“Migration and mental illness” – Dinesh Bhugra and Peter Jones

Dinesh Bhugra and Peter Jones write for Advances in Psychiatric Treatment:

Human beings have moved from place to place since time immemorial. The reasons for and the duration of these migrations put extraordinary stress on individuals and their families. Such stress may not be related to an increase in mental illness for all conditions or to the same extent across all migrant groups. In this paper, we provide an overview of some observations in the field of migration and mental health, hypothesise why some individuals and groups are more vulnerable to psychiatric conditions, and consider the impact of migration experiences on provision of services and care.


Migration is the process of social change whereby an individual moves from one cultural setting to another for the purposes of settling down either permanently or for a prolonged period. Such a shift can be for any number of reasons, commonly economic, political or educational betterment. The process is inevitably stressful and stress can lead to mental illness.

The preparation the migrants undertake, their acceptance by the new host community and the process of migration itself are some of the macro-factors in the origin of mental disorders. The micro-factors include personality traits, psychological robustness, cultural identity, and the social support and acceptance of others in their own ethnic group.

Migration to the UK has had many peaks. In the 20th century there were several: the first was the refugee influx during and surrounding the Second World War; the second occurred in the 1960s, when able-bodied young men and women were recruited from the former colonies in order to fill jobs created by the belated expansion of the post-war economy. A decade later, following the upheaval in East Africa, a large number of individuals migrated with their families, en masse.

Migrants can be classified using several different criteria – such as legal definition. There needs to be a distinction between actual settlers and migrant workers. The reasons for migration as defined by Rack (1982) include both ‘push’ and ‘pull’ factors. Settlers, as well as political exiles, asylum seekers and refugees, may well have to deal with very stringent legal procedures, which will test their psychological stamina. If there are conditions akin to war, refugees may even face rougher times. Factors like language, communication and social networks will play a role in the processes of dealing with initial adversity, settling down and assimilation.

The migratory process can be seen as three stages. The first, pre-migration, is when the individuals decide to migrate and plan the move. The second involves the process of migration itself and the physical transition from one place to another, involving all the necessary psychological and social steps. The third stage, post-migration, is when the individuals deal with the social and cultural frameworks of the new society, learn new roles and become interested in transforming their group (see Fig. 1). Primary migrants may be followed by others. Once they have settled down and had children, the second generation is not a generation of migrants, but it will have some similar experiences in terms of cultural identity and stress.

Fig. 1

Fig. 1

Stages of migration

We must advise a note of caution here. Although we are using the terms ‘migrant’, ‘migration’ and ‘psychological disorders’, these do not explain the heterogeneity inherent within each setting. Not all migrants have the same experiences or even the same reasons for migration and certainly the new societies’ responses are not likely to be similar either.

We shall now review some of the key studies related to specific psychiatric conditions and highlight some key assessment and treatment strategies.


Ödegaard (1932) reported that migrant Norwegians to the USA had higher rates of schizophrenia (with a peak occurring 10–12 years post-migration). This study has been cited frequently as indicating that all migrant groups have high rates of schizophrenia. Sashidharan (1993) argues cogently that this model should not be applied to other ethnic minority groups in the UK without critical evaluation.

Several studies in the 1980s and 1990s showed that rates of schizophrenia were higher among migrant groups to the UK compared to native Whites (see Bhugra, 2000, for a review). Cochrane & Bal (1987) observed that migrants had higher rates of admission than the native population. Similar high rates of schizophrenia have been reported among the migrant populations from The Netherlands (Selten & Sijben, 1994). Whether rates reported are from admission or community, some common themes emerge. First, incidence of schizophrenia in the African–Caribbean population is 2.5–14.6 times higher than in the White population (Harrison et al, 19881997). Second, the rates among Asians are not as elevated and are not consistently high. King et al(1994) reported that the rates among Asians were no different from those in the White population from the same catchment area.

The key difference between these two studies was that, while King et alcollected their data from an area where population density of Asians was thinner, Bhugra et al (1997) collected data from the London Borough of Ealing where, in the Southall catchment area, Asians form 50% of the population.

A number of hypotheses may be put forward to explain these high rates, some of which are listed in Box 1 and examined below. Hypotheses 1–4 were postulated by Cochrane & Bal (1987); Hypothesis 6 is discussed in the section Ethnic density v. social isolation.

 Box 1.

Six hypotheses for the higher incidence of schizophrenia in migrant groups

  1. Sending countries have high rates of schizophrenia

  2. People with schizophrenia are predisposed to migrate

  3. Migration produces stress, which can initiate schizophrenia

  4. Migrants are misdiagnosed with schizophrenia

  5. Different symptom patterns are presented by migrants

  6. Increased population density of ethnic migrant groups can show elevated rates of schizophrenia


Hypothesis 1: Factors in countries of origin

For a considerable period it was believed that the countries from which the individuals had migrated had high rates of schizophrenia, thereby suggesting a biological vulnerability leading to the illness. Until the 1990s very few studies had looked at the rates in sending countries, especially the Caribbean. With the Determinants of Outcome in Severe Mental Disorders (DOSMD) study (Jablensky et al, 1992) and studies from Jamaica (Hickling & Rodgers-Johnson 1995), Trinidad (Bhugra et al, 1996) and Barbados (Mahy et al, 1999), it was observed that rates of narrow-definition schizophrenia were not elevated compared with populations who had migrated. This suggests that biological causation is less likely, although biological vulnerability to environmental exposures cannot entirely be ruled out. Biological factors such as neurodevelopmental abnormalities, pregnancy and birth complications, and genetic vulnerability have not been reported consistently as having different prevalence in ethnic groups.

Hypothesis 2: Predisposition to migrate

The process of ‘selective’ migration has been put forward as a plausible hypothesis to explain the high incidence rates of schizophrenia among migrants, in that more vulnerable people are more restless and rootless. This superficially attractive hypothesis cannot be supported, on several accounts. First, high rates of schizophrenia are found in second generation rather than original migrants. Bhugra et al (1997) reported significantly different rates among older Asian females who were not primary migrants, thereby making it less likely that illness had predisposed them to move. Second, the physical process of migration and dealing with official immigration procedures is such a difficult and stressful task that individuals with schizophrenia are unlikely to complete it. Third, if this were the case, rates would be high among all migrant groups, which they are clearly not.

Hypothesis 3: Migratory stress

The key question here is whether migration itself acts as a stressor and produces elevated rates of schizophrenia, or whether the stressors occur later. If it were a straightforward association then rates of common mental disorders will be elevated among all migrant groups, which is clearly not the case (see below). Furthermore, as Ödegaard (1932) had demonstrated, rates of schizophrenia were elevated more than 10–12 years after migration, thus making it less likely that the actual process of migration is contributory. The finding of raised rates in the second generation but not in the first also argues against this.

However, the stress and chronic difficulties of living in societies where racism is present both at individual and institutional levels may well contribute to ongoing distress. These factors may also interact with social class, poverty, poor social capital, unemployment and poor housing. For example, Bhugra et al (1997) showed that 80% of the African–Caribbeans in their sample of cases of first-onset schizophrenia were unemployed, compared with 40% of Whites and Asians. However, such differential rates of unemployment and high rates of schizophrenia among African–Caribbeans may not be causally related and the high rates of unemployment cannot be explained away by general unemployment only.

Hypothesis 4: Misdiagnosis

The notion that misdiagnosis alone can explain the high rates of schizophrenia among the African–Caribbeans has caught the public and professional imaginations. However, this belief cannot be true. If misdiagnosis is the sole explanation, why is it that Asians are not misdiagnosed as readily, bearing in mind language differences? By using standardised definitions and assessments, as well as operational criteria in research, researchers should be able to reduce any discrepancy in diagnosis. As the same criteria are used in the sending countries, it would be likely that patients in those countries are being misdiagnosed too, and if they are not, why not?

Hypothesis 5: Symptom patterns

A key hypothesis that has been excluded from Cochrane & Bal’s (1987) list is that of symptom patterns of schizophrenia. There is considerable evidence from DOSMD that there are indeed cultural differences between inception rates of narrow- and broad-definition schizophrenias (Jablensky et al, 1992). Rates of narrow-definition schizophrenia vary across cultures within a very narrow band. It is possible that these two subtypes of schizophrenia and different migrant groups show different increases in specific symptoms…

To continue reading the article, follow this link


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s