Is Medicare Advantage Truly Declining or Has It Simply Shifted Focus?

Mike Rawaan

15 minutes ago

8 min read

The headlines about the decline in Medicare Advantage over the past year are misleading.

Medicare Advantage did not stall. It changed who it serves. Between November 2025 and October 2026, Part C enrollment grew 2.4% to 36.0 million members, and the program held about 51% of all Medicare beneficiaries. But one product type delivered nearly all of the growth: Special Needs Plans. Strip out the SNPs, and standard Medicare Advantage lost ground to traditional Medicare.

The largest insurers drove that shift on purpose. They shut down money-losing plans and counties, trimmed supplemental benefits, and adjusted to a federal enforcement wave aimed at how they score patient risk. Think of a retailer that shutters its weakest stores and doubles down on the ones with the best margins. Revenue shifts from some aisles to others, and the headlines post the numbers without explaining why.

The numbers that matter

I compared CMS's November 2025 and October 2026 files from the Medicare Advantage/Part D Contract and Enrollment Data series. Four figures carry the story.

Part C grew 2.4%. Enrollment rose from 35.17 million to 36.02 million members, a gain of 854,000.
Medicare Advantage held its share. It moved from 50.9% to 51.1% of Medicare eligibles, while Fee-for-Service moved from 48.4% to 48.5%.
SNPs delivered 87% of the net gain. SNP enrollment rose 9.6%, from 7.74 million to 8.48 million. Non-SNP Medicare Advantage rose 0.4%, and its share of eligibles fell from 39.5% to 38.9%.
Traditional Medicare grew too. Fee-for-Service added roughly 780,000 members in my files. Chartis and HealthScape found a similar 600,000 gain on a February-to-February basis.

The payor map shifted harder than the totals. Humana added 1.38 million Part C members (+23.7%) and lifted its share from 16.5% to 20.0%. UnitedHealthcare lost 945,000 (-9.2%), and Elevance Health lost 340,000 (-15.2%). Kaiser Permanente added 87,000, and Aetna/CVS finished flat at -25,000. Centene, HCSC, BCBS of Michigan, Highmark and Florida Blue each lost between 11,000 and 123,000 members. Every payor outside that named group combined gained 1.02 million. Devoted Health alone more than doubled to 541,000 members, and SCAN added 152,000.

Cigna shows zero because it no longer files Part C contracts. Cigna sold its Medicare Advantage business to HCSC in March 2025 for $3.3 billion, and the former HealthSpring contracts now carry HCSC as their parent.

Method: payor figures come from the contract-level file and count Local CCP, Regional CCP, PFFS and MSA contracts, including employer group plans. Share and SNP figures come from the state-level and SNP files. Contracts with 10 or fewer members have been omitted.

December 2025 broke a 12-year streak

CMS's monthly enrollment series shows Medicare Advantage growing every month from January 2013 through November 2025. December ended that run. ACASignups' read of the CMS data shows Medicare Advantage slipping about 6,500 members to 35.62 million, while traditional Medicare added about 82,000 to reach 34.26 million. It marks the first month-over-month decline since at least 2013, the first month available in the CMS dataset.

Keep the scale in view. A 6,500-member dip equals 0.02% of the program, and the analyst who flagged it calls it possibly a data anomaly. The streak resumed right away: January added about 96,000 members, and February added another 159,000 while traditional Medicare shed 152,000.

I read December as a ratchet that clicked backward once. It did not signal a retreat, but it showed the program no longer climbs on autopilot. The insurers themselves expected worse. CMS reported last fall that carriers projected a decline, from 34.9 million to 34 million members in 2026. The actual count passed 35 million in February 2026. Members displaced by exits mostly landed in another Medicare Advantage plan, and SNPs absorbed the largest share.

Special Needs Plans carried the growth

SNP enrollment rose 9.6%, from 7.74 million to 8.48 million members. KFF finds the same pattern: SNPs delivered 85% of the net Medicare Advantage increase between 2025 and 2026. The CMS SNP Comprehensive Report lets me break that growth out by plan type.

SNP type	Nov 2025	Oct 2026	Change	% change
Chronic or disabling condition (C-SNP)	1,417,207	1,860,564	+443,357	+31.3%
Dual-eligible (D-SNP)	6,197,821	6,492,413	+294,592	+4.8%
Institutional (I-SNP)	128,620	131,017	+2,397	+1.9%
Total SNP	7,743,648	8,483,994	+740,346	+9.6%

Chronic-condition plans grew fastest. C-SNPs added 443,000 members and supplied 60% of all SNP growth. D-SNPs remain the base at 77% of SNP enrollment, down from 80%. I-SNPs held flat.

D-SNP integration level	Nov 2025	Oct 2026	Change	% change
Coordination-only (CO)	3,508,149	3,368,057	-140,092	-4.0%
Highly integrated (HIDE)	2,303,545	2,315,549	+12,004	+0.5%
Fully integrated (FIDE)	385,206	808,807	+423,601	+110.0%

Fully integrated D-SNPs more than doubled, and policy was a much bigger contributor to that move than consumer demand. CMS ended a Medicare-Medicaid demonstration, which accounted for 224,000 members across 27 contracts, at the end of 2025, and on December 31, 2025, its report to Congress says CMS helped move more than 178,000 members into integrated D-SNPs.

Carriers also chose to grow SNPs. KFF shows UnitedHealth Group lost 582,000 individual-plan members and 219,000 group members between February 2025 and February 2026, yet it added 267,000 SNP members. KFF also points to rebate payments for SNPs that have more than doubled since 2017, and to a 2018 law that made SNPs permanent. Insurers retreated from plans open to everyone and put their growth capital where federal payment and enrollment rules favor them.

Root cause: carriers chose margin over membership

Demand in the MA market is not cooling. The supply side is adopting discipline. Airlines call the same move capacity discipline: they cut the routes that lose money and fill the planes that remain. Medicare Advantage carriers did exactly that, and the proof is in the public records that show large insurers left markets over negative margins.

UnitedHealthcare: CEO Tim Knoll told investors the company would exit plans serving more than 600,000 members, mostly less-managed PPOs. For 2026, it withdrew from 109 counties and about 180,000 members.
Humana: It shed roughly 500,000 members in 2025 and cut its 2026 footprint from 89% to 85% of U.S. counties. For 2027, it plans exits that touch about 600,000 members, and it targets a pre-tax margin of at least 3% in 2028.
Aetna/CVS: CVS Health framed its pivot as margins over membership and warned that Aetna could lose up to 10% of its MA members in 2025.
Cigna left the business in 2024 and sold its MA book to HCSC.

KFF counted 3,373 individual MA plans for 2026, a 9% drop from 3,719, and found that 13% of MA-PD enrollees faced a plan termination, roughly double the prior-year share. Carriers cite two pressures: federal payment that grew slower than costs, and medical utilization that ran above their bids. The cuts hit PPOs hardest. Oliver Wyman finds that 20.8% of general-enrollment PPO members sit in plans that disappear for 2027, versus 8.0% of HMO members. That shouldn't surprise anyone since PPO members hold the broadest networks and plans have fewer levers to manage them.

Carriers did shrink, but they mostly moved capital to where margin lives. Humana still added more than 1 million MA members for 2026, and its growth concentrated in SNPs. That pattern explains why the November 2025 to October 2026 enrollment comparison shows total Part C up while general-population plans stall.

Benefits shrank alongside the footprint. Milliman measured average value added across general-enrollment plans down more than 8% from 2025 to 2026, with non-Medicare-covered benefits down $7 per member per month. KFF reports that the share of plans offering over-the-counter allowances fell from 73% to 66%, and meal benefits fell from 65% to 57%. A plan that offers less and covers fewer counties gives a beneficiary fewer reasons to pick Medicare Advantage over traditional Medicare, and that's what we saw in the December 2025 dip.

Risk-adjustment enforcement rewrote the coding playbook

Most commentary treats OIG reports and DOJ cases as headline risk. I treat them as a change in the price of a revenue line, and here's why: Carriers built a decade of margin on diagnosis capture, and every enforcement action raises the cost of that capture.

OIG audit trail: An HHS-OIG report from October 2024 estimated that diagnoses recorded only on health risk assessments and linked chart reviews drove $7.5 billion in 2023 payments across 1.7 million enrollees. Twenty companies accounted for 80% of that total, and in-home assessments produced almost two-thirds of it. CMS declined two of OIG's three recommendations.
DOJ civil enforcement: Kaiser Permanente affiliates agreed to pay $556 million in January 2026 to resolve False Claims Act allegations that they pushed record addenda to add diagnoses, tied incentives to risk scores, and left unsupported diagnoses in place. DOJ tied the conduct to about $1 billion in payments from 2009 to 2018.
UnitedHealth: The Wall Street Journal reported a DOJ criminal probe in May 2025, and UnitedHealth disclosed that it cooperates with the investigation. CEO Stephen Hemsley then ordered a comprehensive review of how the company assigns risk scores. A Senate Judiciary Committee report in January 2026 described UnitedHealth's aggressive diagnosis-capture strategy.
CMS posture: The Trump administration announced a plan to expand and speed up MA audits in May 2025, and CMS opened an online provider complaint form on January 5, 2026.

I cannot show a clean line from any single investigation to lost members, and anyone who claims one is simply overreaching. UnitedHealth cut 582,000 individual-plan members, but V28, utilization, and exits all landed in the same window. Coincidence? I think not. The strategy shift is visible even where the attribution is not. Carriers that once treated in-home assessments and retrospective chart reviews as a growth engine now face audit, litigation, and a CMS rule that stops counting unlinked chart reviews in 2027. A plan that cannot lean on aggressive coding must earn its margin from benefit design, network management, and care delivery. Carriers that cannot do that exit the market.

Payment policy squeezed the margin from four sides

Enforcement targets what plans did. Payment policy changes what plans can afford to do. Four policy moves hit the same bids at once.

V28 reached full phase-in. CMS replaced the risk model in 2024 and fully implemented V28 in 2026. UnitedHealth executives have acknowledged they underestimated the effect, and the model cut payment for the diagnosis-heavy members carriers once prized.
Rate notices ran thin. For 2027, CMS floated a near-flat increase, then finalized 2.48%, about $13 billion more than this year. Insurers had expected 4% to 6%, and more than 100 organizations urged CMS to raise the rate, citing expected plan exits. CMS also delayed another model recalibration, yet it still expects risk scores to climb about 2.5% on the underlying coding trend.
Part D redesign shifted risk to plans. This one often falls under the radar. The Inflation Reduction Act capped beneficiary drug spending and moved more liability to the plan. Milliman finds that Part D coverage got leaner for 2026, with higher deductibles and brand copays that converted to coinsurance. Avalere Health offers a great analysis on the out-of-pocket impact.
Stars and demonstration programs ended. Humana says its 2027 exits skew toward plans rated 3.5 stars or lower, although executives say stars did not drive the decisions. CMS also closed its MA Value-Based Insurance Design model at the end of 2025, and Becker's reports that plans spent an estimated $86 billion on supplemental benefits in 2025.

Each move looks manageable alone. Together they leave a carrier two choices: cut benefits and markets, or lose money. No surprise that in a capital economy, carriers chose the former, and beneficiaries saw it as a thinner plan that costs the same or more.

The takeaway

Medicare Advantage did not stall. It sorted itself. Carriers dropped the PPOs, rural markets, and rich supplemental benefits that lost money, and they put their capital into special needs plans, where federal payment and enrollment rules favor them. The December 2025 dip marks the moment the sort became visible in the national total.

If executives plan their strategy around the old growth curve, they are planning for a market that no longer exists. Keep your eyes on non-SNP enrollment against traditional Medicare, because that line will tell you whether the general-population product still wins on value.

Before you go, check out my Podcast, The Outer Shell Podcast from Covalence Health, on Apple Podcasts, Spotify, or YouTube.