Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Socioeconomic Drivers of Physical Morbidity Across U.S. Counties: A Spatial Causal Inference Approach

Ranadeep Daw, Hunter Evans, Indrabati Bhattacharya

Abstract

Identifying the causal effects of socioeconomic determinants on population health is of many great interests - from statistical methodology development to public health practitioners and policy developments. The statistical side of the problem needs to address several questions: spatial autocorrelation in both exposures and outcomes, confounding between treatments and covariates, and the need for geographically logical inference. We address these jointly by using spectral basis functions - Moran Eigenvector Maps and ICAR precision matrix eigenvectors - within a doubly robust generalized propensity score estimator for continuous treatments. Applied to 2022 county health data across the U.S. counties, the framework identifies the effect of six chosen predictors on the average physically unhealthy days per month. Possible further applications and methodological extensions are also discussed as future directions from this research.

Socioeconomic Drivers of Physical Morbidity Across U.S. Counties: A Spatial Causal Inference Approach

Abstract

Identifying the causal effects of socioeconomic determinants on population health is of many great interests - from statistical methodology development to public health practitioners and policy developments. The statistical side of the problem needs to address several questions: spatial autocorrelation in both exposures and outcomes, confounding between treatments and covariates, and the need for geographically logical inference. We address these jointly by using spectral basis functions - Moran Eigenvector Maps and ICAR precision matrix eigenvectors - within a doubly robust generalized propensity score estimator for continuous treatments. Applied to 2022 county health data across the U.S. counties, the framework identifies the effect of six chosen predictors on the average physically unhealthy days per month. Possible further applications and methodological extensions are also discussed as future directions from this research.

Paper Structure

This paper contains 10 sections, 3 equations, 2 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Data and Literature Review
  3. Methodology
  4. Data Structure and Assumptions
  5. Spatial Basis Construction
  6. Moran Eigenvector Maps (MEM):
  7. ICAR Basis:
  8. Doubly Robust Causal Estimation
  9. Results
  10. Discussion

Figures (2)

  • Figure 1: Geographic distribution of average physically unhealthy days per month across U.S. counties (2022 County Health Rankings, derived from BRFSS). Darker shading indicates higher morbidity burden. Pronounced clustering in Appalachia, the Deep South, and parts of the rural Midwest motivates the spatial confounding adjustment central to our methodology.
  • Figure 2: Geographic distribution of the six socioeconomic and environmental treatments across U.S. counties. From left to right, top to bottom: unemployment rate, food insecurity, rent burden (American Community Survey, Table B25070), PM 2.5 concentration (EPA via CHR&R), pay gap, and homeownership rate. The visible spatial clustering across all six treatments illustrates why spatial confounding control is necessary prior to causal estimation.