Michael Chernew, PhD, chairs the influential Medicare Payment Advisory Commission (MedPAC), an independent 17-member panel appointed by the U.S. Comptroller General to advise Congress on Medicare policy.
MedPage Today recently interviewed Chernew by phone and email about growing concerns — throughout the country and among commission members — that private Medicare Advantage plans in some ways may be failing their mission to improve access to quality care and lower costs. These plans have become the dominant way beneficiaries receive health benefits.
Chernew, who is also a health policy professor and researcher at Harvard Medical School, said he was speaking from that role.
The interview has been edited for clarity and brevity.
Cheryl Clark: Medicare Advantage plans have been enrolling beneficiaries at astonishing rates, and plans now cover 52% of eligible lives, or almost 32 million people — up from 11 million in 2010 and 33% in 2016. At this pace, 70% or 80% of eligible people will be in an MA plan in a few more years. Eventually, will everyone be required to enroll in a Medicare Advantage plan to access benefits?
Michael Chernew, PhD: For the foreseeable future, I think traditional Medicare will still be an option, but just one that fewer people choose. The pace will slow as fewer people are left to convert. But there are now markets that — in Puerto Rico for example — 92% are in Medicare Advantage plans. There are already other markets outside of Puerto Rico that are 70% and 80%.
Clark: The idea behind MA plans was to control waste, fraud, and abuse, discourage low-value services, and avoid unnecessary and potentially harmful care. But there are growing concerns that Medicare Advantage plan companies have become too big and powerful, with the profit motive often being the determining factor in whether patients get care their doctors say they need. Some have called Part C a Frankenstein. Last year, we began seeing a growing number of large physician groups and whole health systems saying they won’t take some or all of the plans because of prior authorization delays and huge losses.
Chernew: Let me broaden your question. We’ve had HMOs in Medicare since 1973. And the motivation, much as you stated, is that there’s a lot of wasteful care in the American healthcare system. Health plans have the ability to both eliminate that waste, coordinate care, and engage beneficiaries to support better health. In Medicare, private plans do provide care more efficiently, avoiding fragmentation. There’s no way to think about the whole patient in a fee-for-service system.
And what I think is increasingly important is that payments to Medicare Advantage plans get funneled in part to beneficiaries so they have more generous coverage than the traditional Medicare program, which is not a particularly generous benefit package.
Clark: Do you mean like hearing and vision coverage, transportation to the doctor, and meals? Benefits that may not have a direct medical effect?
Chernew: Right, but more basic benefits, like lower outpatient cost sharing and lower deductibles for inpatient care — the gaps in traditional Medicare.
Clark: But are you saying it’s worked out the way it’s supposed to? Your staff study noted that MA plans have cost the Medicare program $613 billion more than beneficiaries in fee-for-service over the last 18 years, and $353 billion of that is in just the last 5 years, $88 billion more expected in 2024 alone. These private plans have been using the risk adjustment factor to score patients 20% sicker than their counterparts in fee-for-service, which means higher capitated revenue. And some people are concerned that we really don’t know if those extra benefits — that free transportation to the doctor or that lower copay for a dental visit that some of this money is paying for — have improved health status.
Chernew: I believe the Medicare Advantage program has successfully changed patterns of care in ways that have reduced overall utilization. They have enabled plans to offer better benefits to beneficiaries, financed by the plans’ efficiencies as well as the payments they’ve received. And there’s a range of technical things — changing the risk adjustment models, which we have already done — and making changes to the quality bonus program. There’s a lot of strategies one could take if one wanted to improve the balance of payments between Medicare Advantage and fee-for-service.
Clark: The Office of Inspector General’s reports have found so much upcoding, even fraud, in many of the leading plans that have deceived the trust fund into paying more than the patients’ actual diagnoses called for. How do we stop that?
Chernew: Although there has been some documented overly aggressive coding — and in some cases I’m sure it was downright fraud — the fee-for-service coding system is imperfect too. Medicare Advantage plans will have more intense coding just because they’re more accurate.
For example, a lot of fee-for-service patients who have diabetes coded in one year might not have that reported in a subsequent year. That the MA plan is more accurate in coding is not fraudulent, but it does contribute to this imbalance between payments. I wouldn’t characterize it as malicious as your question implied. I think on average, Medicare Advantage plans probably do a good job – my guess is a better job — than fee-for-service in treating beneficiaries with chronic conditions.
Clark: Let’s shift to some of the unintended or perhaps unappreciated consequences of majority MA enrollment. Physicians at large academic hospitals that specialize in certain conditions or diseases are excluded from MA networks. Patients can’t get a consult at the Mayo Clinic in Arizona and Florida, and they can’t get cancer care at Sloan Kettering. I have friends who have had symptoms that doctors in their MA networks couldn’t diagnose, but the expert was at UC San Diego. They had to fork over $900 for an out-of-network visit. What do we say to these centers when 70% or 80% of the Medicare population can’t benefit from their expertise? I don’t think many realize that when they enroll.
Chernew: This is a complicated question. Medicare Advantage involves a trade-off. Individuals choose to have a narrow network. The challenge is for CMS to make sure people have good quality care outside of academic medical centers. It’s not the case that if you don’t go to an academic medical center, you’ll get substandard care. But it is complicated when it’s a rare diagnosis.
And academic medical centers could be in network if they negotiate with the plans. But I have a hard time seeing how one accomplishes what one wants to accomplish in the Medicare Advantage space if one requires all the academic medical centers to be in a network, or some version of that.
That said, there are some areas that merit some attention; there’s a penalty if you try to switch out of a Medicare Advantage plan because you could be underwritten in the Medigap market and it may be more expensive to get a Medigap plan.
Clark: It’s not a penalty. It’s often a rejection, right? There are 46 states that allow underwriting for supplemental plans and that usually means getting rejected for common health issues like uncontrolled diabetes or a recent hospitalization.
Chernew: I think this is going to become a growing policy issue. The problem is, if you allow people to switch back to a supplemental plan as soon as they need care, you create adverse selection. If everybody chooses the cheaper MA plan when they’re healthy, and the supplement plans when they’re sick, then the whole market collapses. In the states that prohibit underwriting, supplemental plan premiums are more expensive.
Also, the premise that the fee-for-service system has better quality and is more efficient is actually not true. It’s easy to tell stories of people in the fee-for-service system who didn’t get the care they wanted, got too much care, and were harmed in some way. Also, in Medicare Advantage there’s a lot of heterogeneity. Some plans do a better job than other plans.
Clark: Another consequence of MA plans dominating the market is that there are certain metrics, like readmissions and hospital-acquired conditions, by which we score quality of care for everybody. But those scores are drawn from care provided only to fee-for-service beneficiaries. If those scores are based on care to only a few million patients, can they still be a reliable index to score care for everybody?
Chernew: We’ll need to think about how robust those systems are for a growing MA system. Some of those measures you mention — readmissions for example — are not the best quality measures. Some other programs like MIPS (the Merit-Based Payment Incentive Program) are really lacking.
Clark: Traditional Medicare beneficiaries usually subscribe to a Part D plan, but as more beneficiaries choose Medicare Advantage plans with Part D benefits, the number of companies offering Part D plans is declining. Of course, choosing a Part D plan is impossibly confusing and seniors have to do that every year, so maybe it’s a good thing that there will be fewer Part D options?
Chernew: Of the many concerns I have about Medicare Advantage growth, the impact on the Part D market is low on my list. The other issue, how people choose the right plan, is a bigger concern. The benefits are hard to figure out and systems are hard to compare. It is hard to choose a plan. And even with MA plans with drug benefits, you have to figure out the same issues. You have to figure out the networks, what the benefits mean, and then you take a particular benefit, like dental. There’s a wide range of details.
Clark: I am well aware of the confusion and frustration. I know people who were in tears last fall trying to navigate the system after Scripps Health medical groups told 32,000 MA enrollees they weren’t contracting with any MA plans in 2024 because they were losing too much money. People I’ve known for years were devastated, and some didn’t even realize they were in an MA plan.
Chernew: When that happens, beneficiaries can choose a different MA plan. But that’s asking people to do a lot. They have to find new doctors and go through a lot of paperwork.
Clark: We’ve heard a lot about MA enrollees not being able to figure out which doctors are in network and are actually accepting new patients. Sometimes they’re listed in network but they’re on leave or their practices are full.
Chernew: That’s no doubt an issue. MA plans rely on networks to accomplish what they need to do, and on balance that’s probably a good thing but it makes choice complicated. It’s also true that an MA plan can exclude providers who might not be particularly good physicians.
The concern you raised, that when you have a bunch of competing health plans, some of them may be doing things that you would rather them not be doing, and that their beneficiaries may not be aware of, is a real concern. Is there an important role for some level of government oversight? And how do you balance that oversight with allowing the market to actually work? That will be a continuing objective of policymakers as Medicare Advantage grows, and a more important objective.
Clark: Let’s talk about misleading MA plan marketing, which CMS has tried to rein in, but there are still a lot of misleading claims.
Chernew: People are getting advice from brokers and sometimes it’s great, and sometimes not so great, sometimes given by agents who are rewarded by the plans. CMS is trying to get on top of that problem. The plans are just so complicated and so varied, there may be value in some type of standardizing benefits. It’s hard to make sure everybody gets the information they need. The broader issue you raised is that people don’t know they’re in a Medicare Advantage plan. My father wasn’t even sure.
Clark: Is he in an MA plan?
Chernew: I don’t know. I asked him and he didn’t know. He has a plan through his previous employer.
Clark: Whenever I am interviewing a senior doctor who is not employed, if the opportunity presents itself I ask if he or she has Medicare Advantage or fee-for-service. I have yet to find one in an MA plan, including many Kaiser doctors.
Chernew: I haven’t done a systematic study of that issue.
Clark: Let’s talk about delays and denials of care, prior authorization and the use of algorithms that automatically reject doctors’ requests for referrals. Sen. Richard Blumenthal (D-Conn.) said during a hearing last year that MA is “fine … so long as you don’t need it for the big stuff.”
Chernew: It is a concern and will be an ongoing tension. I remain concerned about excessive prior authorization, inadequate networks, and a whole range of things that remain major issues. There are concerns about specific plans that provide really substandard access and substandard care, and that are being overly aggressive in their claim denials. Patients are not getting the care they need and their experiences are atrocious. There are pockets of real problems. But that tends to not be what you find in a broad swath of evidence.
Clark: You may not have bodies or corpses to count. But what you do have, in my experience, listening to people who have been there, is anxiety as they waited for months for an appointment or approval. And they can’t manage the appeals process. Some people say star ratings for these plans should include a measure of their denial rates.
Chernew: Certainly understanding the rate of denials, and more importantly, what is being denied, would be valuable information. A colleague years ago was upset because his son – in a managed care plan — was in a serious accident and was discharged earlier than the father thought he should be and that he got very bad care. But he later learned that the issue was that the son was at risk of infection and getting out of the hospital to home was much safer.
Clark: Last question. How’s the solvency of the Medicare trust fund in light of the much higher spending going to MA plans?
Chernew: Medicare spending in general has been growing more slowly for a bunch of reasons. There have been some inexplicable reductions in utilization of certain types of care, and a shift to cheaper settings, hospitals to outpatient. But if your question is, can we improve the fiscal solvency of the Medicare program by cutting MA, the answer is absolutely, yes. Would we lose something? Yes. Will we lose a lot? If we cut modestly, I don’t think so. But that remains to be seen.