In the United States, people with multiple chronic health conditions account for 65 percent of all healthcare dollars spent and 95 percent of Medicare dollars. Their care is complex, often requiring expertise from multiple specialist doctors. Some healthcare providers do a better job of coordinating this care than others, yet there are few measures of the quality of coordination at a facility.
A team of researchers at St. Olaf College is looking to change that. This summer, students working with Associate Professor of Economics Ashley Hodgson and Assistant Professor of Economics Tom Bernardin began work to develop a measure to assess the quality of care coordination.
“These patients are going from doctor to doctor to doctor. They might have five doctors, but as a society we don’t have a good way of deciding if a particular facility, insurer, or medical system is good at coordinating that care across doctors. We have no way of knowing that, which is really important given that chronic conditions are so common,” Hodgson says.
The team’s research is part of the Collaborative Undergraduate Research and Inquiry (CURI) program, which provides opportunities for St. Olaf students from all academic disciplines to gain an in-depth understanding of a particular subject by working closely with a St. Olaf faculty member in a research framework.
To begin constructing a quality-of-coordination measure, Ella Hagopian ’20 and Tyler Radtke ’20 are working alongside Hodgson and Bernardin to first identify a group of patients likely to need the most coordination with the use of ICD-10-CM codes — hospital codes that tell of a patient’s diagnosis.
“From these codes we’ve had to find all of those doctors that were part of the patient’s hospital visit and then we have to also find the data that corresponds to the speciality of that physician,” Radtke says. “Because a patient might also see, for example, different cardiologists depending on rotations, we’re also trying to get a measure of the distinct number of specialties.”
They will then use traditional statistics as well as more modern machine learning algorithms in order to create a scoring system that helps researchers estimate the number of unique specialties a patient typically consults with to be treated with the most up-to-date medical care.
“We want the score to be a measure that other researchers might be able to use, because a lot of them don’t have access to the same amount of data that we do,” Hagopian says. “A lot of them just might know the diagnosis codes and not how many different doctors or specialists they see, so this score would be a way for them to measure the complexity without having that information themselves. They could use it in their own research.”
Hodgson and Bernardin are looking to use this summer’s research as a foundation for future work involving the assessment of facilities in their ability to treat patients with chronic conditions.
“As a team we think together about these problems,” Hodgson says. “I feel like the more brains with really different perspectives you have thinking about a problem, the more creative the solutions are going to be, and I always find that with CURI.”