Novel tool predicts progression to severe disease, death in COVID-19 patients

Researchers have recently developed an interactive tool that quickly and accurately provides the probability of a novel coronavirus disease (COVID-19) patient’s progression to severe illness or death based on readily available clinical information.

Called the Severe COVID-19 Adaptive Risk Predictor (SCARP), this novel, easy-to-use clinical prediction tool has undergone internal and temporal validation.

“SCARP has the potential to serve as a quantitative tool to help guide clinicians managing patients hospitalized with COVID-19, whose clinical courses are complex and seemingly unpredictable, and inform hospital operations to best use resources in meeting the ever-changing demand for intensive care,” researchers said.

This retrospective observational cohort study was conducted in five hospitals in Maryland and Washington, DC, in the US. Patients who were hospitalized between 5 March and 4 December 2020 with SARS-CoV-2 confirmed by nucleic acid test and symptomatic disease were included.

The primary data source was a clinical registry for patients hospitalized with COVID-19. Researchers obtained the following data: demographic characteristics, admission source, comorbid conditions, time-varying vital signs, laboratory measurements, and clinical severity.

They also performed random forest for survival, longitudinal, and multivariate (RF-SLAM) data analysis to predict the 1- and 7-day risks for progression to severe disease or death for any given day during the first 14 days of hospitalization.

A total of 3,163 patients admitted with moderate COVID-19 were included in the study, of whom 228 (7 percent) became severely ill or died in the next 24 hours. In addition, 355 (11 percent) patients became severely ill or died in the next 7 days. [Ann Intern Med 2021;doi:10.7326/M20-6754]

The area under the receiver-operating characteristic curve (AUC) of SCARP for 1-day risk predictions for progression to severe disease or death was 0.89 (95 percent confidence interval [CI], 0.88–0.90) during the first week of hospitalization and 0.89 (95 percent CI, 0.87–0.91) during the second week. For 7-day risk predictions, the AUC was 0.83 (95 percent CI, 0.83–0.84) and 0.87 (95 percent CI, 0.86–0.89) during the first and second weeks of hospitalization, respectively.

“The SCARP tool substantially advances the performance and reliability of clinical prediction of COVID-19 severity by using time-varying covariates, highly granular clinical information, and robust survival analysis methods,” researchers said. “It is designed to provide interpretable and personalized risk prediction for severe disease or death in patients hospitalized with COVID-19 at any time in the first 2 weeks of their hospitalization.”

Several prediction tools for COVID-19 have already been described, but these have many limitations. For instance, calculators developed by Liang and colleagues predict severe illness or death that achieved AUCs of 0.88 by using the least absolute shrinkage and selection operator and logistic regression to develop a predictive risk score and 0.91 by using a deep learning survival Cox model. However, these methods depend on baseline variables and have limited tolerance to missing data. [Nat Commun 2020;11:3543; JAMA Intern Med 2020;180:1081-1089]

SCARP, on the other hand, provides accurate prediction of progression to severe disease or death at the time of admission, but requires input of 23 variables including symptoms, Charlson Comorbidity Index score, and a full suite of laboratory values with limited tolerance for missingness and is applicable only on the day of admission, according to researchers. [Ann Intern Med 2021;174:33-41]

“Further studies with national-level, external data sets have been planned for larger-scale validation and generalizability. In addition, work is under way to integrate a simplified version of SCARP into the electronic medical record and assess its utility in clinical practice,” they said.