ORIGINAL�ARTICLE ABSTRACT Objective: To study the risk of psychosis in south Asian communities in Bradford and investigate the role of cannabis as a contributory factor. Study Design: Naturalistic studies based on electronic summary records. Place and Duration of Study: The studies were conducted at the Becklin Centre, St James's University Hospital, Leeds and the University of Leeds, School of Medicine from 2018 to 2020. Material and Method: A service evaluation and research project looking into the role of cannabis included 194 st th patients admitted to acute psychiatry wards at the Becklin Centre between 1 January 2016 and 30 November 2018. Epidemiological study used electronic summary records provided by the Bradford Early Intervention for Psychosis Service of 15-35-year old newly diagnosed cases with first episode psychosis in 2013-15 and local census data to calculate the risks ratios. Results: Compared with indigenous white population, Pakistanis in Bradford had significantly higher risk of psychosis (RR: 1.41, 95% CI 1.07, 1.85*). This trend was also seen in Bangladeshi community (RR 1.72, 95% CI 0.91, 3.28*). Indian community, on the other hand, experienced lower risk (RR 0.54, 95% CI 0.20, 1.27). Conclusion: We found increased risk of psychosis in Pakistani and Bangladeshi communities but not in Indian community. Key Words: Bradford, Ethnic, Psychosis, Risk Ratio. Caribbean in the UK. These studies found that the incidence of schizophrenia was 4-6 times higher than the indigenous white population and was more 5 marked in the second generation. Whilst an increased incidence of schizophrenia has been consistently reported in people of African Caribbean origin who are resident in the UK, the results are less consistent for those of south Asian 6 origin. In the ÆSOP study, the incidence of psychosis, although increased, was raised to a much lesser extent in Asian and non-indigenous white populations. Another study in east London, had found the incidence to be raised in most migrant and minority ethnic groups, however, in Pakistani and Bangladeshi populations, this trend appeared to be 7 evident only in women. Bradford (population 530,000), a Yorkshire city with a rich industrial history of cotton and woollen mills, is home to a number of immigrant communities from the subcontinent including Pakistanis, Bangladeshis and Indians. Pakistanis from Kashmir in the north of Pakistan are the largest ethnic group (110,000) and there are sizeable Indian and Bangladeshi populations. Anecdotally, the prevalence of schizophrenia among Pakistanis in Bradford was said to be 2-5 times higher and since the district has the largest proportion Introduction The association between migrant status and schizophrenia was reported in 1930s among 1 , 2 Norwegian immigrants in New York which suggested that stressful life experiences played a role in the causation of psychosis. An influential paper, 3 in which the Norwegian psychiatrist Odegaard reported the findings of his study in Wisconsin, argued that the increased incidence could be attributed to selective migration of people who were genetically at a higher risk. This was handy munition for anti-immigration politicians of that time and 4 although later studies did not support this claim it stymied further investigations in the field. Interest in the topic was rekindled in the 1980s by findings concerning first and second generation African Migration and Psychosis: Evidence from South Asian Communities in Bradford Cara Beattie , Fraser Ritchie, Alastair Cardno, Tariq Mahmood Correspondence: Dr. Tariq Mahmood Consultant Psychiatrist & Honorary Senior Lecturer Becklin Centre, St James's University Hospital Leeds, United Kingdom E-mail: tariq.mahmood5@nhs.net Department of Psychiatry Becklin Centre, St James's University Hospital Leeds, United Kingdom Funding Source: Leeds & York Partnership NHS Foundation Trust; Conflict of Interest: NIL Received: July 18, 2020; Revised: October 01, 2020 Accepted: October 05, 2020 Migration and PsychosisJIIMC 2020 Vol. 15, No.4 226 (20.3%) of people of Pakistani origin in England, there was an opportunity to study these reports in a 8,9 systematic manner. This was made easier by the existence of an early intervention service (EIPS) in Bradford since 2005, which up until 2016 accepted cases in 15-35 age band and had gathered a substantial case load. According to unpublished EIPS data 29.5% of Bradford EIPS referrals were of Pakistani ethnicity. Migrant status is now a well-recognised risk factor for the development of psychosis; however, its antecedents have not been that well elucidated. Social disadvantage, unemployment and residence in densely populated inner- city areas are applicable 10 to new immigrants however, it is the use of illicit substances which is often under the spotlight. There are reports to suggest that cannabis use has a greater 11 effect in inducing psychosis in urban environments. The evidence, however, is sketchy as there are no systematic studies of migrant populations such as community surveys, possibly due to the difficulty, which any attempt to gather such stigmatising information will encounter. We, therefore, addressed this question indirectly by looking into the drug use histories of patients (ICD-10, F20-29) of various ethnic backgrounds admitted to an acute psychiatry unit in Leeds, a neighbouring Yorkshire city (population 800,000) with sizeable populations of south Asian origin. The findings of this Extended Study Research Project (ESREP) undertaken by two th University of Leeds 4 year medical students are reported in this paper. Material and Methods 8 As reported in an earlier paper by Saleem et al in 2019, EIPS Bradford provided anonymous summary data on the number of 15-35 year old newly diagnosed with first episode psychosis in 2013, 2014 and 2015, grouped by ethnicity, gender and 10-year age bands. Permission to use these data for the current study was granted by the Medical Director of Bradford District Care NHS Foundation Trust. EIPS Bradford uses comprehensive assessment of at risk mental state (CAARMS) interviews to establish the 12 diagnoses according to ICD-10 criteria. Only cases with non-affective first episode psychoses were 13 included. The RevMan 5.3 program was used for analysis and data from the three years 2013-15 were combined. Risk ratios (RRs) with 95% confidence intervals were calculated and RRs with 95% confidence intervals that did not include 1 were regarded as statistically significant at p<0.05, two- tailed. Population data were obtained from 2011 census figures published on the website of the Office of National Statistics. Table I shows the basic demographic data derived from Census 2011, which provides the denominators for relative risk (RR) calculations. Compared with British white in Bradford (23% in 15-35 age band), the three south Asian communities are youthful, as 36 – 48% of their members fall in this age band. Another noteworthy observation is female preponderance in British white, Pakistani and Bangladeshi communities. Indian community, on the other hand, has fewer females. Extended Study Research Project (ESREP) obtained the ethnicity and drug use data from the electronic records of 194 patients admitted with ICD10 (F20- 29) diagnosis of psychosis to four acute adult wards st at the Becklin Centre between 1 January 2016 and th 30 November 2018. Fifteen patients were excluded as no data on cannabis use was available in their electronic case notes. Patients were categorised as either “white British” (n=101, male 72, female 29) or “non-white British” (n=78, male 52, female 26). Parametric and non-parametric tests were used to analyse the data. Research approval was granted by the Leeds & York Partnership Trust Research & Development and ESREP board of the University of Leeds, School of Medicine (R&D 2018-673-SE). Table I: Census 2011 Data for Bradford† 8 †Adapted from Saleem et al Results Table II shows the risk of first episode psychosis for each ethnic group. Compared with UK White population, Pakistanis, and Bangladeshis had higher JIIMC 2020 Vol. 15, No.4 227 Migration and Psychosis incidence rates (50, 71, 87/100000 respectively in 15-35 age band). People of Indian origin, on the other hand, had a lower incidence rate (27/100000). The ESREP carried out by medical students CB and FR found a high rate (53.70%) of cannabis use in all acute patients admitted with a diagnosis of ICD-10 psychosis. Table V compares the rates of cannabis use in white British (47.52%) and non-white British (60.26%) patients and shows a trend towards greater 2 use in non-white British (x =2.864, p 0.091; OR 1.554, 95% CI 0.824,2.933). Table II: Incidence of Psychosis and Risk Ra�o *Sta�s�cally significant at p<0.05 ‡ Trend towards 8 significance †Adapted from Saleem et al Table III shows the risk of psychosis in 10-year age bands for each ethnic group. The risk compared to whites is significantly higher in 25-35 old Pakistanis (RR 1.92, 95% CI 1.29, 2.87) and Bangladeshis (RR 2.64, 95% CI 1.13, 6.19). In contrast to the British White, British-Indian population appears to show a lower risk for this age group. Table III: Risk of Psychosis by Ethnic Group and 10-Year Age Bands *sta�s�cally significant at p<0.05 8 †Adapted from Saleem et al Table IV Shows gender analysis of risk in the four ethnic groups. We see significantly increased risk for Pakistani and Bangladeshi males (RR 1.46, 95% CI 1.16, 1.83; 2.42, 95% CI 1.17, 5.02)) whereas Indian males showed a non-significant lower risk (RR 0.65, 95% CI 0.21, 2.07). Table IV: Risk of Psychosis by Ethnic Group and Gender *sta�s�cally significant at p<0.05 8 †Adapted from Saleem et al Table V: History of Cannabis Use in Acute Psychiatric Admissions ‡ p 0.091 Discussion Higher Risk of Psychosis in Immigrant Communities This naturalistic study confirmed the anecdotal reports of higher incidence of psychosis in Bradford South Asian ethnic groups and found that it is true in the case of British Pakistani and British Bangladeshi communities (RRs 1.41 and 1.72). However, the risk for British Indians is lower (RR 0.54), which is surprising but not unique as it has been reported 14 before. Stillman and McKenzie (2009) report the findings of a study of Tongans, who were selected through a ballot and accepted for migration to New Zealand. Compared with those who were unsuccessful in the ballot, the mental health of migrants, particularly women, was significantly JIIMC 2020 Vol. 15, No.4 228 Migration and Psychosis better. This study, however, was carried out only 11 months after resettlement in New Zealand and it will be interesting to see if the Tongans in New Zealand continue to enjoy the positive effect of migration, which Indian migrants in Bradford have experienced over 50 years. Duration of Untreated Psychosis This is a noteworthy finding that Pakistani and Bangladeshi group has a significantly higher risk in the age band 25-35 years, which suggests late presentation for treatment and is consistent with anecdotal reports that patients are first taken to 15 traditional healers , a practice brought over from their homelands. This practice, unfortunately, 16 prolongs the duration of untreated psychosis which 17 is well known to have an adverse effect on outcome. It may, therefore, pay to promote mental health education amongst ethnic communities and their traditional healers. Gender differential 7 In their east London study Coid et al had found a small increase in the risk of psychosis in Pakistani and Bengali women. We found a higher risk of psychosis in Pakistani and Bengali men, which is more consistent with observations of local mental health practitioners and psychiatrists practising on the subcontinent. Underlying factors Schizophrenia is a serious and enduring illness which causes great suffering to affected individuals and their families and generates a huge societal burden. It is imperative, therefore, that we try to find the causes of increased risk of psychosis in ethnic populations. One factor which distinguishes Bradford Pakistanis from other populations is a high rate of first cousin marriages which is reported to be 18 as high as 70% and has increased from 55% found 19 by Darr & Model in 1988, which was an increase on 33% among the mothers of women included in that 20 study. Dobrussin and colleagues have shown that the risk of developing schizophrenia is elevated in populations with high consanguinity. Therefore, it may be that the higher incidence of schizophrenia among Bradford Pakistani and Bangladeshi communities is partly due to elevated genetic loading associated with cousin marriages. Based on the observation that cannabis use can increase the risk of developing schizophrenia, 21 Hickman et al quantified the theoretical increase in the incidence and prevalence of schizophrenia that may follow a population wide increase in the use of cannabis. Their models predicted that by 2010, cannabis use will lead to 29% increase in incidence of schizophrenia among men. The ESREP project undertaken by Leeds University medical students CB and FR under the supervision of TM and AC addressed this question as to whether cannabis use is a modifiable risk factor for increased psychosis in ethnic populations. They found a trend towards greater cannabis use in non-white British patients, which suggests that a larger study such as a community survey is needed to find a definitive answer to this question. The naturalistic design of these studies has some limitations such as reliance on electronic summary data precludes analysis of individual variations. However, this is offset by the fact that research on populations and themes, which tend not to attract the attention of grant giving bodies, can be carried out by workers such as enthusiastic medical students and junior doctors, who are prepared to give their time out of goodwill to the cause of science. Conclusions Compared with indigenous whites in Bradford, south Asians of Pakistani and Bangladeshi origin are experiencing a higher risk of psychosis, though, the incidence rates do not appear to be as elevated as in those of African and Caribbean origin. However, this does not completely disprove the anecdotal reports of increased prevalence among some ethnic groups which may be due to delayed recognition suggested by elevated incidence rates in 25-35 age band, thus increasing the duration of untreated psychosis and decreasing the chances of better outcome. In view of the above, it will not be out of place to make the customary call for more research to find the definitive answers. However, the fundamental questions such as genetics of schizophrenia may take more time to be answered or they may be un- answerable as they are integral to human condition. In the meantime there is cause for optimism in that some of the underlying factors such as cannabis use are modifiable with health education and adoption of preventive strategies. Acknowledgements We are grateful to the medical directors of Bradford JIIMC 2020 Vol. 15, No.4 229 Migration and Psychosis and Leeds mental health trusts for allowing us to carry out these studies; to Dr Julie Robinson for accommodating the ESREP students in her department; and to Catherine Ding (BDCT) and Tom Evans (LYPFT) for invaluable IT support. REFERENCES 1. Selton JP, Canter-Graee E, Kahn RS. Migration and psychosis. Current Opinion in Psychiatry 2007; 20:111-115. 2. Cantor-Graae E, Selten JP. Schizophrenia and migration: a meta-analysis and review. Am J Psychiatry 2005; 162:12-24. 3. Ødegaard Ø. Emigration and insanity. Acta Psychiatrica Neurologica Scandanavica (Suppl.) 1932; 4, 1-206. 4. Van der Ven E, Dalman C, Wicks S, Allebeck P, Magnusson C, van Os J, Selten JP. Testing degaard's selection hypothesis: a longitudinal cohort study of risk factors for non-affective psychotic disorders among prospective migrants. Psychological Medicine 2015; 45, 727-734. 5. Bhugra D, Leff J, Mallett R, Der G, Corridan B, Rudge S. Incidence and outcome of schizophrenia in Whites, African- Caribbean and Asians in London. Psychological Medicine 1997; 27(4):791-798. 6. Kirkbride JB, Fearon P, Morgan C, et al. Heterogeneity in incidence rates of schizophrenia and other psychotic syndromes: Findings from the 3-Center ÆSOP study. Archives of General Psychiatry 2006; 63(3):250-258. 7. Coid JW, Kirkbride JB, Barker D, et al. Raised incidence rates of all psychoses among migrant groups: findings from the East London first episode psychosis study. Archives of General Psychiatry 2008; 65(11):1250-1258. 8. Saleem M, Ding C, Brewin A, Garnham M, Nazari J, Robinson J, Hosalli P, Inglehearn CF, Cardno AG, Mahmood T. Risk of psychosis in Yorkshire South Asians. Journal of Psychiatric Intensive Care 2019; 15, 2, 117-121; doi:20.20299/ jpi.2019.007. 9. Cheng F, Kirkbride JB, Lennox BR, et al; Administrative incidence of psychosis assessed in an early intervention service in England: first epidemiological evidence from a diverse, rural and urban setting. Psychological Medicine 2011; 41(5):949-58. 10. Bartlett EH, Bourkiza R, Crosland L, Edwards S, Cooke A and Mahmood T. Schizophrenia and bipolar disorder in urban areas. Mental Health & Learning Disabilities Research & Practice 2007; 4: 5-16; https://doi.org /10.5920/ mhldrp.2007.415. 11. Bresnahan M, Begg MD, Brown A, et al. Race and risk of schizophrenia in a US birth cohort: another example of health disparity? International Journal of Epidemiology 2007; 36:751–8. 12. Yung AR, Yuen HP, McGorry PD, et al (2005) Mapping the onset of psychosis: The Comprehensive Assessment of At- Risk Mental States. Australian & New Zealand Journal of Psychiatry 2005; 39(11–12): 964–971.https://doi.org/ 10.1080/j.1440-1614.2005.01714.x. 13. Cochrane Community (2014) Review Manager (RevMan) Version 5.3. The Nordic Cochrane Centre. The Cochrane C o l l a b o ra t i o n , C o p e n h a g e n . h tt p s : / /c o m m u n i t y. cochrane.org /sites/default/files/uploads/inline-files/ RevMan_5.3_User_Guide PDF. 14. Stillman S, McKenzie D, Gibson J (2009) Migration and mental health: Evidence from a natural experiment. Journal of Health Economics 2009, 28, 677-68. 15. Shah I, Khalily MT, Ahmed I, Hallahan B. Impact of conventional beliefs and social stigma on attitudes towards access to mental health services in Pakistan. Community Mental Health Journal, https://doi.org./10.1007/s10597- 018-0310-4. 16. Naqvi HA, Hussain H, Zaman M, Islam M (2009). Pathways to care: Duration of untreated psychosis from Karachi, Pakistan. PLoS One 2009; (4)10:e7409. /journal. Pone. Doi: 10.1371.0007409. 17. Marshall M, Lewis S, Lockwood A, Drake R, Croudace T. Association between duration of untreated psychosis and outcome in cohorts of first episode patients: a systematic review. Archives of General Psychiatry, 62,975- 983.https://doi.org/10.1001/archpsyc.62.9.975. 18. B o r n i n B r a d f o r d . n h s . u k https://www.borninbradford.nhs.uk/. 19. Darr A & Modell B. The frequency of consanguineous marriage among British Pakistanis. Journal of Medical Genetics 1988; 25, 186-190. 20. Dobrussin M, Weitzman D, Levine J, Kremer I, Rietschel M, Maier W, Belmaker RH. The rate of consanguineous marriages among parents of schizophrenic patients in the Arab Bedouin population in Southern Israel. World Journal of Biological Psychiatry 2009; 10(4):334-336. 21. Hickman M, Vickerman P, Macleod J, Kirkbride J, Jones P. Cannabis and schizophrenia: model projections of the impact of the rise in cannabis use on historical and future trends in schizophrenia in England and Wales. Addiction 2007; 102:597–606. JIIMC 2020 Vol. 15, No.4 230 Migration and Psychosis