I was sitting with a model of care (MOC) recently — not skimming it, actually reading it — and it brought back a conversation I had with a colleague that I have not been able to shake.
We were talking about our years leading case management teams inside Special Needs Plans (SNPs). Both of us held senior roles. Both of us supervised the people responsible for coordinating some of the most medically complex, socially fragile members in managed care. When we asked each other the same question, we got the same answer:
We did not remember ever having the MOC in our line of sight.
Not pinned to the wall. Not referenced in team huddles. Not pulled out when a new case manager was onboarded or when a difficult member situation required a reset. The policies and procedures? Yes. The workflows? Certainly. But the MOC — the document that defines why all those policies exist, whom they were designed to serve, and how the health plan envisioned care being coordinated for its most vulnerable enrollees — was largely invisible to the people doing the work.
The MOC Is Not a Compliance Document
That is worth thinking about for a second.
Yes, Centers for Medicare & Medicaid Services (CMS) requires an MOC. Yes, it gets submitted. Yes, it gets scored by the National Committee for Quality Assurance (NCQA) against detailed factor-level criteria. But a MOC is not, or should not be, a filing exercise. It is a plan’s stated theory of how to care for a defined population. It describes the membership. It establishes the goals. It creates the framework within which every clinical decision, every care coordination touchpoint, and every interdisciplinary team meeting is supposed to operate.
When the people leading those teams have never internalized it, something has broken down, not in compliance, but in care.
The Rules Have Changed, and the MOC Must Change With Them
The SNP environment has shifted a lot in the last two years, and those shifts raise the stakes on the MOC in ways that are hard to ignore.
CMS published its calendar year (CY) 2026 final rule in April 2025 and, among other things, codified specific timeframes for all SNPs to complete health risk assessments (HRAs) and individualized care plans. Initial HRAs must be completed within 90 days of enrollment. Individualized care plans must follow within 90 days of the HRA or 90 days after enrollment, whichever is later. Plans must document at least three outreach attempts and a follow-up letter before marking a member as unreachable. These are no longer best practices — they are regulatory requirements.[1]
For Dual Eligible Special Needs Plans (D-SNPs) operating as Applicable Integrated Plans (AIPs), CMS is going further. By 2027, AIPs must issue a single integrated member ID card for both Medicare and Medicaid[2] and conduct a single integrated HRA across both programs.[3] That is a big operational change, one that requires the MOC to describe not just what the plan does for the Medicare benefit, but also how it plans to deliver a unified experience for dually eligible members simultaneously navigating two programs.
At the same time, the termination of the Value-Based Insurance Design (VBID) model at the end of CY 2025 has changed the supplemental benefit picture. As VBID-based approaches to addressing social determinants of health are no longer available, plans should consider leveraging Special Supplemental Benefits for the Chronically Ill (SSBCI) to preserve condition-specific, non-medical benefits that meaningfully support the health of SNP members that qualify.[4]
For many D-SNP members, benefits they had in 2025 look different in 2026, and the MOC is the place where the plan’s rationale for those benefit design choices should be laid out clearly and understood by the care team.
SNP Enrollment Is Growing Fast, and the Mix Matters
There’s a market side to this, too. SNPs now account for roughly 21% of all Medicare Advantage (MA) enrollees, up from 13% in 2018. Between 2024 and 2025 alone, growth in SNPs comprised nearly half of the total increase in MA enrollment.[5] SNPs are no longer a niche product. They are driving a huge share of MA growth.
But the composition of that growth matters. Chronic Condition Special Needs Plan (C-SNP) enrollment surged more than 70% between 2024 and 2025, far outpacing D-SNP growth.[6] CMS has taken notice and has openly expressed concern that some of this C-SNP growth may reflect an effort by MA organizations to route dually eligible members into plan types that avoid the integration requirements and state contracting obligations that come with D-SNP designation. A November 2025 request for information signaled that CMS is actively exploring whether C-SNPs and Institutional Special Needs Plans (I-SNPs) with high dual-eligible concentrations should be subject to requirements similar to those already in place for D-SNPs.[7]
So, what does all this mean for the MOC? If your plan is experiencing rapid enrollment growth, your MOC needs to reflect the population you are serving today — not the one you described two or three years ago. If your membership has shifted in acuity, in dual-eligible mix, in geographic footprint, or in the prevalence of behavioral health or social complexity, and your model has not kept pace, your clinical operations are sitting on top of an outdated document.
Who Is Actually Being Served?
SNPs were built around a premise that certain populations require more than standard managed care. Dual eligibles navigating the fault lines between Medicare and Medicaid. People with severe or disabling chronic conditions. Institutionalized individuals. These are not simply “high utilizers.” They are people whose lives are often marked by medical complexity, social instability, cognitive challenges, and systems that were not designed with them in mind.
The MOC is supposed to reflect that reality. NCQA has been explicit in its CY 2026 scoring guidance: the description of the most vulnerable subpopulation must be clearly differentiated from the general target population. Data must be current, nothing older than three years. Plans renewing after two years of operation must use their own historical data rather than national or proxy benchmarks.[8] CMS wants to see specifics about who is actually in the plan, not a boilerplate population overview lifted from a prior submission.
If a case manager reading the MOC would not understand — with specificity — why this person needs a different kind of care coordination, the document has not done its job.
Your Hardest-to-Reach Members Need the Most Thought-Through Model
There is growing scrutiny from CMS, from advocacy communities, and from the plans themselves on how SNPs engage enrollees who are not actively using services. The member who does not answer the phone. The one who has no documented care needs, but whose life circumstances suggest otherwise. The one who has fallen off the map.
These are often the most vulnerable members in the plan and reaching them effectively requires more than a workflow. It requires an understanding of the population, their barriers, their distrust, the patterns of disengagement that are, in themselves, clinical signals.
The new outreach requirements from CMS makes this concrete. The new rules do not merely say “try to reach the member.” They specify a minimum of three attempts and a written follow-up before a plan can document a member as unable to be reached.[9] That is not a suggestion. It is a compliance requirement — and one that should be reflected in how the MOC describes the plan’s approach to engagement for its hardest-to-reach members.
An MOC that the care team has actually read and absorbed becomes a real resource for exactly these situations. It answers the question: Given what we know about who we serve, what does meaningful engagement actually look like for someone who has none?
That is a question that no policy manual answers alone.
What I Am Thinking About as the Submission Window Approaches
June is coming. Plans are completing or finalizing their MOCs. For many, this process begins with a look at the previously approved model, a round of edits, and a submission.
Here’s what I would push leaders to ask instead:
- Does this document reflect the population we are actually serving today, not the one we described in 2023?
- Have we accounted for the CY 2026 regulatory changes, codified HRA and ICP timelines, new outreach documentation requirements, VBID termination, and the shift to SSBCI, in how our model describes care delivery?
- If we are an AIP D-SNP, does our model lay the groundwork for the 2027 integrated HRA and integrated ID card requirements, or are we treating those as a separate project?
- If a new case manager read this document and nothing else, would they understand the intent behind every workflow we have built?
- Are the people leading interdisciplinary care teams equipped not just to follow process, but to understand the framework that process is meant to serve?
- How does our model speak to the members who are least visible, the ones with no active services, the ones our data shows us least about?
These are not audit questions. They are care questions.
A Different Kind of Review
Ankura works with SNPs on MOC reviews before submission, not to check boxes, but to pressure-test the document against the reality of care delivery and the current regulatory environment.
That means asking whether the model reflects current membership and recent enrollment trends. Whether the population description meets CY 2026 data currency and specificity requirements. Whether vulnerable subpopulations are actually differentiated. Whether a care team leader reading it would come away with a framework for judgment, not just a set of instructions.
It also means asking whether the model accounts for the operational changes hitting SNP care delivery right now: codified HRA and ICP timelines, the end of VBID, stricter integration standards for D-SNPs, and the growing regulatory attention on C-SNP enrollment patterns.
And it means asking whether the document can be operationalized, whether the people responsible for living it have ever been given the chance to understand it.
If you are preparing for a June submission and want a set of experienced eyes on your MOC, eyes that have sat on both sides of this document as health plan leaders and as reviewers, I would like to hear from you.
Because the most important thing about a MOC is not that it gets submitted.
It is that someone, somewhere in your organization, actually uses it.
References
[1] CMS. “Contract Year 2026 Policy and Technical Changes to the Medicare Advantage Program, Medicare Prescription Drug Benefit Program, Medicare Cost Plan Program, and Programs of All-Inclusive Care for the Elderly” (“CMS-4208-F”) at 90 FR 15909. April 15, 2025. 42 CFR 422.101(f) (2025).
[2] CMS-4208-F at90 FR 15793. 42 CFR 422.2267(e)(30)(viii).
[3] CMS-4208-F at90 FR 15793. 42 CFR 422.101(f)(1)(v). See also: Integrated Care Resource Center (“ICRC”). “CMS Issues CY2026 Medicare Advantage and Part D Final Rule.” April 22, 2025.
[4] See, e.g.: CMS. Medicare Advantage Value-Based Insurance Design (VBID) Model to End after Calendar Year 2025: Excess Costs Associated with the Model Unable to be Addressed by Policy Changes. December 16, 2024.
[5] Sachar, A.; Biniek, J.F.; Mohamed, M.; Burns, A. “A Closer Look at the Growing Role of Special Needs Plans in Medicare Advantage.” KFF. September 25, 2025.
[6] Ochieng, N.; Freed, M.; Biniek, J.F.; Damico, A.; Neuman, T. “Medicare Advantage in 2025: Enrollment Update and Key Trends.” KFF. July 28, 2025.
[7] Kopans, D. “Medicare Advantage C-SNPs and I-SNPs Come Under Increased Scrutiny.” DLA Piper. January 21, 2026. See also: CMS., “Policy and Technical Changes to the Medicare Advantage Program, Medicare Prescription Drug Benefit Program, and Medicare Cost Plan Program” (CMS-4212-P) at 90 FR 54982 – 54983. November 28, 2025.
[8] NCQA, “CY 2026 SNP Model of Care Training FAQs” at p. 3, 4. Revised January 2025.
© Copyright 2026. The views expressed herein are those of the author(s) and not necessarily the views of Ankura Consulting Group, LLC, its management, its subsidiaries, its affiliates, or its other professionals. Ankura is not a law firm and cannot provide legal advice.
