Abstract

This study examined the association between daily working hours and accelerometer-based sleep duration, sleep efficiency, and number of awakenings per hour of sleep on the following night among 800 older public sector employees in Finland (mean age 63 years in the first measurement they participated in, 87% women) with 4,818 measurement nights in total. Information on working hours was derived from daily logs and categorized into: 1) 6 h, 2) 7, hours 3) 8 h, 4) 9 h, and 5) 10 or more hours of work. The most common category (i.e. workdays with 8 h of work) was used as the reference category in the analyses. Nights followed by a workday and a free day were analyzed separately. No differences were observed in sleep duration between the reference group and the other working hour categories when the next day was a workday nor when the next day was a free day. After a 6-hour workday, sleep efficiency was on average 1.0 percentage points higher and there were on average 0.13 less awakenings per hour of sleep when compared with the reference category. When the next day was a free day, no differences in sleep quality were observed. Thus, no clear indication of a dose-response relationship between working hours and either duration or quality of sleep was found. Furthermore, future research should further examine the possibility that the association between working hours and sleep is somewhat different depending on whether the workday is followed by another workday or a free day.