Intended for healthcare professionals


Non-affective psychosis in refugees

BMJ 2016; 352 doi: (Published 15 March 2016) Cite this as: BMJ 2016;352:i1279
  1. Cornelius Katona, medical director
  1. Helen Bamber Foundation, Bruges Place, 15-20 Baynes Street, London NW1 0TF, UK
  1. cornelius{at}

Risk is exacerbated by adverse experiences after arrival, including detention, unemployment, and racism

In 2015, 244 million people (3.3% of the world’s population) lived outside their country of origin. This represents an increase of 39% since 2000.1 2 The decision to migrate may be made for economic betterment or (in the case of “refugees”) to escape war, persecution, or natural disaster. Such motives are not of course mutually exclusive. Substantial evidence shows that the risk of non-affective psychosis is increased (by a factor of about 2.5) in migrants compared with the indigenous population.3

In a linked paper (doi:10.1136/bmj.i1030), Hollander and colleagues argue that this increase is due predominantly to exposure to psychosocial adversities.4 They used national register data to carry out a cohort study of more than 1.3 million people in Sweden, in which risk of non-affective psychosis was compared not only between people born in Sweden and migrants to Sweden but also between refugees and non-refugees within the migrant group. They hypothesised that, because of their increased vulnerability to psychosocial adversity, incidence of non-affective psychosis would be particularly high in the refugee group. The study was restricted to relatively young people (born in 1984 or later). Follow-up was to the end of 2011 or to emigration, death, or a diagnosis of non-affective psychosis. The authors’ primary hypothesis was confirmed: incidence rates for non-affective psychosis were 385 per million in those born in Sweden, 804 per million in non-refugee migrants, and 1264 per million in refugees.

The increased rate in refugees compared with other migrants was more pronounced in men and remained evident after allowance for age, baseline disposable income, and baseline population density (reflecting urban or rural origin). It is perhaps of particular importance that the elevated rate in refugees was significant for all geographical areas of origin except sub-Saharan Africa. The authors suggest that this reflects the high rates of social adversity experienced by migrants from sub-Saharan Africa as a whole, as opposed to the more specific increase in adversity experienced by refugees from other parts of the world.

The most obvious implication to be drawn from these findings is that refugees are particularly vulnerable to developing non-affective psychoses and that, as the authors state, there is a “need to take the early signs and symptoms of psychosis into account in refugee populations as part of any clinical mental health service response to current global humanitarian crises.” In my view, however, there are also more fundamental lessons to be learnt.

An important research question raised by Hollander and colleagues’ study concerns the validity of current diagnostic systems. Extensive literature (recently reviewed by OConghaile and DeLisi5) attests to the frequent occurrence of psychotic symptoms in people with a primary diagnosis of post-traumatic stress disorder (PTSD). Evidence from preliminary genome-wide association studies in PTSD and from US veterans suggest significant overlap in genetic vulnerability to schizophrenia and to PTSD.6 7 Cultural and linguistic barriers may further impede definitive diagnosis. The clinical challenges facing mental health professionals working with refugees and asylum seekers should not be underestimated. Diagnostic uncertainties should not, however, result in under-detection and under-treatment of serious mental health problems in this vulnerable group.

As Hollander and colleagues also point out,4 one of the key limitations of their study is the lack of information on post-migration risk factors such as racism and discrimination. Consideration also needs to be given to the challenges that asylum seekers face during what is often a prolonged and distressing process. These factors may include institutional detention, inability to work (and resultant deskilling and loss of self esteem), destitution, and difficulty in accessing health and social care. The relatively “generous” asylum policies practised in Sweden may have reduced the effect of these factors in Hollander and colleagues’ study.

Both detention of immigrants and prolonged uncertainty about immigration are known to increase vulnerability to mental illness as a whole8 9 10—this may well include increased vulnerability to non-affective psychosis. Another aspect highlighted in the context of the current European refugee crisis has been the long and perilous journey many refugees face before reaching the country in which they seek protection.11 12

The potential effect of a fair and responsive asylum system on mental health outcomes in general and incidence of non-affective psychosis in particular remains to be determined. Meanwhile, however, a robust mental health response to the refugee “crisis” must lie in a combination of clinical vigilance, recognition of vulnerability factors, and, above all, a determination to minimise the aggravating effects of post-migration experiences.


  • research, doi: 10.1136/bmj.i1030
  • Competing interests: I have read and understood the BMJ policy on declaration of interests and declare the following interests: I am medical director of the Helen Bamber Foundation, a human rights charity working with victims of torture and other forms of inhumane and degrading treatment; I am the Royal College of Psychiatrists’ lead on asylum and refugee mental health; I have an honorary chair at University College London (the academic base for two of the authors of the linked paper).

  • Provenance: Commissioned; not externally peer reviewed.


View Abstract