Heterogeneous effects of artificial light at night on sleep and mental health: 2SLS augmented geospatial data modeling and geo-correlation analysis

Abstract

Rapid urbanization has led to a significant increase in artificial light at night (ALAN), raising concerns about its potential adverse effects on human health. Yet, evidence on the relationship between ALAN, sleep, and mental health remains limited. In this study, we utilized the Extended Time-Series of Global NPP-VIIRS-like Nighttime Light Data and population distribution maps to estimate population-weighted ALAN at the census tract level across 500 major U.S. cities from 2013 to 2019. We also obtained the prevalence of frequent mental distress and short sleep duration from the SDOH database. A Two-Stage Least Squares (2SLS) model was applied to investigate the causal relationships between ALAN, sleep, and mental health, as well as heterogeneity across differing educational and urbanization levels. The findings revealed that: (1) Population-weighted ALAN was significantly associated with both sleep and mental health, with a ten-fold increase in O-ALAN corresponding to an 8.05 % (±0.04 %) rise in mental distress prevalence and a 4.99 % (±0.07 %) increase in short sleep duration prevalence; (2) higher education levels intensified the negative impact of ALAN on mental health but mitigated its effect on sleep; and (3) higher urbanization levels amplified ALAN's adverse effects on both sleep and mental health. This study is the first to examine the relationship between population-weighted ALAN and sleep and mental health outcomes while accounting for economic endogeneity, offering a comprehensive view of ALAN's impact on health.