Assessing the risk of self-harm: the impact of safety interventions on the predictive validity of risk assessment in secure care

Abstract

Purpose Self-harm is a cause of concern for health-care professionals. The Short-Term Assessment of Risk and Treatability (START) is a short-term assessment instrument used to rate the likelihood of risk behaviours, including self-harm. As result of the assessment, interventions that are implemented to reduce the risk of self-harm may reduce the strength of the predictive validity of a risk assessment tool. The aim of this study was explore the impact of risk management interventions on the capacity of START to predict self-harm. It was predicted that the interventions would weaken the ability of START to predict self-harm in patients who received the intervention. Design/methodology/approach Secondary analysis of routinely collected data in a large sample of women in an inpatient secure care setting. Demographic and clinical information, self-harm episodes, safety management interventions and START assessments were extracted and used to build an anonymous database. Findings START significantly predicted self-harm in those with and without the safety management intervention. However, the strength of the predictive validity was smaller in those who received the intervention compared to those without. Practical implications The results suggest that the implementation of safety management interventions needs to be taken into account when assessing future risk of self-harm. Originality/value To the best of the authors’ knowledge, this is the first study to explore the impact of safety management interventions on the predictive validity of START in a large sample of women.