COVID-19 Risk Calculator Helps Predict Which Hospitalized Patients at Highest Risk

October 14, 2020

Patients who are at highest risk of progression to severe COVID-19 are older patients, those with comorbidities, and those living in nursing homes.

Common risk factors such as nursing home residence and age increase the likelihood of negative outcomes when a hospitalized patient has coronavirus disease 2019 (COVID-19), but other, less-well-known factors can also have a major impact on the course of a patient’s disease, according to a new analysis.

Those risk factors have been used to create a risk calculator which can help clinicians better understand which patients have a higher likelihood of progression once admitted to the hospital with COVID-19.

The report came from investigators at Johns Hopkins University, including corresponding author Brian T. Garibaldi, MD, PhD. It was published in the Annals of Internal Medicine. The data were based on 839 patients who were admitted to 5 hospitals in Maryland and Washington, DC in March and April of this year. The majority of patients (694; 83%) survived their bouts with the disease, and after evaluating the cases of those who survived and those who succumbed, the investigators found patients’ risk factors for severe disease or death varied dramatically from as high as 90% to as low as 5%.

The median age of the patients was 64, and they were split roughly evenly by gender (47% women). The majority of patients were in minority demographic groups that have been shown to be at higher risk of severe disease; 40% were black and 16% were Latinx. However, Garibaldi and colleagues said they did not find a race-based correlation with severity in this data set after adjusting for clinical factors. One in 5 patients (21%) were residents of a nursing home. Half of the nursing home residents admitted to the 5 hospitals ended up dying from the disease.

Forty-five patients already had severe disease when they were admitted to the hospital. Among the other 787, 15% had developed severe disease or died within 12 hours. By the fourth day, 31% had progressed to severe disease or death. The median time from admission to severe disease or death was 1.1 days.

Garibaldi and colleagues said a number of factors correlated with risk of severe disease or death. They included age, nursing home residence, comorbid conditions, obesity, respiratory symptoms, respiratory rate, fever, absolute lymphocyte count, hypoalbuminemia, troponin level, and C-reactive protein level.

They used those data to create a COVID-19 Inpatient Risk Calculator, which is available for free on the university’s website and can be used to assess a particular patient’s risk of progression to severe disease or death.

The number of deaths in the study was insufficient to create a similar calculator focused on risk of death specifically, however, the authors cited a number of risk factors, the biggest of which was death.

Those ages 75 and above had an adjusted hazard ratio of death of 3.56 (95% CI). Living in a nursing home affected risk of death, though the effect only statistically significant if the patient was under age 75 (2.30 HR versus 1.42 HR). Charlson Comorbidity Index scores and Sao2/Fio2 ratios of less than 375 aso lowered a patient’s risk of survival.

Garibaldi and colleagues noted a number of insights from the data. For one, they said strategies to help prevent infection in nursing homes are critical to reducing the number of deaths caused by the illness. They also noted that 1 in 5 patients who died had do not resuscitate/do not intubate orders, which affected the interventions used in those cases.

As for their calculator, the authors said the tool is most effective at predicting progression to severe disease or death in the first 2 days of hospital admission, but they said longitudinal data could be added to the model to strengthen its predictive power at later days.