Counties with large Black population, poverty rate have less access to COVID-19 treatments

Key takeaways:

• Access to COVID-19 treatments was reduced in areas that had larger Black populations.
• Increased social vulnerability scores and more poverty in counties also meant low treatment availability.

Access to COVID-19 therapeutic drugs in U.S. counties decreased if the county had an increased population of Black individuals, according to a research letter published in JAMA Network Open.

Reduced access to COVID-19 treatments also occurred in counties with elevated rates of poverty and uninsured individuals, according to researchers.

“With the end of the COVID-19 Public Health Emergency, these results highlight an important gap in treatment access,” Alyssa Shishkov, BS, research fellow at the National Institute on Minority Health and Health Disparities, and colleagues wrote.

In a cross-sectional study, Shishkov and colleagues evaluated 95,294 courses of COVID-19 therapeutic drugs (mean, 30.7) found through the COVID-19 Public Therapeutic Locator in May 2023 to determine the prevalence of sociodemographic-based disparities in 3,108 U.S. counties.

Per 100,000 people, counties within New England and Kansas, as well across the U.S., had elevated COVID-19 treatment availability (range, 39.08-854.7).

Researchers found evidence of spatial clustering in their sample using global Moran I, and through Anselin local Moran I, they connected counties to one of the five types of clusters: high-high (counties with higher availability), low-low (counties with lower availability), high-low outliers (counties with high levels of availability surrounded by areas with low availability), low-high outliers (counties with low availability surrounded by areas with high availability) and unclustered, which included counties that did not have high or low availability.

Maine, western Kansas, western Nebraska and eastern Montana had high availability of COVID-19 therapeutic drugs, whereas counties within the South, Midwest and Western regions had low availability.

A smaller proportion of Black individuals belonged to high-high counties than unclustered counties (1.92% vs. 7.73%; P < .001), according to researchers.

Compared with unclustered areas, researchers found that counties with high availability had a lower incidence of COVID-19 (crude rate, 27,672.8 vs. 28,910.2 per 100,000 people; P < .05), less poverty (rate, 13.39 vs. 15.14; P < .001), a social vulnerability index (SVI) score closer to 0, which signals less vulnerability (0.36 vs. 0.48; P < .001) and a COVID-19 Community Vulnerability Index (CCVI) score also demonstrating less vulnerability (0.29 vs. 0.48; P < .001).

Notably, household income was lower in the high-high vs. unclustered counties ($53,927.40 vs. $58,842.30; P < .05).

On the other hand, researchers observed that low-low clusters had a higher proportion of Black individuals than unclustered areas (15.15% vs. 7.73%; P < .001).

For areas with low therapeutic drug availability, community characteristics tended to be worse than those observed in unclustered counties, including more poverty (rate, 17.43 vs. 15.14), more uninsured individuals (rate, 11.14 vs. 9.48), greater social vulnerability (SVI score, 0.59 vs. 0.48), lower household income ($54,502.40 vs. $58,842.30) and greater COVID-19 community vulnerability (CCVI score, 0.61 vs. 0.48; P < .001 for all).

Characteristics of high-low outlier counties tended to be similar to those observed for low-low counties. According to researchers, these counties vs. unclustered counties had a higher proportion of Black individuals (14.14% vs. 7.73%; P < .001), greater COVID-19 incidence (crude rate, 31,226.2 vs. 28,910.2 per 100,000 people; P < .001), more poverty (rate, 17.91 vs. 15.14; P < .001), more uninsured individuals (rate, 10.37 vs. 9.48; P < .01), greater social vulnerability (SVI score, 0.59 vs. 0.48; P < .001) and greater COVID-19 community vulnerability (CCVI score, 0.64 vs. 0.48; P < .001).

Household income of high-low outliers was lower than unclustered counties ($53,233.60 vs. $58,842.30; P < .001).

Lastly, low-high outlier counties had a lower proportion of Black individuals than unclustered counties (2.86% vs. 7.73%) but a higher proportion of Hispanic or Latino individuals (22.35% vs. 9.93%; P < .001 for both).

In terms of community characteristics, individuals from low-high outlier counties had a lower COVID-19 community vulnerability (CCVI score, 0.41 vs. 0.48; P < .01) than individuals from unclustered areas.

On the other hand, low-high outlier counties had a higher uninsured rate (12.84 vs. 9.48) and a lower household income ($55,452.70 vs. $58,842.30; P < .01) compared with unclustered counties.

“A study limitation was that size of area-level data could have differential results,” Shishkov and colleagues wrote. “Finer geographic scale, such as Census tracts, should be used for future investigations.”