Yale researchers use ML to predict when physicians would leave their jobs

With physician burnout and their subsequent departure becoming an issue in the healthcare facilities in recent times, Yale researchers said that they have designed a machine-learning programme that could predict which physicians would leave the job and what contributes to their departure.  

The study, published on February 1 in the journal PLOS One, analysed electronic health records (EHRs) of over 300 physicians representing 26 medical specialities over a 34-month period. Data included the time physicians spent using EHRs, clinical productivity measures such as patient volume, and physician demographics, including age and length of employment. During the analysis period, nearly 14% of physicians departed. 

The machine-learning model was able to predict which physicians would depart within six months with 97% accuracy. The artificial intelligence or AI-based model also identified how strongly different variables contributed to turnover risk and what variables changed when a physician transitioned from low to high risk of departure. 

The top four factors that predicted departure risk were how long the physician had been employed, their age, the complexity of their cases and the demand for services, said researcher.

A separate study published in October 2021 also shed light on physician departure concluded that less time spent on the EHR (in particular inbox management) was associated with physician departure, enabled only analyses of linear relationships and several other aspects were overlooked.  

Yale researchers claimed that this machine learning model allowed researchers to observe the challenges facing physicians with more nuance. For example, risk of departure was highest for physicians more recently hired and those with longer tenures but lower for those with middling tenure lengths. Similarly, risk of departure was higher for those up to 44 years old, lower for physicians aged 45- 64, and higher again for those 65 years old or older. 

There were also interactions between variables. For instance, more time spent on EHR activities decreased the risk of departure for physicians who had been on the job for less than 10 years. But for those physicians employed longer, it increased the risk of departure. 

"As physician burnout is an increasingly recognised problem, healthcare systems, hospitals, and large groups need to figure out what they need to do to ensure the emotional and physical health and well-being of the physicians and other clinicians who do the actual caring for patients," Robert McLean, MD, medical director of Northeast Medical Group mentioned in the report. 

"Many healthcare systems already have wellness officers and wellness committees who could have the responsibility of collecting and analysing this data and developing conclusions, which then would lead to implementation plans for changes and hopefully improvements," McLean added. 

The researchers reasoned that as physician turnover is disruptive to patients as well as costly to healthcare facilities, knowledge of contributors to physician departure could help improve job satisfaction and stem costly turnover.