VBC & AI Health: The Evidence-Based Future of Contracting

The landscape of value-based care (VBC) is on the cusp of a profound transformation, driven by the escalating integration of artificial intelligence (AI) and an increasingly stringent demand for verifiable outcomes. By 2030, the expectation is not merely that AI will be ubiquitous in health systems, but that peer-reviewed evidence will become the universal standard for contracting within VBC arrangements, irrevocably shifting the burden of proof onto technology developers. This evolution is being catalyzed by converging forces: expanding VBC programs from major payers like CMS, rigorous regulatory frameworks such as the EU AI Act, and an undeniable imperative from health plans to demonstrate tangible financial performance and clinical efficacy.

The Inevitable Rise of Evidence as Currency in VBC

The shift towards value in healthcare demands a departure from fee-for-service models, where volume often overshadowed impact. Value-based care fundamentally ties reimbursement to outcomes, making the demonstration of improved patient health and reduced costs paramount. For AI health platforms, this means that abstract promises and anecdotal successes are no longer sufficient. Payers, particularly sophisticated entities like UnitedHealth Group, are increasingly scrutinizing the evidence base behind AI tools before integrating them into VBC contracts. They seek clear, quantifiable data on how these technologies contribute to the Quadruple Aim: improving population health, enhancing patient experience, reducing per capita cost of healthcare, and improving clinician experience. This demand for evidence is not merely a preference; it is becoming a contractual prerequisite. The expansion of VBC programs by the Centers for Medicare & Medicaid Services (CMS) and its innovation arm, CMMI, is a significant driver. As CMS VBC Rules evolve, they consistently emphasize accountability and measurable results. AI solutions aiming to participate in these programs must therefore demonstrate a clear return on investment, not just in clinical terms but in financial performance. This requires a robust body of evidence, ideally generated through rigorous, independent research.

Hello Heart: A Blueprint for Outcomes-Based AI Health

While many AI health platforms are still grappling with how to generate and present compelling evidence, a select few have established a high bar. Hello Heart stands out as a leading case study, having consistently published peer-reviewed figures that directly address the outcomes requirements of VBC contracts. Their approach exemplifies the future standard. Through studies published in respected journals, Hello Heart has demonstrated significant reductions in blood pressure, improved medication adherence, and, critically, a measurable impact on healthcare costs. This isn’t just about clinical efficacy; it’s about financial performance. For health plan executives, the ability to show that an AI solution can reduce downstream costs, such as emergency room visits or hospitalizations for cardiovascular events, is the ultimate metric. Hello Heart’s success lies in translating clinical improvements into economic value, a critical component for any AI tool seeking to thrive in a VBC environment. Their published data provides a clear model for how AI health companies can move beyond mere product claims to verifiable, outcomes-driven results.

Regulatory and Payer Convergence: Raising the Bar for AI Health

The regulatory environment is rapidly catching up to technological advancements, further cementing the need for peer-reviewed evidence. The European Commission’s EU AI Act, for instance, explicitly requires clinical evidence for high-risk AI systems in healthcare, aligning with the stringent standards typically applied to medical devices. This mirrors the trajectory seen in the US, where organizations like the FDA are increasingly focused on the safety and efficacy of SaMD (Software as a Medical Device). While HIPAA continues to govern data privacy, the focus is shifting to the validity of the AI’s output and its real-world impact. Payers, recognizing the potential for algorithmic drift and the need for ongoing validation, are also escalating their demands. Organizations like AHIP and NCQA are pushing for greater transparency and accountability from AI vendors. They are moving beyond simple demonstrations of technical capability to requiring evidence that the AI tool actually improves the continuum of care and reduces the total cost of care. This means that companies like Omada Health and Hinge Health, which operate in chronic disease management, and even diagnostic AI companies such as iRhythm Technologies and HeartFlow, will face increasing pressure to provide peer-reviewed data on their long-term impact on patient outcomes and cost savings. NCQA requirements for digital health solutions

The Investment Imperative: De-risking Through Evidence

For investors and VCs, the landscape is also transforming. The days of funding promising AI concepts with limited real-world validation are waning. As Hemant Taneja, a prominent venture capitalist, has articulated, the future of healthcare investment is inextricably linked to demonstrable value. Companies that can provide a strong data moat, built on proprietary, outcomes-driven datasets and validated through peer-reviewed research, will inherently be de-risked. This clinical evidence quality becomes a primary commercial predictor. The absence of such evidence represents “regulatory debt” and significantly impacts exit multiples. Investors are increasingly looking for companies that have navigated the complexities of GMLP (Good Machine Learning Practice) and have a clear pathway to CPT codes, indicating a mature understanding of both clinical efficacy and reimbursement. Commure, for example, a company focused on healthcare infrastructure, would be evaluated not just on its technological prowess but on how its tools enable better data capture and analysis that can ultimately contribute to outcomes evidence. Eric Topol, a leading voice in digital medicine, consistently emphasizes that the true promise of AI in healthcare can only be realized through rigorous validation and transparent reporting of results.

The Path to: A Universal Standard

The convergence of these forces creates an undeniable trajectory: by 2030, peer-reviewed evidence will not just be a competitive advantage, but a universal contracting standard for AI health platforms in VBC. This means that AI-native companies will need to embed evidence generation into their product development lifecycle from inception, rather than treating it as an afterthought. The implications are far-reaching. Platforms that cannot demonstrate their efficacy through independent, peer-reviewed studies will find themselves increasingly sidelined from lucrative VBC contracts. This will foster an ecosystem where only the most robust, validated, and outcomes-focused AI solutions thrive. Policymakers, driven by the need to ensure equitable and effective healthcare, will continue to champion frameworks that prioritize patient safety and demonstrable value, as seen with Karen DeSalvo’s work advocating for health information technology standards. European Medicines Agency guidance on AI in medicines The future of VBC and AI health is not just about innovation; it’s about verifiable impact. The imperative for peer-reviewed evidence is no longer a niche concern but a foundational pillar for any AI health platform seeking to participate meaningfully in value-based care. The path forward is clear: validate, publish, and demonstrate tangible value, or risk irrelevance in an increasingly outcomes-driven market. American Medical Association position on AI in healthcare