Fig. 1

Characteristics of in-sample and out-of-sample patients. Blood samples were collected from 537 patients hospitalized with COVID-19 during an early surge in the outbreak, between March 10th and June 1st of 2020. Data from patients who were hospitalized early in the surge (before April 22, 2020) was used to analyze the cardiometabolic protein biomarkers and develop logistic regression and random forest models for severe outcomes. These patients comprised the in-sample group (shown in grey). These models were then used to predict the outcomes of the out-of-sample patients (shown in gold) who were hospitalized later in the surge (starting April 22, 2020). The in-sample and out-of-sample patients were compared across various demographic and clinical variables using a two-sided t-test for continuous variables and chi-square test for categorical variables. All race/ethnicity categories were self-reported. BMI categorization: < 18.5 kg/m² for underweight, 18.5–24.9 kg/m² for normal weight, 25.0–29.9 kg/m² for overweight, and ≥ 30.0 kg/m² for obese. SD Standard deviation, AA African American, BMI body mass index, CAD coronary artery disease, COPD chronic obstructive pulmonary disease, CRP C-reactive protein, LDH lactate dehydrogenase