Racial/Ethnic Disparities in the Observed COVID-19 Case Fatality Rate Among the U.S. Population

Introduction

In the U.S., minority groups including Blacks/African Americans and Latinx have experienced higher rates of COVID-19 infection and mortality than Whites, as well as higher rates of excess mortality associated with COVID-19.^1–3 Assessing differences in outcomes among those infected is challenging at the national level due to limitations in available data. Studies among subpopulations have often found little or no evidence of racial/ethnic disparities in case fatality. Using data from the Michigan Disease Surveillance System (MDSS) from March 8 through July 5, 2020, Zelner et al. found that Blacks had substantially higher rates of COVID-19 infection and mortality. However, the case fatality rate among Blacks was only modestly higher than that of Whites, suggesting that differences in COVID-19 mortality were driven by differences in the rate of infection.⁴ Similarly, among individuals receiving care in the US Department of Veterans Affairs tested for COVID-19 between February 8 and July 22, 2020, Rentsch et al. found that while Blacks and Hispanics were more likely to test positive than Whites, once infected their 30-day mortality was less and, after adjusting for age, sex and rural/urban residence, was equivalent to that of Whites.⁵

Several studies have examined racial/ethnic differences in hospitalization and/or case fatality rates among health care systems, where access to electronic health records permits adjusting for demographic and clinical covariates. Azar et al. examined COVID-19 positive patients within Sutter Health in northern California from January 1 through April 8, 2020, and found that African Americans had 2.7 times the odds of hospitalization compared to Whites despite adjusting for age, sex, comorbidities and income.⁶ Similarly, Price-Haywood et al. studied COVID-19 positive patients within Ochsner Health in Louisiana and found that although Blacks had roughly twice the odds of hospitalization than Whites, their risk of in-hospital death was similar.⁷ Adjusting for demographics, income, comorbidities and clinical characteristics upon admission had only a modest impact on the estimated difference in mortality. Finally, Ogedegbe et al. examined patients within New York University’s Langone Health system who were tested for COVID-19 between March 1 and April 8, 2020.⁸ In contrast to the previous two studies, the odds of hospitalization among those who tested positive were similar among Blacks, Hispanics and Whites. However, among those hospitalized, Black and Hispanic patients were less likely than White patients to die despite adjusting for demographic characteristics, comorbidities and insurance.

Griffith et al. discuss the potential of collider bias to distort understanding of COVID-19 disease risk and severity.⁹ One potential source of collider bias is restricting analyses to a non-representative subset of the population, such as those hospitalized for an active COVID-19 infection. For example, it is possible that a tendency among Blacks/African Americans to seek care later in the acute setting—perhaps due to structural inequities, biases on the part of providers and prior negative experiences with health care—may partly account for the higher rates of admission observed in some studies.^6,10 If true, focusing on in-hospital mortality could reduce the apparent difference in case fatality between Blacks and Whites due to a larger proportion of Black individuals with COVID-19 being hospitalized. More generally, while studies of selected subgroups can benefit from higher quality data and can answer questions of local interest, it remains important to study COVID-19 risk and severity at the population level.

The purpose of this study was to examine racial/ethnic differences in COVID-19 case fatality rates (CFRs) at the U.S. national level during the first year of the pandemic. We used data from The Atlantic’s COVID Tracking Project (CTP), whose Racial Data Tracker was widely regarded as the most complete source of information on race/ethnicity of COVID-19 cases and deaths during this period. We performed a parallel analysis using the Centers for Disease Control and Prevention (CDC) COVID-19 Case Surveillance Public Use data—an independently compiled and regularly updated individual-level data source. Although less complete than the CTP data, the CDC data contain information on age. Adjusting for age is critical to get an accurate understanding of differences in COVID-19 CFR.¹¹ We focused on the two largest minority groups, Blacks/African Americans and Latinx, and their comparison to Whites, since these categories are reported most completely by a large number of states and territories, and permit approximate comparability between the two datasets.

Methods

Results

The CTP dataset includes 28.4 million cases and 512,627 deaths as of February 28, 2021, corresponding to an overall CFR of 1.8% (Table 1). The CDC dataset contains 22.1 million cases over the same period, however survival status is missing for nearly half (10.4 million). Among the other 11.7 million cases, there were 263,398 deaths corresponding to an overall CFR of 2.3%. Removing those states and territories with problematic data (see Methods) left 235,635 deaths with an overall CFR of 2.0%. Both datasets show a similar pattern in the CFR over time, with the CDC curves running ahead of the CTP curves by approximately 1–2 months, as expected (Figure 1). Based on the CDC data, the CFR decreased from the start of the pandemic through September/October 2020. This decrease was followed by an increase and a subsequent decline, consistent with the estimate of f(.) from the mixed-effects model when fit to the CDC data below (Figure 2).

• Download figure
• Open in new tab

Figure 1. CFR by month and racial/ethnic group, CTP and CDC datasets

• Download figure
• Open in new tab

Figure 2. Estimated overall time trend in CFR from mixed-effects model fit to CDC data, parameterized using restricted cubic spline with 5 knots

The CFR was substantially higher for those 65 and older (13.5%) as compared to those 50–64 (0.8%) and 18–49 (0.1%) (Table 1). No deaths were reported in the CDC dataset for individuals under 18 years of age. In the CDC dataset, the CFRs for Black and Latinx were lower than for Whites throughout the entire year (Figure 1). In the CTP dataset, CFRs for the three racial/ethnic groups were similar from July 2020 through November 2020, though minorities still had lower CFRs than Whites before and after this period. Table 2 (Panel A) shows estimated CFR ratios comparing the two minority groups to Whites, after adjusting for differences in CFR across states. In the CTP dataset, the CFR ratio comparing Blacks to Whites was 0.8 (95% CI [0.7, 0.9]) and comparing Latinx to Whites was 0.4 (95% CI [0.3, 0.5]); the corresponding CFR ratios for the CDC dataset were even lower at 0.5 (95% CI [0.4, 0.6]) and 0.2 (95% CI [0.2, 0.3]), respectively.

The proportion of cases under age 65 was higher among Blacks (89%) and Latinx (93%) than among Whites (80%). These differences were even larger for the proportion of cases under age 50 (69% for Blacks and 78% for Latinx as compared to 58% for Whites). When stratifying by age group, the state-specific CFRs—for states reporting any deaths for a particular age by racial/ethnic group—among those aged 18–49 and 50–64 were higher, on average, among Blacks and Latinx than among Whites (Figure 3). The CFR ratios comparing Blacks to Whites were 3.9 for ages 18–49, 2.1 for ages 50–64 and 0.9 for ages 65 and older; corresponding CFR ratios comparing Latinx to Whites were 9.8, 3.0 and 0.9, respectively (Table 2, Panel A).

• Download figure
• Open in new tab

Figure 3. State-specific CFRs, separately by age group and racial/ethnic group (CDC data)

Results from the mixed-effects model fit to deaths reported in the CDC data are shown in Table 2 (Panel B). Consistent with the CFR ratios reported above, these results demonstrated higher CFRs for minority groups among those under age 65. Specifically, estimated CFR ratios comparing Blacks to Whites were 6.0 (95% CI [3.4, 10.8]) for ages 18–49, 2.7 for ages 50–64 (95% CI [1.8, 4.1]) and 1.0 (95% CI [0.8, 1.1]) for ages 65 and older. The corresponding estimated CFR ratios comparing Latinx to Whites were 4.1 (95% CI [1.9, 8.7]), 2.4 (95% CI [1.5, 3.8]) and 1.2 (95% CI [1.0, 1.4]), respectively. The CFR ratio comparing women to men (not shown) was 0.76 (95% CI [0.72, 0.81).

We observed considerable variation between states for both datasets, both in the CFRs and in the CFR ratios comparing racial/ethnic groups (Figure 3). The value of I² was over 95% in all cases. The estimated variance of the υ_j was 0.31, corresponding to an increase in the CFR of 75% for a state one standard deviation above the mean. Similarly, the estimated variance of the w_jk was 0.26, corresponding to a 67% increase in the CFR for a month one standard deviation above the mean; this variability within state over time is above and beyond that already accounted for by the estimated overall time trend (Figure 2). Figure 4 plots estimates of the υ_j for all 38 states included in the model.

• Download figure
• Open in new tab

Figure 4. Estimated differences in CFR across states (CDC data)

Discussion

While the overall CFR was lower among Blacks and Latinx than among Whites, the CFRs for Blacks and Latinx within age category were higher among those under age 65. This reversal in the direction of the association when stratifying by subgroup is a partial example of what is known as Simpson’s Paradox.²⁰ Although a full reversal of the association was not evident among those 65 or older (with the exception of the mixed-model estimated CFR ratio of 1.2 comparing Latinx to White), it is possible that further adjustment for age within this oldest subgroup would yield a reversal here too. Thus, these data reinforce the point made by Green et al. that COVID-19 CFRs that are not age-specific “may hide more than they reveal.”

We also found evidence of an interaction between age and race/ethnicity such that Black and Latinx disparities in COVID-19 CFR were largest among the youngest ages. To our knowledge, this finding has not been previously reported. This finding deserves further study as it could have substantial implications for health policy—especially for a disease where public attention concerning mortality, at least during the early stages of the pandemic, was primarily focused among the oldest ages.

Our results reflect differences in the observed CFR; that is, the number of deaths divided by the number of reported cases. For COVID-19, this differs substantially from the true, underlying CFR since many infected individuals, especially those who were asymptomatic, were never tested. Moreover, some individuals who tested positive may not appear in the datasets used here. Thus, we expect that much of the change over time, as well as of the substantial variation across states, reflects variation in the rates of testing and reporting. Several studies have found higher testing rates among Blacks and Hispanics than among Whites, including a study of roughly 50 million patients in the Epic health record system.^5,21–23 Higher testing rates among minorities would be expected ceteris paribus to reduce their CFR relative to Whites. Furthermore, by adjusting for differences across time and between states, our results should be robust to confounding due to differences in the racial mix of cases over time or in the distribution of minorities across states.

The CDC dataset contains only 78% of the cases in the CTP dataset, and of these, information on survival is missing for 47% of cases. In addition, among the cases with survival status for the 38 states included in the analysis, race/ethnicity was missing for 41%. Despite these limitations, analysis of this dataset correctly revealed the improvements in COVID-19 survival over the first 6–8 months of the pandemic as well as the established lower risk of death for women relative to men.^24–26 While the overall CFR ratios comparing minorities to Whites were smaller than for the CTP dataset, the basic conclusion is the same.

Finally, though we found evidence of racial/ethnic disparities in CFR among those under age 65, our study was also intended to highlight the limitations in data available on race/ethnicity of COVID-19 cases and deaths at the national level. Specifically, using information on the joint distribution of race/ethnicity and age (as available in the CDC dataset only) reveals an entirely different understanding than when not using it. Chowkwanyun and Reed argue convincingly that while identifying racial/ethnic disparities in COVID-19 is important, a lack of context as provided by richer data “can perpetuate harmful myths and misunderstandings that actually undermine the goal of eliminating health inequities.”²⁷ Acquiring such data from representative samples of the national population is sorely needed.