Outdoor Light at Night and Diabetes Incidence in the Danish Nurse Cohort Study

Abstract

Background/Aim: Recent animal studies suggested that light-at-night (LAN) may impair glucose tolerance in rats and disturb glucose homeostasis. We aimed to evaluate the association between exposure to residential outdoor LAN and diabetes incidence in Denmark. Methods: In the nationwide Danish Nurse Cohort consisting of 23,797 nurses, recruited in 1993 or 1999, we identified 1,065 diabetes incident cases through National Diabetes Register up to 2012. The LAN data was estimated using Global Radiance Calibrated products of the Defense Meteorological Satellite Program (DMSP) for 1996, 1999, 2000, 2002, 2004, 2005, and 2010. The full residential address histories were obtained and the LAN exposure was assigned based on the 1996 data for addresses before 1997, and based on the most recent measure for addresses after 1997. Time-varying Cox regression models were used to estimate hazard ratios (HR) (95% confidence intervals (CI)) for the associations of LAN and diabetes incidence, adjusting for individual anthropometric, lifestyle and environmental covariates, such as residential air pollution (PM2.5 and NO2), and night-time road traffic noise determined by validated models. Results: The estimated HR based on a fully adjusted model for incident diabetes with an interquartile range (IQR) (75.6 nW/cm2/sr) increase in cumulative average outdoor LAN was 1.13 (95% CI: 1.05, 1.21). Based on the fully adjusted model of LAN exposure divided into quintiles, where LAN quintile 1 was reference, we found an increase in diabetes incidence at higher LAN levels: LAN quintile 2: HR: 1.04 (95% CI: 0.86-1.27, p = 0.66), LAN quintile 3: HR: 1.04 (95% CI: 0.85-1.26, p = 0.71), LAN quintile 4: HR: 0.99 (0.81-1.20, p = 0.91), and LAN quintile 5: HR: 1.21 (95% CI: 1.0-1.46, p = 0.053). Conclusion: We found suggestive evidence that outdoor LAN exposure may be associated with diabetes incidence, independent of air pollution and road traffic noise.