Shift work and the onset of type 2 diabetes: results from a large-scale cohort among Japanese workers

Abstract

Data are limited regarding how shift work is linked to the development of type 2 diabetes, especially among workers at high risk of diabetes. We examined the risk of diabetes according to shift-work conditions over several years among Japanese adults. This prospective study enrolled 17,515 workers (age 40–78 years). Shift work was self-reported at annual health examinations over time from 2004 to 2017 and categorized as shift workers or non-shift workers. Diabetes was defined as fasting plasma glucose, random glucose, HbA1c, and self-reported use of antidiabetic medications. The association of shift work and diabetes was quantified using Cox regression. During a follow-up of 8.1 years in median, 2071 incident cases of diabetes were documented. Compared with non-shift work, shift work showed a significantly elevated hazard ratios of developing diabetes. Shift work showed a 19% (95% confidence intervals: 3–37%) higher hazard ratios for diabetes, after adjustment for demographic-, cardiometabolic-, and work-related factors. Further adjustment for lifestyle factors and body mass index did not materially change this association (a 16% increase; 95% confidence intervals, 1–34%). This relationship was replicated among workers with prediabetes. Engaging in shift work may increase the risk of developing diabetes independently of lifestyle factors and body mass index, even among prediabetic workers.