A team of researchers from Johns Hopkins University School of Medicine has developed the severe COVID-19 adaptive risk predictor (SCARP) calculator that reliably predicts the risk of progression to severe disease or death in hospitalized patients with COVID-19. The calculator’s development and performance were described in a recent article published in the Annals of Internal Medicine.

To develop the SCARP calculator, the investigators used data from 3163 patients hospitalized with COVID-19 who were included in the JH-CROWN: COVID-19 Precision Medicine Analytic Platform Registry. The registry houses data from hospitals within the Johns Hopkins Medicine health care system. Data obtained from the registry included demographic characteristics, admission source, comorbidities, time-varying vital signs, laboratory measurements, and clinical severity. The investigators applied random forest for survival, longitudinal, and multivariate (RF-SLAM) data analysis to predict 1- and 7-day risks for progression to severe COVID-19 or death for any day during the first 14 days of hospital admission.

Overall, the median age of the patients in this study was 61 years. A total of 331 patients developed severe COVID-19 in the first 6 hours of hospitalization, and 13 of these patients died. Approximately 7% (n=228) of patients in the study cohort experienced progression to severe disease or death during the next 24 hours. Another 11% (n=355) of patients either became severely ill or died in the following 7 days.


Continue Reading

For SCARP’s 1-day risk predictions for progression to severe disease or death, the area under the receiver-operating characteristic curve (AUC) was 0.89 (95% CI, 0.88-0.90) and 0.89 (95% CI, 0.87-0.91) during week 1 and week 2 of hospitalization, respectively. Additionally, the AUC for 7-day risk predictions for progression to severe COVID-19 or death was 0.83 (95% CI, 0.83-0.84) during the first week of hospitalization and 0.87 (95% CI, 0.86-0.89) during the second week of hospitalization.

Summary decision trees suggest the 1-day risk predication summary tree capture up to 89% of the variance of the 1-day RF-SLAM risk predictions. Also, the 7-day risk prediction summary tree captured up to 90% of the variance of the 7-day RF-SLAM risk predictions of either severe COVID-19 or death.

A limitation of the SCARP tool development process, according to the study investigators, was the use of data from a single health system, which may lead to issues with generalizability.

The researchers concluded that this study demonstrates the SCARP carries with it “the potential to serve as a quantitative tool to help guide clinicians managing patients hospitalized with COVID-19,” particularly in patients “whose clinical courses are complex and seemingly unpredictable.” Additionally, the tool could be used to “inform hospital operations to best use resources in meeting the ever-changing demand for intensive care,” the investigators wrote.

Reference

Wongvibulsin S, Garibaldi BT, Antar AAR, et al. Development of severe COVID-19 adaptive risk predictor (SCARP), a calculator to predict severe disease or death in hospitalized patients with COVID-19. Ann Intern Med. Published online March 2, 2021. doi:10.7326/M20-6754

This article originally appeared on Pulmonology Advisor