A Sentinel Events Approach to Addressing Suicide and Self-Harm in Jail

Statement of the Problem: Suicides are the leading cause of mortality in U.S. jails. As sentinel events (SE), incidents of suicide and self-harm signal a breakdown in systems for ensuring inmate health and safety, yet most jails do not have adequate processes for reviewing these incidents and taking corrective action based on those reviews. The Vera Institute of Justice (Vera) proposes to address this unmet health need by developing and testing a sentinel events approach for reviewing jail suicide and self-harm that builds on current practices in New York City (NYC) and lessons from analogous community settings.

Partnerships: This project is a partnership between Vera, the NYC Department of Correction (DOC), and the Department of Health and Mental Hygiene (DOHMH), and will be guided by a national group of experts in corrections and correctional health care.

Research Design and Methods: Vera will develop, implement, and evaluate a Self-Harm Analysis and Review Protocol (SHARP) for responding to cases of serious self-harm in the NYC jail with the aim of designing a nationally replicable model. Vera will use several methods including: a literature review of SE best practices, a review of recent Morbidity and Mortality Reviews (N = 8-10) to document exiting procedures for investigating suicides, a review of the medical and corrections records of individuals who engaged in serious self-harm (N = 10), a survey of DOC and DOHMH staff (N = 200) to explore barriers to effectively addressing suicide and self-harm, semi-structured interviews with key DOHMH and DOC informants (N = 18-25), an analysis of administrative data to describe predictors of self-harm and evaluate post-implementation changes in the rate of self-harm, onsite observation of SHARP reviews (N = 5), and an analysis of the SHARP action plan reports (N = 10) to assess whether SHARP resulted in tangible changes to jail policies and practices.

Analysis: Jail administrative records and survey data will be analyzed using logistic regression and relative risk ratio models to determine individual and system level predictors of suicide and self-harm. Semi-structured interviews and case files will be analyzed using qualitative analytical methods.

Products, Reports, and Data Archiving: Vera will disseminate findings via a series of publications aimed at policymakers and practitioners, including guidelines and a manual for implementing SHARP in correctional settings, research briefs, conference presentations, and peer-reviewed scholarly articles. Administrative data files, survey data and action plan analyses will be archived according to NIJ guidelines.
ca/ncf