Here, we propose an “Environmental Risk Score (ERS)” as a new tool to examine the risk of exposure to multi-pollutants in epidemiologic research. As a “proof of concept”, we used environmental biomonitoring data from NHANES to illustrate our methodology because it includes a wide range of environmental pollutants from representative U.S. populations and independent data from different cycles enabled us to discover and validate our findings. As outcomes, we examined serum lipid levels including total cholesterol, high-density lipoprotein cholesterol (HDL), low-density lipoprotein cholesterol (LDL) and triglycerides, because these are continuous measures that can be dichotomized at clinically relevant cutoff points, allowing us to evaluate both continuous and binary outcomes. These outcomes were used in the previous EWAS by Patel et al. [6]. We focused on environmental pollutants in this study rather than a broader array of environmental exposures including dietary, behavioral, psychosocial, socioeconomic and neighborhood, and microorganismic factors, which may limit the feasibility and applicability of ERS. Instead, we treated important determinants of lipid outcomes such as age, sex, race/ethnicity, education (an indicator of socioeconomic factor), body mass index (BMI),
