Spatial Regression Analysis of Latitude and Mortality Rates of Multiple Autoimmune Diseases in the United States

Keywords: Autoimmune, spatial regression, latitude, environment
Abstract Type: Poster Abstract

Abstract

Previous research indicates that autoimmune conditions like celiac disease and rheumatoid arthritis are associated with higher latitudes. Our study aimed to determine if latitude-related environmental or climate factors influence mortality rates in individuals with autoimmune conditions. Mortality data were sourced from CDC Wonder for multiple sclerosis (MS), lupus, type 1 diabetes mellitus (DM), and type 2 DM (control). We conducted exploratory stepwise spatial linear regression comparing state latitude to death rates from the three autoimmune diseases, and type 2 DM.The regression included the covariates of smoking rates by state, state BMI rates, average sun exposure, and percent below poverty by state. Univariate analysis had the most consistently low AIC, thus we focus on the univariate results herein. Across the four models, latitude was significantly associated with all three autoimmune conditions (lupus: β= -0.01, p=0.0006; type 1 DM: β= 0.04, p<0.0001; MS: β= 0.04, p<0.0001) and not significant for type 2 DM (β=0.06, p=0.52). Thus, latitude was positively associated with increased mortality rates of type 1 DM and MS, and negatively associated with lupus mortality rates. The negative association with lupus was in contradiction with our initial hypothesis, while the positive association for type 1 DM and MS were as we hypothesized. Future research should investigate more possible contributing factors at a smaller spatial scale to determine if our results are consistent across spatial resolutions.