Objectives Shift work disrupts hormonal rhythms, but evidence linking specific shift patterns to cortisol/melatonin variations remains limited. This study evaluated the associations of a rapid-rotating shift pattern (ie, PAN pattern, shift transitions from afternoon to morning to night shifts within approximately 40 hours) and irregular non-PAN pattern shifts with urinary cortisol and melatonin among Hong Kong nurses, while also assessing the role of sleep on rhythmic hormones. Methods A 7-day actigraphy-based study recruited 201 shift nurses and 100 daytime workers. First-morning voids on the first non-workday postnight shift or postoffice work measured cortisol and melatonin metabolite (aMT6s) levels. Actigraphy measured sleep parameters. Generalised linear models examined the associations. Results PAN pattern nurses slept less than non-PAN nurses (5.8 vs 6.4 hours) and exhibited greater circadian phase delay (16.4 vs 15.9 hours). Non-PAN nurses had the lowest cortisol, cortisone and aMT6s levels, with significantly reduced cortisol levels (β=−0.31, p=0.017) compared with daytime workers. Sleep characteristics strongly influenced hormone levels, with each additional hour of sleep decreasing cortisol levels by 10.3%, while each later hour of wake-up time was associated with 3.9% lower cortisol levels; later sleep midpoints decreased melatonin by 5.1% per hour. A three-way interaction revealed weakened cortisol–sleep duration associations with later wake-up times in PAN nurses (p=0.032). Conclusions Irregular shifts, longer sleep duration and delayed wake-up time significantly reduced wakening cortisol levels, while delayed sleep timing suppressed melatonin secretion. These findings highlight how shift arrangements and sleep characteristics disrupt hormonal rhythms in shift workers. Optimising shift patterns and sleep interventions (eg, prioritising duration/consistency) may mitigate circadian disruption and improve shift worker health.
Li et al. (Wed,) studied this question.