Reciprocal Effects Between Job Stressors and Burnout: A Continuous Time Meta-Analysis of Longitudinal Studies

• Published in: Psychological bulletin 2020-12, Vol.146 (12), p.1146-1173
• Main Authors: Guthier, Christina, Dormann, Christian, Voelkle, Manuel C.
• Format: Article
• Language: English
• Subjects: Adult; Ambiguity; Bias; Burnout; Burnout, Professional - psychology; Cynicism; Depersonalization; Emotional Exhaustion; Emotional Responses; Emotions; Fatigue; Female; Human; Humans; Job Burnout; Job control; Job Satisfaction; Longitudinal Studies; Male; Measurement; Meta-analysis; Middle Aged; Occupational Stress; Occupational Stress - psychology; Publication Bias; Statistical power; Surveys and Questionnaires; Systematic review; Work Engagement; Work Load; Workplace Intervention
• ISSN: 0033-2909; 1939-1455
• DOI: 10.1037/bul0000304

Results from longitudinal studies are ambiguous regarding the direction of effects between job stressors and burnout over time. We meta-analyzed possible reciprocal relations between job stressors and burnout in k = 48 longitudinal studies (N = 26,319), accounting for variation of time intervals in primary studies by using continuous time meta-analysis. Additionally, we analyzed whether country-level job resources (job control and job support; k = 31 European studies, N = 17,747) moderated the effect of job stressors on burnout (stressor-effect) and the effect of burnout on job stressors (strain-effect). Further, we analyzed the replicability of the primary studies by assessing between-study heterogeneity, publication bias, and statistical power. Reciprocal effects between job stressors and burnout exist. The stressor-effect is small, whereas the strain-effect is larger and moderated by job control and job support. Analyses of the different burnout symptoms (emotional) exhaustion and depersonalization/cynicism demonstrated that reciprocal relations between emotional exhaustion and job stressors exist, but depersonalization/cynicism is not directly related to job stressors. Between-study heterogeneity was comparable with other psychological studies, whereas statistical power of primary studies was comparatively large. Conclusions are limited because few primary studies used time intervals of less than 12 months, more than two measurement occasions, and objective measures of stressors. Overall, results imply the need for extended job stress models and new job stress interventions that help employees cope with burnout symptoms. Public Significance Statement This meta-analysis reveals that job stressors and burnout mutually affect each other, with burnout increasing job stressors much more than vice versa. Employees having burnout symptoms should be provided with appropriate resources to avoid increasing levels of job stressors to stop a possible vicious circle between job stressors and burnout and, therefore, prevent the development of critical levels of burnout.