Across ages, ethnicity and race-based discrimination affects both the risk of mental health difficultiesReference Cave, Cooper, Zubrick and Shepherd1 and differential experiences of care. In the UK, differences exist in referral and treatment pathways to mental healthcare, with racially and ethnically minoritised young people more likely to be referred through social care, criminal justice or secondary care than through primary care, affecting the likelihood of accessing appropriate and timely mental health supports.Reference Chui, Gazard, MacCrimmon, Harwood, Downs and Bakolis2,Reference Edbrooke-Childs and Patalay3 London-based research further indicates that Black and Asian young people are more likely to be referred to in-patient and emergency services rather than out-patient mental health supports, compared with White British young people, indicating differential case management in community settings and primary care.Reference Chui, Gazard, MacCrimmon, Harwood, Downs and Bakolis2 In addition to inequalities in referral pathways and triage, further barriers may impede receipt of treatment and reduce the likelihood of remaining in treatment, including insufficient language translation services, lack of cultural humility, limited treatment options offered, and stereotyping and discrimination perpetuating distrust of professionals.Reference Devonport, Ward, Morrissey, Burt, Harris and Burt4,Reference Memon, Taylor, Mohebati, Sundin, Cooper and Scanlon5
At the extreme, if stress is not adequately managed, it has the potential to escalate, increasing the risk of needing emergency treatment. Previous analyses indicate 7.9% of girls and 4.1% of boys aged 11–17 years in England were hospitalised with stress-related presentations.Reference Blackburn, Ajetunmobi, Mc Grath-Lone, Hardelid, Shafran and Gilbert6 Given the differential pathways and barriers to mental healthcare for minoritised young people, there is a need to understand their experiences of care-seeking and hospital admission at the point of crisis. Moreover, existing inequalities research on mental healthcare use and referral pathways has largely focused on internalising symptoms, such as anxiety or depression, despite known variations in manifestations of stress among young people, including emotional, behavioural and physical symptoms. As such, considering the full spectrum of stress-related presentations may provide greater insight into potential bias in pathways and differential access to care.
Aims
We aimed to explore whether there are systematic differences based on race and ethnicity in hospital admissions for stress-related presentations in secondary school-aged young people. This research addresses the evidence gap examining stress experienced by and healthcare use among minoritised young people in the UK. The project aimed to examine rates of unplanned admissions and indicators of admission experience, including the distribution of repeat admissions, duration of admissions and types of stress-related presentations across racial-ethnic groups. We also examined variation based on assigned gender, in an attempt to understand the complex intersectional experience of gender and racialisation.Reference Collins and Bilge7 Given that these outcomes reflect stress-related presentations that were sufficiently severe enough to result in an emergency hospital admission, we examined the experience of those during a period of crisis, which may provide insight into differential pathways or missed opportunities for intervention across groups.
Method
Data and population
We used the Hospital Episode Statistics (HES) Admitted Patient Care (APC) data-set, an administrative hospital database capturing all admissions to National Health Service (NHS) hospitals in England.Reference Herbert, Wijlaars, Zylbersztejn, Cromwell and Hardelid8 HES APC records information on up to 20 ICD-10 diagnostic codes and 24 operation procedure codes (OPCS-4) for each episode of care recorded by trained clinical coders. Secondary school-aged young people aged 11–15 years, residing in England, who had an emergency (unplanned) hospital admission between 1 April 2014 and 15 March 2020 were eligible for study inclusion (Supplementary Fig. 1 available at https://doi.org/10.1192/bjp.2024.123). We selected the age range 11–15 years to reduce differences in referral pathways, as YP in England can choose to leave secondary school for alternative education or training from age 16 years.9
We adopted a birth cohort approach to enable us to examine the ethnic groups as used in the UK Census described below. Disaggregation of racial-ethnic groups is crucial to understanding health inequalities, given the role of differing socio-political-historical factors on health and well-being. Office for National Statistics (ONS) mid-year population estimates by year of age were not available according to these groups; we did not use ETHPOP database denominators for the same reason.
Given the focus on ethnicity and to allow for exploration of any gender differences, individuals with missing data for assigned gender and ethnicity were excluded from the analyses. Individuals with a recorded death in ONS death registrations were also excluded from the analyses, as their use of health services is likely to differ from other young people.
Exposure
Our exposure was racial-ethnic group, recognising young people's recorded ‘ethnicity’ as a proxy for the lived experiences of individuals from minoritised ethnic groups living in a White British majority country. We intentionally use the term racial-ethnic group throughout this paper, given that the variable ‘ethnicity’ in UK administrative data includes both ‘race-based’ (‘White’) and ‘ethnicity-based’ (‘Pakistani’) identifiers. Racial-ethnic groups were coded into 12 groups, merging some groups because of small numbers: White British, White Other (Irish and White Other), Mixed White–Black (White and Black African and White and Black Caribbean), Mixed Other (White and Asian and Mixed Other), Indian, Pakistani, Bangladeshi, Asian Other (Chinese and Asian Other), Black Caribbean, Black African, Black Other and Other (any other ethnic group). Ethnicity values were obtained across all of young people's HES APC records; where values conflicted, a modal value was used. If a young person had missing ethnicity across all of their APC record, we also sought ethnicity in their HES out-patient and birth records.
Outcomes
Outcomes were emergency admission stress-related presentations in HES APC. All emergency admissions in HES APC, regardless of specialty, were included in the analyses. Hospital transfers or admissions within 1 day of discharge were treated as a continuation of the previous admission. An emergency admission was deemed a stress-related presentation if the primary diagnosis related to potentially psychosomatic symptoms (e.g. pain, fatigue) or mental health problems, including internalising, externalising or thought disorders (Supplementary Table 1). Additionally, an admission was counted as a stress-related presentation if self-harm was indicated in any diagnostic position. We used the code-list developed by Blackburn et al,Reference Blackburn, Ajetunmobi, Mc Grath-Lone, Hardelid, Shafran and Gilbert6 which we adapted in line with the ICD-11 and to reflect research suggesting that rather than traditional psychiatric classification systems, mental health disorders (encompassed by general psychopathology) are better captured by the two broad latent dimensions, internalising and externalising, as well as a third dimension, thought disorder, which captures symptoms of psychosis and mania (Supplementary Appendix 1).Reference Carragher, Krueger, Eaton and Slade10–Reference Krueger, Kotov, Watson, Forbes, Eaton and Ruggero12 In addition, to exclude admissions where there was a known medical cause, we expanded the list of exclusion codes indicating a medical or surgical cause for physical health symptoms (Supplementary Appendix 1). We excluded stress-related presentations identified in pregnancy-related admissions, defined in line with previous publications.Reference Harron, Gilbert, Cromwell and van der Meulen13
To capture the complexity and comorbidity that arises in presentations, admissions that included a stress-related presentation in the primary diagnostic position and a different stress-related presentation code in any other diagnostic position were classified as both. For example, if young person had a primary diagnosis of ‘F10 Mental and behavioural disorders due to use of alcohol’ and a diagnosis of ‘F323 Severe depression with psychotic symptoms’ in any other diagnostic position, this would be counted as an externalising and thought disorder related admission.
Cohort characteristics
HES APC documents patient ‘sex’ and age as well as region and area-level deprivation (the English indices of deprivation).14 Given unclear documentation and variation in clinical practice, variable ‘sex’ in patient history was deemed most likely indicative of assigned gender at birth or clinician-determined based on gender presentation, as such reference to ‘gender’ in this paper refers to ‘assigned gender' not ‘gender identity’, which is not yet reliably captured in HES APC data.
Analyses
Rates of stress-related presentations across racial-ethnic groups were calculated per 1000 births, stratified by assigned gender and age, using HES birth records as a population estimate. We calculated the age-specific incidence of a first stress-related presentation and estimated the cumulative incidence of stress-related presentation admissions for young people born in 2003–2004. The proportion of admissions that were stress-related, as well as the types of stress-related presentations diagnosed during admission and the durations of admissions, were compared across racial-ethnic groups. We also compared the proportion of young people with a repeat admission across groups. We calculated the relative risk of repeat admissions between 1 April 2014 and 15 March 2020, longer admission duration (≥3 days), types of stress-related presentations and multiple diagnoses during an admission across racial-ethnic groups compared with all groups combined.Reference Grant, Sullivan, Hebert-Beirne and Acosta15 All data cleaning, preparation and analyses were completed with R software for Windows, version 4.3.0 (R Foundation for Statistical Computing, Vienna, Austria; see https://www.r-project.org/).
Using the available administrative data, it was determined that it is not possible to disentangle the effects of socioeconomic position from historical and ongoing experiences of discrimination, oppression and exclusion, in particular, based on class, race or migration status, which have contributed to lower socioeconomic positioning for minoritised groups and directly affected health outcomes. As such, socioeconomic position is on the same causal pathway, and adjusting for it risks diluting the impact of systemic inequalities on the health outcomes of interest. Therefore, we chose not to adjust for socioeconomic position and report unadjusted results only.
Ethics statement
This study uses de-identified NHS Hospital Episode Statistics data provided to the researchers by NHS England within the terms of a data-sharing agreement (DARS-NIC-393510-D6H1D-v8.10). Study-specific ethical approval was not required.
Results
Summary of cohort and admissions
We identified 210 973 young people aged 11–15 years born in England between 1 April 2003 and 15 March 2009, who had 312 362 emergency admissions between 1 April 2014 and 15 March 2020. Their characteristics are presented in Supplementary Tables 2 and 3.
Combined, 30.9% of all emergency admissions (96 484 of 312 363) were stress-related (Supplementary Table 4). This proportion was highest for White British (32.5%) and lowest for Black African (20.6%) young people.
Rates of stress-related presentation admissions
Rates of admissions varied by age and racial-ethnic group (Fig. 1). At ages 11–13 years, Pakistani, Mixed White–Black race/ethnicity and White British young people had the highest rates of stress-related presentation admissions. At ages 14 and 15 years, Mixed White–Black race/ethnicity and White British young people continue to experience the highest admission rates, with higher rates among their Mixed Other race/ethnicity peers, as the rate slows for their Pakistani peers. Across all ages, White Other, Black African and Indian young people have the lowest admission rates.
Fig. 1 Rates of admission for stress-related presentations (SRPs) per 1000 births for young people aged 11–15 years born April 2003 to March 2009, and admitted to hospital between April 2014 and March 2020, across racial-ethnic groups.
Cumulative incidence
By 2019–2020, of the young people born between April 2003 and March 2004, 21 436 had been admitted at least once with a stress-related presentation. We estimated the cumulative incidence of stress-related admissions among young people between 11 and 15 years born in England between April 2003 and March 2004, to be 5%, highest for Pakistani (5.4%) and White British (5.3%) young people, and lowest for Black African (3.0%) and Asian Other (3.2%) young people (Supplementary Fig. 2, Supplementary Table 6).
Trends in stress-related presentation admissions
The highest proportion of repeat (two or more) admissions were among Pakistani (47.7%), White British (41.4%) and Mixed White–Black race/ethnicity (41.3%) young people, whereas this was less for other minoritised groups, with White Other (35.8%) and Black Other (36.4%) young people least likely to be re-admitted (Fig. 2(a)).
Fig. 2 (a) Trends in the percentage of young people admitted with stress-related presentation with singular and repeat stress-related presentation admissions, across racial-ethnic groups. (b) Trends in the percentage of stress-related presentation admissions by duration of admission in days, across racial-ethnic groups.
Young people from Black Other (16.6%), Bangladeshi (16.3%), Asian Other (15.9%) and Black Caribbean (15.8%) groups experienced a greater proportion of longer admissions (3 days or more) compared with their White British (11.9%) and Indian (11.8%) peers (Fig. 2(b)). Notably, although Bangladeshi and Black Other young people have a correspondingly lower proportion of admissions that lasted <1 day (41.7 and 42.7%, respectively), Black Caribbean and Asian Other young people are also among the highest proportion of admissions lasting <1 day (45.3 and 45.1%, respectively), indicating a differential spread in duration of admissions across racial-ethnic groups.
Types of stress-related presentation admissions
Patterns for types of stress-related presentations were mixed (Fig. 3). Young people from all Asian groups (Indian, Pakistani, Bangladeshi, Asian Other), as well as Black African and Other race/ethnicity groups, were admitted with potentially psychosomatic symptoms more frequently than their White and Mixed race/ethnicity peers. The opposite trend was found for internalising and self-harm presentations. White and Mixed racial-ethnic groups were admitted most frequently with internalising and self-harm concerns, whereas Pakistani, Black African and Indian young people were least likely to be recorded with internalising or self-harm diagnoses. Similarly, White British and Mixed race/ethnicity young people were more frequently admitted with externalising concerns, whereas all Asian groups and Black African young people were admitted least frequently. Young people from all Black groups, Black Other (2.5%), Black Caribbean (1.7%) and Black African (1.4%) were more likely to be diagnosed with thought disorder diagnoses on emergency admission than their White British (0.5%), Mixed race/ethnicity, Indian and Pakistani (0.4%) peers.
Fig. 3 Percentage of stress-related presentation (SRP) admissions with a recorded diagnoses related to (a) psychosomatic, (b) internalising, (c) externalising, (d) thought disorder* and (e) self-harm symptoms, across racial-ethnic groups. Note that y-axis scales differ. *Results for Mixed White–Black race/ethnicity, Mixed Other race/ethnicity and Indian groups with recorded diagnoses related to thought disorder were censored because of the small cell count.
Multiple diagnoses during admissions
During the same stress-related presentation admission, young people from White and Mixed race/ethnicity groups were more likely to be diagnosed with more than one type of stress-related presentation than their South Asian (Indian, Pakistani, Bangladeshi) and Black African peers (Fig. 4).
Fig. 4 Trends in the percentage of stress-related presentation admissions by number of diagnoses recorded during an admission, across racial-ethnic groups.
Stress-related presentation admissions visual summary
The heatmap (Supplementary Fig. 3, Supplementary Table 11) displays the relative risk of each racial-ethnic group (compared with all groups combined) experiencing repeat admissions; admission durations ≥3 days; admissions with recorded diagnoses related to psychosomatic, internalising, externalising and self-harm symptoms; and admissions with more than one stress-related presentation diagnoses during an admission. The relative risk of admissions associated with thought disorder diagnoses is visualised (Supplementary Fig. 4, Supplementary Table 12) for racial-ethnic groups with sufficient cell counts (≥10).
Intersection with gender
Across racial-ethnic groups, females were more likely to be admitted for stress-related concerns than their male counterparts (Supplementary Table 5). Across racial-ethnic groups, the gap between female and male admissions widened steadily with age; however, the extent varied considerably, with White British, Mixed White–Black, Mixed Other and Other race/ethnicity young people experiencing the widest gap from age 13 onward. For the 2003–2004 cohort, among females, the cumulative incidence at age 15 years was highest for White British (7.1%), and lowest for Black African (3.7%) and Asian Other (3.9%). Among males, the cumulative incidence was highest for Pakistani (4.5%) and lowest for Black African (2.4%) and Other race/ethnicity (2.4%) (Supplementary Table 6).
There were also differences in readmissions, duration of admission, type of presentation and multimorbidity (Supplementary Tables 7–10). Across groups, males were admitted with somatic symptoms at a higher frequency than females. Pakistani (96.8%), followed by Indian (95.4%), males experienced the highest proportion of admissions related to psychosomatic presentations, whereas White British females (69.9%) were least likely to be recorded with a psychosomatic-related admission (Supplementary Table 9). The opposite trend was true for admissions associated with internalising and self-harm diagnoses; females were admitted at a higher frequency than males across groups. Females from White and Mixed race/ethnicity groups were admitted with internalising (17.5–20.7%) and self-harm (38.4–43.9%) presentations most frequently, whereas Pakistani males were least likely to be admitted with either internalising (3.4%) or self-harm (3.2%) concerns (Supplementary Table 9).
Discussion
Main findings
Our study is the first to report on rates of mental health-related acute hospital admissions for minoritised young people. We identified important variations in admission patterns across racial-ethnic groups, indicative of differential experiences during triage, assessment, admission thresholds and treatment.
Young people from White British, Mixed White–Black and Other race/ethnicity groups experienced the highest overall admission rates, with White Other and Black African young people having the lowest admission rates. These overall admission rates mask more nuanced differences in the proportions of young people who are ever admitted with a stress-related presentation aged 11–15 years (i.e. cumulative incidence) versus those with higher rates of repeat admissions and/or longer durations of hospital stay. Females were more likely to be admitted with stress-related presentations, experience repeat admissions and remain admitted for longer (relative to males), with the smallest gender gap observed for Indian, Pakistani and Black African young people.
We found variation in the types of stress-related presentations young people from different racial-ethnic groups were admitted with. All Asian groups and Black African young people were admitted more frequently with potentially psychosomatic symptoms, and less frequently with internalising, externalising or self-harm presentations, relative to the White and Mixed race/ethnicity groups. Black Caribbean young people were less likely to be admitted with internalising symptoms than their White and Mixed race/ethnicity peers, but had similar frequency of externalising and self-harm admissions. Black Other young people were markedly more likely to be diagnosed with a thought disorder during an emergency hospital admission than their Mixed race/ethnicity, Indian, Pakistani and White British peers (2.5% v. <1%).
Findings in context
Inequalities in access
Previous research documents inequalities in accessing timely and appropriate mental healthcare, as well as differential treatment experiences, for minoritised young people in the UK.Reference Chui, Gazard, MacCrimmon, Harwood, Downs and Bakolis2,Reference Edbrooke-Childs and Patalay3,Reference Ruphrect-Smith, Davies, Jacob and Edbrooke-Childs16 Given the additional barriers to accessing mental health services and greater likelihood of referrals to in-patient services, minoritised young people may be forced to access emergency services at a point of crisis. Our study of mental health-related emergency hospital admissions identified distinct patterns of admissions across racial-ethnic groups, painting a nuanced picture. At the extremes, White British young people were among the highest levels of readmissions and lowest levels of extended admissions; the opposite was true for Black Other young people. Although White Other young people were least likely to be readmitted and Pakistani young people were most likely to be readmitted, they were equally likely to remain in hospital for admissions of 3 days or more. Mixed White–Black race/ethnicity, Black Caribbean and Asian Other young people experienced similar levels of readmissions as their White British peers, but they continued to experience longer admissions. Notably, not only were Black Caribbean and Asian Other young people more likely to experience longer admissions, they were also among the groups with the highest levels of short admissions (<1 day). It is important to consider that although longer durations may suggest more severe presentations and shorter durations may infer presentations better supported by other services, these variations may also indicate disparities in the perceived relative severity of conditions based on racial-ethnic group. Collectively, these findings indicate structural barriers and potential bias in accessing care that may be driven by inconsistencies in clinical judgement during triage and initial assessment processes, similar to settings in the USA.Reference Aysola, Clapp, Sullivan, Brennan, Higginbotham and Kearney17,Reference Joseph, Kennedy, Landry, Marsh, Baymon and Im18
Inequalities in assessment
Where assessments do not account for variations in articulation or conceptualisation of symptoms cross-culturally,Reference Chapman, DeLapp, Williams, Beidel, Frueh and Hersen19 symptoms may be more or less likely to be missed, misclassified or receive further assessment, resulting in variation in treatment options offered, admission duration as well as diagnoses recorded. We found that young people racialised as Black or Asian were more likely to be admitted with somatic symptoms than their White British peers. As with all mental health presentations, perceptions of discrimination and microaggressions are associated with increased somatic symptoms; however, findings are mixed regarding whether minoritised individuals are more likely to endorse depression-related somatic symptoms than White individuals.Reference Barnes and Bates20,Reference Sauceda, Patel, Santiago-Rodriguez, Garcia and Lechuga21 Ultimately, it may be that minoritised young people experience differential or culturally insensitive assessment, resulting in physiological concerns being misdiagnosed as psychosomatic or alternative presentations being missed or receiving inadequate attention.
Young people from White and Mixed racial-ethnic groups were more likely to be admitted with internalising symptoms, with South Asian groups and Black African young people least likely to be diagnosed with the same. Although this is consistent with results of the Mental Health of Children and Young People (MHCYP) in England Survey,22 which found that anxiety and depressive disorders were most commonly diagnosed in White British and Mixed/Other 5- to 19-year-olds, it contradicts self-reported difficulties.Reference Bains and Gutman23 Among adults, prevalence estimates of internalising mental health conditions vary based on measure, that is, self-report, clinical interview or professionally diagnosed.Reference McManus, Bebbington, Jenkins and Brugha24 Although further research is needed to understand variations across racial-ethnic groups among children and young people, in context, these results continue to raise questions about differential assessment procedures and treatment pathways.
Pathways may also be different for externalising symptoms, including impulsivity, aggression and substance use, which have received less attention in the UK. White British and Mixed race/ethnicity young people were more likely to be admitted with externalising concerns than their peers, mirroring clinical interview outcomes from the MHCYP in England Survey.22 Notably, because of the reductive nature of the racial-ethnic group categories used in the MHCYP in England Survey, it is not clear whether our findings mirror the experiences of Black Caribbean young people who were assessed. However, our findings are in line with self-reported difficulties, indicating that Black Caribbean young people were more likely to report experiencing externalising symptoms.Reference Bains and Gutman23 It may be worth considering whether externalising difficulties displayed by White British and Mixed race/ethnicity young people are more likely to be recognised as indicative of a mental health concern, whereas young people racialised as Black or South Asian, particularly boys, are more likely to be criminalised in UK society.25,26 This aligns with inequalities in referral pathways showing Black and Asian young people are more likely to be referred to mental health services through justice and social care pathways, rather than directly to health services.
Black Other, Black Caribbean and Black African young people were more likely to receive diagnoses related to mania and psychosis on emergency admission than their White British, Mixed race/ethnicity, Indian and Pakistani peers. In particular, Black Other young people were over five times more likely to be diagnosed with thought disorder presentation than White British or Mixed race/ethnicity young people. Diagnosis of psychotic disorders in people aged under 18 years are considered relatively uncommon (0.4–1.7%), and should only be made after long-term monitoring, referral and comprehensive assessment by specialist mental health service teams.27 Failure to account for developmental or cultural factors increases the risk of misdiagnosis.Reference McClellan28 Nevertheless, these results reflect the decades-long disproportionate diagnoses of psychosis in Black adults in the UK,Reference Halvorsrud, Nazroo, Otis, Brown Hajdukova and Bhui29 an increased risk attributed to social and economic disadvantage, structural and interpersonal racism.Reference Bardol, Grot, Oh, Poulet, Zeroug-Vial and Brunelin30,Reference Nazroo, Bhui and Rhodes31
Consistent with previous UK-based findings, we found all Asian groups and Black African young people were less likely to be admitted for self-harm.Reference Farooq, Clements, Hawton, Geulayov, Casey and Waters32 As self-harm is an expression of or an attempt to cope with distress rather than a mental health diagnosis, assessment needs to prioritise understanding the function of the behaviour and carefully consider the underlying mental health difficulties in treatment planning.33 Further, it is well-established that a fraction of young people who self-harm actually present to healthcare services; as such, estimations based on hospital records are likely underestimations of lived experience.Reference Clements, Turnbull, Hawton, Geulayov, Waters and Ness34 Given that Asian and Black African young people were also less likely to receive a secondary diagnosis, taken together, in the context of self-harm, there is a potential for missed indicators of risk, impeding safety and treatment planning.
Further consideration should be given to the ripple effects of the pandemic and school closures, including increased admissions for stress-related presentations,Reference Blackburn, Owen, Downs and Gilbert35 internalising symptoms and rates of self-harm among girls,Reference Trafford, Carr, Ashcroft, Chew-Graham, Cockcroft and Cybulski36 particularly in the context of widening inequalities.
Strengths and limitations
Study strengths include longitudinal and whole nation coverage of acute hospital admissions for young people in England. We refined a previous code-listReference Blackburn, Ajetunmobi, Mc Grath-Lone, Hardelid, Shafran and Gilbert6 (updated in line with the ICD-11) to include the full spectrum of stress-related presentations. However, given this broad and complex definition, misclassification is a concern, including instances where a surgical, medical or physiological cause was present, but not recorded. We used a birth cohort approach to increase specificity in measurement of racial-ethnic groups; however, this resulted in the exclusion of young people not born in England. The relative distributions of racial-ethnic groups in the admissions cohort and birth cohort (our population-level estimate) were generally consistent, except for White Other and Black African young people, who are underrepresented. The mechanism is unclear, but may reflect data collection errors, including clinician-prescribed racial-ethnic group based on physical characteristics, or assumptions based on accent or name. Hospital-recorded ethnicity has not been validated, and there are conceptual limitations in generalising experiences across racial-ethnic groups, as this fails to acknowledge the variations in individuals’ lived experiences.
Implications and conclusions
Discussions around barriers to care often centre stigma and help-seeking behaviours without acknowledging systemic bias and a history of institutionalisation disproportionately experienced by racialised and minoritised communities in the UK. Moreover, it is important to recognise stigma as a learned social response based on historical experience with the function of protecting an individual or group from a negative outcome. By attributing inequalities in access to individual-level factors, it overlooks the structural determinants of mental health and deflects from the differential treatment experienced by those who do access services. Our findings indicate differential experiences based on racial-ethnic group in access to care, experiences of admission and documented diagnoses, which may reflect underlying structural racism in triage, assessment and treatment processes.
It is therefore imperative that anti-racism practices, such as the NHS England Patient and Carer Race Equality Framework,37 be implemented across healthcare settings. Clinicians need to be mindful of differences in how individuals express and make sense of their symptoms, as well as assumptions and potential biases when making clinical judgements, practicing cultural humility.Reference Chapman, DeLapp, Williams, Beidel, Frueh and Hersen19,Reference Lekas, Pahl and Fuller Lewis38 Finally, we need better recording of ‘ethnicity’ in hospital admission records, reflective of an individual's self-identified racial-ethnic group. Further research should explore the referral and admission journey from the community to emergency admission, as well as trends in planned admissions to in-patient services and variations in discharge and referrals to out-patient services. Additional exploration of variation in admission duration and the use of involuntary detainment among young people is warranted, which is known to disproportionately affect Black adults in the UK.39
Supplementary material
Supplementary material is available online at https://doi.org/10.1192/bjp.2024.123.
Data availability
This study uses NHS Hospital Episode Statistics data, provided under the data-sharing agreement (DARS-NIC-393510-D6H1D-v8.10) to the researchers by NHS England. The data do not belong to the authors and cannot be shared by the authors, except in aggregate form for publication. The data is collected by the NHS as part of their care and support. Data can be requested through the NHS Digital Data Access Request Service.
Acknowledgements
We would like to thank Dr Roz Shafran for her contributions to the revised code-list. We gratefully acknowledge all young people whose de-identified data was used in this research.
Author contributions
S.N.C. conceived the study design, to which M.A.J. and R.B. contributed. S.N.C. led on updating the stress-related presentations code-list. S.N.C. carried out data cleaning and analysis, with contributions from M.A.J. S.N.C. led the drafting of the manuscript, with revisions from M.A.J. and R.B.
Funding
S.N.C. is funded by the Medical Research Council Doctoral Training Programme grant MR/N013867/1. This research benefits from and contributes to the National Institute for Health Research (NIHR) Children and Families Policy Research Unit, but was not commissioned by the NIHR Policy Research Programme. The views expressed are those of the author(s) and not necessarily those of the NIHR or the Department of Health and Social Care. R.B. received support from a UK Research and Innovation Fellowship, funded by grant MR/S003797/1 from the Medical Research Council. Research at UCL Great Ormond Street Institute of Child Health is supported by the NIHR Great Ormond Street Hospital Biomedical Research Centre.
Declaration of interest
None.
Open access
eLetters
No eLetters have been published for this article.