of physicians, laboratories, dentists and pharmacists are

will further use the individual databases of the National Surveys on

contracted with CNAM. There are three branches: the public branch,

Household Budget and Consumption conducted by the INS in 2000,

the private branch and the reimbursement branch. The public branch

2005 and 2010.

1

The health system includes: (a) primary health centres or primary health centres and local or district hospitals; (b) regional hospitals; and (c) university teaching hospitals. 2 Medical density in Tunisia is lower than the European average, which is more than 300 physicians per 100 000 people. It is the highest in the Maghreb (Algeria and Libya 120, Morocco 60, Mauritania 10) and occupies the ninth place in the EMRO region behind Lebanon (330), Bahrain (300), Qatar (280), Jordan (260), Egypt (240), Kuwait and Oman (180), Saudi Arabia (160) and ahead of Iran (90), Pakistan (80) , Syrian Arab Republic and Iraq (50 ), Sudan and Yemen (30) (MH, Tunisia Health Map 2010).

4 A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

a

n

k

Economic Brief 2 0 1 4



AfDB

w w w . a f d b . o r g

2. Indicators of Health Status and Use of Health Care Services3

T

here are few studies and statistics on this issue. However, we

because reproductive health concerns the entire population in the

will refer to data known for their relevance and published regularly

same manner.

by the INS. The health status will be assessed through life expectancy

1-

and the infant mortality rate (IMR). These two indicators are particularly

Health status indicators

suited to the health inequalities study (Jusot, 2003). However, they are published in a systematic way only at the national level. The

Differences in life span may be seen as a synthetic indicator of social

mortality rate, which is a poor indicator of health status because it is

differences affecting health throughout the life cycle (Aïach, 2000).

sensitive to the structure of the population by age, is however available

There has been a remarkable increase in life expectancy at birth in

at the governorate level. It will only be used to assess the evolution

Tunisia (Figure 1). From only 58 years in 1956, life expectancy has

of its variation coefficient.4 Health service use will be analysed through

risen to 74.9 years in 2011. The continuous improvement of this

the data on reproductive health because of their availability and

indicator is applicable to both men and women.

Figure 1: Life expectancy by gender (1990-2011)

Male

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1998

1997

1996

1995

1994

1993

1992

1991

1990

78 76 74 72 70 68 66 64

Female

The increase in life expectancy is due to a decrease in mortality

The mortality rate showed a decreasing trend. From 19.1 per thousand

rates in general, and the sharp decline in infant mortality in

in 1960, it dropped to 5.5 per thousand in 2011 (Figure 2). The infant

particular.

mortality rate fell from 120 per thousand births in 1966 to 14 per thousand in 2011 (Figure 2).

3

All the statistics in this section are derived from Tunisian Statistical Yearbooks published by the National Institute of Statistics. For a Y distribution of mortality rates with average  Y, the coefficient of variation, noted CV, is derived from the variance. It is defined as the ratio of the standard deviation σ to the mean mortality rates: CV = σ /Y where σ2 = 1/ N ∑(Yi -Y)2 CV is used to compare the dispersions of distributions with different averages. The higher the CV the more dispersed the distribution will be. 4

5 A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

a

n

k

AfDB 2 0 1 4



Economic Brief w w w . a f d b . o r g

Figure 2: Mortality rate

Figure 3: Infant mortality rate

25.0 20.0 15.0 10.0 5.0 0.0

200 150 100 50 1960 1964 1968 1972 1976 1980 1984 1988 1992 1996 2000 2004 2008

1990 1964 1968 1972 1976 1980 1984 1988 1992 1996 2000 2001 2008

0

Infant mortality is an indicator widely used in international comparisons.

The life expectancy of 74.5 years in 2009 did not exceed 70 years in

It is an indicator of robust health, revealing a country’s development

Kasserine and Tataouine, but reached 77 years in the Tunis or Sfax

level and the quality of its health care system. It depends on several

governorates (Map 1). Similarly, the decline in infant mortality has not

factors, including income, maternal educational level and the

been equally beneficial to all children regardless of their place of birth

effectiveness of preventive care provided to mother and child

(Map 1). In 2009, the infant mortality rate was 17.8 ‰ at the national

(Bouchoucha and Vallin, 2007). The decline in infant mortality rate may

level. In the South, it was 21 ‰, while in the Midwest it rose to 23.6 ‰

be attributed to both factors inherent in the health system (modernization

(Ministry of Regional Development, 2011). Map 1 shows two groups of

and better coverage of the country) and the evolution of Tunisian society

governorates at odds with each other: the first (Tunis, Sousse, Monastir

(improvement in the quality of life and an increase in the living standards

and Sfax) recorded the lowest infant mortality rates (IMR), while the

and the educational level of the population).

second (Kasserine, Sidi Bouzid and Kairouan) had the highest rates.

However, this overall positive trend hides significant disparities between

Lastly, the mortality rate by governorate (indicator sensitive to the age structure

rural and urban areas as well as between socioeconomic groups and

of the population) shows contrasting trends and especially an increase in

between the various governorates. We will focus on the regional aspect

the variation coefficient, indicating a rise in inter-governorate inequality and

of inequality and health status indicators.

greater heterogeneity of living conditions prevailing there (Figure 4).

Figure 4: Coefficient of variation in mortality rates by governorate (1978-2010)

0.25 0.20 0.15 0.10 0.05 0.00 1997 1998 1999 2000 2000 2001 2002 2003 2004 2005 2007 2008 2009 2010

6 A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

a

n

k

Economic Brief 2 0 1 4

2-

Health care use indicators



AfDB

w w w . a f d b . o r g

women’s perception of health and medicine, and it depends on the alternative care available to them and their medical environment.

In general, health care use is related to health condition and inequality

However, it is often subject to social and family control (Gastineau,

in health care use primarily reflects unequal needs. Hence it is not

2003). Belonging to a family or community group, social habits and the

necessarily unfair. It is therefore necessary to analyse the inequalities

environment are likely to influence women's choices in this matter.

observed only when facing the same need. As such, the indicators related to reproductive health are well suited for such an analysis. They help to

With reference to the number of deliveries by governorate, we calculated

identify women who give birth and analyse inequalities between them.

three indicators: the home delivery rate, the hospital delivery rate and the clinic delivery rate.5

Tunisian Statistical Yearbooks each year publish statistics on the use of reproductive health care. We will try to see to what extent the use of

In 2010, the home delivery rate (or rate of medically unassisted childbirth)

such care is egalitarian.

was 7.6%. It varied greatly between governorates. Governorates with high home delivery rates include Monastir, Nabeul and Mahdia, all of which have relatively good health infrastructure (Figure 5).

However, it is important to note that reproductive health care use reflects

Figure 5: Home delivery rate in 2010 by governorate

Zaghouan Kasserine Manouba Sidi Bouzid Kairouan Siliana Mahdia Nabeul Monasr TUNISIA Medenine Tataouine Gaafsa Ben Arous Kebili Sousse Ariana Tozeur Sfax Gabes Le Kef Jendouba Beja Bizerte Tunis

50% 40% 30% 20% 10% 0%

Clinic delivery rate was 12.3%. With the exception of Monastir, it was

private health infrastructure (Figure 6).

generally higher in the major urban centres of the coast which have

Figure 6: Clinic delivery rate in 2010 by governorate

Ben Arous Sfax Nabeul Sousse Tunis Ariana TUNISIA Bizerte Médenine Le Kef Mahdia Monas r Gabes Kairouan Beja Gafsa Jendouba Manouba Sidi Bouzid Siliana Touzeur Kasserine Kebili Zaghouan Tataouine

50% 40% 30% 20% 10% 0%

5

The sum of these three rates is not a unit because some women do not report the place where they gave birth.

7 A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

a

n

k

AfDB 2 0 1 4



Economic Brief w w w . a f d b . o r g

Lastly, hospital delivery is the dominant standard in Tunisia. On

are those where the private sector is important (Sfax, Tunisia...)

average, 67.5% of women give birth in hospital. Governorates

or where the use of medicine is relatively limited (Sidi Bouzid,

where the propensity to give birth in hospitals is lower (Figure 7)

Kasserine…).

Figure 7: Hospital delivery rate in 2010 by governorate

Jendouba Beja Tozeur Kebili Tataouine Gabes Le Kef Siliana Bizerte Gafsa Ariana Medenine Sousse Mahdia Kairouan TUNISIA Sfax Nabeul Tunis Manouba Sidi Bouzid Ben Arous Kasserine Monas r Zaghouan

100% 80% 60% 40% 20% 0%

By comparing the number of antenatal or postnatal visits in the public

show a lower frequency rate for these procedures. These governorates

sector to the total number of births registered in a year (assisted or

are better off in private infrastructure, and many women consult their

unassisted, in hospital or private clinic), the frequency of these procedures

gynaecologist. The number of public postnatal visits is much lower (0.57

may be determined. There is an average of three antenatal visits per

in 2010), but it is just as unevenly distributed among the governorates

delivery. Only Tunis, Sousse, Sfax, Monastir and Mahdia governorates

as the number of antenatal visits (Figures below).

Figure 9: Postnatal visits per delivery

Figure 8: Antenatal visits per delivery

1.5 8 6 4 2 0

1.0 0.5 Tozeur Kebili Zaghouan Kasserine Gabes Sidi Bouzid Gafsa Le Kef Bizerte Siliana Tataouine Ariana Beja Nabeul Kairouan Ben Arous Jendouba Medenine TUNISIA Mahdia Monasr Sfax Sousse Tunis

Kebili Gabes Zaghouan Kasserine Ariana Ben Arous Sidi Bouzid Nabeul Tozeur Mahdia Bizerte Jendouba Tataouine Gafsa Beja Siliana TUNISIA Le Kef Sfax Monasr Medenine Kairouan Sousse Tunis

0.0



More generally, the various governorates do not fall in the same

availability of funding or means of support to make the demand effective;

category for the two indicators, reflecting the importance of community •

and social determinants in health care use. It would be interesting to

existence of an offer or several offers to meet the need.

consider the factors that explain these differences in order to assess the fairness of the system (Fleurbaey and Schokkaert, 2011). The use

When one of the last two elements is absent, access to care becomes

of health care which determines the health status of individuals is

impossible and care will not be provided at the risk of leading to serious

mainly due to the interaction of three determinants:

vital and economic consequences. Disease causes a loss of income and can propel the individual into poverty. As such, these two dimensions



Existence of demand related to the expression of a need for health

deserve special attention because they have a determinant impact on

(disease prevention, disease treatment, reproduction, etc.);

access to care. We will devote the following sections to them.

8 A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

a

n

k

Economic Brief 2 0 1 4



AfDB

w w w . a f d b . o r g

3. Territorial Inequalities in Health Care Facilities

he availability of a health service whether public or private and

hospitals represent approximately 50% of all public sector beds. The

good coverage of the national territory in health infrastructure

health system further includes the polyclinics of the National Social

contribute significantly to the reduction of inequality in access to

Security Fund, hospitals under the Ministry of National Defence and

health care (Gold Zeynep et al., 2009). After an overview of Tunisia’s

the facilities of the Ministry of the Interior and Local Development

health care system, we will analyse the availability of public and private

(Arfa and Elgazzar, 2013).

T

health resources in the 24 governorates and try to build a composite indicator of health care facilities that can help to assess inter-regional

In Tunisia, access to the various levels is open and not by referral. Hence,

inequalities and to monitor their trend.

the first line may start at the primary, secondary or tertiary level. The intense activity of emergency services in hospitals is an example.

1-

Overview of the health care system in 20106 The private health care sector has grown significantly: it accounts for

In Tunisia, the health care delivery system is primarily public although

about 14% of total bed capacity and 70% of advanced technology

there is a growing private sector (Arfa, 2007). Nationally, more than

services. In terms of human resources, it employs 48.3% of physicians

7

86% of hospital beds are in the public sector. The leading care

(55.6% of specialists and 42% of general practitioners), 77.5% of dentists

provider is the Ministry of Health. The public provision of health services

and 81.5% of pharmacists. Private clinics are mostly concentrated in

8

is structured in three levels of care. Primary health care is provided

major coastal urban areas (Arfa and Elgazzar, 2013).

by 2 085 primary health centres (PHC), with 2 923 district hospital beds (consisting of small facilities with an average of 27 beds per

Despite an equalizing trend and geographical accessibility deemed

facility) and maternity centres, which together account for about 15%

acceptable for front-line facilities, the distribution of health services

of public sector bed capacity. This care level implements preventive

in the country is characterized by a certain inequality that should be

health policy. It handles 60% of public sector medical outpatients and

evaluated and corrected.

more than 1.3 million reproductive health visits (perinatal consultations,

2-

contraception, STI, screening for female cancers, etc.). It manages

Trend in the distribution of health facilities

the health activities of all pupils and students at all levels (pre-school, primary, secondary, university, vocational training and others).

To study the trend in the distribution of health facilities in the country,

Level II health care is provided by 33 regional hospitals (RH), which

centres and hospital beds over the past decade. These two indicators

account for 35% of total bed capacity and medical specialists in the

characterize public health provision.9 We will complete them with

we will analyse the allocations of the 24 governorates in primary health

human resource indicators (public sector physicians and paramedical

public sector.

staff). We will also analyse the availability of private practices, Level III health care consists of a network of 24 hospitals and

pharmacies and dental offices in governorates. These three indicators

academic institutions with an average size of 405 beds. These

characterize private health provision.

6

The main source is the 2010 Health Map of Tunisia (Ministry of Health). In 2010, the theoretical public bed capacity is 19 565 beds, while private clinics account for only 3 029 beds. 8 When people need health care, they turn most often to primary health care services, which are the first point of contact with the system. In general, primary health care has a double function. First, it provides preventive and curative support for common diseases. Then it acts as an interface and when necessary, refers patients to higher levels; it facilitates their movement within the health system when more specialized care is needed. 9 We will not discuss any issues related to the efficient use of the infrastructure, (World Bank, 2008). Overall, the potential of district hospitals has been under-utilized because of the weakness of their technical facilities, which limits the scope of diagnostic and therapeutic care management. Regarding regional hospitals, despite generally satisfactory technical facilities, productivity is affected by the lack of specialists, who are more attracted to university hospital careers or private practice. Lastly, UTHs control most of the heavy equipment in the public sector. Skill levels are high, but the sector suffers from consultation congestion, due to the weakness of the second level, as well as an increasingly strong tendency for brain drain to private practice that offers significantly higher income levels (WHO, 2010). 7

9 A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

a

n

k

AfDB 2 0 1 4



Economic Brief w w w . a f d b . o r g

For each indicator, we will refer to the per capita endowment or its

2003 (Figure 10). It increased from 4 795 in 2003 to 5 051 in 2010. This

inverse (the number of inhabitants per unit). We will see if, on average,

trend could mean a more efficient use of PHCs for the benefit of a denser

the indicator improves. In addition, the use of the coefficient of

population. It may also indicate less easy access to these centres. In

variation 10 will reveal whether, overall, the distribution of health

the latter case, the PHC facility, as the access point of the population

resources has become more or less unequal. Comparing the values

to the health system, is fulfilling its role of health prevention and curative

of the indicator at the beginning of the last decade and at its end, for

treatment of common diseases less than before.

each governorate, will allow description of the trend of the governorate. The data used are drawn from the Tunisian Statistical Yearbook for

However, the coefficient of variation of the number of inhabitants per

2006-2010 (Serial No. 53).

PHC according to the governorate shows a downward trend indicating

2-1

in inequalities is due primarily to the deterioration of the situation in

a reduction in inter-governorate inequality (Figure 10). The decrease

Trend in the distribution of public health care facilities

governorates such as Monastir, Bizerte, Sfax, Sousse, Nabeul, Ben 2-1-1. Primary Health Centres (PHCs)

Arous, Ariana and Tunis, as shown in Figure 12. These governorates are relatively well served by Level III and PHCs providing consultations

The number of inhabitants by PHC globally reflects a reversal trend from

almost daily.

Figure 10: Inhabitants per PHC (1998-2010)

Figure 11: CV inhabitants per PHC (1998-2010)

1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010

5100 5000 4900 4800 4700 4600 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010

5100 5000 4900 4800 4700 4600

Figure 12: Inhabitants per PHC by governorate 25 000 20 000 15 000 10 000 5 000 0

2000 Tataouine Kebili Siliana Le Kef Tozeur Beja Zaghouan Mahdia Sidi Bouzid Gafsa Jendouba Kasserine Medenine Gabes Kairouan Ensemble Monasr Bizerte Sfax Sousse Nabeul Manouba Ben Arous Ariana Tunis

2010

10 For a Y expenditure distribution with average  Y, the coefficient of variation (CV) is derived from the variance. It is defined as the ratio of the standard deviation σ to the average of expenses: CV = σ /Y where σ2 = 1/ N ∑(Yi -Y)2 CV is used to compare the dispersions of distributions with different averages. The higher the CV the more dispersed the distribution will be.

10 A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

a

n

k

AfDB

Economic Brief 2 0 1 4



w w w . a f d b . o r g

In this respect, it should be noted that this indicator does not allow

improving the frequency of primary health care consultations rather

proper assessment of the availability of health services in regions

than the multiplication of small health centres.12

because it does not take into account the pace of consultations in 2-1-2. Hospital beds

the PHC. Indeed, primary health centres differ in their type and the pace of medical consultation observed there. In 2010, 1040 of the 2085 PHCs provided at most one day of consultation per week. The

Unlike the first indicator, the bed equipment rate or number of public

2 085 primary health centres are in fact equivalent to 870 full-time

beds per 1000 inhabitants shows a significant increasing trend in the

centres.11 In addition, the pace of consultations is not equally distributed

public provision of care (Figure 13). The number of public beds per 1000

among the PHCs. The proportion of PHCs providing medical

inhabitants increased from 1.74 in 2001 to 1.85 in 2010. Similarly, the

consultation six days per week is only 4.4% in Medenine, 4.8% in

coefficient of variation of the bed rate in governorates decreased (Figure

Tataouine, 9.3% in Tozeur, 8% in Mahdia, 8.6% in Kébili, 8.9% in Sidi

13). This therefore means reduced inter-governorate inequality.

Bouzid and 9.6% in Beja. Thus, in these governorates, most of the population does not have daily access to mobile, community primary

Despite these positive developments, the public bed equipment rate

health care services. Accordingly, efforts should focus more towards

varied in 2010 from 0.4 in Ben Arous to 4 in Tunis (Figure 15).

Figure 13: Bed availability rate (1998-2010)

Figure 14: CV bed availability rate

1.9

0.60 0.50 0.40 0.30 0.20 0.10 0.00

1.8 1.7 1.6 1.5

1998 1999 2000 2001 2002 2004 2005 2006 2007 2008 2009 2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

Figure 15: Public hospital beds per 1 000 inhabitants by governorate

2000 Ben Arous Ariana Sidi Bouzid Kairouan Kasserine Nabeul Jendouba Mahdia Bizerte Medenine Siliana Gabes Tataouine Beja Sfax Kebili Tunisie entère Le Kef Gafsa Monasr Sousse Manouba Zaghouan Tozeur Tunis

4.50 4.00 3.50 3.00 2.50 2.00 1.50 1.00 0.50 0.00

1999

1998

1.4

2010

11 The 2010 Health Map can be used to calculate for each governorate the number of PHC full-time equivalent and the number of inhabitants per PHC full-time equivalent, but this statistic is not available for earlier years. 12 Three evaluations of the primary health care system (1997, 2000 and 2004) were carried out by the Ministry of Health as part of the Health Districts National Development Programme. It is a programme developed since 1994 by the Directorate of Primary Health Care. The overall objective of PNDCS is to make all health units in the country able to manage the health status of the population through a set of preventive, curative, promotional and rehabilitation activities, and ensure coordination within and between sectors involving all health stakeholders. The PNDCS has two specific objectives: firstly, improving the (technical and relational) quality and efficiency of care at the primary health centres (PHC) and the district hospital and secondly, strengthening and involving the population in health management. The main recommendations were: (i) optimizing health delivery by moving from a logic of coverage with infrastructure (number of hospital beds, number of DHCs) to an approach of coverage with effective services (number of medical consultation days offered, range of hospital services with appropriate technical facilities); (ii) improving the dimensions of care quality, for example by adapting opening hours to the rhythm of the patient population.

11 A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

a

n

k

AfDB 2 0 1 4



Economic Brief w w w . a f d b . o r g

Reducing inequality reflects both:

to ensure that at each of its levels, the tasks assigned are carried out in the best conditions. These missions should be clearly specified. Each

l

An improvement in bed equipment in many governorates among the

hospital institution should have a roadmap that allows strategic

least well off, notably: Sidi Bouzid, Kasserine, Siliana, Tataouine, Beja,

management consistent with the whole system.

Kébili, Le Kef, Gafsa, Tozeur... l

a worsening situation in other governorates. The number of beds per

2-1-3.

Public sector physicians

1 000 population decreased in Ariana, Sfax, Monastir, Sousse, Manouba and Zaghouan. These governorates are experiencing strong

There has been a significant decrease in the number of inhabitants per

population growth induced in particular by migration. Almost ten years

physician in the public sector in all governorates with the exception of

of stagnation of the bed equipment rate in Tunis or the reduction

Ariana. This decrease reflects an increase in public health care provision.

thereof in large cities such as Sfax, Monastir, Sousse, Ariana and

Overall, the number of inhabitants per physician dropped from 2176 in

Manouba are all the more disturbing trends since these cities are

2002 to 1569 in 2010 (Figure 16). Until 2008, this trend increased

university teaching hospital centres. The quality of training at the

inequality between governorates. However, since 2008, the inequality

patient’s bedside may be affected by the increasingly less favourable

gap is closing but is still significant (Figure 16). The number of inhabitants

conditions in which it takes place. As such, there is the risk of a vicious

per physician in the public sector varied, in 2010, between 493 in Tunis

circle that reproduces mediocrity. To decongest level III and allow it

and 3377 in Kasserine, a ratio of 1 to 6.8.

to devote more time to training and research missions, level II should be developed.

There is a clear improvement in physician availability in many governorates (Figure 18). However, at the same time, the situation

Indeed, the bed equipment rate analysed below takes into account beds

has changed very little in governorates like Kasserine, Medenine,

at the three levels. It therefore hides disparities within levels. In the absence

Nabeul and Kébili. Yet these governorates were initially less endowed

of detailed statistics for the study period, it was not possible for us to

with physicians.

appraise the paces of matching developments. However, interviews with stakeholders led us to conclude that the weak link in the system is level II.

Five governorates are better provided with public health physicians

Often this level is ineffective or non-existent and therefore needs to be

than the country as a whole. The number of inhabitants per physician

strengthened.

there is less than 1569. They are Tunis, Sousse, Monastir, Sfax (which

Finally, the quest for greater equity in the health system should not result

twice less the number of physicians; the number of physicians per

in levelling from the bottom, or in a substantial carrying forward of

inhabitant is higher than 3 000 in Kairouan, Jendouba, Sidi Bouzid,

activities from a certain level to a higher one. The system should be able

Medenine and Kasserine.

enjoy level III services) and Tozeur. Conversely, five governorates have

Figure 17: CV Inhabitants per public health physician

Figure 16: Inhabitants per public health physician

3000 2500 2000 1500 1000 500 0

A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

2010

2009

2008

k

n

2007

2005

a

12

2006

2004

2003

2002

1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010

0.37 0.36 0.35 0.34 0.33 0.32 0.31 0.30

Economic Brief 2 0 1 4



AfDB

w w w . a f d b . o r g

Figure 18: Inhabitants per public health physician by governorate 5000 4500 4000 3500 3000 2500 2000 1500 1000 500 0

2002

Tunis Sousse Monas r Sfax Tozeur Ensemble Zaghouan Manouba Mahdia Tataouine Ariana Siliana Bizerte Beja Gafsa Kebili Gabes Le Kef Ben Arous Nabeul Kairouan Jendouba Sidi Bouzid Medenine Kasserine

2010

in 2002 to 308 in 2010. The indicator has improved in all governorates

The overall improvement in the availability of physicians veils significant deficits in medical specialists (surgery, obstetrics, ophthalmology,

except Sousse, Monastir and Ariana. The trend of the coefficient of

orthopaedics, anaesthesiology ...).

variation across governorates indicates a gradual reduction of intergovernorate inequalities (Figure 19).

Better coverage of the national territory in level II beds is necessarily However, there are still significant inequalities. The number of

concomitant with better provision of these regions with physicians in general and specialist physicians in particular.

inhabitants per public sector paramedical staff varied in 2010 between 150 in Tunis and 720 in Ariana, a ratio of 1 to 4.8. The six governorates

2-1-4.

Public sector paramedical staff

best equipped with paramedical staff are Tunis, Tozeur, Sousse, Monastir, Gafsa and Kef. The six governorates least equipped are

The number of inhabitants per senior technician is a clear indication of

Zaghouane, Nabeul, Kasserine, Sidi Bouzid, Ben Arous and Ariana,

the increase in public health care provision (Figure 19). It went from 341

as shown in Figure 21.

Figure 19: Inhabitants per paramedical staff

Figure 20: CV Inhabitants per paramedical staff 0.60 0.50 0.40 0.30 0.20 0.10 0.00

360 340 320 300

1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010

1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010

280

13 A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

a

n

k

AfDB 2 0 1 4



Economic Brief w w w . a f d b . o r g

Figure 21: Inhabitants per paramedical staff by governorate 1200 1000 800 600 400

2002

200

2010 Tunis Tozeur Sousse Monas r Gafsa Le Kef Kebili Beja Ensemble Tataouine Sfax Mahdia Bizerte SIliana Gabes Jendouba Manouba Kairouan Medenine Zaghouane Nabeul Kasserine Sidi Bouzid Ben Arous Ariana

0

Once more, better coverage of the territory with level II beds will involve

better. In these areas, even when the number of inhabitants per physician

better provision of these regions with senior health technicians and

is already low, this number has continued to decrease. By contrast, in

nurses.

the hinterland and in the least developed governorates, the propensity of physicians to settle there was low and the number of inhabitants per

2.2

Trend in the distribution of private health care facilities

physician remained high and/or was on the increase. To counteract this spontaneous location of physicians, it is important to negotiate with the

2-2-1.

Private practice offices

medical corps to revise the institutional framework governing the opening of private practice offices and/or to grant special incentives to physicians

The number of private practice offices has increased considerably

who settle in priority areas. The Order of 24 December 2009 in part

throughout the past two decades such that the number of inhabitants

enshrined the idea by providing for compensation for medical

per office has on average been divided by 1.66 in 10 years (Figure 22).

specialists13 practising in the private sector and contracted with health

Inequalities between governorates declined until 2004, but have since

facilities in priority areas (defined by the Order of 1 March 1995 issued

increased (Figure 22). Between 2004 and 2010, the situation improved

by the Prime Minister laying down priority health areas for the granting

in all governorates except Siliana, Sidi Bouzid and Tataouine.

of certain benefits). The effects could be assessed.14 In France, a solution to the “medical desert” phenomenon was the establishment of “medical home”, that is to say, a structure whose main advantage is that of

In 2004, the ratio between the governorate best provided with private practice offices (Tunis) and the least provided governorate (Siliana)

bringing together in one place many practitioners and several specialties

was 1 to 11.6. In 2010, this ratio rose to 14.3 (Figure 24). This problem

with the purpose of saving by pooling and sharing certain costs. However,

is not specific to Tunisia; the same situation prevails in several

these structures instead resulted in the creation of a wider geographical

developed countries, including France where it is called “medical

network of rural and “desert” areas. Proposals have been made to

desert” (Potvin Moquet, Jones, 2010; High Council of Public Health,

hamper the freedom of installation of physicians rather than encourage

2009 and Senate 2013).

them to open offices in areas where medical facilities are scarce. In particular, this means excluding from health insurance physicians who

Thus, in a context where physicians are free to choose their location,

choose to settle in already saturated areas. As a result, since their patients

there was a craze for major urban centres and areas where the

are not reimbursed by social security, it would be impossible for a young

purchasing power of the patient base is higher and the quality of life

physician to build a patient base (Senate, 2013).

13

TND 500 for specialists in surgery and obstetrics and gynaecology, TND 400 for all other specialties. There is concern that physicians may abuse this situation by diverting patients from the hospital to their private practice or by using hospital equipment for private purposes. Ethical standards and rules of governance should be enacted. Very strict controls must be implemented. 14

14 A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

a

n

k

AfDB

Economic Brief 2 0 1 4

Figure 22: Inhabitants per private practice office



w w w . a f d b . o r g

Figure 23: CV Inhabitants/Private practice office

2010

2009

2008

2007

2006

2005

2000

1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010

0

2004

1 000

2003

2 000

2002

3 000

2001

0.70 0.60 0.50 0.40 0.30 0.20 0.10 0.00

4 000

Figure 24: Number of inhabitants per private practice office by governorate 16000 14000 12000 10000 8000 6000 4000 2000 0

2001

Tunis Sfax Ariana Sousse Ben Arous Ensemble Nabeul Medenine Monas r Bizerte Manouba Gabes Mahdia Beja Gafsa Zaghouan Le Kef Kairouan Kebili Jendouba Tozeur Tataouine Sidi Bouzid Kasserine Siliana

2010

2-2-2.

Dental offices

per office was divided on average by 1.76 between 2001 and 2009 (Figure 25). Inequalities between governorates have significantly

The number of dental offices has increased significantly over the

decreased, as shown by the trend of the variation coefficients

last two decades to the extent that the number of inhabitants

(Figure 25).

Figure 26: CV Inhabitants per dental office

Figure 25: Inhabitants per dental office

14 000 12 000 10 000 8 000 6 000 4 000 2 000 0

0.80 0.60 0.40 0.20

15 A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

a

n

k

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009

0.00

AfDB 2 0 1 4

Economic Brief



w w w . a f d b . o r g

Figure 27: Number of inhabitants per dental office by governorate 60000 50000 40000 30000 20000 10000 0

2004

Tunis Ariana Sousse Sfax Ben Arous Nabeul Bizerte Ensemble Monas r Manouba Mahdia Medenine Gafsa Le Kef Gabes Ensemble Jendouba Kairouan Beja Siliana Zaghouan Kebili Sidi Bouzid Tataouine Kasserine Tozeur

2009

Between 2001 and 2009, the situation improved in all governorates

2-2-3.

Pharmacies

and is particularly striking in Zaghouan, Siliana, Sidi Bouzid and Kasserine (Figure 27).

Overall, the number of pharmacies has increased faster than the country’s population, such that the number of inhabitants per pharmacy

In 2004, the ratio between the governorate most provided with dental

has been divided by 1.2 in 10 years (Figure 28). Accordingly, the

offices (Tunis) and the least provided governorate (Zaghouan) was 1 to

inequalities between governorates decreased significantly as shown by

30. In 2009, this ratio dropped to 11 (between Tunis and Kebili). So

the variation coefficient (Figure 28). The situation improved in all

there are still margins to reduce inequalities in dental office availability.

governorates between 2001 and 2010 (Figure 30).

Figure 28: Inhabitants per pharmacy

Figure 29: CV Inhabitants per pharmacy

8 000

0.50

6 000

0.40 0.30

4 000

0.20

2 000

0.10

0

60000

Figure 30: Inhabitants per pharmacy by governorate

50000 40000 30000 20000

2004

10000

2009 Tunis Ariana Sousse Sfax Ben Arous Nabeul Bizerte Ensemble Monas r Manouba Mahdia Medenine Gafsa Le Kef Gabes Ensemble Jendouba Kairouan Beja Siliana Zaghouan Kebili Sidi Bouzid Tataouine Kasserine Tozeur

0

16 A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

a

n

k

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010

0.00

Economic Brief 2 0 1 4

The improvement has been very significant in governorates that were



AfDB

w w w . a f d b . o r g

increased. Hence, it is necessary to develop an appropriate strategy

15

less provided with pharmacies. Despite this, in 2010, the governorate

for addressing this challenge. It would be advisable to review the

most provided with pharmacies (Tunis) had per capita 2.63 times

criteria for opening positions of public health physician at regional

more pharmacies than that least provided (Kasserine). It should be

level and the institutional framework governing practices at the private

noted that the inequalities in the number of pharmacies nationwide

level. Similarly, public-public partnerships (such as agreements

are much lower than inequalities in private practice offices due to very

between academic physicians and regional hospitals) or, failing that,

strict regulations.

public-private partnerships likely to make disadvantaged areas more attractive, could be considered for medical specialists.

2.3

Summary The availability of private medical practitioners is characterized by high

The analysis shows a generally favourable trend that however hides

levels of inequality; the ratio between the most endowed and the least

internal distortions (Table 1 below). The availability of PHCs is declining.

endowed governorates is 14.3. They are followed by dental offices (ratio

Level II, which is the reference for Level I, would not be very effective

of 11.3) and hospital beds (10.7). The most fairly distributed resources

because it is poorly resourced. Therefore, there is an important carry

are pharmacies and paramedical staff. Like in pharmacy, the private

over to Level III, which thus replaces Level II, thereby causing

practice of dentistry and medicine should be better regulated.

inefficiencies. Obviously, there is a need to develop a strategy that strengthens and revitalizes primary health care in the country and

Lastly, the status of health care facilities in one governorate cannot be

enhances Level II.

analysed by reference to a single determinant. As such, all components of the sector and complementarity between the various providers should

Similarly, it has been shown that the trend of all dispersion indicators

be considered simultaneously. Consequently, it seems worthwhile to

is favourable, with the exception of the indicator for physicians,

conduct an analysis of health care provision that integrates all the

whether in the public or private sector. Although the number of

determinants of the provision so as to end up with relatively homogeneous

physicians per inhabitant witnessed a significant drop between 2002

groups. It would further be interesting to develop for each governorate

and 2010 (28.5% for the public health sector and 21% for private

a composite indicator of health care facilities so that a governorate could

practice), the gaps have widened between the better endowed and

gauge its position in relation to other governorates as well as the progress

less endowed governorates, and the variation coefficients have

it may achieve over time.

15 The operation of pharmacies is strictly subject to a numerus clausus, which is established on the basis of five areas according to delegations. It is regularly reviewed to adapt to the new realities of the profession and demographics.

17 A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

a

n

k

AfDB 2 0 1 4



Economic Brief w w w . a f d b . o r g

Table 1: Level and distribution of major health care provision indicators Indicator

Inhabitants /PHC

Hospital beds/ 1 000 inhabitants

Inhabitants/ Physician (public sector) Inhabitants/ Paramedical staff Inhabitants/ Private practice (2004-2010) Inhabitants/ Dental office (2002-2009)

Inhabitants/ Pharmacy

Statistics

2002

2010

2010/2002

Average

4807

5051

+5%

CV

0.94

Max/min

Disadvantaged governorates

Appraisal Unfavourable

Tunis

0.89

Favourable

Ariana

8.63

8.66

Moderate

Ben Arous

Average

1.7

1.85

Favourable

Ben Arous

CV

0.49

0.41

Favourable

Ariana

Max/min

47.4

10.7

High

Sidi Bouzid

Average

2196

1569

Favourable

Kasserine

CV

0.32

0.35

Unfavourable

Medenine

Max/min

5.27

6.85

Moderate

Sidi Bouzid

Average

341

308

Favourable

Ariana

CV

0.44

0.3

Favourable

Ben Arous

Max/min

6.68

4.8

Low

Sidi Bouzid

Average

2128

1681

Favourable

Siliana

CV

0.5

0.6

Unfavourable

Kasserine

Max/min

11.6

14.3

High

Sidi Bouzid

Average

8847

5774

Favourable

Tozeur

CV

0.66

0.58

Favourable

Kasserine

Max/min

24

11.3

High

Tataouine

Average

6756

5604

Favourable

Tozeur

CV

0.39

0.27

Favourable

Kasserine

Max/min

4.4

2.6

Low

Tataouine

+8.8%

-28.5%

-8.7%

-21%

-34.7%

-17%

3-

Health care facilities in 2010: cluster analysis

1. Average distance to get to a regional hospital

3.1

Indicators

3. Inhabitants by PHC

2. Average distance to get to a general hospital 4. Proportion of PHCs providing medical consultation 6 days of 6 To analyse the distribution of health care facilities between the 24 governorates

5. Inhabitants per PHC full-time equivalent (FTE)

in Tunisia, we will refer to a broad set of indicators that characterize such

6. Inhabitants per primary care physician

facilities. These are indicators relating to health infrastructure, human resources

7. Frontline bio-medical laboratory unit per 100 000 inhabitants

in public and private health facilities, and equipment. These indicators are

8. Frontline radiology unit per 100 000 inhabitants

drawn from the Tunisia 2010 Health Map published by the Ministry of

9. Frontline dental chairs per 100 000 inhabitants

Health and/or from the Tunisia Statistical Yearbook published by the INS.

10. Inhabitants per day pharmacy 11. Inhabitants per night pharmacy

3-1-1.

Health infrastructure indicators

12. Private bio-medical laboratories per 100 000 inhabitants 13. Haemodialysis machines per 100 000 inhabitants (public and

These are indicators on:

private)

18 A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

a

n

k

Economic Brief 2 0 1 4

3-1-2.

Common equipment indicators16



AfDB

w w w . a f d b . o r g

This cluster covers 626 PHCs, 21 district hospitals (341 beds), 13 regional hospitals (1 315 beds) 22 university hospitals (92% of the overall with 9 032

1. Hospital bed equipment rate (public and private)

beds) and 4 590 private medical practices out of 6 273 (general practitioners

2. Public hospital bed equipment rate

and specialists). It is characterized by good positioning in terms of access

3. Private bed equipment rate (in clinics)

to hospitals and backed by sustained availability of alternative types of

4. General surgery bed equipment rate

infrastructure and also by poor positioning in terms of the number of

5. Gynaecology and obstetrics bed equipment rate

inhabitants per Primary Health Centre (PHC). Hence, we have:

6. Paediatric bed equipment rate 7. Ophthalmology bed equipment rate

l

The best location in terms of access to regional hospitals and general

l

The largest proportion of PHC providing medical consultation 6 days

11. Anaesthesiology bed equipment rate

l

The highest number of inhabitants per PHC on average;

12. Psychiatric bed equipment rate

l

The highest number of inhabitants per PHC in full time equivalent;

l

A moderate number of inhabitants per primary care physician;

l

The lowest number of primary biomedical laboratory units for

1. Density of physicians (per 100 000 inhabitants)

l

The lowest number of primary radiology units for 100 000 inhabitants;

2. Density of general practitioners (per 100 000 inhabitants)

l

The lowest number of primary health care dental chairs for 100 000

inhabitants)

l

The highest number of inhabitants per day-time pharmacy;

4. Density of general practitioners in the private sector (per 100 000

l

Moderate number of inhabitants per overnight pharmacy;

inhabitants)

l

The highest number of medical analysis laboratories for 100 000

8. ENT bed equipment rate

hospitals (except Médenine due to its low density);

9. Orthopaedic bed equipment rate

a week;

10. Cardiology bed equipment rate

3-1-3.

Human resource indicators

100 000 inhabitants;

inhabitants;

3. Density of general practitioners in the public sector (per 100 000

inhabitants;

5. Density of specialists (per 100 000 inhabitants) 6. Density of public sector specialists (per 100 000 inhabitants)

The highest number of haemodialysis machines for 100 000

l

inhabitants.

7. Density of private sector specialists (per 100 000 inhabitants) 8. Medical density per specialty (all physicians) 9. Density of pharmacists (per 100 000 inhabitants)

Besides Médenine, this cluster is composed of university hospital

10. Density of dentists (per 100 000 inhabitants)

cities.18 It can be observed that there is a predominance of tertiary

11. Density of nurses, nursing aides and senior technicians (per 100 000

care, including emergency services that are particularly in demand

inhabitants).

and are overriding the PHC.19 The question then is not so much

3-2

the motives underlying the people's preference for emergency room

whether or not to increase the density of PHCs but also to understand

Heath Infrastructure Distribution

services to PHCs. Should it be blamed on the overly broad primary Based on health infrastructure indicators in the 24 governorates observed

health network or the discrepancy between its temporal accessibility

in 2010 (Tables 15, 16, 17 and 18 in Annex 1), a dynamic clusters analysis17

and the quality of care it provides?20 These two aspects certainly

was conducted in four clusters (Map 3).

deserve special consideration and it is appropriate to both standardize the availability of infrastructure and upgrade the operation of all

3-2-1- The first cluster (Table 13 in Annex 1) includes the Tunis, Ariana, Ben

structures at all levels. The certification of hospitals would be entirely

Arous, Manouba, Sousse, Monastir, Sfax and Médenine governorates.

appropriate. In this context, an agency for the accreditation and

16 We did not consider indicators for equipment that has a regional scope and serves several governorates, such as the equipment rate for public beds with university status by major region (north, centre and south); the MRI equipment rate by major region; the scanner equipment rate and the equipment rate for other heavy equipment. 17 The method for classifying dynamic clusters is essentially based on the distribution of a population into homogeneous groups (classes or clusters) using the core concept associated with each class. It may involve, as in our study, for example, discovering the main governorates with the closest health facilities.

19 A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

a

n

k

AfDB 2 0 1 4



Economic Brief w w w . a f d b . o r g

certification of health services was established by Decree No. 2012-

an intensive care unit, with the technical equipment needed to support

1709 of 06/09/2012.

different types of emergencies, district hospitals currently perform only a single real hospital function: carrying out eutocic deliveries. Regional

3-2-2- The second cluster (Table 14 in Annex 1) includes the Bizerte,

hospitals that are supposed to provide Level II care, lack specialists (the

Nabeul and Kebili governorates. This cluster is characterized by a

cluster is very poorly staffed with specialist physicians) and equipment.

very average positioning with respect to all criteria relating to public

Furthermore, quality certification will help bring these structures up to

infrastructure (hospitals, PHC ...) or private facilities (pharmacies,

standard. Lastly, it would be appropriate to develop specific incentives

dental offices, laboratories, etc.). It includes 273 PHC, 9 CH and 6

to induce private stakeholders to settle in these governorates.

RH (2 135 beds i.e. 11% of the overall nominal capacity), 709 private practices, 286 pharmacies, 221 dental offices...

3-2-4- The fourth cluster (Table 16 in Annex 1) includes the Jendouba, Kairouan, Kasserine and Sidi Bouzid governorates. This cluster is

3-2-3- The third cluster (Table 15 in Annex 1) includes the Béja, Gabès,

characterized by even low rate of access to hospitals and available

Gafsa, Le Kef, Mahdia, Siliana, Tataouine, Tozeur and Zaghouan

basic infrastructure (with 27 CH, 470 PHC and only 4 RH). Furthermore,

governorates.

the most reduced availability of private facilities can be observed (350 private practices). The shortfall of private health services is probably

These nine governorates have a rather low hospital access rate (due

due to the low standard of living in these governorates and the lack of

to low population density) and limited availability of various types

effective demand for health services.

of infrastructure, especially those that are private-owned. This shortcoming is partly offset by proper positioning in terms of the number

Hence, this fourth cluster comprises all priority governorates in terms

of inhabitants per PHC and per primary care physician.

of infrastructure wherein an intervention to enhance public health coverage would allow coverage similar to the rest of the country, and

Yet these governorates have 46 district hospitals out of a total of 109,

seems to be a necessary step to boost private coverage through the

10 regional hospitals (about 33 throughout Tunisia) and two university

ripple effect. In this regard, it would be wise to develop specific incentives

hospitals (in Mahdia and Zaghouan), with a nominal capacity of 4 133

to induce private stakeholders to settle in these governorates.

beds (21 % of national capacity). The table below summarizes the specificities of each cluster with respect However, while a hospital should have at least one surgical ward and

to health infrastructure.

18

Tunis, Sousse, Monastir and Sfax. In a study on the reasons for recourse to the emergency services of major hospitals in Greater Tunis (Ben Gobrane et al., 2012), the major reasons given by patients are quick and easy access to emergency services, the availability of equipment as compared to PHCs and, for the populations, inappropriate working hours of primary care facilities that work only in the morning. Hence, recourse to emergency services is partly due to the shortcomings of primary care medicine. 20 In most of these structures, consultations are carried out only in the morning. In rural areas, the length of consultations is notoriously reduced given the number of consultations conducted. In urban areas, opening hours do not match the time users are available for consultation. The result is threefold: either unwarranted recourse to hospital emergency services at different levels, or delay in recourse or forced and costly recourse to private primary care facilities. 19

20 A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

a

n

k

Economic Brief 2 0 1 4



AfDB

w w w . a f d b . o r g

Table 2: Distribution of health infrastructure by clusters Characteristics of Allocations Governorates

Cluster 1

Tunis, Ariana, Ben Arous, Manouba, Sousse, Monastir, Sfax and Medenine.

Cluster 2

Bizerte, Kebili et Nabeul

Cluster 3

Beja, Gabes, Gafsa, Le Kef, Mahdia, Siliana, Tataouine, Tozeur et Zaghouan

Cluster 4

Jendouba, Kairouan, Kasserine and Sidi Bouzid

Other front-line structures

Private medical structures

Hospitals

PHC

+

-

-

+

+/-

+/-

+/-

+/-

-

+

+

+/-

+/-

+/-

+/-

-

The analysis below focused on hospital availability in terms of average

located. Moreover, the quality of private technical equipment in Tunis

distance access to a hospital, but did not provide information on the

and the concentration of specialists observed there have fostered the

availability of the said hospitals. Furthermore, to better clarify this aspect,

export of health services. Standardization and certification procedures

we will consider some indicators of bed availability rates.

would allow a greater development of this sector.

3-3

Bed Availability Rate

3-3-2- The second cluster (Table 18 in Annex 1) includes the Sfax, Mahdia, Le Kef and Médenine governorates. This cluster is characterized

Based on the bed availability rate in the 24 governorates observed in 2010

by its leadership in private infrastructure. It ranks first in terms of bed

(Table 19, 20, 21 and 22 in the annex), a dynamic cluster analysis was

availability rates in clinics. However, the number of its public beds is

conducted in four clusters (Listing 2: Map 4).

relatively small (3 567 out of 19 565 beds, i.e. 18%). Overall, it ranks

3-3-1- The first cluster (Table 17 in Annex 1) includes the Tunis,

orthopaedics, ORL, ophthalmology, paediatrics and anaesthesiology.

Sousse, Monastir, Tozeur and Manouba governorates. This cluster is

However, this cluster has the highest number of beds in psychiatric

characterized by:

wards and ranks immediately after the first cluster with respect to

third in terms of hospital beds as well as specialties in general surgery,

specialized beds in gynaecology and cardiology wards.

• The best positioning in terms of bed availability rates and hospital bed availability rates (public beds). This positioning is observed for

In this cluster, Sfax is a university hospital city. Its two university hospital

all specialties except ORL and psychiatry; and

centres21, which are adjacent to each other, suffer various shortcomings.

• The second positioning in terms of clinic beds (private rooms) as

They are congested and overwhelmed by a workload exceeding their nominal capacity and thereby leading to long waiting periods. They

well as public beds in ORL and psychiatry.

service not only the Sfax governorate but also the southern population In this cluster, Tunis, Monastir and Sousse are three university hospitals

of more than 4 million inhabitants, which implies overuse of equipment.

with a total of 6 664 beds (34% of the national capacity), including 2

Because of such anomalies, these UH increasingly face difficulty in

961 by specialty. These UH are regional in scope and assigned to the

meeting academic training, specialization, high-level care and medical

North and Centre, and even national for certain specialties. They service

research needs in good condition. For over a decade, a new UH has

a much larger population than that of the governorate in which they are

been programmed for Sfax but has not yet been implemented.

21 The Habib Bourguiba Hospital (506 beds) is home to the surgical specialty services; the Hedi Chaker Hospital (889 beds), which is older, provides medical pathology services.

21 A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

a

n

k

AfDB 2 0 1 4



Economic Brief w w w . a f d b . o r g

However, Sfax is characterized by a significant private health sector,

previous investment code provided for some tax incentives 23 for

with clinics (mono-disciplinary or polyclinic), radiology centres, medical

equipment.

laboratories and pharmacies. It serves as a medical platform for the South and for the export of health services primarily for Libyans.

3-3-4- Cluster 4 (Table 20 in Annex 1) comprises the less affluent

3-3-3- The third cluster (Table 19 in Annex 1) comprises the Béja, Gafsa,

Nabeul, Jendouba, Kairouan, Kasserine, Sidi Bouzid and Gabes. There

Kébili and Tataouine governorates. This cluster is relatively very well

have, on average, the lowest public bed availability rates and, to a lesser

equipped in terms of public beds but is poorly equipped with respect to

extent, low private bed availability rates. This is true for almost all specialties.

private beds. It ranks second with respect to public infrastructure (1 831

However, the Northern governorates (Ariana, Ben Arous, Zaghouan,

governorates in terms of beds: Ariana, Ben Arous, Zaghouan, Bizerte,

hospital beds in Levels 1 and 2). The significance of the public sector is

Bizerte and Nabeul) seem to be of lower priority due to their proximity to

highlighted by the specialized beds availability rate in ORL, surgery,

Tunis and the importance of the private sector. In the other governorates

orthopaedics, ophthalmology, paediatrics and anaesthesiology. However,

(Jendouba, Kairouan, Kasserine, Sidi Bouzid and Gabes), not only is the

there is a shortfall of equipment for gynaecology, cardiology and psychiatry.

per capita bed ratio low, but a significant proportion of the beds are district hospital beds, and district hospitals are often poorly equipped and poorly

In addition, this cluster is characterized by a relatively underdeveloped

staffed with specialists. The lack of equipment and resources in DHs has

private sector. It is the most underprivileged group in terms of

transformed these centres into intermediate facilities incapable of resolving

clinic beds. The establishment of clinics is governed by clinical

the problems they encounter, thereby causing users to bypass this level

specifications22, considered stringent as compared to that of public

and resort to the private sector or the secondary and tertiary levels, usually

structures (e.g. it is required to have a nurse for two intensive care

in other governorates. Table 3 below summarizes the distribution of beds

beds, whereas there are no set standards for the public sector). The

between the public and private facilities for each cluster.

Table 3: Bed distribution by cluster Governorates

Public Beds

Private Beds

+

+/-

Cluster 1

Tunis, Sousse, Monastir, Manouba and Tozeur

Cluster 2

Sfax, Mahdia, Le Kef and Medenine

+/-

+

Cluster 3

Beja, Gafsa, Kebili et Tataouine

+/-

-

Cluster 4

Ariana, Ben Arous, Zaghouan, Bizerte, Nabeul, Jendouba, Siliana, Kairouan, Kasserine, Sidi Bouzid and Gabes

-

+/-

22 Decree No. 93-1915 of 31 August 1993 to determine structures and specialties, and standards in terms of capacity, equipment and staffing of private health institutions, as supplemented and amended by Decree No. 99-2833 of 21 December 1999 and Decree No. 2001-1082 of 14 May 2001. 23 Decree No. 94-1056 of 9 May 1994, establishing a list of equipment needed for health and hospital institutions that may qualify for the tax incentives under Section 49 of the Investment Incentives Code and the conditions for granting these benefits, as amended and supplemented by Decree No. 98-967 of 27 April 1998 and Decree No. 2006-382 of 6 February 2006.

22 A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

a

n

k

Economic Brief 2 0 1 4

3-4

Human Resources



AfDB

w w w . a f d b . o r g

Tataouine and Tozeur (Table 24 in Annex 1). In the public hospitals o f the 10 governorates, there are only 1 134 physicians (general

Based on human resource indicators from the 24 governorates, reported

practitioners and specialists), i.e. 17% of public health physicians. The

in 2010 (Tables 23, 24, 25 and 26 in Annex 1), a dynamic cluster analysis

private sector has only 601 physicians (308 general practitioners and

was conducted in four clusters (Listing 3: Map 5).

221 specialists) and relatively few dentists.

3-4-1- The first cluster (Table 21 in Annex 1) comprises the Tunis, Ariana,

3-4-3- Ben Arous, Bizerte, Gabes, Mahdia, Manouba, Médenine,

Sousse, Monastir, and Sfax Governorates. Overall, these five

Nabeul and Zaghouan constitute an intermediate group with moderate

governorates have the highest rates of human resource allocation both

human resource allocation for almost all categories. 25% of doctors

in the public sector (3957 to 6723 physicians, i.e. 59% of all physicians

and 27% of paramedical personnel in the public and private sector

in the public sector) and the private sector (47% of physicians in the

have settled in these 8 governorates (Table 22 in Annex 1).

private sector practice in these five governorates), regardless of the area of specialization. 56% of private dentists and 43% of pharmacies

3-4-4- Béja (Table 23 in Annex 1) is a governorate hinged between the

are located in the governorates of this cluster.

second cluster and the cluster of human resource-deficient governorates.

3-4-2- The cluster of human resource-deficient governorates comprises

The table below compares the distribution of human resources in both

Gafsa, Jendouba Kairouan, Kasserine, Kebili, Le Kef, Sidi Bouzid, Siliana,

sectors for the 4 clusters.

Table 4: Distribution of health workers by cluster and by sector Governorates

Public Beds

Private Beds

+

+

Cluster 1

Tunis, Ariana, Sousse, Monastir, Sfax

Cluster 2

Ben Arous, Bizerte, Gabes, Mahdia, Manouba, Medenine, Nabeul and Zaghouan

+/-

+/-

Cluster 3

Beja

+/-

+/-

Cluster 4

Gafsa, Jendouba, Kairouan, Kasserine, Kebili, Le Kef, Sidi Bouzid, Siliana, Tataouine and Tozeur

-

-

Overall, among the three types of indicators of health care provision,

of health care provision. Hence, it is important to summarize

the geographic distribution of health human resources turns out to be

infrastructure, equipment and human resource indicators into a single

the most unequal and reveals a significant concentration on the coast.

indicator.

Despite an increase in physician density, regional disparities have

4Profile of governorates and composite indicator of health care facilities by governorate.

widened. Qualitatively, the inequalities are even more egregious and more than 2/3 of the specialists are concentrated in the coast, not only for rare specialties but also for the most common ones such as gynaecology and paediatrics.

4-1

Profile of governorates

Such regional breakdown reflects the geographic dichotomy that shows

The ranking of the various governorates into homogenous groups with

a clear regional imbalance in favour of the coast to the detriment of the

respect to their medical infrastructure, their availability in terms of beds

North West and Central West of Tunisia. It is important to consider the

and human resources, help to develop a profile for each governorate

impact of this inequality in human resource allocation on the inequality

according to the groups to which it belongs (see Table below).

23 A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

a

n

k

AfDB 2 0 1 4



Economic Brief w w w . a f d b . o r g

Table 5: Assignment of governorates to clusters (Rji) Human Resources

Governorates

Infrastructure

Equipements

MONASTIR

1

1

1

SOUSSE

1

1

1

TUNIS

1

1

1

SFAX

1

1

2

MANOUBA

2

1

1

ARIANA

1

1

4

MEDENINE

2

1

2

BEN AROUS

2

1

4

TOZEUR

4

3

1

MAHDIA

2

3

2

BIZERTE

2

2

4

NABEUL

2

2

4

GABES

2

3

4

KEBILI

4

2

3

LE KEF

4

3

2

ZAGHOUAN

2

3

4

BEJA

3

3

3

GAFSA

4

3

3

TATAOUINE

4

3

3

SILIANA

4

3

4

JENDOUBA

4

4

4

KAIROUAN

4

4

4

KASSERINE

4

4

4

SIDI BOUZID

4

4

4

24 A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

a

n

k

Economic Brief 2 0 1 4



AfDB

w w w . a f d b . o r g

The table shows that three governorates still belong to the most

This indicator is the average of the three scores awarded to each

advantaged cluster irrespective of the criterion applied. They are Tunis,

governorate.

Sousse and Monastir. By contrast, four governorates still belong to the most disadvantaged cluster. They are Jendouba, Kairouan, Kasserine and Sidi Bouzid (Map 6).

If rji is the ranking of the cluster to which governorate j belongs with respect to resource i, score sji may be defined as the inverse of rji

Between these two groups, the other governorates have more or less

sji = 1/rji given that rji = 1, 2, 3, 4

significant shortfalls depending on the type of resources being analysed. The table defines the scope of intervention required by each governorate.

Thus, the score of governorate j for resource i (sji) is considered a unit

Hence, Sfax, for example, requires greater bed availability rates in order

score where the governorate belongs to the higher class. The score

to measure up to Tunis, Sousse and Monastir.

decreases as the governorate moves away from this class.

4-2

The composite indicator of a governorate’s health care provision is equal

Composite indicator of health care provision

to the arithmetic average of its scores. Ij = ∑ (sji)/3 Each governorate, depending on its profile, may be assigned a composite indicator of health care provision.

The outcomes are presented in the table below.

Table 4: Score by resource and composite indicator of health care facilities by governorate. Governorates

Human Resources

Infrastructure

Equipements

Total

MONASTIR

1.00

1.00

1.00

1.00

SOUSSE

1.00

1.00

1.00

1.00

TUNIS

1.00

1.00

1.00

1.00

SFAX

1.00

1.00

0.50

0.83

MANOUBA

0.50

1.00

1.00

0.83

ARIANA

1.00

1.00

0.25

0.75

MEDENINE

0.50

1.00

0.50

0.67

BEN AROUS

0.50

1.00

0.25

0.58

TOZEUR

0.25

0.33

1.00

0.53

MAHDIA

0.50

0.33

0.50

0.44

BIZERTE

0.50

0.50

0.25

0.42

NABEUL

0.50

0.50

0.25

0.42

GABES

0.50

0.33

0.25

0.36

KEBILI

0.25

0.50

0.33

0.36

LE KEF

0.25

0.33

0.50

0.36

ZAGHOUAN

0.50

0.33

0.25

0.36

BEJA

0.33

0.33

0.33

0.33

GAFSA

0.25

0.33

0.33

0.31

TATAOUINE

0.25

0.33

0.33

0.31

SILIANA

0.25

0.33

0.25

0.28

JENDOUBA

0.25

0.25

0.25

0.25

KAIROUAN

0.25

0.25

0.25

0.25

KASSERINE

0.25

0.25

0.25

0.25

SIDI BOUZID

0.25

0.25

0.25

0.25

25 A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

a

n

k

AfDB 2 0 1 4



Economic Brief w w w . a f d b . o r g

The analysis is based primarily on quantitative indicators. However,

while it is true that the choice of the location of pharmacies, given

qualitative aspects are still essential. The same is true for those relating

that it is highly regulated, resulted in a fairly equal space coverage by

to the efficiency of structures. It is obvious that it is equally important

pharmacies, it is worth noting also that such was not the case for

to improve the functioning of the existing infrastructure to make the

dental practices and especially for private medical practices which,

most out of them, as it is to further increase the density of infrastructure

instead of correcting the deficiencies resulting from the lack of public

(World Bank, 2008).

health services, exacerbate existing inequalities. It is then clear that in the absence of a regulatory framework for the opening of private

To this end, a certification of existing structures should be considered

medical practices, it is important to introduce special incentives for

(at least the most important). The missions and responsibilities of each

young doctors to settle in priority governorates, despite the absence

level of care should be laid down, and resources should be allocated

of an adequately solvent demand. It may include conventional

accordingly. Establishment projects should be stopped for each

cooperation between regional hospitals and specialists in private

structure.

practice for the missing specialties.

The analysis revealed that, despite the reduction in public budgets

However, the choices relating to the health sector and the efforts to

allocated to health, there is reduced inequality in the allocation of

better allocate resources to priority areas can be effective only if they

infrastructure to the various regions. This decrease in inequality is partly

form part of a comprehensive local development strategy for these

due to the concentration of government efforts on the most resource-

areas. The reduction of economic, cultural and social gaps between

deprived areas. However, it stems mostly from a reduction of resources

the governorates can only facilitate and strengthen the health reforms.

allocated to the major urban centres with high population growth dynamics. This movement is synonymous with a decline in access to

To better support our recommendations, we would have liked to test a

health care for vulnerable populations in these regions. It is also

panel model which explains the population’s health status by governorate

symptomatic of reduced resources allocated to structures responsible

(life expectancy and infant mortality rate) and by the prevailing health

for the training of future doctors and paramedical personnel.

status in each governorate. We were not able to do so because health indicators are not published at the governorate level. Such work would

The dwindling resources allocated to the public sector and the ensuing

be very informative and could be conducted later.

reduced access to care would be offset by an increase in private sector resources and a greater availability of its services. In fact, there

Reduction of health inequalities may be achieved only by understanding

is an interesting private sector dynamics in its three components:

the determinants, setting the corresponding objective, informing health

private medical practices, dental offices and pharmacies. However,

care professionals and monitoring the achievements.

26 A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

a

n

k

Economic Brief 2 0 1 4



AfDB

w w w . a f d b . o r g

4. Trend in inequalities in health spending in Tunisia between 2000 and 2010

n recent years, the State’s effort in favour of health has been on the

The analysis of the contribution of the various out-of-pocket health

decline. Public health expenditures accounted for 2.7% of GDP in

expenditure items to the total health expenditure inequality and the

1995 and for only 2.3% in 2011. By contrast, private spending by

trends of this contribution between 2000 and 2010 should help to

I

households rose sharply. In 2010, household spending amounted to

better understand some of the causes of the inequality in out-of-

51% of total health expenditures, with approximately 80% being out-

pocket health spending and to possibly identify policies that should

of-pocket payments and 20% corresponding to health insurance

be implemented to reduce this inequality. In this regard, we will use

premiums. In fact, out-of-pocket payments made by households for

the breakdown of the inequality index by source25 (or component).

health purposes account for 41.2% of national health expenditures

This technique indicates which expenditure categories contributed

(Arfa et al., 2013).

most to the formation of inequality. Any policy aimed at reducing health care inequalities should primarily target these categories (Wagstaff and Van Doorslaer, 2000).

The importance of out-of-pocket health expenditure in national health accounts (Arfa et al., 2007 and 2008) implies that the search for greater equity in health should entail efforts to achieve more equitable

Traditionally, inequalities in expenditures are captured through the

distribution of the out-of-pocket spending of households. Therefore,

Lorenz curve and the Gini index that analyse the distribution of health

it is important to evaluate the inequality of out-of-pocket health

expenditures in themselves. However, despite its merits, this approach

expenditures of households and to analyse their trends using inequality

presents only a partial picture of the inequalities in health expenditures.

indicators.

A more comprehensive picture may be obtained through the concentration curve and index, which analyse the distribution of health

To that end, we will refer to data from the national surveys on budget

expenditures in relation to the distribution of the population’s living

and household consumption of 2000, 2005 and 2010. These surveys

standards ranked from the poorest to the richest.

were conducted on a representative sample of households across the country.24 They provide information on individual consumption of

We will start off by focusing on the Gini index, considered as an indicator

goods and services, and therefore make it possible to study the trends

of inequality in health spending, to see how its breakdown by source

in the living standards of households through their expenditures.

can provide information on the formation and trend of inequality. Next,

These data provide information on expenditures per person per year

we will follow the same approach for the concentration index.26 Lastly,

(SPY) and their structure by various expenditure items, especially

we will analyse the inequality in health spending and its trend in light of

those related to health. The data also facilitates the assessment of

the available data.

the degree of inequality of these expenditures and the analysis of their trends. Various health spending items were covered in the

For estimates of the Gini and concentration indices, we use the STATA

consumer surveys (see nomenclature in Annex 1: Health indicators

12 and DASP27 version 2.2 software, which enable us to calculate and

by cluster 2). They correspond to disbursements made by households.

break down the various indices.

24

For the three surveys, the initial sample is drawn from a stratified random sampling conducted in two stages in each governorate. The sample frame consisted of the data files of the general population census (1994 and 2004 respectively). Regarding the 2000 survey, 12 960 households were sampled and 12 249 responded (representing a response rate of 95%). Concerning the 2005 survey, 13 392 households were sampled and 12 317 responded (that is, a response rate of 92%). As for the 2010 survey, of the 13 392 households initially sampled, 11 291 responded (representing a response rate of 85%). 25 The breakdown of inequality indices was introduced in health economics by Wagstaff et al. (2003). 26 The literature on inequalities shows that there is a range of relevant inequality indicators, including the Theil indicator and the log deviation. In addition, other breakdown techniques, such as groups, are interesting and very informative. In subsequent research, we plan to break down the inequalities in health expenditures using various geographic (environment, region, governorate) and socioeconomic (occupational status of the household head, household size, vulnerability. etc.) criteria, and to identify the relative importance of intra-group and inter-group inequalities and their trend. 27 Distributive Analysis Stata Package, developed by Araar and Duclos (2007). PEP, World Bank, UNDP and Laval University. http://dasp.ecn.ulaval.ca/

27 A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

a

n

k

AfDB 2 0 1 4

1-



Economic Brief w w w . a f d b . o r g

Lorenz curve and the Gini index

1.3.

Breakdown of the Gini index by expenditure categories

To describe the distribution of health expenditures as well as the

Analytically, the Gini index has several expressions. Lerman and Yitshaki

degree and origins of their inequality, we start by adopting the Lorenz

(1985) showed that:

classification method and then the method that consists in calculating acceptable inequality indicators.

 G = 2 cov (Y ; F(Y)) / Y

1.1.

Y being the expenditure vector Y : (Y1, Y2, .... Yn) and F(Y) representing

Lorenz Curve

(Equation 1)

the cumulative distribution considered as random variable uniformly distributed Inequality in the distribution of a variable (health expenditure, for

between [0, 1].

example) may be highlighted using the Lorenz curve (Figure 31) which matches the proportion of the population classified by increasing order

The breakdown adopted makes it possible to analyse the contributions

of health expenditures with the share of total health expenditures

to total inequality of the various expenditure categories (consultations,

incurred by that population.

radiological procedures, drug purchases, etc.). It also facilitates the measurement of their specific inequality contribution to the total inequality (Lerman and Yitshaki 1985).

Figure 31: Lorenz Curve for health spending

Let Y= (Y1, Y2, .... Yn) be the distribution of total health expenditures and Yik the expenditure of the person i in item or category k where k=1....K. Y1,Y2,...,Yk are therefore the expenditure components and Y = ∑ Yk (Equation 2) Given the properties of the covariance, we have:  G = 2 ∑cov (Yk ; F(Y)) / Y

(Equation 3)

where cov (Yk ; F(Y)) represents the covariance of the expenditure in item k with the cumulative distribution of total expenditure.

In the case of two distributions X and Y, X dominates Y if the Lorenz curve relating to distribution “X” constantly lies above the curve relating to distribution “Y”. Distribution “X” is then more egalitarian than “Y”.

When cov (Yk , F(Yk)) Yk, is multiplied and divided by cov(Yk ;F(Yk ))

Nevertheless, when two curves intersect, comparison of the inequality

and by Yk, we obtain the rule of breakdown according to the source

trend becomes impossible. Therefore, the digital indicators for evaluating

(or component), that is:

inequality must be calculated as the Gini coefficient.  k] * (Yk /Y) ] G = ∑ [ [cov (Yk ; F(Y)) / cov (Yk ; F(Yk))] * [2cov (Yk ; F(Yk)/ Y

1.2.

GINI coefficient

(Equation 4) Let’s note:

The Gini coefficient seems to be the most popular of the various inequality indices. It is derived from the Lorenz curve (concentration curve) in that it is the ratio of the area between this curve and the first

- the Gini correlation between the component k and the total

diagonal line and the half-square in which the curve lies. The Gini

expenditure:

index lies within the range [0, 1]. The more it tends towards 1, the Rk = cov (Yk ; F(Y)) / cov (Yk ; F(Yk))

more unequal is the distribution of expenditures. On the other hand,

(Equation 5)

when the indicator declines, the distribution of expenditure becomes - the Gini coefficient related to the component k: G k = 2cov

more egalitarian.

28 A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

a

n

k

Economic Brief 2 0 1 4

(Yk ; F(Yk) / Yk

(Equation 6)



AfDB

w w w . a f d b . o r g

living, ranked from the poorest to the richest (horizontal axis). In other words, it links the share of health expenditures to the quintiles of total expenditures (or other variable indicating the standard of living).28

- the proportion of the component k in the total expenditure :  k /Y Sk = Y

(Equation 7)

Then G = ∑ Rk Gk Sk

(Equation 8)

Figure 32: Concentration curve for health spending in 2010

Cumulative distribution of Heath expenditure

The relative contribution of an expenditure source k to the total inequality is: Pk = Rk Gk Sk / G.

(Equation 9)

The sum of the relative contributions of various sources (components) is equal to the unit. The breakdown of the Gini index also helps to determine the marginal effect of variation in each expenditure category k on total expenditure inequality. Let ek be a scalar slightly superior

Population in order of increasing Heath expenditure

to the unit, an increase in expenditures derived from the source k results in the passage to vector ekYk and will involve a variation of

2.2.

G. The variation in the value of G brought about by this change at

Concentration index

the margin of the expenditure category is obtained from the partial The concentration index (Kakwani, 1980), which derives directly from

derivative of G in relation to ek. We show that:

the concentration curve, quantifies the degree of socioeconomic ∂G/ ∂ek = Sk (Rk Gk - G)

(Equation 10)

inequality related to the variable being examined (health expenditure). The concentration index is equal to twice the area between the

∂G/ ∂ek ek is the marginal contribution of source k to total inequality.

concentration curve and the first bisector. If there is no socioeconomic

The relative marginal effect is obtained by dividing the above expression by

and the concentration index is equal to zero. By convention, the

G, that is to say:

concentration index is negative when the concentration curve lies

inequality, the concentration curve is confounded with the bisector

above the first bisector. In this case, health expenditures are highly (∂G/ ∂ek)/ G = ( Rk Gk Sk / G ) - Sk

(Equation 11)

concentrated among the poor. Conversely, the concentration index is positive when the concentration curve lies below the first bisector.

It is clear that the sum of the relative marginal effects is nil, multiplying

Consequently, it is important not to focus on the inequality of health

all the sources of income by e leaves Gini’s global index unchanged.

expenditures as reflected by the Gini index for these expenditures,

2.

Concentration curve and index

the concentration index.

2.1.

Concentration Curve

2.3. Breakdown of the concentration index by expenditure categories

but instead on the inequality of health expenditures as revealed by

The concentration curve (Figure 32) represents the cumulative percentage of health expenditures (vertical axis) associated with the

According to Kakwani (1980), the expenditure concentration index

percentage of the population classified by increasing standards of

X is:

28

The Gini curve appears as a special case of the concentration curve. For the Lorenz curve, the vertical axis and the horizontal axis refer to the same variable. For the concentration curve, the variables are different.

29 A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

a

n

k

AfDB 2 0 1 4



Economic Brief w w w . a f d b . o r g

C = 2 cov (X ; F(Y)) / Y 

3.

(Equation 12)

Kakwani index

X being the health spending vector and Y the vector of total

The comparison of the health expenditures concentration curve with

expenditure: (Y1, Y2, .... Yn) and F (Y) representing the cumulated distribution considered as a random variable evenly distributed

the Gini curve of total expenditures gives an indication of the progressivity of the variables being studied, which, in our case, are health expenditures.

between [0 and 1].

If health expenditures are proportional to the standard of living, then the

The Gini index appears as a special case of the concentration index

and the concentration curve of health expenditures is confounded with

where X=Y.

that of living standards (or Lorenz curve). When the poor spend

inequality in health expenditures is similar to that of living standards,

proportionately less on their health, their share of health spending is That is, X1,X2,...,Xk the expenditure components with X = ∑ Xk

(Equation 13)

lower than their share in the total expenditures. In this case, the Lorenz curve dominates (or is above) the concentration curve. The opposite

The concentration index C may be written as follows (O’Donnell et al.,

phenomenon is observed when the system is regressive.

2008): The Kakwani index is equal to twice the area between the Gini curve C = ∑ Sk Ck

(Equation 14)

and the concentration curve. It is equal to the difference between the health spending concentration index and the Gini index of total

Where Sk = Xk /X represents the specific budget coefficient relating to expenditure item k or the expenditure elasticity Xk in relation to total

expenditures.

expenditure X.

I=C–G

Ck is the concentration index specific to the category k

The value of I lies between -2 and 1. A negative value means that health spending is regressive; the Gini curve is below the concentration

The contribution of category k to total inequality is:

curve. A positive value implies the progressivity of health spending; the Gini curve is above the concentration curve. There is uncertainty

Pk = SkCk/C

(Equation 15)

when the curves intersect. In which case, it becomes necessary, in addition to the graphical analysis, to use the Kakwani index to

This leads to: ∑ Sk Ck /C = 1

distinguish between these cases.

Figure 33: Concentration curves for the health SPY and total SPY in 2010

The sum of the relative contributions of various sources (components) is equal to the unit. The breakdown of the concentration index also helps to determine the contribution of variation in each expenditure category k to total expenditure inequality. Thus, the breakdown of the Gini index as proposed by Lerman and Yitshaki (1985) and that of the concentration index as presented by Kakwani (2000) makes it possible to measure the contribution of an expenditure category to the total inequality. The breakdown also helps to gauge the impact of a marginal increase in a particular expenditure category on total inequality. We will use these breakdown techniques to analyse how the various items of health expenditure have contributed to the inequality in private expenditures on health in Tunisia.

30 A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

a

n

k

Economic Brief 2 0 1 4

4. Trend in and sources of inequality of health SPY: 2000-2005-2010 4.1.



AfDB

w w w . a f d b . o r g

Table 8: Trend in the concentration index and Kakwani index

Overall inequality trends

Gini indices show that inequalities in health expenditures are much

Years

2000

2005

2010

C

0.444437

0.448276

0.410342

G

0.408558

0.414008

0.384674

I

0.035879

0.034268

0.025668

greater than inequalities in total household expenditures (Table 7). In 2010, the Gini curve of health spending revealed a high concentration of health expenditures (Figure 31). Twenty per cent (20%) of the population accounts for over 75% of health spending. The high levels of inequality stem from unequal health status of the population, given that health expenditures are closely linked to the need for care, which is expressed only by those in poor health.

The concentration index shows that inequality in health spending increased between 2000 and 2005, but dropped sharply between

Table 7: Trend in Gini index for total SPY and the health SPY29

2005 and 2010 (Table 5). Despite this decline, inequality in health spending continues to be higher than inequality in total expenditures

Years

2000

2005

2010

(Figure 33).

Total SPY

0.408558

0.414008

0.384674

The Kakwani index is positive. Health expenditures are progressive

Health SPY santé

0.744939

0.715508

0.711574

(Figure 34). The Kakwani index declined between 2000 and 2010, the progressivity of health spending was constrained.

Therefore, it is important to consider not the inequality in health

The scope of the inequalities in health care spending explains why

spending as reflected by the Gini index for these expenses, but the

they should be paid close attention and the need to identify ways and

inequality in health spending as shown by the concentration index.

means of reducing these inequalities.

Figure 34: Budget coefficient of health spending by decile of total expenditures in 2010

7% 6% 5% 4% 3% 2% 1% 0% d1

d2

d3

d4

d5

d6

d7

d8

d9

29

d10

The Gini index for total SPY 2000 is equal to that published in the report of the 2000 consumption survey (p.27). The 2000 and 2005 indices differ from those published in 2012 in the poverty report (p. 23). However, the 2010 index does not differ from the latter.

31 A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

a

n

k

AfDB 2 0 1 4



Economic Brief w w w . a f d b . o r g

4.2. Contribution of expenditure items to total inequality in health spending

special radiology services (scans and MRI), child delivery, medical treatment abroad, long-term care (long-term consultations and medication) and cosmetic surgery (since 2010);

In the national budget and household consumption surveys of 2000, 2005 and 2010, health expenditures are classified into four categories:



Pharmaceuticals (drugs) and such other pharmaceuticals as baby products (talc, soap, etc.), adhesive bandages, etc.;



Routine medical care which includes medical consultations, dental care, paramedical services (x-rays, analyses and nursing



Medical devices, including optical glasses, blood pressure

services) in both public and private institutions. This item also

measuring devices, hearing aids, etc.

includes the use of traditional medicine (healers and medicinal plants);

Therefore, one may wonder to what extent each of these items

Special medical care which includes stays in the hospital

question, we will proceed to break down the health inequalities by

or clinic, surgical procedures, special dental surgery procedures,

category (Table 9).

contributed to changes in health spending inequalities. To answer this •

32 A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

a

n

k

Economic Brief 2 0 1 4



AfDB

w w w . a f d b . o r g

Table 9: Breakdown of health spending inequality by expenditure category

Year

2000

2005

2010

Item

Yk in D

Gini Index

Sk

Concentration Index

Kakwani Index

Gk

RkSkGk/G

Marginal effect

Gk

SkCk/C

Ck-G

Routine medical care

15.471

21.8%

0.877%

22.6%

0.008

0.471

23.14%

0.062442

Special medical care

27.974

39.4%

0.872

40.6%

0.012

0.456

40.78%

0.050442

Pharmaceuticals

26.351

37.2%

0.799

35.3%

-0.019

0.408

34.10%

-0.000558

Medical equipment

1.130

1.6%

0.984

1v5%

-0.001

0.550

1.97%

0.141442

Total

70.927

100%

0.745

100.0%

0.0000

0.444

100.0%

0.035442

Routine medical care

30.287

27.1%

0.846

28.6%

0.015

0.486

29.45%

0.071992

Special medical care

32.054

28.7%

0.888

30.4%

0.017

0.453

29.01%

0.038992

Pharmaceuticals

47.510

42.6%

0.740

39.5%

-0.031

0.417

39.57%

0.002992

Medical equipment

1.742

1.6%

0.977

1.5%

-0.001

0.565

1.97%

0.150992

Total

111.672

100%

0.716

100.0%

0.0000

0.448

100.0%

0.033992

Routine medical care

39.062

27.1%

0.843

28.6%

0.015

0.433

26.04%

0.048326

Special medical care

42.067

29.2%

0.903

32.1%

0.029

0.493

39.99%

0.108326

Pharmaceuticals

60.790

42.2%

0.724

37.8%

-0.044

0.331

31.91%

-0.053674

Medical equipment

2.210

1.5%

0.986

1.5%

-0.000

0.625

2.06%

0.240326

Total

144.251

100%

0.730

100.0%

0.0000

0.410

100.0%

0.025326

33 A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

a

n

k

AfDB 2 0 1 4



Economic Brief w w w . a f d b . o r g

Inequality of the SPY for routine medical care (G = 0.843 and C =

were progressive as evidenced by the Kakwani index and budgetary

0.433 in 2010) was higher than the overall inequality in health

coefficients by decile of total SPY (Figure 35).

expenditures (G = 0.730 and C = 0.410 in 2010). However, on account of their average share in these expenditures (the Sk stood at

Of all health expenditures, pharmaceutical expenditures were the most

27.1% in 2010), their relative contribution to inequality was average

substantial and the least unequal component (Table 6). In 2010, they

(28.6% in 2010). From 2000 to 2010, the relative marginal effect of

accounted for 42.2% of total health expenditures, while their contribution

spending on routine medical care was positive throughout the period,

to the formation of inequality stood at 37.8%, according to the Gini index,

meaning that this source of inequality instead had an inertia effect on

and at 42.2%, according to the concentration index. From 2000 to 2010,

inequalities in health spending. The Kakwani index was positive and

the Gini index specific to pharmaceutical expenditures declined from

routine health care expenditures were progressive. Indeed, in 2010,

0.799 to 0.724.30 The marginal effect of the growth of these expenditures

the budget coefficient of these expenses decreased for the 9th and

on inequality was negative throughout the period analysed, reflecting an equalizing impact. In 2010, pharmaceutical expenditures were regressive

10 deciles (Figure 35).

from the fifth decile (Figure 35) and the Kakwani index was negative. The inequality of the SPY for special medical care was on the uptrend

Expenditures on medical equipment were the smallest component

(GK2000 = 0.872; Gk2010 = 0.903 and = 0.459 CK2000; Ck2010 =

of health expenditures (Table 6), yet they were the most unequal

0.493). Throughout the period, their relative marginal effect remained

(Gk201=0.986 and Ck2010=0.625). In 2010, they accounted for

positive. This development shows that spending on special medical

1.5%, while their contribution to the formation of inequality stood at

care is a source of inequality and its increase helped to worsen the

1.5%, according to the Gini index. These expenditures were highly

inequality in overall care expenditures, given that these expenditures

progressive and had the highest Kakwani index.

Figure 35: Budget coefficients of health spending by decile of total SPY, by category, in 201031

0.03 0.025 0.02 0.015 0.01 0.005 0

Rou ne medical care Special medical care Pharmaceu cals

d2

30 31

d3

d4

d5

d6

d7

d8

Medical equipement

d9 d10

The concentration index went from 0.408 to 0.331 Our calculations from the pooled budget data on the 2010 survey published on site of the National Institute of Statistics (INS) http://www.ins.nat.tn/indexfr.php

34 A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

a

n

k

Economic Brief 2 0 1 4

4.3. Contribution of routine medical care to the total inequality



AfDB

w w w . a f d b . o r g

(G2010 = 0.993 and C2010 = 0.600). This situation reflects the difficulty that poor people have in accessing health care other than the one provided free of charge. The often exorbitant cost of these procedures and care

Routine care expenses include medical consultations, dental care,

would lead to high elasticity and discourage low-income patients from

radiology and biological procedures in both public and private sector

using the services.32 In actual fact, the expenses entailed amount to a

facilities, and the use of traditional medicine (Table 7).

high proportion of their income. The budget coefficient for these

Between 2000 and 2010, the decline in inequalities in routine medical

index is very high, compared to other items of expenditure. There is

care expenses was driven by the decline in inequalities in medical

therefore considerable elbow room for improving the access of poor

consultations (Gk2000 = 0.877; Gk2010 = 0.849 and Gk2010 = 0.443; Ck2010

people to such care.

expenses increases exponentially with income (Figure 36). The Kakwani

= 0.433). In 2010, the specific budget coefficient of these expenditures decreased for the 9th and 10th deciles (Figure 36). The Kakwani index

The trends are similar to those recounted in the literature.

was positive but very small.

Socioeconomic status is identified as a key determinant of health expenditure. Similarly, people of higher socio-economic status generally

The decline in inequalities in spending on medical consultations may be

use more health care services, especially the specialized services, than

attributed to the greater availability of physicians (reflected in the

other people (Allin et al. 2006). In contrast, when people of lower socio-

reduction in the ratio of the number of inhabitants per physicians). The

economic status are not entitled to free care, they are more likely, than

decline is observable despite an increase in inequality in the availability

other people, to forgo health expenditures.

of physicians nationwide. This suggests that when patients do not have easy access to medical services and physicians, they tend to move

For De Looper and Lafortune (2009), despite having higher rates of

towards the latter.

illness, disease and death, poorer or less educated persons often have

Spending on radiological and biological procedures continued to be

services. They make less use of these goods and services, some of

highly unequal (G2010 = 0.947 and C2010 = 0.464). The same is true for

which are very expensive compared to their income. We would expect

dental care, which is generally not considered a priority by the poor

the same to be true for special medical expenses.

difficulties in accessing appropriate specialists and preventive health

Figure 36: Budget coefficients of routine health spending by deciles of total SPY, and by sub-item, in 201033

0.015 Medical consulta ons

0.01

Dental

0.005

X-Rays, Scans and Medical analyses

0 d1 d2 d3 d4 d5 d6 d7 d8 d9 d10

32 33

For these types of care, there is balance and the demand nil. Our calculations from the pooled budget data on the 2010 survey published on site of the National Institute of Statistics (INS) http://www.ins.nat.tn/indexfr.php

35 A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

a

n

k

AfDB 2 0 1 4



Economic Brief w w w . a f d b . o r g

Table 10: Breakdown of inequality in routine health spending by expenditure sub-category

Year

2000

2005

2010

Item

Yk in D

Gini Index

Sk

Concentration Index

Kakwani Index

Gk

RkSkGk/G

Marginal Effect

Gk

SkCk/C

Ck-G

Medical consultations

11.571

16.3%

0.877

16.3%

0.0000

0.443

16.26%

0.03444

Dental care

0.806

1.14%

0.996

1.3%

0.0016

0.607

1.55%

0.19844

X-rays, scans and medical analyses

3.037

4.28%

0.98

4.9%

-0.0062

0.544

5.24%

0.13544

Traditional medicine

0.056

0.08%

0.99

0.08%

-0.0000

0.525

0.09%

0.11644

ROUNTINE MEDICAL CARE

15.471

21.8%

0.877

22.6%

0.008

0.471

23.14%

0.06244

Medical consultations

21.177

19.0%

0.849

19.5%

0.005

0.600

19.66%

0.18599

Dental care

1.506

1.3%

0.995

1.6%

0.003

0.464

2.02%

0.04999

X-rays, scans and medical analyses

7.515

6.7%

0.954

7.4%

0.007

0.822

7.66%

0.40799

Traditional medicine

0.089

0.08%

0.999

0.09%

0.000

0.578

0.10%

0.16399

ROUNTINE MEDICAL CARE

30.287

27.1%

0.846

28.6%

0.015

0.486

29.45%

0.07199

Medical consultations

25.749

17.9%

0.849

18.2%

0.003

0.406

16.08%

0.02133

Dental care

1.640

1.1%

0.993

1.2%

0.001

0.600

1.65%

0.21533

X-rays, scans and medical analyses

11.515

8.0%

0.947

8.9%

0.009

0.464

8.23%

0.07933

Traditional medicine

0.157

0.1%

0.999

0.1%

0.000

0.822

0.07%

0.43733

ROUNTINE MEDICAL CARE

39.062

27.1%

0.843

28.6%

0.015

0.433

26.04%

0.04833

36 A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

a

n

k

Economic Brief 2 0 1 4

4.4.

Contribution of special medical expenditures



AfDB

w w w . a f d b . o r g

Increasing inequalities in special health care services are observed in all expenditure items with the exception of special expenditures for

Les dépenses de soins exceptionnels regroupent les maladies de

radiological services (MRI and scans). Whenever radiological services

longue durée, les dépenses de séjour et chirurgie médicale, les soins

are called for, the special expenditures for the services are most often

dentaires exceptionnels, les dépenses exceptionnelles de radiologie et

waived in hospitals or covered by health insurance, and the patients

les accouchements (Tableau 11).

themselves seldom have to make any payments.

Table 11: Breakdown of inequality in special care spending by expenditure sub-category

Year

2000

2005

Item

Yk in D

Concentration index

Gini index

Sk

Kakwani index

Gk

RkSkGk/G

Marginal effects

Gk

SkCk/C

Ck-G

Hospital stay and medical surgery

8.703

12.3%

0.94

13.5%

0.0123

0.509

14.05%

0.1004

Special dental care

0.710

1.0%

0.99

11.7%

0.0017

0.768

1.73%

0.3594

Special radiology expenses

0.671

0.9%

0.99

1%

0.0008

0.503

1.07%

0.0944

Child Delivery

1.110

1.6%

0.98

1.4%

-0.0017

0.341

1.20%

-0.0676

Treatment abroad

0.215

0.3%

0.99

0.4%

0.0008

0.769

0.53%

0.3604

Long-term illnesses

16.565

23.4%

0.90

23.1%

-0.0024

0.423

22.20%

0.0144

SPECIAL MEDICAL CARE

27.974

39.4%

0.872

40.6%

0.012

0.459

40.78%

0.0504

Hospital stay and medical surgery

11.244

10.1%

0.96

11.3%

0.0125

0.470

10.56%

0.0560

Special dental care

0.837

0.7%

0.99

0.7%

0.00015

0.575

0.96%

0.1610

Special radiology expenses

1.474

1.3%

0.98

1.4%

0.0005

0.430

1.27%

0.0160

Child Delivery

1.793

1.6%

0.98

1.7%

0.00045

0.448

1.61%

0.0340

Treatment abroad

0.034

0.03%

0.99

0.04%

0.0001

0.750

0.05%

0.3360

Long-term illnesses

16.673

14.9%

0.93

15.2%

0.0034

0.437

14.57%

0.0230

SPECIAL MEDICAL CARE

32.054

28.7%

0.888

30.4%

0.017

0.453

29.01%

0.0390

Hospital stay and medical surgery

12.472

8.6%

0.977

10.3%

0.0174

0.515

12.42%

0.1303

Special dental care

1.156

0.8%

0.996

0.8%

0.0006

0.665

1.42%

0.2803

Special radiology expenses

2.101

1.5%

0.981

1.5%

0.0004

0.396

1.55%

0.0113

Child Delivery

1.870

1.3%

0.987

1.3%

0.0004

0.444

0.98%

0.0593

Treatment abroad

0.000

0.0%

0.999

0.0%

-0.000

0.392

0.00%

0.0073

Long-term illnesses

24.464

17.0%

0.933

18.0%

-0.0101

0.485

23.61%

0.1003

SPECIAL MEDICAL CARE

0.009

0.0%

0.999

0.0%

0.000

0.411

12.42%

0.0263

SOINS MEDICAUX EXCEPTIONNELS

42.067

29.2%

0.903

32.1%

0.029

0.493

39.99%

0.1083

2010

37 A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

a

n

k

AfDB 2 0 1 4



Economic Brief w w w . a f d b . o r g

Long-term illnesses had the most substantial marginal effect, accounting

coefficients were strongly on the uptrend, reflecting the magnitude of the

for 17% of total health expenditures in 2010, including expenses

inequalities (Figure 37). The highest Kakwani index observed was for

for hospital stay and medical surgery. The corresponding budget

special dental care and treatment abroad in 2000 and 2005.

Figure 37: Budget coefficients of special health spending by deciles of total SPY and by sub-item, in 201034

0.014 0.012 0.01 0.008 0.006 0.004 0.002 0

Hospital Stay and Surgery Special Dental Care Special Radiology Expenditures Child Delivery Long-term Illnesses

d1 d2 d3 d4 d5 d6 d7 d8 d9 d10

This trend reflects the demographic and epidemiological transitions35

inequality in pharmaceutical expenditures and the weight of these

experienced by the country. The health indicators in Tunisia show

expenditures on the poorest populations, although they are entitled to

an overall decline in the incidence of infectious diseases and a

free or virtually free care and medicines.37

simultaneous increase in the incidence of non-communicable diseases The Kakwani indices were negative or extremely low.

and chronic diseases among aging populations. Analysis of mortality data36 shows that the main causes of death are diseases of the circulatory system, metabolic diseases and cancer. These diseases may

The decline in inequality in pharmaceutical expenditures was due to the

be avoided through primary (healthier lifestyles) or secondary prevention

decrease in inequality relating to these two types of expenditures. Their

(screening, early diagnosis). Their prevention and, above all, treatment

contribution to inequality diminished and their marginal effect was

call for increased spending by households and the community.

negative, pointing to an equalizing effect.

4.5.

Contribution of pharmaceutical expenditures

The decline in inequality of pharmaceutical expenditures may be attributed to the increasingly equal availability of pharmacies nationwide.

The pharmaceutical expenditures include expenses relating to purchase

Faced with the difficulty of accessing physicians, patients - especially

of drugs and those relating to the purchase of pharmaceutical products

the poorer ones and those of the middle class - often resign themselves

(Table 9). In 2010, the corresponding budget coefficients were on the

to self-medication. This is also true for patients who live in areas that are

downtrend depending on the deciles (Figure 38), reflecting the low

under-served in terms of health infrastructure and doctors.

34

Our calculations from the pooled budget data on the 2010 survey published on site of the National Institute of Statistics (INS) http://www.ins.nat.tn/indexfr.php The infant mortality rate dropped from 51.4 ‰ in 1984 to 16 ‰ in 2011. Life expectancy at birth was 74.9 years in 2011 (source: National Institute of Statistics - INS). 36 National Statistics on Medical Causes of Death. Tunis 2009). Research Unit on Aging and Medical Causes of Death – National Public Health Institute. 37 It would appear patients entitled to free drugs are sometimes forced to buy their drugs from private pharmacies due to shortages in hospital pharmacies or Primary Health Centres (PHCs). These shortages are caused by governance problems and would be resolved. 35

38 A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

a

n

k

AfDB

Economic Brief 2 0 1 4



w w w . a f d b . o r g

Figure 38: Budgetary coefficients for pharmaceuticals by deciles of total SPY, and by sub-items, in 201038

0.025 0.02 Drugs

0.015 0.01

Other Pharmaceu cals

0.005 0 d1 d2 d3 d4 d5 d6 d7 d8 d9 d10

Table 12: Breakdown of inequality in spending on pharmaceuticals by expenditure sub-category Concentration Index

Gini Index Year

2000

2005

2010

38

Item

Yk in D

Sk

Kakwani Index

Gk

RkSkGk/G

Marginal Effect

Gk

SkCk/C

Ck-G

Medicines

24.448

34.4%

0.813

33.1%

-0.013304

0.404

31.31%

-0.004558

Other pharmaceuticals

1.904

2.7%

0.963

2.1%

-0.005545

0.462

2.79%

0.053442

PHARMACEUTICALS

26.351

37.2%

0.799

35.3%

-0.019

0.408

34.10%

-0.000558

Medicines

39.783

35.6%

0.782

34.7%

-0.008982

0.435

34.59%

0.020992

Other pharmaceuticals

7.728

6.9%

0.857

4.7%

-0.02157

0.322

4.98%

-0.092008

PHARMACEUTICALS

47.510

42.6%

0.740

39.5%

-0.031

0.417

39.57%

0.002992

Medicines

47.795

33.1%

0.792

32.3%

-0.008436

0.366

28.69%

-0.018674

Other pharmaceuticals

12.995

9.1%

0.788

5.5%

-0.035153

0.202

3.23%

0.182674

PHARMACEUTICALS

60.790

42.2%

0.724

37.8%

-0.044

0.331

31.91%

-0.053674

Our calculations from the pooled budget data on the 2010 survey published on site of the National Institute of Statistics (INS) http://www.ins.nat.tn/indexfr.php

39 A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

a

n

k

AfDB 2 0 1 4



Economic Brief w w w . a f d b . o r g

The analysis focused on the trends in the inequality in health spending

spending in 2010. Medical consultations (16.3% of health spending)

in Tunisia and the contribution of the various items of expenditure to the

and medical devices (1.5% of health spending) also had equalizer

formation of this inequality. The analysis is based on data on total health

effects. These changes may be attributed to the increased availability of

spending per person per year (SPY), as established by the national

physicians and the improved national coverage in terms of pharmacies.

budget and household consumption surveys for 2000, 2005 and 2010. The items where inequality has worsened and produced an inertia effect The trends and sources of inequality are captured mainly through the

were long-term illness (17% of spending), the spending on hospital stay

calculation and breakdown of the Gini index and the concentration index

and medical surgery (8.6%) and paramedical or x-ray, scanner and

by expenditure category. This approach is likely to furnish us with

biology services (8% of health spending). These expenses are related to

information on:

the demographic and epidemiological transition. To reduce the corresponding inequality, government authorities need to adopt specific



The overall inequality in health spending and its trends;

policies targeting the most vulnerable groups among the young and the



the inequality of the SPY in each health care expenditure item

elderly.

and sub-item; •

the contribution of the inequality of each SPY to the total inequality;

Dental care continues to be characterized by unusually high levels of



the marginal effect (equalizer or non-equalizer) of the variation of

inequality and lack of access suffered by disadvantaged classes.

a particular SPY on the total inequality in health expenditure.

Improved coverage of the territory by dental practices and increased public awareness of the importance of oral health should curb one of the

The outcomes showed that overall inequality declined from 2000

causes of this inequality. Similarly, special processing of their dental

to 2010. The breakdown of inequality indicators reveals that this

expenditure refund by health insurance, with no competition with

trend was almost exclusively due to a decrease in inequality in

recurrent expenditures, should contribute to reducing inequalities in

pharmaceutical expenditures, which accounted for 42.2% of health

dental care access.

40 A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

a

n

k

Economic Brief 2 0 1 4



AfDB

w w w . a f d b . o r g

5. General Conclusion

D

2- On the demand side:

espite the progress achieved, health inequalities remain considerable and relatively little known in Tunisia. In light of the

analysis conducted, there is significant elbow room for reducing these

2-1- It is important to reduce financial barriers to health care access by

inequalities.

better targeting the poor who benefit from free medical assistance.

1- From the supply side:

2-2- Pharmaceuticals are a significant drain on the budgets of the poorest households and it is necessary to reduce this weight by

1-1- In the public sector, it is necessary to revitalize primary health

ensuring good governance of public pharmacies.

care by improving the operation 2-2- There is a need to ensure a better collective coverage of long1-2- It is also important to strengthen Level II which seems

term illness, hospital stay and medical surgery, x-rays and scans.

to be the weak link in the system. Better coverage of the territory in

Knowing the profile of households that incur these expenses will make

terms of Level II beds should necessarily go hand-in-hand with the

it possible to better target them, if need be.

provision of more specialized physicians for the poorest regions in light of the demographic and epidemiological transition.

2-3- Dental care continues to be characterized by extremely high inequalities in expenses. Improved coverage of the territory in terms of

1-3- Efforts should be made to ensure that at each level the system

availability of dental practices and greater public awareness of the

performs its assigned tasks under the best possible conditions.

importance of dental health should curb one of the causes of the

These tasks should be clearly defined. Each hospital institution should

inequality. Similarly, a special processing of reimbursement for dental

have a scheme of work that allows for coherent strategic

expenses by health insurance, apart from the recurrent expenses,

management.

should contribute to reducing inequalities in dental care access.

1-4- The specific incentives that were introduced to encourage

3- On the institutional side:

physicians to settle in deserted areas should be evaluated. Publicpublic and possibly private-public partnerships should be instituted.

3-1- It is necessary to aim at reducing social and regional inequalities

Also, it is important to negotiate with corporations an institutional

in health care.

framework to better regulate the opening of private practices. 3-2- There is a need to produce and monitor indicators for assessing the It is necessary to determine measures that should be

progress of specific categories not only at the national level but also at the

implemented to enhance health care delivery at local or regional level,

local level. It is important to conduct periodic surveys on the status of health,

as part of an overall regional development policy.

health care use, or the failure to seek health care for financial reasons.

1-5-

41 A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

a

n

k

AfDB 2 0 1 4



Economic Brief w w w . a f d b . o r g

Bibliography



Aïach P. (2000), « De la mesure des inégalités : enjeux sociopolitiques et théoriques » (Measurement of inequality: Socio-Political and Theoretical Challenges), in Leclerc A et al., Les inégalités Sociales de Santé (Social Inequalities in Health), La Découverte/INSERM.



Aïach P., Fassin D. (2004), « L’origine et les fondements des inégalités sociales de santé » (The origins and foundations of social inequalities in health), Article in French published in La Revue du Praticien, 54.



Allin S., Masseria C., Mossialos E., (2006), “Inequality in health care use among older people in the United Kingdom: an analysis of panel data”. Working Paper No. 1/2006. http://eprints.lse.ac.uk/19262/1/LSEHWP1.pdf

• •

Arfa Ch. and Elgazzar H. (2013), Consolidation and Transparency: Transforming Tunisia’s Health care for the Poor. UNICO Studies Series 4. World Bank. Arfa Ch. and Achouri H. (2008), “Tunisia: Good Practice in Expanding Health care Coverage: Lessons from Reforms in a Country in Transition”, in World Bank, Good practices in health financing lessons from reforms in low– and middle–income countries.



Arfa Ch., Souiden A., Achour N. (2007), National Health Accounts in Tunisia: Resullts for Years 2004 and 2005, Ministry Public Health.



World Bank (2008) Diagnostic Study. Performance of Public Health Institutions in Tunisia. http://documents.banquemondiale.org/curated/fr/2008/10/16262037/etude-diagnostique-performance-des-%C3%A9tablissementspublics-de-sant%C3%A9-en-tunisie



Ben Gobrane H.L et al. (2012), « Motifs du recours aux services d’urgence des principaux hôpitaux du Grand Tunis ». (Reasons for using emergency departments of major hospitals in Greater Tunis), Eastern Mediterranean Health Journal, vol. 18.



Bouchoucha I. and Vallin J. (2007), "Regional inequalities in infant mortality in Tunisia”. The 5th African Population Conference. http://uaps2007.princeton.edu/papers/70240



De Looper M. Lafortune, G. (2009), “Measuring disparities in Health Status and in Access and Use of Health care in OECD Countries”. DELSA/HEA/WD/HWP (2009)2. http://search.oecd.org/officialdocuments/displaydocumentpdf/?cote=DELSA/HEA/WD/HWP(2009)2&doclanguage=en



Fleurbaey M., Schokkaert E. (2011), “Equity in Health and Health Care” CORE DISCUSSION PAPER 2011/26 http://www.ecore.be/DPs/dp_1309869083.pdf



Gastineau, B. (2003), "Social and Economic Patterning of Health among Women / Les facteurs sociaux et économiques de la santé des femmes," in Arber S. and Khlat M.(eds), Paris, CICRED. http://www.cicred.org/Eng/Publications/Books/TunisHealthWomen/TunisGastineau.pdf



Hajem S et al (2011), « Statistique Nationale sur les Causes médicales de Décès -2009 » Institut National de Santé Publique. (National Statistics on Medical Causes of Death – 2009)" Research Unit on Aging and Medical Causes of Death – National Public Health Institute



Jusot F. (2003), « Inégalités Sociales de Mortalité : Effet de la Pauvreté ou de la Richesse » (Social Inequalities in Mortality: Effect of Poverty or Wealth). http://epee.univ-evry.fr/EPEE/colloques/jusotevry200312021.PDF

42 A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

a

n

k

Economic Brief 2 0 1 4



AfDB

w w w . a f d b . o r g



Kakwani, N. C. (1980), Income Inequality and Poverty: Methods of Estimation and Policy Applications. New York: Oxford University Press.



Ladner J. Bailly L. Pitrou I, Tavolacci M.P (2008), « Les patients auto-référés dans les services hospitaliers d'urgence: motifs de recours et comportements de consommation de soins » (Self-referred patients in hospital emergency departments: reasons for use and patterns of health care consumption), in Pratiques et Organisations de Soins (Health Care Practies and Organization), vol. 39,1.



Lerman, R.I. and Yitshaki S. (1985), “Income Inequality Effects by Income Source, a New Approach and Applications: the United States.” Review of Economics and Statistics, Vol. 61.



Maurey H. (2013), Rapport d’information Sénat n°335, Session ordinaire (Senate information report No. 335, Regular Session), 2012-2013 «Présence médicale sur l’ensemble du territoire» (Nationwide Medical Presence). http://www.senat.fr/rap/r12-335/r12-3351.pdf



Ministry of Regional Development (2011), Livre Blanc (White Paper) http://eeas.europa.eu/delegations/tunisia/documents/more_info/livreblanc_devreg_nov11_fr.pdf



Ministry of Health (2010), Health Map 2010 http://www.santetunisie.tn/msp/images/CSfinale2010.pdf.



National Office for the Family and the Population/UNICEF (2007), Multiple Indicator Cluster Survey. MICS Tunisia.



Haut Conseil de la santé publique (Public Health Council) (2009), Les inégalités sociales de santé : sortir de la fatalité. (Social Inequalities in Health: Overcoming the Fatality). www.hcsp.fr/explore.cgi/hcspr20091112_inegalites.pdf



O’Donnell O. et al. (2008), Analyzing Health Equity Using Household Survey Data. A Guide to Techniques and Their Implementation. WBI Learning Resources Series The Wold Bank



WHO (2010), Country Cooperation Strategy for WHO and Tunisia (2010-2014) www.emro.who.int/docs/CCS_Tunisie_2010_FR_14489.pdf, Consulted on 1-3-2101



Or Zeynep et al. (2009), «Inégalités de recours aux soins en Europe. Quel rôle attribuable aux systèmes de santé ? » (Inequalities in health care use in Europe. What role for health systems?), Revue économique, 2009/2 Vol. 60, p. 521-543. DOI: 10.3917/reco.602.0521.



Potvin L., Moquet M.-J., Jones C. (2010), Réduire les inégalités sociales en santé (Reducing Inequalities in Health). Saint-Denis : INPES, coll. Santé en action, 2010.



Van Doorslaer E, Koolman X. (2004), “Explaining the differences in income-related health inequalities across European countries”, Health Economics, 13, 7.



Van Doorslaer E., Koolman X, Jones A. (2004), “Explaining income-related inequalities in doctor utilisation in Europe”, Health economics 13,7.



Van Doorslaer, E., Koolman X, Jones, Puffer F. (2002). “Equity in the use of physician visits in OECD countries: has equal treatment for equal need been achieved?” In Measuring Up: Improving Health Systems Performance in OECD Countries



Van Ourti T. (2004), “Measuring horizontal inequity in Belgian health care using a Gaussian random effects two part count data model”. Health economics, 13, 7.



Wagstaff A., van Doorslaer E., Watanabe N. (2003), “On decomposing the causes of health sector inequalities with an application to malnutrition inequalities in Vietnam”. Journal of Econometrics 112.



Wagstaff A., van Doorslaer E. (2000), “Income inequality and health: what does the literature tell us?” Annual Review of Public Health, 21. 43 A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

a

n

k

A

f

r

i

8

BEN AROUS

c

a

n

D

e

v

e

l

o

p

m

e

n

t

45

B

a

n

k

1

1

Rank of cluster depending on the criteria

22

37

2

20

28

19

1

1

0.263

0.226

0.25

0.044

0.3

0.551

0.76

Proportion of PHC offering medical consultation 6 days per week

4

20414

10640

13144

8425

15030

14847

18529

22299

Inhabit. /PHC FTE (Full-Time Equivalent)

2

5588

6118

5677

4816

9498

7090

6144

5242

Inhabit. per primary care physician

1

0.2

1.14

0.86

2.33

1.54

0.54

0.52

0.6

Frontline medical laboratory unit per 100000 inhabit.

1

0.2

1.14

0.86

2.52

0.88

0.54

0.52

0.2

Frontline radiology unit per 100000 inhabit.

1

0.7

2.94

1.5

4.66

1.54

1.9

1.39

1.61

Frontline dental chairs 100000 per inhab.

1

3877

5883

5575

5923

5845

7229

5203

6385

Inhab. /daytime pharmacy

3

45468

33989

46550

17769

50656

36870

41250

45273

Inhab. /Nighttime pharmacy

1

6.9

3.6

5.6

2.3

0.8

1.9

3.8

4

Medical Laboratory per 100000 inhab.

1

30.1

18.3

23.9

19.4

27.6

19

24.6

23.7

Haemodialysis machines per 100 000 inhab. (Total, public and private)



4

20414

6180

6006

5368

4035

9218

11786

19920

Inhabit. per Primary Centre (PHC)

2 0 1 4

TUNIS

SOUSSE

SFAX

14

219

MEDENINE

MONASTIR

20

17

22

Average distance from a General Hospital (GH)

MANOUBA

20

14

Average distance from a Regional Hospital (RH)

ARIANA

Governorate

Economic Brief

AfDB

w w w . a f d b . o r g

Annex 1: Indicators of Health care Facilities by Cluster Table 13: Health infrastructure indicators: Cluster 1.1

A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

Ranking

NABEUL

KEBILI

BIZERTE

2

24

40

23

3

84

299

86

3

6022

1784

6073

Inhabit. per Primary Centre (PHC)

3

14%

9%

14%

Proportion of PHC offering medical consultation 6 days per week

2

14858

4687

14197

Inhabit. /PHC FTE (Full-Time Equivalent)

3

7682

4432

7699

Inhabit. per primary care physician

2

1.46

1.99

2.01

Frontline medical laboratory unit per 100000 inhabit.

2

1.46

1.99

2.01

Frontline radiology unit per 100000 inhabit.

3

2.79

1.99

2.74

Frontline dental chairs 100000 per inhab.

2

5791

10047

6833

1

31367

30140

36440

Inhab. Inhab. /daytime /Nightpharmacy time pharmacy

2

2.5

1.3

1.8

Medical Laboratory per 100000 inhab.

3

18.5

19.2

24.2

Haemodialysis machines per 100 000 inhab. (Total, public and private)



Average distance from a General Hospital (GH)

2 0 1 4

Governorate Average distance from a Regional Hospital (RH)

AfDB Economic Brief

w w w . a f d b . o r g

Table 14: Health infrastructure indicators, Cluster 1.2

46

B

a

n

k

A

f

r

i

c

a

n

D

e

v

e

36

44

GAFSA

LE KEF

l

o

p

m

e

n

t

B

a

n

59

42

29

3

TOZEUR

ZAGHOUAN

Ranking

40

TATAOUINE

SILIANA

44

32

GABES

MAHDIA

22

Average distance from a Regional Hospital (RH)

BEJA

Governorate

47

k

4

1

3

20%

6%

5%

18%

8%

19%

20%

25%

10%

Proportion of PHC offering medical consultation 6 days per week

1

7809

8295

8141

8116

7979

7998

9316

9726

11067

Inhab. /PHC FTE (Full-Time Equivalent)

1

4736

3996

4177

4500

4833

5832

6379

6821

5777

Inhab. per primary care physician

4

2.93

5.8

4.1

3.85

2.78

3.51

2.66

2.49

1.96

Frontline medical laboratory unit per 100000 inhab.

4

2.35

3.86

3.42

3.85

2.52

2.73

3.25

2.21

1.96

Frontline radiology unit per 100000 inhab.

4

2.93

4.83

4.1

3.42

2.78

3.51

2.66

3.04

2.61

Frontline dental chairs 100000 per inhab.

3

8974

7962

7310

11700

9007

8848

9138

6694

9279

2

28417

34500

29240

29250

30485

36657

42263

36150

38275

Inhab. Inhab. /daytime /Nightpharmacy time pharmacy

3

1.2

0

0.7

0.4

1

0.8

0.6

1.7

1

Medical Laboratory per 100000 inhab.

2

28.2

30.9

25.3

16.7

19.4

21

24.8

16.3

14.4

Haemodialysis machines per 100 000 inhab. (Total, public and private)



3480

4709

2358

2629

3538

2759

3635

3929

3257

Inhab. per Primary Centre (PHC)

2 0 1 4

69

333

307

143

44

206

226

168

127

Average distance from a General Hospital (GH)

Economic Brief

AfDB

w w w . a f d b . o r g

Table 15: Health infrastructure indicators, Cluster 1.3

A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

a

n

Ranking

SIDI BOUZID

KASSERINE

KAIROUAN

4

44

50

46

36

JENDOUBA

2

157

257

98

157

Average distance from a General Hospital (GH)

2

3327

4238

4305

3712

Inhab. per Primary Centre (PHC)

2

9%

16%

17%

11%

Proportion of PHC offering medical consultation 6 days per week

3

13130

12530

13596

12432

Inhab. /PHC FTE (Full-Time Equivalent)

4

3438

5720

8354

8637

Inhab. per primary care physician

3

2.67

2.78

1.97

1.42

Frontline medical laboratory unit per 100000 inhab.

2

2091

2.08

1.97

1.65

Frontline radiology unit per 100000 inhab.

4

2.91

2.31

1.97

1.89

Frontline dental chairs 100000 per inhab.

4

11458

11376

9995

10076

4

41250

54038

43054

70533

Inhab. Inhab. /daytime /Nightpharmacy time pharmacy

4

0.2

0.2

0.5

0.5

Medical Laboratory per 100000 inhab.

4

13.6

11.6

8.8

15.8

Haemodialysis machines per 100 000 inhab. (Total, public and private)



Average distance from a Regional Hospital (RH)

2 0 1 4

Governorate

AfDB Economic Brief

w w w . a f d b . o r g

Table 16: Health infrastructure indicators, Cluster 1.4

48

k

A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

49

B

a

n

k

Ranking

TUNIS

TOZEUR

1

52.7

30.9

26.6

23.9

MONASTIR

SOUSSE

26

MANOUBA

Governorate Beds per 10 000 inhab (pub + private)

1

40.4

30.9

23.9

22.5

26

Hospital beds / 10 000 inhab.

1

12.3

0

2.8

1.4

0

Clinic beds / 10 000 inhab

1

5.03

4.83

1.63

2.93

0

General surgery beds / 10 000 inhab.

1

15.69

10.03

8.13

8.31

0

Gyneco beds/ 10000 WCBA

1

21.6

11.41

7,28

8,81

0

Paediatrics beds/ 10000 children below 15 years

1

1.96

1.93

0.98

0.66

0

Ophthalmology beds /10000 inhab.

2

0.66

0.97

0.67

0.74

0

1

1.73

0

0.98

0.54

3.25

ENR beds Ortho/10000 paedics inhab. beds per 10000 inhab.

1

2.22

0.97

1.18

1.47

0

Cardiology bed per 10000 Inhab.

1

0.99

0.97

0.75

0.66

0.79

1

0

0

0.49

0.62

16.41

Anaesthe- Psychiatric sia and in- beds/10000 tensive inhab. care beds /10000 inhab.

Economic Brief

2 0 1 4

Table 17: Bed availability indicators, Cluster 2.1 •

AfDB

w w w . a f d b . o r g

A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

Ranking

MEDENINE

MAHDIA

LE KEF

2

20.9

16.5

21.7

3

16.6

14.9

21.2

2

4.3

1.6

0.6

Clinic beds / 10 000 inhab.

3

2.68

1.94

2.73

General surgery beds / 10 000 inhab.

2

7.99

6.97

8.32

Gyneco beds/ 10000 WCBA

3

8.89

4.94

7.04

Paediatrics beds/ 10000 children below 15 years

3

0.72

0.25

0.58

Ophthalmology beds /10000 inhab.

3

0.61

0.5

0.58

3

0.77

0.81

0.58

ENR beds Ortho/10000 paedics inhab. beds per 10000 inhab.

2

0.66

0.86

0.17

Cardiology bed per 10000 Inhab.

3

0.53

0.25

0.31

Anaesthesia and intensive care beds /10000 inhab.

2

0.13

0.63

0.58

Psychiatric beds/10000 inhab.



Hospital beds / 10 000 inhab.

2 0 1 4

Governorate Beds per 10 000 inhab (pub + private)

AfDB Economic Brief

w w w . a f d b . o r g

Table 18: Bed availability indicators, Cluster 2.2

50

B

a

n

k

A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

a

Ranking

TATAOUINE

KEBILI

GAFSA

BEJA

Governorate

3

17.7

19.8

23.2

17.9

Beds per 10 000 inhab (pub + private)

2

17.7

18.5

22

17.9

Hospital beds / 10 000 inhab.

4

0

1.3

1.2

0

Clinic beds / 10 000 inhab.

2

2.05

3.98

2.96

2.45

General surgery beds / 10 000 inhab.

3

6.52

7.11

5.95

9.1

Gyneco beds/ 10000 WCBA

2

8.33

8.4

6.66

6.03

Paediatrics beds/ 10000 children below 15 years

2

0.68

0.53

1.33

0.65

Ophthalmology beds /10000 inhab.

1

0.68

0.8

0.33

0.65

2

0.82

0

0.86

0.98

ENR beds Ortho/10000 paedics inhab. beds per 10000 inhab.

4

0

0

0.8

0

Cardiology bed per 10000 Inhab.

2

0.68

0.53

0.18

0.33

Anaesthesia and i ntensive care beds /10000 inhab.

4

0

0

0

0

Psychiatric beds/10000 inhab.

Economic Brief

2 0 1 4

Table 19: Bed availability indicators, Cluster 2.3

51

n

k •

AfDB

w w w . a f d b . o r g

A

f

r

i

c

a

n

D

e

v

e

l

19.1

15.8

GABES

JENDOUBA

o

p

m

e

n

t

52

B

a

n

k

27.9

ZAGHOUAN

4

16.7

SILIANA

Ranking

10.7

14.9

12.8

SIDI BOUZID

NABEUL

KASSERINE

12.8

17

BIZERTE

KAIROUAN

5.5

BEN AROUS

4

27.9

16.7

10.7

12.7

12.3

12.1

14.3

17

16.5

3.8

8.3

Hospital beds / 10 000 inhab.

3

0

0

0

2.2

0.6

0.6

1.5

2.1

0.5

1.7

4.3

Clinic beds / 10 000 inhab.

4

1.47

1.79

1.45

1.46

1.16

1.14

1.77

1.66

2.47

0.45

0

General surgery beds / 10 000 inhab.

4

5.22

4.84

4.57

6.74

4.89

3.77

3.31

7.91

6.83

2.06

3.89

Gyneco beds/ 10000 WCBA

4

4.49

3.95

3.69

3.94

4.08

3.97

2.58

5.17

8.47

0

0

Paediatrics beds/ 10000 children below 15 years

4

0

0.43

0.39

0.27

0.46

0.36

0.35

0.97

0.37

0

0

Ophthalmology beds /10000 inhab.

4

0

0.43

0.29

0.27

0.46

0.36

0.35

0.55

0.37

0

0

4

0

0

0

0.8

0.65

0.54

0

0

0.55

0

0

ENR beds Ortho/10000 paedics inhab. beds per 10000 inhab.

3

0

0.77

0.48

0.4

0

0.54

0.71

0.41

0.73

0

0.6

Cardiology bed per 10000 Inhab.

4

0

0

0.15

0.27

0.23

0.27

0

0.83

0.11

0.21

0.7

Anaesthesia and intensive care beds /10000 inhab.

3

0

0

0

0

0

0.71

0.57

0

0

0

0

Psychiatric beds/10000 inhab. •

12.6

Beds per 10 000 inhab (pub + private)

2 0 1 4

ARIANA

Governorate

AfDB Economic Brief

w w w . a f d b . o r g

Table 20: Bed availability indicators, Cluster 2.4

A

f

r

41.1

33.8

60.5

100.6

1

MONASTIR

SFAX

SOUSSSE

TUNIS

Ranking

i

c

a

n

D

e

v

e

l

o

p

m

e

n

1

Ranking

53

t

B

a

n

k

17.5

22.3

20.1

19.8

28.1

21.5

1

ARIANA

MONASTIR

SFAX

SOUSSSE

TUNIS

ARIANA

Ranking

Governorate Pharmacists / 100000 inhab. private

10.3

3.3

TUNIS

SOUSSSE

4.5

3.3

MONASTIR

SFAX

2.4

ARIANA

Governorate General surgeons /100000 inhab.

29.9

ARIANA

1

202.7

109.7

68.0

74.9

48.0

1

7.5

12.8

10.5

1

29.0

40.9

30.2

1

25.5

35.8

27.0

25.5

18.0

21.3

Dentists /100000 inhab. private

1

1

11.3

3.9

3.7

2.5

6

1

8.0

7.4

8.3

2.4

16.9

4.8

1

33.5

43.2

35.3

27.8

34.9

26.1

1

391.6

666.5

436.3

315.4

400.9

139.2

Paramedical Dentists /100000 Dentists /100000 inhab. staff/100 000 inhab. public total inhab. public

1

7.6

3.6

4.4

2.1

2.4

1

1

21.8

42.6

10.3

22.3

9.7

24.1

Paramedical staff/100 000 inhab. private

10.3

3.9

4.1

2.7

2

1

1

413.4

709.1

446.6

337.7

410.6

163.3

Paramedical staff/100 000 inhab. total

0.5

1.07

1.61

1.62

0



23.3

31.0

19.7

1

13.6

5.4

5.7

3.3

3.8

1

209.0

93.8

94.6

55.3

67.5

Psychiatrists/ 100000 inhab.

1

158.9

98.1

80.6

69.3

64.9

General pract Specialists/ 100000 inhab. 100000 inhab. (total). (total)

2 0 1 4

3.2

8.7

2.2

8.2

6.6

6.6

7.6

17.4

Pharmacists/ Pharmacists/ 100000 inhab. 100000 inhab. public total

1

6.85

3,78

2,84

2,3

2,83

1

367.9

191.9

175.2

124.6

132.3

Physicians / 100000 inhab. (total)

Cardiologists Anaesthetists/ 100000 inhab. /100000 inhab.

1

165.2

82.2

107.2

49.7

84.3

Physicians/ 100000 inhab. (private)

Surgeons/ 100000 inhab.

1

106.9

44.6

60.5

21.5

49.4

Specialists/ 100000 inhab (private).

Ophthalmologists/ 100000 inhab.

1

58.3

37.6

46.7

28.1

34.9

Physicians/ General pract. 100000 inhab. 100000 inhab. (public (private) sector)

Gyneco-obstetriPodiatrists cians /100000 /10000 women of childbea- children ring age (15-49 years)

1

102.2

49.2

34.2

33.8

18.1

Governorate General pract Specialists/ 100000 inhab. 100000 (public) inhab. (public sector)

Economic Brief

AfDB

w w w . a f d b . o r g

Table 21: Human resource indicators: Cluster 3.1

A

f

r

i

c

a

n

D

36.1

31.7

21.7

22.6

45.2

MAHDIA

MANOUBA

MEDENINE

NABEUL

ZAGHOUAN

e

v

e

l

o

p

m

e

n

t

B

54

a

n

k

1.8

0

3.5

2.7

2.3

2

MANOUBA

MEDENINE

NABEUL

ZAGHOUAN

Ranking

2.5

2

1.6

General surgeons /100000 inhab.

MAHDIA

GABES

BIZERTE

BEN AROUS

Governorate

4

29.0

GABES

Ranking

29.6

BIZERTE

3

62.8

34.1

30.1

51.8

51.7

40.9

45.0

35.3

2

1.25

2.17

1.55

1.04

1.5

1.12

1.82

2.36

2

2.9

4.4

3.1

4.6

3.3

2.8

5.7

5.7

Gyneco-obstetriPodiatrists cians /100000 /10000 women of childbearing children age (15-49 years)

2

17.6

11.6

8.3

20.1

15.6

11.9

15.4

9.5

2

2

2.3

3.7

3.1

1.1

2

2.5

3.3

4.2

2

7.0

27.1

24.6

9.0

11.4

20.2

22.5

30.0

2

25.8

55.0

50.7

36.9

34.6

36.8

48.3

68.2

2

1.8

1.9

2

1.4

1

2.2

2

2.6

2

88.6

89.1

80.7

88.7

86.3

77.7

93.3

103.5

3

1.2

2.1

2

0.5

0.5

1.4

2.4

1.6

2

63.9

50.5

47.8

59.7

59.3

45.6

55.4

64.1

2

0

1.2

0.7

0.3

1

0.8

0.9

0.7

2

1.17

0.53

0

13.29

1.21

0.55

0.18

0.17

Psychiatrists/ 100000 inhab.

2

24.6

38.7

32.9

29.0

27.0

32.1

37.9

39.5

General pract Specialists/ 100000 inhab. 100000 inhab. (total). (total)

Anaesthetists/ 100000 inhab.

Physicians/ 100000 inhab. (total)

Cardiologists /100000 inhab.

Physicians/ 100000 inhab. (private)

Surgeons/ 100000 inhab.

Specialists/ 100000 inhab (private).

Ophthalmologists/ 100000 inhab.

18.8

27.9

26.1

27.9

23.2

16.6

25.8

38.3

Physicians/ General pract. 100000 inhab. 100000 inhab. (public (private) sector) •

25.8

General pract Specialists/ 100000 inhab. 100000 inhab. (public) (public sector)

2 0 1 4

BEN AROUS

Governorate

AfDB Economic Brief

w w w . a f d b . o r g

Table 22: Human resource indicators. Cluster 3.2

A

f

r

i

c

a

n

D

e

v

e

20.7

18.0

12.6

BIZERTE

GABES

MAHDIA

l

o

p

m

e

n

t

B

a

Ranking

BEN AROUS

ZAGHOUAN

NABEUL

MEDENINE

2

18.8

15.8

20.5

21.5

16.5

24.8

BEN AROUS

MANOUBA

Pharmacists / 100000 inhab. private

Governorate

2

2.4

1.8

2.5

2.4

3.8

3.3

2.5

1.6

1.4

2

21.2

17.6

23.0

23.9

20.3

15.9

20.5

22.3

26.1

Pharmacists/ Pharmacists/ 100000 inhab. 100000 inhab. public total

2

13.4

8.2

14.9

12.7

9.0

11.9

11.9

17.6

21.1

Dentists /100000 inhab. private

3

3.7

5.3

4.5

2.9

3.8

3.3

3.6

3.3

2.8

2

17.1

13.5

19.4

15.6

12.7

15.1

15.5

20.9

23.9

4

259.5

261.6

221.7

266.1

272.3

302.0

283.0

294.4

175.2

Dentists /100000 Dentists Paramedical inhab. public /100000 inhab. staff/100 000 inhab. public total

2

9.9

1.2

14.2

5.7

10.8

11.1

10.2

12.4

13.5

Paramedical staff/100 000 inhab. private

Economic Brief

2 0 1 4

Table 22: Human resource indicators. Cluster 3.2

55

n

k



AfDB

w w w . a f d b . o r g

A

f

r

i

c

3

31.7

a

n

D

e

v

e

l

o

p

m

4

1.3

e

n

t

B

a

56

n

k

Ranking

BEJA

3

12.7

Governorate Pharmacists / 100000 inhab. private

Ranking

BEJA

Governorate General surgeons /100000 inhab

Ranking

BEJA 2

44.7

2.0

Pharmacists/ 100000 inhab. public

3

1.42

3

14.7

Pharmacists/ 100000 inhab. total

3

2.9

Gyneco-obstetricians Podiatrists /100000 women of /10000 children childbearing age (15-49 years)

3

13.1

3

2

3

9.8

4

3.9

Dentists Dentists /100000 inhab. /100000 private inhab. public

3

2.6

3

74.1

Physicians / 100000 inhab. (total)

3

13.7

Dentists /100000 inhab. total

2

1.6

2

329.5

Paramedical staff/100 000 inhab. public

4

0

3

4.6

2

334.1

Paramedical Paramédical staff/100 000 /100 000 inhab. private hab. total

4

0.33

3

29.4

Psychiatrists/ 100000 inhab.

3

44.7

General pract Specialists/ 100000 inhab. 100000 (total) inhab. (total)

Cardiologists/ Anaesthetists/ 100000 inhab. 100000 inhab.

3

29.4

Physicians/ 100000 inhab. (private sector)

Surgeons/ 100000 inhab.

3

16.3

Specialists/ 100000 inhab. (private sector).

Ophthalmologists/ 100000 inhab.

3

13.1

Physicians/ General pract. 100000 100000 inhab. inhab. (public (private) sector)



Specialists/ 100000 inhab. (public sector).

2 0 1 4

Governorate General pract./ 100000 inhab. (public)

AfDB Economic Brief

w w w . a f d b . o r g

Table 23: Human resource indicators. Cluster 3.3

A

f

r

i

c

a

n

D

e

v

23.9

24.3

25.9

37.2

32.3

24.7

37.2

45.8

52.2

2

JENDOUBA

KAIROUAN

KASSERINE

KEBILI

LE KEF

SIDI BOUZID

SILIANA

TATAOUINE

TOZEUR

Ranking

e

l

o

p

m

e

B

a

n

0.7

0.7

2.3

0.7

1.3

0.7

1.9

3

KASSERINE

KEBILI

LE KEF

SIDI BOUZID

SILIANA

TATAOUINE

TOZEUR

Ranking

1.8

t

KAIROUAN

1.7

JENDOUBA

n

1.8

57

GAFSA

Gouvernorat Chirurgiens Généralistes /100000 hab

35.5

GAFSA

k

4

0.67

0.43

4

3.9

2.1

1.7

4

1.9

1.4

1.3

1

1.2

1.3

0.7

1.4

0.9

1.8

4

1.9

0.7

0.9

0.2

1.6

2

0.7

0.7

0.9

0.9

4

1

0.7

0.9

0.5

0.4

0.7

0.2

1.1

0.9

1.2

3

0

0

0

0

0

0.7

0.2

0.9

0.2

0.3

3

0.97

0

0

0.5

0.39

0

0.19

0.82

0.71

0.3

Psychiatre/ 100000 hab

4

81.2

64.3

57.3

46.3

64.3

63.0

45.3

57.2

51.5

70.4

Physicians/ 100000 inhab. (total)



0.3

1.7

2.3

0.7

1.6

2.1

2.1

1.5

Anesthésistes/ 100000 hab

4

16.4

8.2

10.3

9.7

18.3

13.3

9.7

18.2

15.6

19.8

Specialists /100000 inhab. (total)

2 0 1 4

0.5

0.97

0.67

0.49

0.82

0.91

1.09

4

64.7

56.1

47.0

36.6

46.0

49.8

35.6

38.9

35.9

50.6

General PR actioners/ 100000 inhab. (total)

Cardialogues/ 100000 hab

4

17.4

16.4

11.5

16.0

24.2

20.6

15.7

23.9

19.4

26.3

Physicians/ 100000 inhab. (private)

Chirurgiens/ 100000 hab

4

4.8

6.2

1.7

4.1

10.5

8.0

6.0

9.3

7.3

11.2

Specialists/ 100000 inhab. (private)

Ophtalm./ 100000 hab.

4

12.6

10.3

9.8

11.9

13.6

12.6

9.7

14.7

12.1

15.1

General practitioners /100000 inhab. (private)

Pédiatres/ 10000 enfants

4

63.8

47.9

45.7

30.3

40.1

42.5

29.6

33.2

32.1

44.1

Physicians/ 100000 inhab. (public)

Gynéco-obstétriciens /100 000 femmes en âge de procréer (15 – 49 ans)

4

11.6

2.1

8.5

5.6

7.8

5.3

3.7

8.9

8.3

8.6

Specialists Gouvernorat General pract./ 100000 100000 inhab. (public) inhab. (public)

Economic Brief

AfDB

w w w . a f d b . o r g

Table 24: Human resource indicators. Cluster 3.4

A

f

r

i

c

a

n

D

e

v

12.3

12.5

KAIROUAN

KASSERINE

e

l

o

p

m

e

B

a

n

Ranking

4

17.4

t

TOZEUR

19.2

10.9

14.0

TATAOUINE

n

15.8

58

SILIANA

SIDI BOUZID

LE KEF

15.9

11.3

JENDOUBA

KEBILI

15.1

GAFSA

4

7.7

4.8

7.3

6.8

6.2

8.6

5.1

9.5

8.0

7.1

Pharmacists / 100000 inhab. public

3

5.8

3.4

5.1

3.2

4.3

4.0

2.8

2.9

2.6

3.8

Pharmacists/ 10 000 inhab. total

4

13.5

8.2

12.4

9.9

10.5

12.6

7.9

12.3

10.6

10.9

Dentists/100000 inhab. private

3

492.8

318.1

287.6

199.8

385.4

382.2

215.1

270.3

273.6

397.5

Dentists/100000 inhab. public

4

4.8

1.4

1.7

2.7

9.0

4.6

0.7

2.0

1.7

5.9

Dentists/100000 inhab. total

3

497.6

319.4

289.3

202.4

394.4

386.9

215.8

272.3

275.3

403.4

Paramedical staff/100 000 inhab. public



Pharmacists/ 100000 inhab. private

2 0 1 4

Gouvernorat

AfDB Economic Brief

w w w . a f d b . o r g

Table 24: Human resource indicators. Cluster 3.4

k

Economic Brief 2 0 1 4



AfDB

w w w . a f d b . o r g

Annex 2: Health Expenditure Nomenclature 2000 in 2005 (source : www.ins.nat.tn )

41 ROUTINE MEDICAL CARE

423 MRI SCAN 4231 Mri scan in a public institution 4232 Mri scan in a private institution 4239 Mri scan in a with no indication

411 MEDICAL CONSULTATIONS 4111 Consultations by public institutions 4112 Consultations by private institutions 4119 Consultation with no indication

424 CHILD DELIVERY 4241 Child delivery in a public institution 4242 Child delivery in a private institution 4249 Child delivery with no indication

412 DENTAL CARE 4121 Dental care in public institutions 4122 Dental care in private institutions 4129 Dental care with no indication

425 MEDICAL TREATMENT ABROAD 426 LONG-TERM CARE 4261 Consultations for long-term illnesses 4262 Drugs for long-term illness 4263 Non-classified expenditure on long-term illnesses

413 X-RAY AND ANALYSES 4131 X-ray and medical analyses in public institutions 4132 X-ray and medical analyses in private institutions 4139 X-ray and medical analyses with no indication

43 PHARMACEUTICALS

416 TRADITIONAL HEALER 4161 Traditional healer

431 DRUGS, NURSING 4321 Baby products 4322 Other pharmaceuticals

42 SPECIAL MEDICAL CARE 432 AUTRES PRODUITS PHARMACEUTIQUES 4321 Produits pour bebe 4322 Autre produit pharmaceutique

421 STAY AND SURGERY 4211 Stay and surgery in a public institution 4212 Stay and surgery in a private institution 4219 Stay and surgery with no indication

44 MEDICAL EQUIPMENT 441 MEDICAL EQUIPMENT 4411 Medical eyeglasses

422 SPECIAL DENTAL SURGERY 4221 Special dental care in a public institution 4222 Special dental care in a private institution 4229 Special dental care with no indications

4412 Hearing aids 4419 Other medical equipment

59 A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

a

n

k

A

f

r

i

c

a

n

D

e

v

e

l

o

p

m

e

n

t

B

a

n

k

© 2010 - AfDB - Design, External Relations and Communication Unit/YAL

Related Documents


More Documents from ""