How AI can solve healthcare's equity dilemma | Viewpoint

Medicare Advantage Star ratings have dropped for the third consecutive year, prompting healthcare organizations to realign themselves with the Centers for Medicare & Medicaid Services’ focus on health equity and personalized care. Traditional methods of patient outreach—such as mailing the same reminders to large populations—fail to account for critical personal factors, including language preference, cultural context, and even the time of day that might produce the best engagement. Rather than relying on one-size-fits-all messaging, the article argues that adopting personalized, AI-driven outreach can significantly improve care quality, reduce costs, and increase patient satisfaction—areas that are especially significant when trying to boost Star ratings. By merging AI with behavioral science, healthcare systems can better meet the unique circumstances of every patient, whether it’s an 80-year-old grandmother or her 50-year-old daughter.

However, the article points out that large language models (LLMs) are only part of the solution for closing the equity gap and motivating actual behavior change. Enter “Large Behavioral Models” (LBMs), which consider social determinants of health—like race, geography, digital access, and nutritional factors—in addition to a patient’s past actions and likely future behaviors. By analyzing these diverse data points, LBMs can deploy the right message at the right time, driving longitudinal engagement and better health outcomes. This shift not only promotes equity but also aligns with the broader evolution in healthcare, where AI’s real promise lies in personalized, context-aware interactions that help both patients and health organizations thrive.