RESUMO
BACKGROUND: To investigate the association between long working hours and self-rated health (SRH), examining the roles of potential confounding and mediating factors, such as job characteristics. METHODS: Data were pooled from seven waves (2005-2011) of the Korean Labour and Income Panel Study. A total of 1578 workers who consecutively participated in all seven study years were available for analysis. A generalized estimating equation for repeated measures with binary outcome was used to examine the association between working hours (five categories; 20-35, 36-40, 41-52, 53-68 and ≥ 69 h) and SRH (two categories; poor and good health), considering possible confounders and serial correlation. RESULTS: Associations between working hours and SRH were observed among women, but only for the category of the shortest working hours among men. The associations with the category of shortest working hours among men and women disappeared after adjustment for socioeconomic factors. Among women, though not men, working longer than standard hours (36-40 h) showed a linear association with poor health; OR = 1.41 (95% CI = 1.08-1.84) for 52-68 working hours and OR = 2.11 (95% CI = 1.42-3.12) for ≥ 69 working hours. This association persisted after serial adjustments. However, it was substantially attenuated with the addition of socioeconomic factors (e.g., OR = 1.66 (95% CI = 1.07-2.57)) but only slightly attenuated with further adjustment for behavioural factors (e.g., OR = 1.63 (95% CI = 1.05-2.53)). The associations with job satisfaction were significant for men and women. CONCLUSIONS: The worsening of SRH with increasing working hours only among women suggests that female workers are more vulnerable to long working hours because of family responsibilities in addition to their workload.