Shift work schedule patterns and sleep outcomes: results from the SHIFT2HEALTH survey across eight European countries.

OBJECTIVES: This study aimed to describe shift work patterns in Europe using a data-driven approach and evaluate their impact on adverse sleep outcomes. METHODS: We analyzed 6245 participants from the SHIFT2HEALTH survey across eight European countries, who provided information on work schedules, shift work metrics, sleep quality and timing (MCTQshift) across different shifts, and chronic insomnia (Insomnia Severity Index). We used mixtures of von Mises-Fisher models to identify shift clusters based on shift start and end times and evaluated differences in shift-specific sleep duration and quality using Bayesian multilevel models. Associations of shift metrics with short average sleep duration and chronic insomnia, adjusting for confounders, were evaluated using Bayesian logistic regression. RESULTS: We identified five shift clusters: day, evening and night (mean shift lengths: 7.8, 7.9 and 9.5 hours, respectively), long day and long evening (mean lengths: 10.5 and 12.8 hours, respectively). Sleep duration and quality varied between clusters, with the shortest duration [mean 6.00, 95% credible interval (CI) 5.95-6.06 hours] and poorest quality (29.3% "poor"/"very poor") sleep reported between night shifts. Compared to day workers, greater chronic insomnia odd ratios (OR) were observed among participants with >90% night shifts (OR 1.26, 95% CI 1.01-1.58), rotating shifts (OR 1.42, 95% CI 1.22-1.65) and with longer night work history (OR≥16 years 1.28, 95% CI 1.08-1.52). These variables and longer weekly working hours [OR 6.10h increase 1.13, 95% CI 1.04-1.22) were associated with short sleep. CONCLUSIONS: Data-driven analyses of work schedules may provide a more comprehensive evaluation of the impact of real-world shift work patterns on sleep.