ObjectiveTo determine what, how, for whom and under what conditions individual-focused interventions are effective to improve well-being and decrease burn-out among critical care healthcare professionals.DesignThis study is an umbrella review that used the realist approach, using Realist and Meta-narrative Evidence Synthesis: Evolving Standards guidelines. PsycINFO, Web of Science, CINAHL, MEDLINE, Scopus, ClinicalTrials.gov and ISRCTN databases were searched for published and unpublished systematic reviews and meta-analyses literature between 2016 and 2020. The team appraised and extracted data and identified relationships between content, mechanism and outcomes (CMOs). Theory prepositions were developed using CMOs and were used to refine the existing programme.ResultsA total of 81 interventions from 17 reviews were mapped, including mindfulness interventions, cognitive–behavioural therapy, self-care and coping strategies. The revised programme theory determined that contextual factors such as ethnicity, workload, and work schedules play a crucial role in determining the effectiveness of interventions. Mechanisms including the interventions’ interests, acceptance, and receptivity are also influential in determining engagement and adherence to the intervention. Findings suggest that the solution for burn-out is complex. However, it offers an optimistic view of tailoring and customising one or a combination of interventions, integrating structured education and components of emotional intelligence. Self-care, social support, awareness or mindfulness and self-efficacy are prime components to improve emotional intelligence and resilience for critical care healthcare professionals to improve well-being and decrease burn-out experience.ConclusionsThese findings provide realistic and reliable reporting of outcomes to better support implementation within the ‘real world’. Future research such as seeking validation using expert opinions can provide further in depth understanding of hidden contextual factors, mechanisms and their interactions to provide a greater depth of knowledge ready for application with the critical care population.