Commuting and sick leave: a retrospective longitudinal study among a Belgian military population.

Objectives: In a military context, people often have to deal with long commuting distance. The aim of the current study is to investigate to what extent commuting distances predict sickness absence among a military population.

Methods: The present study is based on a dynamic cohort of active-duty military personnel of the Belgian Defence. A proportional hazard Andersen-Gill model was used to determine the effect of commuting distance on the recurrence of sickness absence, considering relevant covariates. In a second model, an interaction term for commuting and traffic congestion was introduced.

Results: Higher commuting distances led to higher risk of recurring sickness absence, which was 9%, 15%, 8% and 11% higher, respectively, for distances of 5-14 km, 15-29 km, 30-59 km or >60 km compared with a distance of <5 km.A commuting distance of 30-59 km increased the risk with 47%, 33%, 35% and 76% if the reported traffic congestion was, respectively, 5-20%, 20-40%, 40-70% and 70-100% compared with a commuting distance of <5 km and a reported traffic congestion of 0-5%. A commuting distance of >60 km increased the risk with 25% and 76%, respectively, for a reported traffic congestion of 5-20% and 70-100% compared with a commuting distance of <5 km and a reported traffic congestion of 0-5%.

Conclusions: These findings support measures to reduce the distance and frequency of commuting in order to reduce sickness absence, considering for example remote working, a more responsible geographical distribution of the offices or a revision of the employees' job-changing policies.