Indoor and outdoor light at night: Associations with sleep and breast cancer

PROJECT SUMMARY Applicant. The goal of this F31 fellowship application is to support and promote the training of Marina Sweeney, who is currently a pre-doctoral student in the Department of Epidemiology at the University of North Carolina, Chapel Hill. Her long-term research interests focus on elucidating the role of the environment on cancer risk. The applicant will be mentored by Drs. Hazel Nichols and Alexandra White during the F31 training period. Significance. Breast cancer is the most common malignancy diagnosed in women in the US, and known risk factors are estimated to explain only half of incident cases. Risk of breast cancer may be increased by poor quality or insufficient sleep through mechanisms of inflammation and obesity. Outdoor and indoor light at night (LAN) are proposed to disrupt circadian rhythms and thereby negatively impact sleep and possibly increase the risk of breast cancer. Prior studies suggest that outdoor LAN increases breast cancer risk, while studies of indoor LAN have been mixed. Outdoor and indoor LAN have been shown to disrupt sleep, but few large-scale epidemiologic studies have been conducted. Assessments of outdoor or indoor LAN individually may misclassify LAN exposure, and no studies have assessed combined outdoor and indoor LAN exposure in relation to either breast cancer or sleep characteristics. The overall goal of the proposed study is to further our knowledge of the role of LAN in breast carcinogenesis and sleep disruption. Aim 1. Examine the association between outdoor, indoor, and combined LAN and risk of incident breast cancer. Aim 2. Examine the association between outdoor, indoor, and combined LAN and sleep duration and quality. Approach. The Sister Study prospective cohort includes 50,884 women who had a sister with breast cancer but who themselves were breast-cancer free at baseline. Outdoor LAN will be determined from satellite data and linked to baseline geocoded addresses. Indoor LAN was self-reported at baseline. Aim 1 will use Cox proportional hazards models to calculate hazard ratios and 95% confidence intervals for the associations between indoor and outdoor LAN and breast cancer risk. Aim 2 will use weighted log binomial models to calculate prevalence ratios and 95% confidence intervals for associations between LAN and characteristics of sleep duration and quality. Innovation. The proposed project capitalizes on existing data within the Sister Study and will incorporate existing and new geospatial variables (outdoor LAN, neighborhood SES index, area deprivation index, fine particulate matter, green space, noise) in a prospective cohort of >50,000 US women. This will be the first study to assess joint outdoor and indoor LAN in relation to either breast cancer or sleep characteristics. Impact. The proposed study will provide the applicant with further training in epidemiologic methods, including the integration of both geographic and individual exposure data, and provide high-quality mentorship and numerous professional development opportunities.