ABSTRACT
Several population-level studies explored the association between breast cancer (BC) incidence and artificial light-at-night (ALAN), and found higher BC rates in more lit areas. Most of these studies used ALAN satellite data, available from the United States Defence Meteorological Satellite Program (US-DMSP), while, in recent years, higher-resolution ALAN data sources, such as Visible Infrared Imaging Radiometer Suite Day-Night Band (VIIRS-DNB), have become available. The present study aims to determine whether the use of different ALAN data sources may affect the BC–ALAN association. As the test case, we use data on BC incidence rates in women residing in the Greater Haifa Metropolitan Area (GHMA; Israel), matching them with US-DMSP and VIIRS-DNB data on ALAN intensities, and controlling for several potential confounders, including age, fertility, and socio-economic status (SES). Both ordinary least squares (OLS) and spatial dependency models were used in the analysis. ALAN emerged as a stronger predictor of BC rates in models based on better-resolution VIIRS-DNB estimates (t > 6.035; p < 0.01) than in models based on coarser US-DMSP data (t < 4.196; p < 0.01).
