Skip to main content
. 2022 Feb 24;45(5):1099–1106. doi: 10.2337/dc21-2186

Table 1.

Baseline cohort characteristics and death within 30 days of COVID-19 diagnosis

Population (n = 39,616) Deaths (n = 2,242) P value
Sex <0.01
 Male 19,431 (49.1) 1,310 (58.4)
 Female 20,185 (50.9) 932 (41.6)
Age (years) <0.01
 <40 2,728 (6.9) 21 (1.0)
 40–49 4,352 (11.0) 70 (3.1)
 50–59 8,550 (21.6) 236 (10.5)
 60–69 11,128 (28.1) 524 (23.4)
 70–79 8922 (22.5) 801 (35.7)
 ≥80 3,936 (9.9) 590 (26.3)
Race 0.64
 White 21,946 (55.4) 1,268 (56.6)
 Black or African American 10,467 (26.4) 569 (25.4)
 Asian or Pacific Islander 1,098 (2.8) 68 (3.0)
 Other/missing data 6,105 (15.4) 337 (15.0)
Ethnicity <0.01
 Non-Hispanic or Latino 31,084 (78.5) 1,720 (76.7)
 Hispanic or Latino 6,396 (16.1) 363 (16.2)
 Missing data 2,136 (5.4) 159 (7.1)
BMI (kg/m2) <0.01
 <20 822 (2.2) 96 (4.3)
 20–24.9 4,526 (11.4) 374 (16.7)
 25–29.9 10,224 (25.8) 639 (28.5)
 30–34.9 10,111 (25.5) 521 (23.2)
 35–39.9 6,794 (17.1) 325 (14.5)
 ≥40 7,139 (18.0) 287 (12.8)
Preexisting comorbidities
 Myocardial infarction 5,375 (13.6) 510 (22.7) <0.01
 CHF 9,475 (23.9) 972 (43.4) <0.01
 Peripheral vascular disease 8,427 (21.3) 712 (31.7) <0.01
 Stroke 7,060 (17.8) 628 (28.0) <0.01
 Dementia 1,878 (4.7) 290 (12.9) <0.01
 Pulmonary disease 12,638 (31.9) 830 (37.0) <0.01
 Mild liver disease 6,078 (15.3) 347 (15.4) 0.84
 Severe liver disease 1,185 (3.0) 123 (5.5) <0.01
 Renal disease 12,300 (31.0) 1,185 (52.8) <0.01
 Cancer 5,754 (14.5) 476 (21.2) <0.01
 HIV 444 (1.1) <20* 0.07
Medications
 Metformin 10,408 (26.3) 257 (11.5) <0.01
 GLP-1 receptor agonist 3,160 (7.9) 80 (3.6) <0.01
 DPP-4 inhibitor 2,172 (5.5) 97 (4.3) 0.015
 Sodium–glucose cotransporter 2 inhibitor 2,122 (5.4) 44 (2.0) <0.01
 Sulfonylurea 3,653 (9.2) 134 (5.9) <0.01
 Thiazolidinedione 503 (1.3) <20* <0.01
 Insulin 11,172 (28.2) 770 (34.3) <0.01
Severity of illness
 Hospitalization 19,401 (49.0) 2,067 (92.2) <0.01
 Invasive ventilation or ECMO 2,779 (7.0) 1,160 (51.7) <0.01

Data are presented as n (%).

*

Cells with <20 data points were reported as <20 to avoid risk of reidentification.