Introduction Night shift work is associated with adverse health effects. Yet, some persons prefer working permanent night shifts and it is speculated that they tolerate night work better than others. The aim of is to study associations between permanent night work, age and sickness absence. Due to self-selection out of night work over time by those who experience negative effects of night work, we hypothesised that older workers with permanent night work are ‘healthy workers’ with less sickness absence compared to other groups.

Methods Information on working hours, age (20–34 years, 35–49 years and >50 years) and sickness absence was obtained from the Danish Working Hour Database, which contains daily information on starting and ending time of working hours based on payroll data for all employees at Danish public hospitals (2008–2015). For each year with >50 workdays, individual schedules were classified as permanent day, evening or night (>88% of work days with night work), 2-shift (day/evening, day/night or evening/day) or 3-shift (day/evening/night). We applied linear regression with individual as random intercept (participants served as their own controls) for employees (n=5774) with at least one year of permanent night work.

Results Employees aged >50 years had 25.6 sickness days/year when working permanent night, which is more than in all other schedules. In comparison, employees had 9.0 (sd=1.0) fewer days when working permanent day, 7.3 (sd=2.4) for permanent evening, 6.3 (sd=0.9) for day/evening, 3.2 (sd=0.6) for day/night, 5.3 (sd=1.0) for evening/night, and 5.5 (sd=0.7) for day/evening/night. There was no interaction effect between schedule and age.

Discussion Employees had more sickness absence when working permanent night work compared to any other schedule. The association was not modified by age and did therefore not indicate that older employees with permanent night work tolerate night work better than others.