Reply to Associations between light at night and risk of thyroid cancer

Abstract

We thank Xiao et al for their interest in our study and the relevant points that they raise in their letter. We agree that because of the potential role of circadian disruption in thyroid cancer etiology, nightshift work could influence the risk of thyroid cancer. Because nightshift work increases exposure to light at night (LAN), it may have an impact on the association between LAN and thyroid cancer. However, the NIHAARP Diet and Health Study did not collect information on nightshift work; therefore, we were not able to consider this as a potential confounder in our analysis. Moreover, because we used satellitebased estimation of LAN as opposed to individuallevel exposure measurements, uncontrolled confounding by nightshift work is unlikely to explain our findings. We believe that it is important for future studies to use personallevel measures to more accurately assess LAN exposure levels, and nightshift work would be an important potential confounding factor to consider in such studies. Further complicating the issue, however, is that certain nightshift workers have greater access to health care and require routine physical examinations, and this may lead to more common incidental detection and overdiagnosis of thyroid cancer. This may account for some, or all, of the elevated incidence of thyroid cancer observed in medical workers and firefighters, for example. 4 We also agree that future studies should examine the relationship between LAN and thyroid cancer risk in different age groups. By focusing on middleaged to older adults, our study may not be generalizable to younger individuals. As Xiao et al point out, younger women have different levels of exposure to sex steroid hormones, and they also differ in other physiological factors and lifestyle characteristics in comparison with the older women in our cohort. Therefore, it is reasonable to hypothesize that both age and sex may modify the relationship between LAN and thyroid cancer risk.