Environmental mixtures and body mass index in two prospective US-based cohorts of female nurses

We estimated the joint effect of particulate matter <= 2.5 mu m in diameter (PM2.5), nitrogen dioxide (NO2), seasonal temperature, noise, greenness, light at night, and neighborhood socioeconomic status (NSES) on body mass index (BMI) in a mixture context among 194,966 participants from the Nurses' Health Study (NHS) and Nurses' Health Study II (NHSII) over 30 years. BMI was calculated from self-reported weight and height. Single- and multi-exposure generalized estimating equations models were used to estimate the difference in BMI per interquartile range (IQR) increase of environmental factors, and quantile g-computation methods were used to estimate joint associations. In both cohorts, we consistently observed positive associations of BMI with PM2.5 and NO2 concentrations as well as negative associations with light at night and NSES regardless modeling approach. A positive association with noise was only observed in NHS. Negative associations with greenness and winter temperature were only observed in NHSII. Overall, the changes in BMI per quintile increase in all eight exposures were -0.11 (-0.13, -0.08) in NHS and -0.39 (-0.41, -0.37) in NHSII, which were largely driven by air pollution and nighttime noise (18-45 %) in the positive direction and NSES (>70 %) in the negative direction. Future intervention on environmental factors, especially reducing PM2.5, NO2 and noise or improving the NSES, might be helpful to lower BMI.