Individual-level precision diagnosis for coronavirus disease 2019 related severe outcome: an early study in New York

Abstract

Because of inadequate information provided by the on-going population level risk analyses for Coronavirus disease 2019 (COVID-19), this study aimed to evaluate the risk factors and develop an individual-level precision diagnostic method for COVID-19 related severe outcome in New York State (NYS) to facilitate early intervention and predict resource needs for patients with COVID-19. We analyzed COVID-19 related hospital encounter and hospitalization in NYS using Statewide Planning and Research Cooperative System hospital discharge dataset. Logistic regression was performed to evaluate the risk factors for COVID-19 related mortality. We proposed an individual-level precision diagnostic method by taking into consideration of the different weights and interactions of multiple risk factors. Age was the greatest risk factor for COVID-19 related fatal outcome. By adding other demographic variables, dyspnea or hypoxemia and multiple chronic co-morbid conditions, the model predictive accuracy was improved to 0.85 (95% CI 0.84-0.85). We selected cut-off points for predictors and provided a general recommendation to categorize the levels of risk for COVID-19 related fatal outcome, which can facilitate the individual-level diagnosis and treatment, as well as medical resource prediction. We further provided a use case of our method to evaluate the feasibility of public health policy for monoclonal antibody therapy.

Figure 1

Receiver Operating Characteristic (ROC) Curve as an Estimate of Diagnostic Accuracy for COVID-19 related Fatal Outcome in NYS. Figure legend: The area under the ROC curve for age model was 0.78. The area under the ROC curve for combined effect of demographic variables (age, sex, race/ethnicity), clinical presentation/examination, and all chronic co-morbid conditions was 0.85.