Meta-regression analyses of relationships between burnout and depression with sampling and measurement methodological moderators.

Abstract

Despite 35 years of study, burnout researchers have failed to reach a consensus about whether burnout is distinct from depression. This review compiled reports containing zero-order correlations between scores on burnout, depression, and other measures of negative affect (NA) based on (a) reviews published by Kahill (1988), Glass and McKnight (1996), and Bianchi et al. (2015b), and (b) a search of PsycInfo using "depression" and "burnout" as search terms to find relevant studies published after 2014. The resulting data set contained 69 studies with 196 burnout-depression correlations based on 46,191 participants. We found an overall effect size of .492 between scores on burnout and depression measures (essentially equivalent to the .52 value reported in Koutsimani et al.'s, 2019, review) and an effect size of .546 between the Maslach Burnout Inventory emotional exhaustion scale and depression. Similarly, a correlation of .53 between burnout and NA measures is similar in size to the .46 correlation found by Koutsimani et al. Moderator analyses indicated that a larger burnout-depression correlation was associated with a higher proportion of female participants in a study, older participants, participants who had worked longer, and burnout measures with higher reliability estimates. The effects of age and years employed on the burnout-depression relationship suggest that repeated and negative work experiences are required for burnout to develop to the extent that its effects overlap with symptoms of depression. Conceptualizing the empirical relation between burnout and depression as a single point estimate may miss the more complex empirical picture. (PsycInfo Database Record (c) 2022 APA, all rights reserved).