In a recent Substack, I introduced the Friction Index, a single measure that balances the burden of prior authorization against its effectiveness for controlling costs. It combines publicly available CMS data on approval rates, denial rates, and how often denials are overturned on appeal.
The Friction Index (FI) across 157 Medicare Advantage contracts[1] – representing nearly 12 million beneficiaries – is 1.40. Plans with a Friction Index below 1.40 make it easier to get care. In plans with a FI above 1.40, beneficiaries experience more prior authorizations, denials and subsequently, delays in care. Since publishing that analysis, one question has come up more than any other.
Do nonprofit health plans create less friction than for-profit plans?
The nonprofit versus for-profit question is a reasonable one. Nonprofit insurers don’t answer to shareholders or quarterly earnings expectations. If prior authorization functions primarily as a cost-control tool, you might expect nonprofit plans to rely on it less aggressively.
But I wanted to answer the question with data rather than assuming.
When I analyzed the Friction Index by insurers’ tax status, for-profit Medicare Advantage plans have a weighted Friction Index of 1.60. Nonprofit plans average 1.10, roughly 31% lower.
While this felt validating, my 30+ years of industry experience told me that this wasn’t the final answer. An organization’s tax status is not their operating model. Some nonprofit insurers are structured like traditional for-profit insurers. They contract with independent hospitals and physician groups, pay claims, and manage utilization. Others are structured as integrated health systems where the health plan, hospitals, physicians, and clinical records function as one organization. These two models are fundamentally different.
Network Model or Integrated Model: Which One Actually Lowers Friction?
I re-analyzed the data after segmenting the nonprofit plans into two groups:
those with a traditional provider network model. These 29 organizations include several regional Blue Cross Blue Shield plans, Medica, Point32Health, Healthfirst, PacificSource, and Capital District Physicians’ Health Plan,
those with an integrated delivery system where the health plan and care delivery system operate together. These 23 insurers include Kaiser Permanente, Intermountain Health, HealthPartners, Priority Health, Baylor Scott & White, Wellsense and NYC Health + Hospitals.
The results change the story. Nonprofit plans operating in a traditional network model have a Friction Index of 1.70 – essentially the same as for-profit plans (1.60). Whereas integrated plans have a Friction Index of just 0.60.
While tax status certainly influences healthplans’ business practices, this analysis suggests that operating model is what distinguishes high-friction from low-friction plans.
What does “integrated” really mean?
In a traditional network model, providers are largely paid fee-for-service. When they do more, they earn more. Health plans, meanwhile, are responsible for controlling costs. Because those financial incentives are not fully aligned, insurers rely more heavily on prior authorization to confirm that services are clinically appropriate before paying for them. That creates friction.
In contrast, an integrated system isn’t simply a health plan that owns hospitals or physician practices. It’s an organization where the health plan, physicians, hospitals, and clinical information all work together toward the same goals. Physicians, hospitals, and the health plan share responsibility for both the cost and quality of care. Financial incentives are aligned, clinical information is shared across the organization, and everyone benefits from delivering efficient, appropriate care. Prior authorization becomes less necessary – and friction decreases – because the system is already working toward the same objective.
The benefits of an integrated system extend beyond reducing administrative burden. When fewer resources are spent managing prior authorization, more attention can be directed toward coordinating patient care.
One way to see this is through CMS’s Transitions of Care (C20) Star Rating measure. This metric reflects how well patients move from the hospital back to home or another care setting, including receiving clear discharge instructions, having medications reconciled, and completing timely follow-up care. The integrated plans in this analysis consistently outperform both for-profit plans and nonprofit network-model plans on this measure. When the organizations responsible for a patient’s care share information and accountability, patients are less likely to fall through the cracks.
This naturally raises another question.
Isn’t this what value-based care is supposed to achieve?
In theory, yes. Value-based care aligns financial incentives by rewarding providers for better outcomes, higher quality, and lower costs. But the evidence has been mixed. A Commonwealth Fund review found that the most successful value-based care programs weren’t simply those that changed how providers were paid. They were the ones that fundamentally changed how care teams worked together.
That’s an important distinction. In many value-based arrangements I’ve seen, providers still lack real-time access to payer claims data, while payers lack visibility into providers’ clinical information. Financial incentives may be shared, but clinical information often isn’t.
Key Takeaway
The lesson from this analysis is not that every health system should become integrated like Kaiser or Intermountain. Nor is it that nonprofit ownership is inherently better, or that for-profit ownership is inherently worse.
Instead, the data suggest something more practical. More coordinated care and lower friction both emerge when financial incentives and clinical information move together. Full organizational integration is one way to achieve that. Network-model plans might get there too, but they’d need to go beyond simply sharing financial risk and build genuine information exchange with their providers — shared data infrastructure, shared workflows, shared accountability.
So no, nonprofit status alone doesn’t explain why some Medicare Advantage plans create less friction than others. Integration does. And that’s a far more useful thing to know — because unlike tax status, integration is something any plan, for-profit or nonprofit, can actually choose to build.
[1] CMS data is reported at the contract level, which is the unit of analysis used throughout this piece. Contracts generally map to what most people think of as a health plan, but not always one-to-one — a single contract can cover multiple benefit packages, and a single insurer can hold multiple contracts. For readability, I’ll use “healthplan” for the rest of this piece, though the analysis itself stays at the contract level.
Thanks Clif. The data led me in a direction I wasn't expecting.
Terrific analysis!