ABSTRACT

The PhD dissertation is built on data from the PUMA project initiated by professor Tage Søndergård Kristensen, and it summarizes the results of the PhD-project "Burnout in human service work - causes and consequences" carried out during the period March 2003 to February 2005 at the National Institute of Occupational Health, Copenhagen, Denmark. The dissertation includes four papers and had two major aims: 1) to investigate possible causes for burnout, and 2) to evaluate burnout as a predictor for sickness absence. Burnout is a "grassroots" concept introduced in the 1970's as a particular type of prolonged occupational stress that seemed to occur most prominently among human services professionals, with emotional exhaustion as its core symptom. Until start of the new millennium, little was known about causes and consequences for burnout because most studies were cross-sectional.

Data for the PhD-project are based on questionnaire data and stems from baseline (n=1914) and 3-year follow-up (n=1024) of the PUMA study, an ongoing 6-year prospective intervention study in the human services sector. Burnout was measured with a new instrument, the Copenhagen Burnout Inventory (CBI), psychometric properties of which are evaluated as part of the study. As potential causes of burnout, the study evaluated psychosocial work environment factors, which were measured with the Copenhagen Psychosocial Questionnaire (COPSOQ) and with some additional items on client-related work. Sickness absence was measured by self-reported number of sickness absence days and spells during the last 12 months before the baseline and the follow-up survey. Linear regression models and Poisson regression models were used for the analyses.

The major prospective findings regarding causes for burnout were that high possibilities for development, high predictability, high role-clarity, and low role-conflicts at baseline had a protective effect against subsequent burnout (at a 3-year follow-up), after adjusting for several potential confounders and for burnout level at baseline. The analyses showed that burnout measured at baseline predicted both sickness absence days and spells at 3-year follow-up. In addition, increase in burnout levels predicted increases in sickness absence, and decrease in burnout predicted decrease in sickness absence. These associations remained after controlling for several potential confounders, including socio-demographics, type of workplace, socio-economic status, health-related lifestyle, family status, and prevalence of disease.

It can be concluded that the analyses for this PhD dissertation have shown that specific factors in the work environment protect against burnout and that burnout increases the likelihood of sickness absence. This suggests that burnout is preventable and that burnout prevention is of importance for the reduction of sickness absence.