PURPOSE: To compare sex-specific rates of hospital admission and repeat admission following self-harm between ethnic groups in London and test whether differences persist after adjustment for socio-economic deprivation.

METHODS: A population-based cohort of all individuals aged over 11 admitted to a general hospital for physical health treatment following self-harm between 2008 and 2018, using administrative Hospital Episode Statistics for all people living in Greater London.

RESULTS: There were 59,510 individuals admitted to the hospital following self-harm in the 10 year study period, ethnicity data were available for 94% of individuals. The highest rates of self-harm admission and readmission were found in the White Irish group. Rates of admission and readmission were lower in Black and Asian people compared to White people for both sexes at all ages and in all more specific Black and Asian ethnic groups compared to White British. These differences increased with adjustment for socio-economic deprivation. People of Mixed ethnicity had higher rates of readmission. Rates were highest in the 25-49 age group for Black and Mixed ethnicity men, but in under-25 s for all other groups. There were substantial differences in rates within the broader ethnic categories, especially for the Black and White groups.

CONCLUSION: In contrast to earlier UK studies, self-harm rates were not higher in Black or South Asian women, with lower self-harm admission rates seen in almost all ethnic minority groups. Differences in rates by ethnicity were not explained by socio-economic deprivation. Aggregating ethnicity into broad categories masks important differences in self-harm rates between groups.

Original languageEnglish
Pages (from-to)1967-1977
Number of pages11
JournalSocial Psychiatry and Psychiatric Epidemiology
Issue number11
Early online date20 Apr 2021
Publication statusPublished - Nov 2021


Dive into the research topics of 'Variation in rates of self-harm hospital admission and re-admission by ethnicity in London: a population cohort study'. Together they form a unique fingerprint.

Cite this