Hopkins Tool Predicts How Hospitalized COVID Patients Will Fare

BALTIMORE (WBFF) - Johns Hopkins has been at the forefront of tracking and treating COVID during this pandemic.

Now Hopkins doctors have another new tool they’re using that’s helping to predict how patients who are hospitalized will fare.

It's an algorithm doctors can use to predict a patient’s risk of developing a severe case of COVID-19 or even dying from the disease.

The tool is called SCARP which stands for "Severe Covid-19 Adaptive Risk Predictor".

"What is this trying to address is the challenge of the uncertainty around who is going to do just fine with COVID and who is going to get more sick," says Johns Hopkins Dr. Matthew Robinson.

He explains it predicts how a patient will fair over the next day and week by giving the percentage they have for developing severe illness or dying.

Doctors put in a few of the patient’s vital signs and how much respiratory support a patient is receiving.

"We may not have information on whether or not a person has cardiovascular disease just easily at our fingertips, but we do know how fast their heart rate is going," he says.

This tool being used at five Hopkins hospitals in Maryland and DC.

ALSO READ | COVID-19 - The race between more contagious variants and the vaccine in Maryland

Doctors can use it to determine whether a patient should be on a regular floor or in the ICU, also what type of treatment is best.

"I can very easily see who are the sickest patients? Who are the ones who may need closer attention?" he says. "By knowing where you’re going in the next day or week, we can consider other therapeutics that are still hit or miss and how they work."

They developed this looking at data from thousands of patients Hopkins has treated over the last year.

ALSO READ | How Will Progressive Prosecution Impact Juvenile Justice System?

Among 3,163 patients admitted with moderate COVID-19 during this time, 228, or 7%, became severely ill or died within 24 hours. An additional 355, or 11%, became severely ill or died within the first week. Data also were collected on the numbers who developed severe COVID-19 or died on any day within the 14 days following admission.

Overall, SCARP’s one-day risk predictions for progression to severe COVID-19 or death were 89% accurate, while the seven-day risk predictions for both outcomes were 83% accurate.

Dr. Robinson says he’s used it to help ease some patient’s minds.

"I've used it more as a way to reassure patients that you are in the hospital and you’re sick, but we actually don’t think you’re going to deteriorate so much in the next day or the next week based on the information we already have," he says.

This is also helpful for hospitals to plan should patient counts rise.

As more people are hospitalized, it helps predict how many beds they need.

Dr. Robinson says it’s difficult to switch over a unit to a COVID floor if they need to expand so having a sense of those needs is crucial.