Predictors of sickness absence and presenteeism: Does the pattern differ by a respondent’s health?

Objectives: We examine the predictors of sickness presenteeism in comparison with sickness absenteeism. The paper focuses on the effects of working-time match and efficiency demands and differentiates the estimates by a respondent’s self-assessed health. Methods: We use survey data covering 884 Finnish trade union members in 2009. We estimate logit models. All models include control variables such as the sector of the economy and the type of contract. Results: Working-time match between desired and actual weekly working hours reduces both sickness absence and presenteeism in the whole sample that consists of workers with all health levels. The point estimates reveal that working-time match decreases the prevalence of sickness absence by 7% and presenteeism by 8%. However, the estimates that differentiate by a respondent’s health show that this pattern prevails only for those workers who have poor health. Hence, the point estimates for those who have poor health are much larger than the ones for the whole sample. Working-time match reduces the prevalence of sickness absence by 21% and presenteeism by 20% for those workers who have poor health. In contrast, working-time match has no influence whatsoever on the prevalence of work-related sickness for those who have good health. We also find that efficiency demands increase presenteeism 4 in the whole sample. However, additional results reveal that this pattern prevails only for those workers who have good health. Conclusions: The effects of working-time match and efficiency demands on the prevalence of sickness absence and presenteeism are strongly conditional upon a worker’s self-assessed health level. Therefore, the worker’s initial health is an important attribute that has to be taken into account when one is designing appropriate policies to reduce sickness absence and presenteeism.