Workplace violence and its association with sleep disturbances in the Swedish working population

Abstract: Aim: The purpose of this study was to investigate associations between workplace violence and sleep disturbances in the Swedish working population, and to estimate potential dose-response relationships. Methods: Data were derived from two waves (2014 and 2016) of the Swedish Longitudinal Occupational Survey of Health (n=9364). Sleep disturbances were measured with four items of the Karolinska Sleep Questionnaire and one question asked whether the participants were exposed to violence or threats of at the workplace in the past six months. Multiple logistic regression analyses were performed. Results: Exposure to workplace violence was associated with higher odds of concurrent sleep disturbances (cross-sectional approach). Workplace violence was also associated with higher odds of subsequent sleep disturbances, but only when covariates were not adjusted for (prospective approach). Sleep disturbances were associated with higher odds of subsequent workplace violence (reverse approach). A dose-response relationship was found in the cross-sectional approach but not in the prospective and reverse approach. Conclusion: The findings suggested that there was a cross-sectional association between workplace violence and sleep disturbances, and that sleep disturbances were prospectively associated with exposure to workplace violence, but the results did not support a statistically significant prospective association between workplace violence and sleep disturbances.