Network Design for C-SNPs and I-SNPs

What are C-SNPs and I-SNPs?

Special Needs Plans are a category of Medicare Advantage coordinated-care plan that CMS permits to restrict enrollment to specific populations, defined at 42 CFR 422.2 and detailed in Chapter 16-B of the Medicare Managed Care Manual. There are three types. D-SNPs serve beneficiaries dually eligible for Medicare and Medicaid and make up the large majority of SNP enrollment. C-SNPs serve beneficiaries with one or more severe or disabling chronic conditions drawn from a CMS-approved list. I-SNPs serve beneficiaries who need, or are expected to need for 90 days or longer, the level of care provided in a nursing facility or comparable institution.

The distinction matters for network design because each type's members use care differently. A D-SNP network must braid Medicare and Medicaid benefits, a subject we cover in D-SNP network requirements and its evolving integration rules. A C-SNP network must be deep in the specialties that treat its qualifying conditions. An I-SNP network is built around long-term-care facilities and the clinicians who round in them. Same regulatory family, very different blueprints.

Which chronic conditions qualify a C-SNP, and what do they demand?

CMS recognizes 15 SNP-specific chronic conditions that can serve as the basis for a C-SNP. A plan may build around a single condition, a CMS-approved grouping of clinically related conditions, or a broader multi-condition design that CMS reviews for coherence. The recognized conditions are the anchor of the entire network, because they dictate which specialties and services the plan must make abundant rather than merely adequate.

Each condition implies a different clinical spine. A diabetes C-SNP leans on endocrinology, ophthalmology, podiatry, nephrology, and nutrition support; a cardiovascular or chronic-heart-failure plan needs cardiology depth, anticoagulation management, and cardiac rehabilitation; a dementia plan needs neurology, geriatric psychiatry, and robust caregiver and home-based support. The network that satisfies CMS adequacy for a general MA plan will almost never be dense enough in these specialties to serve the C-SNP population well. Designing the specialist mix to the condition is the core intellectual work of a C-SNP build.

• Chronic alcohol and other drug dependence
• Autoimmune disorders (e.g., rheumatoid arthritis, systemic lupus erythematosus)
• Cancer, excluding pre-cancer conditions
• Cardiovascular disorders (e.g., coronary artery disease, arrhythmias, peripheral vascular disease)
• Chronic heart failure
• Dementia
• Diabetes mellitus
• End-stage liver disease
• End-stage renal disease requiring dialysis
• Severe hematologic disorders (e.g., sickle-cell disease, hemophilia, myelodysplasia)
• HIV/AIDS
• Chronic lung disorders (e.g., COPD, asthma, chronic bronchitis, emphysema)
• Chronic and disabling mental health conditions (e.g., schizophrenia, bipolar disorder, major depression)
• Neurologic disorders (e.g., epilepsy, multiple sclerosis, Parkinson's disease, ALS)
• Stroke

How do you design the specialist mix for a C-SNP?

Start from the care journey of the qualifying condition, not from the standard adequacy grid. Map the specialties, ancillary services, and sites of care a member with that condition touches in a typical year, then quantify expected utilization against the enrollment you project. The goal is a network where the high-frequency specialists are close, plentiful, and contracted for the volume the population will actually generate — not a network that technically clears minimums but leaves members waiting months for the cardiologist or nephrologist they see quarterly.

Behavioral health deserves special attention, because chronic-disease populations carry high rates of co-occurring depression, anxiety, and substance-use disorders, and CMS has sharpened its behavioral-health adequacy expectations. Building genuine access here — not just a directory listing — is both a compliance and an outcomes issue, as we discuss in behavioral health adequacy. Because HEDIS and Star measures for these conditions turn on whether members actually receive recommended care, the specialist mix and the quality strategy are the same conversation, not two separate ones.

What makes an I-SNP network different?

I-SNPs invert the usual network-design problem. Their members live in — or are clinically eligible for — skilled nursing facilities, long-term-care nursing facilities, intermediate care facilities for individuals with intellectual disabilities, or inpatient psychiatric facilities. The network's center of gravity is therefore the facility itself and the clinical team that delivers care inside it, typically nurse practitioners and physicians who round on-site, supported by the facility's own nursing staff. CMS eligibility hinges on a 90-day-or-longer institutional level-of-care need, verified through a state assessment tool applied identically to community-dwelling and institutionalized members.

That model reshapes adequacy. CMS allows an I-SNP that operates one or multiple facilities to establish a county-based service area as long as it has at least one long-term-care facility that can accept enrollment and is accessible to county residents. The contracting priority is securing those facility relationships and the on-site provider capacity to serve them, rather than assembling a sprawling outpatient panel. Credentialing the rounding clinicians and standing up the facility agreements is the critical path — and it is unforgiving, which is why we treat credentialing as the critical path in institutional builds.

Because so much care is delivered in place, I-SNPs also lean heavily on telehealth and on specialties that can be delivered remotely. CMS has recognized this: in its December 2024 network adequacy guidance it created exceptions for facility-based I-SNPs that cannot contract with certain specialists and expanded the telehealth arrangements that count toward adequacy, including a 10-percentage-point credit for plans that add qualifying telehealth providers in specialties such as clinical psychology and clinical social work. Designing an I-SNP network without exploiting these provisions leaves both access and compliance on the table.

How does CMS evaluate SNP network adequacy?

SNP networks are held to the same core adequacy framework as other MA plans under 42 CFR 422.116: for each provider and facility specialty type, in each county type, the plan must contract with a minimum number of providers and meet maximum time-and-distance standards. CMS runs these tests against its published criteria using the Health Service Delivery tables plans submit, a process we break down in how CMS measures time and distance.

Where SNPs differ is in emphasis and in the exceptions available to them. The specialties that matter most for a C-SNP are precisely the ones most likely to be thin in a given market, so exception requests — supported by documentation that the plan tried and failed to contract, or that access is achieved another way — become a routine part of a SNP filing. Facility-based I-SNPs have their own dedicated exceptions. Knowing when and how to file a defensible exception is a core competency; our guide to network adequacy exceptions covers the standard CMS will actually accept.

The practical implication is that SNP adequacy is not a one-time hurdle cleared at application. It is a living target that shifts as providers churn, as CMS updates criteria, and as enrollment concentrates in particular counties. Building the monitoring to catch a gap before CMS does — rather than after a filing is rejected — is what separates a plan that scales from one that scrambles every cycle.

Why is the Model of Care the spine of a SNP?

Every SNP must operate under a Model of Care that CMS requires and that the National Committee for Quality Assurance (NCQA) reviews and scores before approval. The Model of Care is not paperwork adjacent to the network — it is the clinical operating system the network exists to execute. It specifies the target population, the staff structure, and the care-management processes that turn a list of contracted providers into coordinated care for a high-need cohort.

Its core components drive network requirements directly. Each member receives a Health Risk Assessment that inventories medical, functional, cognitive, psychosocial, and behavioral needs; those results feed an Individualized Care Plan; and the plan is carried out by an Interdisciplinary Care Team whose members must have demonstrated expertise with the target population. A C-SNP whose Model of Care promises disease-specific specialty coordination must have those specialists under contract to deliver it, and an I-SNP whose Model of Care promises on-site, facility-based care management must have that staffing in place. NCQA scores the Model of Care against published guidelines, and a network that cannot execute it is a network that will not pass.

This is why network design and Model of Care design cannot be sequenced separately. The clinical model defines the population's needs; the network must be provably capable of meeting them; and the Star Ratings that ultimately determine a plan's revenue depend on both working together. Treating the Model of Care as a compliance afterthought is the most common — and most expensive — SNP design mistake.

How do SNP networks differ from standard MA and D-SNP networks?

A standard MA network optimizes for breadth across a general senior population. A SNP network optimizes for depth in a defined cohort. The C-SNP concentrates specialty capacity around qualifying conditions; the I-SNP concentrates around institutions and the clinicians who staff them; the D-SNP concentrates around the intersection of Medicare and Medicaid benefits and the providers, such as behavioral health and long-term services and supports, that dual-eligibles disproportionately need. Each demands a different mix, a different set of exception strategies, and a different Model of Care.

The contrast with D-SNPs is especially worth understanding, because many organizations run more than one SNP type and assume the networks are interchangeable. They are not: a D-SNP's Medicaid-driven requirements and a C-SNP's condition-driven requirements pull the network in different directions, even in the same market. The interplay between Medicaid and Medicare network standards is its own discipline, one we take up in Medicaid versus Medicare Advantage networks. Designing each SNP to its own population — rather than cloning a general network and hoping it clears — is the through-line of every point above.

Sequencing your SNP network build

The order of operations matters. Define the population and Model of Care first, because they determine the specialties and facilities you must prioritize. Map expected utilization to translate that model into concrete contracting targets. Then build the high-frequency, hardest-to-recruit specialties — or, for an I-SNP, the anchor facilities — before you chase the long tail, and stand up adequacy monitoring and exception documentation in parallel so nothing surfaces as a surprise at filing. Working backward from the CMS deadline, credentialing and contracting the critical specialties is almost always the binding constraint.

This is exacting work, and the penalty for getting it wrong is a rejected bid or a plan that cannot deliver on its own Model of Care. Kearny Street Management builds and repairs exactly these networks — matching the specialist mix and facility relationships to the population, filing defensible adequacy and exceptions, and making sure the network can execute the Model of Care the plan promised. If you are standing up a C-SNP or I-SNP, or fixing one that stalled, talk to our team and start with a clear-eyed look at what your population actually requires. Credentialing the right providers on time is often the whole game — see how we treat credentialing as the constraint it is.