Association Between Social Vulnerability Index Theme Percentile Ranks and the Number of Respiratory Viruses Detected in People Who Underwent Multiplex Testing in Nashville, Tennessee, 2018–2022

Abstract

Abstract Background The CDC/ATSDR Social Vulnerability Index (SVI)—an aggregate of 16 variables grouped under four themes—measures the relative social vulnerability of each census tract in the United States. We hypothesized that social vulnerability may be a determinant of health in the context of viral co-detection because community attributes are associated with viral circulation. Therefore, we aimed to assess the association between SVI theme percentile ranks and the number of viruses detected in people who underwent respiratory viral panel testing in Nashville, Tennessee. Methods We systematically retrieved the records of all people who underwent respiratory viral panel testing at Vanderbilt University Medical Center between 01/01/2018 and 12/01/2022. Anyone who had a medical encounter >14 days after a preceding encounter was considered a new, distinct case. We supplemented the data with SARS-CoV-2 test results by any method (antigen-based, molecular, or serologic). We used the United States Census Bureau’s census geocoder to retrieve the census tract by city and state, which we subsequently used to retrieve data from the CDC/ATSDR SVI 2020 database, including the percentile ranks for the four themes (Socioeconomic Status, Household Characteristics, Racial & Ethnic Minority Status, and Housing Type & Transportation). We described the distribution of each theme percentile rank using the mean and standard deviation and used proportional odds models to assess the association between each SVI theme percentile rank (higher values indicate higher vulnerability) and the number of respiratory viruses detected. The odds ratio (OR) for each theme represents groups differing in their SVI theme percentile rank by 10%. Results Of 57,930 people who underwent multiplex testing, we retrieved data from the SVI database for 54,356 (93.8%), of whom 28,369 (52.2%) were male. The median age at presentation was 15.7 years (IQR, 1.7–55.5 years), and the numbers of people in the 0–4, 5–17, 18–64, ≥65 age groups were 20,247 (37.2%), 8,245 (15.2%), 17,566 (32.3%), and 8,298 (15.3%), respectively. The mean percentile ranks for the themes of Socioeconomic Status, Household Characteristics, Racial & Ethnic Minority Status, and Housing Type & Transportation were 0.55 (0.28), 0.51 (0.28), 0.41 (0.25), and 0.47 (0.29), respectively. No viruses were detected in 30,201 people (55.6%), while one virus was detected in 20,325 (37.4%), two in 3,400 (6.3%), three in 408 (0.8%), and four in 22 (0.04%). The number of viruses detected was significantly associated with Socioeconomic Status (OR=1.017 [95% CI, 1.011–1.023]; p<0.001), Household Characteristics (OR=1.009 [95% CI, 1.003–1.015]; p=0.003), and Racial & Ethnic Minority Status (OR=1.051 [95% CI, 1.044–1.058]; p<0.001), although not Housing Type & Transportation (OR=0.999 [95% CI, 0.993–1.005]; p=0.74). Figure 1 shows the mean percentile ranks for the four themes stratified by the number of viruses detected. Figure 1. Mean percentile ranks for the four themes that comprise the CDC/ATSDR Social Vulnerability Index, stratified by the number of viruses detected, for the addresses of 54,356 people who underwent respiratory viral panel testing in Nashville, Tennessee. Error bars indicate 95% confidence intervals. Conclusion Various aspects of social vulnerability were associated with the number of respiratory viruses detected in people of all ages who underwent multiplex testing. The strongest association was observed for racial and ethnic minority status, more so than other well-established social determinants of health, including socioeconomic status. Future research will focus on racial and ethnic disparities that inform this finding.