Compassion satisfaction, resilience and passion for work among nurses and physicians working in intensive care units: A mixed method systematic review

To identify, appraise, and synthesise current evidence on prevalence, correlates, and interventions to enhance compassion satisfaction, resilience, and passion for work among nurses and physicians working in intensive care units.A mixed methods systematic review was conducted. The comprehensive search used the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. Seven databases (MEDLINE, EMBASE, CINAHL, JBI, ProQuest, PsycINFO, and Cochrane Library) were searched for literature published between January 2011 and June 2021. The Mixed Methods Appraisal Tool was used to assess methodological quality. Data from included studies were analysed using a convergent mixed methods design. The protocol was prospectively registered (PROSPERO 2021 CRD42021252051).A total of 37 studies met the inclusion criteria. Most studies reported moderate levels of compassion satisfaction among intensive care health professionals, whereas levels of resilience varied. Compassion satisfaction and resilience were positively correlated, but relationships between compassion satisfaction and resilience and other correlates (personal factors, psychological factors, and work-related factors) were inconsistently reported. Only four interventions aimed to improve compassion satisfaction or resilience among intensive care health professionals. None of the included studies investigated passion for work.Compassion satisfaction, resilience, and passion for work among staff in the intensive care unit are important in the current global COVID-19 pandemic. Health professionals report a moderate level of compassion satisfaction but findings in relation to resilience are mixed. No studies examined passion for work. Further research to determine ongoing psychological wellbeing and professional quality of life and evaluate tailored interventions to support intensive care staff well-being is recommended.