Randomized controlled trials of mental health nurse-delivered interventions: A systematic review

WHAT IS KNOWN ON THE SUBJECT?: Well conducted randomized controlled trials provide the highest level of evidence of effectiveness of healthcare interventions, including those delivered by mental health nurses. Trials have been conducted over the years but there has not been a comprehensive review since 2005, and never one including studies conducted outside the UK. WHAT THE PAPER ADDS TO EXISTING KNOWLEDGE?: The paper provides a comprehensive overview of results from randomized controlled trials of mental health nurse-delivered interventions conducted in the UK, Ireland, US, Australia, New Zealand, or Canada and reported 2005 to 2020. It highlights that the trial evidence is limited and offers only partial evidence for interventions that are central to mental health nursing practice. WHAT ARE THE IMPLICATIONS FOR PRACTICE?: Much mental health nursing practice is not supported by the highest level trial evidence. Mental health nurses need to carefully select evidence on which to base their practice both from the mental health nursing literature and beyond. Mental health nurses and other stakeholders should demand greater investment in trials to strengthen the evidence base.

INTRODUCTION: Nurses are the largest professional disciplinary group working in mental health services and have been involved in numerous trials of nursing-specific and multidisciplinary interventions. Systematic appraisal of relevant research findings is rare.

To review trials from the core Anglosphere (UK, US, Canada, Ireland, Australia, New Zealand) published from 2005 to 2020.

A systematic review reported in accordance with the Preferred Reporting Items for Systematic reviews and Meta-Analyses. Comprehensive electronic database searches were conducted. Eligible studies were randomized controlled trials of mental health nurse-delivered interventions conducted in relevant countries. The risk of bias was independently assessed. Synthesis involved integration of descriptive statistics of standardized metrics and study bias.

Outcomes from 348 between-group comparisons were extracted from K = 51 studies (N = 11,266 participants), Standardized effect sizes for 68 (39 very small/small, 29 moderate/large) statistically significant outcomes from 30 studies were calculable. All moderate/large effect sizes were at risk of bias.

Trial evidence of effective mental health nurse-delivered interventions is limited. Many studies produced few or no measurable benefits; none demonstrated improvements related to personal recovery.

Mental health nurses should look beyond gold standard RCT evidence, and to evidence-based interventions that have not been trialled with mental health nurse delivery.