Written by: Ken Cohen, MD
Chief Medical Officer, Optum Health
Chief Scientific Officer, Health Care Transformation Hub

Published: July 7, 2026

More than any other factor, where you live, and the resources around you have the greatest impact on your overall health.

People living in socially vulnerable communities face predictable and persistent barriers: limited access to primary care, fragmented care coordination and greater reliance on emergency services. As a result, they have significantly higher rates of chronic illness, mental health disorders and premature mortality.

These challenges are rooted in broad social factors — including income, housing stability, transportation and community resources — that don’t just influence health, but shape whether people can access care at all.

Our latest research on more than 1.7 million Medicare beneficiaries uses the CDC’s Social Vulnerability Index to better understand how social risk intersects with care delivery, and to isolate the impact of healthcare payment models — specifically, how providers who were truly accountable for their patients’ outcomes through two-sided risk arrangements compared with those in fee-for-service Medicare.

Across eight clinical and utilization measures, one pattern stood out: Models that align incentives around outcomes consistently outperform those that reward service volume.

Patients in two-sided risk Medicare Advantage experienced:

• 14% to 17% fewer hospital admissions
• 2% to 4% fewer emergency department visits
• 11% to 14% fewer hospitalizations through the ED
• 23% to 30% lower use of high-risk medications

Critically, these improvements were largest in communities with the highest social vulnerability.

That distinction matters. This wasn’t just better performance overall — it was better performance where the need is greatest.

Why does this work? Because two-sided risk models change the way our health system is set up. When providers  are accountable for the total cost and quality of a patient’s care, they invest in prevention, coordinate care more intentionally and intervene before conditions escalate. They identify their at-risk patients sooner, and they engage patients beyond the exam room.

Fee-for-service models are not built for this kind of proactive, whole-person care; they reward services delivered, not outcomes achieved.

If we’re serious about improving healthcare for people with the most obstacles in their way, we have to move beyond simply measuring disparities — and toward scaling the care models proven to reduce them.

The path forward is clear: Design care around the whole person, align incentives with outcomes, and meet patients where they are.