Americans owe at least $220 billion in medical debt, and roughly 100 million adults carry some form of health care debt. Both figures come from KFF, and both are older than most people assume. The $220 billion estimate comes from a KFF analysis of Census Bureau Survey of Income and Program Participation data, published in 2022 and reflecting 2021. The 100 million figure comes from a KFF and NPR investigation, also published in 2022. Every number below is dated and attributed, because medical debt is a field where stale figures get recycled with fresh years attached to them.
The headline number, and what it excludes
The $220 billion KFF estimate counts what households reported owing. It is a floor, not a ceiling, and KFF has been explicit about why. The figure captures debt a household recognizes and reports as medical. It does not capture the bill that got moved onto a credit card, the balance rolled into a home equity line, the loan from a relative, or the payment plan a hospital set up internally. Once a medical bill changes form, it stops being counted as medical debt and starts being counted as consumer credit.
That accounting choice matters more than it sounds. Put a $4,000 emergency room bill on a credit card at 24 percent interest and you have not reduced what you owe. You have reclassified it into a category where the original cause disappears from the statistics.
How many people are affected
The KFF and NPR investigation put roughly 100 million adults as carrying some form of health care debt in 2022. That is a wider definition than the $220 billion balance figure, and that is deliberate. It includes people on payment plans, people who borrowed from family, and people carrying balances on cards specifically because of medical bills.
Two numbers with two definitions describing the same problem is not a contradiction. It is what happens when a category of debt has no single place it lives on a balance sheet.
Insurance does not settle the question
The common assumption is that medical debt is a problem for the uninsured. The premium data complicates that. KFF put the total annual premium for family coverage at roughly $25,000 in 2024, with the worker share above $6,000. That $6,000 is what comes out of a paycheck before a single deductible, copay, or coinsurance payment is made.
Set that against the U.S. Census Bureau figure of about $80,000 for median household income in 2023. The worker share of a family premium alone runs close to eight percent of a median household’s gross income, and that purchase does not buy the absence of a bill. It buys a different bill structure. For a household in the bottom half of the income distribution, a high deductible plan and no plan at all can produce a similar outcome after a serious diagnosis, which is the mechanism that puts insured people into the 100 million.
The Federal Reserve’s Survey of Household Economics and Decisionmaking has tracked a related question for years: whether households could cover a modest unexpected expense with cash. A meaningful share consistently report they could not. A hospital bill is not a modest unexpected expense.
What it does to a household balance sheet
Medical debt behaves differently from most consumer borrowing in one specific way. Nobody shops for it. A mortgage, a car loan, and a credit card balance all involve a decision made in advance with a price known at the time. A medical bill arrives after the service, often months after, frequently at a price the patient could not have learned beforehand and cannot verify afterward.
That sequence breaks the assumption behind most personal finance advice. Budgeting works on known quantities. It does not work on an obligation whose size is determined after the fact by a billing department. Telling a household to plan better for a $30,000 surgery it could not price in advance is not advice.
The wage side of the equation
The federal minimum wage is $7.25 an hour and has not moved since 2009, according to the U.S. Department of Labor. A full-time year at that rate is roughly $15,000 before taxes. Against the KFF family premium figure of about $25,000 a year in total cost, the arithmetic does not resolve. It is not a budgeting problem at that income. The obligation exceeds the income.
Most workers earn more than the federal floor. That is not the point. The point is that the floor sets the reference for the bottom of the wage distribution, and it has held still for more than fifteen years while health costs, housing costs, and childcare costs did not. Organizations working on affordability, including the nonpartisan 501(c)(3) Fight For A Living Wage, keep a running set of the underlying cost-of-living figures for exactly this reason: the wage number on its own says very little without the cost numbers next to it.
Reading these numbers carefully
Three rules make medical debt statistics more useful.
First, check the vintage. The $220 billion and 100 million figures are 2021 and 2022 data published in 2022. Anyone presenting them as current-year measurements is guessing, and you should discount the rest of what they say. Cost growth since then has not been zero, but the honest statement is that the most widely cited national totals are several years old.
Second, check the definition. Balance owed, number of adults affected, and share of households with debt in collections are three different measurements. They do not sum, substitute, or convert into one another.
Third, check the source. KFF, the U.S. Census Bureau, and the Federal Reserve produce the underlying data. A figure with no institution behind it, or one that traces back to a summary of a summary, should not be repeated.
What the data does not tell you
National aggregates say nothing about how the burden is distributed. A $220 billion total is compatible with many different distributions, and the policy implications differ sharply depending on which one is true. State-level variation in Medicaid eligibility, hospital charity care practice, and state collection law all shape who ends up in the total. Anyone drawing a conclusion about a specific state or a specific population from a national number is going past what the number supports.
What the data does support is narrower and still substantial. A very large number of American adults carry debt from health care they received. That debt is frequently invisible in consumer credit statistics because it has changed form. And carrying insurance reduces the chance of a catastrophic bill without removing it.
