P-2055. County-Level Social Vulnerability Index and Infectious Disease Mortality Among Adults in the United States

TL;DR

Counties with higher social vulnerability in the US have higher infectious disease mortality rates, suggesting public health efforts should focus on vulnerable areas.

Contribution

This study demonstrates a strong association between county-level social vulnerability and increased infectious disease mortality rates in the US.

Findings

ID-related mortality rates increased stepwise with higher SVI quartiles, with a 2.8-fold increase in Q4 compared to Q1.

Higher SVI was linked to excess ID mortality across all racial/ethnic groups and geographic regions.

When ID was the underlying cause of death, mortality rates were 2.1 times higher in the most vulnerable counties.

Abstract

Social vulnerability index (SVI) measures social determinants of health in United States (US) counties and has been used to assess the impact of social vulnerability on health outcomes including COVID-19 mortality. It is unknown if county-level SVI negatively impacts mortality from all infectious diseases (ID). We assessed the relationship between the SVI and ID-related age adjusted mortality rates (ID-AAMR) across US counties from 2018 to 2023.

We conducted a retrospective cross-sectional analysis of the multiple cause of death dataset from the CDC WONDER database. The ID-AAMR for all US residents aged ≥20 years who had any ID (ICD-10 codes A00-B99) listed on the death certificate were included. This was linked to the corresponding 2022 SVI score (range: 0 to 1; higher scores indicate greater vulnerability). The dataset was divided into quartiles based on SVI scores (0-0.25: 1st quartile (SVI-Q1), least vulnerable, and 0.75-1 as SVI-Q4, most vulnerable). Counties with death counts < 20 were excluded. The ID-AAMR per 100,000 population and 95% confidence intervals (CI) were estimated for the overall population and stratified by race and census region. The exposure was SVI quartile. The outcome was the rate difference in ID-AAMR between SVI-Q4 and SVI-Q1 and was significant if the CIs did not overlap. We used negative binomial regression to estimate rate ratios (RRs) with SVI-Q1 as the reference. The RR ratios were significant if the CI excluded 1. The analysis was repeated with ID as the underlying cause of death.

Among 2,951 counties, the overall ID-AAMR per 100,000 was 136.5 (CI: 134.8–138.5). There was a stepwise increase in the overall ID-AAMR from 110.8 in SVI-Q1 to 163.4 in Q4 (RR 2.8, CI: 2.6 – 3.2), yielding an excess of 52.6 in SVI-Q4 (Figure 1). Similarly, SVI-Q4 was associated with excess ID-AAMR across all racial/ethnic groups and all geographic regions studied (Table 1) compared to SVI-Q1. When limited to ID as the underlying cause of death, the ID-AAMR increased significantly from 25.9 (CI: 25.1 – 26.8) in SVI-Q1 to 38.5 (CI: 37.5 – 39.5) in SVI-Q4 (RR 2.1; CI: 1.8 – 2.3).

Counties with higher SVI were associated with increased ID-related mortality. Targeting public health efforts towards the most vulnerable could mitigate poor outcomes.

All Authors: No reported disclosures

Figures