Overview of COVID-19 clinical cohorts with investigation of susceptibility by race and ethnicity
| Study . | Location . | Participants (n) . | Diabetes findings . | Comorbidities findings* . | Select laboratory findings** . |
|---|---|---|---|---|---|
| Stokes et al. (58) | U.S. | 599,636 of known race | No correlation study of diabetes to race performed | 33% Hispanic, 22% Black, 1.3% American Indian or Alaska Native, which account for 18%, 13%, and 0.7% of the U.S. population, respectively, suggesting they were disproportionately affected by COVID-19 | Not examined |
| Bhargava et al. (59) | Detroit, MI | 197 | Diabetes more frequent in patients with severe (48.6%) vs. nonsevere infection (30.1%), OR 2.20 (95% CI 1.21–4.0; P = 0.009) in univariate analysis but not multivariate; no correlation study of diabetes to race performed | Obesity, hypertension, congestive heart failure, cerebrovascular disease did not increase univariate OR of severe disease, though CKD did; 82.1% were Black, and Black race was not a risk for severe infection | Elevated Cr and PCT had significant univariate OR >1 for severe disease; elevated Cr from baseline and initial CRP had significant multivariate OR >1 for severe infection |
| Gold et al. (60) | GA | 297; Black hospitalizations (83.2%) were disproportionate to other races, indicating greater disease severity | Diabetes prevalence did not differ significantly in Black patients (41.7%) vs. in patients of other races (32.0%) P = 0.21 | Hypertension more common in Black patients (69.6%) vs. patients of other races (54.0%), P = 0.047; mean BMI higher in Black (31.4%) patients vs. patients of other races (29.6%), P = 0.003; Black patients did not have higher mechanical ventilation use or mortality | Not examined |
| Azar et al. (57) | CA | 1,052 confirmed cases | Diabetes had OR 2.2, P < 0.01, for hospital admission, in multivariate analysis with adjustment for sex, age, comorbidities, socioeconomic factors; no correlation study of diabetes to race performed | Non-Hispanic African Americans had OR 2.7, P = 0.007, for hospital admission vs. non-Hispanic Whites, after adjustment for the same variables as listed for diabetes findings | Not examined |
| Raisi-Estabragh et al. (61) | U.K. | 1,326 positive, 3,184 negative COVID-19 tests from UK Biobank | Diabetes not a risk for susceptibility to positive vs. negative COVID-19 test; no correlation study of diabetes to race performed | Hypertension, high cholesterol not risks for susceptibility to positive vs. negative COVID-19 test; Black, Asian, and minority ethnic group more susceptible to positive vs. negative COVID-19 test, with adjustment for age, sex, BMI, diabetes, hypertension, cholesterol, and socioeconomic factors | Not examined |
| El Chaar et al. (56) | NYC | 4,260 deaths | Diabetes not investigated | Hispanic and Black patients had highest age-adjusted mortality rates per 100,000 (22.8% and 19.8%, respectively, vs. other ethnic groups) corresponding to the groups with the highest obesity rates, 25.7% and 35.4%, respectively, P < 0.05; the two NYC boroughs with highest mortality rates, Bronx (6%) and Brooklyn (5.4%), also had the highest obesity rates, 32% and 27%, respectively | Not examined |
| Millet et al. (62) | U.S. | Nationwide population demographics and COVID-19 deaths | Diabetes prevalence in counties with <13% Black residents was 11.1% vs. 13.9% in counties with ≥13% Black residents, P value not stated; diabetes prevalence did not correlate with counties by COVID-19 cases (rate ratio 0.97 [95% CI 0.92–1.03]) or deaths (rate ratio 1.01 [95% CI 0.88–1.16]), after adjustment for demographics, comorbidities, and socioeconomic factors | Counties with higher Black resident proportions (≥13%) had more COVID-19 cases (rate ratio 1.24, 95% CI 1.17–1.33) and deaths (rate ratio 1.18, 95% CI 1.00–1.40), after adjusting for county-level traits, e.g., age, comorbidities, poverty, and epidemic duration | Not examined |
| Study . | Location . | Participants (n) . | Diabetes findings . | Comorbidities findings* . | Select laboratory findings** . |
|---|---|---|---|---|---|
| Stokes et al. (58) | U.S. | 599,636 of known race | No correlation study of diabetes to race performed | 33% Hispanic, 22% Black, 1.3% American Indian or Alaska Native, which account for 18%, 13%, and 0.7% of the U.S. population, respectively, suggesting they were disproportionately affected by COVID-19 | Not examined |
| Bhargava et al. (59) | Detroit, MI | 197 | Diabetes more frequent in patients with severe (48.6%) vs. nonsevere infection (30.1%), OR 2.20 (95% CI 1.21–4.0; P = 0.009) in univariate analysis but not multivariate; no correlation study of diabetes to race performed | Obesity, hypertension, congestive heart failure, cerebrovascular disease did not increase univariate OR of severe disease, though CKD did; 82.1% were Black, and Black race was not a risk for severe infection | Elevated Cr and PCT had significant univariate OR >1 for severe disease; elevated Cr from baseline and initial CRP had significant multivariate OR >1 for severe infection |
| Gold et al. (60) | GA | 297; Black hospitalizations (83.2%) were disproportionate to other races, indicating greater disease severity | Diabetes prevalence did not differ significantly in Black patients (41.7%) vs. in patients of other races (32.0%) P = 0.21 | Hypertension more common in Black patients (69.6%) vs. patients of other races (54.0%), P = 0.047; mean BMI higher in Black (31.4%) patients vs. patients of other races (29.6%), P = 0.003; Black patients did not have higher mechanical ventilation use or mortality | Not examined |
| Azar et al. (57) | CA | 1,052 confirmed cases | Diabetes had OR 2.2, P < 0.01, for hospital admission, in multivariate analysis with adjustment for sex, age, comorbidities, socioeconomic factors; no correlation study of diabetes to race performed | Non-Hispanic African Americans had OR 2.7, P = 0.007, for hospital admission vs. non-Hispanic Whites, after adjustment for the same variables as listed for diabetes findings | Not examined |
| Raisi-Estabragh et al. (61) | U.K. | 1,326 positive, 3,184 negative COVID-19 tests from UK Biobank | Diabetes not a risk for susceptibility to positive vs. negative COVID-19 test; no correlation study of diabetes to race performed | Hypertension, high cholesterol not risks for susceptibility to positive vs. negative COVID-19 test; Black, Asian, and minority ethnic group more susceptible to positive vs. negative COVID-19 test, with adjustment for age, sex, BMI, diabetes, hypertension, cholesterol, and socioeconomic factors | Not examined |
| El Chaar et al. (56) | NYC | 4,260 deaths | Diabetes not investigated | Hispanic and Black patients had highest age-adjusted mortality rates per 100,000 (22.8% and 19.8%, respectively, vs. other ethnic groups) corresponding to the groups with the highest obesity rates, 25.7% and 35.4%, respectively, P < 0.05; the two NYC boroughs with highest mortality rates, Bronx (6%) and Brooklyn (5.4%), also had the highest obesity rates, 32% and 27%, respectively | Not examined |
| Millet et al. (62) | U.S. | Nationwide population demographics and COVID-19 deaths | Diabetes prevalence in counties with <13% Black residents was 11.1% vs. 13.9% in counties with ≥13% Black residents, P value not stated; diabetes prevalence did not correlate with counties by COVID-19 cases (rate ratio 0.97 [95% CI 0.92–1.03]) or deaths (rate ratio 1.01 [95% CI 0.88–1.16]), after adjustment for demographics, comorbidities, and socioeconomic factors | Counties with higher Black resident proportions (≥13%) had more COVID-19 cases (rate ratio 1.24, 95% CI 1.17–1.33) and deaths (rate ratio 1.18, 95% CI 1.00–1.40), after adjusting for county-level traits, e.g., age, comorbidities, poverty, and epidemic duration | Not examined |
Conditions comorbid with diabetes considered.
Select laboratory findings for significant differences reported in immune cell populations, cytokines, and biomarkers of infection and kidney, liver, and cardiac damage. Changes were reported if there were significant differences in either mean values or in the number of patients above a cutoff value.