Same benefits, fewer members, a determination for each one

The Value-Based Insurance Design model ended after plan year 2025. For most D-SNPs that was not a benefit decision. It was a change of authority, and it happened to nearly the whole market at once.

Milliman puts the shift at roughly 90 percent of D-SNP members in plans offering VBID benefits in 2025, against roughly 15 percent in plans with SSBCI coverage. With VBID terminated, SSBCI prevalence moved to roughly 90 percent.¹

Read as a benefit story that is unremarkable. Plans kept offering transportation, food support, in-home help. Read as an eligibility story it is one of the larger operational changes the D-SNP market has absorbed in years.

What actually changed is the targeting basis

Under VBID, a plan could target dual-eligible members using socioeconomic criteria.¹ Dual status. Low income subsidy status. Facts that already sit in the enrollment file, known on the day the member enrolls, requiring no clinical judgment and no individual assessment.

You made one plan design decision and applied it to a population.

SSBCI works the other way. Eligibility attaches to the chronically ill definition, which requires three separate findings about a specific person: that the enrollee has one or more comorbid and medically complex chronic conditions that is life threatening or significantly limits overall health or function, that the enrollee has a high risk of hospitalization or other adverse health outcomes, and that the enrollee requires intensive care coordination.² The benefit itself must also carry a reasonable expectation of improving or maintaining the health or overall function of that enrollee.²

The unit of decision moved from the population to the person. Everything downstream follows from that.

The eligible population is smaller than it was

This is the part that gets missed, and it is the part a Medicare product lead should care about most.

Chronic condition criteria are narrower than income based targeting. Milliman observes that fewer members qualify under the new model even where benefit categories remain nominally available, which is why aggregate benefit value declined in cases where the advertised menu looks unchanged.¹

So the position a plan is in for 2027 is specific. The same benefits appear in the catalog. The population that can actually receive them is smaller. And the difference between those two numbers is no longer a rounding adjustment, it is a set of individual determinations that either happened and were documented, or did not.

That gap was priced into the bid months ago. It now has to be managed through the coverage year and evidenced afterward. Three different teams, one underlying record.

The documentation obligation arrived with the authority

SSBCI itself is not new. What is new is that it now carries most of the D-SNP supplemental benefit book rather than a slice of it, and the obligations attached to it apply per enrollee.

A plan offering SSBCI must have and apply written policies based on objective criteria for determining enrollee eligibility, must document each eligibility determination whether the enrollee is found eligible or ineligible, must list those written policies and objective criteria on a public facing website, must make the documentation available to CMS on request, and must maintain the specific objective criteria without modification for the full coverage year.³

The CY2027 Final Rule, published April 2, 2026, tightened this further by requiring plans to use objective processes to demonstrate enrollee eligibility and to publicly post their plan-developed SSBCI eligibility criteria.⁴

Records supporting those determinations sit inside a ten year retention obligation and are subject to HHS and Comptroller General inspection for the current contract period and ten prior periods.⁵

Why nobody built for this

Under VBID there was no individual clinical determination to record, because eligibility rested on a status the plan already held. There was nothing to document per member because nothing was decided per member.

So the absence of a determination record across most of this market is not an oversight. It is an accurate reflection of what the previous authority required. The teams running these programs built exactly what the model asked for.

The model changed underneath them, on a one year notice, for close to the entire dual eligible book.

A plan that ran VBID well in 2025 has no determination record to carry into 2026, and no gap in its past practice either. Both things are true.

Three questions worth answering

• How many of your members actually qualify? Not how many are enrolled in a plan that offers the benefit. How many meet the chronically ill test on objective criteria you have written down. The difference between those numbers is your real exposure, in both directions.
• Where does the determination live? If the answer involves a care management note, a vendor portal and a spreadsheet, you have the decision but not the record. Under 42 CFR 422.102(f)(4) the record is the obligation.
• Can you produce the ineligible ones? The regulation requires documenting each determination whether eligible or ineligible. Most operational systems are built to record what was approved. The denials are what show the criteria were applied consistently rather than selectively.

The takeaway

The VBID sunset is usually discussed as a benefit design story, and on the surface it is a quiet one. Underneath, the basis on which a dual eligible member qualifies for a supplemental benefit changed from a status to a judgment, and the obligation to document that judgment landed on nearly every D-SNP in the same plan year.

Plans that treat this as a documentation problem to solve after the fact will spend 2027 reconstructing determinations they made in 2026. Plans that capture the determination when it is made will find the audit shorter and the next bid easier to price.

References

1. State of the 2026 Medicare Advantage industry: Dual-eligible plan valuation and selected benefit offerings. Milliman, March 2026. Approximately 90 percent of D-SNP members in plans offering VBID in 2025 against approximately 15 percent in plans with SSBCI, moving to approximately 90 percent SSBCI prevalence following termination. Figures are member weighted. Also the source for VBID socioeconomic targeting of dual eligible members and for narrower qualification under chronic condition criteria.
2. 42 CFR § 422.102. Supplemental benefits. Paragraph (f)(1) defines a chronically ill enrollee as one who has one or more comorbid and medically complex chronic conditions that is life threatening or significantly limits the overall health or function of the enrollee, has a high risk of hospitalization or other adverse health outcomes, and requires intensive care coordination. Paragraph (f)(1)(ii) carries the reasonable expectation of improving or maintaining the health or overall function standard.
3. 42 CFR § 422.102. Paragraph (f)(4) requires written policies based on objective criteria for determining enrollee eligibility, documentation of each eligibility determination whether eligible or ineligible, public listing of the written policies and objective criteria, availability of that documentation to CMS on request, and maintenance of the specific objective criteria without modification for the full coverage year.
4. Contract Year 2027 Medicare Advantage and Part D Final Rule. CMS fact sheet, April 2, 2026. Confirmed SSBCI changes are the use of objective processes to demonstrate enrollee eligibility and public posting of plan-developed SSBCI eligibility criteria.
5. 42 CFR § 422.504. Contract provisions. Paragraph (d) requires books, records and documentation to be maintained for 10 years. Paragraph (e) sets out HHS and Comptroller General inspection and audit rights for the current contract period and 10 prior periods.

Regulatory text verified against the electronic Code of Federal Regulations on 26 July 2026. This article is general information about benefit operations and is not legal or compliance advice.