Predictive Analytics ID High Risk Patients
By Katie Wike, contributing writer
In Maine, software is searching the statewide EHR in order to identify patients at a high risk of being admitted to the hospital or visiting the emergency room.
Maine’s statewide EHR is now equipped with a tool which allows providers to prevent admissions and emergency room visits for at risk patients. According to the Bangor Daily News, the program runs through records of more than 1.3 million residents in order to red flag those at risk of a visit to the ER. Often, these patients don’t even know they are at risk until their provider contacts them.
Fierce EMR reports the program spearheaded in part by HealthInfoNet, the state's health information exchange. It applies the predictive analytics on the records in its database and gives that information to providers who can then contact the patient and offer support such as educational materials and resources.
“The tool alerts me, but that human element is still so necessary,” Jessica Taylor, a nurse care manager at St. Joseph Internal Medicine said. “It allows me to reach out to the patient and get the story.” For example, one man who showed up on the high risk list had recently lost his health insurance. Taylor was able to help him sign up for insurance under the Affordable Care Act and find his asthma medicine at a more affordable price. He was then no longer on the at-risk list.
“He hasn’t been in the hospital, he hasn’t been in the ED, he hasn’t been here,” Taylor said. “He’s been home with his family.”
Devore Culver, executive director and CEO of HealthInfoNet says the program came about when researchers decided to see if they could make rhyme or reason out of ER admissions. “We started with a fairly simple question,” Culver said. “Can you predict who’s going to show up in the emergency room?”
Using the software, they identified patients who had already been in the ER and worked backwards. The program produced a list of patients expected to return to the ER within six months. Interestingly, it was 74 percent accurate.
A similar program at the University of Pennsylvania Health System integrated software into the system’s EHR just over a year ago. The program identified patients at a risk of readmission within 30 days of their original admission based on how many times they had been admitted in the last year.