Bias in a common health care algorithm disproportionately hurts black patients

Simple tweaks to the machine-learning program could eliminate the disparity, researchers say

doctor and patient

One widely used machine-learning program that helps hospitals identify patients who need extra health services refers fewer black patients than white patients, a study finds.

aldomurillo/E+/Getty Images

A widely used algorithm that helps hospitals identify high-risk patients who could benefit most from access to special health care programs is racially biased, a study finds.