Objectives Night work-induced circadian misalignment is associated with an increased risk of adverse health effects. It has been suggested that permanent night workers can entrain to being awake at night and therefore do not experience substantial circadian disruption in relation to night work. The aim of this study was to investigate the effects of night work on diurnal rhythms of cortisol, melatonin, and testosterone among permanent night workers.

Methods Twenty-three men with permanent night work participated in this observational field study. They collected saliva every 4 hours over a 24-hour period for assessment of diurnal rhythms (mesor, amplitude and acrophase) on three days: the 2nd and 4th consecutive night shifts, and a day off. Within-person regression analyses with repeated measures adjusted for age were used to evaluate differences between the 2^nd and 4^th night shifts and between night shifts taken together versus days off.

Results There were no statistically significant differences between diurnal rhythms on the 2^nd and 4^th consecutive night shifts. We observed differences in levels (mesor and amplitude) of cortisol and melatonin on night shifts compared to days off with the highest levels during work days. The timing (acrophase) of testosterone, but not cortisol and melatonin, differed between night shifts and days off. Thus, observed differences between the various hormonal diurnal rhythms to night work indicate internal desynchronization.

Conclusion Studies of how diurnal rhythms of multiple biological systems are affected by night work and the link to health effects are warranted to further elucidate the mechanisms linking night work to adverse health effects.

• Supplementary material