Source basis includes the CMS ACCESS materials, Shiff/Sutton JAMA Perspective, Liao Health Affairs critique, Brooks et al. OCM evaluation, the accepted-applicant list, and the prior AI long-form analysis.
ACCESS, Chronic Care, and the CMMI Savings Paradox
When Better Care for Under-Managed Patients May Cost More
CMS’s ACCESS model is one of the most intellectually coherent CMMI models in recent years. It is also one of the most revealing. On its surface, ACCESS is a voluntary Medicare demonstration for technology-supported chronic care. It creates new payments for organizations that manage common chronic conditions such as hypertension, dyslipidemia, obesity, prediabetes, diabetes, chronic kidney disease, atherosclerotic cardiovascular disease, chronic musculoskeletal pain, depression, and anxiety.
That description is accurate, but too small. ACCESS is better understood as an attempt to build a new Medicare payment architecture for modern chronic care. It asks whether Medicare can pay for longitudinal, digital, remote, asynchronous, hybrid, and AI-enabled care without simply creating another fee-for-service volume engine.
That is the real policy problem. Chronic disease care increasingly involves software, devices, messaging, remote monitoring, patient coaching, medication management, behavioral support, and algorithmically assisted triage. These activities do not map neatly to traditional office visits. But if CMS responds by creating a separate code for every digital touch, remote transmission, coaching minute, app interaction, algorithmic review, or care-management task, it recreates the old problem in new clothing. The system will optimize for billable activity.
ACCESS tries to move the unit of payment away from activity and toward outcomes.
That is the model’s conceptual leap. CMS is not primarily paying for a visit, a device, a software license, a care-management minute, or a remote data stream. It is paying for measurable improvement or control in a defined chronic disease domain. The model’s central instrument is the Outcome-Aligned Payment: a recurring payment to a Medicare-enrolled care organization, with full payment tied to clinical results.
This is why ACCESS matters even if it remains modest in size. If it succeeds, it could become a template for how Medicare pays for digital chronic care, AI-enabled diagnostics, connected devices, hybrid virtual care, and chronic disease management outside the physician office. It could also become a reference architecture for Medicare Advantage, Medicaid managed care, commercial payers, and ACO-adjacent contracting.
But ACCESS also exposes a deeper problem in CMMI’s statutory design. CMMI models are expected to reduce Medicare spending without reducing quality, or improve quality without increasing spending. That sounds reasonable in the abstract. Yet in under-managed chronic disease, better care often means more care. More medication adherence. More appropriate drugs. More physical therapy. More behavioral health care. More labs. More monitoring. More specialist input. More follow-up. More patient support.
A poorly managed patient can look cheap only because the system has been failing quietly. If ACCESS finds the failure and fixes it, spending may rise.
That is not a technical flaw at the edge of the model. It is the core paradox.
The Problem ACCESS Is Trying to Solve
Original Medicare has never had a clean payment pathway for much of modern chronic care. Fee-for-service Medicare pays for defined services: visits, procedures, tests, drugs, devices, and time-based care-management activities. That works reasonably well when the clinical service is discrete and easily coded. It works poorly when the relevant care is continuous, distributed, digital, behavioral, remote, or team-based.
Chronic disease is not organized around fifteen-minute increments. A patient with hypertension, diabetes, CKD, chronic pain, or depression does not need only episodic appointments. The patient needs ongoing engagement, monitoring, medication adjustment, adherence support, coaching, escalation, de-escalation, and coordination across clinicians. Much of that work happens between visits. Much of it may be performed by nurses, pharmacists, behavioral health clinicians, coaches, care navigators, software, devices, and algorithms. The physician office visit remains important, but it is no longer the natural unit of care.
CMS has tried to patch this mismatch with codes for chronic care management, remote physiologic monitoring, remote therapeutic monitoring, telehealth, and related services. These codes are useful, but they also reveal the problem. Once CMS defines and prices an activity, the system learns to produce the activity. Time thresholds, documentation rules, device transmissions, monthly billing periods, and staff minutes become the commercial object. A model designed to support better care can become a model for generating more billable units.
The Shiff/Sutton JAMA Perspective makes this point unusually clearly. Technology in health care is often directed toward documentation, billing optimization, and shifting clinician time into reimbursable categories. That is not because technology is inherently low value. It is because the payment system rewards monetizable activity. In a fee-for-service environment, technology that helps generate billable services may be adopted faster than technology that quietly prevents worsening blood pressure, improves depression, reduces pain interference, or slows progression from prediabetes to diabetes.
ACCESS tries to change the target of innovation. Under ACCESS, the organization is not paid because it produced more app clicks, more remote readings, more messages, more minutes, or more visits. It is paid because patients improved or reached control on defined measures.
That is the cleanest version of the ACCESS idea: pay for outcome production, not activity production.
The Basic Design
ACCESS is a ten-year CMMI model in Original Medicare beginning in July 2026. It starts with four broad clinical tracks.
The early cardio-kidney-metabolic track includes hypertension, dyslipidemia, obesity or overweight with central obesity, and prediabetes. The cardio-kidney-metabolic track includes diabetes, stage 3a or 3b chronic kidney disease, and atherosclerotic cardiovascular disease. The musculoskeletal track focuses on chronic musculoskeletal pain. The behavioral health track focuses on depression and anxiety.
The breadth of these tracks is important. CMS is not paying for a single app, a single device, or a single digital therapeutic. It is paying organizations to manage clusters of related conditions that share intervention types and outcome measures. A narrow technology-specific payment would reproduce the “one technology, one code” problem. A track-based payment pushes organizations to build care systems.
Participating organizations must enroll in Medicare Part B as providers or suppliers, except for excluded supplier types such as DMEPOS and laboratory suppliers. They must designate a physician clinical director or medical director. They must comply with state licensure, HIPAA, and applicable FDA requirements, or operate within FDA enforcement discretion. Care may be delivered in person, virtually, asynchronously, or through technology-enabled means. ACCESS contemplates clinician consultations, lifestyle support, behavioral support, therapy and counseling, patient education, care coordination, medication management, diagnostic ordering and interpretation, and use or monitoring of FDA-authorized or enforcement-discretion devices and software.
Patients enroll voluntarily, either directly or through referral. Medicare Advantage patients are not included, although MA plans may create similar arrangements. Some beneficiaries attempting to enroll may be randomized into a control group, which is essential for model evaluation. That randomized design matters because ACCESS is not only a payment pathway; it is an experiment. CMS must eventually show whether the model improves quality and what it does to Medicare spending.
ACCESS is also designed to complement traditional care rather than replace it. Patients retain their ordinary Medicare rights and may continue to see any Medicare provider. ACCESS organizations must share care plans and updates with primary care and referring clinicians. Those clinicians may bill a separate co-management payment for documented review and coordination work.
This creates a new kind of Medicare participant: not simply a vendor, not quite an ACO, not a health plan, and not merely a physician practice. The ACCESS organization is a Medicare-enrolled, clinically accountable, technology-supported chronic care entity paid through condition-track outcomes.
Outcome-Aligned Payments: The Core Innovation
The central payment mechanism is the Outcome-Aligned Payment. Participating organizations receive recurring payments for managing a qualifying condition track. Full payment depends on achieving measurable clinical improvement or control relative to the patient’s baseline.
This baseline-relative design is one of the model’s most important features. A model that pays only for absolute control risks rewarding organizations for enrolling easy patients. A model that ignores baseline risks punishing organizations that accept more difficult patients. ACCESS tries to thread the needle. Patients must have at least one uncontrolled measure to qualify for full initial payment, and organizations are paid for meaningful improvement or guideline-informed control.
The model also uses a population or portfolio logic. CMS determines payment based on the share of an organization’s patients who meet outcome targets. An organization can perform well overall even if some individual patients do not improve. That is clinically realistic. Chronic care is noisy. Patients may not respond, may disengage, may have side effects, may face food insecurity or transportation barriers, or may deteriorate for reasons outside the organization’s control. A model demanding success for every patient would be unworkable. A model paying for aggregate performance is more plausible.
This is not full capitation. It is not total-cost-of-care risk. It is condition-track, outcome-contingent payment. CMS is trying to capture some advantages of value-based care without requiring every digital health or chronic care company to become an insurer or ACO.
That makes ACCESS a possible middle layer between fee-for-service and full population risk. Many organizations are not prepared to accept total-cost-of-care risk, but may be prepared to accept outcome risk for a defined domain such as cardiometabolic disease, MSK pain, or behavioral health.
The Financial Architecture: Who Gets Paid, How Much, and for What
The financial architecture of ACCESS deserves explicit attention because it is the hinge between the model’s elegant theory and its practical incentives.
There are three distinct financial layers: payments to ACCESS organizations, payments to traditional clinicians, and spending outside the ACCESS payment.
The first layer is the payment to the ACCESS organization itself. These are the technology-supported chronic care organizations that enroll patients, manage a clinical track, collect baseline and follow-up measures, report data to CMS, and attempt to meet the outcome targets. They may be digital-first companies, hybrid care organizations, medical groups, behavioral health platforms, MSK companies, cardiometabolic care firms, or provider-affiliated entities. But under ACCESS they are not merely vendors. They must become Medicare Part B-enrolled providers or suppliers, accept Medicare compliance obligations, designate physician oversight, and bill ACCESS-specific G-codes.
The ACCESS organization receives an Outcome-Aligned Payment for managing the patient’s qualifying track. The payments are relatively modest annual amounts, paid monthly, and are intended to reflect the lower marginal cost of scalable technology-supported care. For the early cardio-kidney-metabolic track, the initial-year allowed amount is approximately $360 per beneficiary, with a lower continuation-year amount of approximately $180. For the cardio-kidney-metabolic track, the initial-year amount is approximately $420, with a continuation amount of approximately $210. For musculoskeletal care, the initial-year amount is approximately $180, with no continuation period. For behavioral health, the initial-year amount is approximately $180, with a continuation amount of approximately $90.
These are annual allowed amounts, not large capitated payments. They are divided into monthly payments. The beneficiary cost-sharing issue is handled flexibly: ACCESS organizations may elect to waive cost-sharing uniformly using the CMS-sponsored model safe harbor, or they may collect it with required disclosure. CMS’s decision to allow waiver of cost-sharing matters because many digital chronic care organizations will view even a small beneficiary bill as an enrollment barrier.
The second layer is payment to the patient’s ordinary clinicians. ACCESS is supposed to complement traditional care, not replace the patient’s primary care physician, cardiologist, nephrologist, psychiatrist, therapist, orthopedist, or other clinician. To support that integration, CMS created a separate co-management payment for primary care or referring clinicians who review ACCESS updates and document associated coordination activities. This payment is small, roughly in the range of $30 per service, with a small initial setup modifier and limits on frequency. The total amount is therefore not transformative for primary care economics. But it is important as a policy signal: CMS recognizes that traditional clinician review, medication-list reconciliation, problem-list updating, and response to ACCESS care plans are real work.
The third layer is spending outside the ACCESS payment. This is where the model becomes both clinically realistic and financially vulnerable. The ACCESS payment does not include many of the expensive things that better chronic care may require. Drugs, laboratory tests, imaging, durable medical equipment, specialist visits, physical therapy, behavioral health services outside the ACCESS entity, hospitalizations, and other ordinary Medicare services may still be billed by unaffiliated providers under standard Medicare rules. Beneficiaries retain freedom of choice and may continue seeing any Medicare provider.
This distinction is essential. ACCESS pays the chronic care organization to manage and improve outcomes, but it does not make that organization responsible for total cost of care. The ACCESS participant is blocked from stacking ordinary fee-for-service claims for the same aligned beneficiary, but unaffiliated providers can still bill Medicare. That means ACCESS has a partial-risk structure: the organization is financially accountable for delivering its own track management efficiently and meeting outcome targets, but it is not globally accountable for all downstream spending its work may trigger.
CMS is aware of this problem and includes a substitute-spend adjustment for certain duplicative services. For example, if an ACCESS musculoskeletal organization is managing chronic pain and the patient receives certain overlapping outside services, CMS may reduce the ACCESS payment. But this mechanism is necessarily narrow. The clinical boundary between duplicative and appropriate outside care is often ambiguous. A PT evaluation, imaging study, nephrology visit, medication change, behavioral health referral, or specialist consultation may be unnecessary in one patient and essential in another.
This is where the financial design creates the model’s central paradox. The ACCESS organization is paid a modest fixed amount to improve outcomes. But the best way to improve outcomes may be to uncover unmet needs that generate substantial outside spending. The organization may discover that the patient is not taking an expensive but indicated medication, needs physical therapy, needs behavioral health treatment, needs labs, needs home monitoring, needs specialist care, or needs a more intensive clinical workup. Ethically, the right answer may be to activate that care. Financially, that activation may undermine the model’s savings story.
ACCESS therefore has to succeed on two financial fronts at once. The direct payments to ACCESS organizations must be low enough and outcome-contingent enough to avoid becoming a new digital health annuity. At the same time, the model must not discourage the organization from triggering appropriate outside care when that care is clinically necessary. That balance is difficult. If CMS focuses too much on savings, ACCESS may reward low-cost metric management and under-escalation. If CMS focuses too little on total spending, ACCESS may become a platform for identifying patients and increasing outside utilization without producing offsetting value.
The payment design is elegant, but it is not magic. ACCESS pays for outcome production, not total-cost accountability. That may be the right middle ground for technology-enabled chronic care, but it means the model’s financial evaluation must distinguish three very different phenomena: waste reduction, cost shifting, and appropriate care activation. Only the first is true savings. The second is a model failure. The third may be good medicine, even if it costs Medicare more.
Why ACCESS Is Different from Prior Digital Health Payment
ACCESS differs from prior Medicare approaches in several ways.
First, it is not built around CPT code proliferation. CMS is not creating a code for every digital care activity. That matters because activity codes often generate activity inflation.
Second, it is not a device add-on payment. The model does not pay because a device is novel, expensive, connected, or FDA-cleared. Devices and software are tools inside the care model, not the unit of payment.
Third, it is not simply telehealth expansion. ACCESS is broader than live virtual visits. It can include asynchronous care, remote monitoring, coaching, therapy, education, diagnostic interpretation, medication management, and technology-supported longitudinal care.
Fourth, it is not dependent on bespoke ACO-vendor contracting. Risk-bearing entities often struggle to contract with technology-enabled chronic care organizations because savings are condition-specific, attribution is difficult, and transaction costs are high. ACCESS creates a standardized claims-based Medicare pathway that can be used alongside ACO arrangements without requiring every ACO to negotiate every parameter of every vendor relationship.
That last point is especially important. ACCESS could become a payment rail. An ACO may refer patients to an ACCESS organization without negotiating a custom cardiometabolic or MSK savings contract. CMS can incorporate ACCESS expenditures into ACO reconciliation, while the ACCESS participant bills Medicare directly under the model. If this works, it creates a standardized interface between chronic care vendors and the Medicare payment system.
The Fee-for-Service Exclusion: The Guardrail That Makes the Model Real
ACCESS is explicitly an alternative to fee-for-service billing by the ACCESS participant for a given beneficiary during the care period. It is not an add-on. The participant and its affiliates generally may not bill ordinary Medicare FFS claims for other services furnished to that beneficiary during the ACCESS care period. They may bill only ACCESS-specific codes.
This is one of the most important provisions in the model. Without it, ACCESS could become simply another layer of payment on top of traditional billing. An organization could collect outcome-aligned payments while also billing remote monitoring, telehealth, chronic care management, evaluation and management, or other claims for the same beneficiary. CMS appears to understand that such stacking would undermine the model immediately.
The exclusion is beneficiary-specific and participant-specific. An organization may treat one patient under ACCESS and another under ordinary FFS. But it cannot treat the same patient under ACCESS while also billing ordinary Medicare FFS for other services to that same patient during the care period.
Beneficiary choice is preserved. Patients may still see other Medicare providers, and those other providers may bill Medicare normally. That is necessary, but it creates a hard boundary problem. When is outside care complementary, and when is it duplicative? If an ACCESS MSK organization is managing chronic pain, is outside PT duplicative or necessary? If a cardiometabolic ACCESS organization is managing diabetes and CKD, is a nephrology visit duplicative or appropriate? If an ACCESS behavioral health organization is managing depression, is outside psychotherapy duplicative or essential?
CMS tries to address this through substitute-spend adjustments for certain overlapping services. But this will be hard to operationalize. Chronic disease patients often need multiple forms of care. The same outside service may be wasteful in one patient and ethically necessary in another.
This becomes central to the model’s moral and economic challenge.
The Co-Management Payment: Small Dollars, Large Signal
ACCESS includes a modest co-management payment for primary care and referring clinicians who review ACCESS updates and document coordination activities. The dollars are small, but the policy signal is meaningful. CMS recognizes that the traditional clinician’s attention is part of the infrastructure needed for technology-supported chronic care.
Digital health companies often fail because they sit outside the patient’s real care team. They may generate recommendations, dashboards, messages, remote monitoring data, care plans, or risk alerts, but the primary care physician may not know what is happening, may not trust the vendor, may not have time to review the material, or may not be paid to act on it. ACCESS tries to build a bridge. The ACCESS organization must share updates, and the traditional clinician may bill for reviewing and coordinating.
This will not transform primary care economics. But it is a practical design feature. It acknowledges that chronic care cannot be split cleanly between a digital vendor and a physician office. If ACCESS works, it will be because external technology-supported organizations become integrated extensions of traditional care, not parallel systems competing for the patient’s attention.
ACCESS and FDA: A Payment Model as Regulatory Sandbox
ACCESS also has an unusually important FDA dimension. Participants must comply with applicable FDA requirements or operate within enforcement discretion. CMS has also connected ACCESS to FDA’s TEMPO pilot, which may allow certain technology-enabled products to operate under enforcement discretion while real-world performance data are collected in the model context.
This is potentially a major bridge between payment policy and product regulation. Payment, clinical use, real-world evidence, and FDA oversight are usually separate. ACCESS may bring them together. A digital tool, connected device, or AI diagnostic may not need its own Medicare payment code if it becomes part of an outcome-producing care system.
This is especially important for early detection and AI diagnostics. Detection alone often has a business model problem. Finding prediabetes, early CKD, uncontrolled hypertension, depression, or cardiometabolic risk is not enough unless there is a reimbursable pathway for acting on the finding. ACCESS creates that downstream pathway. The diagnostic or device is not the revenue center; the outcome-producing care model is.
That could reshape digital health strategy. The question becomes less, “What code pays for our product?” and more, “What measurable outcome can our product help a Medicare-enrolled care organization achieve?”
The Applicant List: Market Validation, Not Proof
The ACCESS accepted-applicant list provides early market validation. CMS attracted more than a symbolic pilot cohort. The list includes digital-first chronic care companies, behavioral health platforms, cardiometabolic firms, MSK organizations, nutrition and weight-management brands, remote-monitoring companies, virtual care groups, nephrology practices, community organizations, provider-affiliated entities, and multi-track platforms.
This does not prove the model will work. CMS cautions that inclusion is not endorsement and does not guarantee final participation. Medicare enrollment, participation agreements, and final approval remain necessary. But the list shows that ACCESS has created a commercially legible pathway.
The strongest signal is that many accepted organizations have not previously served Medicare beneficiaries. ACCESS is not simply giving existing Medicare providers another billing mechanism. It is inviting nontraditional chronic care organizations into Original Medicare, but under Medicare enrollment, physician oversight, privacy, licensure, reporting, and quality obligations.
The applicant mix also reveals where the market sees opportunity. Cardiometabolic care appears especially attractive. That makes sense. Hypertension, dyslipidemia, obesity, prediabetes, diabetes, CKD, and ASCVD are huge Medicare problems; they have measurable biomarkers; and they map naturally to remote monitoring, medication management, nutrition support, adherence tools, and coaching.
Behavioral health is also heavily represented. Depression and anxiety have validated measures, large access gaps, and a strong virtual-care precedent. But they may be harder to risk-adjust and harder to manage consistently. PHQ-9 and GAD-7 are useful, but mood and anxiety outcomes are affected by pain, loneliness, sleep, cognition, substance use, social stress, and comorbid disease.
MSK is present but may have a more complicated savings story. Chronic pain is common and costly, and a technology-supported MSK program may substitute for some imaging, orthopedic visits, procedures, or prolonged therapy. But MSK care also overlaps with ordinary PT, pain management, imaging, orthopedics, surgery, and behavioral health. The line between substitution and necessary complementary care may be contentious.
The applicant list suggests that ACCESS is not a narrow chronic disease code. Many organizations appear to see it as a general payment chassis for longitudinal virtual or hybrid care.
Lessons from the Oncology Care Model
The Oncology Care Model provides a useful cautionary comparison. OCM was CMS’s first major oncology-focused alternative payment model. It ran from 2016 to 2022 and paid participating oncology practices monthly enhanced oncology services payments while allowing performance-based payments for reducing spending and meeting quality benchmarks.
A 2026 JAMA evaluation found that OCM was associated with modest gross reductions in Medicare spending during six-month cancer treatment episodes. The estimated reduction was about $616 per episode, or 2.1 percent of baseline spending. Savings grew over time and were larger in later performance periods. But after monthly enhanced oncology services payments and performance-based payments were included, the model produced an estimated net loss to Medicare of $639 million over six years. Most utilization and quality measures did not significantly change.
OCM is not ACCESS. Oncology episodes differ from chronic disease tracks. But OCM teaches several important lessons.
First, gross savings are not net savings. A model can reduce claims spending and still lose money once model payments are counted. ACCESS will face the same arithmetic. Its per-beneficiary payments are much lower than OCM’s $160 monthly enhanced oncology services payments, but the principle remains: the model payment itself must be earned back through real clinical value, reduced waste, or both.
Second, savings may take time. OCM’s reductions grew in later years. Chronic disease prevention may take even longer. Improvements in blood pressure, weight, lipids, pain, function, depression, or anxiety may not generate near-term reductions in admissions or ED visits.
Third, quality measurement may not capture everything that matters. OCM may have changed care delivery without producing broad measurable quality gains. ACCESS uses more direct clinical measures, which is an advantage, but it still faces the question of whether measured improvement translates into durable health benefit and lower spending.
Fourth, voluntary models attract selected participants. OCM practices were not random oncology practices, and ACCESS participants will not be random chronic care organizations. Early results may reflect who chose to enter as much as what the model achieved.
The larger lesson is sobering: a model can be clinically serious, operationally complex, and still fail the net-savings test.
Liao’s Critique: The Right Kind of Skepticism
Joshua Liao’s Health Affairs critique is useful because it avoids both hype and reflexive cynicism. He recognizes the real problem ACCESS is trying to solve: Medicare’s fee-for-service architecture is poorly suited to continuous chronic care. But he also emphasizes that payment models inevitably generate optimization behavior.
That is the mature way to assess ACCESS. The question is not whether distortions will occur. They will. Fee-for-service rewards volume. Capitation rewards restraint. Risk adjustment rewards diagnosis capture. Documentation-based quality measurement rewards documentation. Outcome-based models reward selection, measure optimization, and denominator management.
The question is whether ACCESS can generate enough clinical improvement to justify its distortions.
That is the right standard. No payment system is neutral. The fragmented status quo is also a policy choice. ACCESS should not be rejected merely because it will create strategic behavior. But it must be evaluated with a clear view of predictable failure modes.
Liao also points to a crucial distinction between scalable chronic care and intensive multimorbidity management. ACCESS may work best for a middle population: patients with measurable, improvable chronic conditions who can engage with technology-supported care and whose needs can be managed through standardized workflows. It may be less suitable for the sickest, least connected, most socially complex, or most multimorbid beneficiaries.
That distinction matters because the highest-need patients may be the least attractive under the model. The most economically attractive ACCESS patient is not necessarily the neediest patient. It may be the patient who is uncontrolled enough to qualify, reachable enough to engage, likely enough to improve, and unlikely to require expensive downstream care.
This is where the ACCESS optimism meets the ACCESS paradox.
The Savings Paradox: Under-Managed Patients May Need More Spending
ACCESS is expected to improve outcomes and reduce spending. The model’s savings theory has two parts.
First, better clinical control may reduce downstream costs. Better blood pressure, weight, HbA1c, lipids, pain, function, depression, and anxiety may eventually reduce acute utilization, complications, and high-cost care. This is the prevention theory.
Second, ACCESS may substitute for more expensive traditional care. A technology-supported MSK program may reduce some specialty visits, imaging, or procedures. A behavioral health platform may create lower-cost access where in-person care is scarce. A cardiometabolic program may prevent deterioration that would otherwise appear later in high-cost settings.
Both theories are plausible. But both collide with a competing reality: under-managed patients often have unmet needs that cost money to address.
A patient with uncontrolled diabetes may need medication intensification, nutrition counseling, renal testing, eye exams, foot care, CGM, or closer follow-up. A patient with CKD may need appropriate labs, SGLT2 therapy if indicated, nephrology input, and careful medication review. A patient with chronic pain may need PT, behavioral support, imaging, injections, surgery, or medication changes. A patient with depression may need therapy, medication management, safety assessment, and repeated follow-up. A patient with cardiovascular disease may need high-intensity lipid therapy, blood pressure medication changes, adherence support, and lab monitoring.
Some of this care may reduce future spending. Some may improve quality of life without saving money. Some may increase spending indefinitely because the patient is finally receiving appropriate care.
This is not failure. It is medicine.
Consider a simple adherence example. ACCESS may discover that a patient is not taking a clinically important $5,000 medication. Perhaps the patient cannot afford it, misunderstood the prescription, experienced side effects, or never completed a refill. The best intervention may be to restart and maintain therapy. But if adherence improves, Medicare spending rises. Pharmacy costs rise. Monitoring may rise. Follow-up may rise. The patient may be better treated, but the model may look worse financially.
That example reveals the tension. ACCESS organizations are paid to improve outcomes at fixed payment, and the model is expected to reduce overall Medicare costs. Yet the ethically correct response to under-treatment may be to activate expensive care.
This is the central paradox: ACCESS is designed to find and manage chronic disease, but properly managing chronic disease may increase Medicare spending.
Waste Reduction, Cost Shifting, and Need Activation
The word “savings” is too blunt for chronic care. Good chronic disease management does three different financial things, and CMS must distinguish them.
First, it reduces waste. It may prevent avoidable admissions, reduce duplicative imaging, avoid low-value procedures, replace unnecessary visits with lower-cost support, deprescribe harmful drugs, and prevent acute deterioration. This is real savings.
Second, it may shift costs. An ACCESS organization might improve its own economics while pushing services to unaffiliated providers who bill Medicare separately. The ACCESS payment might look efficient while total Medicare spending rises. This is not savings. It is displacement.
Third, it activates need. It may increase appropriate medication use, increase PT, increase behavioral health treatment, increase diagnostic monitoring, increase lab testing, increase specialist referral, and increase patient support. This may raise spending, but it may also represent better care for patients who had been neglected.
These three phenomena can look similar in claims data. Spending changes. Utilization changes. The interpretation is the hard part.
If spending rises because a model induces low-value utilization, that is a problem. If spending rises because the model corrects underuse of high-value care, that may be success. If spending falls because patients avoid unnecessary care, that is success. If spending falls because the organization failed to escalate necessary care, that is not success.
This is the hardest evaluation problem. A narrow net-spending test may penalize care activation. A narrow outcome-measure test may ignore cost shifting. CMS must evaluate whether spending growth represents waste, substitution failure, or newly delivered appropriate care.
For example, increased use of guideline-directed cardiometabolic drugs may raise Part D spending. Increased PT may raise Part B spending. Increased behavioral health care may raise outpatient spending. Better adherence may raise pharmacy costs. These increases could be exactly what good care requires.
The policy question should not be, “Did ACCESS raise or lower spending?” The better question is, “Did ACCESS reduce waste while increasing appropriate care where care had been missing?”
That is harder to answer. It is also the question that matters.
Medicare Advantage as a Mirror Image
Medicare Advantage provides a useful comparison because it shows a different way payment can separate information from obligation.
MA plans are famously incentivized to find and document diagnoses. More diagnoses can increase risk-adjusted payment. The result is a large industry built around coding intensity: chart reviews, home visits, health risk assessments, suspect-condition analytics, and documentation capture. The plan is rewarded when the patient is shown to be sicker.
But the payment system does not automatically require a proportionate treatment response. A plan may document diabetes with complications, vascular disease, depression, COPD, CKD, frailty, or pressure ulcers. The diagnosis increases revenue. Yet the system does not symmetrically say: having found this problem, you must demonstrate that the patient received optimal drugs, PT, nutrition support, behavioral health, wound care, home care, or specialist follow-up.
Quality measures and star ratings provide some counterweight, but the diagnosis-capture incentive is more direct and comprehensive than the treatment obligation.
MA’s failure mode is diagnosis capture without proportional care.
ACCESS has a different failure mode. It is not rewarded for finding diagnoses in the same risk-adjustment sense. It is rewarded for studying patients, engaging them, managing them, and improving measurable outcomes. But it may not be rewarded for triggering expensive care, even when expensive care is the best care.
ACCESS’s failure mode may be care-management capture without full care activation.
The contrast is useful. Medicare Advantage monetizes illness recognition. ACCESS monetizes measurable improvement. MA may over-code without enough treatment. ACCESS may over-monitor or under-escalate when treatment is expensive.
In both cases, the intermediary is rewarded for knowing more about the patient. But the patient’s interest is not merely being known, coded, monitored, or measured. The patient’s interest is receiving the right care.
The ethical question is the same: once the system knows, what is it obligated to do?
Selection: The Ideal ACCESS Patient
Every payment model creates an ideal patient. For ACCESS, the ideal patient is likely to be uncontrolled enough to qualify, reachable enough to engage, stable enough to manage through technology-supported workflows, likely enough to improve on measured outcomes, and not so complex that improvement requires expensive outside care.
That is a narrow sweet spot.
A patient already controlled may not qualify for full initial payment. A patient far from target may be hard to improve. A patient with severe social barriers may not engage. A patient with cognitive impairment may be difficult to manage digitally. A patient needing repeated specialist care may generate outside spending. A patient needing expensive drugs may improve clinically but worsen the model’s financial story.
CMS has built guardrails: baseline-relative targets, uncontrolled-measure requirements, limits on participant-initiated disenrollment, public reporting of risk-adjusted outcomes, rural adjustments, and monitoring. These are important. But no guardrail eliminates selection pressure. Organizations will learn which patients perform well under the model.
That does not invalidate ACCESS. Many “improvable” patients need help. But it should shape expectations. ACCESS may be strongest as a scalable chronic care improvement model for a large middle population. It may be less effective for the most complex, least connected, most socially vulnerable, or most clinically unstable patients.
The danger is not that ACCESS serves the middle well. That would be valuable. The danger is that policymakers mistake that success for solving chronic disease management broadly.
Outcome Measurement: Better Than Activity, Still Imperfect
ACCESS’s greatest strength is that it pays for outcomes rather than activities. Its greatest vulnerability is that measured outcomes can themselves become objects of optimization.
Blood pressure, HbA1c, lipids, weight, pain PROMs, PHQ-9, and GAD-7 are clinically meaningful. They are much better payment targets than app clicks, device transmissions, or staff minutes. But they are still measures. Measures can be selected around, documented around, repeated strategically, and optimized narrowly.
An organization may focus on moving patients across thresholds rather than improving broader health. It may favor patients likely to improve on measured variables. It may focus on short-term score movement rather than durable outcomes. It may improve the measure while leaving unmeasured clinical problems untouched.
Behavioral health illustrates the challenge. PHQ-9 and GAD-7 are validated and widely used, but depression and anxiety scores are affected by life events, loneliness, pain, sleep, substance use, cognition, housing instability, social isolation, and comorbid illness. Risk adjustment can help, but behavioral health outcomes remain noisy.
MSK raises similar issues. Pain intensity, interference, and function are meaningful, but chronic pain is heterogeneous. Some patients need digital coaching; others need PT, behavioral therapy, medication changes, imaging, injections, surgery, or social support. A model that improves PROMs may be valuable, but it may also select patients whose pain is most responsive to low-cost intervention.
Cardiometabolic tracks have more objective biomarkers, but they raise other concerns. Aggressive medication intensification can improve measures but create side effects, hypotension, hypoglycemia, falls, or polypharmacy. Conversely, reluctance to use expensive medications may preserve spending but fail the patient.
Outcome payment is better than activity payment. It is not magic.
What CMS Should Watch
CMS should watch several issues closely.
First, selection. Are ACCESS organizations enrolling patients who are most likely to benefit, or patients most likely to be profitable under the model?
Second, under-escalation. When organizations identify unmet need, do they trigger appropriate care even when it increases spending?
Third, outside spending. Are organizations shifting costs to unaffiliated providers, or are outside services clinically appropriate complements?
Fourth, primary care integration. Are PCPs and referring clinicians meaningfully engaged, or are they receiving electronic clutter in exchange for small payments?
Fifth, durability. Do measured improvements persist after the initial care period, or are they short-term threshold effects?
Sixth, equity. Are rural, low-income, cognitively impaired, socially complex, digitally limited, and multimorbid patients served, or are they quietly bypassed?
Seventh, net financial performance. Do ACCESS payments, co-management payments, substitute-spend adjustments, outside utilization, and downstream savings add up to real value? The OCM experience shows that gross reductions in claims spending may not be enough if model payments and incentive payments exceed savings.
The under-escalation issue is especially important because it may be less visible than overuse. Claims data can show increased imaging, PT, drugs, or specialist visits. It is harder to see the care that should have happened but did not. A model can look financially disciplined because necessary care was never activated.
That is why CMS cannot rely only on claims and summary outcome measures. It will need audits, clinical review, patient-level safety monitoring, referral pattern analysis, medication-use analysis, disenrollment monitoring, and focused studies of underuse.
Political Durability
ACCESS has a politically interesting profile. It is voluntary, outcome-based, technology-forward, prevention-oriented, and focused on chronic disease. It emphasizes patient choice, rural access, private-sector innovation, measurable results, and lower costs. It also expands access to behavioral health and chronic care support.
That gives it some cross-administration durability. It can be described in Republican-friendly language: innovation, competition, technology, private-sector solutions, outcome accountability, and fiscal discipline. It can also be described in Democratic-friendly language: access, prevention, behavioral health, chronic care, equity, and Medicare modernization.
But durability is not guaranteed. ACCESS is a ten-year model beginning in 2026. It will cross administrations. Its survival will depend on early execution, stakeholder support, operational feasibility, and avoidance of visible abuse or runaway spending. The randomized control design will help, but definitive results may take years.
The model’s early applicant pool suggests strong market interest. But market interest can be double-edged. If ACCESS becomes associated with opportunistic vendors, weak oversight, patient selection, or spending growth without visible benefit, it will become vulnerable. If it produces credible outcome improvement, meaningful patient engagement, and tolerable spending results, it could become a major precedent.
Likely Winners
The organizations best positioned for ACCESS are not pure software companies and not ordinary provider groups with little digital infrastructure. The likely winners are hybrid organizations that combine clinical operations, software, patient engagement, measurement infrastructure, and disease-specific workflows.
Cardiometabolic platforms may be especially well positioned because the conditions are common, measurable, and amenable to medication management, nutrition support, monitoring, and coaching. Behavioral health platforms may benefit from access gaps and validated measures, though they face risk-adjustment and engagement challenges. MSK companies may benefit if they can demonstrate substitution for more expensive care while maintaining functional improvement.
Remote-monitoring companies, nutrition and obesity-management firms, nephrology-adjacent groups, virtual primary care organizations, ACO-adjacent service companies, and provider-affiliated digital care teams may also find ACCESS attractive.
AI diagnostic and device companies may benefit indirectly. But the device, algorithm, or software tool is not the payment unit. It must become part of an accountable care model that improves outcomes. That is a strategic shift. A company’s reimbursement question becomes not, “What code pays for our technology?” but, “What outcome can our technology help produce, and who is the Medicare-enrolled organization accountable for producing it?”
Bottom Line
ACCESS is CMS’s attempt to build a new Medicare payment rail for technology-enabled chronic care. Its strongest idea is clear: pay for measurable improvement or control, not for visits, clicks, devices, minutes, or billing activity. Its most important guardrails include the fee-for-service exclusion, physician oversight, Medicare enrollment, FDA and HIPAA compliance, care coordination requirements, public outcome reporting, substitute-spend adjustments, and randomized evaluation.
The model is serious. It addresses a real defect in Medicare’s payment architecture. It may create a new category of Medicare-participating chronic care organizations. It may become a reference model for private payers. It may help CMS and FDA learn how to handle AI-enabled and software-supported care. It may shift digital health strategy away from code capture and toward outcome production.
But ACCESS also exposes the contradiction in CMMI’s savings logic. Many chronic disease patients are under-managed. Under-management may look inexpensive until someone starts managing the patient correctly. Better care may require more medication, more PT, more behavioral health, more monitoring, more labs, more specialist input, and more adherence support. Some of that spending may eventually prevent complications. Some may improve life without saving money. Some may simply be the appropriate cost of treating Medicare beneficiaries well.
That is the central ACCESS paradox: the model is designed to find and manage chronic disease, but properly managing chronic disease may increase spending.
Medicare Advantage offers the mirror image. MA plans are rewarded for finding and documenting illness diagnoses, but not always equivalently rewarded for doing everything needed to treat them. ACCESS organizations are rewarded for studying patients and improving measurable outcomes, but may be financially uncomfortable when the best care requires expensive activation. MA risks coding without care. ACCESS risks management without escalation.
The patient needs neither merely to be coded nor merely to be monitored. The patient needs the right care.
That should be the standard for judging ACCESS. Not whether it is elegant. Not whether it creates a new digital health market. Not even whether it saves money in a narrow window. The real question is whether ACCESS can improve chronic care while resisting the predictable pressure to select easy patients, optimize measures, shift costs, avoid costly escalation, and confuse low spending with good care.
ACCESS should be judged by whether its modest payments to technology-supported care organizations produce real clinical value after accounting for model payments, co-management payments, substitute spending, outside utilization, and the ethically necessary activation of care that under-managed patients should have been receiving all along.
If ACCESS succeeds, it could become one of Medicare’s most important payment innovations of the decade. If it fails, the failure will still teach an important lesson: outcome-based payment is not enough unless the model can distinguish waste from unmet need, savings from underuse, cost shifting from efficiency, and measurable improvement from ethically adequate care.
