Sociodemographic, health and criminological risk factors for post-imprisonment mental health and substance use-related service contacts: A nationwide retrospective cohort study using linked data (RELEASE)

Citation

Kjellgren R, Savinc J, Dougall N, Watson J, Brown A, Parkes T, Kurdi A, Leyland A, Haddow C, Hunt K & Connell C (2026) Sociodemographic, health and criminological risk factors for post-imprisonment mental health and substance use-related service contacts: A nationwide retrospective cohort study using linked data (RELEASE). Social Psychiatry and Psychiatric Epidemiology. https://doi.org/10.1007/s00127-026-03217-x

Abstract
Purpose To quantify differences in health service (community, unscheduled/emergency, inpatient) contact for mental health and substance use pre and post-release from prison, and explore variations between subgroups based on sociodemographic, health and criminological risk factors. Methods We conducted a nationwide retrospective cohort study of mental health and substance use-related health service contacts for everyone released from a Scottish prison in 2015. A whole-systems approach utilising linked administrative health and prison data was used to assess contacts over a four-year follow-up period, across nine health services spanning community, unscheduled/emergency, and inpatient care. The analysis is focused on subgroup variations among people released from prison; we have previously reported findings from a matched population comparison elsewhere [1]. Incidence rate ratios, using person-years of time-in-community as a denominator, with 95% confidence intervals, were estimated to compare community contacts prior to, and following, the 2015 index release. Multivariable negative binomial models were fitted for each service to explore subgroup variation. Results Contact incidence rates increased in the follow-up period in four out of nine services compared to the four years prior. Sociodemographic, health and criminological risk factors contribute to variability, and demonstrate complex patterning, across the nine services. The number of pre-index prison episodes, time spent in prison, and circumstances of index release (e.g., supervision, probation) were independently associated with post-imprisonment contacts. Conclusion Cumulative time in prison and repeated short-term imprisonment, and circumstances of release, contribute to different experiences upon release and variations in post-imprisonment health service contacts for mental health and substance use. Health and criminal justice policymakers must consider how imprisonment and release may exacerbate health needs and service demands, and ensure harms of imprisonment are minimised.

Keywords
Mental Health; Substance use; Imprisonment; Whole-system analysis; Linked data

• Published Version