Ambulance attendances resulting from self-harm after release from prison: a prospective data linkage study
MetadataShow full item record
Objective: Incarcerated adults are at high risk of self-harm and suicide and remain so after release into the community. The aims of this study were to estimate the number of ambulance attendances due to self-harm in adults following release from prison, and to identify factors predictive of such attendances. Methods: Baseline surveys with 1309 adults within 6 weeks of expected release from prison between 2008 and 2010 were linked prospectively with state-wide correctional, ambulance, emergency department, hospital and death records in Queensland, Australia. Associations between baseline demographic, criminal justice and mental health-related factors, and subsequent ambulance attendances resulting from self-harm, were investigated using negative binomial regression. Results: During 4691 person-years of follow-up (median 3.86 years per participant), there were 2892 ambulance attendances in the community, of which 120 (3.9%) were due to self-harm. In multivariable analyses, being Indigenous [incidence rate ratio (IRR): 2.10 (95% CI 1.14–3.86)], having previously been hospitalised for psychiatric treatment [IRR: 2.65 (95% CI 1.44–4.87)], being identified by prison staff as being at risk of self-harm whilst incarcerated [IRR: 2.12 (95% CI 1.11–4.06)] and having a prior ambulance attendance due to self-harm [IRR: 3.16 (95% CI 1.31–7.61)] were associated with self-harm attendances. Conclusions: Ambulance attendances resulting from self-harm following release from prison are common and represent an opportunity for tertiary intervention for self-harm. The high prevalence of such attendances, in conjunction with the strong association with prior psychiatric problems, reinforces the importance of providing appropriate ambulance staff training in the assessment and management of self-harm, and mental health problems more broadly, in this vulnerable population.
Social Psychiatry and Psychiatric Epidemiology
This publication has been entered into Griffith Research Online as an Advanced Online Version.
Psychology not elsewhere classified