Shift work, sleep quality, and hypertension among working adults in Ontario
Background: The epidemiological evidence supporting an association between shift work, and cardiovascular diseases and hypertension is growing. However, few population-based studies have explored the relationship between shift work and hypertension risk and the potential mechanism(s) linking them. Objectives: 1) To determine the association between shift work status and rates of hypertension in a cohort of Ontario working adults during a 12-year follow-up period; 2) To determine whether sleep quality mediates this relationship. Methods: The cohort included Ontario adult respondents of the 2000-01 Canadian Community Health Survey (CCHS), which was linked to health administrative databases within the Institute for Clinical Evaluative Sciences (ICES). Baseline shift work status, sleep quality, and covariate measures were obtained from the CCHS; time of hypertension diagnosis was ascertained from the ICES databases. The inception cohort included individuals who reported working with no pre-existing hypertension. Follow-up was 12 years. We conducted a mediation analysis using the natural effects models that use marginal structural modeling to directly parameterize the natural direct and indirect effects of shift work on the rates of hypertension in a Cox proportional hazard model. Results: Shift workers reported lower frequency of refreshing sleep, more trouble sleeping, and poorer sleep quality overall. Night shift work was not associated with increased hypertension rates in either men (HR=1.15, 95% CI:0. 93-1.43) or women (HR=1.23, 95% CI: 0.96-1.57) during 12 years of follow-up, but the hazard ratios were in the hypothesized direction. Other shift work (including evening shifts, split shifts, irregular schedules, on-call, or other schedules) was associated with increased hypertension rates in men (HR=1.29, 95% CI: 1.03-1.62) during 12 years of follow-up, but this association was not mediated through sleep quality (HR=1.00, 95% CI: 0.99-1.01). Similar results were found among women (HR=1.30, 95% CI: 0.97-1.74) during 12 years of follow-up. Conclusions: Irregular working hours appear to increase hypertension rates. Shift work does disrupt sleep, but sleep quality does not appear to be a mediator between shift work and increased hypertension rates.