Vetter · JAMA 2016 · prospective cohort study · n=189,158

Association Between Rotating Night Shift Work and Risk of Coronary Heart Disease Among Women.

Cited 528 times in the scientific literature.

Level 3 - non-randomized controlled study

Prospective longitudinal cohort study

PubMed 27115377 · doi:10.1001/jama.2016.4454 · record verified 2026-08-30

What was done

A prospective cohort analysis was conducted using data from 189,158 initially healthy female registered nurses across two cohorts: Nurses' Health Study (NHS, n = 73,623, 1988-2012) and Nurses' Health Study 2 (NHS2, n = 115,535, 1989-2013), followed for up to 24 years. Exposure was lifetime history of rotating night shift work (at least 3 night shifts/month in addition to day and evening shifts) assessed at baseline and updated every 2 to 4 years in NHS2. The primary outcome was incident coronary heart disease (CHD), comprising nonfatal myocardial infarction, CHD death, angiogram-confirmed angina pectoris, coronary artery bypass graft surgery, stents, and angioplasty, analyzed via multivariable-adjusted Cox proportional hazards models.

What was found

During 24 years of follow-up, 7,303 incident CHD cases occurred in NHS (mean baseline age 54.5 years) and 3,519 cases in NHS2 (mean baseline age 34.8 years). Increasing years of rotating night shift work was significantly associated with higher CHD risk in both cohorts (numerical hazard ratios and confidence intervals not reported in the abstract). In NHS, the association was stronger in the first half than the second half of follow-up (P = .02 for interaction). In NHS2, longer time since quitting shift work showed a significant decrease in CHD risk among ever shift workers (P < .001 for trend).

Why it matters

This study provides long-term prospective evidence linking rotating night shift work duration to elevated CHD risk, while suggesting cardiovascular risk attenuates after cessation of night shift work.

Limits

The cohort was restricted entirely to female registered nurses, limiting generalizability to men and other occupational shift schedules. Specific effect estimates, hazard ratios, and confidence intervals are omitted from the abstract text.

Cited by