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

Abstract

Listen

Translate article

Introduction: Breast cancer is the most common cancer type, and the most common reason for cancer-related deaths among women worldwide. Animal and epidemiologic studies suggest that exposure to light at night may contribute to increased breast cancer risk through circadian disruption, which involves a decrease in the nocturnal secretion of melatonin followed by disturbed estrogen regulation, however, epidemiological evidence is limited. In this study, we examined the association between exposure to residential outdoor light at night (LAN) and breast cancer incidence in the Danish Nurse Cohort Study. Methods: We followed 20,404 Danish nurses from 1993 or 1999 through 2012. LAN exposure was estimated using the US Defense Meteorological Satellite Program's (DMSP) Operational Linescan System. We collected LAN data (nW/cm2/sr) for 1996, 1999, 2000, 2002, 2004, 2005, and 2010, assigned at the subject's residence during the follow-up. Data on the incidence of breast cancer was obtained from the Danish Cancer Registry. We used time-varying Cox regression models to estimate hazard ratios (HRs) and 95% confidence intervals (CIs), adjusting for reproductive, lifestyle and environmental factors such as residential air pollution, road traffic noise, and greenspace. Results: Of the 20,490 women, 1,081 developed breast cancer in total during 339,633 person-years of follow-up. Based on the fully adjusted model of LAN exposure divided into quintiles, where the lowest LAN quintile was reference (0-13.6 nW/cm2/sr), we found an increase in breast cancer incidence at higher LAN levels: HR: 1.01 (95% CI: 0.83-1.23, p = 0.92), 1.08 (95% CI: 0.88-1.33, p = 0.44), 1.04 (0.83-1.30, p = 0.74), and 0.87 (95% CI: 0.66-1.15, p = 0.32) for the 2nd, 3rd, 4th, and 5th quintiles, respectively. Conclusion: There was a tendency, that higher levels of exposure to LAN were associated with a higher risk of breast cancer although the statistical significance was not observed.