The effectiveness of peer-led interventions to improve work-related psychosocial outcomes and reduce turnover of support workers in residential aged care: A systematic review and meta-analysis

Background: Support workers are central to the delivery of residential aged care, but the workforce is facing increasing work demands and widespread shortages. This contributes to high rates of burnout, decreased job satisfaction and high staff turnover. Peer-led interventions are reported to be effective but it is necessary to use evidence-based interventions to support this key workforce group. Objective: This study aimed to evaluate the scientific evidence on effectiveness of strategies improving psychosocial and turnover-related outcomes for support workers in aged care that could be incorporated into a peer-led intervention. Design: Systematic review and meta-analyses of experimental and quasi-experimental studies. Setting: Residential aged care. Methods: A systematic literature review was conducted using MEDLINE (via PubMed), EMBASE (via Scopus), and CINAHL (via EBSCO). We included studies examining the effectiveness of workplace interventions aiming to reduce aged care support workers' turnover rates and/or improve their work-related psychosocial outcomes (such as work stress, job satisfaction, selfesteem, and other). A number of meta-analyses using a mixed-effects model were performed to calculate standardized mean differences and odds ratios. Results: Fifty-one studies were included: 15 randomised controlled trials (RCTs), 19 non-RCTs and 17 Pre-Post studies. Most of the studies were rated as having 'high' or 'very high risk of bias'. The studies were clustered by intervention type: 1) knowledge-based, 2) interpersonal skills-based, 3) team-building, and 4) self-care. Knowledge-based interventions were the most used approach, with 26 studies in this category, and frequently reported improvements in stress- and satisfactionrelated outcomes. There were twelve interpersonal skills-based and nine team-building interventions, which often reported decreased work stress, staff turnover, and intention to quit. There were four self-care interventions of which only one reported improvements in stress-related outcomes. Meta-analyses showed that only knowledge-based interventions resulted in statistically significant improvements: lower staff turnover rates (OR 0.47, 95 %CI: 0.37, 060), and higher scores for job/life satisfaction (SMD 0.26, 95 % CI: 0.05, 0.46) and staff attitude (SMD 0.23, 95 % CI: 0.05, 0.45). Conclusion: This review found numerous strategies that have been trialled to improve support workers' psychosocial- and turnover-related outcomes. Most studies reported improvements in outcomes. However, our meta-analyses suggest that the effect sizes were small and mostly nonsignificant, with the evidence being of low certainty. The evidence for effectiveness of knowledge-based interventions appears the most convincing, with statistically significant improvements reported for turnover rates, job/life satisfaction and staff attitude. More high-quality studies are needed to consolidate the existing evidence. PROSPERO registration number: CRD42017059007; 02 June 2017. Tweetable abstract: Knowledge-based interventions most promising in improving support workers' outcomes in aged care. #agedcare #staffturnover