Principles of Social Psychiatry

QUALITATIVE RESEARCH METHODS 83 they do not access services? Who should conduct ing selectivity or ‘cherry picking’ from the data to fit the interviews? Should it be an interviewer from the the researcher’s preconceptions. The analytic proce- same ethnic background? We recruited participants dure should be flexible enough to allow emerging through GPs and local organizations for older peo- themes to be included and to modify earlier themes. ple from particular ethnic groups. The interviews The process should include comparison of data be- were held at their chosen venue and in the language tween and within cases and refining the coding strat- of their choice. In order to encourage discussion in egy as a result. Seeking out deviant cases is another participants who were reluctant to access treatment principle of qualitative analysis. By purposefully for depression, the interviewers did not have a looking for data that do not accord with emerging clinical background but were specialists in qualita- patterns the researcher is better able to refine the tive interviewing with older adults. developing theory, to delimit its generalizability and to identify areas for further data collection. Rigour in qualitative analysis is maintained by involving other 7.4.2 Data analysis: preliminary researchers in the coding process and by keeping considerations detailed notes on the process and development of themes. As illustrated above, the method of analysis should be considered before data collection begins. There are four preliminary criteria to consider: 7.4.3 Data analysis: approaches . Does the literature review include sufficient evi- dence to inform the topics to be covered in the There are a number of approaches to data analysis, study? If this is the case then a thematic analysis, depending on the level of abstraction required. Some based on predetermined themes, is likely to enable studies are essentially descriptive, aiming to identify the appropriate level of interpretation of the data. the range of attitudes towards particular treatments or services. In studies of this type, such as service . Where little is known about the phenomenon in evaluation, the domains in which individual opinions question and the study aims to increase understand- are sought may be largely predetermined. Analysis is ing or develop theory then a grounded theory ap- unlikely to proceed beyond the descriptive level since proach [26] to analysis might be appropriate. the researchers are not aiming for high levels of interpretation or theory development but to enable . Does the investigation focus on how the individual service providers to understand the perspectives of the makes sense of their lived experience? Will the various ‘insiders’. There are also more deductive analysis require a high level of interpretation methods of analysis, such as traditional content ana- from the researcher? If the answer to these questions lysis, in which predetermined coding categories are is ‘yes’, Interpretative Phenomenological Analysis applied to the whole dataset. (IPA) [28] is likely to be appropriate. In contrast, Interpretative Phenomenological Ana- lysis [28], as the name suggests, is an explicitly . Where the aim of collecting qualitative data is to test interpretative approach and adopts an idiographic hypotheses on a large dataset, then the more deduc- (case-by-case) approach to analysis. The researchers’ tive method of content analysis might be appropri- theoretical position is therefore central in ‘making ate [29]. This method can be applied to published sense’ of how the participant understands the phe- sources such as newspapers, official documents and nomenon (e.g. the experience of living with an eating visual recordings. disorder from the perspective of the person with the diagnosis and their supporter). IPA prescribes a num- Whichever method is chosen, qualitative analysis ber of steps in the process of analysis: empathic must always be systematic and comprehensive, avoid- reading; critical reflection; making impressionistic

84 PERSPECTIVES AND METHODS notes; applying themes to capture the concepts con- distinguish the processes of grounded theory from tained in the narrative (e.g. loss, coping); clustering of basic thematic analysis: themes into descriptive categories with shared mean- ings; producing a summary table of themes supported . The precise research problem should emerge from by original data from each interview. There are three the data; therefore a preliminary review of the levels of analysis: thematic (descriptive), interpreta- literature is not encouraged. This proceeds along- tive and theory building. Reflexivity is particularly side the iterative (nonlinear) process of theoretical important in IPA because of the central position of the sampling and analysis. researcher in interpreting the data. Thematic analysis, perhaps the most widely used . Constant comparison within and between cases is method, is positioned between IPA and content ana- central to the analysis process as is negative case lysis on the descriptive–interpretative spectrum. How- analysis. Theoretical memos are kept throughout by ever, it is more important to understand the purpose the researcher as a detailed record of developing and procedures of qualitative analysis and to apply theory. them in a systematic way than to adhere to a particular terminology. . There are three levels of analysis – open coding, Grounded theory [26,30] is a theoretical approach organizing the codes into more abstract categories to generating higher levels of understanding of social and theoretical coding – in which the links between phenomena that are derivedfrom a systematic analysis the categories are identified. These are expressed as procedure. There are a number of principles that hypotheses or propositions. 7.5 MIXED METHODS Mixed methods are now the standard way of describ- exploratory or pilot stages of a project to generate or ing studies that combine quantitative and qualitative firm-up hypotheses, which are then tested using quan- methods within a single design. In health research, titative methods. Alternatively, qualitative interview mixed methods are increasingly common. A recent data may be used to develop structured questionnaires analysis found that mixed method studies formed 30% that can yield quantitative data. Implicit in this ap- of projects funded by the UK Department of Health proach is the view that scientific rigour and increasing between 2002 and 2004, up from 15% in 1996 [31]. It certainty in findings will only come via the subsequent is not clear how common such approaches are in use of quantitative methods, with qualitative methods mental health research, but there are no reasons to being only a precursor to the more scientific, quanti- think the trend is different from that seen more tative methodologies. The second broad approach generally. resists any ordering of methods into a hierarchy. The Leaving aside increasingly redundant philosophi- researcher, from this point of view, has a range of cal debates about the compatibility of qualitative and methodological tools at his/her disposal and simply quantitative methods, in broad terms three approaches selects the ones most suited to the task. The final broad to the combined use of both qualitative and quantita- approach is that in which qualitative methods are tive methods can be identified: (1) quantitative re- viewed as the primary, most effective, tools. Within search as senior partner; (2) equal status for both; and this, moreover, it has usually been participant obser- (3) qualitative research as senior partner. vation that has been proposed as the gold standard of The first of these is perhaps the most familiar in research methods on the basis that it offers ‘the most mainstream medical and health research and derives complete form of sociological datum’ ([32], p. 322). from the perception of qualitative methods as parti- There are a number of ways in which methods cularly useful in hypothesis generation. The purpose can and have been practically combined. As noted here is usually to use qualitative methods at the above, qualitative methods may be used to develop

QUALITATIVE RESEARCH METHODS 85 quantitative tools. To take another example, qualita- methods, of clinical practices aimed at investigating tive methods may be used to explain quantitative how and why variations observed in the epidemiolo- results – epidemiology is often not able to explain gical stage came about. In this example, qualitative why associations arise; qualitative methods have the methods are used to illuminate differences observed at potential to penetrate this epidemiological black box. a population level: as Pope and Mays [1] note, it is a A particularly good example of how combining meth- case of qualitative methods ‘reaching the parts other ods can work to capture social processes more fully is methods cannot reach’. given by Pope and Mays [1] in their series on quali- Methods, then, are chosen according to their tative research in the British Medical Journal. They strengths. As has already been suggested, quantitative describe a study by Bloor et al. [33] that set out to methods are particularly strong in enumerating pat- investigate differences in the geographic incidence terns in, for instance, pathways to care. Qualitative of operations on tonsils and adenoids and local methods, again as suggested above, are more effective differences in specialists’ clinical practices. To in capturing processes and what has been termed the address this issue a two-staged approach was adopted, ‘insider perspective’. As Silverman ([34], p. 14) notes, in which the first step was an epidemiological study ‘There are no principled reasons to be qualitative or of regional variations in operations and the second quantitative. It all depends on what you are trying step was a sociological study, utilizing qualitative to do.’ 7.6 CRITICAL APPRAISAL OF QUALITATIVE RESEARCH Given the variety of approaches outlined in this chap- 3. Was the data sufficiently comprehensive in breadth ter, the reader needs to be aware of the criteria for and depth to generate and support interpretation? judging the empirical quality of qualitative research. Was an iterative approach to data collection and There are two principle questions: analysis used? Can the data be audited from field- notes, audio recordings, transcripts? . Was the study designed to address the research ques- 4. Was there a systematic approach to data analysis, tion and objectives appropriately? including efforts to identify contradictory data? Was the analysis corroborated by more than one . Was the study conducted with sufficient rigour to researcher? achieve its aims? 5. Do the results of the analysis look credible and do The description of the methods and outcome of the they justify the conclusions? study should include the following details: 6. Are the limitations of the study clearly described? 1. How were participants selected: were they relevant What contribution does the study make to our to the research question and was their selection previous knowledge of the area? How do the find- reasoned (purposive/theoretical)? If a convenience ings fit with existing theory? If relevant, what are sample was used, was this justified? the implications for service development? 2. Were the data collection methods appropriate? Further guidance on establishing the rigour of What was the rationale for focus group discussion, qualitative research can be found in Barbour [5], Mays individual in-depth interviews or participant and Pope [6], Kuper et al. [8]. observation? Who conducted the fieldwork/data One of the strengths of qualitative research is collection and has the perspective of the researcher flexibility: the iterative approach enables the resear- been adequately considered (reflexivity)? cher to adapt the design in response to data analysis.

86 PERSPECTIVES AND METHODS With this flexibility comes the imperative to justify process. Validity of findings can only be established and describe systematically each step in the research through transparency in reporting. 7.7 CONCLUSION Qualitative methods have long been treated with 8. Kuper, A., Lingard, L. and Levinson, W. (2008) Criti- suspicion in health research, particularly in psychia- cally appraising qualitative research. British Medical tric research. This is decreasing, and there is now a Journal, 337, a1035. widespread acceptance that qualitative methods are 9. Lingard, L., Albert, M. and Levinson, W. (2008) vital to understanding the social processes that under- Grounded theory, mixed methods, and action research. lie mental health problems, their cultural expression British Medical Journal, 337, a567. and patterns of help-seeking from, and engagement 10. Hodges, B. D., Kuper, A. and Reeves, S. (2008) with, mental health services. As qualitative methods Discourse analysis. British Medical Journal, 337, a879. have become more acceptable in health research, so they have become more rigorous and systematic, with 11. Reeves, S., Albert, M., Kuper, A. and Hodges, B. D. (2008) Why use theories in qualitative research? British clearly explicated procedures for sampling, data col- Medical Journal, 337, a949. lection and analysis. As such, qualitative methods 12. Reeves, S., Kuper, A. and Hodges, B. D. (2008) Qua- (singly or in combination with other approaches) have litative research methodologies: ethnography. British become essential tools for more fully understanding Medical Journal, 337, a1020. and responding to the challenges of mental illness. 13. Vidich, A. J. and Lyman, S. M. (2000) Qualitative methods: their history in sociology and anthropology, REFERENCES in The Sage Handbook of Qualitative Research, 2nd edn (eds N. K. Denzin and Y. S. Lincoln), Sage, 1. Pope, C. and Mays, N. (1995) Reaching the parts other Thousand Oaks, California, pp. 37–84. methods cannot reach: an introduction to qualitative 14. Brown, C. and Lloyd, K. (2001) Qualitative methods in methods in health and health services research. British psychiatric research. Advances in Psychiatric Treat- Medical Journal, 311, 42–45. ment, 7, 350–356. 2. Britten, N. (1998) Qualitative research: qualitative 15. Barbour, R. S. (1999) The case for combining qualita- interviews in medical research. British Medical Jour- tive and quantitative approaches in health services nal, 311, 251–253. research. Journal of Health Services Research Policy, 3. Jones, J. (1998) Qualitative research: consensus meth- 4, 39–43. ods for medical and health services research. British 16. Murphy, E., Dingwall, R., Greatbatch, D. et al. (1998) Medical Journal, 311, 376–380. Qualitative research methods in health technology 4. Keen, J. and Packwood, T. (1995) Qualitative research: assessment: a review of the literature. Health Technol- case study evaluation. British Medical Journal, 311, ogy Assessment, 2 (16). 444–446. 17. Denzin, N. K. and Lincoln, Y. S. (2000) The discipline 5. Barbour, R. S. (2001) Checklists for improving and practice of qualitative research, in The Sage Hand- rigour in qualitative research: a case of the tail book of Qualitative Research, 2nd edn (eds N. K. wagging the dog? British Medical Journal, 322, Denzin and Y. S. Lincoln), Sage, Thousand Oaks, 1115–1117. California, pp. 1–28. 6. Mays, N. and Pope, C. (2000) Assessing quality in 18. Lawrence, V., Murray, J., Banerjee, S. et al. (2006) qualitative research. British Medical Journal, 320, Concepts and causation of depression: a cross-cultural 50–52. study of the beliefs of older adults. The Gerontologist, 7. Kuper, A., Reeves, S. and Levinson, W. (2008) An 46 (1), 23–32. introduction to reading and appraising qualitative re- 19. Lawrence, V., Banerjee, S., Bhugra, D. et al. (2006) search. British Medical Journal, 337, a288. Coping with depression in later life: a qualitative study

QUALITATIVE RESEARCH METHODS 87 of help-seeking in three ethnic groups. Psychological 27. Bhugra, D., Baldwin, D. and Desai, M. (1997) Focus Medicine, 36, 1375–1383. groups: implications for primary and cross-cultural 20. Murray, J., Banerjee, S., Byng, R. et al. (2006) Primary psychiatry. Primary Care Psychiatry, 3, 45–50. care professionals’ perceptions of depression in older 28. Smith, J. A. (1996) Beyond the divide between cogni- people: a qualitative study. Social Science and Medi- tion and discourse: using interpretative phenomenolo- cine, 63, 1363–1373. gical analysis in health psychology. Psychology and 21. Kamberelis, G. and Dimitriadis, G. (2005) Focus Health, 11, 261–271. groups: strategic articulations of pedagogy, politics and 29. Weber, R. P. (1996) Basic Content Analysis, 2nd edn, inquiry, in The Sage Handbook of Qualitative Research Sage, Newbury Park. (eds N. K. Denzin and Y. S. Lincoln), Sage, Thousand 30. Charmaz, K. (1997) Grounded theory, in Rethinking Oaks, California. Methods in Psychology (ed. J. A. Smith), Sage, London, 22. Prior, L., Wood, F., Lewis, G. and Pill, R. (2003) Stigma pp. 27–49. revisited, disclosure of emotional problems in primary 31. O’Cathain, A., Murphy, E. and Nicholl, J. (2007) Why, care consultations in Wales. Social Science and Med- and how, mixed methods research is undertaken in icine, 56, 2191–2200. health services research in England: a mixed methods 23. Quirk, A., Lelliot, P. and Seale, C. (2006) The perme- study. BMC Health Services Research, 7, 85. able institution: an ethnographic study of three acute 32. Becker, H. S. and Geer, B. (1969) Participant observa- psychiatric wards in London. Social Science and Med- tion: the analysis of qualitative data, in Issues in icine, 63, 2105–2117. Participant Observation: A Text and Reader (eds G. 24. Lee, D. T., Kleinman, J. and Kleinman, A. (2007) McCall and J. Simmons), Addison-Wesley, Reading, Rethinking depression: an ethnographic study of the Massachusetts. experiences of depression among Chinese. Harvard 33. Bloor, M. J., Venters, G. A. and Samphier, M. L. (1976) Review of Psychiatry, 15 (1), 1–8. Geographical variation in the incidence of operations 25. Skultans, V. (2006) Psychiatry through the ethno- on the tonsils and adenoids: an epidemiological and graphic lens. International Journal of Social Psychia- sociological investigation. Journal of Laryngology and try, 52 (1), 78–83. Otology, 92, 791–801, 883–895. 26. Glaser, B. G. and Strauss, A. L. (1967) The Discovery of 34. Silverman, D. (1997) The logics of qualitative research, Grounded Theory: Strategies for Qualitative Research, in Context and Method in Qualitative Research (eds G. Aldine, Chicago, Illinois. Miller and R. Dingwall), Sage, London.



Part Two Components of the social world



8 The social epidemiology of mental disorder 2 1 Ronald C. Kessler, Philip S. Wang and Hans-Ulrich Wittchen 3 1 Department of Health Care Policy, Harvard Medical School, Boston, Massachusetts, USA 2 Division of Services and Intervention Research, National Institute of Mental Health, Bethesda, Maryland, USA 3 Max Planck Institute for Psychiatry, M€ unchen, Germany This chapter reviews the epidemiological literature problems in the population (www.int/topics/global_ on the prevalence and societal costs of mental dis- burden_of_disease/en). A number of factors account orders. Interest in the costs of illness – not only direct for these results and have important implications treatment costs but the human costs as well – has for the design of treatment programmes for mental increased dramatically over the past decade among disorders: that mental disorders are commonly occurr- health policy analysts as part of the larger movement ing, often begin at an early age, often are quite to rationalize the allocation of treatment resources persistent throughout the life course and often and maximize benefit in relation to cost. Much of have substantial adverse effects on functioning. This the current interest in mental disorders among chapter reviews the epidemiological evidence regard- health policy makers is based on the fact that these ing these points, with a special emphasis on data from disorders have consistently been found in cost-of- the recently completed World Health Organization illness studies to be among the most costly health (WHO) World Mental Health (WMH) Surveys [1]. 8.1 METHODS OF ASSESSING MENTAL DISORDERS IN EPIDEMIOLOGICAL SURVEYS Information about the epidemiology of mental dis- Schedule (DIS) [2], an instrument developed for use in orders has proliferated over the past two decades. The a large community epidemiological survey in the reason for this can be traced to modifications in the US [3] and subsequently used in a number of similar criteria for diagnoses of mental disorders in the DSM surveys in other parts of the world [4]. system, beginning with DSM-III, that made it much The WHO subsequently developed the Composite easier than previously to operationalize diagnostic International Diagnostic Interview (CIDI) [5], which criteria. Fully structured research diagnostic inter- was based on the DIS, in order to have an instrument views appropriate for use by trained lay interviewers that could be used to generate diagnoses according to were subsequently developed for this purpose. The the definitions of both the DSM and ICD systems and first of these interviews was the Diagnostic Interview that could be used reliably in many different cultures Principles of Social Psychiatry, second edition Edited by Craig Morgan and Dinesh Bhugra Ó 2010 John Wiley & Sons, Ltd.

92 COMPONENTS OF THE SOCIAL WORLD throughout the world [6]. As general population sur- structs embedded in existing interviews have low veys were carried out in a number of countries with the relevance to some developing countries. This is not first version of the CIDI, the WHO developed a cross- a problem unique to the CIDI. Prevalence estimates national research consortium to carry out systematic based on other fully structured diagnostic interviews comparisons of CIDI survey results [7]. Results based have similarly implausible values. Methodological on these comparisons led to the expansion and refine- studies are underway to investigate these problems ment of the CIDI and to a new generation of cross- and to determine if modifications can be made to fully national CIDI surveys in the WHO World Mental structured interviews and/or to the recruitment and Health (WMH) Survey Initiative. The latter is an implementation procedures used in community epi- initiative aimed at carrying out and analysing the demiological surveys to generate more plausible pre- results of parallel CIDI surveys in countries through- valence estimates. Caution is needed in interpreting out the world. Twenty-eight countries have completed the results of epidemiological studies carried out using WMH surveys at the time this chapter is being written the CIDI or other fully structured interviews in devel- and close to 200 country-specific reports have been oping countries until these methodological investiga- published from these surveys (www.hcp.med.harvard. tions have resolved these problems. edu/wmh). Although only a small number of cross- Prior to that time, our best hope of obtaining national comparative WMH reports have been pub- accurate data on the population prevalence of mental lished so far [8–10], the first volume in a new series of disorders in developing countries is to use a two-stage WMH books was recently published that provides screening approach. In this approach, a representative very useful comparative data on disorder prevalence sample of community residents is administered a short and treatment [1]. We draw heavily on these data in culturally sensitive battery of screening questions this chapter. designed to determine which people might have a As the CIDI has become so predominant in psy- mental disorder and then the screened positives plus a chiatric epidemiological surveys, a few words need to representative subsample of the screened negatives be said about the extent to which diagnoses based on are administered a rigorous clinician-administered the CIDI are consistent with diagnoses based on research diagnostic interview. Two-phase screening independent clinician-administered research diagnos- studies of this sort, although rare, have been carried tic interviews. Clinical reappraisal studies of the out successfully in developing countries [12]. CIDI original version of the CIDI were quite mixed in this surveys could be carried out using this same approach, regard, some showing concordance of anxiety and with the CIDI used as the first phase and a culturally mood disorder diagnoses with clinical diagnoses to be sensitivesemi-structuredclinical interview used as the low and others moderate to good. Concordance has second phase. When the CIDI symptom-level data are been considerably better for more recent versions of strongly predictive of the clinical diagnoses, predicted the CIDI in clinical reappraisal studies carried out in probabilities of these diagnoses can be made for each Western countries [11]. Both individual-level diag- CIDI respondent and these predicted clinical diag- nostic concordance and consistency of the CIDI noses can be used as the outcomes in substantive anxiety disorder prevalence estimates with prevalence analyses [13]. The use of country-specific calibration estimates based on clinical interviews have been good rules could allow valid cross-national comparisons to in these studies. be made even when the validity of the CIDI varies Much less is known about the clinical relevance of across countries. This approach has not been used up diagnoses based on fully structured diagnostic inter- to now, however, because existing CIDI surveys have views in developing countries. CIDI prevalence esti- not included a sufficiently large number of second- mates in epidemiological surveys in some developing phase clinical reappraisal interviews to make it pos- countries seem implausibly low, raising concerns sible to implement this approach. It would be extre- either that research diagnostic interviews are not valid mely valuable for future CIDI surveys to include a in countries where there is no tradition of public clinical reappraisal phase that would allow this type of opinion research or that the Western diagnostic con- calibration analysis to be carried out.

THE SOCIAL EPIDEMIOLOGY OF MENTAL DISORDER 93 8.2 PREVALENCE OF MENTAL DISORDERS With these cautions as a backdrop, we consider the including implausibly low estimates like 4.8% in prevalence estimates of the mental disorders reported the People’s Republic of China and 5.2% in Israel. in published community epidemiological surveys. We Anotherone-fourthofestimates are higher than 16.7%, focus on anxiety, mood and behavioural disorders, as most of them in developed countries like France, New these are the disorders that have been most commonly Zealand and the US. It is possible that this cross- studied in epidemiological surveys. We also say a few national variation is real. Another possibility, though, words about nonaffective psychoses. Other disorders is that this variation reflects differences in the accuracy that have been the focus of epidemiological study, of diagnostic assessments in the surveys carried out in such as cognitive disorders, somatoform disorders, the different countries. Even when the same interview sleep disorders and personality disorders, are not schedule and field procedures are used across coun- discussed. tries, as in the WMH surveys, differences remain in the Several recent literature reviews have presented severity implied by purportedly similar words trans- detailed summary tables of prevalence estimates lated into different languages, in the rigour with which for individual disorders in the domains we review thesurveyswereimplementedandinthewillingnessof across a number of epidemiological surveys [14–16]. community residents to give honest reports about their A number of patterns are consistent in these reviews. emotions to interviewers. One is that anxiety disorders have consistently been Although some community epidemiological sur- found to be the most prevalent class of mental dis- veys attempt to estimate the prevalence of schizo- orders in the general population. The estimated life- phrenia and other nonaffective psychoses (NAPs), time prevalence of any anxiety disorder averages these estimates are generally recognized as inaccurate approximately 16% and the estimated 12-month pre- due to the high false positive rate of screening ques- valence of any anxiety disorder approximately 11% in tions concerning delusions and the low response rate the surveys carried out in developed countries. The in such surveys of people with a known history of estimated lifetime and 12-month prevalence of any NAP [19]. Clinical reappraisal studies show that only mood disorder, in comparison, average approximately a minority of the respondents classified by fully 12 and 6%, respectively. A smaller number of surveys structured interviews as having NAP are confirmed also study behavioural disorders, such as attention by blinded clinician-administered assessments [20]. deficit/hyperactivity disorder (ADHD), conduct dis- The CIDI’s more comprehensive approach is needed order (CD), oppositional-defiant disorder (ODD) to estimate the prevalence of NAP, as illustrated in a and intermittent explosive disorder (IED) [17,18]. recent national survey in Finland that began with the The estimated lifetime and 12-month prevalence CIDI as a first-phase screen for psychosis, but also of these disorders range approximately between 5 used a number of other first-phase screens that and 10%. included self-reports, medical examinations and There is wide variation around these averages. national registers [21]. A second-phase clinician- This can be seen clearly by considering, as an administered interview was then used to make diag- example, the range of lifetime and 12-month pre- noses. The lifetime prevalence of schizophrenia was valence estimates of anxiety disorders in the WMH estimated to be 0.9%. This estimate is consistent with surveys. Themedianlifetimeprevalenceestimateis the 0.7% estimate obtained in a meta-analysis that somewhat higher for anxiety disorders than in the knit together data from surveys and registries through- larger literature – 14.3% – but the interquartile range out the world to estimate the lifetime prevalence of (QR; 25th–75th percentiles) is very wide, between schizophrenia [22]. The lifetime prevalence of any 9.9 and 16.7%. One-fourth of estimates are below NAP, in comparison, was estimated to be 2.9%. These 9.9%, almost all of them in developing countries, estimates include survey nonrespondents who were

94 COMPONENTS OF THE SOCIAL WORLD known from registry data to have a history of NAP. now shown, however, that the original one-month Roughly 15% of the detected cases were survey requirement makes more sense from an epidemiolo- nonrespondents of this sort. In addition, consistent gical perspective than the more strict duration requi- with previous research, the CIDI failed to detect the rements in that the predictors of lifetime GAD majority of the clinician-diagnosed cases who com- (e.g. childhood adversity, family history of anxiety pleted both interviews, confirming the low validity of disorder, primary comorbid disorders) and the clinical the CIDI in assessing NAP. correlates of GAD (e.g. age of onset, course, role Focusing on individual disorders, specific phobia is impairment, suicidality) are all very similar for cases generally found to be the most prevalent disorder in with one-month durations versus six-month durations. community epidemiological surveys in developed Many of the people with episodes of GAD that do not countries, with lifetime prevalence estimates usually last as long as six months report having many episodes in the 6–12% range and 12-month prevalence esti- in their lifetimes, suggesting that this is a chronic mates in the 4–8% range [23]. Major depressive recurrent episodic disorder characterized by extreme disorder (MDD) is generally found to be the next stress reactivity that triggers episodes of excessive most prevalent lifetime mood disorder, with lifetime worry that remit (in less than six months) and then prevalence estimates usually in the 4–10% range and recur repeatedly over time. The six-month duration 12-month prevalence estimates in the 3–6% requirement results in these people failing to be range [24]. Social phobia is generally found to be the defined as clinically meaningful cases. If these diag- next most prevalent anxiety or mood disorder, with nostic controversies were resolved in the direction of prevalence estimates sometimes approaching those of broadening the criteria for PTSD and GAD, preva- MDD [25]. At the other extreme, bipolar I disorder lence estimates would increase substantially: as much (BPD-I) is usually found to be the least common as 50% in the case of PTSD and as much as 150% in anxiety or mood disorder, with lifetime prevalence the case of GAD. estimates averaging approximately 1% and 12-month A related issue is that considerable evidence exists prevalence averaging 0.6% [26]. Obsessive compul- for clinically significant sub threshold manifestations sive disorder (OCD) is usually found to be the least of many mental disorders that are much more pre- common anxiety disorder, with lifetime prevalence valent than the disorders themselves. Unlike the situa- typically less than 2% and 12-month prevalence of tions with PTSD and GAD, the more general issue is approximately 1% [27]. that many mental disorders appear to be extremes on Controversy exists regarding the appropriate diag- underlying dimensions rather than categorical mani- nostic thresholds for some mental disorders, such as festations that are qualitatively distinct from these post-traumatic stress disorder (PTSD) [28] and gen- distributions [30]. For example, even though OCD is eralized anxiety disorder (GAD) [29]. In both these almost always estimated to be fairly rare in general cases, good evidence exists from epidemiological population surveys, subthreshold manifestations of surveys that one or more particular diagnostic criteria OCD, some of them appearing to be clinically sig- define a much more restrictive set of cases than the nificant, are fairly common [31]. The same is true for other criteria, calling into question the wisdom of bipolar spectrum disorder, where even though the including the restrictive criteria. In the case of GAD, lifetime prevalence of BP-I is estimated to be only for example, the original diagnostic criteria in DSM- about 0.8–1.5%, the combined prevalence of BP-I, III, which required a minimum duration of one month, BP-II and clinically significant subthreshold BPD is was changed in the DSM-III-R and DSM-IV to six likely in the range 4–6% [32]. However, as community months in an effort to reduce the high comorbidity epidemiological surveys have for the most part found in clinical samples (but not, as it was subse- not explored these subthreshold manifestations sys- quently discovered, in community samples) between tematically, we do not currently have good estimates GAD and MDD. The ICD-10 criteria split this dif- of the proportion of the population that would meet ference by requiring a minimum duration of ‘several’ criteria for one or more anxiety and mood spectrum months. Considerable epidemiological research has disorders.

THE SOCIAL EPIDEMIOLOGY OF MENTAL DISORDER 95 8.3 AGE-OF-ONSET DISTRIBUTIONS While the results reviewed in the last section docu- separation anxiety disorder (SAD) have very early ment that mental disorders are highly prevalent, it is AOO distributions. In the WMH data, these disorders also important to examine age-of-onset (AOO) dis- had a median AOO in the range 7–14 and an IQR tributions (the distribution, across people with a between 4 and 20 years. Country-specific AOO lifetime history of a disorder, of when the disorder curves are strikingly consistent. No significant asso- first occurred) for three reasons. The first reason is ciations were observed between the position of the that commonly occurring lifetime disorders might curve and either prevalence of the disorders in the have much less effect on the overall lives of the country or the county’s level of economic develop- people who experience them if they only occur late ment. Behavioural disorders also have early AOO in life. All else equal, earlier-onset disorders are distributions, with the vast majority of onsets in more burdensome unless they not only begin but childhood or adolescence. The other common anxi- also remit early in life. Second, and related, AOO ety disorders (panic disorder, generalized anxiety information allows us to distinguish between lifetime disorder, and post-traumatic stress disorder) have prevalence (the proportion of the population who had considerably later AOO distributions, with medians a disorder at some time in their life up to their age at in the range 25–53. As with earlier-onset anxiety interview) and projected lifetime risk (the estimated disorders, no significant associations were observed proportion of the population who will have the between the position of the curve and either preva- disorder by the end of their life). The estimates lence of the disorders in the country or the country’s reported above were for lifetime prevalence rather level of economic development. The mood disorder than for lifetime risk. Lifetime risk cannot be esti- AOO distributions in the WMH surveys are quite mated directly from community surveys because similartothosefor thelater-onset anxiety disorders, respondents differ in age and, therefore, number of with consistently low prevalence until the early years at risk. Projections of estimated future risk can teens, followed by a roughly linear increase through be made from AOO distributions, however, using late middle age and a declining increase thereafter. standard statistical methods. Third, an understanding The median AOO of mood disorders has a very wide of AOO is important for targeting research on pre- range across countries (ages 25–45), but again with- vention of mental disorders, early intervention with out any consistent association between the shape of prodromal or incipient mental disorders, and primary the curve and either prevalence of the disorders in the prevention of secondary disorders. In the absence of country or the country’s level of economic develop- AOO information, we would have no way of know- ment. Although less data are available on the AOO ing the appropriate age groups to target in these distributions of NAP, available evidence suggests interventions. that median AOO is in the range between the late AlthoughAOO is routinely assessed retrospectively teens and mid-twenties. in community surveys, only a few reports have been A note of caution is needed in interpreting all these published over the years that describe AOO distribu- AOO data, as they are largely based on retrospective tions based on these data. These studies are reviewed lifetime recall and thus are subject to recall bias as elsewhere [33]. Recently, though, comprehensive well as to selection bias. Indeed, somewhat earlier AOO data were published from the WMH sur- AOO estimates are generally found in prospective veys [17]. These data are remarkably consistent across longitudinal studies rather than in the analysis of countries as well as consistent with the AOO data retrospective AOO reports [34]. Nonetheless, these reported in previous studies in showing distinct prospective data are for the most part consistent with AOO patterns for the different mental disorders. Some the AOO distributions seen in the retrospective WMH anxiety disorders, most notably the phobias and data.

96 COMPONENTS OF THE SOCIAL WORLD We noted above that AOO distributions can be used class-specific proportional increase in projected life- to generate projections of lifetime risk: the proportion time risk versus prevalence was associated with mood of the population that will experience a given disorder disorders. This is because of the comparatively late at some time in their lives. These estimates will onset of mood disorders compared with most other necessarily be higher than estimates of lifetime pre- mental disorders. The proportional increases were valence, as they include not only all lifetime-to-date comparable for GAD and PTSD, but very low for cases but also some number of anticipated future other anxiety disorders, again reflecting the typically onsets. The issue of interest is the size of this projected later ages of onset of the former than the latter. number of future onsets. Estimates of projected life- It is noteworthy that most fear-related anxiety dis- timeriskofanyDSM-IVdisorderintheWMHdataare orders and most impulsive behavioural disorders have roughly one-third higher than the estimates of lifetime considerably earlier AOO distributions than mood prevalence-to-date [17]. This means that 3–4 people disorders or NAP. This early onset, coupled with the in the populations of these countries are likely to fact that significant associations exist between these develop a first mental disorder at some time in the early-onset disorders and the subsequent first onset of future for every ten people who already had a disorder. other mental and substance use disorders, has led The highest risk-to-prevalence ratios (57–69%) were some commentators to suggest that aggressive treat- found in countries exposed to sectarian violence, such ment of child–adolescent anxiety and behavioural as Israel, Nigeria and South Africa. Excluding these disorders might be effective in preventing the onset three, no strong difference in risk-to-prevalence of the secondary mental and substance disorders that ratios was found for less developed countries versus are associated with the vast majority of serious mental developed countries. Not surprisingly, the highest illness [35]. 8.4 COURSE OF ILLNESS The course of illness, like AOO, has been much less is because as many as one-third of lifetime cases have well studied in epidemiological surveys than has only a single episode and another one-third have prevalence. Indeed, few direct questions about the episodes that recur only rarely. course of illness were included in most community More detailed analyses of these ratios could be epidemiological surveys of mental disorders prior to carried out by breaking them down separately for the WMH surveys. However, the fact that mental subsamples defined by age at interview or by time disorders are seen as often being quite persistent since first onset, but we are unaware of any published adds to the judgement that they have such adverse research that has reported such analyses. Our own effects. Objective assessment of this persistence can preliminary analyses of this sort in the WMH data be obtained by comparing estimates of recent pre- suggest, however, that although 12-month to lifetime valence (variously reported for the year, six months prevalence ratios for most mental disorders decline or one month before interview) with estimates of with increasing age, this decline is fairly modest after lifetime prevalence. The 12-month to lifetime preva- mid-life, suggesting that mental disorders are often lence ratios for anxiety and mood disorders are typi- quite persistent over the entire life course. The few cally in the range 0.4–0.6, with the ratio always long-term longitudinal studies that exist in represen- somewhat higher for anxiety disorders than mood tative samples yield results consistent with this con- disorders. These ratios are lower for behavioural clusion and suggest that this persistence is due to a disorders and NAP. In the case of behavioural dis- recurrent intermittent course that often features wax- orders, this is because these disorders often remit in ing and waning of episodes of different comorbid adolescence or early adulthood. In the case of NAP, it disorders [36,37].

THE SOCIAL EPIDEMIOLOGY OF MENTAL DISORDER 97 8.5 COMORBIDITY Comorbidity among mental disorders is quite com- distinct class of disorders in the DSM and ICD systems mon, with up to half of people with any lifetime and to suggest that a more useful organizing disorder meeting criteria for two or more such disor- scheme in the upcoming DSM-V and ICD-11 revi- ders [38]. Factor analytic studies of diagnostic comor- sions would be one that distinguished between fear bidity consistently document separate internalizing disorders and distress disorders, with the latter includ- and externalizing factors in which anxiety and mood ing not only GAD and possibly PTSD but also uni- disordershavehighfactorloadingsontheinternalizing polar depression and dysthymia [40]. The argument dimension and behavioural and substance disorders for a class of fear disorders has some support in havehighloadingsontheexternalizingfactor[39].The neurobiological research based on investigation of internalizing dimension, furthermore, sometimes is fear brain circuitry [41]. The possibility also exists found to have secondary dimensions that distinguish that future research might lead to OCD being distin- between fear disorders (panic, phobia) and distress guished from either fear disorders or distress disorders disorders (depression, dysthymia, GAD) [40]. as part of a spectrum of impulse-control disorders These results have recently been used to call into based both on evidence of differential comorbidity question the codification of anxiety disorders as a and differences in brain circuitry [42]. 8.6 THE SOCIETAL COSTS OF MENTAL DISORDERS Early-onset mental disorders are significant predictors expenditures, impaired functioning and reduced long- of the subsequent onset and persistence of other evity, but most of this work has been done in the mental and substance use disorders as well as of a US [48,49]. The magnitude of the cost estimates in wide range of physical disorders [43,44]. It is impor- these studies is staggering. For example, Greenberg tant to note that these predictive associations are part et al. [49] estimated that the annual total societal costs of a larger pattern of associations that has been of active anxiety disorders in the US over the decade documented between early-onset mental disorders of the 1990s exceeded $42 billion. This estimate and a much wider array of adverse life course out- excludes the indirect costs of early-onset anxiety comes that might be conceptualized as societal costs disorders through adverse life course outcomes (e.g. of these disorders, including reduced educational the documented effects of child–adolescent anxiety attainment, early marriage, marital instability, and disorders in predicting low educational attainment and low occupational and financial status [45–47]. It is consequent long-term effects on lower income) and unclear if these associations are causal, i.e. if inter- through increased risk of other disorders (e.g. anxiety ventions to treat early-onset mental disorders would disorders predicting the subsequent onset of cardio- prevent the subsequent onset of the adverse outcomes vascular disorder). with which they are associated. As a result, it is not Although comparable studies of the societal costs possibleto state unequivocally that these outcomes are of mental disorders have been carried out in few other consequences of mental disorders. It would be very countries, a recent study of the comparative impair- valuable, however, from a public health perspective to ments in role functioning caused by mental disorders have long-term evidence to evaluate this issue based and commonly occurring chronic physical disorders on experimental treatment effectiveness studies. in the WMH surveys documented that common A considerable amount of research has been carried mental disorders have substantial adverse effects on out to quantify the magnitude of the short-term soci- functioning in many countries around the world [9]. etal costs of mental disorders in terms of health-care This analysis made use of the fact that a set of

98 COMPONENTS OF THE SOCIAL WORLD commonly occurring physical disorders were assessed 91 comparisons in developing countries. Nearly all of in the WMH surveys with a standard chronic disorders these higher mental-than-physical impairment ratings checklist. Respondents with the ten most commonly were statistically significant at the 0.05 level and held reported such disorders were asked to report the extent in within-person comparisons (i.e. comparing the to which each such disorder interfered with their reported impairments associated with a particular ability to carry out their daily activities in both pro- mental–physical disorder pair in the subsample of ductive roles (i.e. job, school, housework) and social respondents who had both disorders). Comparable roles (i.e. social and personal life). The same questions results were obtained for severe disability ratings. about disorder-specific role impairments were also Furthermore, a similar pattern held when treated asked of respondents with each of the mental disorders physical disorders were compared with all (i.e. treated assessed in the surveys, the ten most commonly or not) mental disorders to address the concern that the occurring of which were compared to the ten most more superficial assessment of physical than mental common physical disorders. disorders might have led to the inclusion of subthres- Of the 100 logically possible pairwise disorder- hold cases of physical disorders that might have low specific mental–physical comparisons, mean impair- disability. Mental disorders were found in the same ment ratings were higher for the mental than physical analyses to be associated with more days out of role disorder in 91 comparisons in developed and also for than physical disorders. 8.7 TREATMENT OF MENTAL DISORDERS Although a detailed analysis of epidemiological data makes treatment cost-effective. Experimental epide- on the treatment of mental disorders is beyond the miological studies are needed to answer this question. scope of this chapter, we briefly note that epidemio- A study of this sort focused on screening and treating logical studies consistently find that only a minority of depression among working people was carried out in people with 12-month mental disorders received treat- the US in conjunction with the US WMH survey [52]. ment for these disorders within the year of interview, The intervention group was found to have signifi- even in developed countries [10]. The situation is, as cantly higher job retention and hours worked than one might guess, dramatically worse in developing controls at both six and twelve months after the countries, where only a small minority of people with intervention. The financial benefits of these interven- serious mental disorders receive treatment [50]. Ra- tion effects (in terms of hiring and training costs, tionality in the allocation of treatment resources is disability payment and salaries paid for sickness indicated by the fact that both the probability of absence days) were substantially higher than the costs receiving treatment and, among patients, the intensity of treatment, documenting that the intervention was a of treatment are strongly related to seriousness of human capital investment for employers. Replications illness. However, adequacy of treatment, as indicated of this intervention experiment are currently under- by concordance of treatment (in terms of number of way in other WMH countries, including Australia and visits and types of medication used) with published Japan. Extensions of the intervention guided by WMH treatment guidelines, is low even for serious cases in results documenting significant workplace effects of treatment in developed countries [51]. bipolar depression and adult ADHD are also under- Given the enormous societal costs of mental dis- way. Ongoing analyses of the WMH data are also orders, a question can be raised whether expansion of being used to search for other intervention targets detection, treatment and treatment quality improve- that can be used to evaluate the effects of treatment ment initiatives might be able to reduce the adverse in reducing the burdens associated with mental societal effects of mental disorders to an extent that disorders.

THE SOCIAL EPIDEMIOLOGY OF MENTAL DISORDER 99 8.8 CONCLUSIONS The epidemiological results reviewed here document disorders receives treatment in most countries and that mental disorders are commonly occurring, often even fewer receive high-quality treatment. This situa- have an early age of onset and often are associated tion has to change. A good argument could be made with significant adverse societal costs. The important based on data from controlled studies of treatment public health question raised by these results is effectiveness that an expansion of treatment would whether these societal costs can be reversed with be a human capital investment opportunity from a best-practices treatment. The preliminary results societal perspective as well as from an employer reported in the last paragraph argue much more per- perspective. Further epidemiological studies are suasively than the naturalistic survey findings that needed to refine naturalistic analyses of the adverse mental disorders are actual causes rather than merely effects of mental disorders in an effort to target correlates of impaired role functioning. Based on experimental interventions that can demonstrate the these results, it appears that mental disorders are not value of expanded treatment to address the enormous only common but also consequential from a societal global burden of mental disorders, with a special perspective throughout the world. Yet, as noted in the emphasis aimed at increasing the scant data that exist last section, only a minority of people with mental on this issue in the developing world. ACKNOWLEDGEMENTS Preparation of this chapter was supported, in part, Interview Schedule: its history, characteristics and by the following grants from the US Public validity. Archives of General Psychiatry, 38, 381–389. Health Service: U01MH060220, R01DA012058, 3. Robins, L. N. and Regier, D. A. (eds) (1991) Psychiatric R01MH070884 and R01DA016558; and by Robert Disorders in America: The Epidemiologic Catchment Wood Johnson Foundation Grant 044780. Portions of Area Study, The Free Press, New York. the chapter appeared previously in Kessler, R. C., 4. Horwath, E. and Weissman, M. M. (2000) The Aguilar-Gaxiola, S., Alonso, J., Chatterji, S., Lee, epidemiology and cross-national presentation of ¨ S. and Ust€ un, T. B. (2009) The WHO World Mental obsessive-compulsive disorder. Psychiatric Clinics of North America, 23, 493–507. Health (WMH) surveys, Die Psychiatrie, 6, 5–9. Permission to use the material has been given by 5. Robins, L. N., Wing, J., Wittchen, H. U. et al. (1988) The composite international diagnostic interview. An Schattauer GmbH, Stuttgart, Germany. The views and epidemiologic instrument suitable for use in opinions expressed in this report are those of the conjunction with different diagnostic systems and in authors and should not be construed to represent the different cultures. Archives of General Psychiatry, 45, views of any of the sponsoring organizations, agencies 1069–1077. or US Government. 6. Wittchen, H. U. (1994) Reliability and validity studies of the WHO-Composite International Diagnostic Interview (CIDI): a critical review. Journal of REFERENCES Psychiatric Research, 28, 57–84. 7. WHO International Consortium in Psychiatric 1. Kessler, R. C. and U ¨ st€ un, T. B. (eds) (2008) The WHO Epidemiology (2000) Cross-national comparisons of World Mental Health Surveys: Global Perspectives on the prevalences and correlates of mental disorders. the Epidemiology of Mental Disorders, Cambridge Bulletin of the World Health Organization, 78, University Press, New York. 413–426. 2. Robins, L. N., Helzer, J. E., Croughan, J. L. et al. (1981) 8. Nock, M. K., Borges, G., Bromet, E. J. et al. (2008) National Institute of Mental Health Diagnostic Cross-national prevalence and risk factors for suicidal

100 COMPONENTS OF THE SOCIAL WORLD ideation, plans and attempts. British Journal of the WMH Surveys, in The WHO World Mental Health Psychiatry, 192, 98–105. Surveys: Global Perspectives on the Epidemiology of 9. Ormel, J., Petukhova, M., Chatterji, S. et al. (2008) Mental Disorders (eds R. C. Kessler and T. B. U ¨ st€ un), Disability and treatment of specific mental and physical Cambridge University Press, New York. disorders across the world. British Journal of 19. Saha, S., Chant, D. and McGrath, J. (2008) Meta- Psychiatry, 192, 368–375. analyses of the incidence and prevalence of 10. Wang, P. S., Aguilar-Gaxiola, S.,Alonso, J. et al. (2007) schizophrenia: conceptual and methodological issues. Use of mental health services for anxiety, mood, and International Journal of Methods in Psychiatric substance disorders in 17 countries in the WHO World Research, 17, 55–61. Mental Health surveys. Lancet, 370, 841–850. 20. Kessler, R. C., Birnbaum, H., Demler, O. et al. (2005) 11. Haro, J. M., Arbabzadeh-Bouchez, S., Brugha, T. S. The prevalence and correlates of nonaffective psychosis et al. (2006) Concordance of the Composite in the National Comorbidity Survey Replication International Diagnostic Interview version 3.0 (CIDI (NCS-R). Biological Psychiatry, 58, 668–676. 3.0) with standardised clinical assessments in the WHO 21. Perala, J., Suvisaari, J., Saarni, S. I. et al. (2007) World Mental Health surveys. International Journal of Lifetime prevalence of psychotic and bipolar I Methods in Psychiatric Research, 15, 167–180. disorders in a general population. Archives of 12. Silove, D., Bateman, C. R., Brooks, R. T. et al. (2008) General Psychiatry, 64, 19–28. Estimating clinically relevant mental disorders in a 22. Saha, S., Chant, D., Welham, J. et al. (2005) A rural and an urban setting in postconflict Timor systematic review of the prevalence of schizophrenia. Leste. Archives of General Psychiatry, 65, 1205–1212. PLoS Medicine, 2, e141. 13. Kessler, R. C., Abelson, J., Demler, O. et al. (2004) 23. Silverman, W. K. and Moreno, J. (2005) Specific Clinical calibration of DSM-IV diagnoses in the World phobia. Child and Adolescent Psychiatric Clinics of Mental Health (WMH) version of the World Health North America, 14, 819–843, ix–x. Organization (WHO) Composite International 24. Judd, L. L. and Akiskal, H. S. (2000) Delineating the Diagnostic Interview (WMHCIDI). International longitudinal structure of depressive illness: beyond Journal of Methods in Psychiatric Research, 13, clinical subtypes and duration thresholds. 122–139. Pharmacopsychiatry, 33, 3–7. 14. Somers, J. M., Goldner, E. M., Waraich, P. et al. (2006) 25. Furmark, T. (2002) Social phobia: overview of Prevalence and incidence studies of anxiety disorders: a community surveys. Acta Psychiatrica Scandinavica, systematic review of the literature. Canadian Journal of 105, 84–93. Psychiatry, 51, 100–113. 26. Benazzi, F. (2007) Bipolar II disorder: epidemiology, 15. Waraich, P., Goldner, E. M., Somers, J. M. et al. (2004) diagnosis and management. CNS Drugs, 21, 727–740. Prevalence and incidence studies of mood disorders: a systematic review of the literature. Canadian Journal of 27. Fontenelle, L. F., Mendlowicz, M. V. and Versiani, M. Psychiatry, 49, 124–138. (2006) The descriptive epidemiology of obsessive- compulsive disorder. Progress in Neuropsychophar- 16. Wittchen, H. U. and Jacobi, F. (2005) Size and macology and Biological Psychiatry, 30, 327–337. burden of mental disorders in Europe – a critical review and appraisal of 27 studies. European 28. Mylle, J. and Maes, M. (2004) Partial posttraumatic Neuropsychopharmacology, 15, 357–376. stress disorder revisited. Journal of Affective Disorders, 17. Kessler, R. C., Aguilar-Gaxiola, S., Alonso, J. et al. 78, 37–48. (2008) Lifetime prevalence and age of onset 29. Ruscio, A. M., Chiu, W. T., Roy-Byrne, P. et al. (2007) distributions of mental disorders in the World Mental Broadening the definition of generalised anxiety Health Survey Initiative, in The WHO World disorder: effects on prevalence and associations with Mental Health Surveys: Global Perspectives on the other disorders in the National Comorbidity Survey Epidemiology of Mental Disorders (eds R. C. Kessler Replication. Journal of Anxiety Disorders, 21, 662–676. and T. B. U ¨ st€ un), Cambridge University Press, New 30. Brown, T. A. and Barlow, D. H. (2005) Dimensional York, pp. 511–521. versus categorical classification of mental disorders in 18. Kessler, R. C., Aguilar-Gaxiola, S., Alonso, J. et al. the fifth edition of the Diagnostic and Statistical (2008) Prevalence and severity of mental disorders in Manual of Mental Disorders and beyond: comment

THE SOCIAL EPIDEMIOLOGY OF MENTAL DISORDER 101 on the special section. Journal of Abnormal 42. Whiteside,S. P., Port, J. D. and Abramowitz, J. S. (2004) Psychology, 114, 551–556. A meta-analysis of functional neuroimaging in 31. Matsunaga, H. and Seedat, S. (2007) Obsessive- obsessive-compulsive disorder. Psychiatry Research, compulsive spectrum disorders: cross-national and 132, 69–79. ethnic issues. CNS Spectrums, 12, 392–400. 43. He, Y., Zhang, M., Lin, E. H. et al. (2008) Mental 32. Skeppar, P. and Adolfsson, R. (2006) Bipolar II and the disorders among persons with arthritis: results from the bipolar spectrum. Nordic Journal of Psychiatry, 60, World Mental Health surveys. Psychological Medicine, 7–26. 38, 1639–1650. 33. Kessler, R. C., Amminger, G. P., Aguilar-Gaxiola, S. 44. Ormel, J., Von Korff, M., Burger, H. et al. (2007) et al. (2007) Age of onset of mental disorders: a review Mental disorders among persons with heart disease – of recent literature. Current Opinion in Psychiatry, 20, results from World Mental Health surveys. General 359–364. Hospital Psychiatry, 29, 325–334. 34. Wittchen, H.-U., Lieb, R., Schuster, P. et al. (1999) 45. Kessler, R. C., Berglund,P. A.,Foster, C. L.et al. (1997) When is onset? Investigations into early developmental Social consequences of psychiatric disorders, II: stages of anxiety and depressive disorders, in teenage parenthood. American Journal of Psychiatry, Childhood Onset of ‘Adult’ Psychopathology: 154, 1405–1411. Clinical and Research Advances (ed. J. L. Rapoport), 46. Kessler, R. C., Foster, C. L., Saunders, W. B. et al. American Psychiatric Press, Washington, DC, (1995) Social consequences of psychiatric disorders, I: pp. 259–302. educational attainment. American Journal of 35. Kendall, P. C. and Kessler, R. C. (2002) The impact of Psychiatry, 152, 1026–1032. childhood psychopathology interventions on subsequent 47. Kessler, R. C., Walters, E. E. and Forthofer, M. S. substance abuse: policy implications, comments, and (1998) The social consequences of psychiatric recommendations. Journal of Consulting and Clinical disorders, III: probability of marital stability. Psychology, 70, 1303–1306. American Journal of Psychiatry, 155, 1092–1096. 36. Angst, J. and Vollrath, M. (1991) The natural history of 48. Greenberg, P. E. and Birnbaum, H. G. (2005) The anxiety disorders. Acta Psychiatrica Scandinavica, 84, economic burden of depression in the US: societal 446–452. and patient perspectives. Expert Opinion in 37. Bruce, S. E., Yonkers, K. A., Otto, M. W. et al. (2005) Pharmacotherapy, 6, 369–376. Influence of psychiatric comorbidity on recovery and 49. Greenberg, P. E., Sisitsky, T., Kessler, R. C. et al. recurrence in generalised anxiety disorder, social (1999) The economic burden of anxiety disorders phobia, and panic disorder: a 12-year prospective in the 1990s. Journal of Clinical Psychiatry, 60, study. American Journal of Psychiatry, 162, 427–435. 1179–1187. 50. Wang,P. S.,Aguilar-Gaxiola, S., Alonso, J.et al. (2008) 38. Kessler, R. C. (1995) Epidemiology of psychiatric Recent treatment of mental disorders in the World comorbidity, in Textbook in Psychiatric Epidemiology Mental Health Survey Initiative, in The WHO World (eds M. T. Tsuang, M. Tohen and G. E. P. Zahner), John Mental Health Surveys: Global Perspectives on the Wiley & Sons, Inc., New York, pp. 179–197. Epidemiology of Mental Disorders, Cambridge 39. Kendler, K. S., Prescott, C. A., Myers, J. et al. (2003) University Press, New York, pp. 541–554. The structure of genetic and environmental risk factors 51. Wang, P. S., Lane, M., Olfson, M. et al. (2005) Twelve- for common psychiatric and substance use disorders in month use of mental health services in the United men and women. Archives of General Psychiatry, 60, States: results from the National Comorbidity Survey 929–937. Replication. Archives of General Psychiatry, 62, 40. Watson, D. (2005) Rethinking the mood and anxiety 629–640. disorders: a quantitative hierarchical model forDSM-V. 52. Wang, P. S., Simon, G. E., Avorn, J. et al. (2007) Journal of Abnormal Psychology, 114, 522–536. Telephone screening, outreach, and care management 41. Knight,D.C.,Nguyen,H.T.and Bandettini,P.A.(2005) for depressed workers and impact on clinical and work The role of the human amygdala in the production productivity outcomes: a randomised controlled trial. of conditioned fear responses. Neuroimage, 26, Journal of the American Medical Association, 298, 1193–1200. 1401–1411.



9 Families and psychosis Juliana Onwumere, Ben Smith and Elizabeth Kuipers Institute of Psychiatry, King's College London, London, UK 9.1 INTRODUCTION Significantchangestotheprovisionofcareandtreatment These discussions have tended to shift between forpeoplewithlong-termmentalhealthproblemshasled examining the impact of mental health problems on toamuchgreaterfocusonfamilies.Amongworkingage carers (e.g. carer burden) or the impact of carers on adults (i.e. 18–65 years), the literature on mental health illness course and patient outcomes (e.g. expressed problems and families has tended to concentrate on emotion). This chapter reviews evidence across psychosisandthiswillbereflectedinthecurrentchapter. both areas and examines the efficacy of family work. However, it is also the case that there is increasing In line with recent neurobiological models of psy- evidence that issues related to mental health problems chosis (e.g. see Reference [6]), we continue to hold within families can extend across different diagnostic the view that families do not cause schizophrenia. groups including dementias, affective and eating disor- Families, however, do play an instrumental role in ders [1–4] and severe physical health conditions [5]. improving our understanding of psychosis and can For decades, the ‘family’ has remained at the heart facilitate optimal outcomes, for patients and carers of discussions on schizophrenia spectrum disorders. themselves. 9.2 FAMILIES AS A RESOURCE High levels of social disadvantage are found in indi- duringthe early phasesofthe illness [13,14] and among viduals with psychosis [7]. Moreover, social dysfunc- minority ethnic groups [15]. The large numbers of tion is a key characteristic of psychosis that is present individuals livingwithresidualsymptomscoupledwith during the early phases of the illness [8]. Many patients high rates of relapse and problematic recovery means tend to have smaller social networks [9,10]; relatively that the role of an informal carer can often represent a few will have partnerships or become parents [11]. long-term commitment [16]. However, a large majority live with and/or maintain Following the onset of psychosis, which typically close contact with informal carers [12], particularly occurs during late adolescence, carers can encounter Principles of Social Psychiatry, second edition Edited by Craig Morgan and Dinesh Bhugra Ó 2010 John Wiley & Sons, Ltd.

104 COMPONENTS OF THE SOCIAL WORLD significant role changes that they often feel ill- phases of the illness [27,28]. Carers can provide equipped to manage [17]. The evidence suggests that different types of support, serve as advocates and carers tend to be first-degree relatives (e.g. parents) or respond to areas of unmet need. Across the mental the partner of the identified patient [18]. They are also and physical health literature, the evidence also sug- likely to be middleaged (i.e.over 50 years old) [19,20]. gests that patients tend to obtain better outcomes (e.g. Since caregiving is usually a female-dominated acti- improved quality of life, decreased symptoms and vity [21], and patients are more frequently male, carers length of inpatient days) when they have of individuals with psychosis in clinical services carers [11,29]. For example, Schofield et al. [25] ob- and research studies are usually mothers, who are served that patients with schizophrenia and substance providing care for their adult male sons [22,23]. misuse problems, who also had carers, spent signifi- Although historically family carers were incor- cantly fewer days as an inpatient compared with rectly identified as having a causal role in psychosis, matched patient groups without carer contact. Recent more recently the role of a carer is recognized as being findings from a study of individuals with first-episode critical in maintaining thewell-being of patients living psychosis suggested that lower levels of positive in the community [24,25]. Carers can play a signifi- symptoms were linked to supportive social environ- cant role in improving a patient’s treatment adher- ments. The support received specifically from ence [26], monitoring their mental state and identify- family members was related to reduced hospital ing incipient signs of deterioration, and accessing admissions in the three years following the initial appropriate services, particularly during the early episode [30]. 9.3 THE IMPACT OF CAREGIVING Research evidence spanning five decades attests to the experience burden in their caregiving role [41]. Carer negative impact of caring for an individual with burden is universal and has been observed in diverse psychosis on a carer’s physical and emotional ethnic and cultural groups drawn from Europe, Africa, health [31–33]. As a group, carers tend to report poor America, the Middle East and Asia [22,23,42–44]. It physical health and they are more likely to experience has been recorded in different carer subgroups includ- sleep disturbances [34]. Like patient groups, carers ing spouses and siblings [45,46]. Evidence of carer also have an increased risk of mortality and medical burden has also been confirmed in a broad range of hospitalizations [35,36]. mental health disorders including bipolar affective Treudley [37] provided some of the earliest data disorder [47], depression [48] and obsessive-compul- documenting the negative impact of caregiving on sive disorders [49]. carer mental health and quality of life. The negative The negative impact of psychosis on carer well- impact was commonly described as carer or family being is also commonly measured and described in ‘burden’. Carer burden was measured along two di- terms of stress levels and mood disturbance. There are mensions: objectiveand subjectiveburdens. Objective robust contemporary findings that confirm signifi- burden related to the tangible and observable effects of cantly higher rates of depression, anxiety and general the patient’s mental health problems on the family, distress in carers [50]. Moreover, levels of burn-out such as disrupted domestic routines, and constraints among carers are usually high (e.g. emotional exhaus- on carers’ social and leisure activities [38]. In contrast, tion) and not dissimilar to levels reported by profes- subjective burden was related to a carer’s negative sional psychiatric personnel [51]. Carer distress and appraisal of their circumstances. It reflected the ne- burden can vary during different illness phases but can gative psychological sequelae of the illness on the be enduring and continue beyond the end of the carer, such as feelings of loss, guilt, shame and caregiving relationship [52]. Higher levels of distress anger [39,40]. Data drawn from more recent investi- have also been observed among carers of patients who gations of carers suggest as many as 80% of carers have recently become unwell and those who have

FAMILIES AND PSYCHOSIS 105 recently been admitted to hospital [17,53]. Similar to be affected by mental health problems. Therefore, patient groups, carers of patients with psychosis are carers may be fulfilling a caring role for more than often very socially isolated and stigmatized [54,55]. one family member and/or may have mental health Many carers have been subject to episodes of verbal problems of their own. It seems important to recognize and physical violence from the individuals they care that most carers will usually take on their new role in for [56,57], and recent evidence suggests that as many the absence of specialist knowledge and training or as one-third of carers of patients with psychosis additional resources. The role and its demands are meet ‘caseness’ criteria for post-traumatic stress often subsumed within their regular duties, which can disorder [58]. take their toll on carers’ emotional and physical well- The incidence and prevalence rates of psychosis being. It is therefore not uncommon for carers to and other mental health problems mean that it is not express ambivalence in their feelings towards the uncommon for more than one member of a family to patient, their illness and their carer role. 9.4 CARER BURDEN AND PATIENT SYMPTOMS There have been many empirical investigations of pa- studies have identified links with both positive and tients’ clinical characteristics to determine whether negative symptoms [13,62]. Overall, however, carers specific symptoms are linked to carer burden and dis- tendtoreportgreaterdistressandburdenwhencaringfor tress; thus far, findings remain equivocal. Difficulties in individuals with poorer social and role functioning and the measures employed and samples tapped have con- sometimes embarrassing and disruptive behaviours. tributed to the inconsistency [59]. Thus, positive symp- These are often areas of difficulty that are not perceived toms (e.g. auditory verbal hallucinations) have been as directly linked to symptoms or illness. In a recent linked with significantly higher levels of carer distress study that involved a two-year follow-up of patients and burden [60]. Likewise, negative symptoms (e.g. admitted for their first time, reductions in carer burden underactivity, social withdrawal) have also been asso- were associated with improvements with a patient’s ciated with carer burden and distress [43,61]. Some psychosocial functioning [63]. 9.5 CARER COPING STYLES AND CAREGIVING Levels of carer burden and stress have also been linked psychosis recruited from five European countries. to coping styles. Carers of patients with psychosis During a 12-month follow-up, levels of objective and tend to use different coping strategies (e.g. ignoring, subjective burden decreased in carers who employed avoidance) that are often employed through a process fewer emotion focused coping strategies (e.g. avoid- of trial and error [33,64,65]. Historically, the coping ance of the patient) and had higher levels of practical literature has made distinctions between ‘emotion’ and support from their social network. Scazufca and Kui- ‘problem’ focused coping [66]. The former reflects pers [68] reported that higher rates of carer burden coping designed to diminish the negative emotional were observed in the period prior to and after inpatient impact of the stressor using strategies such as avoi- admission for carers who used more avoidant styles of dance and denial. In contrast, problem focused coping coping. Coping research also suggests that carers describes the direct action an individual adopts to report lower burden when they employ coping styles change the situation, such as seeking social support characterized by ignoring and acceptance (e.g. indif- and problem solving. ference) [64]. Recent evidence from an exploratory There is growing evidence of links between carer study suggests that parents with a stronger history of burden, distress and emotion focused coping. Maglia- psychosis in the family, defined in terms of having no et al. [67] studied 159 caregivers of patients with more than one first-degree relative with psychosis,

106 COMPONENTS OF THE SOCIAL WORLD also tended to display less effective coping styles [69]. toms and levels of burden, have been linked to reduced It is also of note that carers’ positive evaluations about distress levels and positive attitudes towards their coping efficacy, independent of patient symp- patients [70]. 9.6 EXPRESSED EMOTION AND FAMILIES The relationship between psychosis and carers has carer behaviours, which include self-sacrifice, pre- been studied extensively through the concept of occupation with the patient’s illness and overprotec- expressed emotion (EE). EE is said to represent a tion. Positive remarks are unambiguous positive carer’s evaluation of their relationship quality with the statements about the patient’s personality, skills and patient [71]. The study of EE has a long history in adult attributes. Warmth reflects carer expressions of em- mental health and commenced with the work of pathy, sympathy, concern and understanding towards George Brown and colleagues in the 1950s and the patient. EE ratings are based on content and 1960s (e.g. see References [72] to [74]). Their inves- prosodic aspects of speech, including emphasis and tigations of patient outcomes following community tone. Therefore, ratings are derived from what is discharge showed that patients experienced more said and the way it is said [78]. The computation of relapses and readmissions when they returned to live high and low EE is based on scores from the in large hostels or with parents and spouses [72,73]. criticism, hostility and EOI subscales only, since Moreover, higher levels of relapse were observed for these scales have historically proved to be the most patients who returned to households with carers with predictive of patient outcomes [79,80]. High EE ‘high emotional involvement’ rather than ‘low emo- ratings are assigned when carers score above-thresh- tional involvement’. Patient relapse was indepen- old levels of criticism, hostility and EOI on the CFI. dently predicted by high levels of carer emotional While EE ratings are based on one person (the involvement. The term EE and the individual compo- person who spends the most time with the patient), nents of EE were developed in one of their later the final ratings are thought to reflect the emotional studies [75]. Brown and colleagues also developed climate of the whole household. the Camberwell Family Interview (CFI), a semi-struc- High EE is common among carers [81] and tured audiotaped interview, to evaluate the patient– observable during the early stages of the ill- caregiver relationship and derive EE ratings [76,77]. ness [54,82,83]. There is evidence to suggest that EE comprises five different subscales: criticism, cultural membership can influence carers’ emotional hostility, emotional overinvolvement, warmth and reactions and understanding of patients’ symp- positive comments. Criticism reflects an unfavour- toms [84,85]. Although rates do vary across studies, able remark and/or expression about the patients’ high EE has been observed in carers from diverse behaviour or personality. Hostility refers to extreme racial and cultural groups, including those drawn from aspects of criticism and is most commonly voiced as Asia and the Middle East [86–88]. EE levels are not a rejecting remark or negative expression about the stable and changes can usually be observed over a nine patient in general rather than a specific behaviour. to 12 month period, particularly reductions from Emotional overinvolvement (EOI) reflects various original high EE levels [82,89]. 9.7 EXPRESSED EMOTION AND PATIENT OUTCOMES EE remains an important concept in mental health due predictive relationship between high carer EE and to its ability to predict patient outcomes. Although patient relapse [79,80]. Bebbington and Kuipers [79] relapse definitions have varied between studies there analysed 25 worldwide EE outcome studies and are significant and robust findings in support of a revealed a 50% relapse rate for patients living in high

FAMILIES AND PSYCHOSIS 107 EE households compared with 21% for patients in low studies, predominately from America, suggests that EE households. Overall, the risk of relapse for patients the links between carer EE and poor patient outcomes in high EE households was 2.5 times higher than risk may not be uniform across all cultural groups [96,97]. levels observed for patients from low EE households. For instance, Rosenfarb et al. [85] found that higher The association between high EE and relapse was levels of carer criticism and intrusive behaviours were similar for patients receiving regular medication and in fact associated with improved patient outcomes in for those who did not. While high levels of face-to- African American families compared with their white face contact between patients and high EE caregivers participants. increased the risk of relapse, high levels of contact In terms of specifying the exact mechanisms, it has with a low EE relative served as a protective factor. been argued that high EE leads to adverse patient Similar results were reported by Butzlaff and Hooley outcomes via increased patient arousal and affective [80], who conducted a meta-analysis of 27 EE- disturbances [98]. Kuipers et al. [18] observed, as outcome studies. They observed an effect size of hypothesized, that patients with carers with high EE 0.31, which was equivalent to a relapse rate of 65% also tended to have higher levels of anxiety and in high EE households and 35% in the low EE group. depression. Moreover, carers’ critical comments were The negative consequences of high EE on patient predictive of patient anxiety. Likewise, low carer self- outcomes are not limited to the short term (e.g. esteem and poorer coping were associated with cri- 12 months). For example, Marom et al. [90] con- tical comments. Barrowclough et al. [81] have also ducted aseven-year follow-upof carers of 108 patients found carer critical comments to be associated with with psychosis living in Israel; higher rates of read- negative self-evaluations in patients. missions, earlier first and second readmissions, and In addition to predictive links with relapse, EE longer inpatient stays were associated with carers attitudes remain of clinical interest as they provide reporting higher levels of criticism compared with an important pathway to understanding how carers low criticism carers. Likewise, Huguelet et al. [91] make sense of the patient’s illness. There are many reported higher levels of relapse and readmissions in studies that confirm the importance of these carer patients from high EE compared with low EE house- attributions. For instance, carer criticism and hostility holds as part of a five-year follow-up. The association tends to be associated with a carer’s perception that a between EE and patient relapse is also evident during patient is able to control their symptoms and pro- the early phases of psychosis, although results thus far blems. In contrast, carers with high levels of emotional have been equivocal. Some studies have reported a overinvolvement are more inclined to perceive significant association [92,93], while others have patients as having little or no control over their failed to do so [94]. Thus far, the predictive power symptoms and problems (see Reference [99] for a of EE ratings appears stronger for patients with a review). Overall, these results have led some resear- longer duration of illness [80]. The predictive link chers to suggest that high EE behaviours are probably between high EE and poorer patient outcomes has also best conceptualized as strategies employed by carers been demonstrated in diverse cultural groups [84,95]. to cope with and influence behavioural change among However, recent evidence from a small handful of patients. 9.8 POSITIVE FAMILY RELATIONSHIPS Data from a Finnish adoptee study confirmed that health carers who express positivity towards patients family environments defined as positive and healthy are more likely to assist patients in their recov- conferred protection against developing schizophre- ery [102]. However, the positive dimensions of EE nia spectrum disorders in children with high genetic (i.e. warmth and positive remarks) and positive family risk [100]. Similar findings have been reported by environments and relationships have received much Schiffman et al. [101]. We also know that mental less attention from researchers so far.

108 COMPONENTS OF THE SOCIAL WORLD From their review of EE studies, Bebbington and carer warmth was associated with illness course, but Kuipers [79] confirmed that good patient outcomes onlyintheMexicanAmericansample.Thus,following were predicted by carer warmth and positive com- admission,MexicanAmericanpatientswithpsychosis ments. Carer warmth has been associated with im- who returned to households that were high in warmth proved outcomes for patients with long-established were less likely to relapse, whereas warmth had no illness courses [79,103,104] and more recently for significant impact on illness course for the Anglo patientsintheprodromalandat-riskphases[105–107]. American patients. The authors conclude that warmth The evidence suggests that warmth also provides pro- appearedtoserveasaprotectivefactorfortheMexican tection against relapse particularly in some cultural American sample, while high levels of criticism con- groups [97,108] and is linked to fewer symptoms and tinued to serve as a risk factor for Anglo Ameri- improved social functioning in patients at risk of cans [97]. Recent findings from a study of patients at developing psychosis [105]. For example, Bertrando ultra high risk for psychosis and with recent onset et al. [103] conducted a nine-month follow-up of 42 symptoms revealed that carers’ positive remarks and Italian inpatients and their caregivers. Significantly warmth towards patients were positively linked with fewerreadmissionswerereportedamongpatientsfrom their own improved problem-solving skills. Carers ‘high warmth’ compared with ‘low warmth’ house- with good problem-solving skills were also likely to holds. Carer warmth was significantly associated with have patients with similar skills. Patients with good lower rates of relapse in both low and high EE house- problem-solving skills at baseline were more likely to holds.InarecentstudythatexaminedEEandrelapsein show improved social functioning and fewer positive Mexican American and Anglo American samples, symptoms at six-month follow-up [106]. 9.9 PATIENT PERCEPTIONS OF CAREGIVING RELATIONSHIPS Historically, EE studies have been limited to examin- was a significant predictor of a patient’s negative ing carer attitudes and their links to patient outcomes. perceptions. The significant associations were not In the last decade, however, there have been a growing accounted for by the patient’s mood or psychotic number of studies that have sought to explore how symptoms [112]. Similar findings were reported by patients perceive carers’ attitude towards them and Scazufca et al. [113], who observed that patients from the degree of overlap between the two [109,110]. high EE households perceived higher levels of carer These studies have also been driven by more recent criticismthanpatientswithlowEEcarersatthepointof cognitive models of psychosis, which posit that admission. Patient perceptions of carer attitudes have appraisals are important aspects of patient reactions also been linked to patient outcomes [114]. Among to unusual experiences and stress [6,111]. patientswithpsychosis,Tompsonetal.[110]observed In an investigation of recently relapsed patients, that perceived carer criticism, rather than observed results indicated that patient perceptions of carer carer EE, predicted patient relapse at twelve-month criticism were positively linked with carer ratings of follow-up. This finding was particularly evident for criticism, hostility and high EE. Moreover, high EE black and minority ethnic groups. 9.10 WORKING WITH INDIVIDUALS WITH PSYCHOSIS AND THEIR FAMILIES It is well documented that carers want to have more to be involved in their care [117]. Moreover, staff information about psychosis including management recognize the importance of working with families of specific problem behaviours, and to be more in- where an individual has psychosis [118]. The impor- volved in treatment decisions [115,116]. Patients tance of supporting individuals with psychosis and would also welcome greater opportunities for carers their families is reflected by the inclusion of family

FAMILIES AND PSYCHOSIS 109 (carer) interventions in the UK evidence-based treat- manuals have been developed to meet the varied ment guidelines for schizophrenia [119]. Similar clinical needs of carer and patient groups. Thus, treatment recommendations have been made in the family interventions can be delivered in settings US [120,121]. Over the last thirty years, a number of designed for individual families [131] or large multi- studies have documented the benefits of family work ple family groups [133]. Families can be seen at for patients and carers. home or within treatment clinics and sessions may Family interventions have proven efficacy in sig- not always routinely include the patient [134], nificantly reducing patient relapse rates and the although the evidence is stronger when they are negative impact of psychosis on carers [122,123]. included. Some manuals have been specifically They can reduce carers’ negative attitudes and developed for families during the early phases [135]. positively influence their willingness to continue However, despite the strong evidence base [136] and providing patient care [124,125]. The effectiveness national treatment recommendations [119], the pro- of family interventions in routine services has vision of family interventions within routine clinical also been confirmed [126–128]. The updated NICE services is extremely low [137]. Moreover, the Schizophrenia Guidelines propose that family inter- evidence suggests that independent of formal inter- ventions should be on offer to all individuals with ventions, any contact between professionals and psychosis in regular family contact and at least 10 families is usually limited and in the form of tele- therapy sessions should be offered over a period of at phone calls during times of crises [137]. These least three months and up to one year [119]. circumstances are disappointing, particularly given In addition to having a positive and nonblaming the role played by carers in patient functioning. stance towards families, the evidence-based family Difficulties with provision of family interventions intervention manuals comprise key therapeutic have been linked to issues related to staff training activities such as problem solving, communication and their attitudes towards families [137], organiza- skills, relapse prevention and psychoeducation tional barriers (e.g. heavy workload of staff) [138] (e.g. see References [129] to [132]). The different and reluctance expressed by carers and/or patients. 9.11 SUMMARY AND CONCLUSION Schizophrenia spectrum disorders can be severe attitudes are strongly linked to poorer patient out- and long term, and have an adverse impact on an comes, including elevated rates of relapse. Moreover, individual’s social network. Many patients, however, carers with critical patient attitudes tend to believe do remain in contact with families, particularly during patients can control their symptoms and are less likely the early phases of the illness. Carers provide an to attribute patient difficulties to an illness. Patients invaluable source of community support and social can accurately perceive carer attitudes, particularly contact. However, caregiving can negatively affect the those that are negative. These perceptions have also health and well-being of carers; a majority report been associated with poorer outcomes. More recent stress and burden within their role. In addition to research has looked at positive family reactions and coping with problematic patient behaviours, carers confirmed, in contrast, that these can be helpful in must grapple with a broad range of issues, including maintaining recovery. Evidence-based family inter- those related to diminished finances, restrictions on ventions have been developed to alleviate some of the their freedom and leisure time, and emotional issues strains and difficulties found within such families. such as loss and grief. Like patients, carers are often These focus on developing more positive coping, socially isolated, have difficulties coping and may be support, communication patterns, problem solving stressed and depressed themselves. These problems and cognitive reappraisals when someone in the fa- can then manifest themselves in carers’ negative and mily has psychosis. Despite having a well-developed critical attitudes towards patients. However, negative evidence base [119] showing that such family

110 COMPONENTS OF THE SOCIAL WORLD intervention can prevent relapse, access to such thera- 5. Magliano, L., Fiorillo, A., De Rosa, C. et al. (2005) pies remains limited currently by a lack of trained staff Family burden in long-term diseases: a comparative and adequate infrastructure. The only exceptions to study in schizophrenia vs physical disorders. Social this are in some early intervention services (e.g. see Science and Medicine, 61 (2), 313–322. References [10], [139] and [140]). 6. Garety, P. A., Bebbington, P., Fowler, D. et al. (2007) Mental healthcare systems have limited resources. Implications for neurobiological research of cognitive As a consequence, informal carers will continue to be models of psychosis: a theoretical paper. Psychologi- cal Medicine, 37, 1377–1391. major partners in the care and support of people with long-term mental health problems, particularly in the 7. Morgan, C., Kirkbride, J., Hutchinson, G. et al. (2008) Cumulative social disadvantage, ethnicity and first- initial stages. Moreover, carers often want to be more episode psychosis: a case-control study. Psychological involved and to establish better communication path- Medicine, 38 (12), 1701–1715. ways with service providers. However, long-term 8. Addington, J., Penn, D., Woods, S. W. et al. (2008) mental health problems such as schizophrenia or Social functioning in individuals at clinical high risk depression can have a negative effect on families for psychosis. Schizophrenia Research, 99 (1–3), and relationships. If carers are negatively affected by 119–124. their role and lack support from services, they are 9. Berry, K., Wearden, A. and Barrowclough, C. (2007) less likely to maintain their caregiving role. In Adult attachment styles and psychosis: an investiga- turn, patient outcomes will be adversely affected tion of associations between general attachment styles and additional professional support is likely to be and attachment relationships with specific others. So- required. It follows that family interventions, deter- cial Psychiatry and Psychiatric Epidemiology, 42, mined by level of need and embedded within a 972–976. social–cultural context, should remain a priority for 10. Stanghellini, G. and Ballerini, M. (2007) Criterion B all services. These should be offered not just as early (social dysfunction) in persons with schizophrenia: the interventions, but also for the middle and later phases puzzle. Current Opinion in Psychiatry, 20, 582–587. of problems [141]. 11. Fleury, M. J., Grenier, G., Caron, J. and Lesage, A. (2008) Patients’ report of help provided by relatives and services to meet their needs. Community Mental REFERENCES Health Journal, 44 (4), 271–281. 12. Parabiaghi, A., Lasalvia, A., Bonetto, C. et al. (2007) 1. Coen, R. F., O’Boyle, C. A., Coakley, D. and Lawlor, Predictors of changes in caregiving burden in people B. A. (2002) Individual quality of life factors distin- with schizophrenia: a 3-year follow-up study in a guishing low-burden and high-burden caregivers of community mental health service. Acta Psychiatrica dementia patients. Dementia and Geriatric Cognitive Scandinavica, 116 (s437), 66–76. Disorders, 13 (3), 164–170. 13. Addington, J., Coldham, E. L., Jones, B. et al. (2003) 2. Whitney, J., Haigh, R., Weinman, J. and Treasure, J. The first episode of psychosis: the experience of (2007) Caring for people with eating disorders: factors relatives. Acta Psychiatrica Scandinavica, 108 (4), associated with psychological distress and negative 285–289. caregiving appraisals in carers of people with eating 14. Fisher, H., Theodore, K., Power, P. et al. (2008) disorders. British Journal of Clinical Psychology, 46, Routine evaluation in first episode psychosis services: 413–428. feasibility and results from the MiData project. Social 3. Perlick, D., Clarkin, J. F., Sirey, J. et al. (1999) Burden Psychiatry and Psychiatric Epidemiology, 43 (12), experienced by care-givers of persons with bipolar 960–967. affective disorder. British Journal of Psychiatry, 15. Guada, J., Brekke, J. S., Floyd, R. and Barbour, J. 175, 56–62. (2009) The relationships among perceived criticism, 4. Tolin, D. F., Frost, R. O., Steketee, G. and Fitch, K. E. family contact, and consumer clinical and psychosocial (2008) Family burden of compulsive hoarding: results functioning for African-American consumers with of an internet survey. Behaviour Research and Ther- schizophrenia. Community Mental Health Journal, apy, 46 (3), 334–344. 45 (2), 106–116.

FAMILIES AND PSYCHOSIS 111 16. Harrow, M., Grossman, L. S., Jobe, T. H. and Herbener, initiation: a qualitative study of family members’ per- E. S. (2005) Do patients with schizophrenia ever show spectives. Comprehensive Psychiatry, 49, 530–536. periods of recovery? A 15-year multi-follow-up study. 28. Morgan, C., Abdul-al, R., Lappin, J. M. et al. (2006) Schizophrenia Bulletin, 31 (3), 723–734. Clinicalandsocialdeterminantsofdurationofuntreated 17. Martens, L. and Addington, J. (2001) The psycholo- psychosis in the AESOP first-episode psychosis study. gical well-being of family members of individuals British Journal of Psychiatry, 189 (5), 446–452. with schizophrenia. Social Psychiatry and Psychiatric 29. Garety, P. A., Fowler, D. G., Freeman, D. et al. (2008) Epidemiology, 36 (3), 128–133. Cognitive behavioural therapy and family intervention 18. Kuipers, E., Bebbington, P., Dunn, G. et al. (2006) for relapse prevention and symptom reduction in Influence of carer expressed emotion and affect on psychosis: randomised controlled trial. British Journal relapse in non-affective psychoses. British Journal of of Psychiatry, 192, 412–423. Psychiatry, 188, 173–179. 30. Norman, R. M. G., Malla, A. K., Manchanda, R. 19. Hirst, M. (2005) Carer distress: a prospective, popula- et al. (2005) Social support and three-year symptom tion-based study. Social Science and Medicine, 61 (3), and admission outcomes for first episode psychosis. 697–708. Schizophrenia Research, 80 (2–3), 227–234. 20. Lelliott, P., Beevor, A., Hogman, G. et al. (2003) 31. Da Roza Davis, J. M. and Cowen, P. J. (2001) Bio- Carers’ and users’ expectations of services – carer chemical stress of caring. Psychological Medicine, 31, version (CUES-C): a new instrument to support the 1475–1478. assessment of carers of people with a severe mental 32. Kuipers, E. and Bebbington, P. (2005) Research on illness. Journal of Mental Health, 12 (2), 143–152. burden and coping strategies in families of people 21. Hooyman, N. R. and Gonyea, J. G. (1995) Family with mental disorders: problems and perspectives, caregiving, in Encyclopedia of Social Work, 19th edn in Families and Mental Disorder: From Burden to (ed. R. L. Edwards), National Association of Social Empowerment (eds N. Sartorious, J. Leff and J. J. Workers, Washington, DC, pp. 951–959. Lopez), John Wiley & Sons, Ltd, Chichester, 22. Grandon, P., Jenaro, C. and Lemos, S. (2008) Primary pp. 217–234. caregivers of schizophrenia outpatients: burden and 33. Wynaden, D. (2007) The experience of caring for a predictor variables. Psychiatry Research, 158 (3), person with mental illness: a grounded theory study. 335–343. International Journal of Mental Health Nursing, 16, 23. Tang, V. W. K., Leung, S. K. and Lam, L. C. W. (2008) 381–389. Clinical correlates of the caregiving experience for 34. Wilcox, S. and King, A. C. (1999) Sleep complaints in Chinese caregivers of patients with schizophrenia. older women who are family caregivers. Journal of Social Psychiatry and Psychiatric Epidemiology, 43 Gerontology: Psychological Sciences, 54B, 189–198. (9), 720–726. 35. Gallagher, S. K. and Mechanic, D. (1996) Living with 24. Czuchta, D. M. and McCay, E. (2001) Help-seeking the mentally ill: effects on the health and functioning for parents of individuals experiencing a first episode of other household members. Social Science and Med- of schizophrenia. Archives of Psychiatric Nursing, 15 icine, 42, 1691–1701. (4), 159–170. 36. Schulz, R. and Beach, S. R. (1999) Caregiving as a risk 25. Schofield, N., Quinn, J., Haddock, G. and Barrow- factor for mortality: the caregiver health effects study. clough, C. (2001) Schizophrenia and substance misuse Journal of American Medical Association, 282, problems: a comparison between patients with and 2215–2219. without significant carer contact. Social Psychiatry 37. Treudley, M. B. (1946) Mental illness and family and Psychiatric Epidemiology, 36 (11), 523–528. routines. Mental Hygiene, 15, 407–418. 26. Garcia, J. I. R., Chang, C. L., Young, J. S. et al. (2006) 38. Thompson, J. E. H. and Doll, W. (1982) The burden of Family support predicts psychiatric medication usage families coping with the mentally ill: an invisible among Mexican American individuals with schizo- crisis. Family Relations, 31, 379–388. phrenia. Social Psychiatry and Psychiatric Epidemiol- 39. Jungbauer, J., Wittmund, B., Dietrich, S. and Anger- ogy, 41 (8), 624–631. meyer, M. C. (2003) Subjective burden over 12 months 27. Bergner, E., Leiner, A. S., Carter, T. et al. (2008) in parents of patients with schizophrenia. Archives of The period of untreated psychosis before treatment Psychiatric Nursing, 17 (3), 126–134.

112 COMPONENTS OF THE SOCIAL WORLD 40. Lauber, C., Eichenberger, A., Luginbuhl, P. et al. 51. Angermeyer, M. C., Bull, N., Bernert, S. et al. (2006) (2003) Determinants of burden in caregivers of pa- Burnout of caregivers: a comparison between partners tients with exacerbating schizophrenia. European Psy- of psychiatric patients and nurses. Archives of Psy- chiatry, 18 (6), 285–289. chiatric Nursing, 20 (4), 158–165. 41. Magliano, L., Marasco, C., Fiorillo, A. et al. (2002) 52. Brown, S. and Birtwistle, J. (1998) People with schi- The impact of professional and social network support zophrenia and their families – fifteen-year outcome. on the burden of families of patients with schizophre- British Journal of Psychiatry, 173, 139–144. nia in Italy. Acta Psychiatrica Scandinavica, 106 (4), 53. Boye, B. and Malt, U. F. (2002) Stress response 291–298. symptoms in relatives of acutely admitted psychotic 42. Ostman, M. and Hansson, L. (2004) Appraisal of patients: a pilot study. Nordic Journal of Psychiatry, caregiving, burden and psychological distress in re- 56 (4), 253–260. latives of psychiatric inpatients. European Psychiatry, 54. Kuipers, E. and Raune, D. (2000) The early devel- 19 (7), 402–407. opment of expressed emotion and burden in the 43. Ukpong, D. I. (2006) Demographic factors and clinical families of first onset psychosis, in Early Interven- correlates of burden and distress in relatives of services tion in Psychosis (eds M. Birchwood, D. Fowler and users experiencing schizophrenia: a study from south- C. Jackson), John Wiley & Sons, Ltd, Chichester, pp. western Nigeria. International Journal of Mental 128–140. Health Nursing, 15, 54–59. 55. Lee, C. and Gramotnev, H. (2007) Transitions into and 44. Karanci, A. N. and Inandilar, H. (2002) Predictors of out of caregiving: health and social characteristics of components of expressed emotion in major caregivers mid-age Australian women. Psychology and Health, of Turkish patients with schizophrenia. Social Psy- 22, 193–209. chiatry and Psychiatric Epidemiology, 37 (2), 80–88. 56. Kjellin, L. and Ostman, M. (2005) Relatives of psy- 45. Friedrich, R. M., Lively, S. and Rubenstein, L. M. chiatric inpatients – do physical violence and suicide (2008) Siblings’ coping strategies and mental health attempts of patients influence family burden and par- services: a national study of siblings of persons ticipation in care? Nordic Journal of Psychiatry, 59 with schizophrenia. Psychiatric Services, 59 (3), (1), 7–11. 261–267. 57. Vaddadi, K. S., Gilleard, C. and Fryer, H. (2002) Abuse 46. Jungbauer, J. and Angermeyer, M. C. (2002) Living of carers by relatives with severe mental illness. Inter- with a schizophrenic patient: a comparative study of national Journal of Social Psychiatry, 48 (2), 149–155. burden as it affects parents and spouses. Psychiatry – 58. Barton, K. and Jackson, C. (2008) Reducing symptoms Interpersonal and Biological Processes, 65 (2), of trauma among carers of people with psychosis: pilot 110–123. study examining the impact of writing about caregiv- 47. Chadda, R. K., Singh, T. B. and Ganguly, K. K. (2007) ing experiences. Australian and New Zealand Journal Caregiver burden and coping: a prospective study of of Psychiatry, 42 (8), 693–701. relationship between burden and coping in caregivers 59. Perlick, D. A., Rosenheck, R. A., Kaczynski, R. et al. of patients with schizophrenia and bipolar affective (2006) Components and correlates of family burden in disorder. Social Psychiatry and Psychiatric Epide- schizophrenia. Psychiatric Services, 57 (8), 1117–1125. miology, 42 (11), 923–930. 60. Wolthaus, J. E. D., Dingemans, P. M. A. J., Schene, 48. Jenkins, J. H. and Schumacher, J. G. (1999) Family A. H. et al. (2002) Caregiver burden in recent-onset burden of schizophrenia and depressive illness – schizophrenia and spectrum disorders: the influence of specifying the effects of ethnicity, gender and social symptoms and personality traits. Journal of Nervous ecology. British Journal of Psychiatry, 174, 31–38. and Mental Disease, 190 (4), 241–247. 49. Wilbram, M., Kellett, S. and Beail, N. (2008) Com- 61. Dyck, D. G., Short, R. and Vitaliano, P. P. (1999) pulsive hoarding: a qualitative investigation of partner Predictors of burden and infectious illness in schizo- and carer perspectives. British Journal of Clinical phrenia caregivers. Psychosomatic Medicine, 61, Psychology, 47, 59–73. 411–419. 50. Singleton, N., Maung, N., Cowie, A. et al. (2002) 62. Roick, C., Heider, D., Toumi, M. and Angermeyer, M. Mental Health of Carers, Office forNational Statistics, C. (2006) The impact of caregivers’ characteristics, London. patients’ conditions and regional differences on family

FAMILIES AND PSYCHOSIS 113 burden in schizophrenia: a longitudinal analysis. Acta course of schizophrenic illness. British Journal of Psychiatrica Scandinavica, 114 (5), 363–374. Preventive Social Medicine, 16 (2), 55–68. 63. Moller-Leimkuhler, A. M. and Obermeier, M. (2008) 75. Brown, G. W., Birley, J. L. T. and Wing, J. K. (1972) Predicting caregiver burden in first admission psychia- Influence of family life on the course of schizophrenic tric patients 2-year follow-up results. European disorders: a replication. British Journal of Psychiatry, Archives of Psychiatry and Clinical Neuroscience, 121 (3), 241–258. 258 (7), 406–413. 76. Brown, G. W. and Rutter, M. (1966) The measurement 64. Birchwood, M. and Cochrane, R. (1990) Families cop- of family activities and relationships: a methodologi- ing with schizophrenia – coping styles, their origins and cal study. Human Relations, 19, 241–263. correlates. Psychological Medicine, 20 (4), 857–865. 77. Rutter, M. and Brown, G. W. (1966) The reliability and 65. Huang, X. Y., Sun, F. K. and Fu, Yen (2008) The coping validity of measures of family life and relationships experiences of carers who live with someone who has in families containing a psychiatric patient. Social schizophrenia. Journal of Clinical Nursing, 17, 817–826. Psychiatry, 1, 38–53. 66. Folkman, S. and Lazarus, R. S. (1980) An analysis of 78. Kuipers, L. (1979) Expressed emotion: a review. coping in a middle-aged community sample. Journal British Journal of Social and Clinical Psychology, of Health and Social Behavior, 21 (3), 219–239. 18, 237–243. 67. Magliano, L., Fadden, G., Economou, M. et al. (2000) 79. Bebbington, P. and Kuipers, L. (1994) The predictive Family burden and coping strategies in schizophrenia: utility of expressed emotion in schizophrenia – an 1-year follow-up data from the BIOMED I study. aggregate analysis. Psychological Medicine, 24 (3), Social Psychiatry and Psychiatric Epidemiology, 35 707–718. (3), 109–115. 80. Butzlaff, R. L. and Hooley, J. M. (1998) Expressed 68. Scazufca, M. and Kuipers, E. (1999) Coping strategies emotion and psychiatric relapse – a meta-analysis. in relatives of people with schizophrenia before and Archives of General Psychiatry, 55 (6), 547–552. after psychiatric admission. British Journal of Psy- 81. Barrowclough, C., Tarrier, N., Humphreys, L. et al. chiatry, 174 (2), 154–158. (2003) Self-esteem in schizophrenia: relationships 69. St-Hilaire, A., Hill, C. L. and Docherty, N. M. (2007) between self-evaluation, family attitudes, and symp- Coping in parents of schizophrenia patients with tomatology. Journal of Abnormal Psychology, 112 (1), differing degrees of familial exposure to psychosis. 92–99. Journal of Nervous and Mental Disease, 195 (7), 82. Patterson, P., Birchwood, M. and Cochrane, R. (2005) 596–600. Expressed emotion as an adaptation to loss – prospec- 70. Garcia, J. I. R., Hernandez, B. and Dorian, M. (2009) tive study in first-episode psychosis. British Journal of Mexican American caregivers’ coping efficacy asso- Psychiatry, 187, S59–S64. ciations with caregivers’ distress and positivity to their 83. Raune, D., Kuipers, E. and Bebbington, P. (2004) EE relatives with schizophrenia. Social Psychiatry and at first episode psychosis: investigating a carer Psychiatric Epidemiology, 44 (2), 162–170. appraisal model. British Journal of Psychiatry, 184, 71. Scazufca, M. and Kuipers, E. (1996) Links between 321–326. expressed emotion and burden of care in relatives 84. Kopelowicz, A., Lopez, S. R., Zarate, R. et al. (2006) of patients with schizophrenia. British Journal of Expressed emotion and family interactions in Mexican Psychiatry, 168 (5), 580–587. Americans with schizophrenia. Journal of Nervous 72. Brown, G. W. (1959) Experiences of discharged and Mental Disease, 194 (5), 330–334. chronic schizophrenic mental hospital patients in var- 85. Rosenfarb, I. S., Bellack, A. S. and Aziz, N. (2006) ious types of living group. Milbank Quarterly, 37, Family interactions and the course of schizophrenia 105–131. in African American and white patients. Journal of 73. Brown, G. W., Carstairs, G. M. and Topping, G. (1958) Abnormal Psychology, 115 (1), 112–120. Post-hospital adjustment of chronic mental patients. 86. Dorian, M., Ramirez Garcia, J. I., Lopez, S. R. and The Lancet, 272 (7048), 685–689. Hernandez, B. (2008) Acceptance and expressed emo- 74. Brown, G. W., Monck, E. M., Carstairs, G. M. and tion in Mexican American caregivers of relatives with Wing, J. K. (1962) The influence of family life on the schizophrenia. Family Process, 47 (2), 215–228.

114 COMPONENTS OF THE SOCIAL WORLD 87. Li, Z. and Arthur, D. (2005) A study of three measures 99. Barrowclough, C. and Hooley, J. M. (2003) Attribu- of expressed emotion in a sample of Chinese families tions and expressed emotion: a review. Clinical Psy- of a person with schizophrenia. Journal of Psychiatric chology Review, 23 (6), 849–880. and Mental Health Nursing, 12 (4), 431–438. 100. Tienari, P., Wynne, L. C., Sorri, A. et al. (2004) 88. Mottaghipour, Y., Pourmand, D., Maleki, H. and Genotype–environment interaction in schizophrenia- Davidian, L. (2001) Expressed emotion and the spectrum disorder – long-term follow-up study of course of schizophrenia in Iran. Social Psychiatry and Finnish adoptees. British Journal of Psychiatry, Psychiatric Epidemiology, 36 (4), 195–199. 184, 216–222. 89. Scazufca, M. and Kuipers, E. (1998) Stability of 101. Schiffman, J., LaBrie, J., Carter, J. et al. (2002) expressed emotion in relatives of those with schizo- Perception of parent–child relationships in high-risk phrenia and its relationship with burden of care and families, and adult schizophrenia outcome of off- perception of patients’ social functioning. Psycholo- spring. Journal of Psychiatric Research, 36 (1), 41–47. gical Medicine, 28 (2), 453–461. 102. Weisman, A. G., Gomes, L. G. and Lopez, S. R. (2003) 90. Marom, S., Munitz, H., Jones, P. B. et al. (2005) Shifting blame away from ill relatives – Latino famil- Expressed emotion: relevance to rehospitalization in ies’ reactions to schizophrenia. Journal of Nervous schizophrenia over 7 years. Schizophrenia Bulletin, 31 and Mental Disease, 191 (9), 574–581. (3), 751–758. 103. Bertrando, P., Beltz, J., Bressi, C. et al. (1992) 91. Huguelet, P., Favre, S., Binyet, S. et al. (1995) The use Expressed emotion and schizophrenia in Italy – a of expressed emotion index as a predictor of outcome study of an urban-population. British Journal of Psy- in first admitted schizophrenic patients in French chiatry, 161, 223–229. speaking area of Switzerland. Acta Psychiatrica Scan- 104. Ivanovic, M., Vuletic, Z. and Bebbington, P. (1994) dinavica, 92, 447–452. Expressed emotion in the families of patients with 92. Barrelet, L., Ferrero, F., Szigethy, L. et al. (1990) schizophrenia and its influence on the course of illness. Expressed emotion and first-admission schizophrenia Social Psychiatry and Psychiatric Epidemiology, 29 nine-month follow-up in a French cultural environ- (2), 61–65. ment. British Journal of Psychiatry, 156 (3), 357–362. 105. O’Brien, M. P., Gordon, J. L., Bearden, C. E. et al. 93. Stirling, J., Tantam, D., Thomas, P. et al. (1991) (2006) Positive family environment predicts improve- Expressed emotion and early onset schizophrenia: a ment in symptoms and social functioning among one year follow-up. Psychological Medicine, 21, adolescents at imminent risk for onset of psychosis. 675–685. Schizophrenia Research, 81, 269–275. 94. Jarbin, H., Grawe, R. W. and Hansson, K. (2000) 106. O’Brien, M. P., Zingberg, J. L., Ho, L. et al. (2009) Expressed emotion and prediction of relapse in ado- Family problem solving interactions and 6-month lescents with psychotic disorders. Nordic Journal of symptomatic and functional outcomes in youth at ul- Psychiatry, 54, 201–205. tra-high risk for psychosis and with recent onset psy- 95. Marom, S., Munitz, H., Jones, P. B. et al. (2002) chotic symptoms: a longitudinal study. Schizophrenia Familial expressed emotion: outcome and course of Research, 107, 198–205. Israeli patients with schizophrenia. Schizophrenia 107. McFarlane, W. R. and Cook, W. L. (2007) Family Bulletin, 28 (4), 731–743. expressed emotion prior to onset of psychosis. Family 96. Kopelowicz, A., Zarate, R., Gonzalez, V. et al. (2002) Process, 46 (2), 185–197. Evaluation of expressed emotion in schizophrenia: a 108. Breitborde, N. J. K., Lopez, S. R., Wickens, T. D. et al. comparison of Caucasians and Mexican-Americans. (2007) Toward specifying the nature of the relation- Schizophrenia Research, 55 (1–2), 179–186. ship between expressed emotion and schizophrenia 97. Lopez, S. R., Hipke, K. N., Polo, A. J. et al. (2004) relapse: the utility of curvilinear models. International Ethnicity, expressed emotion, attributions, and course Journal of Methods in Psychiatric Research, 16 (1), of schizophrenia: family warmth matters. Journal of 1–10. Abnormal Psychology, 113 (3), 428–439. 109. Cutting, L. P., Aakre, J. M. and Docherty, N. M. (2006) 98. Kuipers, E. (2006) Family interventions in schizophre- Schizophrenic patients’ perceptions of stress, ex- nia: evidence for efficacy and proposed mechanisms of pressed emotion, and sensitivity to criticism. Schizo- change. Journal of Family Therapy, 28 (1), 73–80. phrenia Bulletin, 32 (4), 743–750.

FAMILIES AND PSYCHOSIS 115 110. Tompson, M. C., Goldstein, M. J., Lebell, M. B. et al. Database of Systematic Reviews, Issue 4, Art. No.: (1995) Schizophrenic-patients’ perceptions of their CD000088, DOI: 10.1002/14651858.CD000088.pub2. relatives’ attitudes. Psychiatry Research, 57 (2), 123. Pfammatter, M., Junghan, U. M. and Brenner, H. D. 155–167. (2006) Efficacy of psychological therapy in schizo- 111. Garety, P. A., Kuipers, E., Fowler, D. et al. (2001) A phrenia: conclusions from meta-analyses. Schizophre- cognitive model of the positive symptoms of psycho- nia Bulletin, 32, S64–S70. sis. Psychological Medicine, 31 (2), 189–195. 124. Berglund, N., Vahlne, J. O. and Edman, A. (2003) 112. Onwumere, J., Kuipers, E., Bebbington, P. et al.(2009) Family intervention in schizophrenia – impact on Patient perceptions of caregiver criticism in psychosis: family burden and attitude. Social Psychiatry and links with patient and caregiver functioning. Journal Psychiatric Epidemiology, 38 (3), 116–121. of Nervous and Mental Disease, 197 (2), 85–91. 125. Mueser, K. T., Sengupta, A., Schooler, N. R. et al. 113. Scazufca, M., Kuipers, E. and Menezes, P. R. (2001) (2001) Family treatment and medication dosage Perception of negative emotions in close relatives by reduction in schizophrenia: effects on patient so- patients with schizophrenia. British Journal of Clin- cial functioning, family attitudes, and burden. ical Psychology, 40, 167–175. Journal of Consulting and Clinical Psychology, 114. Lebell, M. B., Marder, S. R., Mintz, J. et al. (1993) 69 (1), 3–12. Patients’ perceptions of family emotional climate and 126. Barrowclough, C., Tarrier, N., Lewis, S. et al. (1999) outcome in schizophrenia. British Journal of Psychia- Randomised controlled effectiveness trial of a needs- try, 162, 751–754. based psychosocial intervention service for carers of 115. Simpson, A. (1999) Creating alliances: the views of people with schizophrenia. British Journal of Psychia- users and carers on the education and training needs of try, 174, 505–511. community mental health nurses. Journal of Psychia- 127. Kelly,M.and Newstead, L.(2004) Family intervention tric and Mental Health Nursing, 6 (5), 347–356. in routine practice: it is possible! Journal of Psychia- 116. Biegel, D. E., Song, L. Y. and Milligan, S. E. (1995) A tric and Mental Health Nursing, 11, 64–72. comparative-analysis of family caregivers perceived 128. Magliano, L., Fiorillo, A., Malangone, C. et al. relationships with mental-healthprofessionals. Psychia- (2006) Patient functioning and family burden in a tric Services, 46 (5), 477–482. controlled, real-world trial of family psychoeduca- 117. Walsh, J. and Boyle, J. (2009) Improving acute tion for schizophrenia. Psychiatric Services, 57 (12), psychiatric hospital services according to inpatient 1784–1791. experiences. A user-led piece of research as a means 129. Barrowclough, C. and Tarrier, N. (1992) Families to empowerment. Issues of Mental Health Nursing, of Schizophrenic Patients: Cognitive Behavioural 30 (1). Intervention, Chapman & Hall, London. 118. Slade, M., Holloway, F. and Kuipers, E. (2003) Skills 130. Falloon, I. R. H., Boyd, J. L. and McGill, C. W. (1984) development and family interventions in an early Family Care of Schizophrenia, Guildford, London. psychosis service. Journal of Mental Health, 12 (4), 131. Kuipers, E., Lam, D. and Leff, J. (2002) Family Work 405–415. for Schizophrenia: A Practical Guide, Gaskell Press, 119. NICE (2009) Schizophrenia Guideline (Update, 2009) London. www.nice.org.uk/guidance/index.jsp?action¼byId&o¼ 132. McFarlane, W. R., Lukens, E., Link, B. et al. (1995) 11786. Multiple-family groups and psychoeducation in the 120. American Psychiatric Association (2004) Practice guide- treatment of schizophrenia. Archives of General Psy- line for the treatment of patients with schizophrenia. chiatry, 52 (8), 679–687. American Journal of Psychiatry, 161 (2 suppl.), 1–56. 133. McFarlane, W. R. (2002) Multifamily Groups in the 121. Lehman, A. F., Lieberman, J. A., Dixon, L. B. et al. Treatment of Severe Psychiatric Disorders, Guildford (2004) Practice guideline for the treatment of patients Press, New York and London. with schizophrenia, second edition. American Journal 134. Barrowclough, C., Haddock, G., Tarrier, N. et al. of Psychiatry, 161 (2), 1–56. (2001) Randomized controlled trial of motivational 122. Pharoah, F., Mari, J., Rathbone, J. and Wong, W. (2006) interviewing, cognitive behavior therapy, and family Family Intervention for Schizophrenia, Cochrane intervention for patients with comorbid schizophrenia

116 COMPONENTS OF THE SOCIAL WORLD and substance use disorders. American Journal of 138. Fadden, G. (2006) Training and disseminating family Psychiatry, 158 (10), 1706–1713. interventions for schizophrenia: developing family 135. Addington, J. and Burnett, P. (2004) Working with intervention skills with multi-disciplinary groups. families in the early stages of psychosis, in Psycholo- Journal of Family Therapy, 28 (1), 23–38. gical Interventions for Early Psychosis (eds J. F. M. 139. Kuipers, E., Holloway, F., Rabe-Hesketh, S. and Ten- Gleeson and P. D. McGorry), John Wiley & Sons, Ltd, nakoon, L. (2004) An RCT of early intervention in Chichester, pp. 99–116. psychosis: Croydon Outreach and Assertive Support 136. Pfammatter, M., Junghan, U. M. and Brenner, H. D. Team (COAST). Social Psychiatry and Psychiatric (2006) Efficacy of psychological therapy in schizo- Epidemiology, 39, 358–363. phrenia: conclusions from meta-analyses. Schizophre- 140. Gleeson, F. M., Cotton, S. M., Alvarez-Jimenez, M. et al. nia Bulletin, 32 (S1), S64–S80. (2009) A randomised controlled trial of relapse preven- 137. Kim, H. W. and Salyers, M. (2008) Attitudes and tion therapy for first episode psychosis patients. Journal perceived barriers to working with families of persons Clinical Psychiatry, 24, el–el10 (pii:ej08,04407). with severe mental illness: mental health 141. Kuipers, E. (2008) The case for early, middle and late professionals’ perspectives. Community Mental intervention for psychosis. Invited commentary. World Health Journal, 44 (5), 337–345. Psychiatry, 7, 158–159.

10 Culture and its influence on diagnosis and management 1 Dinesh Bhugra and Susham Gupta 2 1 NIH Biomedical Research Centre and Institute of Psychiatry, King's College London, London, UK 2 East NHS Foundation Trust, London, UK 10.1 INTRODUCTION We are all born into a culture, absorb its premises as not all of its components will carry the same emphasis. we grow up, and consciously or unconsciously use The constituents of culture include language, folk cultural norms and parameters throughout our lives. tales, arts, morals, religious values, taboos, diet and The relationship between clinician and patient lies at so on. In effect, culture programmes societies by the heart of any health care delivery system. Both framing the ways in which shared life experiences clinician and patient are influenced by culture in are interpreted and what they mean both at the micro- presenting and understanding the diverse experiences level of the individual and family and at the macro- of distress. Culture influences not only models of level of the wider society. In some respects, cultures illness, i.e. how people make sense of and frame their can be said to have ‘personalities’, and both changes in experiences, but also help-seeking behaviour and how and movement between cultures can lead to conflict health care systems (broadly defined) are funded and and distress. accessed. Furthermore, for clinicians, an understand- This chapter provides an overview of the role of ing of the cultural values, norms and taboos held by culture in the clinical encounter. Relevant epidemio- patients and their carers may improve the therapeutic logicaldata willbereferredtobut will notbe alliance and subsequent engagement. Consequently, analysed. Furthermore, the prevalence and variation the concept of culture, and its constituent components, of mental disorders by cultural groups will not be help us considerably in understanding how individuals discussed here – the focus will be on broad cultural understand and cope with distress, and their interac- factors and their influence on diagnosis and manage- tions with health care systems. ment. Broad issues related to the role of culture and The way culture is constituted and constructed help-seeking behaviour and some principles of is multifaceted and complex. In any given culture managing patients across cultures will be explored. Principles of Social Psychiatry, second edition Edited by Craig Morgan and Dinesh Bhugra Ó 2010 John Wiley & Sons, Ltd.

118 COMPONENTS OF THE SOCIAL WORLD The clinician–patient interaction and factors influ- and some general guidelines for practice will be encing such interactions will be discussed in detail, offered. 10.2 DEFINITIONS Kirmayer [1] suggests that the construction of culture He offers a perspective that takes into account the offers one way of conceptualizing differences impact of colonization, by noting that so-called between people, bringing together race, ethnicity and primitive ways of life may well have been protective ways of life under a broad rubric to examine the against mental disorder on the basis that insanity impact of social knowledge, institutions and practices was apparently rare in such societies. Comparative on health, health care and healing. psychiatry came into its own in the middle of the There are well over a hundred definitions of cul- twentieth century. ture [2]. Tylor [3] offered the definition of culture as Boas [7] saw culture as an entity that embraces the ‘that complex whole which includes knowledge, social habits of a community; individual behaviours belief, art, law, morals, customs and any other capa- and practices are shaped by the habits and activities bilities and habits acquired by man as a member of of the group or wider community. On the other hand, society’. The implication is that culture is acquired and in contrast to Tylor, culture is defined by Kroeber that there is a totality of culture [4]. Itis this totality that and Kluckholn as consisting of patterns (both expli- anthropologyhasoftensoughttounderstand,reflecting cit and implicit) of behaviour (acquired and trans- an appreciation that culture is more complex than mitted by symbols), traditional (historically derived simple behavioural outputs. There is no doubt that the and selected) ideas, and especially their attached development and progression of anthropology in values. They indicate that culture may both be a the UK and the US has taken different paths. In the product of action as well as conditioning elements of nineteenth century, British anthropologists followed further action. Thus external factors and behaviours the Empire across the globe and studied ‘natives’, become internalized and, as Kroeber and Kluckholn whereas American anthropologists tended to focus on suggest, culture sets standards for behaviour and Amerindian cultures and tribes. Kirmayer [1] points ideologies. Others have also argued that these stan- out that the roots of cultural psychiatry can be traced to dards can be taken as a guide for acting and inter- the very beginnings of modern psychiatry. From preting the acts of others [8]. The transmissibility of travellers’ impressions to participation as observer- culture and its values lies at the heart of other participants, the focus remained on attempts to under- definitions, such as the one provided by Hughes [9]. stand the ‘other’. It is no surprise, given this, that Cultural processes transmit across generations psychiatry and anthropology often intersected, for and produce behaviours that may be normal for a example, in the work of W. H. R. Rivers. particular culture but abnormal for others. There is Broad uses of the term ‘culture’ include culti- a social component to culture that is determined by vation (cultivating crops and friendships, and in social transmission of values that in turn influence medicine the striking example is that of bacterial social behaviours. culture), collective identity and the anthropological DSM-IV [10] defines culture as meanings, values, meaning as described above [5,6]. Professions have systems and behavioural norms that are learned and cultures (a collective identity, shared practices and transmitted in the dominant society and within its norms) as do organizations (whether these are insti- social groups. Furthermore, culture powerfully influ- tutions such as clubs or health care organizations). ences cognitions, feelings and concepts of self, along Psychiatry itself has been a product of a specific with diagnostic processes and treatment. More recent culture and carries its own cultural meanings within definitions [11] also see culture in broad terms, as a itself. An admirable historical account of the develop- common heritage, learned set of beliefs, norms ment of culture has been provided by Kirmayer [1]. and values. Culture is thus a spiritual, physical and

CULTURE AND ITS INFLUENCE ON DIAGNOSIS AND MANAGEMENT 119 behavioural system that refers to what individuals aspects of culture may be easier to define and specify learn, retain and utilize on a regular basis. Material than spiritual aspects. 10.3 CULTURE AND ITS ROLE IN THE GENESIS OF DISTRESS Tseng [12] notes that several hypotheses have been religious delusions may be influenced not only by proposed, from a social psychiatric perspective, to one’s own religious world view but also by the explain observed variations in rates of mental disorder majority religions around the individual. As by social and cultural groups, from theories of social Tseng [12] points out, the intensity and the degree drift to theories of social cohesion. Tseng very of modification of symptomatology will be influ- cogently argues that there is little doubt that social enced to the degree that these disorders are seen as and cultural factors influence psychopathology in an atypical, subtypes or variations of disorders officially individual, but notes that the exact underlying pro- recognized in Western classificatory systems. cesses remain unclear, and outlines six different pos- sibilities through which culture can contribute to . Some behaviours are universal across cultures psychopathology: whereas others may become exaggerated due to specific cultural influences. Tseng [12] suggests . Cultures can be directly pathogenic and cause psy- that such examples are where in Malaysia indivi- chiatric disorders. For example, cultural ideas and duals provide social entertainment through being beliefs can contribute to stress, which in turn pro- provoked on social occasions and acting as clowns duces psychopathology. By contributing to an and in Japan warrior suicide (hara-kiri). individual’s stress, culture can add a further layer of anxiety and depression. Such a contribution is . Culture influences the frequency of occurrence of seen, for instance, in patients with loss anxiety. certain mental disorders in a population. These occurrences may change in response to changing . Culture can influence some individuals, who may factors in the culture. Tseng [12] goes on to illustrate have a pre-existing vulnerability, in such a way that this pathofacilitative effect of culture in cases of their reactions to stress are pathological. This is obsession with body weight and body shape, leading seen, for example, in the Malaysian notion of amok, to dieting patterns and pathological eating patterns. where a vulnerable individual is ‘expected’ to attack Increased or binge drinking is another example. others to demonstrate his masculinity. . Culture influences beliefs and understandings of . Culture can modify symptoms. The contents of disorders and moulds individuals’ reactions to dis- delusions, hallucinations, obsessions and phobias are order. Culture dictates how individuals perceive and all influenced by cultural and social factors. In the understand pathology and how they react to it. 1950s and 1960s delusions related to being poisoned Tseng [12] uses post-traumatic stress disorder as by mustard gas were not uncommon and these were an example. He argues that here society perceives replaced by those to do with space and NASA, and the disorder and reacts to its consequences with then with those related to computers. The contents of either sympathy or ignorance of the problem. 10.4 CULTURE AND MENTAL DISORDERS Itisapparentthatnotallmentaldisordersareinfluenced disorders but pathoplastic effects may be stronger. byculturetothesamedegree.Thepathogeniceffectsof Similarly, in culturally influenced syndromes, patho- culture may be somewhat limited in organic mental genic, pathoplastic and pathoreactive factors may play

120 COMPONENTS OF THE SOCIAL WORLD an important role [12,13]. In some cultures rates against self-harm so it is very difficult to be absolutely of suicide are high whereas in others these are low. certain of the exact rates. Similarly, in Mediterranean Some of the discrepancy can be explained by religious cultures, alcohol consumption may be high butrates of factors and proscription of self-harm. However, it is binge drinking and alcohol dependence may not, also likely that some cultures have legal proscriptions because culture normalizes such consumption. 10.5 CULTURE AND ETHNICITY Ethnicity is a narrower concept than culture. Ethnicity Isawij [17] identified twelve characteristics of eth- is often self-ascribed and refers to a sense of nicity of which the top five are: common ancestral individual identity and belonging based on common origin; same culture or customs; religion; race or origins and shared symbols and standards for beha- physical characteristics; and language. There may be viour [14]. Ethnicity, ethnic identity and ethnic groups an element of objectivity to this. However, the use of are often used to describe assumed characteristics ethnic categories in epidemiological studies is often that may ‘identify’ the individual to others [15]. not uniform. The UK Office of Population Consensus Thernstrom et al. [16] see ethnicity as both an internal andSurveys(OPCS–nowcalledtheOfficeofNational sense of distinctiveness and an external perception of Statistics or ONS) introduced ethnic categories for the distinctiveness. These authors point out that the fea- first time in the 1991 UK census. In subsequent cen- tures of common geographic origin, language or suses, the number of categories has increased and dialect, and religion that transcend community bo- perhaps become more sophisticated. The Social undaries are characteristics of ethnic groups. However, Sciences Research Council [18] defined ethnic groups as Bhugra and Bahl [15] suggest, this definition aspast-oriented groupidentification,based onconcep- lacks rigour. tions of cultural and social distinctiveness. 10.6 RACE, RACISM AND ITS EFFECTS Race emerged as a concept in the nineteenth century to are several types of racism – dominative, aversive, refer to certain assumed physical characteristics of cultural, pre- and post-reflecting gut racism, mis- groups [4]. Gaw [4] notes commonalities and overlaps sionary racism, paternalistic or colour blind racism. between concepts of race, ethnicity and culture. Racism in clinical situations may lead to under- Racism (i.e. discrimination of one group against diagnosis, missed diagnosis or misdiagnosis, under- another on the basis of perceived physical differences) treatment or overtreatment. Race is both a biological is related to one group holding and exercising power and a social label. Clinicians from a minority group through control of resources, including economic and may experience racism from patients and staff who educational. may belong to the majority group, thereby creating Racism is not a new phenomenon. The creation further tensions and strains in the therapeutic relation- of the ‘other’ is crucial in coming to terms with ship and the power within such a relationship. one’s own identity. Bhugra and Bhui [19] argue that Institutional racism has been defined as the sys- blunt applications of the word ‘racism’ perpetuate tematic enforcement of discrimination and mainte- conceptual confusion. Racism may operate through nance of the status quo by legal, cultural, religious, overt beliefs and actions of the individual (active educational, economic, political, environmental and racism) or through less conscious attitudes in military institutions of society – racism thus becoming society as a whole (not offering housing, education, an institutionalized form of personal attitudes [19]. etc.) – aversive racism [20]. As this suggests, there Any attempt to understand and eradicate racism must

CULTURE AND ITS INFLUENCE ON DIAGNOSIS AND MANAGEMENT 121 focus on three levels: individual, intragroup and inter- controversial area (see Reference [21–25]). To blame group. Patients’ clinical experience has to be seen not every misfortune experienced by black and minority only in the socioeconomic context but also in the ethnic communities on institutional racism is as facile political context. The debate about whether psychiatry as to say that institutional racism is an exaggerated and psychiatric services are institutionally racist is by response to everything. Bhugra and Bhui [19] recom- no means over. In the UK at least this remains a mend steps for change. 10.7 MIGRATION Migration is a universal phenomenon and has (e.g. migration within the same country) such changes occurred throughout human history. A migrant is may well be limited. Westermeyer [26] goes on to note defined as someone who changes their place of resi- that geographical distance can also affect the migrant’s dence for any purpose and for any period of time [26]. ability to satisfy emotional needs by returning home at The duration of ‘migration’ becomes a problem in this intervals. However, with recent changes in commu- definition. If someone goes away for a month’s holi- nication and social support styles (see Chapter 36) the day in another country, would that count as migration? sense of alienation and isolation may shift. When Migrants are classified according to a number of large differences exist between the new and the old potential factors, e.g. characteristics of migrants, residence, ecological changes related to unfamiliar motivation for migration and duration of migration. systems, climate, modes of transportation, housing Certain occupations are inextricably linked with style and neighbourhood will all influence response migration, such as military personnel, anthropolo- to migration [26]. Communication and linguistic styles gists, diplomats and students. It is inevitable that the may add to the stress, and vocational changes may add impact of migration will differ across each occupa- to the frustration as well-educated and trained migrants tional group, as will reasons and exact duration of may not find jobs in the field they want to, thereby migration. Another factor that is often not studied is causing a discrepancy in aspiration and achievement the likelihood of return and circumstances of such an and producing a lowering of self-esteem. Understand- event. ing the new structures and learning about new social Demographic characteristics (e.g. age, gender, roles will take time and may also be influenced by being primary migrant) and reasons for migration will generational differences. Legal status and challenges all play a role in experiencing and in dealing with may need to be overcome. Culture conflict may occur stresses related to migration. The individual’s in terms of one’s own cultural values coming into response to stress will depend on the reason for conflict with others, and also within one’s own migration and whether the migration was voluntary culture, as some individuals may move faster in the or involuntary. Whether the individual migrates alone process of acculturation. The differences between a or with others may well add to the stress. Those who primary migrant (one who is the ‘pioneer’ and are forced to migrate or who are extruded from one migrates first) and those who follow will be multi- culture may find that adjustment to the new culture is faceted. As Westermeyer [26] points out, the reasons essential for survival. Thosewho migrate for economic for moving to a new place are not necessarily reasonsmaywell facedifferentstressors.Timetakento the same reasons for staying in the same place. migratemayalsoaffectadjustment.Forexample,those Rationales (and related stress) will change over time. migrating as refugees and asylum seekers may be held Meszaros [27] suggested that migrantsmay respond in a suspended state for months before being given to the act of migration by being overenthusiastic, leave to remain, thereby adding to the stress being inhibited,critical,perplexedorhyperactive.Berry[28] experienced. Some relocations are likely to involve suggests that migrant adjustment styles may be to do extensive linguistic, ecological, political, technologi- with adjustment (movement towards), reaction cal, social and family changes, whereas with others (movement against) and withdrawal (movement

122 COMPONENTS OF THE SOCIAL WORLD away). Westermeyer [26] proposes that for a migrant’s Table 10.1 Possible questions for migrants’ migration adjustment four factors should be considered: the experiences migrant’s personality, degree of difference or similarity, How long ago did the subject migrate? attitude of the new community and age of the migrant. At what age did the subject migrate? Bhugra [29,30] suggests that at different stages of Motives for migration? migration, coping strategies will include different Difficulties in migration and its perceived reversibility? strands and varying stressors. Therefore assessment Preparedness? at different stages of migration will need to be mod- How great was the difference between the two cultures? ified accordingly. After migration, settling down with Experience before and during the journey? Migrated alone or in a group? others from the same culture may have its additional Intentions at the beginning about duration of stay? stressors. For sociocentric individuals from socio- Attitude to culture of new country? centric countries, settling down in an egocentric The role of the new society in helping the subject to adjust? country (if not surrounded by sociocentric indivi- Previous life experiences? duals) may be more stressful. Communities may feel stress both at individual and group levels. Post-migration, there may be two peaks related to an increase in psychopathology. First, immediately migration and its sequalae. Clinical assessment of after arrival, when the individual may be finding it migrants is no different from that of other ethnic difficult to adjust and, second, 10–12 years later when groups, but in addition a clear assessment of migration the individual has settled down but may not have and its process may provide additional dividends in achieved as much as he/she expected to. Again, this planning any intervention. Some questions on how to discrepancy in achievement and aspiration may lower address migration are provided in Table 10.1. self-esteem and lead to depression or other mental The attitude of the clinician, their awareness of their disorders [31]. own prejudices and a nonjudgemental approach can Preparation for migration may alleviate some provide the basis for a good rapport. The mistrust potential stressors. Learning about the new country and feelings of alienation that migrants may be carry- and its customs, languages and weather may help the ing with them must be dealt with before any shift can individual to adjust and also allow them to engage and occur in treatment or the therapeutic alliance. From teach others to cope. observation of migrants’ clothes and behaviour, which Migration is not a unitary experience and migrants are likely to be influenced strongly by their culture, to are not a homogeneous group. However, they can linguistic differences and affect, the clinician must not suffer any or all of the mental disorders that non- take anything for granted. Social support systems may migrants may do. The role of culture becomes much influence the process of acculturation, which in more significant in migrants, especially if their accul- turn will affect the clinical encounter. Loss of control turative processes are still in the process of transition. as an explanation of distress may enable some indi- Various terms such as ‘culture shock’, ‘culture con- viduals to cope better if they see it is an external cause flict’ and ‘cultural bereavement’ have been used in the (i.e. it is in the stars or their fate) rather than an migrant groups to understand their experiences of internalized one. 10.8 RACIAL LIFE EVENTS AND PSYCHIATRIC MORBIDITY Life events are universal and their association with recent [32]. These authors suggest that there seems onset of certain psychiatric disorders and relapse of to be an increasingly robust body of evidence high- others is well known. However, interest in racial life lighting the role of chronic stress and difficulty in events (life events that the patient or their carers the aetiology of common mental disorders across attribute to their race or ethnicity) is relatively different cultural and ethnic groups. It is difficult to

CULTURE AND ITS INFLUENCE ON DIAGNOSIS AND MANAGEMENT 123 find the exact rates of race-related crimes, but as racial and racism (see Reference [32]). The relationship discrimination is not randomly distributed in society between race, racism and related life events and their these events can add to chronic difficulties. There are impact on the mental health of individuals is complex several scales that measure experiences related to race and deserves to be explored further. 10.9 ACCULTURATION Acculturation is the process of cultural and psycho- include behaviour. The levels of acculturation will logical change that takes place as a result of contact rely on the degree of exposure, distance between the between cultural groups and their individual mem- two cultures and the degree of willingness on the part bers [33]. This contact initially was a result of of the individual to change [36]. Although accultura- colonization, military invasions, migration and tion occurs when two cultures come together and both sojourning [34]. However, in the last decade or so, will change to some degree, it is inevitable that the with increased access to international media and nondominant culture is likely to change more. It is globalization, the process has become not only more helpful, therefore, for the clinician to be aware of evident but increasingly more rapid. The changes in acculturation and cultural identity so that the indivi- both cultures coming into contact mean that both dual patient’s symptoms and distress can be under- psychological and cultural shifts occur at individual, stood in the context of their culture and suitable group and institutional levels. interventions set in place. At an individual level, behavioural changes may Bhugra [37] suggests that cultural identity is also well be associated with cognitive changes such that very closely linked with the concepts of the self, which either immediately or in the long run attitudes and are very clearly influenced by cultures and cultural even the identity of individuals will shift. Berry [34] values. The single most important question asked of provides a helpful framework for understanding the an individual in order to understand their identity is processes and consequences of acculturation. Two ‘Who am I?’ [38]. The perception of the individual is cultures coming into contact are unlikely to be of the linked with what they see as the context in which the same level. Invariably one is going to be a dominant question is being asked, and yet the identity can be culture and the other one nondominant. At a cultural psychological and fluid. The independent and inter- level the nondominant culture may therefore be de- dependent construal can thus be measured through cultured or it may be assimilated. Furthermore, cognitive schemas (which may be more difficult to depending upon a number of factors, the nondominant measure) and behavioural and motivational schemes culture may become acculturated and its members (which may be easier to measure). The concept of the may become bicultural. Linton [35] argues that self is at the core of an individual’s being and its change occurs both at covert and overt levels. The assessment, in line with changes related to accultura- covert aspects focus on knowledge, psychological tion, is helpful in understanding what is abnormal and states, attitudes and values, and the overt aspects what may well be pathological. 10.10 CULTURAL IDENTITY Cultural identity is multilayered and is shaped by, patients, psychopathology related to thought disorder among others, race, ethnicity, culture, language, gender, may not be easily clarified if the patient’s primary age, sexual orientation,religious values, socioeconomic language is not used in assessment. This also allows status, migration and acculturation [39]. The language an understanding of the level of acculturation. The the patient chooses to communicate in is a significant clinician must explore the degree of comfort the patient part of the individual’s cultural identity. In bilingual feelswithanylanguage,their knowledge of bothwritten

124 COMPONENTS OF THE SOCIAL WORLD and spoken language and their preferred choice. How- effort must be madeto ask about discriminatory actions, ever, the patient may feel embarrassed by the fact that perceptions of such discrimination must also be dis- they are not able to speak the language of the new cussed. Cultural explanations of the individual’s illness society; hence the clinician must be sensitive to their mustbediscussedwiththepatientandtheirfamilies(see needs.Levelsofacculturationcanbeexploredbyasking Reference [40]). Cultural identity also includes reli- about what they feel about their culture of origin, their gious beliefs and attitudes, which may change after the attitudes and beliefs, their response to the values of the individual has come into contact with a new culture. new culture, interactions with their own community and Explanatory models (as discussed below) and their also with others from other communities. Experiences assessment mayalsoenablethe cliniciantoget an ofdiscriminationmustalsobeexplored.Althoughevery overview of cultural identity. 10.11 CULTURE AND CLINICIAN–PATIENT INTERACTIONS: EXPLANATORY MODELS Weiss and Somma [41] point out that the idea of to prototypic personal history and associations that explanatory models of illness is often interpreted diff- collectively characterized the illness at a particular erently by various professions and professionals. The point of enquiry’. Bhui and Bhugra [43,44] argue that first notions of explanatory models appeared in the there is a need for understanding explanatory models 1970s but the most influential proponent is Kleinman so that adherence to treatment can be strengthened, [42], who defined explanatory models as notions about especially if the patient’s and family’s explanatory an episode of sickness and its treatment that are models are at variancewith those of the clinicians. The employed by all those engaged in the clinical process. understanding of the patient’s explanatory models Explanatorymodelsofpatients,theircarersandthoseof does not need to be complicated. Bhuiand Bhugra [44] professional carers needtobestudiedandunderstoodso suggest that even if explanatory models differ mark- that patients and their carers are engaged in the thera- edly between the clinicians and the patient, compro- peutic process and the therapeutic alliance improved. mise is possible. Through understanding and making Explanatory models are used to explain not only the patient feel listened to and allowing the patient to perceived causes or aetiology but also understand- hang on to their views dialogue is made easier. If an ings of symptoms, pathophysiology, course of sick- impasse is reached, other strategies such as involving ness and treatment. Explanatory models differ from the family or community leaders may be used. general beliefs about sickness and are marshalled in Through such an exploration, permission is granted, response to a particular illness (see Reference [42]) which encourages and reassures the patient that it is and will therefore vary. Eliciting explanatory models possible to hold on to cultural values and views. provides an understanding of patients’ and their Explanatory models have been assessed in epide- carers’ models about what they call the problem, miological studies using qualitative instruments such what they see as its cause, why it appeared when it as the EMIC [45] and the shorter SEMI [46]. Frame- did, how the symptoms affect the individual and works of explanatory models have been applied across those around them, an understanding of the severity ethnic groups [47] and cultural groups [48], in child of the problems, foreboding about the symptoms, psychiatry [49] and among the elderly [50], thus problemsasa result of thesymptoms, whether treat- indicating that it is possible to use this model across ment is needed and if so what treatment, and what conditions and age groups. results are expected from the treatment. As Weiss and Somma [41] emphasize, although Weiss and Somma [31] examined 677 Medline illness explanatory models have developed in differ- references with the term ‘explanatory model’ or ent ways with different formulations and concepts, ‘explanatory models’ in their titles in 2006 and noted they remain an important area of interest in cultural that explanatory models were usually described psychiatry. Bhugra and Cochrane [51] caution that with reference ‘to a set of cognitive explanations, heterogeneity of migrant groups and their models symptomatic, emotional and social experiences and of perceived ill health are all important factors in

CULTURE AND ITS INFLUENCE ON DIAGNOSIS AND MANAGEMENT 125 help-seeking, as is their world view. Tseng [13] pro- in some cases, following acculturation, attitudes, poses that explanatory models change with industria- beliefs and explanations will change. Sociocentric lization. More traditional societies have supranatural beliefs, with an external locus of control, may lead or natural models whereas more ‘modern’ or indus- to some protection against distress [36]. Patients will trialized societies may have mixed biological, psy- seek help from whichever source maps on to their chological and social models. Thus it is inevitable that understandings of the episode of illness. 10.12 DISEASE VERSUS ILLNESS Eisenberg [52] differentiated between disease and ill- after all, abnormalities in the structure and functioning ness. This distinction highlights not only the problems of body organsand systems. Thus illness is theshaping of the patient’s perceptions and expectations but also of biological and psychological symptoms into beha- those of the clinician. Kleinman [42] suggests that vioural and social responses that affect not only the disease (literally dis-ease) refers to a malfunctioning patient but also those around the individual. Societies of biological and/or psychological processes whereas and cultures have set standards for what is normal and illness includes secondary personal and social abnormal and this is where the role of the psychiatrist responses to a primary malfunctioning (consequence in understanding and remaining within the social ofdisease).Illnessthusinvolvesprocessesofattention, norms is a significant one. Cultural factors contribute perception, affective responses, cognition and valua- differentially to the causation, formation and interpre- tion (of symptoms and role impairment) directed at the tation of abnormal behaviours. For a clinician to disease and its malfunction. Eisenberg [52] points out engage a patient effectively and form a therapeutic that patients suffer from illnesses and seek treatment alliance, a knowledge of differentiations between dis- for these whereas doctors (by virtue of their training ease and illness and an exploration of explanatory and experience) diagnose and treat diseases that are, models to understand this distinction is crucial. 10.13 CULTURE, DIAGNOSIS AND CLINICAL CARE 10.13.1 Culture of the individual or initiate management. It may be that the clinician is carrying out an assessment of compulsory detention In the therapeutic encounter between the patient and and the culture of the organization may dictate how, the clinician, their respective cultures and world views where and when such encounters take place. In addi- play a key role in therapeutic engagement and improv- tion, the notion of the culture of the organization allows ing the therapeutic alliance. The idioms of distress both the patient and the clinician to explore and mould employed by the individual are sanctioned by their their own responses in a way that enhances joint work- culture. The patient’s expectations of the therapeutic ing. Along with geographical accessibility to the orga- encounterwilldictatewhatinformationishandedtothe nization, emotional accessibility by the patient is a clinician and what information is held back. Patients useful concept that clinicians and service providers are interested in getting better quicker and thus carry need to be aware of. This in turn indicates an openness models that may not match those of the clinician. on the part of the organization and a willingness to be culturally sympathetic to patients and their carers. 10.13.2 Culture of the organization 10.13.3 Culture of the clinician The setting where therapeutic encounters take place is also important. However, often not much attention is Not only do the cultural values and cultural prejudices paid to this. The purpose of the therapeutic encounter of the clinician influence the interaction, other factors may be to offer a diagnosis, commence investigations such as age, previous experience and gender,

126 COMPONENTS OF THE SOCIAL WORLD Doctor (signs) Patient (symptoms) Class Training Race Gender Culture Concepts of Expectations Expectations Concepts of disease illness Folk culture Knowledge Education Age Experience Race Experience Training Class Gender Social status Social status Figure 10.1 Therapist–client interaction Cultural Verbal Communication Verbal Cultural affilliation Non-verbal Non-verbal affilliation and influencing factors along with explanatory models, may play a role (see cultures. Similarly, in some cultures, being in close Figure 10.1). Although the figure shows an equal and physical proximity indicates closeness, whereas in balanced interaction, in reality it is not. The power lies others it maybeseenasinvadingone’s personal with the clinician who can deprive the patient of their space. In initial clinical contact, Westermeyer’s [26] liberty, thereby creating an underlying and unstated advice is worth remembering – he suggested that tension. An awareness of both verbal and nonverbal clinicians should ignore their initial clinical instincts communications in a specific culture will go a long regarding the patient’s gesticulations, affect, dress way towards establishing a therapeutic alliance. andmannerismssothatnomistakesare made in The patient’s physical placement and activity reaching assumptions. during the interview, i.e. nonverbal communication Mental state examination and its interpretation along with facial expressions and gesticulations, depends on the knowledge of the patient’s early reflect the influence of culture [26]. Hall [53] socialization, culture of origin, patient’s educational described the diverse means by which members of level and literacy, and the acculturation process [26]. different cultures communicate with one another For something simple like assessing orientation, not nonverbally(afieldtermedkinesics) and use space only calendars vary but seasons and time of the day (a field called proxemics). There are often simila- may not always be clear. It has been shown that in ritiesinthe waythese twofields areusedbymem- many rural areas time is measured by which pro- bersof thesamecultureandhow thesediffer from gramme is on television. Similarly, proverbs used to other cultures. It is likely that one or both of these evaluate abstract thinking will be very heavily influ- fields may change for individuals who come into enced by the culture and language. Education and life contact with other cultures, as a result of the process experiences will play a significant role in commu- of acculturation. Westermeyer [26] also notes that nication across cultural boundaries. Educated kinesic and proxemic behaviours do vary between patients, irrespective of their culture, may well have American psychiatric patients in comparison with more in common with clinicians than those who are American nonpsychiatric subjects, and thus if other illiterate. While reaching a clinical diagnosis of cultures are added to this, interactions clearly mental illness in a patient from a different culture, become much more complex and multifaceted. the clinician must take into account not only the Some cultures will see direct eye contact as demon- culture of the patient and that of the organization but strating honesty and openness, but the same eye also their own prejudices and world view. They must contact can indicate anger or disrespect in other recognize whether a specific cluster of symptoms,

CULTURE AND ITS INFLUENCE ON DIAGNOSIS AND MANAGEMENT 127 signs and resulting behavioural changes are inter- keep changing, and the clinician needs to be aware of preted by the patients, their families and community this possibility. Furthermore, patients and their as pathological or not. families may still be using folk remedies that may Cultures are never static and the dynamic process interact with Western allopathic treatments. In addi- of acculturation at both individual and community tion, even among dominant communities, folk models levels means that the patient’s world view may well of care may exist and need to be explored. 10.14 CULTURE AND PSYCHOTHERAPY Psychotherapy across cultures brings with it chal- the side of the therapist. The type of communication lenges of its own. Embedded within such a challenge and interpretation will be influenced by this power are the role of gender and age. Deeply held beliefs, disparity. In settings where family work is being values and concepts of the therapist and the patient are carried out and families have different cultures, the very strongly influenced by a number of factors. relationship will become different. The therapeutic Depending upon the type of psychotherapy, the inter- alliance needs to be understood in a number of dimen- action between the therapist and the patient will vary. sions in psychotherapeutic settings. As the therapist For example, in some cultures, due to strong notions of contributes to this alliance by virtue of their age, shame, sitting in a group situation and talking of experience, training, gender and culture, the patient family problems may prove anathema to the patient, brings material to the sessions that is equally affected whereas in others, letting it all ‘hang out’ will not only by these factors. The patient’s capacity to form a be acceptable but highly valued. Social activities such therapeutic alliancewill be influenced by their cultural as going on dates or preparing for such an eventuality alliances and acceptance of different schools of may be problematic for some cultures where, for thought. Some Western egocentric therapies may be example, restrictions in social interactions may apply. anathema to some patients and appropriate adjust- Psychotherapy itself is a cultural construct. Com- ments may need to be made. With cognitive or beha- bining this with social constructs of race, ethnicity and vioural therapies, the essential homework that is often culture adds several dimensions, which may make it part of these may carry connotations of control that difficult for the therapist to carry out assessment and some patients may find difficult to deal with. The three therapy. When the patient and the therapist are from components of this alliance are the therapist, the two different cultures – whether one belongs to the patient and the disorder itself [56]. In the patient’s majority culture or whether both belong to minority case a process of organizing information and reflect- cultures – this will determine perceptions and engage- ing on it along with negotiations of conflict are ment from both sides. Moodley and Palmer [54] point important factors. As the therapeutic relationship out that even though sociocultural and political ideas progresses, the initial directiveness may give way to of race, culture and ethnicity have been considerations joint involvement and working, which may be less in counselling psychotherapy they have also often acceptable to the patient. In particular, specific tech- been marginalized or ignored. Bhugra and Bhui [55] niques used in each therapy may have to be modified. suggest that although psychotherapies are based on Behavioural tasks may be seen as intrusive or pushy, differing theoretical standpoints, the majority relieve producing feelings of frustration. On the other hand, distress using models of therapy developed in the West total flexibility and passivity on the part of the thera- using Western perspectives of the mind, illness, emo- pist may be interpreted by the patient as incompe- tions and healing. These authors argue that ethno- tence, since the patient may be looking for clearer centrism on the part of the therapist may well lead to direction. disengagement on the part of the patient. Embedded in Psychotherapy in many forms is a Western such a dialectic is the imbalance of power, which is on invention, especially in its egocentric form. In other

128 COMPONENTS OF THE SOCIAL WORLD cultures, using religious discourses and scriptures, to look for ethnic matching it would make this slim suggestions can be made to change lifestyles and deal possibility even rarer. The relationship process with stress. Whatever the model of therapy, Bhugra between the therapist and the patient needs to be and Bhui [55] assert that the patient’s motivation and explored further, especially if they are from different expectations are important. The patient’s image of the cultures; not only the outcomes but also the process of therapist may deploy that of a master, and each therapy needs to be investigated. Psychotherapists are encounter is seen through this prism. Both patient similar to anthropologists in that their work often and therapist have to agree on the rationale for therapy leads to the same insights at a personal level [57]. that can be converted into a therapeutic plan. All effective treatments, whether these are Western Ethnic matching of the therapist and the patient is style psychotherapies or traditional healing, must be not as important as is their shared world view. As it is, based on the right kind of relationship between the psychotherapies are not easily available to patients patient and the therapist with the right kind of from minority cultures, at least in the UK; if we were ambience. 10.15 CULTURE AND PHARMACOTHERAPY It is not proposed to discuss details of ethnopsycho- Amongnonbiologicalfactorsthatmayplayarole,Lin pharmacology or pharmacodynamics and pharmaco- etal.[60]pointoutthattheexpectationeffect(including kinetics of drugs in this chapter (for details see the placebo effect, clinician ideology, past experiences, References [58] and [59]). Only a few principles will religioustaboos,etc.)powerfullydeterminesthesuccess be highlighted here, especially those that are going to or failure of any pharmacological treatment. be useful in understanding social psychiatry. Lin Use of complementary medicine in some cultures is et al. [60] suggest that treatment decisions are more prominent, especially if such systems of med- often not individualized but the choice of medication icine emerged from these countries. For example, the and dosages are largely based on ‘trial and error’ Ayurvedic system of medicine in India and the system practice. There is no doubt that ethnicity and culture of traditional medicine in China are used extensively, powerfully determine the individual’s pharmacologi- both in those countries and among the diaspora. cal response [61]. These responses are a result of both Patients may also believe that medicines from natural environmental and genetic factors and perhaps an sources are nontoxic and nonaddictive or noninva- interaction between the two. Lin et al. [60] note that, sive [62] and perhaps safer. Thus these medications of even greater importance, the success of any therapy are used commonly in less severe and chronic mental (including pharamacotherapy) depends upon the rela- disorders such as sleep, anxiety or neurasthenia. tionship between the patient and the therapist. Sig- Sociocultural, illness and biological factors all nificant variations in drug response across ethnic affect an individual’s attitudes towards medica- groups exist. Dietary habits affect pharmacokinetics tion [63]. Health explanatory models can play a subtle as these change the body’s ability to absorb, distribute role in engaging the patient with the clinician. These and metabolize medication. After migration, with authors suggest that stigma, attitudes of significant resulting changes in diet as part of the acculturation others and individual preferences all play a role [64]. process, rates of metabolism of different compounds Cognitive and behavioural effects of medication also change. Smoking also influences absorption and across cultural groups have not been studied and metabolism of some drugs. In addition, drug–drug deserve to be. The occurrence of placebo response, interaction may lead to interference with absorption, as well as demand for pills from certain cultural synergy of therapeutic pharmacological effects (or groups, indicates that different individualized path- that of side effects) and interactions with drug ways into care deserve to be explored so that any care metabolism [62]. provided will be culturally accepted.

CULTURE AND ITS INFLUENCE ON DIAGNOSIS AND MANAGEMENT 129 10.16 CONCLUSION The impact of culture on the mental health of indivi- 9. Hughes, C. C. (1992) Culture in clinical psychiatry, in duals, their idioms of distress and help-seeking cannot Culture, Ethinicity and Mental Illness (ed. A. C. Gaw), be underestimated. When both the patient and the APA Press, Washington, DC, pp. 1–41. clinician come from different cultures, their interac- 10. APA (1994) DSM-IV, American Psychiatric Associa- tion depends upon a number of factors. Equally, the tion, Washington, DC. process of interaction plays a significant role in the 11. DHSS (2001) Mental health, culture, in Race and therapeutic alliance and the success of the therapy, Ethnicity – A Report of the Surgeon General,US whether it is psychotherapy or pharmacotherapy. DHSS, Rockville, Maryland. Patients’ experiences and perceptions play a signifi- 12. Tseng, W.-S. (2007) Culture and psychopathology, in cant role in the therapeutic engagement. Cultural Textbook of Cultural Psychiatry (eds D. Bhugra and K. S. Bhui), Cambridge University Press, Cambridge. psychiatry, although existing for several decades, is only now coming together as a discipline that fits into 13. Tseng, W.-S. (2001) Handbook of Cultural Psychiatry, Academic Press, San Diego, California. and sits alongside the study of the social aspects of aetiology and management of mental disorders. 14. Harwood, A. (ed.) (1981) Ethnicity and Medical Care, Harvard University Press, Cambridge, Massachusetts. 15. Bhugra, D. and Bahl, V. (1999) Ethnicity – issues of definition, in Ethnicity: An Agenda for Mental Health REFERENCES (eds D. Bhugra and V. Bahl), Gaskell, London, pp. 1–6. 16. Thernstrom, S., Orlov, A. and Handlin, O. (eds) (1980) 1. Kirmayer, L. (2007) Cultural psychiatry in historical TheHarvardEncyclopaediaofAmericanEthnicGroups, perspective, in Textbook of Cultural Psychiatry (eds Harvard University Press, Cambridge, Massachusetts. D. Bhugra and K. S. Bhui), Cambridge University 17. Isawij, W. (1974) Definitions of ethnicity. Ethnicity, Press, Cambridge. 1, 111–124. 2. Kroeber, A. and Kluckholm, C. (1952) Culture: a 18. SSRC (1974) Comparative Research on Ethnicity, critical review of concepts and definitions, Papers of Social Sciences Research Council, New York. the Peabody Museum (cited in A. C. Gaw, 2001), 19. Bhugra, D. and Bhui, K. S. (1999) Racism in psychia- in Concise Guide to Cross-Cultural Psychiatry, Ame- try: paradigm lost and paradigm regained. Interna- rican Psychiatric Publishing, Washington, DC. tional Review of Psychiatry, 11, 236–243. 3. Tylor, E. (1871) Primitive Culture, Estes and Lauriat, 20. Dolan, B., Polley, K., Allen, R. and Norton, K. C. Boston, Massachusetts. (1991) Addressing racism in psychiatry. International 4. Gaw, A. C. (2001) Concise Guide to Cross-Cultural Psy- Journal of Social Psychiatry, 37, 71–79. chiatry, American Psychiatric Publishing, Washington, 21. Singh, S. P. (2007) Institutional racism in psychiatry: DC. lessons from inquiries. Psychiatric Bulletin, 31, 5. Eagleton, T. (2000) The Idea of Culture, Blackwell, 363–365. Oxford. 22. Murray, R. M. and Fearon, P. (2007) Searching for 6. Kuper, A. (1999) Culture: The Anthropologist’s racists under the psychiatric bed: Commentary on.. . Account, Harvard University Press, Cambridge, Institutional racism in psychiatry. Psychiatric Bulletin, Massachusetts. 31, 365–366. 7. Boas, F. (1982) Summary of the work of the committee 23. Patel, K. and Heginbotham, C. (2007) Institutional in British Columbia, in A Franz Boas Reader: The racism in mental health services does not imply racism Shaping of American Anthropology (ed. G. Stocking in individual psychiatrists: Commentary on... Institu- Jr), University of Chicago Press, Chicago, Illinois, tional racism in psychiatry. Psychiatric Bulletin, 31, pp. 1883–1911. 367–368. 8. Goodenough, W. H. (1961) Comments on a cultural 24. McKenzie, K. and Bhui, K. (2007) Better mental revolution. Daedalus, 90, 521–528. healthcare for minority ethnic groups – moving away

130 COMPONENTS OF THE SOCIAL WORLD from the blame game and putting patients first: Com- 41. Weiss, M. and Somma, D. (2007) Explanatory models mentary on.. . Institutional racism in psychiatry. Psy- in psychiatry, in Textbook of Cultural Psychiatry (eds chiatric Bulletin, 31, 368–369. D. Bhugra and K. S. Bhui), Cambridge University 25. Singh, S. P. (2007) Institutional racism in psychiatry: Press, Cambridge. author’s response. Psychiatric Bulletin, 31, 370. 42. Kleinman, A. (1980) Patients and Healers in the 26. Westermeyer, J. (1989) Psychiatric Care of Migrants: Context of Culture, University of California Press, A Clinical Guide, American Psychiatric Press, Berkeley, California. Washington, DC. 43. Bhui, K. S. and Bhugra, D. (2002) Explanatory models 27. Meszaros, A. F. (1961) Types of displacement reactions for mental distress: implications for clinical practice among the post revolution Hungarian immigrants. and research. British Journal of Psychiatry, 181, 6–7. Canadian Psychiatric Association Journal, 6, 9–19. 44. Bhui, K. S. and Bhugra, D. (2004) Communication with 28. Berry, J. (1976) Human Ecology and Cognitive Styles, patients from other cultures: the place of explanatory Sage Halstead, New York. models. Advances in Psychiatric Treatment, 10, 29. Bhugra, D. (2000) Migration and cchizophrenia. Acta 474–478. Psychiatrica Scandinavica, (suppl. 407), 68–73. 45. Weiss, M. G., Doongaji, D., Sidhartha, S. et al. (1992) 30. Bhugra, D. (2004) Migration and mental health. Acta The explanatory model interview catalogue (EMIC). British Journal of Psychiatry, 160, 819–830. Psychiatrica Scandinavica, 109 (4), 243–258. 46. Lloyd, K. R., Jacob, K. S., Patel, V. et al. (1998) The 31. Bhugra, D., Bhui, K., Mallett, R. et al. (1999) Cultural identity and its measurement. International Review of development of the Short Explanatory Model Interview Psychiatry, 11, 244–249. (SEMI) and its use among primary care attenders with common mental disorders. Psychological Medicine, 28 32. Bhugra, D. and Ayonrinde, D. (2001) Racial life events (5), 1231–1237. and psychiatric morbidity, in Psychiatry in Multicul- 47. Sheikh, S. and Furnham, S. (2000) A cross-cultural tural Britain (eds D. Bhugra and R. Cochrane), Gaskell, study of mental health beliefs and attitudes towards London, pp. 91–111. seeking professional help. SPPE, 35, 326–334. 33. Redfield, R., Linton, R. and Herskovits, M. J. (1936) 48. McCabe, R. and Priebe, S. (2004) Explanatory models Memorandum on the study of acculturation. American of illness in schizophrenia: comparison of four ethnic Anthropologist, 38, 149–152. groups. British Journal of Psychiatry, 185, 25–30. 34. Berry, J. (2007) Acculturation and identity, in Textbook 49. Ho, T. P. and Luk, C. (1997) Comparison of child of Cultural Psychiatry (eds D. Bhugra and K. S. Bhui), psychiatric patients in hospital and community clinics Cambridge University Press, Cambridge. in Hong Kong. General Hospital Psychiatry, 19, 35. Linton, R. (1945) The Cultural Background of Person- 362–369. ality, Appleton-Century Croft, New York. 50. Pang, K.-Y. (1998) Symptoms of depression in elderly 36. Bhugra, D. (2005) Cultural identities and cultural con- Korean immigrants: narration and the healing process. gruency: a new model for evaluating mental distress in Culture Medicine and Psychiatry, 22, 93–122. immigrants. Acta Psychiatrica Scandinavica, 111 (2), 51. Bhugra, D. and Cochrane, R. (2001) Psychiatry in a 84–93. multi ethnic context, in Psychiatry in Multicultural 37. Bhugra, D. (2001) Acculturation, cultural identity and Britain (eds D. Bhugra and R. Cochrane), Gaskell, mental health, in Psychiatry in Multi-Cultural Britain London. (eds D. Bhugra and R. Cochrane), Gaskell, London. 52. Eisenberg, L. (1977) Disease and illness: distinction 38. Cousins, S. D. (1989) Culture and self perception in between professional and popular ideas of sickness. Japan and the United States. Journal of Personality and Culture Medicine and Psychiatry, 1, 9–23. Social Psychology, 56, 124–131. 53. Hall, E. T. (1959) The Silent Language, Doubleday, 39. Ton, H. and Lim, R. F. (2006) The assessment of Garden City, New York. culturally diverse individuals, in Clinical Manual of Cultural Psychiatry (ed. R. F. Lim), APPJ, Washington, 54. Moodley, R. and Palmer, S. (2006) Race culture and DC. other multiple constructions: an absent presence in psychotherapy, in Race Culture and Psychotherapy 40. Lim, R. F. (2006) Clinical Manual of Cultural Psy- (eds R. Moodley and S. Palmer), Routledge, London. chiatry, APPJ, Washington, DC.

CULTURE AND ITS INFLUENCE ON DIAGNOSIS AND MANAGEMENT 131 55. Bhugra, D. and Bhui, K. S. (2006) Psychotherapy 60. Lin, K.-M., Chen, C.-H., Yu, S.-H. and Wang, S.-C. across the cultural divide, in Race Culture and Psy- (2008) Culture and Ethnicity in Psychopharmacology chotherapy (eds R. Moodley and S. Palmer), Routle- (eds C.-H. Ng et al.), Cambridge University Press, dge, London. Cambridge. 56. Rush, L. (1985) The therapeutic alliance in short term 61. Lin, K.-M. and Poland, R. E. (1995) Ethnicity culture directive therapies, in APA Annual Review 4 (eds R. E. and psychopharamacology, in Psychopharmacology: Hales and A. Frances), APA, Washington, DC, Fourth Generation of Progress (eds F. E. Bloom and D. pp. 562–572. J. Kupfer), Raven Press, New York. 57. Bhugra, D. and Tantam, D. (1998) Psychotherapy, cul- 62. Yu, X. (2008) Complementary medicine in mental ture and ethnicity, in Clinical Topics in Psychotherapy disorders, in Ethno-Psychopharamacology (eds C.-H. (ed. D. Tantam), Gaskell, London, pp. 241–255. Ng, K.-M. Lin, B. S. Singh and E. Chiu), Cambridge 58. Lin, K.-M., Poland, R., Nagasaki, G.et al. (eds) (1993) University Press, Cambridge. Psychopharmacology and Psychobiology of Ethnicity, 63. Ng, C.-H. and Klimidis, S. (2008) Cultural Factors APA Press, Washington, DC. and the Use of Psychotropic Medication (eds C.-H. Ng 59. Ng, C.-H., Lin, K.-M., Singh, B. S. and Shiu, E. (eds) et al.), Cambridge University Press, Cambridge. (2008) Ethnopsychopharamacology, Cambridge Uni- 64. Bhugra, D. (2002) Migration and depression. Acta versity Press, Cambridge. Psychiatrica Scandinavica, (suppl. 418), 67–72.