Medicare Advantage is heading into 2027 with a striking split-screen: lower average premiums and broad headline availability, but fewer plan choices and a stronger push by major insurers toward networks that give patients less freedom to go outside them.

The shift matters because roughly half of people on Medicare now receive benefits through private Medicare Advantage plans. On October 1, UnitedHealthcare, CVS Health’s Aetna and Humana outlined 2027 offerings that emphasize health maintenance organizations, reduced service areas or a smaller national footprint. Reuters reported that UnitedHealthcare will leave some markets with heavy enrollment in preferred provider organizations, while Aetna is expanding HMOs and dropping coverage in two states. Humana said it will offer plans in about 2,600 counties, or just over 80 percent of U.S. counties, down from 85 percent in 2026.

Those announcements do not mean Medicare Advantage is broadly collapsing. They do show that a $0 or low premium cannot, by itself, answer the most important question for many beneficiaries: whether their doctors, hospitals and prescription drugs will remain covered on acceptable terms.

Lower premiums, fewer choices

The Centers for Medicare & Medicaid Services projects that the weighted average monthly Medicare Advantage premium will fall 16.5 percent, from $14.37 in 2026 to $12 in 2027. CMS estimates 34 million people will enroll, representing 47.4 percent of Medicare beneficiaries, although the agency says plan projections have historically understated final enrollment.

At the same time, the national number of plans is expected to edge down from 5,553 to about 5,532. CMS says more than 99 percent of beneficiaries will have at least one Medicare Advantage option and 97 percent will have 10 or more. About eight in 10 current enrollees should be able to stay in the same plan at the same or a lower premium.

Those measures describe availability, not sameness. A KFF analysis of CMS landscape files found that the average beneficiary will have 28 Medicare Advantage plans with drug coverage in 2027, down from 32 in 2026. Across all generally available Medicare Advantage plans, the average falls to 35 from 39. KFF also found sharp geographic differences: beneficiaries in nine states and the District of Columbia average fewer than 10 Medicare Advantage drug-plan choices, while those in 26 states and Puerto Rico average at least 20.

The premium projection is therefore genuine but incomplete. It does not measure deductibles, copayments, maximum out-of-pocket limits, drug formularies, prior-authorization rules or network breadth. Reuters’ review of the federal landscape data likewise found that insurers are concentrating on margin recovery rather than membership growth after higher medical use and rising costs pressured their Medicare businesses.

Why insurers are favoring HMOs

HMOs generally require members to use contracted clinicians and facilities except in emergencies. PPOs also have networks but usually cover some out-of-network care at a higher cost. For insurers, tighter networks can reduce prices and steer patients toward selected providers. For beneficiaries, they can trade flexibility for lower premiums and supplemental benefits.

UnitedHealthcare said about 95 percent of Medicare-eligible people in its footprint will have access to an HMO in 2027. The company also published a provider notice confirming service-area reductions and plan discontinuations effective January 1, with affected members receiving nonrenewal notices dated October 2. Humana’s 2027 announcement highlights $0 in-network primary-care visits and laboratory services in eligible plans, as well as dental, vision and hearing benefits, while limiting its footprint to 45 states and Washington, D.C.

Insurers say the redesign responds to rising medical and drug costs, increased use of care and federal payment changes. CMS, however, estimates its 2027 payment policies will produce a net average increase of 2.48 percent, or more than $13 billion, in payments to plans. Both statements can be true: an industry-wide increase does not ensure that every contract, county or benefit package is profitable, and the CMS figure incorporates several payment components rather than an insurer’s complete cost trend.

What the evidence says about network access

The best national network evidence predates the 2027 offerings, so it cannot prove that next year’s redesign will worsen access. It does establish why the change deserves scrutiny.

Using 2022 provider directories and Medicare physician data, KFF found that the average Medicare Advantage enrollee’s plan included 48 percent of the physicians available to traditional Medicare beneficiaries in the same area. One-fifth of enrollees were in plans covering no more than 32 percent of those physicians, while another fifth were in plans covering at least 63 percent. The study also found that PPOs had broader networks than HMOs on average.

That analysis measures directory breadth, not appointment availability, clinical quality or health outcomes. Directories can contain errors, and the presence of a physician does not show whether the doctor accepts new patients. KFF also cautioned that smaller networks may still be adequate, while a broad list may fail to include the particular specialist a patient needs. Its value lies in documenting wide variation—both between counties and among plans offered in the same county—not in establishing that every narrow network is harmful.

The Medicare Payment Advisory Commission’s March 2026 report describes the underlying policy bargain: Medicare Advantage plans can offer integrated drug coverage, supplemental benefits and a ceiling on annual out-of-pocket spending that traditional Medicare lacks, while enrollees accept provider networks and utilization management. MedPAC also noted that 55 percent of eligible beneficiaries were enrolled in Medicare Advantage in 2025 and that plans received an estimated $537 billion, excluding drug payments, making network performance a central issue rather than a niche concern.

Federal rules require network-based Medicare Advantage organizations to maintain enough appropriate providers to meet their members’ needs. CMS’s network-adequacy framework evaluates access against the pattern of care in a plan’s service area. Compliance with those minimums, however, is not the same as preserving an individual patient’s established clinician or preferred academic medical center.

The practical test comes during open enrollment

Open enrollment runs from October 15 through December 7. The evidence supports comparing total coverage, not assuming that a lower premium equals a better or worse plan. Current enrollees can use the Annual Notice of Change and Medicare Plan Finder to check whether a plan is ending, whether a replacement is an HMO or PPO, and how premiums, copays, drug coverage and annual spending limits will change.

Provider checks require more than a plan name. Beneficiaries with ongoing care should verify the specific physician, hospital and facility in the 2027 directory and confirm participation with the provider. People receiving complex treatment may also need to examine referral requirements, prior authorization and continuity-of-care provisions. Those comparisons are especially important for beneficiaries whose plan is being discontinued; KFF estimates that UnitedHealth and Humana together accounted for 46 percent of Medicare Advantage enrollment in 2026.

The 2027 market is not accurately summarized as either a premium victory or an access crisis. Federal projections point to lower average premiums and many available plans. Insurer filings and independent analyses point to fewer choices, selective withdrawals and tighter networks. The same design may benefit one enrollee through coordinated care and predictable costs while burdening another who depends on clinicians outside the contracted network.

The unresolved question is how those changes will affect timely access, continuity and out-of-pocket spending after enrollment begins. That outcome will require updated provider data, complaints, disenrollment patterns and claims—not marketing materials alone.