Editor's Note
In this study, machine learning algorithms accurately identified cancer patients who were at risk of 6-month mortality.
Of 26,525 cancer patients analyzed, machine learning models based on structured electronic health record data accurately predicted short-term mortality risk with good discrimination and positive predictive value.
When the gradient boosting algorithm was applied in real time, 100 of 171 patients classified as high risk were deemed appropriate for a conversation with their oncologists regarding serious illness.
These models could facilitate timely conversations between patients and physicians about treatment and end-of-life preferences, the researchers say.
