CreditJoshua Lott for The New York Times
In the first congressional hearing held on “Medicare for all” in April, Michael Burgess, a Republican congressman from Texas and a physician, called such a proposal “frightening” because it could limit the treatments available to patients.
The debate over Medicare for all has largely focused on access and taxpayer cost, but this raises a question that hasn’t gotten much attention: What treatments would it cover?
A good starting place for answers is to look at how traditional Medicare currently handles things. In one sense, there are some important elements that Medicare does not cover — and arguably should. But a little digging into the rules governing treatments also reveals that Medicare allows a lot of low-value care — which it arguably should not.
Many countries don’t cover procedures or treatments that have little medical value or that are considered too expensive relative to the benefits. American Medicare has also wrestled with the challenge of how to keep out low-value care, but for political reasons has never squarely faced it.
You might remember the factually misguided “death panel” attack on the Affordable Care Act, which preyed on discomfort with a governmental role in deciding what health care would or would not be paid for. (This discomfort also extends to private plans, exemplified by the backlash against managed care in the 1990s.)
Perhaps as a result, Americans don’t often talk about what treatments and services provide enough value to warrant coverage.
Two classes of Medicare
You can divide current Medicare coverage into two layers.
The first is relatively transparent. Traditional Medicare does not cover certain classes of care, including eyeglasses, hearing aids, dental or long-term care. When the classes of things it covers changes, or is under debate, there’s a big, bruising fight with a lot of public comment. The most recent battle added prescription drug coverage through legislation that passed in 2003.
Over the years, there have also been legislative efforts to add coverage for eyeglasses, hearing aids, dental and long-term care — none of them successful. Some of these are available through private plans. So a Medicare for all program that excluded all private insurance coverage and that resembled today’s traditional Medicare would leave Americans with significant coverage gaps. Most likely, debate over what Medicare for all would cover would center on this issue.
But there is a second layer of coverage that receives less attention. Which specific treatments does Medicare pay for within its classes of coverage? For instance, Medicare covers hospital and doctor visits associated with cancer care — but which specific cancer treatments?
This second layer is far more opaque than the first. By law, treatments must be “reasonable and necessary” to be approved for Medicare coverage, but what that means is not very clear.
Local versus national
We think of Medicare as a uniform program, but some coverage decisions are local. What people are covered for in, say, Miami can be different from what people are covered for in Seattle.
Many treatments and services are covered automatically because they already have standard billing codes that Medicare recognizes and accepts. For treatments lacking such codes, Medicare makes coverage determinations in one of two ways: nationally or locally.
Although Medicare is a federal (national) program, most coverage determinations are local. Private contractors authorized to process Medicare claims decide what treatments to reimburse in each of 16 regions of the country.
In theory, this could allow for lots of variation across the country in what Medicare pays for. But most local coverage determinations are nearly identical. For example, four regional contractors have independently made local coverage determinations for allergen immunotherapy, but they all approve the same treatments for seasonal allergy sufferers.
There are more than 2,000 local coverage determinations like these. National coverage decisions, which apply to the entire country, are rarer, with only about 300 on the books.
Clinical trials
When Medicare makes national coverage decisions, sometimes it does so while requiring people to enter clinical trials.
It has been doing this for over a decade. The program is called coverage with evidence development, and its use is rare. Fewer than two dozen therapies have entered the program since it was introduced in 2006. But it allows Medicare to gather additional clinical data before determining if the treatment should be covered outside of a trial. To be considered, the treatment must already be deemed safe, and it must already be effective in some population. The aim is to test if the treatment “meaningfully improves” the health of Medicare beneficiaries.
Only one therapy (CPAP, for sleep apnea) that entered this process has ever emerged to be covered as a routine part of Medicare. The others are in a perpetual state of limbo, neither fully covered nor definitively not covered. CAR-T cell therapy, a type of cancer immunotherapy, which appears to be very successful but is also very expensive, is one of the most recent to enter this process.
Unwillingness to talk about cost-effectiveness
Despite the complexity of all these coverage determination methods — local, national, contingent on clinical trials — the bottom line is that very few treatments are fully excluded from Medicare, so long as they are of any clinical value. And this suggests that it’s not very likely that Medicare for all would deny coverage for needed care.
A 2018 study in Health Affairs found only 3 percent of Medicare claims were denied in 2015. And traditional Medicare doesn’t limit access to doctors or hospitals either, as it is accepted by nearly every one. (This is in contrast with Medicare Advantage.)
Medicare has a troubled history in considering cost-effectiveness in its coverage decisions. Past efforts to incorporate it have failed. For example, regulations proposed in 1989 were withdrawn after a decade of internal review.
As a result, Medicare covers some treatments that are extremely expensive for the program and that offer little benefit to patients. The Medicare Payment Advisory Commission recently studied this in detail. In a 2018 report to Congress, it noted that up to one-third of Medicare beneficiaries received some kind of low-value treatment in 2014, costing the program billions of dollars. If Medicare for all followed in traditional Medicare’s path, it could be wastefully expensive.
The United States has had a historical unwillingness to face cost-effectiveness questions in health care decisions, something many other countries tackle head-on. Some Americans favor Medicare for all because it would make the system more like some overseas. And yet, in choosing not to consider the value of the care it covers, Medicare remains uniquely American.
Elsa Pearson, M.P.H., is a policy analyst with the Boston University School of Public Health. You can follow her on Twitter at @epearsonbusph.