Abstract
Purpose
Studies have found that the Mental Health Act is used disproportionally among minoritised ethnicities. Yet, little research has been conducted to understand how the intersectionality of ethnicity with sociodemographic factors relates to involuntary admission. This study aimed to investigate whether an association between ethnicity and involuntary hospitalisation is altered by variations in service-users’ sociodemographic positions.
Methods
A retrospective cohort study using records from the South London and Maudsley identified 18,569 service-users with a first episode of hospitalisation in a 13-year period. Logistic regression was used to calculate odds ratios for involuntary hospitalisation across ethnicities while adjusting for sociodemographic (age, gender, area-level deprivation, homelessness, and migration) and clinical factors (psychiatric diagnosis and HoNOS scores). Interaction analysis was conducted to identify intersectional effects between ethnicity and sociodemographic variables, potentially modifying the odds ratios of involuntary admission across ethnic groups.
Results
Increased odds of involuntary hospitalisation compared to White British service-users were observed among 10 of the 14 ethnicities, with around, or just under twice the odds observed for Asian Chinese, Black African, and Black Caribbean. Women were found to have increased odds of involuntary admission. Significant interactions were present between ethnicity and age, area-level deprivation, homelessness, and migration in the unadjusted models. These effect modifications were not significant after adjustment for confounders.
Conclusions
Ethnic inequalities were observed in involuntary hospitalisation among service-users on first admission. No evidence of intersectional effects was present when adjusting for sociodemographic and clinical factors. Further research needs to identify the mechanisms causing the inequalities.
Supplementary Information
The online version contains supplementary material available at 10.1007/s00127-025-02898-0.
Keywords: Race, Detention, Intersectionality, Inequity, Disparity
Introduction
Mental health act and involuntary hospitalisation
Involuntary psychiatric hospitalisation features heavily in modern psychiatry, a medical and legal practice with the potential to infringe on civil liberties [1] and an individual’s right to self-determination [2]. Some service-users have described detainment under the Mental Health Act (MHA) as necessary to protect them at their most vulnerable [3], whilst others have described the intervention as coercive, distressing, and disempowering [4]. Clinical outcomes are also mixed, with some studies reporting positive results [5, 6], while others found worsening rates of suicide, length of stay, and rates of readmissions [7].
Ethnicity and involuntary hospitalisation
Evidence from systematic reviews and meta-analyses show that minoritised ethnicities are at higher risk of involuntary hospitalisation when compared to their White ethnic peers, both within the UK [8–10] and internationally [11]. In the UK, studies with services-users of secondary mental healthcare with psychosis receiving outpatient and inpatient care [12, 13] and studies with only inpatient care population [14, 15] show that Black African and Black Caribbean people have over twice the odds for involuntary admission compared to White British service-users. In contrast, studies investigating Asian ethnic groups yielded mixed results, with some showing no differences [12, 16, 17], whilst others, including a meta-analysis, have found South Asian people have a 1.3 times greater risk of involuntary admission [10, 11, 14, 18]. The contrasting findings in this group may be due to differences between studies in sample sizes, adjustments, or restricting the cohort by diagnosis. Fewer studies examined involuntary hospitalisation in East Asian communities, and evidence suggests East Asian service-users have over twice as likely to be admitted involuntarly compared to White service-users [11, 18]. There is also a paucity of studies examining inequalities in individuals of mixed ethnic backgrounds, due to small sample sizes and statistical power. When mixed ethnic groups were included, common problems involved not describing the make-up of the category [19] or including an aggregated category for all mixed ethnic groups [12].
Intersectionality
Variations in sociodemographic factors may interact with ethnicity to modify the risk of involuntary hospitalisation. Intersectionality, a term coined by Crenshaw [20], describes a multidimensional analytic framework where the experience of multiple overlapping identities of gender, race, and class contributes to power imbalances that are greater than the sum of their parts. This approach broadens an understanding of inequity, replacing a single-axis system that focuses on the most privileged group members and marginalises the multiply burdened. Despite the need for an intersectional approach to the study of ethnic inequalities being identified [9, 11], the literature remains scarce in investigating if the interaction with sociodemographic variables can influence the relationship between ethnicity and involuntary hospitalisation.
Studies focusing on the main effect of sociodemographic characteristics (i.e., without adopting an intersectional approach) show evidence of associations between many factors and involuntary hospitalisation. For example, a meta-analysis by Walker et al. [21] found that men (of all ethnicities) had a greater risk of involuntary hospitalisation than women. However, the recent meta-analysis by Barnett et al. [11] reports that study samples with a higher percentage of women had stronger associations with involuntary admission among Black Caribbean, Black unspecified and South Asian ethnicities (no interactions were tested in other ethnicities). Further, studies have found evidence of associations between age and involuntary hospitalisation, such that younger service-users had the highest risk [10, 22]. But, a recent meta-analysis showed no interactions between Black Caribbean ethnicity and age for involuntary admission, with no interactions being tested for other ethnicities [11]. Neighbourhood deprivation has been associated with compulsory admission, with some studies suggesting a dose-response relationship [10, 21]. Furthermore, migrants were also more likely to be involuntarily hospitalised when compared to the native groups [11, 23–25].
Objectives and hypothesis
The current study investigated potential ethnic inequalities in involuntary admission in the first episode of hospitalisation and whether the intersectionality of ethnicity with age, gender, area-level deprivation, homelessness, and migration alters the associations between ethnicity and involuntary, compared to voluntary, hospitalisation. Firstly, based on previous literature, we hypothesised that minoritised ethnic groups had an increased risk of involuntary hospitalisation compared to the White British ethnic group. Secondly, we hypothesised that there is an interaction between ethnicity and key sociodemographic factors, such that minoritised ethnic service-users who are younger, male, from areas of higher deprivation, homeless, or migrants had an exacerbated risk of involuntary hospitalisation.
Methods
Study design and setting
This retrospective cohort study used routinely collected patient information from the South London and Maudsley National Health Service Foundation Trust (SLaM) [26]. SLaM, one of Europe’s largest secondary mental health providers, has fully electronic health records (EHRs) since 2006 and provides care for people living in the London boroughs of Lambeth, Southwark, Lewisham, and Croydon [27].
The Clinical Record Interactive Search (CRIS) system was set up in 2007 and was used to extract data sourced from EHRs in SLaM, allowing researchers access to de-identified information in structured and unstructured fields [28]. In this study, data were extracted exclusively from structured fields. CRIS dataset received approval from the Oxford C Research Ethics Committee (18/SC/0372). This project gained approval from the service-user-led CRIS oversight committee (ref. 19–066).
Participants
The cohort inclusion criteria consisted of service-users that (a) had a first episode of voluntary or involuntary admission under the MHA that started or finished during the observation period of 01/01/2008 to 31/05/2021; (b) a personal address or be registered with a general practitioner (GP) in the SLaM catchment area within the observation period, or lived in London at the time of admission; (c) over the age of 18 at time of the first admission; (d) stayed at least one night in hospital. Individuals with missing data on ethnicity, age, or gender were excluded from the cohort at the point of entry.
Exposure and outcome
Ethnicity was divided into self-ascribed categories following the NHS classifications. Due to small numbers, the ethnic groups of Mixed ethnicity White and Asian were merged into ‘other Mixed background’, and the categories of White Gypsy/Irish Traveller, Other ethnic group– Arab and any Other ethnic group were merged to form ‘Other ethnic background’.
The outcome of interest was involuntary hospitalisation under the MHA Sects. 2, 3, 4, or 5(2) applied within two days of the first admission to SLaM during the observation period. Involuntary hospitalisation was compared to voluntary hospitalisation.
Demographics and clinical factors
The sociodemographic factors investigated included age, gender, area-level deprivation, homelessness, and migration. Adjustment for clinical factors comprised of psychiatric diagnoses and Health of the Nation Outcome Scale (HoNOS) items.
Area-level deprivation was calculated from the English Indices of Multiple Deprivation (IMD) and divided into quintiles [29]. The version of the IMD score recorded (2007, 2010, 2015 or 2019) was based on the service-user’s address closest to the date of admission. Homelessness was calculated as whether the individual had been homeless at the point of admission, in the previous year, or had a risk assessment mentioning unstable housing recorded 12 months before or 28 days after admission.
Migration was ascertained if: (a) a language other than English was listed as their first language; (b) an interpreter was documented as needed; (c) country of birth was not listed as the UK; (d) there was evidence of an asylum application.
Psychiatric diagnoses were assessed as per the ICD-10 categories and were recorded within 28 days prior and 28 days post-admission. The Health of the Nation Outcome Scale (HoNOS) is an instrument conceived to assess the social and physical functioning of service-users with mental illnesses [30]. The HoNOS score used was retrieved hierarchically: (a) within 28 days before admission; (b) if unavailable, then within 28 days after; (c) if unavailable, then within 12 months before admission.
Statistical analysis
We used multivariable logistic regression to estimate odds ratios (OR) of involuntary hospitalisation among minoritised ethnicities compared to White British service-users. We estimated the OR in unadjusted models and models adjusting for sociodemographic and clinical factors. We tested statistical interaction in logistic regressions to assess the possible effects of the intersectionality of ethnicity with each of the sociodemographic factors in unadjusted and fully adjusted models. This was done using contrast analyses, which compare the model with and without the interaction term. Some variables contained missing data, namely, area-level deprivation (11.9%), migrant status (32.1%), and HoNOS items (18.2% - 21.8%). To preserve statistical power, missing data were categorised as undetermined. Data were analysed using STATA version 15 [31].
Results
Descriptive data
From a total number of 356,056 service-users with records on 31/05/2021, 18,569 service-users were identified as meeting the inclusion criteria for the study (Fig. 1). Of this cohort, 34.7% had been hospitalised involuntarily (Table 1).
Fig. 1.
Flow diagram of study cohort selection
Table 1.
Demographic and clinical characteristics of service users with a first episode of admission in slam within the observation period
| Total N (% of total sample) a | Voluntarily admitted for inpatient care on first admission (% per characteristic) | Admitted under Sects. 2, 3, 4 or 5(2) on first admission (% per characteristic) | |
|---|---|---|---|
| Total | 18,569 (100) | 12,132 (65.3) | 6437 (34.7) |
| Ethnicity | |||
| White British | 8240 (44.4) | 6,367 (77.3) | 1873 (22.7) |
| White Irish | 513 (2.8) | 385 (75.1) | 128 (25.0) |
| Other White background | 1852 (10.0) | 1109 (59.9) | 743 (40.1) |
| Black African | 2031 (10.9) | 981 (48.3) | 1050 (51.7) |
| Black Caribbean | 1145 (6.1) | 569 (49.7) | 576 (50.3) |
| Black British / Other Black background | 1887 (10.2) | 966 (51.2) | 921 (48.8) |
| Asian Bangladeshi | 90 (0.5) | 48 (53.3) | 42 (46.7) |
| Asian Indian | 287 (1.6) | 183 (63.8) | 104 (36.2) |
| Asian Pakistani | 159 (0.9) | 92 (57.9) | 67 (42.1) |
| Asian Chinese | 130 (0.7) | 62 (47.7) | 68 (52.3) |
| Asian British / Other Asian background | 661 (3.6) | 398 (60.2) | 263 (39.8) |
| White and Black African | 83 (0.5) | 57 (68.7) | 26 (31.3) |
| White and Black Caribbean | 227 (1.2) | 159 (70.0) | 68 (30.0) |
| Other Mixed background | 171 (0.9) | 115 (67.3) | 56 (32.8) |
| Other ethnic background | 1093 (5.9) | 641 (58.7) | 452 (41.4) |
| Sociodemographic variables | |||
| Age | |||
| 18–24 | 3006 (16.2) | 1792 (59.6) | 1214 (40.4) |
| 25–34 | 4571 (24.6) | 2968 (64.9) | 1603 (35.1) |
| 35–49 | 5835 (31.4) | 4048 (69.4) | 1787 (30.6) |
| 50–64 | 2958 (15.9) | 1932 (65.3) | 1026 (34.7) |
| 65–99 | 2199 (11.8) | 1392 (63.3) | 807 (36.7) |
| Gender | |||
| Male | 10,343 (55.7) | 6795 (65.7) | 3548 (34.3) |
| Female | 8226 (44.3) | 5337 (64.9) | 2889 (35.1) |
| Area-level deprivation | |||
| 1st quintile (least deprived) | 2437 (13.1) | 1719 (70.5) | 718 (29.5) |
| 2nd quintile | 3367 (18.1) | 2184 (64.9) | 1183 (35.1) |
| 3rd quintile | 3566 (19.2) | 2304 (64.6) | 1262 (35.4) |
| 4th quintile | 3607 (19.4) | 2296 (63.7) | 1311 (36.4) |
| 5th quintile (most deprived) | 3387 (18.2) | 2171 (64.1) | 1216 (35.9) |
| Undetermined | 2205 (11.9) | 1458 (66.1) | 747 (33.9) |
| Homeless | |||
| No | 15,058 (81.1) | 9278 (61.6) | 5780 (38.4) |
| Yes | 3511 (18.9) | 2854 (81.3) | 657 (18.7) |
| Migrant status | |||
| No | 7513 (40.5) | 5615 (74.7) | 1898 (25.3) |
| Yes | 5099 (27.5) | 2938 (57.6) | 2161 (42.4) |
| Undetermined | 5957 (32.1) | 3579 (60.1) | 2378 (39.9) |
| Psychiatric diagnosis (ICD-10) | |||
| Mental disorders due to known physiological conditions (F01-F09) | |||
| No | 17,724 (95.5) | 11,596 (65.4) | 6128 (34.6) |
| Yes | 845 (4.6) | 536 (63.4) | 309 (36.6) |
| Mental and behavioural disorders due to psychoactive substance use (F10-F19) | |||
| No | 15,562 (83.8) | 9581 (61.6) | 5981 (38.4) |
| Yes | 3007 (16.2) | 2551 (84.8) | 456 (15.2) |
| Schizophrenia, schizotypal, delusional, and other non-mood psychotic disorders (F20-F29) | |||
| No | 14,919 (80.3) | 10,632 (71.3) | 4287 (28.7) |
| Yes | 3650 (19.7) | 1500 (41.1) | 2150 (58.9) |
| Affective psychosis (F30.2, F31.2, F31.5, F32.3, F33.3) | |||
| No | 17,786 (95.8) | 11,747 (66.1) | 6039 (34.0) |
| Yes | 783 (4.2) | 385 (49.2) | 398 (50.8) |
| Mood disorder (F30– F39, except the affective psychosis codes) | |||
| No | 15,737 (84.8) | 9925 (63.1) | 5812 (36.9) |
| Yes | 2832 (15.3) | 2207 (77.9) | 625 (22.1) |
| Anxiety, dissociative, stress-related, somatoform, and other nonpsychotic disorders (F40-F48) | |||
| No | 17,038 (91.8) | 10,892 (63.9) | 6146 (36.1) |
| Yes | 1531 (8.2) | 1240 (81.0) | 291 (19.0) |
| Behavioural syndromes associated with physiological disturbances and physical factors (F50-F59) | |||
| No | 18,313 (98.6) | 11,923 (65.1) | 6390 (34.9) |
| Yes | 256 (1.4) | 209 (81.6) | 47 (18.4) |
| Disorders of adult personality and behaviour (F60-F69) | |||
| No | 17,837 (96.1) | 11,540 (64.7) | 6927 (35.3) |
| Yes | 732 (3.9) | 592 (80.9) | 140 (19.1) |
| Intellectual disabilities (F70-F79) | |||
| No | 18,459 (99.4) | 12,064 (65.4) | 6395 (34.6) |
| Yes | 110 (0.6) | 68 (61.8) | 42 (38.2) |
| Pervasive and specific developmental disorders (F80-F89) | |||
| No | 18,496 (99.6) | 12,088 (65.4) | 6408 (34.7) |
| Yes | 73 (0.4) | 44 (60.3) | 29 (39.7) |
| Behavioural and emotional disorders with onset usually occurring in childhood and adolescence (F90-F98) | |||
| No | 18,534 (99.8) | 12,108 (65.3) | 6426 (34.7) |
| Yes | 35 (0.2) | 24 (68.6) | 11 (31.4) |
| Health of the Nation Outcome Scales (HoNOS) | |||
| Overactive, aggressive, disruptive, or agitated behaviour | |||
| No problem | 5763 (31.0) | 4252 (73.8) | 1511 (26.2) |
| Minor problem | 3294 (17.7) | 1978 (60.1) | 1316 (40.0) |
| Mild problem | 2991 (16.1) | 1581 (52.9) | 1410 (47.1) |
| Moderately severe problem | 2052 (11.1) | 888 (43.3) | 1164 (56.7) |
| Severe to very severe problem | 1092 (5.9) | 393 (36.0) | 699 (64.0) |
| Undetermined | 3377 (18.2) | 3040 (90.0) | 337 (10.0) |
| Non-accidental self-injury | |||
| No problem | 9436 (50.8) | 4872 (51.6) | 4564 (48.4) |
| Minor problem | 1744 (9.4) | 1108 (63.5) | 636 (36.5) |
| Mild problem | 1552 (8.4) | 1148 (74.0) | 404 (26.0) |
| Moderately severe problem | 1446 (7.8) | 1153 (79.7) | 293 (20.3) |
| Severe to very severe problem | 986 (5.3) | 807 (81.9) | 179 (18.2) |
| Undetermined | 3405 (18.3) | 3044 (89.4) | 361 (10.6) |
| Problem drinking or drug taking | |||
| No problem | 9186 (49.5) | 5504 (59.9) | 3682 (40.1) |
| Minor problem | 1404 (7.6) | 819 (58.3) | 585 (41.7) |
| Mild problem | 1765 (9.5) | 1049 (59.4) | 716 (40.6) |
| Moderately severe problem | 1672 (9.0) | 1038 (62.1) | 634 (37.9) |
| Severe to very severe problem | 900 (4.9) | 591 (65.7) | 309 (34.3) |
| Undetermined | 3642 (19.6) | 3131 (86.0) | 511 (14.0) |
| Cognitive problems | |||
| No problem | 8694 (46.8) | 5645 (64.9) | 3049 (35.1) |
| Minor problem | 2771 (14.9) | 1528 (55.1) | 1243 (44.9) |
| Mild problem | 2047 (11.0) | 1057 (51.6) | 990 (48.4) |
| Moderately severe problem | 1175 (6.3) | 611 (52.0) | 564 (48.0) |
| Severe to very severe problem | 429 (2.3) | 219 (51.1) | 210 (49.0) |
| Undetermined | 3453 (18.6) | 3072 (89.0) | 381 (11.0) |
| Physical illness or disability problems | |||
| No problem | 9140 (49.2) | 5284 (57.8) | 3856 (42.2) |
| Minor problem | 2291 (12.3) | 1393 (60.8) | 898 (39.2) |
| Mild problem | 1996 (10.8) | 1261 (63.2) | 735 (36.8) |
| Moderately severe problem | 1230 (6.6) | 817 (66.4) | 413 (33.6) |
| Severe to very severe problem | 444 (2.4) | 298 (67.1) | 146 (32.9) |
| Undetermined | 3468 (18.7) | 3079 (88.8) | 389 (11.2) |
| Problems associated with hallucinations or delusions | |||
| No problem | 5619 (30.3) | 4411 (78.5) | 1208 (21.5) |
| Minor problem | 1824 (9.8) | 1132 (62.1) | 692 (37.9) |
| Mild problem | 2913 (15.7) | 1454 (49.9) | 1459 (50.1) |
| Moderately severe problem | 3029 (16.3) | 1361 (44.9) | 1668 (55.1) |
| Severe to very severe problem | 1722 (9.3) | 693 (40.2) | 1029 (59.8) |
| Undetermined | 3462 (18.6) | 3081 (89.0) | 381 (11.0) |
| Problems with depressed mood | |||
| No problem | 4077 (22.0) | 1653 (40.5) | 2424 (59.5) |
| Minor problem | 3267 (17.6) | 1748 (53.5) | 1519 (46.5) |
| Mild problem | 3794 (20.4) | 2561 (67.5) | 1233 (32.5) |
| Moderately severe problem | 2807 (15.1) | 2184 (77.8) | 623 (22.2) |
| Severe to very severe problem | 1183 (6.4) | 929 (78.5) | 254 (21.5) |
| Undetermined | 3441 (18.5) | 3057 (88.8) | 384 (11.2) |
| Problems with relationships | |||
| No problem | 4936 (26.6) | 3087 (62.5) | 1849 (37.5) |
| Minor problem | 3590 (19.3) | 2160 (60.2) | 1430 (39.8) |
| Mild problem | 3740 (20.1) | 2207 (59.0) | 1533 (41.0) |
| Moderately severe problem | 2025 (10.9) | 1155 (57.0) | 870 (43.0) |
| Severe to very severe problem | 655 (3.5) | 376 (57.4) | 279 (42.6) |
| Undetermined | 3623 (19.5) | 3147 (86.9) | 476 (13.1) |
| Problems with activities of daily living | |||
| No problem | 6825 (36.8) | 4292 (62.9) | 2533 (37.1) |
| Minor problem | 3369 (18.1) | 1993 (59.2) | 1376 (40.8) |
| Mild problem | 2864 (15.4) | 1647 (57.5) | 1217 (42.5) |
| Moderately severe problem | 1494 (8.1) | 802 (53.7) | 692 (46.4) |
| Severe to very severe problem | 453 (2.4) | 275 (60.7) | 178 (39.3) |
| Undetermined | 3564 (19.2) | 3123 (87.6) | 441 (12.4) |
| Problems with living conditions | |||
| No problem | 7989 (43.0) | 4893 (61.3) | 3096 (38.8) |
| Minor problem | 2570 (13.8) | 1534 (59.7) | 1036 (40.3) |
| Mild problem | 1773 (9.6) | 1021 (57.6) | 752 (42.4) |
| Moderately severe problem | 1053 (5.7) | 603 (57.3) | 450 (42.7) |
| Severe to very severe problem | 1135 (6.1) | 685 (60.4) | 450 (39.7) |
| Undetermined | 4049 (21.8) | 3396 (83.9) | 653 (16.1) |
| Problems with occupation and activities | |||
| No problem | 5932 (32.0) | 3681 (62.1) | 2251 (38.0) |
| Minor problem | 3315 (17.9) | 1981 (59.8) | 1134 (40.2) |
| Mild problem | 3259 (17.6) | 1894 (59.0) | 1365 (41.9) |
| Moderately severe problem | 1453 (7.8) | 857 (59.0) | 596 (41.0) |
| Severe to very severe problem | 623 (3.4) | 371 (59.6) | 252 (40.5) |
| Undetermined | 3987 (21.5) | 3348 (84.0) | 639 (16.0) |
Note. Undetermined category represents missing data in the variable
a Percentages may not add up to 100% due to rounding to one decimal place
The largest ethnic group in the cohort was White British (44.4%), The largest age group in the cohort was 35–49 years old (31.4%), and the sample mean age was 42 years. The gender distribution was 55.7% male. In addition, 18.9% of the cohort was classified as homeless, and 27.5% had been categorised as migrants (Table 1).
Ethnicity and involuntary hospitalisation
The ethnicities with the highest proportion of involuntary admission were Asian Chinese (52.3%), followed by Black African (51.7%), Black Caribbean (50.3%), and Black British or other Black background (48.8%). The lowest proportion of involuntary admissions was in the White British ethnicity (22.7%). In the unadjusted logistic regression model (Table 2), all minoritised ethnicities, except for White Irish and Mixed ethnicity White and Black African, were more likely to be involuntarily admitted than the White British group.
Table 2.
Multivariable logistic regression of unadjusted and fully adjusted associations with involuntary admission under MHA sects. 2,3,4, or 5(2) at first hospital admission
| OR (95% CI) | |||
|---|---|---|---|
| Variable | Unadjusted | Fully adjusted | p-value |
| Ethnicity | |||
| White British | Reference group | ||
| White Irish | 1.13 (0.92–1.39) | 1.01 (0.79–1.28) | 0.942 |
| Other White background | 2.28 (2.05–2.53) | 1.49 (1.29–1.73) | < 0.001 |
| Black African | 3.64 (3.29–4.03) | 1.80 (1.57–2.07) | < 0.001 |
| Black Caribbean | 3.44 (3.03–3.91) | 1.63 (1.40–1.89) | < 0.001 |
| Black British / Other Black background | 3.24 (2.92–3.60) | 1.75 (1.55–1.98) | < 0.001 |
| Asian Bangladeshi | 2.97 (1.96–4.51) | 1.68 (1.04–2.72) | 0.034 |
| Asian Indian | 1.93 (1.51–2.47) | 1.43 (1.08–1.91) | 0.013 |
| Asian Pakistani | 2.48 (1.80–3.41) | 1.67 (1.16–2.41) | 0.006 |
| Asian Chinese | 3.73 (2.63–5.28) | 2.28 (1.53–3.41) | < 0.001 |
| Asian British / Other Asian background | 2.25 (1.91–2.65) | 1.47 (1.20–1.79) | < 0.001 |
| White and Black African | 1.55 (0.97–2.47) | 0.94 (0.55–1.60) | 0.816 |
| White and Black Caribbean | 1.45 (1.09–1.94) | 1.08 (0.78–1.51) | 0.635 |
| Other Mixed background | 1.66 (1.20–2.29) | 1.18 (0.82–1.71) | 0.374 |
| Other ethnic background | 2.40 (2.10–2.73) | 1.66 (1.41–1.96) | < 0.001 |
| Age | |||
| 18–24 | 1.53 (1.40–1.68) | 1.24 (1.11–1.38) | < 0.001 |
| 25–34 | 1.22 (1.13–1.33) | 1.10 (1.00-1.21) | 0.054 |
| 35–49 | Reference group | ||
| 50–64 | 1.20 (1.10–1.32) | 1.23 (1.10–1.38) | < 0.001 |
| 65–99 | 1.31 (1.18–1.46) | 1.00 (0.87–1.15) | 0.977 |
| Gender | |||
| Male | Reference group | ||
| Female | 1.04 (0.98–1.10) | 1.11 (1.03–1.19) | 0.007 |
| Area-level deprivation | |||
| 1st quintile (least deprived) | Reference group | ||
| 2nd quintile | 1.30 (1.16–1.45) | 1.10 (0.97–1.25) | 0.154 |
| 3rd quintile | 1.31 (1.17–1.47) | 1.08 (0.95–1.23) | 0.260 |
| 4th quintile | 1.37 (1.22–1.53) | 1.09 (0.96–1.24) | 0.177 |
| 5th quintile (most deprived) | 1.34 (1.20–1.50) | 1.08 (0.95–1.23) | 0.242 |
| Homeless | |||
| No | Reference group | ||
| Yes | 0.37 (0.34–0.40) | 0.62 (0.55–0.71) | < 0.001 |
| Migrant status | |||
| No | Reference group | ||
| Yes | 2.18 (2.02–2.35) | 1.15 (1.02–1.29) | 0.020 |
Note. Fully adjusted column includes adjustment for all variables in the table, as well as adjustment for psychiatric diagnosis and Health of the Nation Outcome Scales items (as seen in Table 1). p-value corresponds to fully adjusted analysis
In the fully adjusted model (Table 2), increased odds of involuntary over voluntary hospitalisation were observed in the majority of ethnicities compared to the White British group. These were among service-users who are Asian Chinese (adjusted OR = 2.28, 95% CI: 1.53–3.41); Black African (aOR = 1.80, 95% CI: 1.57–2.07); Black British or any other Black background (aOR = 1.75, 95% CI: 1.55–1.98); Asian Bangladeshi (aOR = 1.68, 95% CI: 1.04–2.72); Asian Pakistani (aOR = 1.67, 95% CI: 1.16–2.41); Other ethnic background (aOR = 1.66, 95% CI: 1.41–1.96); Black Caribbean (aOR = 1.63, 95% CI: 1.40–1.89); Other White background (aOR = 1.49, 95% CI: 1.29–1.73); Asian British or any other Asian background (aOR = 1.47, 95% CI: 1.20–1.79); and Asian Indian (aOR = 1.43, 95% CI: 1.08–1.91).
No significant associations with involuntary hospitalisation were observed when comparing the White British group to Other Mixed ethnic background (aOR = 1.18, 95% CI: 0.82–1.71); Mixed ethnicity White and Black Caribbean (aOR = 1.08, 95% CI: 0.78–1.51); White Irish (aOR = 1.01, 95% CI: 0.79–1.28); and Mixed ethnicity White and Black African groups (aOR = 0.94, 95% CI: 0.55–1.60).
Sociodemographic characteristics and involuntary hospitalisation
With regards to sociodemographic variables, the main findings were that:
The only age groups with higher odds of involuntary admission were 18–24 and 50–64 years old, compared to the reference group of 35–49, the age group with the lowest proportion of involuntary admission.
Women had higher odds of involuntary admission when compared to men, although the magnitude of the effect was very small.
There were no significant associations for any deprivation quintiles with involuntary admission.
Homeless service-users had a decreased odds of involuntary admission compared to those with a home. Instead, these individuals were more likely to be voluntarily admitted.
Individuals classified as migrants had an increased odds of involuntary admission compared to people without this status.
Interaction analysis
Contrast analyses and stratified analyses for the unadjusted models (Table 3 and Supplementary Material Tables S3-S7) suggested that the relationship between ethnicity and involuntary hospitalisation changed depending on four key sociodemographic variables. These were: (a) age [χ2 (df = 56, N = 18569) = 85.03, p = 0.007], (b) area-level deprivation [χ2 (70, 18569) = 103.39, p = 0.006], (c) housing situation/homelessness [χ2 (14, 18569) = 67.22, p < 0.001], and (d) migrant status [χ2 (28, 18569) = 48.19, p = 0.01]. In the analyses adjusted for all sociodemographic variables and the clinical factors of psychiatric diagnosis and HoNOS (Table 3), there was no longer an interaction between ethnicity and these sociodemographic variables in relation to involuntary hospitalisation in the contrast analyses. Due to the exploratory nature of the study, we present details of the interactions observed in crude models, due to their relevance to the use of the MHA, in line with other statistics [22].
Table 3.
Interaction analysis of ethnicity and sociodemographic variables with involuntary admission as the outcome using contrast analysis to compare models with and without the interaction terms specified below
| Unadjusted model a | Fully adjusted model b | |||||||
|---|---|---|---|---|---|---|---|---|
| Interaction terms | N | χ2 | df | p-value | N | χ2 | df | p-value |
| Ethnicity and Age | 18,569 | 85.03 | 56 | 0.007 | 18,569 | 66.71 | 56 | 0.155 |
| Ethnicity and Gender | 18,596 | 20.76 | 14 | 0.108 | 18,596 | 14.29 | 14 | 0.428 |
| Ethnicity and Area Deprivation | 18,569 | 103.39 | 70 | 0.006 | 18,569 | 86.25 | 70 | 0.091 |
| Ethnicity and Homelessness | 18,569 | 67.22 | 14 | < 0.001 | 18,569 | 15.55 | 14 | 0.342 |
| Ethnicity and Migrant Status | 18,569 | 48.19 | 28 | 0.010 | 18,569 | 22.42 | 28 | 0.762 |
Note. N = total sample. χ2 = chi-squared statistic value. df = degrees of freedom
a Unadjusted model represents the unadjusted interactions of ethnicity with the key sociodemographic variable
b Fully adjusted model represents the adjusted interaction of ethnicity with the key sociodemographic variable, adjusting for all other sociodemographic variables in the table as well as psychiatric diagnosis and Health of the Nation Outcome Scales items
In crude models, interaction analysis indicated that the relationship between ethnicity and involuntary admission varied by age group. Odds ratios for disparities in involuntary admission (comparing minoritised ethnic groups to White British) were strongest in the younger age groups (18–24 years old) for service-users of Other White, Black African, Black Caribbean, and Black British ethnicities (Supplementary Table S3).
Other White background aged 65–99 [(OR65-99 = 1.47, 95% CI: 1.02–2.10) vs. (OR18-24 = 2.60, 95% CI: 1.97–3.24), interaction term p = 0.014].
Black African aged 65–99 [(OR65-99 = 1.84, 95% CI: 1.18–2.88) vs. (OR18-24 = 4.08, 95% CI: 3.19–5.21), interaction term p = 0.002]
Black Caribbean aged 65–99 [(OR65-99 = 2.00, 95% CI: 1.53–2.62) vs. (OR18-24 = 4.17, 95% CI: 2.85–6.11), interaction term p = 0.002]
Black British aged 50–64 [(OR50-64 = 2.45, 95% CI: 1.84–3.26) vs. (OR18-24 = 3.93, 95% CI: 3.15–4.92), interaction term p = 0.011]
By contrast, for Asian British service-users, the odds ratio for the disparities in involuntary admission was higher in the middle-aged group (35–49 years old) [(OR35 − 49 = 3.19, 95% CI: 2.40–4.23) vs. (OR18 − 24 = 1.82, 95% CI: 1.23–2.70), interaction term p = 0.024].
In the unadjusted interaction analysis, the relationship between ethnicity and involuntary admission varied depending on the level of deprivation in the area where the person resided. For Black Caribbean service-users, disparities in involuntary admission (i.e. odds ratios comparing minoritised ethnic groups to White British people) were higher for those living in the most deprived areas [(ORquintile 5 = 4.14, 95% CI: 3.15–5.43) vs. (ORquintile 1 = 2.24, 95% CI: 1.43–3.51), interaction term p = 0.022] (Supplementary Table S5). However, among other ethnic groups the odds ratio for involuntary admission was lower for those living in more deprived areas. Namely:
Black British service-users [(ORquintile 4 = 2.56, 95% CI: 2.05–3.21) vs. (ORquintile 1 = 4.18, 95% CI: 2.97–5.90), interaction term p = 0.019].
Asian Pakistani service-users [(ORquintile 4 = 0.73, 95% CI: 0.24–2.17) vs. (ORquintile 1 = 3.02, 95% CI: 1.47–6.18), interaction term p = 0.033].
White and Black Caribbean mixed-ethnicity service-users [(ORquintile 4 = 0.71, 95% CI: 0.34–1.48) vs. (ORquintile 1 = 3.66, 95% CI: 1.51–8.87) interaction term p = 0.005].
In the interaction analysis of crude models, we observed that the relationship between ethnicity and involuntary admission varied depending on homelessness status. Interestingly, the main effect of homelessness was protective against involuntary admission. However, for some minoritised ethnicities, disparities in involuntary admission (i.e. odds ratios comparing minoritised ethnic groups to White British) were higher in the homeless group than in the non-homeless group (Supplementary Table S6). Namely, this effect was observed among:
Other White service-users [(ORhomeless = 4.57, 95% CI: 3.54–5.91) vs. (ORnot homeless = 2.03, 95% CI: 1.80–2.29), interaction term p < 0.001]
Black African service-users [(ORhomeless = 7.34, 95% CI: 5.47–9.84) vs. (ORnot homeless = 3.08, 95% CI: 2.76–3.43), interaction term p < 0.001]
Black Caribbean service-users [(ORhomeless = 4.96, 95% CI: 3.26–7.54) vs. (ORnot homeless = 2.98, 95% CI: 2.61–3.41), interaction term p = 0.024]
Asian Chinese service-users [(ORhomeless = 13.05, 95% CI: 4.48–38.03) vs. (ORnot homeless = 2.93, 95% CI: 2.03–4.24), interaction term p = 0.01]
Other ethnic background service-users [(ORhomeless = 5.16, 95% CI: 3.53–7.53) vs. (ORnot homeless = 2.01, 95% CI: 1.75–2.32), interaction term p < 0.001]
In the main analysis, migrants were at a higher risk for involuntary admission. However, in crude models, interaction analysis indicated that the relationship between ethnicity and involuntary admission varied depending on migrant status. The disparity in involuntary admission between people of Black African heritage and White British was higher (almost double the odds ratio) in the non-migrant group compared to the migrant group [(ORnon−migrant = 6.41, 95% CI: 4.48–9.16) vs. (ORmigrant = 3.38 95% CI: 2.27–5.03), interaction term p = 0.019] (Supplementary Table S7).
Discussion
This retrospective cohort study examined the ethnic inequalities in voluntary and involuntary hospital admission in 18,569 service-users on their first admission to SLaM, adjusting for several sociodemographic and clinical factors. A further aim was to identify any intersectional effects of ethnicity with sociodemographic factors concerning involuntary admission. The study found evidence that compared to White British people there was a higher likelihood of involuntary hospitalisation in most minoritised ethnicities, but not in White Irish, Mixed ethnicity White and Black Caribbean, Mixed ethnicity White and Black African, and Other Mixed ethnic background. On analysis of interaction in the unadjusted models, there were interactions between ethnicity and age, deprivation, homelessness, and migrant status in relation to involuntary hospitalisation, but these differences were no longer significant in the fully adjusted models.
Ethnicity
This study identified inequalities among minoritised ethnicities and involuntary hospitalisation that align with previous studies’ findings [8, 10, 11, 16, 32–34]. A possible explanation for this may be due to reduced access to resources driven by patient and service-level disadvantages, that affect ethnic minorities, leading to adverse pathways into mental health care and potentially higher levels of unmet care needs [9, 11, 35–38]. Interestingly, these associations do not hold for individuals of any Mixed ethnicity in the present study, suggesting these disadvantages in access to care may be minimised when one parent is of White British ethnicity. White Irish people also did not have an increased likelihood of involuntary admission.
The ethnic group with the strongest association with involuntary admission was the Asian Chinese group, with over twice the odds for involuntary admission compared to White British people. The magnitude of differences is similar to what was observed in a recent meta-analysis [11]. Black African service-users were almost twice as likely to be involuntarily admitted - a finding that is consistent across several studies [11, 13, 15]. The increased likelihood of involuntary admission among service-users of Black Caribbean and South Asian descent also aligns with previous studies [11].
Sociodemographic variables
Age
The present study identified an association between age and involuntary hospitalisation, observed in the age groups of 18–24 and 50–64, as compared to people aged 35–49. The current study’s findings of the greatest risk of involuntary admission in the youngest age category of 18–24 are in keeping with that of Weich et al. [10] observed using national data. However, there is some divergence in previous literature that reports no associations [39–41], possibly due to these studies operationalising age as a continuous variable.
Gender
This study agrees with previous literature that reports an increased likelihood of involuntary hospitalisation among women [41, 42]. However, this association is not always seen across studies, as some have found evidence of increased risk in males, including in a recent literature review and meta-analyses [10, 21, 43–45]. Exploring the differences across literature suggests that these inconsistencies may be due to differences in analytical and recruitment methods, such as not adjusting for potential confounders [43] or restricting the cohort to one diagnostic group [45].
While the present study did not find interactions in either the unadjusted or fully adjusted models, findings in previous literature have suggested that gender plays a role in modifying risk for minoritised ethnic groups. For example a study by Mann et al. [12] observed, in stratified adjusted models, a greater likelihood of involuntary admission in women over men across many minoritised ethnicities when compared to White British people, although no interaction analyses were conducted. Similarly, a recent meta-analysis also observed that a higher proportion of women in a study was a predictor for involuntary hospitalisation among those who are Black Caribbean, ‘unspecified’ Black and South Asian people [11]. The higher likelihood for women may be related to unaddressed specific healthcare needs [45, 46], compounded disempowerment in negotiating mental healthcare [35, 48], and barriers to accessing services [49].
Area-level deprivation
The findings for area-level deprivation differed from other studies. While this study did not find any associations between area-level deprivation and involuntary hospitalisation after adjusting for sociodemographic and clinical factors, other studies were able to find evidence of this [10, 50].
Housing situation / homelessness
In this study, we observed that homeless service-users were less likely to be there involuntarily admitted compared to those who were not homeless. In the unadjusted interaction and stratified analyses, we observed that for some minoritised ethnicities, disparities in involuntary admission (i.e. odds ratios comparing minoritised ethnic groups to White British) were higher in the homeless group than in the non-homeless group. In the fully adjusted model, no significant interactions between ethnicity and homelessness were observed. The finding of the main association of reduced involuntary admission among those who are homeless was unexpected, as a previous meta-analysis had found that homelessness was not a predictor for involuntary admission [21]. The protective effect observed in the present study may be due to mental health services having a reduced threshold for requiring inpatient care for those without a home, as it can provide food and shelter to a population associated with a greater risk of premature mortality [51].
Migrant status
Regarding migration, the findings from this study are in line with others, as migrants were at higher risk of involuntary hospitalisation than voluntary hospitalisation [11, 23–25]. This increased risk may be due to challenges migrants face, such as worsening social climates and acculturation barriers [52, 53], barriers to accessing care [49], and institutional racism due to their visible minority status [54, 55].
Limitations and strengths
To accommodate for missing data, higher-level categories of migration status were created by amalgamating different data sources, and an undetermined category was used to maintain power in analysis. Additional sociodemographic variables such as marital and employment status could not be included in the analysis due to a large amount of missing data, potentially resulting in residual confounding. Although information on sociodemographic variables and involuntary hospitalisation were recorded simultaneously, it is unlikely that the type of hospitalisation has influenced the sociodemographic profile. In contrast, the clinical variables of HoNOS may suffer from imperfect adjustment as the item used may have been recorded within 28 days after admission if there was no assessment before admission. Thus, for some service-users, their admission status may influence the assessment of symptoms on HoNOS. Also, this cohort and the SLaM catchment area had a higher proportion of people in the lowest socioeconomic groups, which does not accurately represent the spread within England [26]. Further, this study divided IMD scores into quintiles, which may mask subtle associations, potentially resulting in loss of information via aggregation. Involuntary admission was directly compared to voluntary, and examining this outcome as dichotomous may ignore the realities of how voluntary admission can occur under coercion [56] and how some service-users admitted involuntarily may still possess some level of ‘voluntariness’ [40].
The study’s strengths included utilising a large, diverse, and representative cohort, owing to the near-monopoly of mental health care SLaM has within the catchment. In part due to the use of EHRs, this study was also able to investigate and adjust for several sociodemographic and clinical risk factors identified in previous literature. Further, the study did not restrict the cohort by psychiatric diagnosis, whereas many others had limited participants to those with a psychotic disorder. In addition, due to the large sample size of the cohort, ethnicity could be captured on a granular level, allowing for the expression of heterogeneity between ethnicities, as opposed to grouping ethnicities under larger, often dissimilar classifications.
Implications
The findings of this study show that the majority of minoritised ethnicities have a greater likelihood of being admitted for inpatient care involuntarily, compared to White British service users. These disparities are not explained by other sociodemographic factors, psychiatric diagnoses, or the severity of clinical symptoms at the time of admission. This inequality ought to be the focus of policies and interventions designed to improve previous care experiences and other mechanisms contributing to inequalities in the likelihood of involuntary admission [8, 9, 11, 15, 21, 25–38, 48].
The absence of significant interactions between ethnicity and specific sociodemographic factors, when considering the impact of all sociodemographic and clinical factors, limits our ability to provide further recommendations for clinical practice. Although a few exploratory findings were observed in the unadjusted models, some of the findings may warrant further investigation The lack of observation of statistically significant findings when considering interactions between only two sociodemographic factors at a time might be a methodological limitation. Future work could employ latent class analyses (LCA), longitudinal study designs, and other methodologies that differently address the multidimensionality of intersectionality. A multifactorial approach like LCA may better capture the interaction of multiple intersecting underprivileged social identities [57]. Qualitative studies with service users and mental health professionals suggest higher risk perceptions for young black men, associated with higher involuntary admissions [58]. We recommend future studies to employ analytical approaches that capture the interaction of multiple factors.
Conclusion
This study provides evidence of ethnic inequalities as after adjustment for various sociodemographic and clinical factors, evidence of an increased likelihood of involuntary hospitalisation, was found among 10 out of 14 minoritised ethnicities compared to white British people. The study also observed an increased likelihood of involuntary hospitalisation among service-users who were younger, migrants, or women– with the finding in the latter being unexpected based on previous literature. The findings also suggest that there was no evidence of major intersectional effects after controlling for sociodemographic and clinical factors.
Electronic supplementary material
Below is the link to the electronic supplementary material.
Acknowledgements
The authors wish to acknowledge the contribution of Craig Colling, who extracted the data from SLaM’s electronic health records using Clinical Record Interactive Search. For the purposes of open access, the authors have applied a Creative Commons Attribution (CC BY) licence to any Accepted Author Manuscript version arising from this submission.
Author contributions
RA developed the objective of the study in collaboration with RDH and DFF. RA conducted all the analyses and wrote the first draft of the manuscript. SW, PN, JD, RP, MK, KB, RDH and DFF provided input on the study design and writing of the manuscript. All authors contributed to the final version of the manuscript.
Funding
This work utilised the Clinical Record Interactive Search (CRIS) platform funded and developed by the National Institute for Health and Care Research (NIHR) Maudsley Biomedical Research Centre at South London and Maudsley NHS Foundation Trust and King’s College London, and a joint infrastructure grant from Guy’s and St Thomas’ Charity and the Maudsley Charity. RDH and DDF receive salary support from the National Institute for Health and Care Research (NIHR) Maudsley Biomedical Research Centre at South London and Maudsley NHS Foundation Trust and King’s College London. DFF work was supported by the Lankelly Chase Foundation, which funded the emerging work of the Synergi Collaborative Centre (a 5-year national initiative to build a knowledge hub on ethnic inequalities and multiple disadvantages in severe mental illness in the UK). SW is funded by a National Institute for Health Research (NIHR) Clinical Doctoral Fellowship DRF-2017–10–147. PN is funded by the NIHR and the Applied Research Collaboration Kent, Surrey, and Sussex (ARC KSS) as part of his Doctoral Research. JD is supported by the NIHR Clinician Science Fellowship award (CS-2018–18-ST2-014) and has received support from a Medical Research Council (MRC) Clinical Research Training Fellowship (MR/L017105/1). RP has received grant funding from the National Institute for Health and Care Research (NIHR301690) and the Medical Research Council (MR/S003118/1). The views expressed are those of the authors and not necessarily those of the NHS, the NIHR, the Department of Health and Social Care, Lankelly Chase Foundation, or other funders.
Data availability
No datasets were generated or analysed during the current study.
Declarations
Competing interests
In the past 5 years, for works not related to this manuscript, RP has participated in a Scientific Advisory Board for Boehringer Ingelheim, has received grant funding from Janssen, and has received consulting fees from Holmusk, Akrivia Health, Columbia Data Analytics, Clinilabs, Social Finance, Boehringer Ingelheim, Bristol Myers Squibb, Supernus, Teva and Otsuka. In the past 5 years, for works not related to this manuscript, RDH has received research funding from Roche, Pfizer, Janssen and H Lundbeck. In the past 5 years, for works not related to this manuscript, DFdF received funds from the NIHR Maudsley Biomedical Research Centre, the UK Department of Health and Social Care, Janssen Research & Development LLC, and H. Lundbeck A/S.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Daniela Fonseca de Freitas and Richard D. Hayes contributed equally to this work.
References
- 1.Høyer G (2000) On the justification for civil commitment. Acta Psychiatrica Scand Supplement 101(399):65–71. 10.1111/j.0902-4441.2000.007s020[dash]16.x [PubMed] [Google Scholar]
- 2.Välimäki M, Leino-Kilpi H, Helenius H (1996) Self-determination in clinical practice: the psychiatric patient’s point of view. Nurs Ethics 3(4):329–344. 10.1177/096973309600300406 [DOI] [PubMed] [Google Scholar]
- 3.Department of Health and Social Care (2018) Modernising the Mental Health Act: increasing choice, reducing compulsion. Retrieved from https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/778897/Modernising_the_Mental_Health_Act_-_increasing_choice__reducing_compulsion.pdf
- 4.Akther SF, Molyneaux E, Stuart R, Johnson S, Simpson A, Oram S (2019) Patients’ experiences of assessment and detention under mental health legislation: systematic review and qualitative meta-synthesis. BJPsych Open 5(3):1–10. 10.1192/bjo.2019.19 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Kjellin L, Candefjord I-L, Machl M, Westrin C-G, Eriksson K, Ekblom B, stman O (1993) Coercion in psychiatric care: problems of medical ethics in a comprehensive empirical study. Behav Sci Law 11(3):323–334. 10.1002/bsl.2370110309 [Google Scholar]
- 6.McEvoy JP, Applebaum PS, Apperson LJ, Geller JL, Freter S (1989) Why must some schizophrenic patients be involuntarily committed? The role of insight. Compr Psychiatr 30(1):13–17. 10.1016/0010-440x(89)90113-2 [DOI] [PubMed] [Google Scholar]
- 7.Kallert TW (2008) Coercion in psychiatry. Curr Opin Psychiatry 21(5):485–489. 10.1097/YCO.0b013e328305e49f [DOI] [PubMed] [Google Scholar]
- 8.Bhui K, Stansfeld S, Hull S, Priebe S, Mole F, Feder G (2003) Ethnic variations in pathways to and use of specialist mental health services in the UK: systematic review. Br J Psychiatry 182(FEB):105–116. 10.1192/bjp.182.2.105 [DOI] [PubMed] [Google Scholar]
- 9.Halvorsrud K, Nazroo J, Otis M, Hajdukova B, E., Bhui K (2018) Ethnic inequalities and pathways to care in psychosis in England: A systematic review and meta-analysis. BMC Med 16(1):1–17. 10.1186/s12916-018-1201-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Weich, S., McBride, O., Twigg, L., Duncan, C., Keown, P., Crepaz-Keay, D.,… Bhui,K. (2017). Variation in compulsory psychiatric inpatient admission in England: a cross-classified,multilevel analysis. The Lancet Psychiatry, 4(8), 619–626. 10.1016/S2215-0366(17)30207-9 [DOI] [PubMed]
- 11.Barnett P, Mackay E, Matthews H, Gate R, Greenwood H, Ariyo K, Smith S (2019) Ethnic variations in compulsory detention under the mental health act: a systematic review and meta-analysis of international data. Lancet Psychiatry 6(4):305–317. 10.1016/S2215-0366(19)30027-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Mann, F., Fisher, H. L., Major, B., Lawrence, J., Tapfumaneyi, A., Joyce, J.,… Johnson,S. (2014). Ethnic variations in compulsory detention and hospital admission for psychosis across four UK Early Intervention Services. BMC Psychiatry, 14(1), 1–12. 10.1186/s12888-014-0256-1 [DOI] [PMC free article] [PubMed]
- 13.Morgan C, Mallett R, Hutchinson G, Bagalkote H, Morgan K, Fearon P, Leff J (2005) Pathways to care and ethnicity. I: sample characteristics and compulsory admission: report from the ÆSOP study. Br J Psychiatry 186:281–289. 10.1192/bjp.186.4.281 [DOI] [PubMed] [Google Scholar]
- 14.Bowers L, Jones J, Simpson A (2009) The demography of nurses and patients on acute psychiatric wards in England. J Clin Nurs 18(6):884–892. 10.1111/j.1365-2702.2008.02362.x [DOI] [PubMed] [Google Scholar]
- 15.Lawlor C, Johnson S, Cole L, Howard LM (2012) Ethnic variations in pathways to acute care and compulsory detention for women experiencing a mental health crisis. Int J Soc Psychiatry 58(1):3–15. 10.1177/0020764010382369 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Gajwani R, Parsons H, Birchwood M, Singh SP (2016) Ethnicity and detention: are black and minority ethnic (BME) groups disproportionately detained under the mental health act 2007? Soc Psychiatry Psychiatr Epidemiol 51(5):703–711. 10.1007/s00127-016-1181-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Singh SP, Burns T, Tyrer P, Islam Z, Parsons H, Crawford MJ (2014) Ethnicity as a predictor of detention under the mental health act. Psychol Med 44(5):997–1004. 10.1017/S003329171300086X [DOI] [PubMed] [Google Scholar]
- 18.Rotenberg M, Tuck A, Ptashny R, McKenzie K (2017) The role of ethnicity in pathways to emergency psychiatric services for clients with psychosis. BMC Psychiatry 17(1):1–11. 10.1186/s12888-017-1285-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Borschmann RD, Gillard S, Turner K, Lovell K, Goodrich-purnell N, Chambers M (2010) Demographic and referral patterns of people detained under sect. 136 of the mental health act (1983) in a South London mental health trust from 2005 to 2008. Med Sci Law 50(1):15–18 [DOI] [PubMed] [Google Scholar]
- 20.Crenshaw K (1989) Demarginalizing the intersection of race and sex: A black feminist critique of antidiscrimination doctrine, feminist theory, and antiracist politics. Univ Chic Legal Forum 1(8):39–52. 10.4324/9780429499142-5 [Google Scholar]
- 21.Walker S, Mackay E, Barnett P, Sheridan Rains L, Dalton-Locke C, Trevillion K, Johnson S (2019) Clinical and social factors associated with involuntary psychiatric hospitalisation: a systematic review, meta-analysis, and narrative synthesis. Lancet Psychiatry 6:1039–1053. 10.1016/S2352-4642(21)00089-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.NHS Digital (2021) NHS Digital. Mental Health Act Statistics, Annual Figures. Retrieved from https://files.digital.nhs.uk/ED/8F6815/ment-heal-act-stat-eng-2020-21-summ-rep.pdf
- 23.Curley A, Agada E, Emechebe A, Anamdi C, Ng XT, Duffy R, Kelly BD (2016) Exploring and explaining involuntary care: the relationship between psychiatric admission status, gender and other demographic and clinical variables. Int J Law Psychiatry 47:53–59. 10.1016/j.ijlp.2016.02.034 [DOI] [PubMed] [Google Scholar]
- 24.Kelly BD, Emechebe A, Anamdi C, Duffy R, Murphy N, Rock C (2015) Custody, care and country of origin: demographic and diagnostic admission statistics at an inner-city adult psychiatry unit. Int J Law Psychiatry 38:1–7. 10.1016/j.ijlp.2015.01.001 [DOI] [PubMed] [Google Scholar]
- 25.Rodrigues, R., Macdougall, A. G., Zou, G., Lebenbaum, M., Kurdyak, P., Li, L.,…Anderson, K. K. (2019). Risk of involuntary admission among first-generation ethnic minority groups with early psychosis: A retrospective cohort study using health administrative data. Epidemiology and Psychiatric Sciences, 29(59), 1–8. 10.1017/S2045796019000556 [DOI] [PMC free article] [PubMed]
- 26.Perera G, Broadbent M, Callard F, Chang CK, Downs J, Dutta R, Stewart 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(3):1–22. 10.1136/bmjopen-2015-008721 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Stewart, R., Soremekun, M., Perera, G., Broadbent, M., Callard, F., Denis, M.,…Lovestone, S. (2009). The South London and Maudsley NHS Foundation Trust Biomedical Research Centre (SLAM BRC) case register: development and descriptive data. 10.1186/1471-244X-9-51 [DOI] [PMC free article] [PubMed]
- 28.Fernandes AC, Cloete D, Broadbent MTM, Hayes RD, Chang CK, Jackson RG, Callard F (2013) Development and evaluation of a de-identification procedure for a case register sourced from mental health electronic records. BMC Med Inf Decis Mak 13(1):1. 10.1186/1472-6947-13-71 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Department of Communities and Local Government (2010) The English Indices of Deprivation 2010. Retrieved from https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/6320/1870718.pdf
- 30.Wing, J., Beevor, A., Curtis, R. H., Park, S. B., Hadden, S., & Burns, A. (1998). Health of the Nation Outcome Scales (HoNOS). British Journal of Psychiatry, 172, 11–18. 10.4324/9781003076391-177 [DOI] [PubMed] [Google Scholar]
- 31.StataCorp (2017) Stata statistical software: release 15. StataCorp LLC, TX [Google Scholar]
- 32.Audini B, Lelliott P (2002) Age, gender and ethnicity of those detained under part II of the mental health act 1983. Br J Psychiatry 180:222–226. 10.1192/bjp.180.3.222 [DOI] [PubMed] [Google Scholar]
- 33.Singh SP, Croudace T, Beck A, Harrison G (1998) Perceived ethnicity and the risk of compulsory admission. Soc Psychiatry Psychiatr Epidemiol 33(1):39–44. 10.1007/s001270050020 [DOI] [PubMed] [Google Scholar]
- 34.Singh SP, Greenwood NAN, White S, Churchill R (2007) Ethnicity and the mental health act 1983. Br J Psychiatry 191:99–105 [DOI] [PubMed] [Google Scholar]
- 35.Lawrence V, McCombie C, Nikolakopoulos G, Morgan C (2021) Ethnicity and power in the mental health system: experiences of white British and black Caribbean people with psychosis. Epidemiol Psychiatric Sci 30(12):1–7. 10.1017/S2045796020001043 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Morris RM, Sellwood W, Edge D, Colling C, Stewart R, Cupitt C, Das-Munshi J (2020) Ethnicity and impact on the receipt of cognitive-behavioural therapy in people with psychosis or bipolar disorder: an english cohort study. BMJ Open 10(12). 10.1136/bmjopen-2019-034913 [DOI] [PMC free article] [PubMed]
- 37.Freitas, D. F. De, Kadra, G., Megan, S., Shetty, H., Broadbent, M., Patel, R.,…Khondoker, M. (2022). Ethnic inequalities in clozapine use among people with treatment– resistant schizophrenia: a retrospective cohort study using data from electronic clinical records. Social Psychiatry and Psychiatric Epidemiology, (0123456789). 10.1007/s00127-022-02257-3 [DOI] [PMC free article] [PubMed]
- 38.Freitas DF, Walker S, Nyikavaranda P et al (2022) Ethnic inequalities in involuntary admission under the mental health act: an exploration of mediation effects of clinical care prior to the first admission. Br J Psychiatry 222:27–36. 10.1192/bjp.2022.141 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Aguglia A, Moncalvo M, Solia F, Maina G, Aguglia A, Moncalvo M, Maina G (2016) Involuntary admissions in Italy: the impact of seasonality. Int J Psychiatry Clin Pract 20(4):232–238. 10.1080/13651501.2016.1214736 [DOI] [PubMed] [Google Scholar]
- 40.Balducci PM, Bernardini F, Pauselli L, Tortorella A, Compton MT (2017) Correlates of involuntary admission: findings from an Italian inpatient psychiatric unit. Psychiatria Danubina 29(4):490–496. 10.24869/psyd.2017.490 [DOI] [PubMed] [Google Scholar]
- 41.Gou, L., Zhou, J., Xiang, Y., Zhu, X., Correll, C. U., Ungvari, G. S.,… Wang, X. (2014). Frequency of Involuntary Admissions and Its Associations With Demographic and Clinical Characteristics in China. Archives of Psychiatric Nursing, 28(4), 272–276. 10.1016/j.apnu.2014.04.002 [DOI] [PubMed]
- 42.Opjordsmoen, S., Friis, S., Melle, I., Haahr, U., Johannessen, J. O., Larsen, T.K.,… McGlashan, T. H. (2010). A 2-year follow-up of involuntary admission’s influence upon adherence and outcome in first-episode psychosis. Acta Psychiatrica Scandinavica, 121(5), 371–376. 10.1111/j.1600-0447.2009.01536.x [DOI] [PubMed]
- 43.Ielmini M, Caselli I, Poloni N, Gasparini A (2018) Compulsory versus voluntary admission in psychiatry: an observational study. Minerva Psichiatrica 59(3):129–134. 10.23736/s0391-1772.18.01967-2 [Google Scholar]
- 44.Cougnard A, Kalmi E, Desage A, Misdrahi D, Abalan F, Brun-Rousseau H, Verdoux H (2004) Factors influencing compulsory admission in first-admitted subjects with psychosis. Soc Psychiatry Psychiatr Epidemiol 39(10):804–809. 10.1007/s00127-004-0826-5 [DOI] [PubMed] [Google Scholar]
- 45.Canova Mosele PH, Figueira C, Antônio G, Bertuol Filho A, Ferreira de Lima JAR, Calegaro VC (2018) Involuntary psychiatric hospitalization and its relationship to psychopathology and aggression. Psychiatry Res 265(March):13–18. 10.1016/j.psychres.2018.04.031 [DOI] [PubMed] [Google Scholar]
- 46.Jankovic J, Parsons J, Jovanović N, Berrisford G, Copello A, Fazil Q, Priebe S (2020) Differences in access and utilisation of mental health services in the perinatal period for women from ethnic minorities - A population-based study. BMC Med 18(1):1–12. 10.1186/S12916-020-01711-W/FIGURES/2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Wellesley Wesley, E., Patel, I., Kadra-Scalzo, G., Pritchard, M., Shetty, H., Broadbent,M.,… de Freitas, D. F. (2021). Gender disparities in clozapine prescription in a cohort of treatment-resistant schizophrenia in the South London and Maudsley case register. Schizophrenia Research, 232, 68–76. 10.1016/j.schres.2021.05.006 [DOI] [PubMed]
- 48.Tseris EJ, Hart B, E., Franks S (2022) My voice was discounted the whole way through: A gendered analysis of women’s experiences of involuntary mental health treatment. Affilia - Feminist Inq Social Work 37(4):645–663. 10.1177/08861099221108714 [Google Scholar]
- 49.Nyikavaranda P, Pantelic M, Jones CJ, Paudyal P, Tunks A, Llewellyn CD (2023) Barriers and facilitators to seeking and accessing mental health support in primary care and the community among female migrants in Europe: a feminisms systematic review. Int J Equity Health 22(1):1–21. 10.1186/s12939-023-01990-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.Keown P, Murphy H, McKenna D, McKinnon I (2018) Changes in the use of the mental health act 1983 in England 1984/85 to 2015/16. Br J Psychiatry 213(4):595–599. 10.1192/bjp.2018.123 [DOI] [PubMed] [Google Scholar]
- 51.Yohanna D (2013) Deinstitutionalization of people with mental illness: causes and consequences. Am Med Association J Ethics 15(10):886–891 [DOI] [PubMed] [Google Scholar]
- 52.Frost DM (2020) Hostile and harmful: structural stigma and minority stress explain increased anxiety among migrants living in the united Kingdom after the brexit referendum. J Consult Clin Psychol 88(1):75–81. 10.1037/ccp0000458 [DOI] [PubMed] [Google Scholar]
- 53.Choy B, Arunachalam K, G. S, Taylor M, Lee A (2021) Systematic review: acculturation strategies and their impact on the mental health of migrant populations. Public Health Pract 2. 10.1016/j.puhip.2020.100069 [DOI] [PMC free article] [PubMed]
- 54.Terhune J, Dykxhoorn J, Mackay E, Hollander AC, Kirkbride JB, Dalman C (2020) Migrant status and risk of compulsory admission at first diagnosis of psychotic disorder: A population-based cohort study in Sweden. Psychol Med 1–10. 10.1017/S0033291720002068 [DOI] [PMC free article] [PubMed]
- 55.McKenzie K, Bhui K (2007) Institutional racism in mental health care. BMJ 334(7595):645. 10.1136/bmj.39161.665579.BE [DOI] [PMC free article] [PubMed] [Google Scholar]
- 56.Katsakou C, Marougka S, Garabette J, Rost F, Yeeles K, Priebe S (2011) Why do some voluntary patients feel coerced into hospitalisation? A mixed-methods study. Psychiatry Res 187(1–2):275–282. 10.1016/j.psychres.2011.01.001 [DOI] [PubMed] [Google Scholar]
- 57.Goodwin L, Gazard B, Aschan L, MacCrimmon S, Hotopf M, Hatch SL (2018) Taking an intersectional approach to define latent classes of socioeconomic status, ethnicity and migration status for psychiatric epidemiological research. Epidemiol Psychiatric Sci 27(6):589–600. 10.1017/S2045796017000142 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 58.Department of Health and Social Care (2019) Focus groups: a qualitative exploration of perspectives on the Mental Health Act and people of African and Caribbean descent. Retrieved from https://assets.publishing.service.gov.uk/media/5c6596e140f0b676df6e4746/Independent_Review_of_the_Mental_Health_Act_1983_-_supporting_documents.pdf
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
No datasets were generated or analysed during the current study.