Skip to main content Accessibility help
×
Home

Information:

  • Access
  • Open access

Actions:

      • Send article to Kindle

        To send this article to your Kindle, first ensure no-reply@cambridge.org is added to your Approved Personal Document E-mail List under your Personal Document Settings on the Manage Your Content and Devices page of your Amazon account. Then enter the ‘name’ part of your Kindle email address below. Find out more about sending to your Kindle. Find out more about sending to your Kindle.

        Note you can select to send to either the @free.kindle.com or @kindle.com variations. ‘@free.kindle.com’ emails are free but can only be sent to your device when it is connected to wi-fi. ‘@kindle.com’ emails can be delivered even when you are not connected to wi-fi, but note that service fees apply.

        Find out more about the Kindle Personal Document Service.

        Change in incidence rates for psychosis in different ethnic groups in south London: findings from the Clinical Record Interactive Search-First Episode Psychosis (CRIS-FEP) study
        Available formats
        ×

        Send article to Dropbox

        To send this article to your Dropbox account, please select one or more formats and confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your <service> account. Find out more about sending content to Dropbox.

        Change in incidence rates for psychosis in different ethnic groups in south London: findings from the Clinical Record Interactive Search-First Episode Psychosis (CRIS-FEP) study
        Available formats
        ×

        Send article to Google Drive

        To send this article to your Google Drive account, please select one or more formats and confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your <service> account. Find out more about sending content to Google Drive.

        Change in incidence rates for psychosis in different ethnic groups in south London: findings from the Clinical Record Interactive Search-First Episode Psychosis (CRIS-FEP) study
        Available formats
        ×
Export citation

Abstract

Background

A higher incidence of psychotic disorders has been consistently reported among black and other minority ethnic groups, particularly in northern Europe. It is unclear whether these rates have changed over time.

Methods

We identified all individuals with a first episode psychosis who presented to adult mental health services between 1 May 2010 and 30 April 2012 and who were resident in London boroughs of Lambeth and Southwark. We estimated age-and-gender standardised incidence rates overall and by ethnic group, then compared our findings to those reported in the Aetiology and Ethnicity of Schizophrenia and Other Psychoses (ÆSOP) study that we carried out in the same catchment area around 10 years earlier.

Results

From 9109 clinical records we identified 558 patients with first episode psychosis. Compared with ÆSOP, the overall incidence rates of psychotic disorder in southeast London have increased from 49.4 (95% confidence interval (CI) 43.6–55.3) to 63.1 (95% CI 57.3–69.0) per 100 000 person-years at risk. However, the overall incidence rate ratios (IRR) were reduced in some ethnic groups: for example, IRR (95% CI) for the black Caribbean group reduced from 6.7 (5.4–8.3) to 2.8 (2.1–3.6) and the ‘mixed’ group from 2.7 (1.8–4.2) to 1.4 (0.9–2.1). In the black African group, there was a negligible difference from 4.1 (3.2–5.3) to 3.5 (2.8–4.5).

Conclusions

We found that incidence rates of psychosis have increased over time, and the IRR varied by the ethnic group. Future studies are needed to investigate more changes over time and determinants of change.

Introduction

A higher incidence of psychotic disorders has been consistently reported among black and some other minority ethnic groups, particularly in northern European countries (Fearon et al., 2006; Coid et al., 2008; Veling, 2013). These rates are generally higher compared with rates in the countries of origin or heritage (Jablensky et al., 1992b; Hickling and Rodgers-Johnson, 1995; Bhugra et al., 2000; Selten et al., 2001; Selten et al., 2002). This suggests that social and environmental factors in host countries underpin the high rates. In addition, incidence rates are not similarly elevated across all minority ethnic groups. For example, in the UK the highest reported rates are for black ethnic groups, with incidence rate ratios (IRR) ranging from 2.1 to 18.2 for black Caribbean (Lloyd et al., 2005; Cheng et al., 2011) and from 2.5 to 11.9 for black African (Lloyd et al., 2005; Kirkbride et al., 2008). In the Netherlands, reported relative risks (RRs) of psychosis range from 2.3 to 7.8 for Moroccan populations (Veling et al., 2006; Zandi et al., 2010) and 1.4 to 4.0 for Surinamese populations (Selten et al., 2001; Veling et al., 2007). It is unclear whether these rates have changed over time.

Early research on rates of psychotic disorders in minority ethnic groups was characterised by considerable methodological heterogeneity. For example, many studies used hospital admissions to identify those with a first episode psychotic disorder, which meant that findings were confounded by other factors, such as mode of contact, since not all individuals with a psychotic disorder are admitted to hospitals (Kendell et al., 1993). Such a design may introduce systematic bias that may exacerbate differences between ethnic groups. More recent studies have brought a measure of methodological rigour, such as population-based studies that include both hospital and community patients (Fearon et al., 2006; Coid et al., 2008; Cheng et al., 2011; Kirkbride et al., 2017). Further, some previous studies grouped all black patients and therefore do not allow estimates of rates within quite different black Caribbean and black African groups.

While some studies have estimated the incidence of psychosis specifically among black Caribbean populations (Hutchinson et al., 1996; Harrison et al., 1997), to date, only three epidemiological studies in the UK have investigated incidence rates and rate ratios among black African and black Caribbean groups separately, namely the Aetiology and Ethnicity of Schizophrenia and Other Psychoses (ÆSOP), the East London First Episode Psychosis and the Social Epidemiology of Psychoses in East Anglia studies (Fearon et al., 2006; Coid et al., 2008; Kirkbride et al., 2017). All have reported an increased incidence of psychosis in black African and black Caribbean populations compared with white British populations.

There are of course considerable differences between the different migrant populations. For example, some minority ethnic groups are more integrated into UK society than others. For instance, the migration of black African people to the UK is more recent compared with black Caribbeans, whose migration to the UK was most notable during the 1940s and 1950s (Mason, 1995; Chamberlain, 2002). This is an important marker of population change, which may have an impact on the mental health of the population. Given the limited research investigating rates of psychosis among black African and black Caribbean populations separately, far less is known about whether the elevated rates persist in these two groups. If the elevated incidence is persistent, this further strengthens the urgent need for public health strategies to address these disparities.

ÆSOP was a large prospective incidence study of psychotic disorders among minority ethnic groups between 1997 and 1999 in three catchment areas in England, namely, southeast London, Nottingham and Bristol (Fearon et al., 2006). The sample comprised first episode cases aged 16–64 years, the majority (66%) of whom were drawn from the southeast London site. This study found that the incidence of psychosis was higher among minority ethnic groups compared with white British. ÆSOP provides a methodological template to re-examine the evidence on incidence among ethnic minority populations, particularly the black African and black Caribbean groups in southeast London. In this current study, we addressed the question of whether the increased incidence of psychosis among these groups had changed 10 years later.

We therefore set out to compare rates in 2010–2012 in southeast London with those reported in 1997–1999 as part of the ÆSOP study. We used data from one of the largest anonymous electronic health record systems in Europe, covering an ethnically diverse population in the UK to estimate the incidence of psychosis by age, gender and ethnicity over a two year period, modelled on methods used in previous face-to-face epidemiological studies such as ÆSOP.

Methods

Study design and population at risk

Using the South London and Maudsley NHS Foundation Trust (SLaM) Biomedical Research Centre (BRC) Clinical Records Interactive Search (CRIS) system (Stewart et al., 2009), we sought to identify all individuals with a first episode psychotic disorder (in ICD-10: F20-F29, F30-F33 psychotic codes) who presented to mental health services in the SLaM between 1 May 2010 and 30 April 2012 and who were resident in the London boroughs of Lambeth and Southwark. The first year of this study was conducted as part of the European Union Gene-Environment Interactions study (Jongsma et al., 2018).

Study setting and participants

The study was conducted within two inner city areas in southeast London, UK. The areas covered were the London boroughs of Lambeth (total population, 303 086) and Southwark (total population, 288 283) (ONS, 2011b), served by SLaM. These areas have large minority ethnic populations, principally black African (28.0%) and black Caribbean (15.7%) (ONS, 2011b). SLaM is one of the largest mental health providers in Europe serving a population over 1.3 million across four south London boroughs (Stewart et al., 2009). Since April 2006, full electronic health records have been operational in SLaM. Between 2007 and 2008, the CRIS was built, which provides a fully anonymised copy of SLaM electronic records (Perera et al., 2016). CRIS contains over 300 000 patient records from hospital and community services. The clinical information documented in CRIS appears in two forms i.e. structured fields (e.g. dates, demographic and diagnosis) and unstructured free text fields (e.g. clinical information in medical notes and correspondence) (Perera et al., 2016). We interrogated the CRIS database for clinical and demographic information to screen all potentially eligible patients for inclusion in the study.

Case identification

First episode psychosis cases were ascertained through a 3-stage manual screening of CRIS clinical records between May 2010 and April 2012. Firstly, a combination of information from the CRIS structured and free text fields was used to ascertain probable cases of psychosis. We used the structured language query (Netz et al., 2001; Tulloch, 2013) to interrogate and extract information in CRIS based on our inclusion criteria i.e. using defined search terms (date, postcode, age, symptoms-psychos*; psychot*, delusion*, voices, hallucinat* and diagnosis). This returned a set of patient records, which were individually screened by a team of researchers using the Screening Schedule for Psychosis (Jablensky et al., 1992a). Secondly, two primary researchers reviewed all the included cases from the first stage screen to ensure cases met all inclusion criteria. Thirdly, discrepant or ambiguous cases were resolved by consensus with the principal investigator (CM) (Oduola et al., 2019).

Inclusion and exclusion criteria

Inclusion and exclusion criteria were the same as those used in the Aetiology and Ethnicity in Schizophrenia and Other Psychoses study (Fearon et al., 2006). Cases were included if they were: resident in the London boroughs of Lambeth or Southwark; aged 18–64 years (inclusive); experiencing psychotic symptoms, during the study periods, as assessed by the Screening Schedule for Psychosis (Jablensky et al., 1992a). Exclusion criteria were: (a) evidence of psychotic symptoms caused by an organic cause, (b) transient psychotic symptoms resulting from acute intoxication and (c) evidence of the previous contact with services for psychotic symptoms.

Data collection and instruments

We screened for psychotic symptoms using the Screening Schedule for Psychosis (Jablensky et al., 1992a). Socio-demographic data were collected using the Medical Research Council Socio-demographic Schedule (MRC-SDS) (Mallett, 1997). The MRC-SDC classifies ethnic groups according to the UK Office for National Statistics (ONS) ethnic categories (ONS, 2011b). The Personal and Psychiatric History Schedule (WHO, 1996), adapted for case notes, was used to determine the date of onset and the Operational Criteria Checklist for Psychotic Illness (McGuffin et al., 1991) to assess psychotic symptoms.

Reliability

A number of steps were taken to ensure the reliability of screening and data collection procedures. Firstly, researchers were trained in the inclusion and exclusion criteria and the application of these. Interrater reliability was assessed for case identification, whereby researchers who were blind to each other's ratings, swapped their screened cases and repeated the screening procedure. A κ of 0.78, p < 0.001 was achieved, indicating substantial agreement.

We verified patients' residence using the Public Health England postcode widget (Public Health England, 2004). We then linked the middle layer super output area (MSOA) data for the catchment area (ONS, 2011b) to individual patients to ensure their addresses mapped to the boundaries of the study area. The MSOA is a geographic hierarchy designed for reporting of small area statistics that therefore provides neighbourhood-level information (including postcodes) which fits within the boundaries of a local authority (ONS, 2011b) By linking the patients' addresses to the MSOA, we were able to confirm that cases were resident in the study catchment area.

Data

Ethnicity was self-ascribed and recorded in clinical records. Where this information was missing, ethnicity was ascribed independently by researchers using all available information from the free-text field in clinical records, including country of birth, nationality, language spoken at home, parents' country of birth, geographical region (e.g. Saharan and sub-Saharan Africa) and religious group, as recommended by the Office for National Statistics (ONS, 2011a). A high inter-rater reliability was achieved between three researchers, who independently extracted and rated ethnicity information on 89 cases (κ score = 0.87, p < 0.001), indicating substantial agreement.

We coded ethnicity using the MRC-SDC (Mallett, 1997) according to the 18 categories used in the 2011 census. For analytical purposes, we collapsed the ethnic groups into seven categories to match those used in the ÆSOP study (Fearon et al., 2006) as follow: white British, black Caribbean (black Caribbean and other black), black African, Asian (Indian, Pakistani, Bangladeshi, Chinese), white non-British (white Irish, white Gypsy, white Other), other (Arab, Any Other Ethnic group) and mixed (all mixed groups).

Ethical approval

The CRIS system was approved as an anonymised dataset for secondary analysis by the Oxfordshire Research Ethics Committee (reference 08/H0606/71). Local approval for this study was obtained from the CRIS Oversight Committee at the BRC SLaM (reference: 09-041).

Statistical analysis

Stata version 12 (StataCorp, 2011) was used to analyse the data. Populations were estimated from the 2011 UK Census and stratified by age (5 year age-band i.e. 18–19; 20–24; 25–29; 30–34; 35–39; 40–44; 45–49; 50–54; 55–59; 60–64), gender and ethnicity. Age-gender-standardised incidence rates of psychotic disorders were calculated by direct standardisation, using the population of England and Wales in 2001, as was used in the ÆSOP study. Crude IRR were calculated and then adjusted for confounders (age and gender) using Poisson regression. Finally, we stratified by age-band and gender and estimated IRR for each ethnic group using white British as the reference group.

Results

Sample

Our searches of CRIS retrieved 9109 potentially eligible patients who presented to services during the 2-year study period. Of these, 558 met our inclusion criteria, 8549 screened negative (i.e. 5324 did not have a psychotic disorder, 2956 had previous episodes of psychosis and 359 were either not resident in the study area or were older or younger than the age of inclusion) (see online Supplementary Fig. 1). In our sample, the mean age was 33.3 (s.d. = 10.6) and there was a similar proportion of men (52.3%) and women (47.3%) (Table 1). Compared with the population at risk in Lambeth and Southwark, cases were younger and more were from black Caribbean and black African ethnic groups.

Table 1. Demographic characteristics of CRIS-FEP study population and crude and age-gender standardised incidence per 100 000 person-years with 95% CI of all psychoses

Incidence rates of psychosis by age and gender

The overall age-gender-standardised incidence rate of psychotic disorders was 63.1 per 100 000 person-years at risk (PY) (95% confidence interval (CI) 57.3–69.0). There was no difference in incidence rates between men (64.5 per 100 000 PY) and women (61.8 per 100 000 PY). Figure 1 shows incidence rates by age and gender. Rates were higher among men up to age 35–39 years, while women had higher rates than men from age 40 years onward.

Fig. 1. Incidence rates of all psychoses by age and gender: CRIS-FEP study.

Incidence rates and rate ratios by ethnic group

The overall age-and-gender standardised incidence rate of psychotic disorders reported in the ÆSOP southeast London site was 49.4 per 100 000 (95% CI, 43.6–55.3) (Kirkbride et al., 2006), which is modestly lower than the present CRIS-FEP study. Table 2 shows the comparisons between ÆSOP and the present study in the incidence rate and adjusted rate ratios of psychotic disorders by the ethnic group. In the ÆSOP study, higher incidence rates were reported among all minority ethnic groups compared with the white British group (20.2 per 100 000). These were most markedly raised for black Caribbean (140.8 per 100 000; adj. IRR 6.7; 95% CI 5.4–8.3), black African (80.6 per 100 000; adj. IRR 4.1; 3.2–5.3) and ‘other’ (55.0 per 100 000; adj. IRR 2.6 95% CI 1.7–3.9) groups. Rates were modestly raised for the Asian (31.6 per 100 000; adj. IRR 1.5 95% CI 0.9–2.4), white non-British (33.1 per 100 000; adj. IRR 1.6; 95% CI 1.1–2.2) and ‘mixed’ (45.9 per 100 000; adj. IRR 2.7; 95% CI 1.8–4.2) groups.

Table 2. Comparisons between ÆSOP and CRIS-FEP for age-gender standardised incidence per 100 000 person-years with adjusted IRR for all psychoses

a ÆSOP was conducted from 1997 to 1999 (Fearon et al., 2006), the CRIS-FEP study was conducted from 2010 to 2012.

p ⩽ 0.1; *p ⩽ 0.05; **p ⩽ 0.01.

These findings have changed to varying extents in the present CRIS-FEP study. The most striking was that rates for the white British group were higher (44.3 per 100 000 PY), while for the black Caribbean group (100.4 per 100 000 PY; adj. IRR 2.8; 95% CI 2.1–3.6) these were notably lower than those reported in the ÆSOP study. Also notable was the reduced magnitude of risk observed among the ‘mixed’ group from nearly three-fold (adj. IRR 2.7; 95% CI 1.8–4.2) to less than two-fold (adj. IRR = 1.4; 95% CI 0.9–2.1). While rates have increased for the Asian and white non-British groups, the rate ratios were broadly in keeping with the ÆSOP study. For the black African group, there was very little change in their RR from (adj. IRR 4.1; 95% CI 3.2–5.3) to (adj. IRR 3.59; 95% CI 2.8–4.5). By contrast, rate and rate ratios were considerably higher for the ‘other’ group (126.4 per 100 000; 95% CI 85.5–167.2) in the present study, but these need to be treated cautiously given the wide CIs.

Table 3 compares adjusted IRR in minority ethnic groups between the ÆSOP and CRIS-FEP studies, stratified by gender. In the ÆSOP study, raised IRR for black Caribbean, black African, ‘mixed’ and ‘other’ ethnic groups were present for both men and women. Whilst elevated rates by the ethnic group were also observed in the CRIS-FEP study, there were striking differences by gender when compared with the ÆSOP study. For example, there were notable reduction in the RRs for black Caribbean men: from nearly six-fold (adj. IRR 5.6; 95% CI 4.2–7.5) to less than threefold (adj. IRR 2.7; 95% CI 1.8–4.1), black Caribbean women: from eight-fold (adj. IRR 8.1; 95% CI 5.9–11.1) to less than threefold (adj. IRR 2.6; 95% CI 1.8–3.8) and black African women: from four-fold (adj. IRR 4.2; 95% CI 2.8–6.4) to less than threefold (adj. IRR 2.8; 95% CI 2.0–4.0). There was no change for black African men.

Table 3. Comparisons between ÆSOP and CRIS-FEP for adjusted IRR with 95% CI in ethnic minority groups for all psychoses, stratified by gender

p ⩽ 0.1; *p ⩽ 0.05; **p ⩽ 0.01.

Age-specific IRR by ethnic group

Table 4 shows age-specific IRR by the ethnic group; we compared these to those reported in the ÆSOP study. Findings from ÆSOP showed that IRR were elevated for both black African and black Caribbean populations across all age groups. However, our findings from the CRIS-FEP study indicated that these were only elevated for the black Caribbean and black African groups in the age-bands 20–24, 25–29, 30–34, 35–39 and 40–44 years. Moreover, in the CRIS-FEP study, patients aged 25–29 and of Asian (adj. IRR 2.3; 95% CI 1.1–5.1) and mixed (adj. IRR 3.3; 95% CI 1.5–7.4) ethnic groups also had elevated rates. In addition, older (55–59 years) Asian patients (adj. IRR = 4.6; 95% CI 1.0–20.9) had higher rates, but no other ethnic minority group showed an increase in rate ratios in this particular age group.

Table 4. Age-specific and gender-adjusted IRR with 95% CI in ethnic minority groups for all psychoses: comparisons between ÆSOP and CRIS-FEP

p ⩽ 0.1; *p ⩽ 0.05; **p ⩽ 0.01.

Discussion

Main findings

We carried out a comprehensive study of the treated incidence of psychotic disorders using a large database of electronic health records of people who presented to secondary mental health services. Our results suggest that the overall incidence of psychotic disorders in southeast London has gone up between 1996–1999 and 2010–2012 and that rates have changed over time in all ethnic groups. The first, and perhaps most striking, findings were that rates were lower in the black Caribbean and higher in the white British populations; as a consequence, the rate ratios have narrowed. These changes were more marked by gender. That is, the magnitude of rate ratios particularly reduced for black Caribbean men and women and for black African women.

Relationship with other previous studies

In keeping with some recent studies of incidence of psychosis in the UK, our results confirm that rates of psychosis are higher in black African and, to a lesser degree, black Caribbean populations, compared with white British population (Coid et al., 2008; Kirkbride et al., 2017). Regarding age and gender, our findings are consistent with previous studies (Coid et al., 2008; Cheng et al., 2011). We found that rates were at their peak among men between the ages 18–35 years, which mirrors previous findings (Fearon et al., 2006; Kirkbride et al., 2006; Coid et al., 2008). We observed elevated rates among women over the age of 39; the peak risk of psychosis for women has been reported to be close to the time of menopause (Grigoriadis and Seeman, 2002). Our findings also suggest that rate incidence of psychosis is not in declining, in keeping with some (Boydell et al., 2003; Kirkbride et al., 2009; Kirkbride et al., 2012), but not other, earlier studies (Geddes et al., 1993; Munk-Jorgensen and Mortensen, 1993).

Methodological considerations

The primary methodological consideration is the comparability of methods of case ascertainment in AESOP and CRIS-FEP. It is possible that bias arising from differences in case ascertainment between the two studies (i.e. face-to-face v. case register) may explain the apparent changes in rates and rate ratios over time. Indeed, case register studies do tend to produce higher estimates of incidence. However, several points and lines of reasoning suggest this is unlikely to fully account for our findings. For example, it is the case register studies from Canada, the Netherlands and Scandinavian countries, that use linked national registers and rely on clinical diagnoses, that tend to produce higher estimates (Hogerzeil et al., 2014; Anderson et al., 2018; Jongsma et al., 2019); they do not allow for researchers to check case records to determine caseness and therefore may be over-inclusive. As noted earlier, in CRIS-FEP, we painstakingly reviewed records of every potential case to determine inclusion, with borderline cases being decided by consensus, thereby closely mirroring the approach used in ÆSOP. We also used identical inclusion and exclusion criteria in the two studies. Further, the changes in incidence we observed go in different directions for different ethnic groups (e.g. increase for white British, decrease for black Caribbean). It is not clear why the use of electronic records to identify cases, would over-include some and under-include other groups. While not discounting the possibility that using electronic records, rather than face to face screening of services, may explain some of the reported changes, we think it unlikely to be the sole explanation. Given what we know about the influence of environmental factors on psychoses and variations in rates across populations, it is reasonable to expect that rates will change over time. Our findings suggest that there are variations by the ethnic group in changes over time and, in doing so, challenge the usual framing of this issue, which implies elevated rates in minority ethnic groups are universal and static over time. There are other methodological considerations. For instance, our findings for the ‘Other’ ethnic group need to be considered with caution. It is a highly heterogeneous group and includes some newly added ethnic groups in the 2011 UK Census, e.g. Arab, and so it is possible that the denominator data may not be accurate; therefore, this finding may be artificially inflated if under enumeration is present in the denominator population. In addition, patients in our Asian group are heterogeneous, which in addition to Indian, Pakistani and Bangladeshi people also included those of Chinese origin.

A further potential limitation is in the lack of adjustment for socioeconomic variables, which may possibly explain differences in incidence rates by the ethnic group. However, where investigators have adjusted for socioeconomic factors in previous studies, this made little difference to the estimates (Kirkbride et al., 2008; Hollander et al., 2016; Kirkbride et al., 2017). Further, we deliberately based our case identification on broadly defined psychosis so as not to miss cases. Further research exploring rates of psychotic disorders by diagnosis is needed.

Interpretations

It is possible that changes in incidence rates are related to changes in the demography of the source population, in mental health service provision, and in the distribution of environmental risk factors over time in the catchment area. The white British population in Lambeth and Southwark changed notably between the censuses of 2001 and 2011, when the proportion of White British people decreased from 50% to 39% (ONS, 2011b). This population change has potentially significant implications for changes in overall incidence rates (Kendell et al., 1993). Furthermore, the catchment area of southeast London contains some of the most socioeconomically deprived wards in England and Wales. Inevitably, many white British people are exposed to and may experience similar levels of socioeconomic disadvantages experienced by the minority ethnic groups (Social Mobility Commission, 2016), which may explain the higher rates in this group compared with 10 years ago. We demonstrated this in our earlier paper, where we found that the proportion of white British patients with first episode psychosis in southeast London who were unemployed rose from 53.7% to 63.9% between 1997–1999 and 2010–2012 (Oduola et al., 2019).

Further, UK government policy changes and major new investments in mental health services in the last two decades (Dept. of Health and Social Care, 1999) may have impacted on incidence rates of psychosis. For example, early intervention (EI) for psychosis services was established in southeast London in the early 2000s (Craig et al., 2004), the latter stage of the ÆSOP study. A central tenet of the EI services is early detection and reduction in delays to receiving treatment for people at an early stage in the psychotic illness. EI services are also known to work collaboratively with other agencies such as the criminal justice system and emergency rooms to identify people at the early phase of psychosis (Jarrett et al., 2012). Consequently, these services may identify more patients than was previously possible, which may explain the overall higher incidence rates. However, this is unlikely to fully explain the observed changes. As noted above, changes in rates varied by the ethnic group; therefore, to fully explain, for example, increased rates in the white British group and decreased rates in the black Caribbean, EI services would need to be more engaging for white people; they would also have to be less engaging than mainstream services for black Caribbean (but not black African).

Another possibility is that this increase is linked to the rise in the use of high-potency cannabis, which has been linked to the onset of psychosis (Marconi et al., 2016; Murray et al., 2017). In particular, a recent case-control study of incidence of psychosis in south London, showed that cases who used skunk-like cannabis daily had up to five times increased odds of psychotic disorder compared with those who never used cannabis (Di Forti et al., 2015). Around the same time of our study, the samples of cannabis seised by the Metropolitan Police in London area had higher concentration of tetrahydrocannabinol than those in the late 1990s when AESOP was carried out (Freeman et al., 2014; Di Forti et al., 2015).

The reduced incidence of psychosis in the black Caribbean population that we observed may be explained by considering generational status. Although not measured specifically here, the majority of the black Caribbean group in our sample and in the UK population are second or third generation, since migration among black Caribbean populations into the UK was highest post World War II, mostly in the 1950s (Chamberlain, 2002; Coid et al., 2008; Jones, 2011). Therefore, it is possible that social networks for the black Caribbean group may have improved, which has been documented to be a protective factor from mental illness (Bhugra and Becker, 2005). We showed in our recent study that black Caribbean patients are now more likely to live with family and friends (41.7% v. 61.1%), compared with 15 years ago (Oduola et al., 2019). In addition, the lower IRR among the black Caribbean group are partly explained by the higher incidence rate that we observed among the white British group in this study compared with the ÆSOP study. Since they are the reference group, a rise or drop in the incidence rate of psychosis for the white British groups will affect the IRR in the other ethnic groups.

It is also possible that other sociodemographic changes in London may have influence rates of psychosis in some ethnic groups. According to the Indices of Deprivation (2015), London has relatively low levels of deprivation in education (Ministry of Housing and Local Goverment, 2015). Data from our recent study showed that the proportion of first episode psychosis patients with university level of education rose from 14.9% in AESOP to 27.5% in CRIS-FEP (Oduola et al., 2019), which may suggest an improvement in this well documented social risk factor of psychosis. However, this may at most only partly explain our findings, since elevated rates are still present in some minority ethnic groups but not others, as discussed above.

Furthermore, the fact that we observed little or no change in the overall magnitude of risk of psychosis among black African patients may be explained in a number of ways. First, for the black African patients, it is possible that the well documented indicators of social disadvantage (Morgan et al., 2009), isolation (Reininghaus et al., 2008) and discrimination (Reininghaus et al., 2010) may be driving the excess of psychosis observed among the black African and other minority ethnic populations in our study. According to the Office for National Statistics (2005), there has been an increase of 2.4% in the black African population in the catchment area since the 2001 census (ONS, 2011b), indicating this is an active migrant group. It has also been reported that recent migrants may be predisposed to such experiences on arrival in the new environment since there may be tension with regard to cultural bereavement and culture shock i.e. a discrepancy between expectations and achievement in the host country (Bhugra and Becker, 2005). A recent study from Sweden illuminated this issue, where the authors investigated the rates of schizophrenia and non-affective psychosis in refugee and non-refugee migrants compared with native Swedish populations. They found that refugee migrants had higher rates of psychosis than non-refugee migrants, but the risk was particularly greater for those from sub-Saharan Africa irrespective of refugee status (Hollander et al., 2016). Despite not controlling for country of birth or generation status in this study, our findings here for the black African patients are consistent with previous studies.

Despite the increase in the white non-British population in the UK, particularly since the expansion of the European Union in 2004 and ease of migration from Eastern Europe, we found weak evidence of increased risk of psychosis between the white non-British and white British ethnic groups. Our findings in this group are in keeping with two recent studies (Kirkbride et al., 2017; Schofield et al., 2017), which both found that non-British white ethnic groups in the UK and migrants from elsewhere in Europe to Denmark were not at increased risk of psychosis. This may also be because the white non-British patients may experience a less stressful acculturation process and lower perceived discrimination (Schofield et al., 2017) within Europe.

Conclusions

Our results suggest that incidence rates of psychosis are still elevated among minority ethnic groups. However, the magnitude of IRR varies considerably by the ethnic group. The findings here also highlight that black African and other ethnic groups remain disproportionately at higher risk of psychosis compared with their white British counterparts. Our findings suggest that concerted efforts are needed to ameliorate health outcomes for our minority populations, and could also help inform commissioners, policymakers and healthcare providers in allocating resources to delivery effective mental health services and public health strategies. While our study has shed light on possible demographic changes over time to explain change in incidence of psychotic disorders, future studies are needed to investigate more changes over time and determinants of change.

Supplementary material

The supplementary material for this article can be found at https://doi.org/10.1017/S0033291719003234.

Acknowledgements

We are grateful to Mr Matthew Broadbent, Dr Alex Tulloch and Mr Hitesh Shetty for advice on usage of and data extraction from BRC-CRIS.

Author contributions

SO, CM, FB, CGA, RM and TC conceived the study. The data extraction was carried out by SO, FB, JT and CGA. SO analysed data and drafted the manuscript with guidance from CM, JDM and TC. All authors were involved in the interpretation of the data and in commenting on and revising drafts of the paper.

Financial support

This work was supported by UK Medical Research Council (Ref: G0500817), the Wellcome Trust (Grant Number: WT087417), the European Union [European Community's Seventh Framework Program (grant agreement No. HEALTH-F2-2009-241909) (Project EU-GEI)] and the Department of Health via the National Institute for Health Research (NIHR) Specialist Biomedical Research Centre for Mental Health award to South London and Maudsley NHS Foundation Trust (SLaM) and the Institute of Psychiatry at King's College London. JD is a clinician scientist fellow funded by the Academy of Medical Sciences working together with the Health Foundation.

Conflict of interest

None.

References

Anderson, KK, Norman, R, MacDougall, AG, Edwards, J, Palaniyappan, L, Lau, C and Kurdyak, P (2018) Estimating the incidence of first-episode psychosis using population-based health administrative data to inform early psychosis intervention services. Psychological Medicine 49, 19.
Bhugra, D and Becker, MA (2005) Migration, cultural bereavement and cultural identity. World Psychiatry 4, 1824.
Bhugra, D, Hilwig, M, Mallett, R, Corridan, B, Leff, J, Neehall, J and Rudge, S (2000) Factors in the onset of schizophrenia: a comparison between London and Trinidad samples. Acta Psychiatrica Scandinavica 101, 135141.
Boydell, J, Van Os, J, Lambri, M, Castle, D, Allardyce, J, McCreadie, R and Murray, R (2003) Incidence of schizophrenia in south-east London between 1965 and 1997. The British Journal of Psychiatry 182, 4549.
Chamberlain, M (2002) Caribbean Migration: Globalized Identities. London: Routledge.
Cheng, F, Kirkbride, JB, Lennox, BR, Perez, J, Masson, K, Lawrence, K, Hill, K, Feeley, L, Painter, M, Murray, GK, Gallagher, O, Bullmore, ET and Jones, PB (2011) Administrative incidence of psychosis assessed in an early intervention service in England: first epidemiological evidence from a diverse, rural and urban setting. Psychological Medicine 41, 949958.
Coid, JW, Kirkbride, JB, Barker, D, Cowden, F, Stamps, R, Yang, M and Jones, PB (2008) Raised incidence rates of all psychoses among migrant groups: findings from the East London first episode psychosis study. [Erratum appears in Arch Gen Psychiatry. 2009 Feb;66(2):161]. Archives of General Psychiatry 65, 12501258.
Craig, TK, Garety, P, Power, P, Rahaman, N, Colbert, S, Fornells-Ambrojo, M and Dunn, G (2004) The Lambeth Early Onset (LEO) Team: randomised controlled trial of the effectiveness of specialised care for early psychosis. British Medical Journal 329, 10671069.
Dept. of Health and Social Care (1999) National service framework for mental health Available at: https://www.gov.uk/government/publications/quality-standards-for-mental-health-services.
Di Forti, M, Marconi, A, Carra, E, Fraietta, S, Trotta, A, Bonomo, M, Bianconi, F, Gardner-Sood, P, O'Connor, J, Russo, M, Stilo, SA, Marques, TR, Mondelli, V, Dazzan, P, Pariante, C, David, AS, Gaughran, F, Atakan, Z, Iyegbe, C, Powell, J, Morgan, C, Lynskey, M and Murray, RM (2015) Proportion of patients in south London with first-episode psychosis attributable to use of high potency cannabis: a case-control study. The Lancet. Psychiatry 2, 233238.
Fearon, P, Kirkbride, JB, Morgan, C, Dazzan, P, Morgan, K, Lloyd, T, Hutchinson, G, Tarrant, J, Fung, WL, Holloway, J, Mallett, R, Harrison, G, Leff, J, Jones, PB and Murray, RM (2006) Incidence of schizophrenia and other psychoses in ethnic minority groups: results from the MRC AESOP Study. Psycholological Medicine 36, 15411550.
Freeman, TP, Morgan, CJ, Hindocha, C, Schafer, G, Das, RK and Curran, HV (2014) Just say ‘know’: how do cannabinoid concentrations influence users' estimates of cannabis potency and the amount they roll in joints? Addiction 109, 16861694.
Geddes, JR, Black, RJ, Whalley, LJ and Eagles, JM (1993) Persistence of the decline in the diagnosis of schizophrenia among first admissions to Scottish hospitals from 1969 to 1988. British Journal of Psychiatry 163, 620626.
Grigoriadis, S and Seeman, MV (2002) The role of estrogen in schizophrenia: implications for schizophrenia practice guidelines for women. Candian Journal of Psychiatry 47, 437442.
Harrison, G, Glazebrook, C, Brewin, J, Cantwell, R, Dalkin, T, Fox, R, Jones, P and Medley, I (1997) Increased incidence of psychotic disorders in migrants from the Caribbean to the United Kingdom. Psychological Medicine 27, 799806.
Hickling, FW and Rodgers-Johnson, P (1995) The incidence of first contact schizophrenia in Jamaica. The British Journal of Psychiatry 167, 193196.
Hogerzeil, S, van Hemert, A, Rosendaal, F, Susser, E and Hoek, H (2014) Direct comparison of first-contact versus longitudinal register-based case finding in the same population: early evidence that the incidence of schizophrenia may be three times higher than commonly reported. Psychological Medicine 44, 34813490.
Hollander, AC, Dal, H, Lewis, G, Magnusson, C, Kirkbride, JB and Dalman, C (2016) Refugee migration and risk of schizophrenia and other non-affective psychoses: cohort study of 1.3 million people in Sweden. British Medical Journal 352, 1030.
Hutchinson, G, Takei, N, Fahy, TA, Bhugra, D, Gilvarry, C, Moran, P, Mallett, R, Sham, P, Leff, J and Murray, RM (1996) Morbid risk of schizophrenia in first-degree relatives of white and African-Caribbean patients with psychosis. British Journal of Psychiatry 169, 776780.
Jablensky, A, Sartorius, N, Ernberg, G, Anker, M, Korten, A, Cooper, J, Day, R and Bertelsen, A (1992 a) Schizophrenia: manifestations, incidence and course in different cultures a World Health Organization Ten-Country Study. Psychological Medicine. Monograph Supplement 20, 197.
Jablensky, A, Sartorius, N, Ernberg, G, Anker, M, Korten, A, Cooper, JE, Day, R and Bertelsen, A (1992 b) Schizophrenia: manifestations, incidence and course in different cultures. A World Health Organization Ten-Country Study. Psychol Med Monogr Suppl 20, 197.
Jarrett, M, Craig, T, Parrott, J, Forrester, A, Winton-Brown, T, Maguire, H, McGuire, P and Valmaggia, L (2012) Identifying men at ultra-high risk of psychosis in a prison population. Schizophrenia Research 136, 16.
Jones, P (2011) Schizophrenia & Psychotic Disorders: current epidemiological highlights. Cutting Edge Psychiatry in Practice 1, 311.
Jongsma, HE, Gayer-Anderson, C, Lasalvia, A, Quattrone, D, Mule, A, Szoke, A, Selten, JP, Turner, C, Arango, C, Tarricone, I, Berardi, D, Tortelli, A, Llorca, PM, de Haan, L, Bobes, J, Bernardo, M, Sanjuan, J, Santos, JL, Arrojo, M, Del-Ben, CM, Menezes, PR, Velthorst, E, Murray, RM, Rutten, BP, Jones, PB, van Os, J, Morgan, C and Kirkbride, JB (2018) Treated incidence of psychotic disorders in the multinational EU-GEI Study. JAMA Psychiatry 75, 3646.
Jongsma, HE, Turner, C, Kirkbride, JB and Jones, PB (2019) International incidence of psychotic disorders, 2002–17: a systematic review and meta-analysis. Lancet Public Health 4, e229e244.
Kendell, RE, Malcolm, DE and Adams, W (1993) The problem of detecting changes in the incidence of schizophrenia. British Journal of Psychiatry 162, 212218.
Kirkbride, JB, Fearon, P, Morgan, C, Dazzan, P, Morgan, K, Tarrant, J, Lloyd, T, Holloway, J, Hutchinson, G, Leff, JP, Mallett, RM, Harrison, GL, Murray, RM and Jones, PB (2006) Heterogeneity in incidence rates of schizophrenia and other psychotic syndromes: findings from the 3-center AESOP study. Archives of General Psychiatry 63, 250258.
Kirkbride, JB, Barker, D, Cowden, F, Stamps, R, Yang, M, Jones, PB and Coid, JW (2008) Psychoses, ethnicity and socio-economic status. British Journal of Psychiatry 193, 1824.
Kirkbride, JB, Croudace, T, Brewin, J, Donoghue, K, Mason, P, Glazebrook, C, Medley, I, Harrison, G, Cooper, JE, Doody, GA and Jones, PB (2009) Is the incidence of psychotic disorder in decline? Epidemiological evidence from two decades of research. [Erratum appears in Int J Epidemiol. 2009 Dec;38(6):1753]. International Journal of Epidemiology 38, 12551264.
Kirkbride, JB, Errazuriz, A, Croudace, TJ, Morgan, C, Jackson, D, Boydell, J, Murray, RM and Jones, PB (2012) Incidence of schizophrenia and other psychoses in England, 1950–2009: a systematic review and meta-analyses. PLoS One 7, e31660.
Kirkbride, JB, Hameed, Y, Ioannidis, K, Ankireddypalli, G, Crane, CM, Nasir, M, Kabacs, N, Metastasio, A, Jenkins, O, Espandian, A, Spyridi, S, Ralevic, D, Siddabattuni, S, Walden, B, Adeoye, A, Perez, J and Jones, PB (2017) Ethnic minority status, age-at-immigration and psychosis risk in rural environments: evidence from the SEPEA study. Schizophrenia Bulletin 43, 12511261.
Lloyd, T, Kennedy, N, Fearon, P, Kirkbride, J, Mallett, R, Leff, J, Holloway, J, Harrison, G, Dazzan, P, Morgan, K, Murray, RM and Jones, PB (2005) Incidence of bipolar affective disorder in three UK cities: results from the AESOP study. British Journal of Psychiatry 186, 126131.
Mallett, R (1997) MRC Sociodeomgraphic Schedule. Institute of Psychiatry, King's College London: Section of Social Psychiatry.
Marconi, A, Di Forti, M, Lewis, CM, Murray, RM and Vassos, E (2016) Meta-analysis of the association between the level of cannabis use and risk of psychosis. Schizophrenia Bulletin 42, 12621269.
Mason, D (1995) Race and Ethnicity in Modern Britain. Oxford University Press: London.
McGuffin, P, Farmer, A, Harvey, I. (1991) A polydiagnostic application of operational criteria in studies of psychotic illness: development and reliability of the opcrit system. Archives of General Psychiatry 48, 764770.
Ministry of Housing & Local Goverment (2015) Indices of Deprivation 2015. Available at https://www.gov.uk/government/statistics/english-indices-of-deprivation-2015.
Morgan, C, Fisher, H, Hutchinson, G, Kirkbride, J, Craig, TK, Morgan, K, Dazzan, P, Boydell, J, Doody, GA, Jones, PB, Murray, RM, Leff, J and Fearon, P (2009) Ethnicity, social disadvantage and psychotic-like experiences in a healthy population based sample. Acta Psychiatrica Scandinavica 119, 226235.
Munk-Jorgensen, P and Mortensen, PB (1993) Is schizophrenia really on the decrease? European Archives of Psychiatry and Clinical Neuroscience 242, 244247.
Murray, RM, Englund, A, Abi-Dargham, A, Lewis, DA, Di Forti, M, Davies, C, Sherif, M, McGuire, P and D'Souza, DC (2017) Cannabis-associated psychosis: neural substrate and clinical impact. Neuropharmacology 124, 89104.
Netz, A, Chaudhuri, S, Fayyad, U and Bernhardt, J (2001) Integrating data mining with SQL databases: OLE DB for data mining. In Data Engineering, 2001. Proceedings. 17th International Conference on, pp. 379387.
Oduola, S, Craig, TKJ, Das-Munshi, J, Bourque, F, Gayer-Anderson, C and Morgan, C (2019) Compulsory admission at first presentation to services for psychosis: does ethnicity still matter? Findings from two population-based studies of first episode psychosis. Social Psychiatry and Psychiatric Epidemiology 54, 871881.
ONS (2011 a) Ethnic group, national identity and religion: measuring equality: A guide for the collection and classification of ethnic group, national identity and religion data in the UK. Office for National Statistics. Available at https://www.ons.gov.uk/methodology/classificationsandstandards/measuringequality/ethnicgroupnationalidentityandreligion.
ONS (2011 b) Official labour market statistics. ONS. Available at https://www.nomisweb.co.uk/reports/lmp/la/1946157253/report.aspx?c1=2013265927&c2=1946157256.
Perera, G, Broadbent, M, Callard, F, Chang, C-K, Downs, J, Dutta, R, Fernandes, A, Hayes, RD, Henderson, M and Jackson, R (2016) Cohort profile of the South London and Maudsley NHS Foundation Trust Biomedical Research Centre (SLaM BRC) Case Register: current status and recent enhancement of an Electronic Mental Health Record-derived data resource. BMJ open 6, e008721.
Public Health England (2004) Postcode lookup widget. Available at https://www.ndtms.org.uk/emids/cgi-bin/ons_locale.cgi.
Reininghaus, UA, Morgan, C, Simpson, J, Dazzan, P, Morgan, K, Doody, GA, Bhugra, D, Leff, J, Jones, P, Murray, R, Fearon, P and Craig, TK (2008) Unemployment, social isolation, achievement-expectation mismatch and psychosis: findings from the AESOP Study. Social Psychiatry and Psychiatric Epidemiology 43, 743751.
Reininghaus, U, Craig, TK, Fisher, HL, Hutchinson, G, Fearon, P, Morgan, K, Dazzan, P, Doody, GA, Jones, PB, Murray, RM and Morgan, C (2010) Ethnic identity, perceptions of disadvantage, and psychosis: findings from the AESOP study. Schizophrenia Research 124, 4348.
Schofield, P, Thygesen, M, Das-Munshi, J, Becares, L, Cantor-Graae, E, Agerbo, E and Pedersen, C (2017) Neighbourhood ethnic density and psychosis – Is there a difference according to generation? Schizophrenia Research 95, 501505.
Selten, JP, Veen, N, Feller, W, Blom, JD, Schols, D, Camoenie, W, Oolders, J, van der Velden, M, Hoek, HW, Rivero, VM, van der Graaf, Y and Kahn, R (2001) Incidence of psychotic disorders in immigrant groups to The Netherlands. British Journal of Psychiatry 178, 367372.
Selten, J-P, Cantor-Graae, E, Slaets, J and Kahn, RS (2002) Ødegaard's selection hypothesis revisited: schizophrenia in Surinamese immigrants to the Netherlands. American Journal of Psychiatry 159, 669671.
Social Mobility Commission (2016) Ethnicity, gender and social mobility. Available at https://www.equallyours.org.uk/social-mobility-commission-report-ethnicity-gender-and-social-mobility/.
StataCorp L (2011) STATA-IC (Version 12) College Station. Texas: StatCorp LP.
Stewart, R, Soremekun, M, Perera, G, Broadbent, M, Callard, F, Denis, M, Hotopf, M, Thornicroft, G and Lovestone, S (2009) The South London and Maudsley NHS Foundation Trust Biomedical Research Centre (SLAM BRC) case register: development and descriptive data. BMC Psychiatry 9, 51.
Tulloch, AD (2013) Using SQLCRIS. In NIHR Biomedical Research Centre, South London and Maudsley NHS Foundation Trust. London: South London and Maudsley NHS Foundation Trust.
Veling, W (2013) Ethnic minority position and risk for psychotic disorders. Current Opinion in Psychiatry 26, 166171.
Veling, W, Selten, J-P, Veen, N, Laan, W, Blom, JD and Hoek, HW (2006) Incidence of schizophrenia among ethnic minorities in the Netherlands: a four-year first-contact study. Schizophrenia Research 86, 189193.
Veling, W, Selten, JP, Susser, E, Laan, W, Mackenbach, JP and Hoek, HW (2007) Discrimination and the incidence of psychotic disorders among ethnic minorities in The Netherlands. International Journal of Epidemiology 36, 761768.
WHO (1996) Personal and Psychiatric History Schedule. Geneva: World Health Organisation.
Zandi, T, Havenaar, JM, Smits, M, Limburg-Okken, AG, van Es, H, Cahn, W, Algra, A, Kahn, RS and van den Brink, W (2010) First contact incidence of psychotic disorders among native Dutch and Moroccan immigrants in the Netherlands: influence of diagnostic bias. Schizophrenia Research 119, 2733.