Clinical Risk Score in COVID-19 Predicts Severity, Informs Interventions


A risk score that predicts critical illness in patients hospitalized with COVID-19 could help inform allocation of clinical resources.

coronavirus, test, covid-19, risk, score

A newly developed and validated system to score risk for patients with coronavirus 2019 (COVID-19) becoming critically ill, recently described in JAMA Internal Medicine, could improve outcomes by informing decisions on allocating resources and initiating aggressive treatment.

The scoring system, winnowed down from 72 clinical variables of potential relevance to 10 that were validated to predict likelihood of patient deterioration, has been made available as a web-based calculator.

The clinical risk score was developed by Jianxing He, MD, PhD, Guangzhou Institute of Respiratory Health, China State Key Laboratory of Respiratory Disease, and colleagues of the China Medical Treatment Expert Group for COVID-19.

"We aimed to construct a risk prediction score based on a nationwide cohort of Chinese patients with COVID-19 to help identify patients at the time of hospital admission who are likely to develop critical illness," He and colleagues wrote.

There have been several other systems for prognostic clinical risk prediction developed for COVID-19, but a recent review found that only 1 provided data on validation.

He and colleagues gathered medical records for 1550 patients admitted with laboratory confirmed COVID-19 diagnosis from 575 hospitals in 31 provincial administrative regions between November 21, 2019, and January 31, 2020. A validation cohort of 710 patients was obtained from 4 different sites. Their analysis was conducted between February 20 and March 17, 2020.

Potential predictive variables from patient characteristics at hospital admission were drawn from clinical signs and symptoms, imaging results, laboratory findings, demographic variables and medical history. The researchers initially identified 72 variables for analysis, from which 19 remained as predictors.

From these, logistic regression analysis yielded 10 variables that were independently statistically significant predictors of critical illness and now comprise the risk score: chest radiography, age, hemoptysis, dyspnea, unconsciousness, number of comorbidities, cancer history, neurtrophil-to-lymphocyte ratio, lactate dehydrogenase, and direct bilirubin.

The accuracy of the score was measured by the area under the receiver operating characteristic curve (AUC); with an AUC in the principal cohort of 0.88 (95% CI, 0.85-0.91), and of 0.88 (95% CI, 0.84-0.93) in the validation cohort.

The score calculator provides a percentage probability of risk for the patient's condition to deteriorate to critical illness, rather than predicting the levels of severity.

"We deliberately did not categorize risk into low-, moderate-, and high-risk groups," investigators explained, "as we believe that clinicians are better informed by calculating the risk estimate for each individual patient and making decisions based on local or regional conditions."

He and colleagues suggested, for example, that the scoring could help determine which patients should be most closely monitored and considered for aggressive treatment or admission to the intensive care unit (ICU). They noted that in regions with good access to clinical and supportive care, outcomes might be optimized by providing more aggressive intervention to patients with moderate risk.

In areas with high case volume and/or limited resources, however, the choice may be made to reserve ICU beds and ventilators for those at higher risk.

"Estimating the risk of critical illness could help identify patients who are and are not likely to develop critical illness," investigators wrote, "thus supporting appropriate treatment and optimizing the use of medical resources."

Recent Videos
© 2024 MJH Life Sciences

All rights reserved.