Differences in State Reporting of COVID-19 Data by Race and Ethnicity Across the United States

Gracyn Travitz¹, Melissa White², Kimberly S. Johnson³, Julius Wilder⁴

¹Beaver College of Health Sciences, Public Health, Appalachian State University, Boone, NC, USA
²Trinity College, Duke University, Durham, NC, USA
³Division of Geriatrics, Department of Medicine, Duke University School of Medicine and Geriatric Research Education and Clinical Center, Durham Veterans Affairs Medical Center, Durham, NC, USA
⁴Division of Gastroenterology, Department of Medicine, Duke University School of Medicine and Duke Clinical Research Institute, Durham, NC, USA

Correspondence: Julius Wilder, Division of Gastroenterology, Department of Medicine, Duke University School of Medicine and Duke Clinical Research Institute, Durham, NC, USA.

Introduction

African American, Latino, and Indigenous populations are disproportionately affected by COVID-19 [1]. Accurate reporting of COVID-19 data by race and ethnicity across states is essential to understand variation in disease burden and inform testing and vaccine strategies that prioritize populations at greatest risk [2].

Methods

Between February 8-15, 2021, state health department websites and COVID-19 dashboards were searched to determine reporting of COVID-19 cases, deaths, and vaccinations by race and ethnicity using US Census designations (five race categories: White, African American (AA)/Black, Asian, American Indian and Alaska Native (AI/AN), Native Hawaiian or Other Pacific Islander (NHOPI) and two ethnicity categories: Hispanic, Non- Hispanic as defined by the federal Office of Management and Budget) [3].

Results

Ninety-five percent (47/50) of states reported COVID-19 cases, and 88% (44/50) reported deaths using race categories, even if combined with ethnicity categories (Figure 1). A slightly smaller proportion of states reported COVID- 19 cases (44/50, 88%) and deaths (43/50, 86%) by ethnicity, even if combined with racial categories (Figure 2). Eighty-eight percent (44/50) of states reported cases and, s88% (43/50) reported deaths by both race and ethnicity. Race and ethnicity were determined as two separate categories; thus, there was no impact on the results if any state-reported race and ethnicity together. Of the states that reported COVID-19 data by both race and ethnicity, only 43% (19/44) of those reporting cases and 42% (18/43) of those reporting deaths included all of the five races and two ethnicity designations of the US Census (Figure 3). NHOPI was the most commonly excluded group; only 40% (20/50) of states listed NHOPI in case data and 38% (19/50) in death data. Eleven states combined Asian and NHOPI into one category. In comparison, 94% (47/50) of states included White and Black/AA groups in case data and 88% (44/50) in death data. A smaller proportion of states included AI/AN and Asians in case (34/50, 68% for both groups) and death (62% and 66%respectively) data. Even when reported, race/ethnicity data were often incomplete. Over two-thirds of states reported the number of COVID-19 cases/deaths with missing race data, and over half reported the number with missing ethnicity data.

Ninety-two percent of states (46/50) had a COVID-19 vaccination dashboard (Figure 4), and 72% (33/46) of those included data on administration by race and 67% (31/46) by ethnicity (Figure 5). Race and ethnicity were determined as two separate categories. Of the states that reported vaccine data by race, 67% (22/33) included each of the five-race designations of the US Census (Figure 6). Of the 11 states reporting vaccine data for only some racial groups, all included White and Black/AA groups. However, only 4 of the 11 states included AI/AN, and only 4 included NHOPI. Sixty-six percent (33/50) of states reported the number of vaccines with missing race and ethnicity data.

Discussion

Most states reported COVID-19 case and death data by race and/or ethnicity. However, less than half included all census categories. Approximately 30% did not report vaccine data by race and ethnicity, and one-third of states that provided these data did not use all Census categories.

There was variation in reporting of COVID-19 data by racial groups, with NHOPI frequently excluded. The absence of data reporting for some racial and ethnic groups and incomplete or missing data for others hinder efforts to identify variation in disease burden across populations. Our findings do not account for collecting data that are not publicly reported on websites by state officials. We do not know if the absence of reporting of COVID-19 data for some racial groups is due to the small representation of those groups across states or other factors.

Conclusion

This analysis demonstrates heterogeneity in state practices for publicly reporting COVID-19 cases, deaths, and vaccinations by race/ethnicity. Consistent and accurate reporting of COVID-19 data with standardized categories for race and ethnicity is vital for targeted interventions to improve outcomes and facilitate equitable care for minority populations.

Funding

This research was conducted as part of the Duke Center for REACH to Advance Healthcare Equity (REACH Equity) Summer Undergraduate Research Program (RESURP). REACH Equity is supported by the National Institute on Minority Health and Health Disparities (NIMHD) of the National Institutes of Health under Award Number U54MD012530. In addition, RESURP is supported by funding from the Duke School of Medicine and Duke Health.