Sleep duration and progression to diabetes in people with prediabetes defined by HbA1c concentration

Abstract

Aims: To evaluate the association between sleep duration and the risk of progression to diabetes among people with prediabetes, defined by HbA1c values. Methods: We conducted a cohort study in 17 983 adults who underwent health check-up examinations, including assessments of sleep duration and quality. Diabetes was defined as either HbA1c ≥48 mmol/mol (6.5%), or the use of antidiabetic medication. Time-dependent proportional hazards models were used to evaluate the association between sleep duration and the risk of progression to diabetes. Results: During 31,582 person-years of follow-up, 664 incident cases of diabetes were identified; the incidence rate was 21.0 per 1000 person-years. The multivariate adjusted hazard ratios for progression to diabetes in people with sleep durations of ≤5, 6 and ≥8 h compared with 7 h were 1.68 (95% CI 1.30–2.16), 1.44 (95% CI 1.17–1.76) and 1.23 (95% CI 0.85–1.78), respectively (P for quadratic trend <0.001). This association was partially mediated by biomarkers of adiposity, fatty liver and insulin resistance. Conclusion: In this large study in young and middle-aged adults with prediabetes, we found an association between short sleep duration and the risk of progression to diabetes. Our findings suggest that sufficient sleep duration is important for delaying or preventing the progression of prediabetes to diabetes.