One In Five Doctors Say: “No New Medicare Patients”

Happy Older Couple in Beach ChairsIf you learned that 93 percent of non-pediatric primary care physicians took Medicare patients and 94 percent took patients with private insurance, you would likely conclude that Medicare is doing just fine.

Unfortunately, such data do not describe physicians’ behavior at the margin, which is what will determine future access to Medicare. The Kaiser Family Foundation/Commonwealth Fund 2015 National Survey of Primary Care Providers also asks which physicians are not accepting new patients: 21 percent are not taking new Medicare patients and 14 percent are not taking new privately insured patients. That is, the proportion not taking new Medicare patients is 1.5 times greater than the proportion not taking new privately insured patients.

I would encourage the Kaiser Family Foundation and Commonwealth Fund to add a little more detail to the survey, by breaking Medicare down into traditional Fee-For-Service (in which physicians are paid according to a fee schedule dictated by government) or Medicare Advantage (in which they are paid fees negotiated with private insurers), and by breaking “privately insured” into categories (especially patients on Obamacare exchanges).

It would also be interesting to learn whether the physicians who are saying “enough” to new Medicare patients are more likely to be in new payment models (Accountable Care Organizations) which frustrate them. (Another survey reports a wide dispersion of opinions about such reforms among physicians.)

Nevertheless, even this level of detail indicates baby boomers aging in to Medicare will have increasingly difficult access to care. The idea of premium support (whereby Medicare beneficiaries receive subsidies to choose lightly regulated private plans), which NCPA advocates, should become more politically acceptable – if not unavoidable.

Comments (32)

Trackback URL | Comments RSS Feed

  1. John Fembup says:

    Overall percentages aside, it’s worth noting that virtually all individuals who age-into Medicare because they turn 65, are new Medicare patients.

    These 65-year-olds are way disproportionately affected by physicians’ reluctance to accept new Medicare patients.

    How many you might ask? Among others, Pew Research has the answer. Since 2011, baby boomers have been aging in to Medicare at the rate of about 10,000 a day, or between 3.5 and 4 million per year. That’s the equivalent of about 8% of today’s entire Medicare population. The rate of aging-in is projected to continue at or above this level for the next 15 years.

  2. Michael Gorback says:

    I wouldn’t be surprised to learn that it was mostly primary care that wasn’t taking new Medicare. That trend has been going on for years. The time required for older patients with more complicated medical problems vs the reimbursement just doesn’t work.

    Some specialties can’t refuse Medicare because of patient demographics. For example, few adult cardiology practices could survive without Medicare.

    Some specialties have to accept Medicare because they’re hospital-based: ER, Radiology, Anesthesia, Pathology, Hospitalists, etc.

    Right now Medicare is my preferred payer. A lot of the big five insurers are paying essentially Medicare rates (UHC was paying me less than Medicare rates before I dropped the contract). These plans also carry the administrative overhead of an HMO or PPO (e.g., pre-authorization), continually change their rules and policies, and are generally unpleasant to work with. If you understand Medicare’s rules and color inside the lines it’s the most hassle-free.

  3. Michael Gorback says:

    No new HMO patients either.

    East Texans at odds with health insurance and limited providers in network

    Confusion and frustration are the emotions felt by thousands of East Texans signing up for health insurance during this year’s open enrollment.

    A huge number of independent East Texas physicians have opted out of accepting certain health insurance plans offered by the marketplace. From patients to providers, the change has caused headaches for all parties involved.

    Full story at

    Short version: No PPOs available on the Exchange and the docs don’t accept HMO plans.

    • Ron Greiner says:

      The best way to describe Obamacare’s 3rd Open Enrollment popularity plunge is – nose-dive!

      • Yes, but when you go to get your nose repaired, with the specialist be covered by the plan?

        • Ron Greiner says:

          Also John is it a covered expense. It would be covered if it were an accident but this looks like this nose-dive is self-inflicted to me which of course would be a non-covered expense even if the medical provider was in network.

          If Hillary argues it’s not self-inflicted but merely incompetent leadership I think we have a covered expense again.

        • Michael Gorback says:

          Closed reduction of a nasal fracture is about $2,000 and open reduction maybe $4,000 so depending on your deductibles and co-pays it might not matter who’s in-network.

          I understand the point you’re making but the way things are going we’ll all just have catastrophic coverage anyway. I’m calling a spade a spade and getting a Bronze HMO with an HSA for 2016.

          • Ron Greiner says:

            That’s smart Michael. 95% of my HSA clients have a $100 deductible on accidents because it’s cheap and permitted. The consumer likes it because they can’t comprehend getting sick but they can comprehend an accident.

            Of course the number 1 way to lose a child is accidents.

            Just ask when you enroll, “How much is accident coverage?” Or don’t, but you would at least know the cost. Like I said it is usually cheap if it makes a difference.

  4. Barry Carol says:

    According to an article in Health Affairs some months back, most Medicare Advantage plans pay providers between 100% and 103% of Medicare rates while a few try to pay a bit below Medicare rates. Hospitals are unable to cost shift to private Medicare Advantage plans like they can do to the commercially insured population below the age of 65. The reason is pretty obvious. Either they accept Medicare rates from standard FFS Medicare or they accept Medicare rates or something very close to them from Medicare Advantage plans. Unless the MA members are significantly more expensive to serve as Michael Gorback suggests was the case with United Healthcare, providers, especially hospitals, are likely to be indifferent between the two for the most part.

  5. Michael Gorback says:

    “Unless the MA members are significantly more expensive to serve as Michael Gorback suggests was the case with United Healthcare,”

    I didn’t say that. You’ve conflated two separate observations (but no harm done).

    Does anyone know if MA patients run higher costs? It seems to me that if you anticipate high medical expenses an MA plan would be better since most plans cover Part D and all of them by law have out of pocket caps. Even though you pay a higher premium and sacrifice flexibility and availability of providers, it seems to me it would be attractive to the high-cost patients.

    • Barry Carol says:

      Michael –

      By more expensive to serve, I was referring to your earlier comment about the need to seek pre-authorization in many cases as opposed to the comparatively hassle free process under standard FFS Medicare as long as you color inside the lines. As long as you are paid fee for service as opposed to capitation or shared risk / shared savings contract, you shouldn’t care how much healthcare a given individual uses at least from a financial standpoint.

      Medicare Advantage is especially popular among lower income seniors because it eliminates the need for a supplemental plan which can be quite expensive and, as you noted, offers an out-of-pocket maximum amount by law. Insurers tell me, though, that the average risk score for Medicare Advantage members is between 0.80 and 0.85 with 1.0 defined as average risk. This suggests that the MA population at large incurs below average healthcare costs.

      Some less scrupulous MA plans go to considerable lengths to avoid sick people with strategies ranging from holding information meetings on the second or third floor of non-handicapped accessible buildings to deliberately not contracting with the best heart and cancer centers. While CMS has a pretty sophisticated system to calculate individual risk scores that are used to determine MA premiums, at the end of the day, their system still overpays for the healthy and underpays for the sick. So, there is still an economic incentive for the insurers to attract as many healthy people as possible and do everything they can to minimize the number of sick members they wind up with.

      • Michael Gorback says:

        What they tell you isn’t what they tell Uncle Sam. There have been several instances, including a whistleblower suit, where MA plans have upcoded severity scores or reported diagnoses that the patients didn’t have. I think the scandal runs the gamut of carriers, including UHC, Humana, BCBS, Wellpoint, etc. NBER put the amount of fraudulent overbilling in the billions. I tried to find the paper but their site is down at the moment.

        • dennisbyron says:

          barry carol

          you might want to check your source for the “2- to 3-story non-handicap-accessible building” claim. Nothing to do with Medicare, where is there a 2- to 3-story non-handicap-accessible public building in the US 25 years after passage of the ADA?

          Michael Gorbach

          You don’t need NBER for alleged upcoding statistics. See this GAO report (see I think it turns out that GAO found that IF there is such upcoding, it amounted to under about 1% of public Part C Medicare health plan spending. (IF is upper case because the GAO also said that it could be that FFS Medicare claims are undercoded because there is no incentive to correctly code FFS Medicare claims.)

          • Michael Gorback says:

            Upcoding: Evidence from Medicare on Squishy Risk Adjustment
            Michael Geruso and Timothy Layton
            NBER Working Paper No. 21222
            May 2015, Revised October 2015
            JEL No. H42,H51,I1,I13,I18

            Upcoding—manipulation of patient diagnoses in order to game payment systems—has gained significant attention following the introduction of risk adjustment into US insurance markets. We provide new
            evidence that enrollees in private Medicare plans generate 6% to 16% higher diagnosis-based risk scores than they would generate under fee-for-service Medicare, where diagnoses do not affect payments.

            Our estimates imply upcoding generates billions of dollars in excess public spending annually and significant consumer choice distortions. We show that coding intensity increases with vertical integration, reflecting a principal-agent problem faced by insurers, who desire more intense coding from the physicians
            with whom they contract.

            • dennisbyron says:

              The abstract you reprint says “billions” but note that it was revised in October after being released in May. The authors made a huge error in May 2015, not factoring in that a law had been passed 7 years earlier in 2008 to correct for the problem that showed up effectively only in one year, 2007. In the October 2015 version the authors — who were called out in the healthcare press for their unbelievable mistake ( — added footnotes citing the law change and citing the 2013 GAO report to which I referred you.

              Both sources — GAO and these authors — now agree that IF upcoding is an issue it relates to only about a percent of public Part C health plan spending.

              (By the way, I am not defending upcoding of any type. I am just referring you to the GAO report, the source of the primary data, rather than an embarrassingly flawed interpretation of that data. In my opinion, get this coding thing out of both Medicare and Obamacare. Give everyone on Medicare — except low income — the same premium support.)

              • Barry Carol says:

                “Give everyone on Medicare — except low income — the same premium support”

                While that idea would be easy to explain and administer, the fact is that per capita Medicare spending varies by as much as almost three times around the country. This is due to a combination of variation in medical input costs, differences in physician practice patterns, the health of the population and fraud / upcoding. ACA exchange premiums also vary a lot for similar coverage for the same reasons though fraud is probably not nearly as big a factor as it is for FFS Medicare and unmanaged Medicaid.

                In many rural areas where other living costs, especially housing and local property taxes, are low, healthcare costs are high due, in part, to the lack of competition and the need to pay more to attract doctors to rural areas that most doctors and their spouses don’t particularly want to live in.

                I think any premium support program should include what I call a circuit breaker provision that caps an individual or family’s contribution toward the premium at 10% of income for those with income above 300% of the FPL and a sliding scale percentage below that down to zero at 150% of FPL income or less. Income verification would need to be strict and robust and penalties for cheating should be stiff.

              • Michael Gorback says:

                I note that the “corrected” report still uses “billions”.

                The Center for Public Integrity reported on aggregate CMS audits of five insurers and found that roughly 50% of bills for 1,005 patients were too high, that risk scores were inflated in 80%, and that for 20% of patients overpayments were at least $5,000 too high. Extraopolate that.

                There continues to be a steady stream of whistleblower suits alleging upcoding abuse.

                It is also well-known that risk scores soar after enrollment in MA plans.

                • dennisbyron says:

                  Clearly you prefer to depend on propaganda rather than research. OK

                  • Michael Gorback says:

                    Thanks for starting my day off with a laugh.

                  • Michael Gorback says:

                    BTW, the paper you cite lists MA overpayments as follows:

                    “According to GAO’s estimates, the amount of the excess payments to MA plans after accounting for CMS’s adjustments was $0.6 billion in 2010, between $1.1 billion and $1.6 billion in 2011, and between $1.5 billion and $2.9 billion in

                    Billions with a “B”, and increasing.

                    They also said,

                    “Our work confirms that differences in diagnostic coding caused risk scores for MA beneficiaries to be higher than those for comparable beneficiaries in Medicare FFS in 2010, 2011, and 2012.

                    Furthermore, GAO did NOT actually audit charts. They compared MA to FFS. I can assure you that FFS charges are inflated too. I can only conclude from this report that MA is even more dishonest than FFS. Quelle surprise!

                    The “propaganda” cited by CPI used CMS chart audit data obtained by FOIA actions. Over 1,000 actual chart reviews.

          • Michael Gorback says:

            “it could be that FFS Medicare claims are undercoded because there is no incentive to correctly code FFS Medicare claims”

            What? How about the incentive to collect more money?

            • dennisbyron says:

              I don’t believe that’s how it works for FFS Medicare. If I read the GAO report correctly, the fee is the same no matter what the health of the FFS Medicare patient. But again, I am just repeating what it says in the definitive 2013 GAO report on this subject.

              • Michael Gorback says:

                No, the fee is not always the same.

                In anesthesia they get paid more for sicker patients. There are 5 ASA classes and IIRC you can bill extra for classes 3,4, and 5.

                For E&M services you have to document the level of complexity of decision-making, which usually entails listing all the concurrent diseases that one would have to take into consideration.

                For instance, if I want to do an epidural steroid injection on a healthy 20 year old the complexity is less than for a 70 year old with glaucoma, heart failure, and diabetes. All of those conditions can be exacerbated by steroid administration and I can justify a higher code.

                Speaking of anesthesia, the fees have gotten so low that many anesthesia groups don’t accept ANY insurance. That’s a very unpleasant surprise to anyone who has surgery since the anesthesia bill from an out of network group will be astronomical. I have seen EOBs where the anesthesia bill was 2-3 times my own. One of my patients showed me an EOB where the ENT got $500 for a tonsillectomy and the anesthesia group got over $1,000.

                This is flying under the radar for now but one of these days . . .

                I can’t say that I blame them. Medicare now pays $17 per unit. A low risk procedure that takes an hour would generate 3 units to start the anesthetic and then 1 unit per 15 minutes. Roughly $100/hour. That’s not sustainable.

    • Barry Carol says:

      For someone like myself with a long history of heart disease, I like standard FFS Medicare better because it ensures maximum provider choice. At least in the NYC area, lots of the better providers don’t accept many of the MA plans but they all seem to accept standard FFS Medicare. At least that has been my experience so far and I’m finishing my fourth year in the program.

      The supplemental and stand-alone Part D plans aren’t cheap but luckily for me, I can afford them. I’m also subject to the IRMAA surcharge which I would still have to pay even if I chose a cheap MA plan. I wind up joking that it costs my wife and me a lot of money to make healthcare feel free at the point of service.

      If regular Medicare offered a combined Part A and Part B OOP maximum of $10,000 or less, I would probably drop the supplemental plan which will cost the two of us $4,865 for 2016. That’s for Plan F, which is the most comprehensive and expensive.

  6. Devon Herrick says:

    I looked at Medicaid acceptance. The categories are somewhat confusing. But it appears one-third of primary care physicians do not accept Medicaid. More than one-in-two are not accepting new Medicaid patients.

  7. Wanda J. Jones says:

    John and Fellow Experts:

    Have you noted that not only is all this hard to follow for you who aleady know a lot, but there is at least informatikon about providers. What the public does not get is performance information about health plans:

    –% of denials of new coverage
    –% denials of a particular procedure
    –% of Physicians accepting their coverage
    –% of annual expenditures that were by the enrollee, not the plan.

    And so on. Can you all promote this?

    Wanda J. Jones
    San Francisco

  8. Perry says:

    An recent JAMA article about who is taking what in terms of new patients: