Background: Self-harm is a strong predictor of future suicide, but little is known about self-harm among the homeless population. The study aim was to estimate the incidence of self-harm among the homeless population and to assess factors associated with self-harm.
Methods: Data on self-harm presentations to 34 hospital emergency departments in Ireland were collected by the National Self-Harm Registry Ireland (NSHRI). Index presentations between 2010 and 2014 were included for the homeless and fixed residence populations. Incidence rates of self-harm were calculated using NSHRI data and census estimates. Factors associated with self-harm and repeated self-harm were analysed by multivariable-adjusted logistic regression.
Results: The age-standardised incidence rate of self-harm was 30 times higher among the homeless (5572 presentations per 100,000) compared with those with a fixed residence (187 presentations per 100,000). Homeless people had significantly higher odds of being male (OR 1.86, 95%CI 1.56-2.23), presenting with self-cutting (vs. overdose, OR 2.15, 95%CI 1.74-2.66) and having psychiatric admission (vs. general admission, OR 2.43, 95%CI 1.66-3.57). Homeless people had higher odds of self-harm repetition within 12 months (vs. fixed residence, OR 1.46, 95%CI 1.21-1.77). The odds of repetition were significantly increased among homeless who engaged in self-cutting (vs. overdose, OR 1.76, 95%CI 1.17-2.65) and did not receive psychiatric review at index presentation (vs. reviewed, OR 1.54, 95%CI 1.05-2.26).
Limitations: The study only reflects self-harm presenting to hospital, and assumes no change in homelessness status after index presentation. Residual confounding may affect the results.
Conclusion: There is a disproportionate burden of self-harm among the homeless. Targeted preventive actions are warranted.
Keywords: Epidemiology; Homeless; Incidence; Overdose; Self-cutting; Self-harm.
Copyright © 2017 Elsevier B.V. All rights reserved.