Table 2. Demographics and clinical characteristics of validation cohort with comparisons to primary cohort subsets.

Statistical analyses of validation cohort demographics include comparisons between validation and healthy groups for age, sex, and ethnicity using a multinomial logistic regression that includes the variation from influenza-infected individuals. Comparisons between the validation cohort and both patient groups from the primary cohort (COVID-19 and influenza) were assessed using a multivariate logistic regression that included demographic and clinical variables. Results are reported as the corresponding P value and, when significant, include the OR. The validation cohort serves as the reference condition in all analyses, African American serves as the reference for ethnicity comparisons, and the negative indication serves as the reference for all categorical clinical characteristics. A significant OR > 1 indicates a positive association between the comparator group and the variable (e.g., patients with COVID-19 from the primary cohort were more likely to be intubated than validation cohort patients with COVID-19), whereas a significant OR < 1 indicates a negative association (e.g., healthy controls were likely to be younger than validation cohort patients). P values were adjusted for multiple testing by controlling the false discovery rate. SD denotes standard deviation, and IQR denotes interquartile range. The “immunocompromised” comorbidity was not included in the validation-influenza comparison logistic regression due to complete separation across groups and was instead tested using Fisher’s exact test.

	SARS-CoV-2 validation (n = 89)	Validation-primary COVID-19 comparison	Validation-healthy comparison	Validation-influenza comparison
Demographics
Means ± SD (range) age, in years	61 ± 17 (19–92)	P = 1, N.S.	P < 0.001, OR = 0.87	P = 0.041, OR = 0.93
Female	39% (35/89)	P = 1, N.S.	P = 1, N.S.	P = 1, N.S.
Ethnicity
African American	74% (66/89)	–	–	–
White	25% (22/89)	P = 1, N.S.	P = 0.069, N.S.	P = 1, N.S.
Other	1% (1/89)	P = 1, N.S.	–	P = 1, N.S.
Clinical characteristics
Mean (IQR) body mass index	28.6 (24–33)	–		–
Mean (IQR) symptom duration at study enrollment, in days	7.5 (2–9)	P = 0.958, N.S.		P = 0.635, N.S.
Hospital admission	94% (84/89)	P = 0.958, N.S.		P = 0.043, OR = 0.046
ICU admission	48% (43/89)	P = 0.10, N.S.		P = 1, N.S.
Intubation and mechanical ventilation	27% (24/89)	P < 0.001, OR = 19.03		P = 0.893, N.S.
In-hospital death	17% (15/89)	P = 0.509, N.S.		P = 1, N.S.
Comorbidities
Immunocompromised	8% (7/89)	P = 1, N.S.		P = 0.347, N.S.
Chronic lung disease	16% (14/89)	P = 0.017, OR = 5.9		P = 0.049, OR = 11.69
Chronic heart failure	15% (13/89)	P = 1, N.S.		P = 0.817, N.S.
End-stage renal failure	2% (2/89)	P = 1, N.S.		P = 1, N.S.
Diabetes mellitus	44% (39/89)	P = 1, N.S.		P = 1, N.S.
Active cancer	3% (3/89)	P = 0.893, N.S.		P = 0.387, N.S.