Designed Wrong, Built Again: A Field Report on Stagnation and the Case for Structural Innovation

Executive Summary

Forty years of well-funded, well-intentioned effort has not produced durable, self-sustaining healthcare systems, in the developing world or in the United States. This piece argues that the persistent failure is not a knowledge problem or a funding problem. It is a structural design problem: the economic architecture underneath both global health aid and domestic US payment reform was never built to sustain the outcomes it was meant to produce. International aid programs that deliver real results collapse when donor enthusiasm fades.

US alternative payment models, from CMMI’s ACO initiatives to BPCI to Direct Contracting, produce promising pilots that never achieve scale because the dominant fee-for-service architecture absorbs and neutralizes them. Patients in rural Kenya and patients in rural Indiana face different problems with the same root cause: no one designed a business model that makes their care economically self-sustaining. The person who pays the price for that failure is not a payer, a policy architect, or a hospital CFO. It is the patient in the waiting room, and the physician who is no longer there to see them. The case for a genuinely new architecture built around aligned financial incentives, digital infrastructure, and new asset classes has never been stronger. Or more overdue.

How to Read This Piece

This piece is written through the lens of the patient, because at some point every reader becomes one. It covers two markets, the United States and the developing world, not as separate problems but as two expressions of the same structural failure. Act I examines how the US system failed the patient, through the lens of payment reform, commercial insurance, Medicare Advantage, and the physicians being driven out of medicine. Act II examines how the global aid system failed the patient, and introduces the physician who has spent 25 years inside both systems. Act III makes the case for what needs to change, why the moment is now, and what Snark Health is building to address each failure. Readers with a specific context, US healthcare, global health, or investment and policy, will find their entry point clearly marked. All roads lead to the same destination: a system that finally works for the patient and the physician.

Healthcare is the one thing every human being on earth will eventually need, and the one thing that virtually every system built to deliver it has managed to make expensive, inaccessible, or both. This is not for lack of trying. The past forty years have produced an extraordinary volume of effort: billions of dollars in international aid, decades of US federal payment reform, landmark legislation, Nobel-caliber research, and the sincere commitment of thousands of intelligent people working inside institutions that genuinely wanted to do better. The outcomes have been, with notable exceptions, underwhelming.

The standard explanation is that healthcare is complicated, that the problems are uniquely resistant to solution, that progress is necessarily slow, that the next pilot program or policy initiative will finally crack what the last one could not. This piece argues something different: that the recurring failure is not a knowledge problem, a funding problem, or an execution problem. It is a design problem. The economic architecture underneath both global health aid and domestic US healthcare was never built to sustain the outcomes it was meant to produce, and no amount of institutional goodwill, government spending, or philanthropic capital fixes a broken architecture. It only delays the reckoning.

What follows is a field report on that failure, across two markets that look nothing alike on the surface and share the same structural wound underneath. And, at the end, a case for why the window to do something genuinely different is open right now, and what it would actually take to walk through it.

Act I: How the US System Failed the Patient

Value-Based Healthcare: Promising in Theory Since Approximately Always
Origins: The 1980s Fee-for-Service Problem

American healthcare in the 1980s operated on a model that rewarded volume with almost sublime indifference to outcomes. Hospitals billed for procedures. Physicians billed for visits. Insurers paid, passed costs to employers, employers passed costs to employees, and everyone marveled at how expensive everything was becoming while doing approximately nothing structurally different.

The academic critique was well-developed by the mid-1980s. Michael Porter would eventually put a name to the alternative, "value-based healthcare," but the diagnosis was familiar long before the branding: the system paid for inputs, not outputs, and predictably produced expensive inputs with inconsistent outputs.

The 1990s–2000s: Managed Care, Then Managed Retreat

The managed care movement of the 1990s was the first serious structural attempt to change this, and it managed to make nearly everyone miserable enough to abandon it without quite solving the underlying problem. HMOs became synonymous with denied claims and restricted networks. The political backlash was swift and thorough.

What followed was a decade of relatively modest tinkering: disease management programs, pay-for-performance pilots, episode-of-care bundling experiments. The ACA in 2010 embedded value-based payment language throughout its architecture and created Accountable Care Organizations with genuine enthusiasm.

Policymakers described a future where providers would be paid for keeping populations healthy rather than treating them when sick.

The 2010s–Present: Pilots All the Way Down

What followed was a proliferation of pilots. The Center for Medicare and Medicaid Innovation (CMMI) launched model after model: bundled payments for joint replacements, primary care transformation initiatives, direct contracting experiments. Results were mixed, timelines were long, and the models that showed promise were often modified, suspended, or quietly wound down before achieving meaningful scale.

Meanwhile, the commercial insurance market continued paying largely on volume. Hospitals that had invested in value-based infrastructure found themselves straddling two incompatible economic logics simultaneously. Alternative payment models remained a minority of overall healthcare spend. The dominant fee-for-service architecture, despite universal agreement that it was wrong, proved extraordinarily durable.

The honest summary is this: thirty-plus years after the problem was clearly identified, the majority of healthcare transactions in the United States still reward the wrong things. The field produced impressive journals, influential conferences, and a generation of consultants. It produced less in the way of a fundamentally different business model.

Value-based care has been the future of healthcare for so long that it may hold the record for the longest-running pilot program in American industrial history.

The US Parallel: Good Institutions, Broken Architecture

It would be a mistake to frame the structural problem as one that belongs exclusively to the developing world. The United States has spent the better part of three decades attempting to engineer a more rational healthcare system from within, and the results tell a strikingly similar story. Different context, different institutions, same architecture failure.

The Center for Medicare and Medicaid Innovation (CMMI) was created in 2010 with explicit congressional authority and a $10 billion mandate to test payment and delivery models that would reduce costs and improve quality. It has launched dozens of models since, including Accountable Care Organizations under the Medicare Shared Savings Program (MSSP), Bundled Payments for Care Improvement (BPCI), Direct Contracting, Primary Care First, and more. The intent was serious. The institutional backing was real. The results have been, at best, modest and uneven.

The MSSP has produced ACOs that generate modest shared savings in some years and modest losses in others, with the majority of participating organizations never advancing beyond the lowest-risk tracks. BPCI generated genuine pockets of efficiency, including Dr. Selke’s program, but participation remained voluntary, savings were often captured by large health systems with existing infrastructure advantages, and the model never achieved the scale or permanence its architects envisioned. Direct Contracting, the most ambitious attempt to move primary care toward full capitation, was redesigned mid-implementation after political pressure and ultimately rebranded as the ACO REACH model with a narrowed scope.

Even the most celebrated integrated delivery systems, Kaiser Permanente, Geisinger, and Intermountain, have demonstrated that the value-based model can work within carefully controlled environments, and have spent decades failing to export that model anywhere else. Their success depends on vertical integration, captive patient populations, and geographic concentration. It does not travel. It does not scale to the fragmented, fee-for-service-dominated markets where the majority of Americans actually receive care.

The diagnosis is the same on both sides of the Atlantic and on both sides of the income divide. In Sub-Saharan Africa, care systems break down because donor capital is finite and local economic engines to sustain delivery were never designed. In the United States, care systems fail to reform because the dominant fee-for-service architecture is so deeply embedded in the financial incentives of hospitals, insurers, and physician groups that voluntary alternative payment models remain perpetually at the margin.

Patients in rural Kenya and patients in rural Indiana face structurally different problems that share the same root: no one built a business model that makes their care economically self-sustaining.

The problem is not that well-intentioned institutions lack the data, the clinical knowledge, or the policy tools. CMMI had all three. So did the Gates Foundation. So did Kaiser. The problem is that none of them were designed to replace the underlying economic architecture. They were designed only to work around it. That distinction is everything.

The Commercial Insurance Market: Optimized for Everything Except Care

The commercial insurance market controls the majority of non-elderly healthcare spending in the United States, covers roughly 154 million Americans through employer-sponsored plans, and has spent the last four decades demonstrating that it is extraordinarily good at one thing: generating revenue for insurers. Its relationship with the underlying goal of that revenue, affordable access to high-quality care, is considerably more ambiguous.

The structural logic of commercial insurance is not, at its core, designed to deliver healthcare. It is designed to price and transfer risk. An insurer that successfully enrolls a healthy population, limits its exposure through network design and prior authorization, and exits unprofitable markets has done exactly what its business model requires. That this behavior is often directly at odds with patient access, physician sustainability, and population health outcomes is not a failure of the system. It is the system working as designed. Administrative overhead in the US commercial insurance market consumes an estimated 25–35 cents of every premium dollar, a figure with no defensible clinical justification. What it does have is a business justification: utilization management, claims adjudication, network negotiation, and the apparatus of prior authorization exist not to improve care but to manage cost exposure on behalf of the insurer. The patient and the physician absorb the friction. The insurer captures the margin.

Prior authorization, the requirement that physicians obtain insurer approval before delivering care, is perhaps the most visible symptom of a system that has inverted its own purpose.

The American Medical Association’s annual surveys document the consequences consistently: physicians and their staff spend an average of 14 hours per week managing prior authorization requests. More than a third of physicians report that prior authorization has led directly to a serious adverse event for a patient in their care. Approval rates for initial requests run high, suggesting that the primary function of the process is not clinical gatekeeping but delay, attrition, and administrative burden sufficient to reduce utilization without the legal exposure of outright denial.

For independent and small-practice physicians, precisely the doctors most likely to serve rural, underinsured, and lower-income populations, the administrative cost of navigating commercial insurance requirements is an existential pressure that has accelerated the very consolidation and access reduction the system nominally exists to prevent.

Medicare Advantage: The Hybrid That Borrowed the Worst of Both Worlds

Medicare Advantage was, in theory, exactly the structural innovation reformers had been asking for: private-sector efficiency applied to a public-sector population, with capitated payment intended to incentivize prevention and value rather than volume. Enrollment now covers more than half of all Medicare beneficiaries. What it has also delivered is a case study in how misaligned financial incentives at the design level cannot be corrected by good intentions at the operational level.

Medicare Advantage plans are paid a risk-adjusted capitated rate by CMS based on predicted health costs. The logical consequence, documented by the HHS Office of Inspector General and the Medicare Payment Advisory Commission, has been systematic upcoding of patient diagnoses to inflate risk scores and capitation payments, without corresponding improvement in care. Overpayments to Medicare Advantage plans have been projected to reach $76 billion in 2026.

Meanwhile, the tools commercial insurers use in employer-sponsored lines have migrated wholesale into Medicare Advantage: prior authorization rates in MA plans exceed those in traditional Medicare, network restrictions limit specialist access, and mid-year formulary changes create medication disruptions for chronically ill populations. The promise of the managed care model, that capitation would drive prevention and efficiency, has, at scale, largely produced risk score optimization applied to a population that was sold the program as an enhancement of their Medicare benefit.

Between 2000 and 2024, average employer-sponsored family health insurance premiums grew about 350% from approximately $6,400 to over $25,000 annually, nearly fourfold, significantly outpacing both wage growth and general inflation. The result is a category actuaries call the “underinsured”: Americans who technically have coverage but face cost-sharing obligations so substantial that they delay or forgo care in ways that are clinically and financially indistinguishable from being uninsured. The underinsured population in the United States is, by multiple survey estimates, larger than the uninsured population. Coverage has become a product that is nominally present and functionally absent for a material fraction of the population it is supposed to protect.

Medicare Advantage promised to reconcile private-sector efficiency with public-sector obligation. What it delivered, at scale, was private-sector financial engineering applied to a public-sector population, with the government absorbing the overpayment and the patient absorbing the access restrictions. The commercial insurance market did not fail to reform. It was never designed to reform. Optimization for shareholder return and optimization for patient outcomes are not the same objective, and when they conflict, the business model has a clear answer.

What This System Does to Doctors, and to the Patients Who Need Them

The following reflects the perspective of Dr. Hank Selke, Co-Founder and CEO of Snark Health, writing as a practicing internist with over 25 years of direct patient care experience in the United States and Kenya.

When a physician leaves medicine, the immediate story is an economic one: the math stopped working, the administrative burden became unbearable, the reimbursement no longer justified the overhead.

But the downstream story is a patient safety one. Every physician who exits the system is a panel of patients who will wait longer, travel farther, or forgo care entirely. The two failures are not sequential. They are simultaneous. The system that has made independent practice economically irrational has also made access to a physician structurally unreliable for the patients that physician would have served. You cannot separate the doctor’s problem from the patient’s problem. They are the same problem.

I have watched the same conversation happen dozens of times over two decades. A physician, talented, committed, trained at significant personal cost and sacrifice, sits across from me and explains why they are leaving medicine. Not because they stopped caring about patients. Because the system has made caring for patients economically irrational. The administrative burden has consumed the hours that used to go to clinical work. The prior authorization queue has replaced the relationship with the patient as the primary demand on their attention.

The payment arrived, eventually, partially, after appeals, while the accounts receivable aged and the overhead compounded. They did the math. It stopped working.

Physicians in the United States are leaving medicine at rates that should alarm anyone who depends on the system, and that means everyone. Burnout surveys consistently find that more than half of practicing physicians report symptoms of professional burnout, with administrative burden identified as the primary driver in virtually every study.

The average physician spends more time on administrative tasks than on direct patient care. A significant and growing fraction of that time is spent on prior authorization alone, a process that, as documented above, exists primarily to delay and attrite utilization rather than to protect clinical quality.

The consolidation pressure compounds this. Independent practice has become economically precarious for a growing share of physicians, not because independent physicians deliver worse care (the evidence strongly suggests the opposite) but because the administrative infrastructure required to survive in the current billing and insurance environment is prohibitively expensive at the individual practice level. Physicians join large health systems not because they want to be employed but because they cannot afford the overhead of independence. CMS itself has recognized this dynamic and has articulated a strategic interest in reversing it: the evidence base shows that independent physicians deliver higher quality, lower cost care than their employed counterparts.

The system’s own data condemns its own structural incentives.
I have also watched this from the other side of the world. In rural Kenya, the physician shortage is not a product of administrative burden driving doctors out of practice. It is a product of a financing system that never built the economic infrastructure to pay physicians reliably in the first place. The structural problem looks different. The outcome is the same: patients who need care cannot access a physician, because the system that is supposed to connect them has failed at the level of basic economics.

Whether the failure mechanism is a prior authorization queue in Indiana or the absence of a payment rail in Kabula, the result for the patient is identical. The doctor is not there.

The system has been extraordinarily effective at making the delivery of care expensive for everyone except the intermediaries. Patients pay more and receive less. Physicians earn less and work harder on tasks that have nothing to do with medicine. Governments spend more and get worse outcomes. The only party for whom the current architecture is working is the one that was never supposed to be the point.

Act II: How the Global System Failed the Patient

International Aid: A Monument to Good Intentions and Modest Results

The 1980s: The Golden Age of Optimism

The 1980s were a remarkable time for international aid, remarkable, that is, in how confidently the field believed it had figured things out. Structural adjustment programs prescribed by the IMF and World Bank promised to unlock growth across Sub-Saharan Africa, South Asia, and Latin America. Billions flowed in the form of conditional loans, food packages, and technical advisors from wealthy countries who were absolutely certain they understood the problems of places they had visited briefly on organized tours.

The era produced an industry: development economists, NGO professionals, donor bureaucracies, and a cottage market in earnest white papers. The metrics were simple: dollars disbursed, clinics built, vaccines delivered. Whether any of this actually improved lives in a durable way was a question politely deferred to the next funding cycle.

The 1990s–2000s: Complexity Acknowledged, Behavior Unchanged

By the 1990s, the field had absorbed enough failure to get more sophisticated about it. The "Washington Consensus" fell out of fashion. "Ownership" and "sustainability" became buzzwords. There were summits, many summits. The Millennium Development Goals gave everyone a shared to-do list and the satisfying feeling that naming a problem was meaningfully close to solving it.

HIV/AIDS catalyzed a genuine mobilization: PEPFAR, the Global Fund, and bilateral commitments delivered antiretroviral therapy to millions who would otherwise have died. This is not nothing. But even here, the delivery model remained stubbornly traditional: international experts, vertical programs, and infrastructure that depended entirely on continued donor enthusiasm rather than any self-sustaining local economy of care.

The prevailing assumption, rarely stated aloud, was that aid would flow indefinitely, and that recipient countries needed not business models but benefactors.

The 2010s–Present: More Data, Same Architecture

The last decade brought randomized controlled trials to development economics, a genuinely useful corrective. We now have rigorous evidence about what does and does not work at the intervention level: the efficacy of bed nets, cash transfers, specific training programs. What remained largely unchanged is the underlying structure: wealthy-country capital funding services in low-income countries, mediated by a layer of organizations whose primary competitive advantage is access to that capital.

Digitization arrived and was celebrated. Mobile money, telemedicine pilots, community health worker apps, all promising, all funded by grants, all scaling approximately as fast as donor attention spans allowed. The question of how these innovations would survive without philanthropic subsidy was largely tabled, year after year, in favor of publishing the results and applying for the next round.

The aid sector's most durable innovation over forty years may be the PowerPoint deck summarizing why the last intervention didn't scale.

Shared Goals, Different Paths: Snark Health and the Gates Foundation

Organizations like the Bill & Melinda Gates Foundation have invested substantially in global health access, and the alignment on ultimate objectives is real. The Gates Foundation's stated mission, improving lives by expanding access to health and reducing the burden of preventable disease and death, shares the same north star as Snark Health: ensuring that fewer people face bankruptcy, preventable deterioration, or premature death because they could not access affordable, quality care.

This is not a small or trivial point of agreement. It represents a genuine convergence on what matters.

Where the paths diverge is in the mechanism. The Gates Foundation pursues this vision through philanthropic capital, donor mobilization, and the leverage of grant-making to catalyze systems change. It is an approach that has delivered real results and will continue to do so, within the structural constraints of donor-dependent funding.

Snark Health is building something different: a business model designed to make affordable, high-quality care structurally sustainable without requiring philanthropic subsidy as a permanent condition. The goal is not to criticize the philanthropic model, which serves a genuine function, but to demonstrate that care access can be engineered into economic viability rather than permanently dependent on the generosity of foundations and governments.

One model requires continuous donor enthusiasm. The other requires a working business. Snark Health is betting on the latter being more durable.

Act III: The Physician Who Has Lived Both Failures

A Principal Investigator Who Has Actually Done the Work

It would be easy, at this point in the argument, to dismiss any proposed solution as yet another group of optimists with a deck and a theory of change. Dr. Hank Selke brings something more specific to the table. He is a board-certified Internal Medicine and Infectious Disease physician with 25+ years of frontline clinical experience in both the United States and Kenya.

In 1998, he established a non-profit organization in Kabula, a rural village in Kenya, focusing on health, education, and community infrastructure.

The organization is still active today, and its school now ranks as the top-performing in the region. Through his Volunteer Kenya initiative, more than 1,000 volunteers from 21 countries were recruited to participate, contributing over $3 million USD toward AIDS education, healthcare, and primary school construction in the local community. This is not a technology team that has identified East Africa as a market opportunity. Dr. Selke has been part of these communities for a quarter century.

Beyond Kenya, Dr. Selke led the Bundled Payments for Care Improvement (BPCI) initiative across 12 hospital medicine programs and 170 physicians in Northern California, where his physician organization assumed $2 billion in risk on cost of care directly with CMS. Under his leadership, readmission rates were reduced from 13% to 7%, post-acute utilization was reduced by 14%, and sepsis core measure compliance improved from below 40% to between 82% and 100%. That combination of frontline global health credibility and large-scale federal payment reform experience is not something that can be manufactured. It is the foundation on which Snark Health is built.

Dr. Selke holds a Master of Medical Management from the USC Marshall School of Business, a credential built on the conviction that the intersection of medicine, finance, and systems design is where the most consequential healthcare problems get solved. His subsequent leadership as Medical Director of Advisory Services for a national physician organization, overseeing inpatient utilization management and developing the standardized educational curriculum and process improvement frameworks deployed to Medical Directors nationally, gives him a rare vantage point: he has not only practiced inside the broken system, he has built the management infrastructure that runs inside it.

That conviction extends beyond the theoretical. In March 2025, through a partnership with GivePower, solar equipment arrived at the Kabula farm to begin buildout of a solar-powered electricity system for the clinic and school Dr. Selke founded in 1998. The project pairs solar generation with Bitcoin mining to improve unit economics at the last mile, with mining revenue controlled and managed locally by the community in Kabula, not directed by an outside donor or foundation. The clinic has served the community since 2002. The goal now is to make it financially self-sustaining on its own terms. If the model performs as designed, it is replicable and scalable across similar last-mile settings.

This is not a thought experiment about decentralized infrastructure replacing donor dependency. It is already under construction.

As Principal Investigator and lead author on a prospective cluster randomized controlled trial published in the Journal of Acquired Immune Deficiency Syndromes (2010), Dr. Selke demonstrated that community-based HIV care, delivered not by physicians but by people living with HIV/AIDS themselves, could produce clinical outcomes equivalent to standard clinic-based care, at half the rate of clinic visits. The study, conducted in rural western Kenya, enrolled HIV-positive adults stable on antiretroviral therapy and randomized them between a community-delivery model supported by Personal Digital Assistants and standard monthly clinic visits.

The findings were unambiguous: no significant differences in viral load, CD4 counts, opportunistic infections, or ART regimen stability. The intervention group simply needed to show up to a clinic half as often, which increased capacity to serve more patients in a resource constrained environment. The paper, "Task-shifting of antiretroviral delivery from health care workers to persons living with HIV/AIDS: clinical outcomes of a community-based program in Kenya," has since been cited more than 200 times, an indication that the field found it meaningful enough to build on.

What Dr. Selke's research demonstrated is exactly what the aid sector spent decades failing to operationalize: that task-shifting with appropriate digital support can deliver equivalent clinical quality at lower cost in resource-constrained settings. The pilot worked. The question was always what came next.

His affiliation with Indiana University School of Medicine and the collaborative infrastructure around the Academic Model Providing Access to Healthcare (AMPATH) program in Kenya provided institutional rigor to what was, at its core, a fundamentally disruptive idea: that the most expensive person in the room does not need to be involved in every transaction.

What Actually Needs to Change: New Models, New Asset Classes

The Structural Problem Neither Field Has Solved

Both international aid and value-based healthcare have demonstrated something important by their persistent underperformance: the problem is not primarily one of knowledge or intent. We understand what good care looks like. We have evidence about what works. The problem is structural, and the economic architecture underneath existing systems does not reward the outcomes everyone claims to want.

Aid programs that build clinics cannot pay for them permanently. Value-based care models cannot thrive in markets dominated by fee-for-service contracting. Pilots remain pilots because the business model of scale was never designed, only the intervention itself.

New Business Models

The transition from "interesting pilot" to "durable system" requires building economic structures that align revenue with outcomes. This is not a theoretical proposition. It is a design challenge. What does a care delivery organization look like when it earns money by keeping people healthy and out of expensive settings, rather than by treating them expensively when they arrive sick?

The answer almost certainly involves departing from the assumption that healthcare organizations should look like 20th-century institutions funded by 20th-century mechanisms. It requires building for sustainability from the first day of design rather than hoping that a successful pilot will attract sufficient ongoing philanthropy.

New Asset Classes: Data and Digital Assets

Perhaps more consequentially, the next generation of sustainable care systems will be built on asset classes that did not exist when the current infrastructure was designed. Structured clinical data, collected at scale, longitudinally, and across populations, has genuine economic value that has barely begun to be realized through legitimate, patient-centered models. Digital assets and tokenized incentive structures offer mechanisms for aligning the interests of patients, doctors, and payers in ways that traditional contracts cannot.

These are not speculative additions to otherwise conventional business models. They are the foundation of what a genuinely new architecture looks like, one where the data generated by delivering care creates economic returns that fund more care, where digital infrastructure reduces marginal cost as it scales, and where financial incentives are structurally pointed in the right direction rather than requiring constant regulatory pressure to nudge them there.

Forty years of evidence suggests that good intentions, smart people, and adequate funding are necessary but not sufficient conditions for durable health systems. What has been missing is the business model, and the new asset classes that make it viable.

The Window Is Open

Structural problems rarely have windows. The architecture of a failing system tends to persist long past the point at which its failure is obvious, because too many participants have adapted their economics to the dysfunction. Replacing it requires not just a better design but a moment when the conditions for change converge: a policy shift that disrupts the incumbent equilibrium, a technology infrastructure that makes the alternative viable, and a market with unmet demand large enough to reward a new entrant at scale. Those conditions are converging now, simultaneously, on both sides of the problem this piece has documented.

The Policy Shift: OBBBA and the Largest HSA Expansion in History

Effective January 1, 2026, the One Big Beautiful Bill Act dramatically expanded Health Savings Account eligibility in ways that directly address the structural failures documented throughout this piece. All ACA Bronze and Catastrophic plan holders, precisely the population most likely to be functionally underinsured despite nominally having coverage, can now open HSAs. More consequentially, Direct Primary Care monthly membership fees of up to $150 per individual and $300 per family per month are now payable directly from HSA funds.

This is not an incremental adjustment. It is the largest expansion of the HSA addressable market since the instrument was created, and it creates, for the first time, a financially viable, pre-tax pathway for patients to fund a DPC membership that routes their primary care entirely outside the commercial insurance network. The HSA market currently holds over $159 billion in assets under management. The OBBBA expansion materially increases the population eligible to participate in that market, and creates a new category of HSA-eligible expenditure, DPC, that directly addresses the physician consolidation and access problems documented above. A patient with an HSA-funded DPC membership, a high-deductible plan for catastrophic coverage, and a payment platform that enforces IRS compliance automatically has a viable alternative to the commercial insurance model. That alternative did not exist at scale before January 1, 2026.

The Technology Infrastructure: Solana, Stablecoins, and Settlement at Scale

The second condition, technology infrastructure, has also reached the maturity threshold required for this application. Solana, the blockchain network on which Snark Health’s payment architecture is built, now processes thousands of transactions per second at costs measured in fractions of a cent, with settlement finality in under a second. This is not speculative capacity. It is production infrastructure operating at scale, capable of handling healthcare payment volume without the latency, cost, or opacity of legacy banking rails.

USD-pegged stablecoins have crossed the threshold from experimental instrument to functioning payment infrastructure. Hippocratic Coin® USD (HTCUSD), Snark Health’s purpose-built healthcare stablecoin on Solana, enables physician payment in under 60 seconds, replacing the days-to-weeks settlement timeline of legacy HSA banks with instant, on-chain finality and an immutable, IRS-ready audit trail. Every transaction is anchored to Solana mainnet with multi-signature release, creating the compliance record that traditional HSA banks produce with paper receipts and manual review. The infrastructure that makes 60-second physician payment possible, IRS-compliant, and auditable did not exist at the necessary maturity level five years ago. It does now.

The Emerging Market Infrastructure: M-Pesa and the Mobile Money Foundation

On the emerging markets side, the technology infrastructure condition has also been met, and has been for longer than most Western observers appreciate. M-Pesa, the mobile money platform that launched in Kenya in 2007, now serves over 51 million users across East Africa. It has demonstrated, at continental scale, that populations without access to traditional banking infrastructure can transact digitally, reliably, and at low cost using mobile phones. The financial inclusion problem that aid programs spent decades trying to solve through banking sector development has been substantially addressed by a mobile payment layer that operates entirely outside the legacy banking system.

HTCUSD, denominated in USD and settled on Solana, is designed to sit on top of that mobile money foundation, interoperable with M-Pesa rails, stable in value relative to the currency in which healthcare is priced, and not dependent on the donor enthusiasm or government budget cycles that have caused every prior attempt at sustainable emerging market health financing to eventually stall. The infrastructure is ready. The question is whether anyone has built the right product on top of it. That is the question Snark Health is answering.

The Synthesis

Whether the lens is international aid or domestic value-based care, the diagnosis converges on the same conclusion: the sector needs real alternatives built around genuinely new economic structures. Not slightly modified versions of what failed. Not pilots that assume perpetual subsidy. Not payment reform that tiptoes around the dominant fee-for-service architecture.

What is required is the construction of care systems that are economically self-reinforcing, where delivering better outcomes at lower cost generates the revenue that funds continued operation and growth. Data assets and digital infrastructure are not optional enhancements to this vision. They are the mechanism by which the unit economics of care access can be transformed at scale.

What Snark Health Is Building and Why It Addresses Each Failure

Snark Health is not proposing to tear out and replace the existing healthcare infrastructure. That approach has been tried, in various forms, by every reform initiative documented in this piece, and the incumbent architecture has absorbed and neutralized each one.

Instead, Snark Health layers on top of existing infrastructure, immediately improving efficiency and returning value to patients, physicians, and governmental payers, while simultaneously building the financial rails and data architecture that make genuinely new payment models possible over time. The platform creates the conditions for risk-based contracting where patients, physicians, and payers all have skin in the game and are incentivized toward optimal utilization.

That is a different design principle than anything that has been tried before at scale.

On the US side, the HSA administrative failure, where patients lose 20–30% of HSA value to manual review, paper documentation, and IRS compliance friction, is addressed by Snark Clinical Intelligence, a three-layer AI substantiation engine that reads receipts, matches CPT and ICD-10 codes against the full IRS Publication 502 rule base at 99%+ confidence, and releases payment via HTCUSD directly to the physician. The prior authorization analog in HSA compliance disappears. The administrative burden that makes independent practice economically unviable shrinks. Physicians practicing under a Direct Primary Care model, now HSA-eligible under the OBBBA, receive payment in under 60 seconds with an immutable on-chain audit trail that satisfies IRS requirements without a single paper receipt. The unit economics of independent practice change. The physician who was doing the math and finding it didn’t work finds a different answer.

The commercial insurance network problem, narrow networks that restrict access and extract rent from every transaction between patient and physician, is addressed by routing around it at the primary care level entirely. Snark Health’s platform makes out-of-network care cheaper and easier to access, while simultaneously creating the infrastructure to enforce accountability on narrow networks and force them to compete. By squeezing inefficient rent-seeking intermediaries and accelerating payment cycles, the platform makes underinsured patients viable to serve for independent physicians who currently cannot afford to see them. The addressable patient population expands. The physicians who need the revenue find it. The middleman who was extracting value from both is no longer necessary for the transaction to occur.

On Medicare Advantage, where CMS has spent billions subsidizing a private-sector utilization management apparatus that delivered risk score optimization instead of care improvement, Snark Health’s position is direct: we can replace that function. By assuming risk directly with CMS, applying AI-driven optimal utilization management rather than denial-based cost reduction (a distinction Dr. Selke understands from having built utilization management infrastructure and trained Medical Directors across national health systems), and settling transactions on-chain with full transparency, Snark Health can deliver lower costs for the government, lower costs for patients, and better earnings for physicians simultaneously. This is not a marginal improvement on the Medicare Advantage model. It is a structurally different model that does not require financial engineering to generate its returns, only better-aligned incentives and infrastructure capable of enforcing them.

In emerging markets, Hippocratic Coin® (HTC) addresses the donor-dependency problem at its root. Rather than requiring philanthropic capital to fund care delivery indefinitely, HTC functions as a limited-supply, healthcare-denominated store of value, an asset whose value is underpinned by demand for the one commodity no population can opt out of consuming.

Patients in East Africa can save for future healthcare expenses in an asset that holds value and is purpose-built for the transaction they need to make. Governments can hold HTC’s sovereign reserve tranche, 25% of total supply designed to be sold directly to national governments, as a supplementary hedge against healthcare cost volatility and currency risk, alongside traditional reserve instruments. The care economy becomes self-sustaining not because a foundation decided to fund it this year, but because the asset structure aligns the financial interests of everyone in the system with the delivery of care.

What Snark Health Addresses: The Structural Mapping

HSA administrative failure: Snark Clinical Intelligence AI engine enforces IRS Pub. 502 compliance at 99%+ confidence, eliminating the manual review and documentation friction that costs patients 20–30% of HSA value.

Physician payment delays and income erosion: HTCUSD stablecoin settlement on Solana pays physicians in under 60 seconds with an immutable on-chain audit trail, replacing days-to-weeks legacy bank timelines and eliminating accounts receivable drag.

Physician consolidation pressure and practice viability: Platform reduces administrative burden to the point where independent DPC practice is economically viable. Physicians can start alongside existing practice and migrate progressively, aligned with CMS’s own strategy of increasing independent physician participation.

Fee-for-service dominance and network restrictions: HSA + DPC combination routes around both at the primary care level. Snark is the financial infrastructure that makes that routing work at scale, while building the data rails for future risk-based contracting where patients, physicians, and payers all have aligned incentives.

Medicare Advantage financial engineering and CMS overpayment: Snark Health can assume risk directly with CMS, delivering lower costs for government, lower costs for patients, and better physician earnings through AI-driven optimal utilization, not denial-based cost reduction. A structurally different model, not a marginally improved one.

Donor-dependent emerging market care and government healthcare cost volatility: HTC (limited-supply healthcare store of value) and HTCUSD (USD-pegged stablecoin) create a self-sustaining financial infrastructure for care in emerging markets. The sovereign reserve tranche allows national governments to hedge healthcare cost volatility with a purpose-built asset. The system does not require continued philanthropic subsidy to function.

The clinical knowledge exists. The evidence base is robust. The technology is available and production-ready. The policy window opened January 1, 2026. What has been missing is the organizational will and the credibility to build a business model worthy of the problem. In the next piece, we will introduce Snark Health in full: the first AI-native HSA trustee, the physician payment infrastructure, the DPC integration, Hippocratic Coin®, and the path to assuming risk directly with CMS. No legacy bank can replicate this without rebuilding from scratch. No pure tech company has our clinical credibility or regulatory patience. No health system has our financial architecture. And no one else has spent 25 years in the communities this platform is built to serve.

Resources

Learn more about Snark Health:
https://linktr.ee/snarkhealth

Join the waitlist for early access to Snark Health App:
https://snarkhealth.com/waitlist

Follow us on X(Twitter):
https://x.com/snarkhealth

Join our Discord Community:
https://discord.com/invite/V7psjrGwmC

References

Reference List Peer-Reviewed Research

1. Selke HM, Kimaiyo S, Sidle JE, et al. Task-shifting of antiretroviral delivery from health care workers to persons living with HIV/AIDS: clinical outcomes of a community-based program in Kenya. J Acquir Immune Defic Syndr. 2010;55(4):483-490.
2. Himmelstein DU, Campbell T, Woolhandler S. Health care administrative costs in the United States and Canada, 2017. JAMA. 2020;323(6):569-571. doi:10.1001/jama.2019.19714
3. Sinsky C, Colligan L, Li L, et al. Allocation of physician time in ambulatory practice: a time and motion study in 4 specialties. Ann Intern Med. 2016;165(11):753-760. doi:10.7326/M16-0961
4. Furukawa MF, Kimmey L, Jones DJ, Machta RM, Guo J, Rich EC. Consolidation of providers into health systems increased substantially, 2016-18. Health Aff (Millwood). 2020;39(8):1321-1325. doi:10.1377/hlthaff.2020.00017

Government and Regulatory Sources

5. Centers for Medicare & Medicaid Services. Medicare Advantage and Part D Contract and Enrollment Data. Baltimore, MD: CMS; 2024. https://www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and-Reports/MCRAdvPartDEnrolData
6. Medicare Payment Advisory Commission. Report to the Congress: Medicare Payment Policy. Washington, DC: MedPAC; March 2024.
7. US Department of Health and Human Services, Office of Inspector General. Medicare Advantage: Questionable Use of Health Risk Assessments Associated with Diagnoses That Increased Payments to Plans. Washington, DC: HHS OIG; 2021. OEI-03-17-00471.
8. Patient Protection and Affordable Care Act, Pub L No. 111-148, §3021, 124 Stat 119 (2010).
9. One Big Beautiful Bill Act of 2025, Pub L No. [confirm bill number]. §[confirm HSA section]. (2025).

Survey and Annual Report Sources

10. American Medical Association. 2023 AMA Prior Authorization Physician Survey. Chicago, IL: AMA; 2024.
11. Medscape. Medscape Physician Burnout & Depression Report 2024. New York, NY: WebMD Health; 2024. https://www.medscape.com/slideshow/2024-lifestyle-burnout-6016865
12. Kaiser Family Foundation. Employer Health Benefits Survey 2024. San Francisco, CA: KFF; 2024. https://www.kff.org/health-costs/report/2024-employer-health-benefits-survey/
13. Collins SR, Radley DC, Baumgartner JC, Bhupal HK. Trends in Private Insurance, Medicare, Medicaid, and the ACA, 2008-2020. New York, NY: Commonwealth Fund; 2021.
14. DEVIR America; HSA Council. 2024 Year-End Devir America HSA Market Statistics. Brookfield, WI: DEVIR; 2025.

Institutional and Program Sources

15. Academic Model Providing Access to Healthcare (AMPATH). Program Overview and Research Publications. Indianapolis, IN: Indiana University School of Medicine; 2024. https://www.ampathkenya.org/
16. GivePower Foundation. Program Documentation: Kabula Community Solar Initiative. San Francisco, CA: GivePower; 2025.
17. Safaricom PLC. Annual Report and Financial Statements 2024. Nairobi, Kenya: Safaricom; 2024. https://www.safaricom.co.ke/investor-relations/annual-reports