Burnout as a predictor of self-reported sickness absence among human service workers:Prospective findings from three year follow up of the PUMA study

Aim: To investigate whether burnout predicts sickness absence days and sickness absence spells in human service workers. Method: A total of 824 participants from an ongoing prospective study in different human service sector organisations were eligible for the three year follow up analysis. Burnout was measured with the work related burnout scale of the Copenhagen Burnout Inventory. Sickness absence was measured with self-reported number of days and spells during the last 12 months before the baseline and the follow up survey. A Poisson regression model with a scale parameter was used to account for over dispersion. A linear regression model was used for analysing changes in burnout and absence between baseline and follow up. Results: Burnout was prospectively associated with both sickness absence days and sickness absence spells per year. Differences in sickness absence days varied from a mean of 5.4 days per year in the lowest quartile of the work related burnout scale to a mean of 13.6 in the highest quartile. An increase of one standard deviation on the work related burnout scale predicted an increase of 21% in sickness absence days per year (rate ratio 1.21, 95% CI 1.11 to 1.32) after adjusting for gender, age, organisation, socioeconomic status, lifestyle factors, family status, having children under 7 years of age, and prevalence of diseases. Regarding sickness absence spells, an increase of one standard deviation on the work related burnout scale predicted an increase of 9% per year (rate ratio 1.09, 95% CI 1.02 to 1.17). Changes in burnout level from baseline to follow up were positively associated with changes in sickness absence days (estimate 1.94 days/year, SE 0.63) and sickness absence spell (estimate 0.34 spells/year, SE 0.08). Conclusion: The findings indicate that burnout predicts sickness absence. Reducing burnout is likely to reduce sickness absence.