Phenotype prediction for Patients with COVID-19

Age (years)

Cough

Fever

Diarrhea

Cancer

Diabetes

Hypertension

This calculator was based on a multivariable prediction model with seven predictors, including one demographic variable: patient’s age; three symptom variables: the presence of cough, the presence of fever, and the presence of diarrhea; three comorbidity variables: the presence of cancer, the presence of diabetes, and the presence of hypertension.

There are a total of seven latent class analysis-defined phenotypes of COVID-19 patients. See Wang et al. (2020) for the detailed features of the different phenotypes.

Reference: Wang, X., Jehi, L., Ji, X., and Mazzone, P. (2020). Phenotypes and Subphenotypes of Patients with COVID-19: a Latent Class Modeling Analysis.

Subphenotype prediction for hospitalized patients with COVID-19

Age (years)

Cancer/COPD/emphysema

Creatinine on admission (mg/dL)

Albumin on admission (g/dL)

CRP on admission (mg/L)

WBC on admission (× 10^9/L)

This calculator was based on a multivariable prediction model with six predictors, including a patient’s age, the presence of cancer or COPD or emphysema, the values of creatinine, albumin, c-reactive protein (CRP), and white blood cell (WBC) count on hospital admission, respectively.

There are a total of 5 latent class analysis-defined subphenotypes of hospitalized patients with COVID-19. See Wang et al. (2020) for the detailed features of the different subphenotypes.

Reference: Wang, X., Jehi, L., Ji, X., and Mazzone, P. (2020). Phenotypes and Subphenotypes of Patients with COVID-19: a Latent Class Modeling Analysis.