Duke Health researchers found the current medical standards for assessing stroke risk perform worse for Black patients than white patients.
The study, published Jan. 24 in JAMA, evaluated existing algorithms and two methods of artificial intelligence aimed at predicting a person's risk of stroke in the next 10 years. All algorithms were worse at assessing the risk for Black people than for white people, regardless of gender.
"Disparities can potentially become propagated by these algorithms, and things could get worse for some people, which may lead to inequity in treatment decisions for Black versus white adults," Michael Pencina, PhD, a corresponding author of the study and director of AI Health at Durham, N.C.-based Duke University School of Medicine, said in a news release. "We need to improve data collection procedures and expand the pool of risk factors for stroke to close the performance gap of algorithms between Black and white adults."