ASCO22: Chris Manz, MD on AI in oncology
Artificial intelligence as a tool to increase discussions between patients and their oncology team about care goals can improve end-of-life care, according to research at led by Dana-Farber's Chris Manz, MD and presented at #ASCO22 .
Historically, patients with cancer often receive aggressive care near the end of life that is not consistent with their goals and their wishes. We conducted a clinical trial where we have identified patients with cancer who had a poor prognosis using a machine learning algorithm. We then delivered nudges to their oncology clinicians to encourage a discussion with the patient about their goals and wishes. We previously reported that we were successful at more than doubling the rate of these goals of care conversations in an early phase of this trial. At this year's Asco meeting, we present results showing that we're able to maintain these higher rates of goals of care conversations for nearly a year. Importantly, we're also able to change the care that patients receive near the end of life, such that patients were one third less likely to receive chemotherapy or immunotherapy in the last two weeks of life. Generally, we think that chemotherapy and immunotherapy given in the last couple of weeks of life is not beneficial. So this represents a big improvement in an important quality metric. This study demonstrates that by increasing discussions between patients and their oncology team about goals of care, we can improve the care that patients receive near the end of life
