How Much Does Long-Term Care Really Cost?

Long-term care is the houseguest nobody invited, nobody planned for, and who somehow ends up with the biggest bill at the end of dinner.

It is the largest retirement expense most Americans never write down. Not because it is small. Because it is awkward.

Here is the headline number everybody quotes: a private nursing home room now runs a national median of $129,575 a year, per CareScout's 2025 Cost of Care Survey.

And here is the thing about that number.

It is basically useless.

That is a completely different problem, with a completely different answer. Let us go build it.

CareScout surveyed providers from July to November 2025, contacting more than 211,000 long-term care providers and completing roughly 16,000 surveys. Here is the 2025 board:

Look at that adult day care line for a second. Twenty-five grand a year, which is the cheapest thing on the board by a mile, and almost nobody talks about it. It is the service most families discover third, after they have already burned money on options one and two.

Now the warning label: these are national medians. Not quotes. Not promises. Your actual number depends on your state, your city, your hours, your facility and, above all, your duration.

Everyone says it. Almost nobody prices it.

Home care is genuinely cheaper when you need a little help. It becomes the most expensive option on the entire board when you need a lot. At $35 an hour, the math turns on you fast:

Read the bottom rows again. At 80 hours a week, staying home costs more than the most expensive nursing home in the country's median. And 80 hours is not even full coverage. That is under half a week.

Which brings us to the thing that blindsides families:

24-hour care is not an hourly rate problem. You cannot hire one heroic person to work 168 hours a week. You need shifts, which means multiple caregivers, agency margins, overtime structures, minimum-hour requirements and a scheduler. Multiplying $35 by 24 by 365 is a fun party trick, not a quote.

So if your plan is "age in place," go do one piece of homework: estimate the hours. Ten is a plan. Sixty is a portfolio event.

Assisted living sits between independent living and a nursing home. Your own apartment, plus help with personal care, meals, transportation and medications depending on the community and your needs.

National median: $6,200 a month, or $74,400 a year.

Now the catch that costs families thousands.

The monthly rate is frequently not the bill. Many communities include some services in the base fee and charge separately for higher care levels and add-ons. CareScout specifically tells you to read the residency agreement for additional care charges, fees, deposits and the rules governing future price increases.

So $74,400 is a benchmark, not an invoice. Your actual number can be meaningfully higher.

But here is the part that goes the other way, and almost every panicked family gets this wrong:

Assisted living is not $74,400 of brand new spending. It replaces a bunch of spending you were already doing.

So the real question is not "can my portfolio produce $74,400." It is "can my portfolio produce the difference," which in this illustration is closer to $48,000.

That is still a lot. It is also 36% less terrifying than the headline, and it changes what a family decides to do.

Semi-private: $315 a day, $114,975 a year. Private: $355 a day, $129,575 a year.

Multiply those by duration and the whole retirement conversation changes:

Those assume prices never move, which they will. But notice the shape of the risk.

The difference between one year and five years is half a million dollars. Same person. Same facility. Same daily rate. The only variable that changed is time.

Which is exactly why "I could cover a year" is such a dangerous thing to tell yourself. Nobody needs care for a year on purpose.

Here is where the statistics get genuinely interesting, and where most scare-marketing conveniently stops.

A widely cited HHS and Urban Institute analysis found that about 52% of Americans reaching 65 will develop a disability serious enough to need long-term services and supports. Roughly a coin flip.

But: most will need help for less than two years. And about one in seven will face a disability lasting more than five years.

A more recent HHS analysis from January 2025 frames it the same way, and adds the number worth tattooing somewhere: about one in five Americans turning 65 today will have long-term care costs exceeding $200,000.

So there are really three futures:

This reframes the whole exercise. You are not planning for a certainty. You are buying protection against a tail.

The 2015 HHS and Urban Institute model estimated Americans turning 65 would incur an average of about $138,000 in future long-term services and supports costs, while noting most people need help for a relatively short period. It also estimated about 17% would spend at least $100,000 out of pocket.

How can the average be $138,000 if most people spend far less?

Because averages are what happens when you put a few catastrophes in a blender with a lot of nothing.

The average and the median are off by a factor of eleven. Not one person in that group actually spent the average.

Planning around the average means over-preparing for the seven people who barely need anything and badly under-preparing for the one who needs everything.

Which is why you build scenarios, not a single number.🚫 The single most expensive misunderstanding in America

Say it with me. Medicare does not pay for long-term care.

Medicare.gov says it flatly. Long-term care, also called custodial care or long-term services and supports, is not covered. That includes help with bathing, dressing and using the bathroom.

What Medicare does cover is short-term skilled care after qualifying circumstances. For 2026 that looks like this:

Do the arithmetic on days 21 to 100. That is 80 days at $217, which is $17,360 out of your pocket during the stay everyone believes is "covered."

And then on day 101 the cliff arrives. Not because you got better. Because the calendar ran out.

Here is the version that actually ruins people, and it happens constantly: a parent is in a facility, Medicare is paying, everyone exhales. Then the care is reclassified from skilled to custodial. Nothing about the parent changed. The category changed. And the bill lands on the family with about 48 hours of warning.

Medicaid is the actual long-term care system in this country, whether or not anyone designed it that way.

KFF reported that Medicaid paid 44% of long-term institutional care costs in 2023, while 37% was paid out of pocket and 18% came from other public and private sources. KFF's 2024 data show roughly $185.1 billion in fee-for-service Medicaid spending across nursing facilities, ICF/IID facilities and home health and personal care in fiscal 2024.

That is not a safety net. That is the main floor.

But Medicaid is not a birthday present you get for turning old. Eligibility is financial and functional, and the rules vary by state. In practice, an enormous number of people pay privately, spend down, and then qualify.

"Medicaid will cover me eventually" is a description of what happens after your money is gone. It is not a funding plan. It is the absence of one.

Here is the part families find out at the worst possible moment.

Federal rules require states to seek estate recovery for certain Medicaid benefits received by people 55 or older, including nursing facility services and certain home and community based services.

Now, calm down. This does not mean "Medicaid takes everyone's house." There are real protections and exceptions, including rules involving surviving spouses and certain children, and states must have procedures for hardship waivers. Implementation varies enormously by state.

The planning point is simpler than the fear: public assistance can have estate consequences. Learn how your state handles it before you assume Medicaid quietly substitutes for private money with no downstream effect.

Long-term care for couples is a fundamentally different puzzle, because one person can need $125,000 a year of care while the other still needs to buy groceries and keep the lights on for another 20 years.

Medicaid has federal spousal impoverishment protections designed for exactly this. CMS rules allow certain amounts of income and resources to be protected for the spouse remaining in the community, and CMS issued updated 2026 standards in April 2026.

So the question is never just "what does the nursing home cost." It is:

"How much income and assets does the healthy spouse need to keep living a normal life?"

Consider two households facing the same $125,000 bill:

Same number. Completely different emergency.

There is no single American long-term care market. There are fifty of them, and they do not agree.

CareScout's 2025 state medians for private nursing home rooms ran from roughly $7,604 a month in Texas at the low end of its published examples to $18,488 a month in Oregon at the high end.

A gap of more than $130,000 a year for the same broad category of care. Over three years, that is nearly $400,000 of difference produced by nothing but geography.

Which means your long-term care risk is partly a location decision you are making right now, probably without realizing it. And it interacts nastily with the other thing people plan around location for: being near family. The cheap state is often not where your kids are.

You may not need care for 10, 20 or 30 years. The $129,575 will not be waiting patiently at $129,575.

Illustrations, not forecasts. Actual long-term care inflation has bounced around a lot. CareScout reported that in 2025 nursing home private room costs rose just 1% nationally while assisted living rose 5%, describing 2025 as showing early signs of moderation after the sharp post-pandemic increases.

The lesson is not "assume 4%." It is: pick an explicit assumption and write it down, instead of quietly freezing today's price for three decades and calling it a plan.

People underestimate this expense because they file it mentally under "medical," and they have insurance for medical.

But long-term care is mostly not treatment. It is help with living.

Picture someone medically stable who simply cannot safely bathe, dress, cook, use the bathroom, move around the house, manage medications or be alone. Nothing is being treated. Everything needs doing.

Insurance pays for treatment. Nobody is treating anything here. That single category error is why Medicare and ordinary health insurance leave this risk completely untouched.

Stop asking what care costs. Ask what you would have to fund.

Care cost, minus reliable income, minus insurance and benefits, equals your gap.

Look at rows one and two. Same person, same community, same year. Counting the expenses that disappear when you leave the house takes the gap from $29,400 to about $3,000.

That is not accounting sleight of hand. That is the difference between a family that panics and a family that plans.

And then look at row three. That is the tail risk in one line: to fund a private nursing home indefinitely from a portfolio alone, you need roughly $2.1 million. Which is exactly why almost nobody funds it from a portfolio alone, and why duration matters more than price.

For most American families, this is simply true, whether or not anyone says it out loud.

A household with $500,000 invested and a $500,000 home has a completely different care picture from a household with $500,000 invested and no equity. Same portfolio. Very different resilience.

And there is a quiet efficiency here people miss: if one person moves into assisted living, the house does not need to keep being a house. It can be sold, rented, or borrowed against. The asset that was costing you $26,000 a year to maintain becomes the thing that pays the new bill.

That does not mean the house is automatically care money. It means long-term care planning that ignores the balance sheet and stares only at the brokerage account is planning with one eye shut.

The biggest reason Americans underestimate this cost is that a huge share of care is delivered by family members, and nobody invoices for it.

A spouse absorbs the household tasks. A daughter shows up three times a week. A son becomes the medical transport department. Somebody moves in.

None of that shows up on a long-term care bill. All of it has a cost.

HHS research has emphasized that long-term care financing involves not only formal paid care but substantial unpaid care from family and friends.

So "my family will help" is a legitimate part of a plan. It is not a substitute for one. Somebody is still paying. The payment is just denominated in someone else's career.

Traditional long-term care insurance exists precisely to move this risk off your balance sheet. It can cover qualifying care at home, in assisted living, in nursing facilities and elsewhere depending on the policy. The NAIC notes policies differ significantly in covered services, benefit limits, inflation protection, premiums and other provisions.

Which means a $200,000 benefit is not a $200,000 benefit. Check all of this:

On inflation protection, run the numbers. A $200 per day benefit sounds respectable today. Twenty years out against 3% care inflation, it covers a fraction of what it appears to:

A policy without inflation protection does not shrink. The world around it grows. Same outcome.

The other route is self-insurance, which is a fancy way of saying "keep the risk." Bigger portfolio, more liquidity, assets you could sell. If care never happens, the money stays yours or your heirs'. If it does, you eat all of it.

Notice the middle row of "best for." Self-insurance genuinely works at both ends of the wealth spectrum and works worst in the middle, where there is enough money to lose and not enough to absorb it. That is the group that most needs to actually make a decision, and most often makes none.

Forget the universal targets you see online. There is no correct amount to set aside, because the same $150,000 event lands completely differently on different households.

Instead, run these three:

Then ask one question for each: what would this do to the rest of my retirement?

Because that is the real test. Watch what care does when it lands on top of ordinary spending:

A plan that was perfectly sustainable at 5% is drawing at 18%. That is not a budget problem. That is a countdown.

That last one is the quiet killer. Every good option in this article requires being early. Not one of them can be executed from a hospital discharge meeting.

Here is the range to carry in your head:

But the annual number was never the story. Roughly half of people reaching 65 will need care. Most will need it for under two years. About one in five will face more than $200,000, and about one in seven will need care for more than five years.

So the long-term care problem is not that every retiree spends half a million dollars. It is that a meaningful minority faces a bill large enough to rewrite everything, and none of them knew in advance that they were the minority.

Four questions get you most of the way to a plan:

Answer those and the headline price finally becomes useful, because you will be looking at the only number that was ever yours: what remains after income, insurance and benefits, and what that remainder does to the retirement you spent forty years building.

See you next issue. 🪙

This is general education, not financial, tax, legal, or medical advice. Cost figures are national or state medians from published surveys, not quotes, and actual pricing varies enormously by provider, city, care level and year. All inflation, withdrawal, portfolio and offset illustrations are hypothetical and not forecasts. Medicare, Medicaid, spousal impoverishment and estate recovery rules are detailed, change over time and differ by state. Insurance policy terms vary significantly. Verify anything you would act on with Medicare.gov, your state Medicaid agency, a licensed elder law attorney and a licensed insurance or financial professional.

Sources: CareScout 2025 Cost of Care Survey; Medicare.gov long-term care coverage rules and 2026 Part A skilled nursing facility cost sharing; U.S. Department of Health and Human Services and Urban Institute long-term services and supports research, including the January 2025 policy analysis; KFF Medicaid long-term care spending and payer share data; CMS spousal impoverishment standards for 2026; federal Medicaid estate recovery requirements; National Association of Insurance Commissioners long-term care insurance guidance.