There are thousands of data points about hospitalized patients from the machines that monitor heart rate, blood pressure, temperature and more – far too much information for the human brain to process quickly. A predictive model in Epic helps clinicians make sense of all of that data so they can intervene faster when a patient’s condition is likely to worsen. Now, organizations are finding that this predictive model is effective for patients with COVID-19, too.
More than 50 health systems already use the Deterioration Index predictive model. Epic released an update to help them measure the model’s performance for patients with COVID-19, and now 21 organizations including Parkview Health, ProMedica Health System, Confluence Health, and North Oaks Health System are using the model for more than 16,000 of these patients.
The model gathers factors from a patient’s chart, such as age, vital signs,and lab results, and puts them into a single risk score. A study from six different health systems shows that the model effectively identifies when patients with COVID-19 might decline.
Parkview Health found that 75% of hospitalized patients who had a risk score in the middle zone were eventually transferred to the ICU. In more than 40% of cases where a patient needed a ventilator, the model showed a high risk more than three hours in advance. Clinicians use that information to help them determine which patients need to be closely monitored, and which patients need to be transferred to a hospital with open ICU beds.
“In times like this that are unprecedented in U.S. health care, you really do the best you can with the numbers you have, and err on the side of patient care,” said Mark Pierce, CMIO for Parkview Health.
Epic community members can learn more about using the model to identify patients at risk for deterioration in the Managing Coronavirus Disease 2019 (COVID-19) With Epic white paper, which is updated regularly with recommended build and workflows.