Evaluating the Safe Steps for De-escalation: A protocol for a mixed concurrent control study in acute mental health units

Abstract

There is a shared goal of organising reform efforts in mental health services to eliminate restrictive practices and improve therapeutic relationships. However, evidence on high-quality, culturally safe, co-produced, and strengths-based interventions and evaluations is limited, especially for complex interventions centred on therapeutic responding. In response, a multi-centre, mixed concurrent control study is underway to evaluate the Safe Steps for De-escalation, a multi-component intervention focused on a structured framework for mental health nurses’ therapeutic responses to emotional distress and interpersonal conflict in acute adult mental health inpatient units. The aims of this evaluation were: 1) What is the effectiveness of Safe Steps in reducing restrictive practice events and duration and physical injuries? 2) Does Safe Steps improve people’s service experience, perceived staff action towards violence prevention, and nurses’ professional quality of life and emotionally intelligent workplace behaviours? 3) What factors influence the successful implementation of Safe Steps? It is hypothesised that: a) intervention sites will demonstrate more significant decreases in restrictive practice events and duration and physical injuries, compared to within-group baseline and control group, and b) measures of people’s experiences and perceptions and nurses’ outcomes and behaviours will improve, compared to within-group baseline. Safe Steps has three components: i) a structured de-escalation framework, ii) an in-person and online training programme, and iii) a regular conduct of strengths-based, data-informed restrictive practice review meetings. The control group will be usual care. Other outcomes include nursing intervention clusters, their associations with various outcomes, and factors influencing intervention implementation and restrictive practice use. There is no randomisation, but inverse probability weighting will be applied. The sample sizes were determined through power analyses and supporting evidence on saturation in qualitative research. Various quantitative and qualitative data treatments and measures will be undertaken to minimise research biases.