Promoting Equity In Clinical Decision Making: Dismantling Race-Based Medicine.

Area of Science:

• Health equity
• Artificial intelligence in medicine
• Algorithmic bias

Background:

• Racial and ethnic biases in healthcare algorithms are a growing concern.
• Race-based medicine, using race as a proxy for biology, leads to harmful treatment disparities for minoritized groups.

Purpose of the Study:

• To address racial and ethnic biases in artificial intelligence algorithms used in clinical decision-making.
• To propose a race-aware approach to mitigate health disparities.

Main Methods:

• Analysis of current AI applications in US healthcare.
• Review of policy changes and algorithm development innovations.
• Recommendation for a race-aware clinical decision support framework.

Main Results:

• Existing AI algorithms often embed race-based medicine, causing inappropriate and unjust treatment patterns.
• Progress is being made through policy changes and technological innovation.

Conclusions:

• A race-aware approach incorporating social determinants of health is essential to dismantle race-based medicine.
• Sustained stakeholder commitment is required to increase trial diversity, broaden precision medicine, improve education, and develop culturally responsive care policies.