Artificial Intelligence (AI) modeling can help health care providers figure out which patients are most likely to miss appointments and how to best reach out to them to make sure they don't, according to a study by MetroHealth and Case Western Reserve University.
When staff combined AI modeling with follow-up calls from schedulers, patients were nearly 40% more likely to show up for their appointments, MetroHealth said in a media release.
“Leveraging this technology with creation of a standardized telephone outreach approach and simple, streamlined documentation allows our front-end staff to do this work during their busy clinic day alongside their other duties," said Metrohealth’s Jessica Higginbotham, a co-author of the study, published in the Journal of General Internal Medicine. "This is a win for the system and our patients.”
Using patient data, researchers built an AI model in Epic, MetroHealth’s electronic health record system, to predict the chances a patient would miss an appointment, according to MetroHealth. The study targeted adult internal medicine patients who had a probable no-show rate of 15% or greater.
Between January and September of 2022, a random selection of those patients received phone calls from MetroHealth schedulers, the release said. If patients indicated barriers that may prevent them from showing up for their appointments, schedulers offered resources such as transportation or telehealth options.
The approach resulted in improved show rates for patients, especially those who were Black, according to the hospital. Although race and ethnicity were not included as variables in the AI model, MetroHealth has a sizable Black patient population.
“We want to provide an outreach mechanism that is fair and provides a level of equity,” the study’s lead author, MetroHealth’s Dr. Yasir Tarabichi, said. “People at a higher risk for not showing up are the ones we are helping the most. Minority patients were more likely to pick up the phone when called, and we met them where they were.”
The study is significant because racial and ethnic disparities in missed appointments are common in safety-net health systems like MetroHealth, according to the release. These disparities in access risk harming the very patients safety-net systems are designed to serve.
While patients already receive appointment reminders through text messages and other means, researchers found that additional outreach proved beneficial because of the wide disparities in access to technology in various communities throughout Cleveland.
For the researchers, this study is only the beginning of applications for AI in healthcare.
“A lot of people think AI is going to instantly revolutionize health care, but it has to be put through paces, validated and tested,” Tarabichi explained. “Models must be implemented in ways that do not make disparities worse. This study hopefully provides a roadmap for health care systems thinking about implementing AI.”
The research project was supported by the Clinical and Translational Science Collaborative of Cleveland, a collaborative among Case Western Reserve and its affiliated hospital systems, which includes MetroHealth.