key: cord-0290436-g790wszs
authors: Xu, J. J.; Chen, J. T.; Belin, T. R.; Brookmeyer, R. S.; Suchard, M. A.; Ramirez, C. M.
title: National Population-Level Disparities in COVID-19 Mortality Across the Intersection of Race/Ethnicity and Sex in the United States
date: 2021-09-05
journal: nan
DOI: 10.1101/2021.08.29.21262775
sha: eda6752382cf67250e8d74570a0d8b062dda6f29
doc_id: 290436
cord_uid: g790wszs

Males and certain racial/ethnic minority groups have borne a disproportionate burden of COVID-19 mortality in the United States, and substantial scientific research has sought to quantify and characterize population-level disparities in COVID-19 mortality outcomes by sex and across categories of race/ethnicity. However, there has not yet been a national population-level study to quantify disparities in COVID-19 mortality outcomes across the intersection of these demographic dimensions. Here, we analyze a publicly available dataset from the National Center for Health Statistics comprising COVID-19 death counts stratified by race/ethnicity, sex, and age for the year 2020, calculating mortality rates for each race/ethnicity-sex-age stratum and age-adjusted mortality rates for each race/ethnicity-sex stratum, quantifying disparities in terms of mortality rate ratios and rate differences. Our results reveal persistently higher COVID-19 age-adjusted mortality rates for males compared to females within every racial/ethnic group, with notable variation in the magnitudes of the sex disparity by race/ethnicity. However, non-Hispanic Black, Hispanic, and non-Hispanic American Indian or Alaska Native females have higher age-adjusted mortality rates than non-Hispanic White and non-Hispanic Asian/Pacific Islander males. Moreover, persistent racial/ethnic disparities are observed among both males and females, with higher COVID-19 age-adjusted mortality rates observed for non-Hispanic Blacks, Hispanics, and non-Hispanic American Indian or Alaska Natives relative to non-Hispanic Whites.

We examine a dataset from the National Center for Health Statistics (NCHS) that was publicly released on 15 April 2021, comprising COVID-19 death counts (ICD-10 code U07.1 (21) as an underlying or multiple cause of death) stratified by race/ethnicity, sex, and age for the year 2020 (22) . Also included are counts of various underlying health conditions (e.g., diabetes mellitus, Alzheimer's disease, influenza and pneumonia) of the COVID-19 decedents in the dataset, but we do not consider these factors in this paper. The sex categories are male and female, and the following 7 racial/ethnic groups are recorded: non-Hispanic White (NH White), non-Hispanic Black (NH Black), Hispanic, non-Hispanic Asian (NH Asian), non-Hispanic American Indian or Alaska Native (NH AIAN), non-Hispanic Native Hawaiian or Other Pacific Islander (NH NHOPI), and Unknown. The following set of age groups are used: 0-17, 18-29, 30-49, 50-64 , and 65+. Death counts within race/ethnicity-sex-age strata between 1 and 9 are suppressed for confidentiality reasons to avoid the possibility that individuals would be identifiable. The NCHS dataset comprises 381,415 total non-suppressed COVID-19 deaths, but 1,620 of them have unknown race/ethnicity but known age and sex. Furthermore, there are 4 sex-age strata with unknown race/ethnicity that have suppressed death counts, and there are 6 sex-age strata with known race/ethnicity that have suppressed death counts. In this paper, we combine the NH Asian and NH NHOPI racial/ethnic groups into a single racial/ethnic group that we call non-Hispanic Asian or Pacific Islanders (NH Asian/PI) and combine their suppressed and non-suppressed death counts as appropriate. The NCHS data is included in the Supplementary Material as File S1.

To obtain the denominators for the mortality rates we calculate, we make use of estimated 2019 U.S. national population counts stratified by race/ethnicity, sex, and age from CDC WONDER (23) . The racial/ethnic categories in the CDC WONDER data include the racial/ethnic groups we consider in our analysis, and the CDC WONDER age groups are integers from 0 to 84 and a catch-all 85+ age group. The estimated population counts are appropriately collapsed over age groups to conform to the age groups in the NCHS data. The CDC WONDER data is included in the Supplementary Material as File S2.

As mentioned previously, 1,620 non-suppressed COVID-19 deaths in the NCHS data have unknown race/ethnicity, and 10 race/ethnicity-sex-age strata have suppressed death counts, each of which is an integer between 1 and 9. Our general estimation strategy is to calculate upper and lower bounds for the mortality rates of interest, which subsequently determine the upper and lower bounds for the associated mortality RR's and RD's. The upper bounds for the age-specific and age-adjusted mortality rates are obtained by (i) appropriately combining COVID-19 death counts with unknown race/ethnicity and COVID-19 death counts for the racial/ethnic group of interest, and (ii) assuming the maximal number (9) of deaths corresponding to suppressed death counts for the racial/ethnic group of interest or those with unknown race/ethnicity. The lower bounds for the age-specific and age-adjusted mortality rates are obtained by (i) assuming COVID-19 death counts with unknown race/ethnicity all do not correspond to the racial/ethnic group of interest, and (ii) assuming the minimal number (1) of deaths corresponding to suppressed death counts for the racial/ethnic group of interest. The upper and lower bounds for the age-specific and age-adjusted mortality rate ratios and differences are subsequently obtained from the appropriate quotients and differences of the upper and lower bounds for the age-specific and age-adjusted mortality rates. The mathematical details can be found in Appendix A. Hence, we communicate our results in terms of intervals prescribing the upper and lower bounds for the mortality rates, RR's, and RD's of interest. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted September 5, 2021. ; https://doi.org/10.1101/2021.08.29.21262775 doi: medRxiv preprint

All calculations in our analysis are performed using the R programming language. The code used in our analysis is available upon reasonable request from the corresponding author. Table 1 contains the theoretical bounds for the mortality rates for each race/ethnicity-sex-age stratum and the male-to-female mortality RR's and RD's for each race/ethnicity-age stratum. From Table 1 , males have higher COVID-19 mortality rates than females within almost all race/ethnicity-age strata. For example, among individuals aged 50-64, the age-adjusted mortality rates among NH White men and women are (6.34-6.45) and (3.57-3.61) per 10,000, respectively; the age-adjusted mortality rates among NH Black men and women are (22.1-22.8 ) and (13.3-13.5) per 10,000, respectively; the age-adjusted mortality rates among Hispanic men and women are (30.2-30.7) and (12.4-12.6 ) per 10,000, respectively; the age-adjusted mortality rates among NH Asian/PI men and women are (10.4-11.8) and (3.33-3.74) per 10,000, respectively; and the age-adjusted mortality rates among NH AIAN men and women are (29.9-40.2) and (19.6-22.7) per 10,000, respectively. Table 2 presents the theoretical bounds for the COVID-19 age-adjusted mortality rates for males and females stratified by race/ethnicity along with their corresponding age-adjusted male-to-female mortality RR's and RD's. To complement Table 2, Figure 1 graphically illustrates the male and female age-adjusted mortality rates stratified by race/ethnicity, and Figure 1 graphically illustrates the age-adjusted male-tofemale mortality RR's and RD's stratified by race/ethnicity. Males have higher age-adjusted mortality rates than females within each racial/ethnic group considered, with some differences quite notable. On the ratio scale, the age-adjusted mortality rate for males is over 60% higher than that of females among NH Asian/PI's and over 80% higher among Hispanics. On the difference scale, the male-female differences in age-adjusted mortality rates are highest among Hispanics (12.7-13.8 deaths per 10,000) and NH Blacks (7.48-8.58 deaths per 10,000).

The notable variation in the magnitudes of the sex disparity across different racial/ethnic categories is consistent with the findings of Rushovich et al. The male-female disparity is greater among NH Blacks and NH Asian/PI's than among NH Whites on both the ratio and difference scales, and the male-female disparity is greater for Hispanics than both NH Whites and NH Blacks on both the ratio and difference scales. The male-female disparity is greater among NH Asian/PI's than among NH Blacks on the ratio scale, but the reverse is true on the difference scale. While the level of detail in the NCHS data was sufficient to allow us to ascertain that the COVID-19 age-adjusted mortality rate for males is greater than that for females among NH AIAN's, the widths of the corresponding age-adjusted male-to-female mortality RR and RD intervals obscures comparisons of the magnitude of the male-female disparity between NH AIAN's and the other racial/ethnic groups.

A second way to characterize the disparities is to compare racial/ethnic groups within sex categories. To this end, Table 3 presents the theoretical bounds for the COVID-19 age-adjusted mortality rates for NH Whites, NH Blacks, Hispanics, NH Asian/PI's, and NH AIAN's stratified by sex along with their corresponding age-adjusted mortality RR's and RD's, anchoring comparisons to NH Whites. From Table  3 , the magnitudes of the racial/ethnic disparities in COVID-19 mortality are often substantial but vary by sex. Disparities between NH Blacks and Hispanics relative to NH Whites are greater among males than among females on both the ratio and difference scales. Among males, age-adjusted mortality rates for NH Blacks are over twice that of NH Whites, and age-adjusted mortality rates for Hispanics are over 2.5 . CC-BY-NC-ND 4.0 International license It is made available under a perpetuity.

is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint Table 1 . Theoretical bounds for the (i) mortality rates in each race/ethnicity-sex-age stratum, (ii) maleto-female mortality rate ratios for each race/ethnicity-age stratum, and (iii) male-to-female mortality rate differences for each race/ethnicity-age stratum. Quantities are calculated with respect to data on COVID-19 deaths from the National Center for Health Statistics for the year 2020. Mortality rates are expressed per 10,000 people.

times that of NH Whites. Among females, age-adjusted mortality rates for NH Blacks and Hispanics are over 75% higher than that for NH Whites. Remarkably, age-adjusted mortality rates for NH AIAN's are over twice that of NH Whites among both males and females. Among males, the COVID-19 age-adjusted mortality rate is lowest among NH Whites, followed by NH Asian/PI's. Among females, however, the order is switched: the COVID-19 age-adjusted mortality rate is lowest among NH Asian/PI's, followed by NH Whites. Among males, the COVID-19 age-adjusted mortality rate for Hispanics is notably greater than that of NH Blacks, but the COVID-19 age-adjusted mortality rates between NH Blacks and Hispanics are indistinguishable (overlapping intervals) among females. Among females, the COVID-19 age-adjusted mortality rate is highest among NH AIAN's, but among males, the COVID-19 age-adjusted mortality rate for NH AIAN's is higher than those of NH Whites, NH Blacks, and NH Asian/PI's but is indistinguishable from that of Hispanics. Interestingly, COVID-19 age-adjusted mortality rates are higher for NH Asian/PI's than NH Whites among males but lower among females.

. CC-BY-NC-ND 4.0 International license It is made available under a perpetuity.

is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted September 5, 2021. ; https://doi.org/10.1101/2021.08.29.21262775 doi: medRxiv preprint Figure 1 . (A) Theoretical bounds for the age-adjusted mortality rates for each race/ethnicity-sex stratum, (B) Theoretical bounds for the age-adjusted male-to-female mortality rate ratios by race/ethnicity, (C) Theoretical bounds for the age-adjusted male-to-female mortality rate differences by race/ethnicity. Quantities calculated from data on COVID-19 deaths from the National Center for Health Statistics for the year 2020. Mortality rate differences are expressed per 10,000 people.

Stark contrasts in COVID-19 age-adjusted mortality rates are also observed between certain subgroups defined by race/ethnicity and sex. For example, COVID-19 age-adjusted mortality rates among Hispanic males are over 3 times that of NH White females and over 4 times that of NH Asian/PI females. Furthermore, although COVID-19 age-adjusted mortality rates are higher among males within all racial/ethnic groups, females among some racial/ethnic groups have greater COVID-19 age-adjusted mortality rates than males in other racial/ethnic groups: NH Black, Hispanic, and NH AIAN females all have higher COVID-19 age-adjusted mortality rates than NH White and NH Asian/PI males, important details that are masked when analyzing COVID-19 mortality data disaggregated by sex but not also by race/ethnicity. Indeed, the age-adjusted mortality rates for both NH Black and Hispanic females are over 40% higher than that of NH White males, and the NH AIAN female age-adjusted mortality rate is over 60% higher than that of NH White males. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint 

This is the first analysis quantifying population-level disparities in COVID-19 mortality across race/ethnicity and sex jointly at the U.S. national level. Having demonstrated empirically that males have higher COVID-19 age-adjusted mortality rates within each racial/ethnic group, certain racial/ethnic minority groups (non-Hispanic Blacks, Hispanics, and non-Hispanic AIAN's) have higher COVID-19 age-adjusted mortality rates relative to non-Hispanic Whites among both males and females, and sex differences in COVID-19 age-adjusted mortality rates vary markedly in magnitude by race/ethnicity, researchers must consider multiple pathways (e.g., biological, behavioral, social), that could be responsible for driving these observed differences in COVID-19 mortality outcomes across the intersection of race/ethnicity and sex. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted September 5, 2021. ; https://doi.org/10.1101/2021.08.29.21262775 doi: medRxiv preprint

The persistent one-sided and male-biased disparity within racial/ethnic groups lends support to the theory that biological differences between males and females play an important role in driving the pronounced differences in COVID-19 mortality rates by sex (24, 25, 26, 27) . Indeed, there has been extensive scientific research in this domain. An abundance of attention has been focused on differences in angiotensinconverting enzyme-2 (ACE2) (28, 29, 30) expressions, transmembrane serine protease 2 (TMPRSS2) (31, 32) expressions, and T-cell response (33, 34, 35) between males and females and their relationship to observed disparities in COVID-19 mortality outcomes by sex. Differential rates of certain pre-existing health conditions by sex have also been cited as a likely contributing factor to the differential rates of COVID-19 mortality by sex. For instance, U.S. men have higher rates of coronary artery disease (36, 37) , diabetes (38) , and liver disease (38) , which are risk factors for COVID-19-related severe outcomes (39) . Men in the U.S. also have higher rates of hypertension (40) , which some studies have indicated may also be a risk factor for COVID-19-related severe outcomes (39, 41, 42) .

Behavioral differences between men and women have also been widely hypothesized as a contributing factor to the observed male-female disparities in COVID-19 mortality. For example, men have higher rates of cigarette smoking (43) , which is a risk factor for COVID-19-related severe outcomes (39) . Men are less likely to wear face masks than women (44, 45, 46, 47, 48) , and some studies have suggested that masks can reduce the severity of a COVID-19 infection by reducing the viral inoculum (49, 50, 51, 52) . Proper hand washing may also reduce the viral inoculum of a COVID-19 infection (53) , and men are much less likely to practice proper hand washing behaviors (54, 55, 56) . More broadly, women have been estimated to be approximately 50% more likely than men to practice non-pharmaceutical health-protective behaviors (e.g., proper hand washing, face mask wearing, surface cleaning) in the context of respiratory infectious disease epidemic and pandemic settings, according to a 2016 meta-analysis (57) .

Our results also demonstrated that NH Blacks, Hispanics, and NH AIAN's have substantially higher COVID-19 age-adjusted mortality rates compared to NH Whites among both males and females, which is concerning from a health equity perspective. The disproportionately high COVID-19 mortality burden experienced by these racial/ethnic groups is well-documented, and many theories have been proposed to explain these disparities. For example, differences in the rates by race/ethnicity of pre-existing health conditions identified as risk factors for severe COVID-19 illness (39) have been theorized as contributing to observed disparities in COVID-19 mortality outcomes by race/ethnicity. Blacks have substantially higher rates of diabetes and hypertension relative to NH Whites (58, 59) , as well as an overall higher rate of obesity (60) . Hispanics have a considerably higher rate of diabetes than NH Whites (61) as well as a higher overall rate of obesity (62) . NH AIAN's have nearly three times the rate of diabetes as NH Whites (63) as well as higher overall rates of coronary heart disease and hypertension (64) . Health insurance coverage rates among NH Blacks, Hispanics, and NH AIAN's have historically lagged behind NH Whites and NH Asians (65) , which generally contributes to poorer health outcomes and higher rates of chronic health conditions, some of which are listed above. In fact, a Gallup-West Health study conducted early in the U.S. COVID-19 epidemic (1-14 April 2020) found that 14% of non-White survey respondents would avoid treatment for a suspected COVID-19 infection due to cost of care compared to only 6% of Whites (66) . Poor housing conditions, inextricably linked with overcrowding and lack of adequate plumbing and sanitation, have been found to be associated with higher COVID-19 incidence and mortality (67) , and NH Blacks and Hispanics are more likely to reside in moderate to severe substandard housing than NH Whites (68) . Blacks and Hispanics are also more likely to live in areas with higher rates of industrial air pollution (69) , which some studies have suggested may increase the risk of COVID-19 mortality (70, 71, 72) . is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted September 5, 2021. ; https://doi.org/10.1101/2021.08.29.21262775 doi: medRxiv preprint

The notable variation in the magnitudes of the male-female disparities across race/ethnicity (or equivalently, the notable variation in the magnitudes of the racial/ethnic disparities across sex) indicates that factors related to social determinants of health (73, 74) -some examples of which are listed above-play an important role in driving disparities in COVID-19 mortality outcomes across population subgroups defined by race/ethnicity and sex. For example, the sex disparity in the COVID-19 age-adjusted mortality rates among Hispanics is notably larger than those among NH Whites and NH Blacks in terms of both ageadjusted mortality RR's and RD's. A possible reason for such heterogeneity is the fact that Hispanic men are vastly overrepresented in essential industries such as farming/agriculture, construction, and transportation (75, 76) that required continued employment during the COVID-19 pandemic, subjecting them to increased risks of COVID-19 exposure. Indeed, a 2021 study in California (U.S.) indicated that these industries are among the occupational sectors with the highest associated excess mortality attributable to COVID-19 (77) . Our results are consistent with this hypothesis, with markedly greater male-to-female mortality RR's and RD's observed among Hispanics compared to NH Blacks and NH Whites in the prime working age groups (Table 1) . Moreover, even though COVID-19 age-adjusted mortality rates are higher for males compared to females within every racial/ethnic group, this pattern does not hold across all racial/ethnic groups. The fact that NH Black, Hispanic, and NH AIAN females all have higher COVID-19 age-adjusted mortality rates than NH White and NH Asian/PI males invalidates the potential view that observed disparities by sex are entirely or mostly a consequence of innate biological differences that are constant across racial/ethnic groups. Indeed, the perspective that factors related to social determinants of health primarily drive the racial/ethnic disparities in COVID-19 mortality outcomes themselves is widely propounded in the scientific community (78, 79, 80, 81, 82, 83) . Nevertheless, precisely how biological factors and social processes both individually and jointly influence the nature and extent of population-level disparities in COVID-19 mortality outcomes across the intersection of race/ethnicity and sex is a complex scientific problem that warrants further research, but a comprehensive investigation is beyond the scope of this paper.

Finally, population-level disparities research is important in order to identify preventable morbidity and mortality and improve our knowledge of factors that contribute to disparities in health outcomes between population subgroups, with the ultimate hope of discovering interventions that will reduce them (84) . This study highlights some of the limitations of previous analyses quantifying population-level disparities that relied on public datasets reporting limited socio-demographic characteristics associated with COVID-19 outcomes. Such low-dimensional data inherently limits our ability to understand of the nature of disparities in COVID-19 outcomes between population subgroups and does little to inform how such disparities might be reduced or eliminated. Sophisticated disparities research is only possible with COVID-19 data disaggregated by multiple key variables, but such publicly available datasets in the U.S. have been and continue to be limited. Examples of stratifying variables of practical interest include but are not limited to race/ethnicity, sex, age, educational attainment, occupation, income, religious affiliation, household composition, underlying medical conditions, and geographical residence. We urge U.S. federal, state, and local governmental and health authorities to prioritize the public availability of accurate, detailed, and real-time multi-dimensional data on socially-relevant patient/decedent characteristics associated with COVID-19 outcomes to facilitate research illuminating health disparities between population subgroups so that they can be addressed.

TRB has received support from NIH/NCATS grant UL1 TR001881 and NIH/NIMH grant P30 MH058107 in addition to funding outside the scope of this work from the Patient Centered Outcomes Research Institute 9 . CC-BY-NC-ND 4.0 International license It is made available under a perpetuity.

is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint 

JJX contributed to the conceptualization of the project, development of the methodology, data analysis, visualization, and writing the original draft of the manuscript. JJX, JTC, TRB, RSB, MAS, and CMR contributed to reviewing and editing of the manuscript.

This research received no external funding.

The datasets analyzed for this study can be found in the Supplementary Material, available for download in the online version of the article. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted September 5, 2021. ; https://doi.org/10.1101/2021.08.29.21262775 doi: medRxiv preprint APPENDIX A

The upper bound for the age-specific mortality rate for race/ethnicity r, sex s, and age a is obtained using the following equation:

where d rsa is the number of non-suppressed COVID-19 deaths corresponding to race/ethnicity r, sex s, and age group a; I rsa is the number of suppressed death counts corresponding to race/ethnicity r, sex s, and age group a; and n rsa is the CDC WONDER 2019 U.S. national population estimate corresponding to race/ethnicity r, sex s, and age group a. Note that if I rsa > 0, then d rsa = 0 except in the case of NH Asian/PI's, where d rsa can be greater than 0 because we combine the NH Asian and NH NHOPI racial/ethnic groups in our analysis. Also, note that I rsa can be only be 0 or 1 except in the case of NH Asian/PI's, where it can also attain the value of 2 because we combine the NH Asian and NH NHOPI racial/ethnic groups in our analysis.

The lower bound for the age-specific mortality rate corresponding to race/ethnicity r, sex s, and age a is obtained using the following equation:

The age-adjusted mortality rates by race/ethnicity and sex are calculated using direct age adjustment (85) . The upper bound for the age-adjusted mortality rate for race/ethnicity r and sex s is obtained using the following equation:

where n a is the CDC WONDER 2019 U.S. national population estimate corresponding to age group a.

The lower bound for the age-adjusted mortality rate corresponding to race/ethnicity r and sex s is obtained using the following equation:

. CC-BY-NC-ND 4.0 International license It is made available under a perpetuity.

is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint × n a a n a .

(4)

Theoretical bounds for the age-specific and age-adjusted mortality RR's and RD's can be obtained in a straightforward manner from the appropriate upper and lower bounds for the corresponding individual mortality rates.

To obtain the theoretical bounds for the age-specific mortality RR/RD comparing race/ethnicity-sex-age stratum (r, s, a) to stratum (r,s,ã) (reference stratum), the upper bound equals the upper bound of the mortality rate for stratum (r, s, a), given in Equation (1), divided by/minus the lower bound of the mortality rate for stratum (r,s,ã), given in Equation (2):

The lower bound of the age-specific mortality RR/RD comparing race/ethnicity-sex-age stratum (r, s, a) to stratum (r,s,ã) equals the lower bound of the mortality rate for stratum (r, s, a) divided by/minus the upper bound of the mortality rate for stratum (r,s,ã):

RD (min) rsa,rsã = R (min) rsa − R (max) rsã .

The upper bound of the age-adjusted mortality RR/RD comparing race/ethnicity-sex stratum (r, s) to stratum (r,s) (reference stratum) equals the upper bound of the age-adjusted mortality rate for stratum (r, s), given in Equation (3), divided by/minus the lower bound of the age-adjusted mortality rate for stratum (r,s), given in Equation (4) 

17 . CC-BY-NC-ND 4.0 International license It is made available under a perpetuity.

is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted September 5, 2021. ; https://doi.org/10.1101/2021.08.29.21262775 doi: medRxiv preprint RD (max) adj,rs,rs = R (max) adj,rs − R (min) adj,rs .

The lower bound of the age-adjusted mortality RR/RD comparing race/ethnicity-sex stratum (r, s) to stratum (r,s) equals the lower bound of the age-adjusted mortality rate for stratum (r, s) divided by/minus the upper bound of the age-adjusted mortality rate for stratum (r,s):

rs,rs = R (min) adj,rs

RD (min) adj,rs,rs = R (min) adj,rs − R (max) adj,rs .

. CC-BY-NC-ND 4.0 International license It is made available under a perpetuity.

is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted September 5, 2021. ; https://doi.org/10.1101/2021.08.29.21262775 doi: medRxiv preprint

Centers for Disease Control and Prevention. COVID-19 Hospitalization and Death by Race/Ethnicity

Demographic Trends of COVID-19 cases and deaths in the US reported to CDC

Assessing differential impacts of COVID-19 on Black communities

Racial and Ethnic Disparities in Population-Level Covid-19 Mortality

Variation in racial/ethnic disparities in COVID-19 mortality by age in the United States: A cross-sectional study

Racial and Ethnic Disparities in Years of Potential Life Lost Attributable to COVID-19 in the United States: An Analysis of 45 States and the District of Columbia

Racial Disparities in Coronavirus Disease 2019 (COVID-19) Mortality Are Driven by Unequal Infection Risks

Racial disparities in COVID-19 mortality across Michigan, United States

Population-Level Disparities in COVID-19: Measuring the Independent Association of the Proportion of Black Population on COVID-19 Cases and Deaths in US Counties

National Disparities in COVID-19 Outcomes between Black and White Americans

Coronavirus disease 2019 (COVID-19) mortality and neighborhood characteristics in Chicago

Excess mortality in men and women in Massachusetts during the COVID-19 pandemic

Outcome Disparities Among Men and Women With COVID-19: An Analysis of the New York City Population Cohort

Sexspecific impact of severe obesity in the outcomes of hospitalized patients with COVID-19: a large retrospective study from the

Male-Female Disparities in Years of Potential Life Lost Attributable to COVID-19 in the United States: A State-by-State Analysis

Gender-based disparities in COVID-19 patient outcomes

Sex Differences in COVID-19 Hospitalization and Mortality

Male Sex, Severe Obesity, Older Age, and Chronic Kidney Disease Are Associated With COVID-19 Severity and Mortality

Risk Factors for Mortality in Patients with COVID-19 in New York City

Sex Disparities in COVID-19 Mortality Vary Across US Racial Groups

Centers for Disease Control and Prevention. New ICD-10-CM code for the 2019 Novel Coronavirus (COVID-19)

AH Provisional COVID-19 Deaths and Contributing Conditions by Sex, Race and Hispanic Origin, and Age

Bridged-Race Population Estimates

Male sex identified by global COVID-19 meta-analysis as a risk factor for death and ITU admission

Biological sex impacts COVID-19 outcomes

Considering how biological sex impacts immune responses and COVID-19 outcomes

Higher mortality of COVID-19 in males: sex differences in immune response and cardiovascular comorbidities

ACE2 expression and sex disparity in COVID-19

Men more vulnerable to COVID-19: explained by ACE2 on the X chromosome?

Hypothesis: Sex-Related Differences in ACE2 Activity May Contribute to Higher Mortality in Men Versus Women With COVID-19

TMPRSS2 and COVID-19: Serendipity or Opportunity for Intervention?

TMPRSS2 activity may mediate sex differences in COVID-19 severity

Sex differences in immune responses that underlie COVID-19 disease outcomes

The known unknowns of T cell immunity to COVID-19

Mucosal-associated invariant T cell responses differ by sex in COVID-19

Age-adjusted percentages (with standard errors) of selected diseases and conditions among adults aged 18 and over, by selected characteristics: United States

Estimated Hypertension Prevalence

Hypertension is associated with increased mortality and severity of disease in COVID-19 pneumonia: a systematic review, meta-analysis and meta-regression

Hypertension is a clinically important risk factor for critical illness and mortality in COVID-19: A meta-analysis

Centers for Disease Control and Prevention. Current Cigarette Smoking Among Adults in the United States

Who is wearing a mask? Gender-, age-, and location-related differences during the COVID-19 pandemic

Gender differences in preventing the spread of coronavirus

Gender differences in COVID-19 attitudes and behavior: Panel evidence from eight countries

The effect of messaging and gender on intentions to wear a face covering to slow down COVID-19 transmission

The COVID-19 Responses of Men vs

Facial Masking for Covid-19 -Potential for "Variolation" as We Await a Vaccine

Surgical Mask Partition Reduces the Risk of Noncontact Transmission in a Golden Syrian Hamster Model for Coronavirus Disease 2019 (COVID-19)

Syrian hamsters as a small animal model for SARS-CoV-2 infection and countermeasure development

Does Virus Dose or Load Predict How Sick You Get With COVID-19?

Importance of non-pharmaceutical interventions in lowering the viral inoculum to reduce susceptibility to infection by SARS-CoV-2 and potentially disease severity

Hand Washing Practices in a College Town Environment

Characteristics Associated with Adults Remembering to Wash Hands in Multiple Situations Before and During the COVID-19

Men wash their hands much less often than women and that matters more than ever

A Meta-Analysis of the Association between Gender and Protective Behaviors in Response to Respiratory Epidemics and Pandemics

United States Department of Health and Human Services

United States Department of Health and Human Services

United States Department of Health and Human Services

United States Department of Health and Human Services

United States Department of Health and Human Services

United States Department of Health and Human Services

United States Department of Health and Human Services

Changes in Health Coverage by Race and Ethnicity since the ACA

14% With Likely COVID-19 to Avoid Care Due to Cost

Association of poor housing conditions with COVID-19 incidence and mortality across US counties

Environmental Health Disparities in Housing

Regional variation in environmental inequality: Industrial air toxics exposure in US cities

Air pollution and COVID-19 mortality in the United States: Strengths and limitations of an ecological regression analysis

Regional and global contributions of air pollution to risk of death from COVID-19

The Effects of Air Pollution on COVID-19 Infection and Mortality-A Review on Recent Evidence

Impact of Social Determinants of Health on the Emerging COVID-19 Pandemic in the United States

United States Department of Health and Human Services

Hispanics and Latinos in industries and occupations

Household Data Annual Averages: Employed persons by detailed industry, sex, race, and Hispanic or Latino ethnicity

Excess mortality associated with the COVID-19 pandemic among Californians 18-65 years of age, by occupational sector and occupation

COVID-19 and the impact of social determinants of health. The Lancet

The Detrimental Impact of the COVID-19 Crisis on Health Equity and Social Determinants of Health

Covid-19 and the social determinants of health

COVID-19 and Racial/Ethnic Disparities

Social Determinants of Health and Health Disparities: COVID-19 Exposures and Mortality Among African American People in the United States

The Relationship Between Social Determinants of Health and Racial Disparities in COVID-19 Mortality

The Importance of Evaluating Health Disparities Research