Long-term exposure to outdoor artificial light-at-night and its associations with natural-cause and cause-specific mortality: a register-based cohort study in the Netherlands.

BACKGROUND

Outdoor artificial light-at-night (ALAN) is increasingly recognized as a potential environmental health risk. Epidemiological evidence mainly comes from incidence studies, whereas cause-specific mortality, reflecting disease occurrence and survival, has rarely been examined.Estimated associations may depend on the scale of ALAN exposure assessment, including grid-level and buffer-based measures.

OBJECTIVES

We aimed to (1) assess the associations between long-term ALAN exposure and natural-cause and cause-specific mortality; (2) evaluate the sensitivity of these associations across various spatial scales of the ALAN assessment; and (3) examine whether some socio-demographic population groups are more vulnerable.

METHODS

We conducted a register-based cohort study of 6,325,139 adults in the Netherlands (2013-2023). We linked satellite-derived average ALAN exposure over 2013-2023 to residential addresses using 50-m grid values and 300-1,000 m buffers. Mortality included natural causes, cerebrovascular disease (CeVD), coronary heart disease (CHD), type 2 diabetes (T2D), obesity, breast cancer (BC), and prostate cancer (PCa). Associations were estimated using Cox proportional hazards models adjusted for individual-level characteristics and environmental exposures, with effect modification assessed via interaction terms and stratification.

RESULTS

ALAN exposure was positively associated with natural-cause mortality and with mortality from T2D, obesity, BC, and PCa. The most robust association was for T2D (HR = 1.06, 95% CI: 1.02-1.10) and obesity (HR = 1.07, 95% CI: 1.00-1.15). Associations for BC and PCa were positive only after adjusting for environmental exposures. No associations were found for CHD or CeVD mortality. Estimates were attenuated with larger buffers. Stronger associations were observed among middle-aged adults, those with low and middle levels of education, low-income groups, women, non-Dutch, and non-cohabiting populations.

CONCLUSIONS

Long-term ALAN exposure was associated with metabolic-cause mortality (notably T2D), with stronger associations in some socio-demographic groups, suggesting differential vulnerability. Effect estimates varied with the spatial scale of exposure assessment and with adjustment for other environmental exposures.