The United States continues to face rising healthcare costs, with Federally Qualified Health Centers (FQHCs) disproportionately affected by the complexity of the populations they serve. Traditional payment models often fail to reflect these realities, limiting value-based care. This article describes a 4-year value-based care pilot at Waianae Coast Comprehensive Health Center, developed with a nonprofit Medicaid health plan and supported by artificial intelligence (AI) and natural language processing (NLP) analytics. The pilot aligned payment with population complexity and total cost of care management using predictive risk and impactability analytics embedded in clinical workflows. By integrating claims data with NLP-derived insights from unstructured clinical notes, the model identified high-risk patients and delivered targeted lifestyle and enabling services, including food-as-medicine interventions. In the first year, services delivered to 884 patients reduced total costs by 806, 208 while maintaining access and quality. A produce prescription program generated net savings of 118 per patient per month alongside improvements in hemoglobin A1c. Over 4 years, the approach served more than 3000 patients, generating approximately 4 million in sustained savings. These findings demonstrate that value-based care can succeed in FQHCs when supported by accurate risk adjustment, advanced analytics, and payer–provider collaboration, positioning lifestyle medicine as a cost-effective strategy.
Bettini et al. (Thu,) studied this question.