Protocol: Psychosocial Interventions Addressing Mental Health Outcomes Among Hospital-Based Healthcare Workers in...

Healthcare workers employed in hospital settings face significant occupational stressors that place them at elevated risk for a range of mental health outcomes, including burnout, depression, anxiety, post-traumatic stress disorder, secondary traumatic stress, compassion fatigue, and moral injury. These challenges carry serious consequences for healthcare worker well-being, patient safety, care quality, and workforce retention. Despite growing recognition of this crisis and recent federal legislation allowing funding to address healthcare worker mental health, no comprehensive synthesis exists identifying which psychosocial interventions are most effective for this population in the United States.

This systematic review and meta-analysis aims to examine the effects of psychosocial interventions on mental health outcomes among healthcare workers employed in United States hospital settings, to identify moderators or intervention effectiveness, and to provide actionable evidence to inform program development, resource allocation, and policy implementation.

We will search eight electronic databases and complete an extensive grey literature search to identify studies published from 2010 to present. Eligible study designs include randomized controlled trials (RCTs), cluster randomized controlled trials (CRCTs), and quasi-experimental designs (QEDs) with a control group examining psychosocial interventions delivered within or through hospital-affiliated systems in the United States. Participants will include hospital-based healthcare workers who provide direct clinical care. Primary outcomes include burnout, depression symptoms, anxiety symptoms, post-traumatic stress disorder (PTSD), secondary traumatic stress, vicarious trauma, compassion fatigue, and moral injury, assessed using validated measures. Secondary outcomes include turnover intention and turnover rates, extracted when reported. Two reviewers will independently screen studies and extract data. Risk of bias will be assessed using the ROB 2 tool for RCTs and the ROBINS-I V2 tool for QEDs. Where sufficient data are available, meta-analyses will be conducted using a random-effects model with robust variance estimation (RVE) to account for dependent effect sizes. Heterogeneity will be assessed using the Q statistic and I² statistic, interpreted alongside 95% prediction intervals and visual inspection of forest plots. Moderator analyses will explore intervention type, healthcare worker role, treatment duration, delivery format, and facilitator type as potential sources of heterogeneity.