The Rise and Fall of HMSA Payment Transformation

HHTF Staff · 2026-05-27

Additional data since publication

HHIP holds no data from the Hawaiʻi All-Payer Claims Database. References to it in this piece are corrected below.

HMSA’s Payment Transformation was Hawaiʻi’s major primary care payment experiment from roughly 2016 to 2026. It began with an appealing promise: move away from fee-for-service, pay for value, support population health, and give primary care physicians some flexibility to care for patients outside the narrow limits of office visits. By 2019, the model had national attention. A JAMA article examined HMSA’s population-based primary care payment system and its association with quality measures in Hawaiʻi, and the implication was that Hawaiʻi might be showing the country where primary care payment should go.

That is not where the story ended. Payment Transformation should now be understood less as a model for the nation than as a warning. It tried to fix a real weakness in fee-for-service, but it did not build the primary care capacity needed to make the new model work. For many patients, access worsened. Emergency spending rose. Administrative burden increased. Quality metrics created incentives to avoid patients whose lives made good scores harder to achieve. Now HMSA is moving away from the model while affected practices are being pushed back toward fee-for-service, even as Hawaiʻi is being asked to consider the next generation of risk-based payment models.

The original insight behind Payment Transformation was not wrong. Fee-for-service has serious flaws. It often fails to pay for medication adjustment, lab review, chronic disease follow-up, phone calls, portal messages, care coordination, caregiver support, specialist triage, and hospital follow-up. Primary care should not be reduced to a series of office visits, because much of the work that keeps patients stable happens between visits. A payment model that only rewards face-to-face encounters can miss the work that prevents more expensive care later.

The problem is that replacing fee-for-service with a fixed monthly payment does not automatically create the staff, time, or infrastructure needed to manage a population. A payment model does not hire a nurse. It does not create a care manager, reopen a closed panel, make a specialist available on a neighbor island, or keep a rural physician in practice. Those things require actual investment in clinics.

By 2022, the physician experience was no longer unclear. A study in the Hawaiʻi Journal of Health & Social Welfare evaluated physician perspectives on Payment Transformation and found that 77 percent of respondents were unhappy with the program, while only 10.1 percent reported being happy. The same study found that 60.6 percent of respondents reported a decrease in overall income. In open-ended responses, 95.7 percent were negative. Reported themes included harm to primary care, increased administrative burden, decreased quality of patient care, decreased physician reimbursement, preference to treat healthier patients, harm to private practice, harm to newer practices, ignored physician concerns, and worsening of Hawaiʻi’s physician shortage.

Those findings matter because they show that physician concerns were not simply resistance to change. The argument against Payment Transformation was not that fee-for-service was perfect. It was that quality measures are not the same thing as access, and payment redesign is not the same thing as building primary care capacity. A model can improve selected measures while making clinics more administratively burdened, financially fragile, and less able to care for complex patients. The lesson from the 2019 national attention is not that the analysis was useless. It is that selected quality measures could not answer the larger question: could patients actually get primary care, and could clinics survive while providing it?

Some Payment Transformation payments were reportedly as low as about $19 per patient per month. At that level, the model could not reliably support the population-health work it expected primary care to perform. A clinic cannot hire staff, answer messages, review labs, coordinate care, manage complex patients, and keep panels open on payment that does not cover the work. The payment may have been labeled population health, but the work still had to be done by real people in real clinics.

That is the difference between primary care investment and financial risk transfer. Investment means dollars reach the clinic and build capacity. Risk transfer means practices are held accountable for a population without enough payment, staff, data, or control to care for that population well. Payment Transformation should be judged by which one it became.

The results should be judged by access, not by theory. Available HHIP review of all-payer summary data found that primary care claims were essentially flat from 2016 to 2023, while emergency claims rose 14.1 percent and combined paid amounts for primary care and emergency services rose 36.9 percent. (Retired: no source for these trend figures could be found.) The same review found that available HHIP data could not independently verify HMSA’s reported urgent care trend because urgent care was not a distinct category in the summary table. That limitation matters. If HMSA is using internal data to justify ending Payment Transformation or moving into a new model, it should release the definitions, baseline years, patient populations, codes, and trend tables behind its claims.

Even without the urgent care data, the available pattern raises the right question. If primary care contact was flat or falling while emergency use and spending rose, the system was not producing the promised value. It was allowing care to move into more expensive settings. That does not prove Payment Transformation caused the entire problem. Hawaiʻi already had primary care shortages, Medicaid underpayment, rural access barriers, and broken workforce incentives before HMSA launched this model. But it does mean Payment Transformation did not solve the problem it most needed to solve.

The administrative burden was not incidental. Payment Transformation required attribution, reporting, quality measures, denominator management, exception requests, and payer-specific platforms. Many clinics hired a full-time staff person to do nothing but Coreo, attribution, and metric compliance. That salary was not buying more appointments. It was the price of surviving the model. Some of the work may have been clinically useful, but it still consumed staff time, physician attention, and clinic resources.

The documents show the machinery. The Payment Transformation Review Request Form required providers seeking an exception to use a specific HMSA process, while denominator exclusions for performance measures were routed through Coreo. A quality measure tip sheet on review of chronic conditions describes staff pulling care gaps and chronic condition information from Coreo, then documenting or printing face sheets to remind providers. These are not abstract burdens. They are clinic workflows, staff time, and physician attention diverted into payer-specific compliance work.

Payment Transformation also created incentives to avoid difficult patients. I saw this directly. My practice absorbed families who could not get care because they did not vaccinate their children. It was not only unvaccinated children. It was also adults with serious mental illness, unstable lives, and medical noncompliance that was not simply a matter of choice. I also heard physicians discuss the financial risk of caring for patients whose blood pressure, diabetes, or preventive-care metrics could lower quality scores.

This was not mainly about greed. It was about survival. When payment is too low and quality metrics punish practices for caring for patients with complex lives, practices learn to protect their metrics. The patients most likely to lose access are often the ones who most need care. A payment model that rewards clean metrics without protecting access for complicated patients can make the system look better while making care harder to obtain.

Now HMSA is moving affected practices back toward fee-for-service. That does not solve the underlying problem. The Medicaid fee schedule is inadequate, and FQHCs do not have unlimited capacity to absorb every Medicaid patient in Hawaiʻi. Independent primary care clinics, pediatric practices, family physicians, internists, and rural practices must also be able to participate. If Medicaid payment is too low for those clinics to keep their doors open to Medicaid patients, the network shrinks.

The Medicaid example makes the math plain. If a basic Medicaid primary care visit pays roughly $35, then four visits in an hour generate about $140. That revenue has to support the clinician, medical assistant, front desk, billing, rent, malpractice coverage, EHR, supplies, prior authorization, lab follow-up, portal messages, and care coordination. The math does not work. A clinic cannot make up the difference indefinitely by working harder.

When Medicaid patients cannot find primary care, the system does not save money. It buys more expensive care later. Blood pressure goes unmanaged. Diabetes worsens. Asthma flares. Infections progress. Depression and substance use destabilize. Medications are stopped, duplicated, or taken incorrectly. Abnormal labs are not followed. Patients end up in urgent care, emergency departments, ambulances, hospitals, imaging suites, and specialist offices. That is not fiscal discipline. It is deferred spending with worse outcomes.

HMSA’s increased reimbursement for Neighbor Island providers is appropriate and appreciated. Rural and geographically isolated care is harder and more expensive to provide. But that increase should be the beginning of a broader correction, not the end. Primary care remains underpaid across Hawaiʻi, and rural physicians often provide a broader scope of care because fewer specialists and services are nearby.

The contract issue also belongs in this discussion because payment terms determine access. The PPO participating physician agreement appears to allow HMSA to adjust its Schedule of Maximum Allowable Charges with 60 days’ written notice and states that the determination of charges is at HMSA’s sole discretion after internal review. That may give HMSA broad contractual discretion, but it raises the central question for providers facing a July 1 transition: what exactly are the new terms, and when did meaningful notice begin?

A vague announcement that a transition is coming should not be treated as meaningful notice of the payment arrangement that will determine whether clinics can survive. Providers need the applicable fee schedule, effective dates, payment policies, treatment of between-visit work, appeal rights, termination rights, and any new participation requirements. Without those terms, practices cannot plan staffing, panels, patient communication, or financial viability.

This is why Hawaiʻi should not move from Payment Transformation into the next model without an honest evaluation. OneHealth, global budgets, and other risk-sharing arrangements may be presented as the next solution. But if they are built on the same mistake, they will repeat the same failure at a larger scale. Changing the name of the model will not fix a system that still refuses to pay enough for primary care capacity.

A better model would start with a Primary Care Capacity Test. It would ask whether the model keeps clinics open, improves appointment availability, expands open panels, reduces administrative burden, supports rural scope of practice, and makes complex patients sustainable to care for. It would ask whether the model keeps patients out of avoidable urgent care and emergency care. It would also ask whether primary care dollars reach frontline clinics, rather than being absorbed by administrative layers.

Payment Transformation rose because it promised to fix a real problem. It fell because changing payment is not the same as building capacity. Hawaiʻi should learn from that before accepting the next model. Pay for capacity before you cap spending.