Philadelphia-based Penn Medicine researchers recently found that a machine-learning algorithm that predicts mortality risk in cancer patients quadrupled the rates of end-of-life care conversations with patients.
The study, published in JAMA Oncology on Jan. 12, included 20,506 patients with cancer. The algorithm identified high-risk patients and sent email or text "nudges" to physicians to start serious illness conversations. The nudges resulted in more than 40,000 patient encounters, making it the largest study of machine learning-based interventions focused on serious illness care in oncology, according to a Jan. 12 Penn Medicine news release.
After a 24-week follow-up period, researchers found the following:
- Conversation rates increased from 3.4 percent to 13.5 percent among high-risk patients.
- The use of chemotherapy or targeted therapy in the final two weeks of life decreased from 10.4 percent to 7.5 percent.
- The intervention had no effect on other end-of-life metrics, including hospice enrollment or length of stay, inpatient death or intensive care unit use.
- There was an increase in conversations about goals of care among patients who were not flagged as high-risk by the algorithm. The increase was observed in all patient demographics but was larger among Medicare beneficiaries, suggesting the intervention could help close disparity gaps.
"Communicating with cancer patients about their goals and wishes is a key part of care and can reduce unnecessary or unwanted treatment at the end of life. The problem is that we don't do it enough, and it can be hard to identify when it's time to have that conversation with a given patient," study senior author Ravi Parikh, MD, an oncologist and associate director of the Penn Center for Cancer Care Innovation at Philadelphia-based Abramson Cancer Center, said in the release.