Shift work and mental health: a systematic review and meta-analysis Abstract Background Shift work is common. However, research findings are mixed regarding the impact of shift work on mental health. This systematic review sought to provide a comprehensive summary of existing research examining the association between different types of shift work and mental health. The review included large-scale, non-occupation-specific research. Methods Four electronic databases PubMed, PsycINFO, Web of Science and SCOPUS were searched to identify studies that reported on the statistical association between shift work and mental health and that used population-based samples. Two reviewers extracted information about study characteristics and data on the association between shift work and mental health. A meta-analysis was performed for longitudinal studies adopting a ‘broad binary’ measure of shift work. Results Thirty-three studies were included in the final review—10 cross-sectional studies, 22 longitudinal studies, and 1 study that included both. Findings were grouped based on whether the measure of shift work focussed on: (1) night/evening work, (2) weekend work, (3) irregular/unpredictable work schedule, or (4) a broad binary measure. There was a reasonable level of evidence that overall, when a broad binary measure was adopted, shift work was associated with poorer mental health—this finding was supported by the meta-analysis results. There was also some evidence that irregular/unpredictable work was associated with poorer mental health. There was less evidence for night/evening and minimal evidence for weekend work. Inconsistencies in study methodology, limited contrasting and combining the results. Conclusions The association between shift work and mental health is different across types of shift work. The evidence is strongest for a broad binary, general measure of shift work and for irregular or unpredictable shift work. There is a need for continued research that adopts consistent and clear measures of shift work.