Different rates of mental illness have been reported in ethnic groups in the UK (Nazroo, 1997). Early work was criticised because of methodological flaws but more rigorous studies have confirmed high community prevalence rates of depression in both South Asian and African—Caribbean populations (Nazroo, 1997), high incidence and prevalence rates of psychosis in African—Caribbean groups (see Bhugra & Cochrane, 2001, for review), and higher rates of suicide in some South Asian groups (Neeleman et al, 1997) compared with the White British population. Similarly high rates have not been reported in the countries of origin of these groups (Hickling & Rodgers-Johnson, 1995; Patel & Gaw, 1996), which has led to a search for possible causes within the UK.
The search for biological causes has not been fruitful. For instance, an association between biological risk factors and the rates of psychosis in African—Caribbean people has not been demonstrated (Sharpley et al, 2001). A number of social and service-related risk factors have been proffered to explain differences in illness rates, including socio-economic status, the role of psychiatry in social control, the validity of European illness models in ethnic minority groups, and the use of universalist rather than relativist approaches to psychopathology and diagnosis. These have rarely been investigated in depth and may be better studied using qualitative approaches rather than the quantitative epidemiological approaches that are currently relied on.
One social risk factor frequently identified by service users and increasingly by academics is racism (Sharpley et al, 2001).
Racism is a form of discrimination that stems from the belief that groups should be treated differently according to phenotypic difference. It is widespread in the UK (Modood et al, 1997).
Racism has many forms; direct attack is less common than perceived discrimination in interpersonal communication, or inequity in the receipt of services or justice. It is easier to measure discriminatory acts such as racist attacks, but some believe that everyday minor incidents or slights (micro-aggressions) and the perception that society is discriminatory may have a greater impact on the individual's health (Laveist, 1996). Measurement of perceived racism is complicated by its possible overlap with paranoid ideation and an external locus of control. However, ‘paranoia’ may represent a healthy coping strategy in a discriminatory environment (Sharpley et al, 2001).
The impact of discrimination is influenced by individual factors (such as socio-economic status, skin colour, and coping style), context (for example, where the incident happens, the extent of integration within an area, and the history of the minority group) as well as macro-economics, political ideologies and history (King & Williams, 1995). Longitudinally, racism produces and perpetuates socio-economic difference, and so controlling for this in analyses may decrease a valid association.
LINKS BETWEEN RACISM AND MENTAL ILLNESS
Despite this complexity there have been efforts to investigate possible links between racism and illness.
Research has mainly conceptualised racism as a stressor. An individual's perception of society as racist and the experience of everyday minor acts of discrimination are thought to constitute a chronic stressor. Individual, more overtly racist acts are considered as life events (acute stressors) that are superimposed on this chronic stress (Bhugra & Cochrane, 2001).
In the USA, interpersonal discrimination has been associated with increased rates of hypertension, depression and stress; poorer self-rated health; and more reported days spent unwell in bed (Krieger, 2000). In the UK, both Burke (1984) and Fernando (1984) have documented relationships between depression and life events thought to be due to racism. Burke reported a 1.5-fold increased incidence of depression in a community sample of ‘West Indians’ living in Birmingham compared with Whites. However, this research has been criticised because of poor diagnostic reliability and outmoded analysis (Bhugra & Cochrane, 2001). There are case reports (but no clinical syndrome) describing the development of post-traumatic stress disorder after racist attacks.
Gilvarry et al (1999) investigated life events in African and African—Caribbean patients with psychosis; these patients were as likely to suffer life events as Whites but more likely to attribute them to racism (Gilvarry et al, 1999). Recent qualitative work has reported that patients of Caribbean origin with psychosis were more likely to attribute their problems to racism than to their mental illness (Chakraborty et al, 2002).
The Fourth National Survey of Ethnic Minorities provided UK evidence of a cross-sectional association between interpersonal racism and mental illness (Karlsen & Nazroo, 2002). A nationally representative sample of 5196 persons of Caribbean, African and Asian origin were asked about racial discrimination in the preceding year. Those who had experienced verbal abuse were 3 times more likely to be suffering from depression or psychosis. Those who had experienced a racist attack were nearly 3 times more likely to suffer from depression and 5 times more likely to suffer from psychosis. Those who said their employers were racist were 1.6 times more likely to suffer from a psychosis. There is no published longitudinal research that has investigated an association between racial discrimination and mental illness.
The ecological level
In the USA, when ethnic minority groups form a smaller proportion of the population in an area, they are more likely to suffer from mental illness (Laveist, 1996); Halpern (1993) partially replicated this in the UK. Boydell et al (2001) reported a 2-fold increase in the incidence of psychosis in people from ethnic minority groups in London wards with a low percentage of ethnic minority inhabitants compared with those living in areas with high ethnic minority population densities. Similarly, Neeleman et al (2001) reported that emergency attendance for parasuicide in African—Caribbean and South Asian patients was related to ethnic population density. This relationship was complex. An inverted U-shaped graph with the relative rate of presentation with attempted suicide by ethnic minority groups being lowest at the extremes of residential segregation best fits the data. These effects may reflect complex interactions between exposure to discrimination, social support, socio-economic factors and social capital.
Community-level racist attitudes may be related to mental illness in minority groups. A US study reported a dose-response relationship between the level of racial disrespect (the belief that the plight of African Americans was their own fault rather than a complex socio-economic problem) on a state-by-state basis and all-cause mortality in African Americans (Kennedy et al, 1997).
There has been discussion about institutional racism in medicine in the UK since the Macpherson report into the death of Stephen Lawrence (McKenzie, 1999). Institutional racism is often indirect. An institution may not set out to discriminate but through its rules, may have this very effect. Fernando (1991) has argued that since European psychiatry developed when racist doctrines were rife in Western culture, the ideology of racism became incorporated into it as a discipline. He concludes that the emphasis on an individualised pathology, with insufficient attention paid to social pressures such as race and culture, renders psychiatry a racist institution.
In the UK, widespread discriminatory social policy may influence the rates of mental illnesses, their presentation and outcome. Institutional discrimination is also reflected in the lack of research for an effective response to these societal influences which, in turn, perpetuates social disparity. Community groups in the UK claim that much has been published about increased rates of illness, but there have been few interventions. Interventions largely occur at a health service level, although a public health approach is likely to be more effective in decreasing incidence rates.
MEDICALISING POLITICAL STRUGGLE
Mental health research into the effects of racial discrimination runs the risk of medicalising appropriate social struggle and distress. Focusing on those discriminated against in this way may only serve to maintain the institution's power over the victimised group, while running the risk of stereotyping the group's identity as nothing more than a response to racism. It has been argued that there should be a closer examination of those bodies that discriminate, rather than their victims. Sashidharan (1993) has voiced concern that focusing on psychological differences between Blacks and Whites rather than on the power disparities inherent in a predominantly racist society serves only to reinforce the idea of racial differences.
If racial harmony is considered the aim of a civilised society, then action towards producing this should not depend on proving that racism is an ill — this is already agreed. Although we have highlighted the effects of racial discrimination, we recognise that oppression and the intolerance of differences target numerous groups, such as elderly people and women. Our specificity does not imply superiority.
There is research linking racial discrimination to mental illness but in the UK there has been little rigorous scientific work to support this. Recent cross-sectional research provides strong evidence of an association between perceived racial discrimination, psychosis and depression in ethnic minority groups. However, there are no longitudinal studies to support a causal relationship.
If racism is a cause of mental illness in ethnic minority groups, a public health approach may be needed to counter this. Acknowledgement and understanding of institutional racism in psychiatry and linked services such as housing, benefits and education could form the basis for secondary and tertiary preventive efforts. More research will be needed on this subject in order for health care trusts and practitioners to develop a competency in understanding and working with these issues.
- Received June 14, 2001.
- Revision received October 17, 2001.
- Accepted November 8, 2001.
- © 2002 Royal College of Psychiatrists