1997 - The Quality Of Life Of Patients With Paranoid Schizophrenia In London And Berlin

() I{ I ( i I '\ \ I

1\1. Heinze

I' \ J>I R

. R.E. Ta)"lnr . S. Pri{'ht,

G. Thornicroft

Thequalityof life of patientswithparanoidschizophrenia in London andBerlin

Accepted: 23 July 1996

Abstract This study compared the subjective and objective quality of life and needs of patients with pam. noid schizophrenia between inner city areas in Berlin (69 patients) and London (75 patientsl. Quality of life was assessed by means of the Lancashire Quality of Life Profile (German version Berliner Lebensqualitat-

sprofi), and need was quantified using the Camberwell Assessment of Need (German version Berliner Bedurfnisinventar). The hypotheses tested were that although Berlin patients may rate more highly on objective quality of life measures, the subjective quality of life would be similar as patients would judge their quality of life against their local expectations. The findings supported the first part of the hypothesis as on the objective measures the Berlin group was significantly better off financially and in living conditions. and had significantly fewer material needs. However, despite having more severe psychopathology. the Berlin groups' scores on global subjective quality of life were also higher. On particular life domains. subjective quality of life did not always reflect objective measures and sometimes went in the reversedirection. We concluded that the relationship between subjective and objecti,'e quality of life is complex. and great caution must be exercised in making quality of life comparisons between different cultures.

Introduction

Although international research In the ('\'aluaIH)n or community care is performed increasingly often. it is ----.-------. M. Heinze (B I' S. Pneoe

Departmentof SocialPs~chi~1r\. Platanenallee

I'relc

1'111\ l'r
19. D"14()~() Berllll. Gcrlll.1I11

R E. Taylor G. Thornicrof! PRISM. Institute of PSYl"hratfl. Dc Crc'J'I~II\ Denmark HIli. L()ndpn Sf ~ 1\.\I . I' K

I';II~.

Ilcrllll,

not yet clear how far these results are comparable and can be generali7,ed. In particular. it is unclear to what extenl thc populations of paticnts being treated in the l"()lllmunity care systems of different countries are similar This study is an initial attempt to explore these questions by comparing quality of life and needs data of patients with paranoid schizophrenia (lCD 10: F20.0X) [I6J in two metropolitan catchment areas: the Nunhead sector of the Camberwell district of London and the Charlottenburg district of Berlin. Both districts are developing psychiatric community care systems and both have been the subject of several studies [3.4, 7, 11- I 5], making them suitable for a comparative study. Both are metropolitan areas of capital cities with roughly the same population (Camberwell = 210.000: Charlottenburg = 180:(00). Nevertheless. in their so: cial characteristics they differ suhstantially: Camberwell has a high proportion of minority ethnic groups and is 0ne of the most socially deprived areas in Britain. Charlottenburg, on the other hand. is a district with a mixed structure of de\'cloped middle-class areas and some socially deprivcd areas and with fewer ethnic minority inhahitants. A mcthodological difficulty inherent in any attempt to make such a comparison between t o countries is that both the social care and health care systems will differ. making it difficult to attrihute the cause of differences found. OttI' study focused upon the question: are quality of fife and needs of patienls ith paranoid schizophrenia different in the two districts? Our hypothesis was that the nhjective data ab()ut the standard of living should sho significant differences that would not be explained by differences in the psychopathological status uf the dients. Furthermore, we hypothesized that whilst the oojecti\'e data should show substantial differences. Ihe <.uojective data would not differ. The laller hypoIhc<.i<. as ha<;ed <)n Ihe assumplion that patients would compare themsel\'es 10 tlwc;e around them. rather than C\ ;IIuate their life sitllati(ln q ;111 d ;11d I ~. (). J-t ]

according to <;ome0bjecli,'e

.....

;I dl.I!!I1P'"

Method

,,[ 1'.II;lnpl
'dll1"rhrcl1';I

11('1)10

~~OOXI

"~r~

111-

d"ded 1111111<(l'I11I';III','n III ""Il'l .<' IIll'rea'~ the c"mparat>lht~ ,,[ th~ r;ltlent gll'''I'' (hll\ dala [r"11\ fi'e 'Cl'lions "f the Lancashlle 1)II.dll' "f lIre 1'1,,[ile "c.:n'ratlpn. 11\II1g cl'ndllloos. finances. 1.0nd()n ~amrl~ '''nod hie al1d 'cnlllt~ \\Cle used fpr compallson. Fwm Ihe IIc.-ClIrlpfile<. <'11" the rllq 4u~qll'n nf each section (i.e. whether The !\:unh~ad ~ecl()r c>fthe Camner"elllhqlll"l 'en lIT 1\ Ihe 'Un.ll'(1 there "i" a pwhlem al all answered from the users' perspective) was c>fa larger prospective ~tud)' !the PRiSM ~tudyl 0f a rerre~entallve IIlclllded. resullllll! 111a ,n'r~ summll1g up all problems rated. and ~amr1e of ratienl~ with r~ych0~i~ Th~ PRISM ~lIId\ 11,)p~ r~r0flCd ,"1-,,'1'''' \\Cle dClI'l'd f,'r malcllal. sOCIal and rsychopalhololllC;11 In detail in future raper~lls e\amll1ing the 1111r;II"I"fIII11"d Ul'Il1g."" prt'}\lclu" communlly menIal health teiims. l.ne fl'r an.le rallelll~ II'Al E rll'm Iht rCm;lIlIlIl!! data. tht gcneral satisfactlc>n wilh life and team) and the other for continuing care pallel1ts IPACl teaml. The the sum lIected a~ rart oflhe PRISM ~tudy. II1lZ the sall~facllon scales. sh()\, cd a non-normal distributic>n which involved interviewing a ~ample of 20(, psychl'til' pall~nt~ and th~ref(\I~ ,,~r~ l'IImrared u~ing Ihe Mann- Whilney U test. living in South-east London whc> had heen Idenllfied \ ia a case, Th~ calel!ol1cal 'allahles \\ere cC\mpared using the chi-square finding technique. A subsample of 75 patlent~ "as avallahle wllh leSI Scores are rer(.rted as means wilh their respective standard quality of life and needs data. The~e data wer~ t'(\lIect~d at nas~line de\'latl('ns. (hefore the introduction of th~ community mental health leamsl In order to excludc faclors that are known to inRuence oulcome when mental health services were still e~~entially hospital ha~~d liigC. gCIHJcr ;lnd cthnic l'lI(!inl different comparisons were perBecause patients were found by a case identlficallon t~chOlque. 111'1 form~d Besidt~ wmraring nC>lh!!rc>upsas a whole. we stratified hy all were in contact with the psychiatric servic~~. tlhnoc grour. Py g~nder and Py age ( :t ~ years). We also defined rall< of ratients wh<,m. in addilion to malChinll by lIender and age. \\ t al~,. matched t>~r~ychorathol,'g~ and social funclionins. (BPRS Berlin sample :t 5 and GAF :t ~I These matched couples were statistically Ireal~d a~ rairs ,HId their data. wer~ compared using Ihe Mann. In B~rlin all patients under treatmenl at the derartment of sc>c..d Whllney I' lest for related '"mples and the McNemar test. psychiatry were asked to participate in a larger study on qualil~ (.f life and need. This care system is oriented towards prC>\'lding longterm treatment for all patients in Charlollenburg with severe and chronic mental illnesses. It includes three partial hosrilalizati(.n programmes. various out-plltient facilities Rnd communily-ba~ed Results services such as a day-care centre. a drop-in centre and sheltered accommodation. These different components function as a network, with patients easily transferred from one form of care to another Sociodemographic characlerislics accordin'g to their current condition and needs. Interviews were completed for 89% of patients in the care system. The drop-out was The data of 69 pa'tient~ from Berlin and 7S patients due either to patients refusing to take part 1!I°...oI.heing una\'ailahl~ (3%) or being so unwell that their mental ~Iat~ rendered them una hie from London were used for comparison. The mean ((\ co-operate (I %1. duration of the illness was 17.8 years in London (SO Instruments Diagnosis was made by OPCRIT [17J rallng c>fca~~ nnt~s and SCA N IPSE) 01 interviews in th~ Lc>ndon grour and h~ consultanl diagnoses according to ICD I0 re~earch dlagno~'ic l'lItella in Ihe Berlin group. In both groups. p~ychopalhc>logy wa~ a~sessed using Ihe Brief Psychiatric Raling Scale (BPRSI (9) and the Glohal Asses~ment of Functioning Scale (GAFI [I) For evaluallon c>fthe subjec. live perception of the quahl)' of life and need.lhe 1993 \'~r~lc>nsof Ihe Lancashire Quality of Life Prc>file(LQ(,U [I!J and Ih~ Camoer"dl o\,sessment of Need ICANI (JOJ Wert u,~d h'r Iht LI'lId('n grour and their respective German inslrument, h.r Ihe Berhn group: Ih~ Berliner Lebensqualit3tsprofil (BELPI and tht Btrhner Btdiirfnisln"entar(BeBII (11). The LQoL has 11 ~ectlons nn dlff~rcnt a~rects "f life (i.e work. leisure. religion. finances. legal and safety IS'UCS.family relations. social relations and healthl. Each seCllon c(}nq~ts of que~tions about the objeclive situation and of ~atlsfacti()n ratings for the aspect 10 quesllon using a scale frc>m J \(1 ., The CA N ;I~,~,~~s. h,.th In the view of the client and the case manager. If ther~ are an~ problems 10 22 areas of life. If the patienl c>rcarer idenllfies a prpp. lem. enquiry is made as 10 Ihe htlp b~1I11.! received f,'r 11 ~R!lStical analysis .\< Ih~ I'KIS~1 ~tud\' and the Hcrhn '1'I'\Cl" spread

,.f ratlent~

\\llh

'l"cre

lIIen\;l1

.nd"d,'..

hl'l,,,,,'c,'II\''''''

111111"'. "111\ 1'..1'1'11" \\ Ilh

11.9: range 0-521 and IS years in Berlin (SO 8.7; range 3- 4~1 There was a higher. but not significant, proportion of female~ (48% \'s 32%) in the Berlin group. The gender difference did not account for the difference in the duration of Illness. The groups were comparable in age (mean 43.6 year~. SD 10.6 in Berlin; mean 42.9 ye;Jrs. SO D.2 in Londonl. The London distributi()n was wider (range: 17-75 years) because the Berlin department (range: :!4-69 yearsl does not Ireat ad(llescent or l)lder patients. The groups showed a highly ~ignificant difference in the BPRS total score~ (mean ,~74, SD I ~.1 in Berlin; mean 28.8. SD III.X in London: P < (1001 I. The Berlin group had generally higher ratings in the BPRS subscales [Asthenia' 6.:!(SO :!9) cf. 3.Y(SD 1.5);Anxiety Depression: 9.9 (SD 4.4) cf. 1<.4(SO 4.81: Anergia: 9.1 (SO 4.01 cL :; 9 (SO 301: Hostility: :;.0 (SO 2.61 cc. 4.7 (SO 2.41: Thought disorder: 7.1 (SO 3.6) cc. 6.6 (SO 4.01]. \\'ith the first three differences being significant. The GA r ~cores were not significantly different (mean 58.3. SI) 12 X in Berlin: mean 51\.5.SD 17.2 in Londonl. The pr('r"lljnll \\'ho \\'ere 111emhers pr ethnic minorities (7"" III Berlin. (~' ~ 11UIIII

.~)"" If) I.lIllth.nl

differed

significantly

294

"

Subjective perception of well-being and pf()hlem~ The answer to the question: "How sati~fied are you with your life in general?"" (Tablc I) showed a significant difference, with a satisfaction ,'a'lue ~)f 4.~ (SD 1.5) in Berlin and 4.2 (SD \.131 in Llllldon. This means that Berlin patients in general were more satisfied with their quality of life. We saw the same differences in the needs assessment. In London, pa tients indicated significantly more problems. Although the London group scored a higher number of problems on all three areas of need, only the difference in material needs reached significance, but not the social and psychopathological need subscores. There was a lower subjective rating of all items of quality of life in the British group. The quality of life subscores indicated that this might be due to differences in the material aspect of life and therefore confirmed that it was in the material aspect of life that the most substantial differences between the two groups were demonstrated.

Quality of life in different life domains In occupation there was a higher rate of people (Table 2) working or otherwise being occupied in Berlin, but this difference was not significant. There was no difference in the satisfaction with occupation, both groups showing a high satisfaction with their occupation despite the overall lower London satisfaction rates. There was a wide and significant difference in the satisfaction with being unemployed or not having an occupation. In finances all objective measures demonstrated highly significant differences favouring the Berlin group. Despite the enormous objective differences (income: Berlin DM 1376 (SD 707), London DM 859 (SD 519) per month; DM 1.00 = £0.45) the financial satisfaction rate did not differ significantly and was similar to the general satisfaction rate. In living conditions more or less the same numbers of patients were able to live independently, but London patients were more likely to want to move and they shared their apartments with more people. In this area we found very high satisfaction rates for the Berlin group and larger differences between Berlin and London. In particular, satisfaction with privacy was higher in Germany, possibly reflecting the high occupancy of Table 1 Results of [leneral subjective ratings: general satisfaction rated on scale of I to 7 (LQoL and BELP; 1 = can't be worse. 7 = can't be belterl; sums of problems rated in needs questionnaire by patients (CAN and BEBI. high figure indlcallng more problems 10 this field I

General satisfactIOn with quality of life Sum of problems rated Matenal prohlems rated Social problems rated Psychopa

t h, ,1('[11<.:;.1

prohlem< rated

..

L<.'\fHj(lnapartments. In this area satisfaction differences generally reflected objecti,'e differences. The area of social I(le was complex and cannot be summarizcd by anyone trcnd in the data. More Berlin patients had a partner and thcy could not imagine li,'ing without friends, while more London patients had contact with relatives. In general, the Berlin group was more satisfied with their social life, but the differences on these ratings were not significant and they differed less than the general satisfaction difference. Finally, in s('curi,y, there was no difference in satisfaction with the security generally, but there was a significant difference in the satisfaction with the security in the local area. This indicates that the Camberwell district included more local areas that were perceived to be unsafe.

Analysis of matched groups Matching the two groups by gender and age (known to influence outcome) we identified two groups each of 55 patients (36 males, 19 females) with a mean age of 41 years (SD years 11.1). The same tests as for the overall samples yielded essentially similar results. Only the items satisfaction with the ability to relate socially (P value: 0.015; previous value: 0.076) and the sum of psychopathological problems rated (P value: 0.015; previous value: 0.071) now showed significant differences. The proportion of patients with a steady partner and the sum of material problems rated were no longer significant. In order to reduce the influence of ethnicity, data on white Europeans only were analysed sc:parately. The 64' patients from Berlin (34 male, 30 female) and the 49 patients from London (37 male, 12 female) included in this comparison had an average age of 44.3 years in Berlin and 46.0 years in London. Significant differences were now also found in the rate of patients living independently (P value: 0.003; previous value: 0.32), satisfaction with family contacts (P value: 0.04; previous \'alue: 0.1), the sum of social problems rated (P \'alue: 0.014: previous value: 0.064), and the duration of illness (P value: < 0.001; previous value: 0.24). The proportion of patients having a partner or having a frequent contact with family members now no longer differed significantly, nor did the depression subscore of the BPRS or satisfaction with the privacy in the living arra ngcments.

Berlin

London

Test value

4.8 SD 15

4.2 SO 17

('

7 SO H 0.7 SO 0.1< 14 SO I 5 OQSDIO

5.1 SO o I 0 SO 0.8 I.!<SI) 15 IISDOQ

l: = 1752.0 ti 1922.5 20I!t5 l' l' ='20.WO

------...------.

= 20650 = =

P value <0.05 < 0.01 < 0.05 n.s. (0.0641 n,,100711

, . .,

4....

:?l} _..

Tabl 2 Results of quality of life ratings: objective and subjective data (satisfaction rated on 1-7 scale) from quality of life questionnaires (LQoL and BELP) for the Berlin and London sample

Ikrlln Occupation Has occupation Satisfaction with occupallon (n '"' 27:n = 20) Satisfaction wilh nol having an occupation (n 42'n co 471

Woo 48S0

IJ

46 SO 1.9

L<'nd<'n

Test value

21100 4.7 SD 17

-

n.s.

-

n.s.

7SD

]7

tI

= 685.0

P value

< 0.05

=

Flnanc('s Income (OM/monlh, Receives benefits Lacked money last year Satisfaction with finances Lit';ng Lives independently

No. of persons patient lives with Wants to move Satisfaction with hVlng conditions Satisfaction with privacy there Satisfaction with the expectation to live there for a long tim Sociallif(' Has a partner Contact with relatives more than once a week Could imagine living without friends Has at least one friend Satisfaction with partnership (n

- 32/" - 10)

Satisfaction with family relations Satisfaction with ability to relate socially Satisfaction with no of friends Security Was accused or charged Was victim of assault Satisfaction with security Satisfaction with security in local area

859 SO 519 84% 49% 3.8 SO 1.6

U "" 1207.0 25.80 Chi' Chi' so 7.09 U '" 2140.0

< 0.001 < 0.001 <0.01 n.s. (0.069)

70o 0.

68%

IJ SO 2.9 35% 5.0 SD 1.6

3.3 SO 5.8 58% 4.3 SO 1.6

U

n.s. < 0.001 < 0.01 < om

5.5 SO 1.3 5.1 SO 1.7

45S0 1.7 4.1 SO 1.8

U U

30% 190.;'

15% 36%

Chi' .. 4.98 Chi' = 5.05

19%

52%

Chi'

7]% 4.9 SO 1.5

71% 4.7 SO 1.8

4.8 SO 1.6 5.1 SO 1.1

4.7 SO 1.6 4.7 SO 1.4

-

4.8 SO 1.6

4.4 SO 1.6

U

4 °'0 13% 47S0 1.5 53 SO J.2

8% 4.4 SO 1.6 4.2 SO 1.5

1376S0707 440/.0 28... 4.3 SO 1.7

10%

-

= 1595.0

Chi' li

c

6.99

= 1917.09

- 1576.5 -

1693.0

= 16.42

-

< 0.001 < 0.01

< 0.05 <0.05 < 0.001

n.s.

-

n.s.

U = 2068.0

= 2095.5

-

-

U .. 1465.5

n.s. n.s. (0.076) n.s. (0.10)

n.s. n.s. n.s. < 0.001

BPRS. GAF). In predicting general satisfaction. three Matching patients not only by age and gender, factors [age (beta = 0.21),the question of whether pabut also by severity of psychopa thology. genera ted ten pairs (eight male, two female) with an average age tients had at least one friend (beta = 0.22) and GAF (beta = 0.21)] were included in the equation, resulting of 42 years, a BPRS of 31 and a GAF of 60. This comparison, despite the reduction in confounding fac- in a multiple r of 0.39. For the sum of problems rated, tors. yielded no further information because of the a multiple r of 0.47. with gender (beta = 0.26),income (beta = 0.2I) and G AF (beta = 0.25) as significant prerestricted sample size. with only the sum of the material dictors. was found. In predicting satisfaction with living problems rated still showing a significant difference IP value: 0.043). conditions. only 0ne significant predictor was identified In a final step. we tested whether the inOuence of the with a multiple r of 0.22. and this was whether patients lived in London or in Berlin. London/Berlin difference in patients' subjective ratings was explained by objective factors. The groups showed the greatest difference in satisfaction with living conditions. general satisfaction and the sum of problems Discussion rated. These three criteria were each taken as dependent variables in a step-wise forward multiple regresData from the comparison of these two groups were sion analysis. All socio-demographic and objective consistent with our hypothesis that the objective stanclinical data from the study were included as independent variables (e.g. gender. age. income. 0ccupalion. . dard or living. or chronic schiwphrenic patients differs

...t.

greatly between Berlin and London, DIITl.'rl.'nccs\\l'rC marked in the areas of fin..nces ..nd li\'ing I.:nnditi,)n\. probably due to the differing support gl\cn by Ihc British and German social systems. Thai a higher pn'portion of patients in Berlin Ii\'cd in parlncrships and could not imagine living wiIhnut friend, suggcsts t ha I the social anchoring of the Berlin paticnts was bettcr. On the other hand. patients in London had more contact with their relatives, which might he rclated to their living closer to family members. ie, sharing thc same house. It seems that the German group. despite exhibiting higher psychopathology, achieved a better quality of life, as well as a similar global social functioning, because of their better socio-economic conditions. The different level of psychopathology makes the quality of life differences even more striking. If patients with a similar psychopathology were compared. the objective quality of life differences would probably be even wider. However, the samples were drawn from different study settings: the Berlin patients were being treated in a model service for severely ill and chronic patients. The London group was a communit)' sample of people with psychosis, and therefore included more post-acute and less chronic states. Also, the interviews in Berlin and London were performed for different purposes. The Berlin patients had been in the treatment system for a longer time and had been interviewed in order to evaluate long-term care. In London, the interviews were performed as part of a larger study of representative cases of psychosIs in a population-based study. The data were collected as the baseline stage. when mental health services were still essentially hospital based, and not all patients were in contact with mental health services at the time they were interviewed. Our hypothesis that the differences in subjective ratings of quality of life would not be directly relaled to the objective differences was not supported fM all life areas. There was a lower satisfaction with quality of life in the London group in g.eneral. and we found both patterns of relationship between subjective and objective ratings; in some areas (occupation. finances. social life) the differences in the subjective ratings were less than the differences in the objectivc data. e\en though the latter differences sometimes seemed dramatic. Here one can observe that the subjective perception did not automatically follow the objective situation. sug.gesting that there are other factors influencing satisfaction with quality of life. We have discussed these possible other factors elsewhere [5. 6. J I]. In the area of living conditions. however. the subjective ratings not only followed the trends of the objective data. hut showcd a widcr difference. In summary. it appears that thc objective parameters were influential on the subjecti\e perception of life quality. but were not the only influencc. The single observation that the satisfaction \\'jlh beine. unemployed differed so much bctween Ll'ndpl1 0IIHI13crlin could be explained h~ Ihe fOletthOlIthl" (;erl1lal1 ,(,erOlI

..cellr!!y systcm ()ITl.'rsa higher level of support for IInemplpycd paticnts, Thelc \\'Crl.'\pme mctlwoological problems in comparing thc \wo groups, Although it was not statistically ~ignificanl. thcrc was a diffcrence in the sex ratio be. t\\ccn thc IWll groups: IlOwc\cr. this was unlikely to affcct the quality of life results as a separate analysis showed no scx difference in quality of life. The districts of Camberwell and Charlottenburg are not totally comparable. and so some items of the questionnaires used in this study were not included in the statistical comparison. This study was neccessarily naturalistic. and any such study making international comparisons would share the same problems of heterogeneity within the samples in terms of age. gender, ethnicity, etc. The strategy adopted here of matching for these potentially important variables meets the problem of small numbers and. therefore, the likelihood of type 2 errors with no differences being found. Thus. this study and the shortcomings of the matched pair approach illustrated s()me of the methodological limitations that are associated with any naturalistic comparison of community care samples in different countries'. The other inherent prohlem is that between two countries both the health service and the social infrastructure will differ making it impossible to tease out the influence of each on any outcome measure. Single item comparisons are essentially problematic. For example, the differences in income should be corrected for the standard of living in the two cities. In the social field, the understanding of the importance or family relations. partnership or friendship might be quite different in London and Berlin. and this should be taken into account in the comparison. Methodologically. it is impossible to account for all such influences. In particular. broad cultural differences may not be quantifiable. Matching of parameters classically regarded as highly influential (gender. age) did not help to cxplain the differences. but the imporlance of ethnicity \\'as demonstrated. A numher of differences achieved significance when comparing white Europeans only. and so we can assume that the differences were masked in the carlier comparisons because of ethnic heterogeneity. Similarly. partnership and family contact were no longer significant when minority ethnic group patients were excluded. Living and social conditions in general seemed to be different within the London group betwecn while Europeans and minority ethnic groups. and the differences between the Berlin and London population might be smaller for white patients in both cities than between different ethnic groups in one city. We have to assume that the differences that were found in the general comparison (apart from partnerships and ml're frcquent family contact in London) were mainly due 10 different livinl! conditions of the ethnic minorItle.. or the diffcrent- cultural ..urToundings of ethnic !!IPUp" We \\'ollld like to draw al1enti()n to the finding -\11;11the SUI11(,f 111:111:1'1011 pl'l,bkrn,

ralcu

\tdl ,llOwcd

significant differences undcr the cOnOllll\nS of Ihl' matched-pairs comparison. This indicates that the differences between the material aspect of life 4uality was .1 rl")~ust tinding independent of age. gender or rsych()ratlh)lo~n. The multiple regression analyses wcre Iwt a~le to predict \'ery large proportions (If the \'ariance In any of the three dependant variables. It is interesting that with the dependant variable "satisfaction with li\ing conditions". the variable Berlin London had some Impact that was not explained by any of the other \'ariables examined in our study, Although the multiple r was modest. this finding indicated that the local surroundings patients live in are .relevantto the interpretation of the results of quality of life research. In conclusion, we demonstrated that the relationship between subjective and objecti\'e Quality of Life is not straightforward. and that when comparing such data between different cultures. complex inlc'ractions of faclors exist that must ,be taken into account. In order for this to be possible there is a need for morc studics in which such international comparisons are madc. and further discussion of the inherent mcthodological difficulties is inevitable.

References 1. AmeriCilnPsychiatric Association« 1987)DSM-III-R: diagnostic and statistical manual of mental disorders, 3rd edn. revised, American Psychiatric Association. Washington. DC 2. Bowling A (1991) Measuring health: a re\'iew of quality of life measurement scales, Open University Press. Milton Ke)'nes 3. BrewinCR. Wing J (1993)The MRC Needs for Care Asst:ssmenl: progress and controversies. Psychol Med ::3: IOR-841 4. Brugha TS. Wing KJ. Brewin CR, MacCanhy B. Mangt:n S. Lesague A. Mumford J (1988) The problems of people 10 longterm psychiatric day care, Psychol Med 18: 443-456

~. (iru, ler~ T. Prod'e S 11(N-11 I !'er~' a~se'smenl (If rsychl31nc treatmellt rc~ull, ;11I<1 p,,,hlem~..f a 'y~lematll: e\amination fin (;<:1111.11I1 1',~dll"ll I'la\ ~I "" 'I~ (. Iklllle ~I. I'lIehe S II'IIJ~I The CI'ncepl or need m ps~chlaln, re~e3r.'h IIn German I FllnJam rs\chlatr Q: ~:! (\0 1 \Iullen M. Mark~ I. C'\I1I)(I11\ .I. AudIO' B t IIJ':I:!I Home ha'ed l'are and ~tandarJ I\(\~pltal care f(\r pallC:nls "Ith 'e\ere l11enlal IlIne'~ a ranJoml~ed n.ntrnlled IlIal BMJ 3~: ~~91~4 R Olivcr .Ir J 114Q11 The ~ocial care directi\ e: de\ el(\pment of a quahty (,f life profile r(lf use m communit)' ser\'ices for the menially ill. Sot Work Soc Sci Re\' :I: 5 -45 'I. 0, crall' 11:. (illl h;Im DH 11'11<1\1 The Bnef Psychilltric Ralinll Scales IBPRSI recent developments in ascertainment and scalinll. Psychopharmacol Bull ~4' 97 - 99 10. Phelan M. Slade M. Thornicroft G. Dunn G. Holloway F.

Wykes T. Strarhdee G. Loflus L. McCrone P. Hayward P (l99~1 The Camherwell Assessment of Need (CAN): the \'alidlty and rehahihl)' of an instrument to assess the needs of peuple: with ~e\'ere mental illness. Br J Psychiatry 167: 51<9 W5 11. Pnebc S. Gruylers T. Hein7e M. Hofl'mann Cm. Jiikel A (19951 SUh.lecll\'e criteria ror e\'aluallon of psychialric care - methods ,for 1I~~eSMncnt III re~cllrdl and routine care (in Germanl, Ps\'chialr Prax :!:!: 1-1(1 144 I:! !'riehe S. Grll)'lt:rs T fI '19,11Tht: ((,le of Ihe helping alliance m psychIatric community cart: a prospecti';e stUdy, J Nerv Ment Dis 1111:552 ~57

Li, Prit:bc S. Gruyters T (I':I9~1Patients' assessment of treatment predicting OUlcome.Schizophr Bull 21: 87-94 14, Skantze K. Maim U. Dencker SJ. Ma)' PRA, Corrigan P (1992) Comparison of qualilY of life with standard of living in schizophrenic out-patients. Br J Psychiatry 154: 77-82 15. Steinhart I. Prieve S (1992) Prediction of hospitalisation within a psychiatric

community

care

system

-

a five year

study,

Soc

Psychiatry Psychiatr Epidemiol 27: 270-273 16, World Health Orllanization (19911 Tenth revision of the International C'lassification of Diseases (lCD-IO) 17. MacGuffin P. Farmer A. Harvey I (19911 A polydiagnostic applicati(\n of orerational criteria in Studies of psychotic illness: development and rt:liabilily of the OPCRIT system, Arch Gen Psychiatry 48: 7(,4 ,.770