Association between accelerometer-measured irregular sleep duration and type 2 diabetes risk: a prospective cohort study in the UK Biobank

Objective: To evaluate the association between irregular sleep duration and incident diabetes in a UK population over 7 years of follow-up.

Research Design and Methods: Among 84,421 UK Biobank participants (mean age: 62 years) who were free of diabetes at the time of providing accelerometer data in 2013-2015 and prospectively followed until May 2022, sleep duration variability was quantified by the within-person standard deviation (SD) of 7-night accelerometer-measured sleep duration. We used Cox proportional hazard models to estimate hazard ratios (HRs) for incident diabetes (identified from medical records, death register, and/or self-reported diagnosis) according to categories of sleep duration SD.

Results: There were 2,058 incident diabetes cases over 622,080 person-years of follow-up. Compared with sleep duration SD ≤30 minutes, the HR (95% CI) was 1.15 (0.99, 1.33) for 31-45 minutes, 1.28 (1.10, 1.48) for 46-60 minutes, 1.54 (1.32, 1.80) for 61-90 minutes, and 1.59 (1.33, 1.90) for ≥91 minutes, after adjusting for age, sex, and race. We found a nonlinear relationship (p-nonlinearity: 0.0002), with individuals with a sleep duration SD >60 versus ≤60 minutes having 34% higher diabetes risk (95% CI: 1.22, 1.47). Further adjustment for lifestyle, co-morbidities, environmental factors, and adiposity attenuated the association (HR comparing sleep duration SD >60 vs. ≤60 minutes: 1.11; 95% CI: 1.01, 1.22). The association was stronger among individuals with lower diabetes polygenic risk score (PRS; p-interaction≤0.0264) and longer sleep duration (p-interaction≤0.0009).

Conclusion: Irregular sleep duration was associated with higher diabetes risk, particularly in individuals with a lower diabetes PRS and longer sleep duration.