Objective Previous studies indicate that evening workers have increased risk of long-term sickness absence. The aim of the present study was to examine if evening work is associated with higher risk of sickness absence compared to daytime work.

Methods The Danish Working Hour Database (DWHD) contains administrative payroll data from all Danish regions from 2007 to 2013. It includes 249,218 unique participants with 224,485,710 observations with starting and ending times for each worked shift and sickness absence on daily basis. For the present explorative study we selected all employees in a single hospital in 2009 (6,254 unique participants; 83% women; age: 40.8 years (SD = 11.9 years); 1,461,690 observations). The dataset was further restricted by excluding short (≤3 hours) and long (≥9 hours) shifts and observations with sickness absence or non-day/evening (e.g. night) shifts during the past 28 days. A sensitivity analysis with evening workers as their own controls, was carried out by excluding participants, who never worked evening.

Evening work was defined as one or more evening shifts (≥3 hours between 18:00–23:00) during the past 28 days. The reference was dayshift (≥3 hours between 06:00–20:00). Using a multi-state model we analysed recurrent transitions between three states: work, sickness absence and temporary out of work e.g. holiday. Hazard ratios (HR) for sickness absence were calculated by Cox proportional hazards models adjusted for age and sex.

Results Analyses showed that evening work was associated with lower risk of sickness absence (HR: 0.84, 95% CI: 0.73–0.96). The sensitivity analysis yielded similar results (HR: 0.87, 95% CI: 0.72–1.06).

Conclusion When comparing periods with evening work to periods of day work the present study showed that evening work was associated with a decreased occurrence of sickness absence. This is in contrast to previous studies which compare evening workers with day workers.