April 1, 2026
On March 2026, CMS opened the Request for Applications for the Long-term Enhanced ACO Design (LEAD) Model, the formal successor to ACO REACH and the most ambitious accountable care initiative the Innovation Center has ever launched. Applications close May 17, 2026. The model starts January 1, 2027, and runs through December 31, 2036—a full decade without benchmark rebasing.
That last detail is the one that should make every ACO executive stop scrolling. No rebasing for ten years means the savings you generate in year one compound into year ten. It also means the losses you fail to prevent do the same thing. LEAD is a bet on operational infrastructure, not on gaming the reconciliation cycle.
If your ACO still runs population health on spreadsheets and quarterly chart reviews, this article is your wake-up call.
LEAD introduces several structural changes that raise the bar for participating ACOs:
Participants choose between Professional and Global risk tracks, with 50% or 100% shared savings and losses. Unlike MSSP’s incremental glide path, LEAD expects ACOs to take meaningful downside risk from the outset. The upside: capitated, prospective population-based payments that give you capital to invest in care delivery transformation.
LEAD creates a distinct benchmark and trend factor for high-needs patients—dual-eligibles, homebound beneficiaries, and complex chronic populations. ACOs with more than 40% high-needs beneficiaries qualify for a lower minimum beneficiary threshold. This is a direct invitation to organizations serving Medicaid-Medicare crossover populations, but it requires you to actually identify, stratify, and manage those patients at scale.
LEAD introduces new flexibilities for integrating specialists into ACO care pathways and funding health promotion activities. Translation: CMS wants to see coordinated, whole-person care delivery, not just primary care visit volume.
The elimination of benchmark rebasing over the model’s lifetime is LEAD’s most consequential design choice. In MSSP, ACOs that reduce spending see their benchmarks ratchet down, punishing success. LEAD removes that perverse incentive. But it also means your initial benchmark is the one you live with. Getting your risk adjustment, coding accuracy, and population stratification right before January 2027 is not optional.
Let’s be direct. The median ACO in America today cannot meet LEAD’s operational requirements with its current technology stack. Here is why:
Most ACOs run risk stratification quarterly using claims data that is 60-90 days old. By the time a rising-risk patient surfaces in your analytics, they have already had the avoidable ED visit or inpatient admission. LEAD’s total cost accountability means you pay for that lag—literally.
Quality measures under value-based contracts require closing care gaps—Annual Wellness Visits, cancer screenings, A1c checks, medication adherence. Most ACOs still rely on care coordinators manually pulling EHR reports, cross-referencing payer rosters, and making phone calls. The math does not work at 10,000 beneficiaries. It certainly does not work at 50,000.
LEAD’s high-needs carve-out is a massive opportunity, but dual-eligible patients are the hardest population to manage. They span Medicare and Medicaid systems, have fragmented data across multiple payers, and often have social determinants that clinical data alone cannot capture. Without a unified longitudinal record, you are flying blind.
To compete in LEAD, your ACO needs four capabilities operating in near-real-time:
Move from quarterly retrospective models to concurrent risk adjustment that incorporates diagnoses documented during the performance year. CMS has signaled this approach in LEAD’s design—your analytics need to match. AI-driven risk models that ingest clinical, claims, and ADT data continuously can flag rising-risk patients weeks before they become high-cost.
Care gap closure cannot remain a manual workflow. You need systems that automatically reconcile clinical data against quality measure specifications, identify open gaps at the patient level, and trigger outreach—whether that is a care coordinator call, a patient portal message, or a scheduling workflow.
This is exactly what we built at Zynix for PBACO, a large accountable care organization managing Medicare beneficiaries across multiple states. PBACO deployed Zynix’s AI-driven care gap engine to automate Annual Wellness Visit identification and outreach. The result: a measurable increase in AWV completion rates within the first performance year, directly impacting their quality score and shared savings position.
For the high-needs population track, you need a single longitudinal view that merges Medicare claims, Medicaid encounters, pharmacy data, lab results, and social determinant indicators. Eternal Health, a Zynix customer focused on dual-eligible populations, uses our platform to unify exactly this kind of fragmented data. Without that unification, you cannot accurately stratify risk for the population that LEAD specifically incentivizes you to serve.
LEAD’s 10-year horizon means your care management model needs to scale without linearly scaling headcount. HCN, one of the largest care management organizations in the country, partnered with Zynix to operationalize AI-assisted care management workflows that allow a smaller team to manage a larger, more complex patient panel. The platform automates task routing, surfaces priority patients, and tracks intervention outcomes—turning care management from an art into an engineered process.
LEAD does not exist in isolation. CMS launched the Transforming Episode Accountability Model (TEAM) on January 1, 2026, placing roughly 740 hospitals under mandatory bundled payment accountability for surgical episodes. Early analyses from Brandeis University and the Institute for Accountable Care suggest up to two-thirds of participating hospitals may lose revenue under TEAM.
For ACOs evaluating LEAD, the TEAM overlap matters. If your attributed beneficiaries receive surgical care at TEAM-participating hospitals, your total cost of care benchmark will reflect those episode costs. ACOs that can coordinate with TEAM hospitals on post-acute utilization, readmission prevention, and discharge planning will have a structural cost advantage.
This is another area where data infrastructure separates winners from losers. Union Health, a Zynix customer, uses our platform to track HEDIS quality measures and coordinate care transitions across acute and post-acute settings. That kind of cross-setting visibility is precisely what LEAD-plus-TEAM coordination requires.
ACOs evaluating LEAD should understand the technology market context. At HIMSS 2026 in March, Innovaccer launched Flow Capture, an autonomous medical coding solution. Abridge raised $300M and is expanding beyond ambient documentation into clinical decision support. Epic continues building native AI capabilities into its EHR.
The pattern is clear: point solutions for documentation, coding, and clinical AI are proliferating. What remains scarce is the integrated data and workflow layer that ties these capabilities into operational ACO performance—risk stratification, care gap closure, quality reporting, and financial reconciliation in a single platform.
That integration layer is what Zynix builds. CHOA, Children’s Healthcare of Atlanta, deployed Zynix to close pediatric care gaps across a complex, multi-site system. The challenge was not a lack of clinical AI tools. It was the absence of a unified data backbone that could reconcile clinical records, identify open gaps, and route them to the right care team. That is the same challenge every LEAD applicant will face at scale.
If your ACO is evaluating the LEAD Model RFA, here is a concrete action plan:
The LEAD Model RFA closes May 17, 2026. The organizations that win will not be the ones with the best application narratives. They will be the ones whose data infrastructure is already operating at the level LEAD demands.
If you want to see how Zynix powers ACO performance at scale—from risk stratification to care gap closure to quality reporting—request a technical demo at zynix.ai.