Where VBC Actually Fails
The value-based care conversation at the board level is largely a strategy conversation. Which arrangements to pursue, which payers to engage, what the attributed population looks like, how the shared savings mechanics work. These are important conversations, and health systems and MSOs are generally getting better at them. The strategic analysis is cleaner than it was five years ago.
The execution conversation is harder. It surfaces most visibly in the performance reviews where the expected shared savings didn’t materialize, the quality benchmarks came up short, or the attribution was lower than projected. When those results land, they almost always trace back to the same failure mode: the operational infrastructure to execute the VBC strategy was never actually built.
VBC doesn’t fail in the boardroom. It fails in the back office.
Program Specificity
Before describing what the infrastructure requires, a clarification that matters: MSSP, ACO REACH, Medicare Advantage delegated risk, and commercial upside-only arrangements have almost nothing in common operationally. MSSP operates on claims-derived attribution, retrospective reconciliation, and CMS-defined quality measure sets. ACO REACH has different beneficiary alignment rules, different benchmark methodology, and higher risk exposure. MA delegated risk involves member-level encounter data, plan-specific formulary and authorization rules, and risk adjustment mechanics that run on RAPS and EDPS submissions to CMS. Commercial upside-only arrangements often share none of this structure and vary by payer contract. A VBC infrastructure strategy that doesn’t distinguish between these programs isn’t a strategy — it’s a category label.
The Infrastructure VBC Actually Requires
Attribution management — knowing who is in your attributed population, which providers have been assigned, and how that population changes month to month — is the foundation. Without accurate attribution, every downstream function is working with the wrong denominator.
The attribution mechanism deserves a more explicit frame than it usually gets. In claims-based attribution, a patient attributes to a provider when the plurality of their primary care visits in the measurement period are billed under that provider’s NPI. That means a provider who is not enrolled — or who is enrolled at a different practice address than where they’re seeing patients — will have their claims landing outside the attribution pool. The patient never attributes. The panel is wrong before care management starts. Every credentialing and enrollment gap is simultaneously a fee-for-service problem and a VBC attribution problem, and they compound.
Risk adjustment and HCC capture is the element most frequently missing from VBC infrastructure discussions, and it’s typically the largest single lever in a risk arrangement. In any arrangement with prospective risk-based payment — MA delegated risk, MSSP Track 1+ or above, ACO REACH — the benchmark that determines your savings or losses is calibrated to a risk score. Incomplete HCC documentation doesn’t just reduce the accuracy of the risk prediction — it reduces the benchmark, making your actual performance look worse on a risk-adjusted basis even when the clinical outcomes are the same. An organization that closes quality gaps but fails to document the chronic conditions driving its population’s risk is leaving money in the benchmark. Systematic HCC capture — prospective chart review, provider documentation improvement, annual wellness visit protocols oriented toward complete problem list documentation — is a financial function as much as a clinical one.
Quality benchmark tracking requires a system that connects clinical documentation to measure criteria and projects year-end performance from incomplete data. The word “real time” is misapplied here and in almost every VBC context: claims feeds run 60 to 90 days lagged, and HEDIS measurement is retrospective. The organizations that consistently hit their quality targets aren’t operating on live data — they’re running projection models against lagged data and identifying, at any point in the year, which measures still have closeable gaps. The discipline is knowing how much of the year remains, which patients are reachable, and which measures can move from the current baseline. That’s harder than a real-time dashboard, and it’s the actual competency that separates high performers.
Shared savings reconciliation — confirming that your performance resulted in the savings calculation you expected — requires someone who can read the payer’s reconciliation methodology, audit the attribution and encounter data the payer used, and dispute discrepancies when they arise. Payer reconciliation models are not always correct, and the methodology documentation is rarely transparent. Organizations that don’t review the calculation accept the number they’re given.
What Is Outsourceable — and What Isn’t
Closing the execution gap requires being precise about which elements of VBC infrastructure can be delegated and which can’t. Attribution management, quality data plumbing, HCC capture workflows, reconciliation auditing, and enrollment-attribution coordination are functions that can be built and operated by a partner who already has the infrastructure — and the value of a partner who has already solved those problems at scale is real.
Care redesign, panel management, and the clinical protocols that drive quality measure performance can’t be outsourced in any meaningful sense. An operations partner can tell you that your diabetic population has a gap rate on HbA1c testing — they can’t decide how your clinical teams will close it. The organizations executing VBC well are the ones that are clear about that boundary: they acquire the operational infrastructure so that their clinical leadership can focus on the decisions that only they can make.
Before committing to the next value-based arrangement, map the operational infrastructure you’d need to execute it — attribution management, HCC capture workflows, quality tracking, shared savings reconciliation — and assess whether that infrastructure exists at the program-specific level the arrangement requires.