Ethnicity, social inequality, and mental illness

The relative prevalence and treatment of mental illness among different ethnic groups in Britain is probably one of the most controversial issues in the field of health variations. The Policy Studies Institute, in a study commissioned by the Department of Health, has tackled these complexities and openly addressed the difficulties in the cross cultural assessment of mental illness.¹

The study is based on a national community survey of 5196 people of Caribbean or Asian origin and 2867 white Britons. Ethnicity was assigned on the basis of country of family origin, though the limitations of this approach are acknowledged.² In a two stage interviewing process, initial assessment of mental health relied on structured questionnaires: a cut down clinical interview schedule³ for neurotic disorders and the psychosis screening questionnaire⁴ for psychotic disorders. Second stage interviewing was conducted by ethnically and linguistically matched interviewers using the appropriate translation of version 9 of the present state examination.⁵ A major omission was the absence of the somatisation section of the clinical interview schedule. Similarly, no account was taken of non-Western categories of distress.⁶ However, inclusion criteria were as wide as possible in an attempt to minimise false negatives. The psychosis screening questionnaire has a high sensitivity and specificity but its positive predictive value is poor because the prevalence of psychosis is low and it misses people with a psychotic illness in remission.⁶

Studies of ethnicity and mental illness have previously focused on rates of treated mental illness, primarily in hospital settings, and with an inevitable emphasis on psychosis. Relatively little work has been done in primary care (where 95% of mental illness is treated) and even less in community settings. Hospital based research has consistently shown raised rates of schizophrenia among African Caribbeans compared with the white population.^8–10 In the Policy Studies Institute survey Caribbeans again had a higher rate of psychosis (13 per 1000) than any other group but less than twice that found among whites (8 per 1000). All the differences in rates of psychotic mental illness in this survey were found among women. Caribbean men had the same rate as white men. This finding might accurately reflect community prevalence rates or it may be due to systematic underenumeration and higher attrition rates among Caribbean men, differences in validity and reliability of screening, or differences in pathways to care and treatment of white and Caribbean men.

For the first time Caribbeans were confirmed to have higher rates of depression than whites. However, Caribbeans with depression were far less likely to report receiving medication from their general practitioner. This suggests that depression among this group needs to be better identified and treated within primary care.¹¹

Rates of mental illness among Asians were low, particularly for Bangladeshi and Chinese people, which may be due to the cultural limitations of Western measures of mental illness. Among Asians who were born or received secondary school education in Britain, rates of mental illness were similar to those in their white counterparts. Although young Asian women are more likely to die from suicide than other groups, this study found that they were no more likely to feel suicidal.

Crucially, after adjustment for social status, those in lower social classes had higher rates of mental illness across all groups. Differences in material standard of living made at least some contribution to higher rates of depression and psychosis among Caribbean respondents. White and South Asian single mothers had particularly high rates of mental illness, with a 10% prevalence of depression. Those who were married or cohabiting had the lowest rates. Caribbean single mothers did not, however, have raised rates and the lowest rates were found among single women without young children. These findings suggest that further modelling of the data is required to investigate the effects of socioeconomic and sociodemographic variables and to confirm the findings on psychosis. Such analyses are under way (J Nazroo, personal communication). Further research will be needed to establish the best methods of addressing the role of racism.

Frank Dobson, the secretary of state for health, has stated his commitment to improving the health of black and minority communities and to creating health action zones to tackle health inequalities. The Policy Studies Institute study provides much of the basic epidemiological data to underpin policymaking in these areas. Further research is needed into the recognition and treatment within primary care of common mental disorders among ethnic minorities. Finally, these data suggest that too narrow a focus on ethnicity alone might lead to a downplaying of the important relations between mental illness, ethnicity, gender, and social inequality.