Reminders Based on Machine-Learning Algorithms May Improve End-of-Life Care for Patients With Cancer

Get Permission

Electronic nudges delivered to health-care clinicians based on a machine-learning algorithm capable of predicting mortality risk quadrupled rates of conversations with patients about their end-of-life care preferences, according to a new study published by Manz et al in JAMA Oncology. The study also found that the machine learning–triggered reminders significantly decreased the use of aggressive chemotherapy and other systemic therapies during end-of-life care—which research shows may be associated with poor quality of life and side effects that can lead to unnecessary hospitalizations in patients’ final days.


When cancer advances to an incurable stage, some patients may prioritize treatment that will extend their lives, while others may prefer a care plan designed to minimize pain or nausea. Talking to patients about their prognoses and values can help clinicians develop care plans that are better aligned with each individual’s goals, but it’s essential that these discussions happen before patients become too ill. 

“This study demonstrates that we can use informatics to improve care at [the] end of life,” highlighted senior study author Ravi B. Parikh, MD, MPP, FACP, Assistant Professor of Medicine as well as Medical Ethics and Health Policy at the Perelman School of Medicine, and Director of the Penn Center for Cancer Care Innovation at the Abramson Cancer Center at the University of Pennsylvania. “Communicating with patients [who have cancer] 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,” he emphasized.

Preliminary Results From 2020

Dr. Parikh and his colleagues previously demonstrated that a machine-learning algorithm could identify patients with cancer who were at high risk of mortality within the next 6 months. They paired the algorithm with behavioral-based nudges in the form of emails and text messages to prompt clinicians to initiate serious illness conversations during appointments with high-risk patients. The preliminary results of the study—published by Manz et al in JAMA Oncology in 2020—showed that the 16-week intervention tripled the rates of these conversations.

New Study Findings

The new, randomized clinical trial represented an important step for artificial intelligence (AI) in oncology as the first randomized trial of a machine learning–based behavioral intervention in cancer care. The study—which included 20,506 patients treated for cancer—involved over 40,000 patient encounters, making it the largest study of a machine learning–based intervention focused on serious illness care in oncology.

The findings showed that after a 24-week follow-up period, conversation rates nearly quadrupled from 3.4% to 13.5% among high-risk patients. Further, the use of chemotherapy or targeted therapy in the final 2 weeks of life decreased from 10.4% to 7.5% among patients who died during the study. The intervention did not have an impact on other end-of-life metrics—including hospice enrollment or length of stay, inpatient death, or end-of-life intensive care unit use.

Notably, the researchers also observed that conversations about goals of care increased with patients who weren’t flagged by the algorithm as high-risk—suggesting that the nudges caused clinicians to change their behaviors across their practice. The increase was observed in all patient demographics, but was larger among Medicare beneficiaries, which suggests that the intervention may help rectify a disparity in conversations about serious illness.


Building on the results of this study, the researchers expanded the same approach to all oncology practices within the University of Pennsylvania Health System and are currently analyzing those results. Additional plans for future studies include pairing AI algorithms with a prompt for early palliative care referrals and using the algorithm for patient education.

“While we significantly increased the number of dialogues about serious illness taking place between patients and their clinicians, still less than half of patients had a conversation,” underscored Dr. Parikh, concluding that “We need to do better because we know patients benefit when their health-care clinicians understand each patient’s individual goals and priorities for care.”

Disclosure: The research in this study was supported by the National Institutes of Health and the Penn Center for Precision Medicine. For full disclosures of the study authors, visit


The content in this post has not been reviewed by the American Society of Clinical Oncology, Inc. (ASCO®) and does not necessarily reflect the ideas and opinions of ASCO®.