Specialty value-based care has entered a new phase — one where the success or failure of a contract can hinge on surprisingly small analytical definitions. A slight change to attribution logic, exclusion criteria, or qualifying encounters can materially alter both the economics and operational reality of a specialty risk arrangement.

For years, most value-based care infrastructure has been designed around primary care models. Attribution logic is comparatively broad, populations are relatively stable, and analytics workflows tend to focus on total cost of care, preventive utilization, and longitudinal population management.

Specialty value-based care is fundamentally different.

Why Specialty VBC Requires a Different Analytics Model

Unlike primary care VBC, where attribution and analytics are often designed around broad longitudinal populations, specialty value-based care requires much narrower, clinically precise population definitions. In primary care models, attribution may depend simply on whether a patient saw a PCP within a defined timeframe. In specialty care, attribution often depends on highly nuanced combinations of diagnoses, procedures, treatment histories, provider engagement patterns, exclusions, and episode criteria.

In specialty risk arrangements, viability is often driven by highly specific definitions: qualifying diagnoses, exclusions, treatment pathways, lookback windows, medication utilization, and even the exact encounters that count toward attribution. Small adjustments to these parameters can significantly impact both the attributed population and the overall economics of the contract, affecting PMPM calculations, risk exposure, utilization patterns, and operational workflows.

The analytics themselves are also fundamentally different. Primary care VBC typically focuses on broad total-cost-of-care management, preventive care gaps, chronic disease burden, and generalized risk stratification. Specialty VBC, by contrast, requires disease-specific modeling that may incorporate staging, treatment sequencing, infusion utilization, imaging appropriateness, medication regimens, progression tracking, and specialty-specific quality and utilization patterns.

As a result, specialty VBC organizations require a level of analytical flexibility and transparency that most traditional population health or generalized VBC analytics platforms were never designed to support. Static attribution models and retrospective reporting are insufficient when organizations need to dynamically model payor-specific contract structures, compare multiple scenarios simultaneously, and operationalize those definitions across ongoing clinical and financial workflows.

That complexity increases dramatically when organizations are negotiating with multiple payors simultaneously, each with slightly different assumptions about what defines the “target” population.

The Problem Ursa Set Out to Solve

Working alongside specialty care organizations pursuing full-risk and advanced value-based arrangements, Ursa recognized that traditional actuarial and population health tooling could not adequately support the realities of specialty contracting. Most approaches could generate isolated analyses, but they were not designed for iterative, real-time negotiation workflows where attribution logic, exclusions, qualifying encounters, and financial assumptions evolve continuously across payors.

What emerged was a new kind of analytical infrastructure: a unified, parameterized specialty value-based care model capable of dynamically simulating and comparing complex contract scenarios within a single underlying data framework.

The challenge was not simply building a configurable model. Parameterization itself is not new. The challenge was designing a contract engine capable of handling the multivariate realities of specialty value-based care.

The Multivariate Nature of Specialty Attribution

In practice, specialty VBC attribution behaves less like a linear pipeline and more like an interconnected system of dependencies. A single parameter — such as what qualifies as a specialty encounter, how long a lookback period lasts, or which diagnoses qualify for attribution — may simultaneously affect eligibility logic, attribution persistence, exclusion handling, and downstream financial calculations.

Early oncology use cases quickly demonstrated that even modest changes to attribution criteria could create cascading downstream effects throughout the model. A tweak to a qualifying diagnosis definition or provider encounter requirement did not simply alter one metric — it could affect multiple stages of attribution logic simultaneously. That insight ultimately shaped the architecture of what became known internally at Ursa Health as the “Multiverse” — a framework designed to compare multiple attribution and contract scenarios against a shared patient population in real time.

The Multiverse: Scenario Modeling at Scale

Rather than rerunning isolated models each time assumptions change, the Multiverse enables organizations to evaluate multiple scenarios side by side within the same underlying data structure. Instead of running one scenario, saving the output, reconfiguring the model, and rerunning another, organizations can compare multiple scenarios side by side at every step of the attribution journey.

This distinction matters enormously.

Most analytics systems expose only final outputs: who was attributed, what the PMPM was, or how risk scores changed. The Multiverse allows organizations to understand why those outcomes changed by comparing attribution behavior step-by-step across multiple scenarios simultaneously. That level of transparency transforms contract negotiations from theoretical discussions into collaborative, evidence-based conversations grounded in reproducible data. That transparency becomes especially important in specialty VBC, where clinical, financial, and operational teams must align around the same definitions in order to execute effectively.

Beyond Contracting: Operationalizing Specialty Risk

Importantly, the value of this infrastructure extends beyond contracting. Once contracts are finalized, the same logic and attribution framework can continue supporting operational workflows, outreach prioritization, engagement reporting, reconciliation, and ongoing performance management. Organizations can validate payor rosters, identify attribution discrepancies, prioritize patient engagement, and operationalize risk management relying on the same foundational model used during negotiations.

The Future of Specialty Value-Based Care Analytics

The future of specialty risk management will not be built on static dashboards or generalized VBC tooling alone. It will require highly configurable, deeply transparent analytical systems capable of modeling nuanced clinical and financial realities in real time.

And while oncology represents one of the clearest examples of this complexity, the underlying challenge is not unique to cancer care. Cardiology, nephrology, gastroenterology, neurology, and other specialty domains are all moving toward increasingly sophisticated risk arrangements where attribution precision, clinical nuance, and contract flexibility directly influence financial sustainability.

Specialty value-based care ultimately requires organizations to define reality with precision: who belongs in the population, what costs matter, which clinical signals drive attribution, and how those assumptions evolve across payor relationships. Organizations that can operationalize that complexity without losing transparency or trust will be best positioned to scale specialty risk successfully.

As specialty value-based care continues to evolve, the organizations that succeed will not just be those with more data, but those capable of operationalizing analytical precision at scale. That is the broader vision behind Ursa’s specialty value-based care infrastructure: not simply building contract models but creating adaptable analytical frameworks capable of supporting the full lifecycle of specialty risk arrangements in a transparent, scalable, and operationally actionable way.