The Medicare Stack

I locked in Medicare this spring. Here's the full picture — the plan I chose, the coverage I'm layering on top for travel, what I considered and rejected, and where I'm honestly still uncertain. The whole thing is a stack of decisions, not one, and the parts interact in ways that aren't obvious when you walk in cold.

The short version: Original Medicare (Parts A and B), High-Deductible Plan G as my Medigap supplement, a zero-premium Part D plan, BCBS Global Solutions Single Trip Platinum for international coverage, and a standalone medical evacuation membership I've carried for years. Total annual cost lands around $6,600.

I'll walk through each choice, including the ones where a reasonable person would have gone the other way.

Ground Rules Before the Choices

Medicare has four moving parts and they don't combine intuitively.

Part A covers hospital stays, skilled nursing, and hospice. If you paid into the system for 10+ years, the premium is zero. Part B covers outpatient care — doctor visits, labs, imaging, almost everything that doesn't involve an overnight admission. The 2026 standard premium is $202.90 per month, with a $283 annual deductible. Medicare then pays 80% of approved charges and you pay the other 20% — with no cap. That last detail is what sends people running to buy a supplement.

Part D is prescription drugs. Part C is Medicare Advantage, which replaces Parts A and B with a private plan. I'll get to both.

Two things matter more than the premium itself. First, if your income is above $109,000 individual or $218,000 joint (based on your tax return from two years ago), you pay an IRMAA surcharge on top. It's a cliff, not a slope — one dollar over the threshold and your premium jumps a bracket. The first-bracket hit is roughly $81/month on Part B and $14/month on Part D, for the full year. That's a thousand dollars for a dollar of extra MAGI two years ago. If you're anywhere near a threshold, tax-loss harvesting and charitable contributions in the calendar year that matters for the lookback are worth real attention. Second, there's a critical enrollment window at 65 where you can buy a Medigap supplement without answering health questions. Miss that window, and switching later in most states requires medical underwriting. You may be declined.

That second point is the lever that shapes almost everything below.

Why I Didn't Take Medicare Advantage

Medicare Advantage is the fastest-growing part of the program — around 52% of enrollees are now in it. The premiums are low, sometimes zero, and the marketing is ubiquitous. I didn't seriously consider it.

Medicare Advantage is managed care. It works through a network and a defined service area, typically a metro region or a set of counties. If you live there, it can work fine. If you don't, or if you spend months on the road, it's the wrong product. Out-of-network non-emergency care ranges from expensive to uncovered. For someone who's been on the road for eleven years and intends to keep going, the product-market fit is zero.

There's a technical wrinkle worth knowing about here. CMS rules allow a Medicare Advantage plan to involuntarily disenroll you if you've been outside the service area for more than six consecutive months — some plans have "visitor" or "traveler" programs that extend tolerance to twelve months, but most don't. Whether plans actually enforce this against members who are traveling rather than permanently relocated is less clear. I haven't found documented cases of nomads getting kicked out for extended absence, and anecdotally, plans seem more focused on people who've moved than people who are roaming. But the risk isn't zero, the rule is real, and the consequence if it did happen is severe: you'd revert to Original Medicare outside your guaranteed-issue window for Medigap, which means underwriting, which means possible denial for the supplement you'd want. Worth understanding as a theoretical exposure even if it's not a practical epidemic.

There's a second reason that matters more than the travel one. Medicare Advantage works well for healthy people and breaks down for sick ones. That's a brutal sentence but it matches what the data and the regulators and the inspectors general have been saying for years. These plans have broad authority to require prior authorization, and they use it. They deny care that Original Medicare would pay for. They delay. They defer. They negotiate. A 2022 HHS Office of Inspector General audit found that Medicare Advantage plans were denying services that met Medicare coverage rules, and denying payments for care that should have been paid. That's not fringe reporting; it's the government auditing itself.

The cruel irony is that the product is cheapest when you don't need it and most expensive — financially and in every other way — when you do. Zero premium looks great until the moment a cardiologist says you need an imaging study and the plan says not yet. Every copay, every network restriction, every prior-auth form is designed to slow spending. When you're well, you never touch those mechanisms. When you're not, they become your full-time job.

I've heard the same wisdom from enough people to trust it: Medicare Advantage is a great deal right up until you actually need care, at which point it becomes a mess of delay, defer, deny, and copays. The premium savings don't survive contact with a real medical problem.

The honest counter is that if you live in one place, your doctors are in-network, you're healthy, and IRMAA pushes your Part B plus Medigap premium into painful territory, the cash-out-the-door comparison can favor Advantage. Maybe. My read is that the people making that comparison are assuming the healthy years will keep coming, and the whole point of insurance is the years when they don't. Skip it.

Medigap: Plan G vs. High-Deductible Plan G

Medigap plans are standardized. Every Plan G at every insurer covers the same things. What varies is price and customer service. Plan G covers essentially everything Medicare leaves unpaid except the Part B deductible — once you have it, your exposure for Medicare-covered care in the U.S. is close to zero.

High-Deductible Plan G is the same plan with one wrinkle: you pay $2,950 of Medicare cost-sharing out of pocket in a calendar year before the supplement kicks in. After that, coverage is identical to standard Plan G. Not similar. Identical.

So the question collapses to: does the premium saving pay for the deductible risk?

In my zip code (Jacksonville, Florida, for reasons I'll explain), the 2026 quotes are: Standard Plan G at $250/month ($3,000/year), HD-G at $60/month ($720/year).

The premium gap is $2,280. The deductible is $2,950. In a worst-case year I pay $720 in premium plus $2,950 in deductible — $3,670 total — which is $670 more than Plan G would have cost me that year. In a healthy year, I keep most of the premium savings. Four good years pay for one bad one.

The math says HD-G. I took HD-G.

Where the Math Gets Weaker

Three real counterarguments I want to name, not wave away.

The deductible drifts up fast. The HD-G deductible was $2,200 six years ago. It's $2,950 now — a 34% increase. It resets annually based on a government inflation calculation and insurers have no say in it. If healthcare inflation continues compounding faster than general inflation, the deductible keeps widening while the premium saving might not. I'm betting that the premium gap scales roughly with the deductible, but that's a bet.

Chronic conditions eat the deductible every year. If I'm diagnosed with rheumatoid arthritis, Crohn's, an autoimmune condition that requires ongoing infusions — any of the modern expensive drug regimens administered in an outpatient setting — I'd hit the deductible every year indefinitely. In that scenario HD-G would cost the deductible plus a small premium forever. One analysis I read made the reasonable point that for someone who develops a chronic utilization pattern, HD-G might cost tens of thousands more over a decade than standard Plan G. Worth respecting.

One nuance most people miss: the expensive drugs administered in a clinic — chemo infusions, many biologics, immunotherapy — are billed under Part B, not Part D. That means Medicare pays 80%, you owe 20%, and under HD-G your exposure is the full $2,950 deductible before the supplement picks up. The new Part D out-of-pocket cap doesn't help you here because these drugs aren't Part D drugs. HD-G turns the $2,950 deductible into your effective annual stop-loss for those treatments, which is actually useful — but only once per year, every year, as long as the treatment continues.

The switching trap. If HD-G becomes the wrong plan — because utilization has spiked — moving to standard Plan G in most states requires medical underwriting. By definition, the moment you want to switch is the moment you might not qualify. The choice I'm making is therefore more permanent than the premium math implies.

I'm going in with my eyes open. I'm reasonably healthy, I believe the asymmetric bet is the right bet in my situation, and I've identified a lever for the switching trap that I'll get to below.

And Plan N, Briefly — But With a Nomad-Specific Caveat

Plan N is usually pitched as the compromise between G and HD-G — lower premium than G, no large deductible, some office-visit copays, and one gap: Medicare Part B excess charges. Plan N leaves you on the hook for up to 15% extra when a provider doesn't accept Medicare assignment. Most of the "who cares, nobody charges those" takes I've read are written for people who stay close to home and use local providers who almost all take assignment.

A nomad is a different story. If I ever need to go somewhere specifically for treatment — to MD Anderson in Houston for a cancer workup, or to Mayo Rochester for something complicated — a nomad's natural move is to just pack up and go. That's a large part of the freedom. The destination health systems that draw patients from across the country are also the ones that most commonly don't accept Medicare assignment. Mayo Clinic in Arizona still bills excess charges. Mayo Rochester does not charge excess to Minnesota residents — state law prohibits it — but does charge them to patients from other states, which is to say, to patients like me. Mayo Florida has moved to accept assignment. MD Anderson's Texas hospital accepts assignment. The mapping is specific and it changes.

The point is that the people most likely to actually incur Part B excess charges are the ones willing to travel to the best destination hospital for a serious diagnosis. Nomads are exactly those people. "Plan N is fine, nobody charges excess" is advice written for someone whose treatment options end at the county line. If you'd drop everything and fly to Rochester or Houston for the right specialist, the excess-charge gap in Plan N is a real cost and Plan G (or HD-G) closes it.

Where the G-to-HD-G premium gap is narrow, Plan N still deserves a look. In my case the premium gap between G and HD-G is wide enough that N doesn't pencil out even ignoring the excess-charge angle. Your zip code might say otherwise. Get quotes for all three, and when you compare them, price the excess-charge gap against how you'd actually make decisions if you got a diagnosis that had a clear best-in-country specialist.

The Domicile Problem Nobody Warns You About

Here's the part that does not show up in any Medicare guide I've read. Medigap premiums are set by zip code. The variation is not small.

I ran Medicare.gov's plan comparison tool across Florida. Key West premiums for the identical HD-G product were materially higher than Jacksonville — same state, same insurer, same plan letter. The rate differences between zip codes were larger than most people's entire decision margin between Plan G and HD-G. If you're nomadic and choosing where to establish residency, choosing wrong costs you real money for the rest of your life.

A few mechanics that matter:

You need a real residential address, not a mail service. Commercial mail forwarding addresses — St. Brendan's Isle, Escapees, the big RV-community services — are flagged by insurers because everyone else uses them too. I've heard enough stories from nomads who got a Medigap application questioned or a claim hung up because the address on file pinged as a known commercial mail drop. This is a place to be defensive, not clever. I use a friend's genuinely residential address in Florida. I picked the friend partly for the zip code.

If you don't have the right friend, Savvy Nomad (savvynomad.io) is one option worth looking at. They provide a residential address in an RV/mobile home community rather than a mail drop, handle state residency filing, and offer a premium tier that includes a lease and utility bill. The catch: if five hundred other nomads are using unit numbers at the same underlying property, insurer software will eventually flag it as a commercial mail receiving agency regardless of the technical distinction. A real friend's house with a different last name on the mailbox is a quieter answer, if you have that option. A nomad-residency service is a reasonable answer if you don't, but don't assume it's invisible.

Commit to the state fully. Florida driver's license, Florida voter registration, Florida vehicle registration where applicable, estate planning documents executed under Florida law. A tax return pointing to Florida while everything else points to North Carolina is a mixed picture that creates exposure — both for Medigap purposes and, more consequentially, for state income tax. That second problem can be a bigger financial issue than the Medicare one if your home state decides you never actually left.

Understand the lever, but don't over-trust it. Three states — New York, Connecticut, and Vermont — allow Medigap plan switches at any time without medical underwriting. (Massachusetts has a narrower annual window, Feb 1 to Mar 31.) If the HD-G bet goes wrong later — if utilization spikes, if a chronic condition appears, if standard Plan G would clearly serve me better — I have the option to establish residency in one of those three states, switch plans, and then relocate back. For someone already living nomadically, the friction is much lower than it would be for someone with a house and a job.

But this is a harder maneuver than it reads. Guaranteed-issue protection in those states applies to actual residents — not tourists and not people whose mail happens to forward there. Establishing "intent to reside" generally takes a real lease, utilities, a driver's license change, voter registration, and a sustained physical presence. Doing that while managing a cancer diagnosis or the onset of a chronic condition is a genuine burden. The lever exists, but I'm not counting on being able to pull it gracefully. I'm counting on it as an option I can think about before things go wrong, not a rescue I can summon after.

Part D: Buy the Cheapest One and Reassess Every Year

Part D is the most flexible of the Medicare decisions and the one people most commonly overpay for.

The mental model: Part D plans cover the drugs on their formulary at the tiers they've set, with their deductible, for their premium. All four variables change every year. You can change your plan during open enrollment (October 15 to December 7) and the new plan starts January 1. This is a one-year commitment, not a lifetime commitment.

I take a handful of generics. None of them are expensive anywhere. I enrolled in a zero-premium Part D plan. The math: a Part D plan with a $40 monthly premium costs $480 a year; the same generics cost me around $4 each at Mark Cuban's Cost Plus Drugs or via GoodRx. The premium-plan math only works if you're on something that actually benefits from being in a formulary with decent coverage.

If I get prescribed something expensive mid-year, two moves are available. First, the plan has a formulary exception process — I can request that a specific drug be covered for the rest of the year. Second, at the next open enrollment I pick a plan whose formulary matches my current regimen. Medicare.gov's Plan Finder compares plans against your actual drug list and ranks them by total annual cost. It's one of the most useful tools the government runs.

One footnote that caught me late: there's a Part D IRMAA surcharge that tracks the Part B one — same thresholds, same two-year lookback. At the lowest IRMAA bracket it's about $14 a month, stacked on top of whatever the plan itself costs. It even applies to zero-premium plans. Annoying, but it's there.

One more thing worth knowing, and it's actually good news. The Inflation Reduction Act capped total out-of-pocket spending on Part D drugs at $2,100 in 2026, up from $2,000 in 2025. That's a hard federal cap on deductibles, copays, and coinsurance combined — once you hit it, you pay zero for covered Part D drugs the rest of the year. This makes the zero-premium-plan strategy safer than it used to be. Even if you get unexpectedly prescribed something expensive mid-year, your annual exposure on the Part D side is bounded. The only ways to get burned are if the drug isn't on the formulary at all (exception process handles that) or if it's a Part B drug, which doesn't count toward this cap.

International Coverage: Where I'm Still Adjusting

Medicare pays nothing outside the United States. The foreign travel emergency benefit in Plan G exists, but it's capped, has its own deductible, and only applies to emergencies that begin in the first 60 days of a trip. For anyone spending more than a couple of months abroad, it's not a coverage strategy.

I went with BCBS Global Solutions Single Trip Platinum. The decision felt clearer when I started writing it down than it feels now, so let me be honest about it.

What Single Trip Platinum is: a real international medical policy covering pre-existing conditions, with a $1,000,000 medical maximum and $500,000 emergency evacuation built in. Deductible options run from $0 to $500. It's designed for U.S. residents traveling abroad — which matters because it lets me keep my Florida domicile and Medicare intact. Quote range for my situation is around $2,900 a year, give or take.

What Single Trip Platinum is not: a 12-month policy. The product covers trips up to 182 days. BCBS allows what they describe as "a policy extension or a one-time consecutive re-enrollment while overseas," which in practice gets a second six-month period back-to-back. But the carrier's published materials describe the re-enrollment as a separate event, not a seamless extension of the original policy — and that distinction has real consequences.

Here's the thinnest point of failure in this whole plan, and I want to name it clearly. If I develop a condition during the first six-month policy — the kind of thing that would be a pre-existing condition on a new application — I don't know with certainty whether that condition is covered on the re-enrollment, or excluded from it. A true "medically necessary renewal" expat plan like Cigna Global cannot drop you or exclude a condition that arose during coverage. A "one-time consecutive re-enrollment" on a single-trip product might not offer that protection. The Platinum plan's pre-existing condition coverage at application is conditioned on having a primary U.S. health plan (Medicare, in my case), which satisfies the formal requirement — but whether a condition that first appeared under policy #1 is grandfathered into policy #2 is not something I've seen in writing. I've been told verbally that it works. I'm treating the verbal confirmation as useful and the written policy language as dispositive. Anyone following me through this decision should get the answer in writing, in the policy document, before wiring money.

So my current setup is: one six-month policy, then the consecutive re-enrollment for the balance of the year, then a brief return to the U.S. and a new policy cycle the following year. This matches the nomad pattern I already run, so it's not a meaningful lifestyle constraint.

One note for clarity: everything in this post is about my coverage. Lisa isn't 65 yet, so she's on COBRA from a former employer. That's a separate and time-limited situation, and none of these Medicare decisions apply to her today. When she ages in, we'll run this analysis again for her — and the numbers will be different, because Medigap premiums are priced by age as well as zip code.

I still consider this decision provisional. Cigna Global Individual Health is the main alternative — a true expat product priced around $2,700 a year with a $10,000 deductible. What Cigna Global gets you that BCBS may not: guaranteed renewability with no ability to drop you or exclude conditions that developed during coverage. That's the protection that matters most once you've been on the product for a couple of years. The reasons I didn't pick it: Cigna Global is built for expats with a fixed foreign address, not nomads; pre-existing condition terms at application are less favorable; and the renewability language, while marketed as lifelong, has details in the contract I'd want a second read on before committing. If the BCBS re-enrollment mechanics turn out to exclude conditions I develop during the first six months — or if our travel pattern shifts toward longer stays in fewer countries — Cigna Global moves back up the list immediately.

I want to be clear about one counter-argument to my whole international strategy: some travelers skip international medical insurance entirely, self-insure for small stuff, and carry only a medical evacuation membership. In much of the world, cash-pay medical care is so cheap that a catastrophic event is the only thing you're really insuring against, and evacuation handles the worst of that scenario by flying you to a country where quality and cost are both good. It's not a crazy position. I'm not taking it because I've seen too many "minor" situations escalate — a bike accident, a GI bug that turned into a hospitalization, a chest pain workup overseas. The premium buys me the ability to not negotiate care at the worst possible moment.

Medical Evacuation: Separate and Non-Negotiable

I've carried a standalone medical evacuation membership for years. That piece of the stack isn't new and isn't changing. Evacuation coverage is cheap relative to what a medical transport actually costs — a real air ambulance from the other side of the world runs well into six figures — and it's the one coverage that can't be reconstructed after the fact. You either have it or you don't.

The structure matters more than the brand. I want a membership that pays the transport bill directly rather than reimbursing me later, covers "medically necessary" transport rather than only life-threatening emergencies, and doesn't play games about which hospital they'll transport me to. There are a few operators that meet that bar. There are a lot that don't.

The Stack, Put Together

Here's what this costs annually, broken out by layer:

If I'd picked standard Plan G instead of HD-G, the total would run about $2,280 higher per year — $8,935. Over five healthy years, that's north of $11,000 in additional premium for coverage I'd statistically not fully use.

Bottom Line

Medicare feels complicated on the way in and isn't complicated once you understand the structure. There are four parts, one supplemental layer, an enrollment window that matters disproportionately, and a geographic component nobody talks about. That's it.

The delta between a good decision and an okay decision here is smaller than most people think. You're going to pay somewhere between $6,000 and $10,000 a year for this stack regardless of what you choose. The HD-G-versus-Plan-G decision is real money, but it's not life-changing money. The decision that would actually be life-changing is getting the initial enrollment window wrong, or committing to Medicare Advantage and discovering later that your doctors aren't in-network, or skipping international coverage on the theory that nothing will go wrong on the road.

The usual advice is to find a good independent broker. That advice is mostly right, with one caveat for anyone considering HD-G: brokers don't love writing HD-G because the premium is low and so is the commission. You may have to push through mild resistance to get HD-G quoted at all. That friction tells you something about whose interest the broker is optimizing for.

Make the choices. Don't assume the structure is more complicated than it is. And check your zip code before you pick a friend's couch to theoretically sleep on.