Article Text

Download PDFPDF

Young black females in three UK cities have higher rates of self-harm than other ethnic groups but are less likely to be referred for psychiatric care
  1. Simon Hatcher
  1. Department of Psychological Medicine, The University of Auckland, Auckland, New Zealand

Statistics from



What is the age- and gender-specific prevalence of self-harm by ethnic group in UK cities, and does access to specialist assessment and care vary according to demographic factors?


The general population (aged 16–64 years) served by five hospitals in three UK cities. Data on the population in the hospitals' catchment areas was obtained from the 2001 Census.


General hospitals in Manchester, Derby and Oxford, UK; January 2001 to December 2006.


Computerised records were used to identify cases of self-harm presenting to emergency departments during the study period. Data on ethnicity were collected by the assessing clinician or were recorded from hospital patient record systems and were available on 75% of patients. Participants classified as ‘South Asian’, ‘Black’ or ‘White’; other ethnic groups were excluded from this study due to low numbers. The proportion of individuals re-presenting with self-harm within 12 months of their index case was assessed.


Emergency department presentation with intentional self-poisoning or self-injury (prevalence per 1000 person years). No exclusions were made according to motivation or degree of suicidal intent.



Prospective cohort study.

Main results

There were 33 314 episodes of self-harm recorded in 20 574 individuals aged 16–64 years during the study period. Of these individuals, 14 997 fell into the three ethnic groups being assessed and 8401 resided in postcodes in the city catchment areas of the hospitals. When data were pooled across the three cities, there was no difference in the rates of self-harm among South Asian females aged 16–34 years compared to white females of the same age (rate ratio (RR) 0.99, 95% CI 0.87 to 1.11). Black females in this age group had a higher rate of self-harm than white females (RR 1.70, 95% CI 1.46 to 1.98). The effect remained after sensitivity analysis which included all individuals with missing ethnicity data as white, although the magnitude of the effect was reduced (RR 1.21, 95% CI 1.04 to 1.40). In patients aged 16–64 years, rates of self-harm repetition were lower in South Asian and black individuals than in white individuals (South Asian: RR 0.5, 95% CI 0.4 to 0.7; black: RR 0.6, 95% CI 0.4 to 0.8). There were lower rates of specialist psychiatric assessment in black compared with white individuals (RR 0.9, 95% CI 0.8 to 0.98). Both black and South Asian patients had lower rates of psychiatric referral than white patients (South Asian: RR 0.5, 95% CI 0.5 to 0.6; black: RR 0.6, 95% CI 0.5 to 0.8).


In three UK cities, there is a higher rate of presentation to emergency departments of cases of self-harm in young black females than in young white or South Asian females. Black and South Asian patients are less likely to receive referral for specialist psychiatric assessment and follow-up services than white patients. However, the highest rates of self-harm repetition occurred in white patients.


Only emergency department attendees residing in the hospitals' city catchment areas were included in the prevalence analyses. All individuals in the three ethnic groups being analysed were included in the assessment of characteristics and clinical outcome, regardless of area of residence.

Abstracted from

OpenUrlAbstract/FREE Full Text


At the start of the 21st century, psychiatry has started to ask sophisticated questions about culture, including how culture affects the outcome and management of common problems like self-harm. This paper is another one to come from a monitoring system for self-harm set up in three English cities and demonstrates the usefulness of this approach.

The authors found that young black females had higher rates of self-harm but paradoxically received less psychiatric care. Before accepting this, there are several issues to consider. First, accurate identification and recording of ethnicity in emergency departments is poor and there was no check on the accuracy of the data. Second, the classification of culture was simplistic (eg, it did not consider complexities such as children of mixed marriages, first vs later generations of immigrants or the varieties of culture within a ‘white’ culture).

There are two ways of thinking about culture. One is to categorise people into cultural groups such as Black, White or Indian. These are characterised by a shared set of rules that members adhere to. Unfortunately, this often results in unhelpful and inaccurate cultural stereotypes and rarely tells the clinician anything useful about what to do with the person in front of them. The second more interesting way to think about it is to see culture as a dimension on a scale of ‘sense of belonging’. People may belong to many things including a place, a religion, a family story or a shared history, for example. This also generates useful ideas about what questions to ask clinically and can lead to effective interventions to increase sense of belonging. For the problems of self-harm and suicide the sense of not belonging has been around since at least Durkheim's time and Joiner1 has recently been writing about ‘failed belongingness’. So for clinicians assessing people in the emergency department who have presented with self-harm while it may be important epidemiologically to know that young black women have higher rates of suicide, a more fruitful line of enquiry may be to ask “Where do you belong?”



  • Sources of funding Department of Health/NHS R&D funding programme.


  • Competing interests None.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.