Relationship between commuting and health outcomes in a cross-sectional population survey in southern Sweden

BACKGROUND The need for a mobile workforce inevitably means that the length of the total work day (working and traveling time) will increase, but the health effects of commuting have been surprisingly little studied apart from perceived stress and the benefits of physically active commuting. METHODS We used data from two cross-sectional population-based public health surveys performed in 2004 and 2008 in Scania, Sweden (56% response rate). The final study population was 21, 088 persons aged 18-65, working > 30 h/week. Duration (one-way) and mode of commuting were reported. The outcomes studied were perceived poor sleep quality, everyday stress, low vitality, mental health, self-reported health, and absence from work due to sickness during the past 12 months. Covariates indicating socioeconomic status and family situation, overtime, job strain and urban/rural residency were included in multivariate analyses. Subjects walking or cycling to work < 30 min were used as a reference category. RESULTS Monotonous relations were found between duration of public transport commuting and the health outcomes. For the category commuting > 60 min odds ratios (ORs) ranged from 1.2 - 1.6 for the different outcomes. For car commuting, the relationships were concave downward or flat, with increasing subjective health complaints up to 30-60 min (ORs ranging from 1.2 - 1.4), and lower ORs in the > 60 min category. A similar concave downward relationship was observed for sickness absence, regardless of mode of transport. CONCLUSIONS The results of this study are concordant with the few earlier studies in the field, in that associations were found between commutation and negative health outcomes. This further demonstrates the need to consider the negative side-effects of commuting when discussing policies aimed at increasing the mobility of the workforce. Studies identifying population groups with increased susceptibility are warranted.