Effects of ethnic density on the risk of compulsory psychiatric admission for individuals attending secondary care mental health services: evidence from a large-scale study in England.

Orla Mcbride,Jan Scott,David Crepaz-Keay,Scott Weich,Eva Cyhlarova,Helen Parsons,Liz Twigg,Patrick Keown,Craig Duncan,Kamaldeep Bhui

doi:10.1017/s0033291721001768

Abstract

Black, Asian and minority ethnicity groups may experience better health outcomes when living in areas of high own-group ethnic density - the so-called 'ethnic density' hypothesis. We tested this hypothesis for the treatment outcome of compulsory admission. Data from the 2010-2011 Mental Health Minimum Dataset (N = 1 053 617) was linked to the 2011 Census and 2010 Index of Multiple Deprivation. Own-group ethnic density was calculated by dividing the number of residents per ethnic group for each lower layer super output area (LSOA) in the Census by the LSOA total population. Multilevel modelling estimated the effect of own-group ethnic density on the risk of compulsory admission by ethnic group (White British, White other, Black, Asian and mixed), accounting for patient characteristics (age and gender), area-level deprivation and population density. Asian and White British patients experienced a reduced risk of compulsory admission when living in the areas of high own-group ethnic density [odds ratios (OR) 0.97, 95% credible interval (CI) 0.95-0.99 and 0.94, 95% CI 0.93-0.95, respectively], whereas White minority patients were at increased risk when living in neighbourhoods of higher own-group ethnic concentration (OR 1.18, 95% CI 1.11-1.26). Higher levels of own-group ethnic density were associated with an increased risk of compulsory admission for mixed-ethnicity patients, but only when deprivation and population density were excluded from the model. Neighbourhood-level concentration of own-group ethnicity for Black patients did not influence the risk of compulsory admission. We found only minimal support for the ethnic density hypothesis for the treatment outcome of compulsory admission to under the Mental Health Act.

Highlights

The prevalence of psychiatric disorders, and in particular psychotic disorders, is considerably higher (∼1.5–3 times) among ethnic minority groups living in the UK compared to the White British majority (Kirkbride et al, 2012; Morgan, Knowles, & Hutchinson, 2019; Qassem et al, 2015; Weich et al, 2004)
Significant differences in the distribution of age, gender, compulsory admission, area-level deprivation and population density were evident across the ethnic groups
We sought to test empirically the ethnic density hypothesis as it relates to the risk of compulsory admission to psychiatric hospitals under specific sections of the Mental Health Act (MHA) in England for 1.2 million patients who received treatment provided by National Health Service (NHS)-funded mental health services during 2010–2011

Summary

Introduction

The prevalence of psychiatric disorders, and in particular psychotic disorders, is considerably higher (∼1.5–3 times) among ethnic minority groups living in the UK compared to the White British majority (Kirkbride et al, 2012; Morgan, Knowles, & Hutchinson, 2019; Qassem et al, 2015; Weich et al, 2004). Asian and minority ethnicity groups may experience better health outcomes when living in areas of high own-group ethnic density – the so-called ‘ethnic density’ hypothesis. We tested this hypothesis for the treatment outcome of compulsory admission. Multilevel modelling estimated the effect of own-group ethnic density on the risk of compulsory admission by ethnic group (White British, White other, Black, Asian and mixed), accounting for patient characteristics (age and gender), area-level deprivation and population density. Asian and White British patients experienced a reduced risk of compulsory admission when living in the areas of high own-group ethnic density [odds ratios (OR) 0.97, 95% credible interval (CI) 0.95–0.99 and 0.94, 95% CI 0.93–0.95, respectively], whereas White minority patients were at increased risk when living in neighbourhoods of higher own-group ethnic concentration (OR 1.18, 95% CI 1.11–1.26).

Methods

Results

Conclusion