A cardiovascular risk prediction study of 3,241 adults found that removing race from clinical models paradoxically harmed Black patients despite improving statistical fairness metrics. When researchers compared traditional race-inclusive models against those using social determinants of health or clinical factors alone, the race-neutral approaches showed better population-level equity scores but generated concentrated clinical harms specifically among Black participants. The social determinants model systematically underpredicted risk and led to overtreatment, while the clinical-only model created four cases of untreated cardiovascular disease with no corresponding benefits. This finding challenges the growing movement toward race-neutral medical algorithms, revealing a critical disconnect between statistical fairness measures and real-world clinical outcomes. The research exposes how well-intentioned equity reforms can backfire when evaluated only through population averages rather than examining impacts on the most vulnerable groups. For cardiovascular prevention affecting millions of Americans, this suggests health systems need more nuanced approaches than simply removing race from algorithms. As this is a preprint awaiting peer review, these provocative findings require validation before informing policy decisions about clinical prediction tools.
Race-Neutral Heart Risk Models Harm Black Patients Despite Equity Metrics
📄 Based on research published in medRxiv preprint
Read the original research →⚠️ This is a preprint — it has not yet been peer-reviewed. Results should be interpreted with caution and may change following peer review.
For informational, non-clinical use. Synthesized analysis of published research — may contain errors. Not medical advice. Consult original sources and your physician.