How machine learning models can predict if patients will show up for appointments

Researchers from Boston Children’s Hospital were able to predict with 83% accuracy if patients were going to be a no-show at the time of their appointment.

The findings, published in Nature, were a result of a machine learning algorithm that could help healthcare providers reduce the number of no-show appointments. The method led to a false alert rate of less than 17%.

While no-shows are costly for healthcare providers, these missed appointments also can be detrimental to the patient's health. Discontinuity of care and late presentation to care can lead to adverse outcomes, in addition to inefficient use of medical resources in hospitals and clinics. Predicting if patients are going to show up could enable healthcare providers to make interventions to limit the risks of no-shows. 

The retrospective study was conducted in an academic pediatric teaching hospital with a 20% no-show rate. Researchers designed their predictive model with three challenges in mind: missing patient data, local weather information and developing an interpretable approach that explains how a prediction is made for each individual patient.

The machine learning method worked even if some patient data was missing. The group used a data imputation strategy, based on a collection of missingness-type indicators, to address the frequent missing information in patient records. Including patient records that had missing information significantly improved the predictive accuracy when compared to a baseline approach that can only be trained and assessed on patients for whom complete information is available.

The output model could be used by healthcare providers to identify patients who could benefit from a text, call or email reminder about their appointment. The no-show predictions could also be used to optimize scheduling systems and minimize patient wait time.

“Our results suggest that choosing the day of the week and time of day that would be easier for patients and their parents to come to their medical appointments, and moreover using a language service and choosing a day with likely nicer weather would help reduce no-shows,” wrote first author Dianbo Liu, of the Boston Children’s Hospital and department of pediatrics at Harvard Medical School, and colleagues. “These effects were shown to be the most related to working hours of parents, traffic in the city, and potentially school schedules of the patients.”

 

Amy Baxter

Amy joined TriMed Media as a Senior Writer for HealthExec after covering home care for three years. When not writing about all things healthcare, she fulfills her lifelong dream of becoming a pirate by sailing in regattas and enjoying rum. Fun fact: she sailed 333 miles across Lake Michigan in the Chicago Yacht Club "Race to Mackinac."

Around the web

With generative AI coming into its own, AI regulators must avoid relying too much on principles of risk management—and not enough on those of uncertainty management.

Cardiovascular devices are more likely to be in a Class I recall than any other device type. The FDA's approval process appears to be at least partially responsible, though the agency is working to make some serious changes. We spoke to a researcher who has been tracking these data for years to learn more. 

Updated compensation data includes good news for multiple subspecialties. The new report also examines private equity's impact on employment models and how much male cardiologists earn compared to females.

Trimed Popup
Trimed Popup