Medicare Payments Data Dump Reveals Massive Scope of Federal Health Care Billing System


Medicare is a $500 billion program that's been identified for years as high-risk by the Government Accountability Office for its susceptibility to improper payments—everything from mistaken billing to overpayments to dubious upcoding to outright large-scale fraud. And yet since the program's inception, the public, including most medical policy researchers, has never had anything like complete access to its payment data.

That changed yesterday, when Medicare released a trove of data on its 2012 payments to physicians for anyone, or at least anyone with enough computing horsepower, to see. The single-year data set is anonymized to protect patient information, and certain rare procedures—those performed on fewer than 11 patients—don't show up. It's not for everyone. The set is big enough that it's not easy to work with on a typical desktop computer; but for researchers, policymakers, and journalists who can arrange for database storage and access, it's a pretty big deal. 

It's going to take a while to fully process all this information, but a couple things stand out already from the stories that have been written so far. One is the sheer scale of the payments involved. The data set doesn't cover anywhere close to the entire Medicare program, but it offers a look at $77 billion worth of payments to 880,000 medical professionals in the year 2012. From that group, The Washington Post notes, about 4,000 physicians billed the program more than $1 million. And a handful billed in excess of $10 million.

It won't surprise many people that the highest billers are concentrated in the sunny state of Florida. The state has a heavy concentration of seniors. It's also a haven for Medicare fraud. And the data suggests a possible correlation between unusually high billing and payment funny business. According to The Post, "three of the top 10 earners already had drawn scrutiny from the federal government, and one of them is awaiting trial on federal fraud charges."

But bad billing, including fraud, isn't just a matter of a few folks at the top. The program is rife with bad payment. In fiscal year 2012, the fee for service portion of Medicare made $29.6 billion in improper payments—meaning that about 8.5 percent of all payments made were problematic somehow. And that was a pretty good year, relatively speaking, the product of a decline from 10.8 percent in 2009. The improper payment rate shot back up to 10.1 percent in 2013.

The point is Medicare blows tens of billions of dollars on bad payments every year, some of which is a result of administrative ineptitude, and some of which is just people scamming the system. But until now, most people have never really had an idea of what the overall picture of that payment system looks like. This isn't all the information we might want—it would be great, for example, to have more years available so we could see how payment trends change over time. But it's a good start. 

NEXT: Florida Trying to Make Microbreweries Sell, Buy Their Own Beer Back from Distributors

Editor's Note: We invite comments and request that they be civil and on-topic. We do not moderate or assume any responsibility for comments, which are owned by the readers who post them. Comments do not represent the views of or Reason Foundation. We reserve the right to delete any comment for any reason at any time. Report abuses.

  1. And the much maligned ACA has several cost control measures for Medicare.

    1- Readmission penalties (already saving $15 billion/yr)
    2- MCO’s
    3- DEATH PANELS! (useless procedure panel)

    (still not a net winner though)

    1. If by “MCO”, you mean ACO, that program is failing. Near-final data on the first full “fiscal” year of the program shows that around 75% of ACOs failed to reach their savings targets and won’t get any incentive payments.

      Readmission penalties are like trying to drive a finishing nail with a sledgehammer mounted on spaghetti.

      Both programs share a common trait: they suffer from profound design flaws, especially when applied to the vast array of providers in the US.

      1. Yes, I meant ACO.

        Was thinking Managed Care.

        1. Was thinking

          Nobody believes that.

        2. Managed Care is what they called ACO’s when they failed the first time. But the rebranding will definitely make them work.

      2. I heard that many of the CMO ACO providers couldn’t have met the reporting requirements, much less used data to affect care in a meaningful sense. Is that FUD by my industry (tech consulting)?

        1. Dunno about the reporting requirements. That hasn’t been an issue for ours, really.

          The data? Oh, hells yeah. CMS is months and months behind on turning data back to ACOs. You can’t manage patient care with 6 month old data.

          And don’t get me started on the colossal clusterfuck of the patient attribution methodology, or the fact that their baseline cost methodology was thoroughly botched.

  2. Protect our seniors!


  3. A provider submits a claim to Medicare, and Medicare pays. Only later do they catch up and see if that service was appropriate and actually provided. It’s like they have a sign on their building saying, “Defraud us–It’s easy!”

    1. In most cases, I don’t think they ever show up to see if the service was appropriate or actually provided.

  4. A prog recently used Medicare as Exhibit A in his argument in favor of big government. Then again, that was the same guy who told me that as a Libertarian, I presumably favored drug prohibition.

    1. as a Libertarian, I presumably favored drug prohibition.

      Well, yeah. Libertarians are like soCons only more to the right. Everyone knows that.

    2. as a Libertarian, I presumably favored drug prohibition.

      Seriously? Let me guess, this prog favors some kind of drug legalization and just reflexively assumes anyone they identify as ideological opponents must hold the opposing view?

  5. You can’t take those 7 and 8 figure payments to doctors at face value. I haven’t dug through the database, but those could be to what are, in essence, group practices. Hard to know.

    There’s fraud in Medicare, sure. But a lot of Medicare “fraud” is either a good faith attempt to comply with a fabulously complex, contradictory and ambiguous paymetn system and/or a disagreement on the judgement calls in medicine.

    1. I’ve told this before, but our ambulance company had Medicare claims for transporting stroke victims rejected. The administrator said we had to indicate what kind of stroke. However, EMTS and paramedics can’t determine that–that’s up to doctors at a hospital with proper equipment. It was ridculous.

    2. R C Dean is correct. “Fraud” is such a loosely defined term in the Medicare world. Anyone who provides a lot of Medicare services will unintentionally commit fraud. The system is far too byzantine to achieve any other result.

      1. Fraud is a loosely defined term whenever the government uses it. To the government, it seems to mean any time you ask us for a payment that we at any time now or in the future decide we don’t want to pay.

    3. RC has it right. The problem here is that there is no way for the government to run what amounts to a giant insurance company for every person over 65 in the US. A private sector company that did that would be hugely inefficient. A government agency trying to do that is just mind boggling.

    4. Also, a lot of Medicare “fraud” (and even more so Medicaid fraud) comes under the category of “I can’t make a profit with your price controls if I only billed the services that I did, so I’m going to figure out a way to charge the correct amount.”

      Now third party payment makes some of that fraud, but the large number of doctors who decline to participate (especially in Medicaid) means that there’s something there.

      Cf. this story of a ACA supporting woman who found out that no nearby doctors accept her new Medicaid coverage.

    5. but those could be to what are, in essence, group practices.

      Bingo. There are a lot of Medicare/Medicaid billings that go through an Identified Provider for other practitioners. A doc who supervises ten mid-levels (Nurse Practitioner/Physician Assistants) might have all their billings ascribed to his or her name. Even where I work, in an ER, my name as the Medical Director is the default billing physician for any chart that is missing the physician billing code. So maybe I’m even on that list. I don’t know and I don’t care.

  6. WSJ: Medicare Paid One Doctor More Than $20 Million in 2012
    Florida Ophthalmologist Salomon Melgen Has Been Fighting Government for Years Over Overbilling Allegations…..reno64-wsj

    1. Every state has a Medicare fraud unit. I interned at one in their records section in the 1980s. The common thread seemed to be clinics and independent pharmacies set up in sketchy urban areas. And lots of them were only caught because a rival doctor or pharmacist dropped a dime. The customers didn’t care.

      1. When I worked at FLDOH, we’d get an email every week with healthcare professionals got their license pulled. Nurses get popped for drug habits, docs get popped for billing fraud, as a rule.

        1. The problem is Medicare is program that can be easily gamed by doctors. What I’d like to know is how does Medicare fraud compare to private insurance fraud?

          1. Private insurance has a smaller level of fraud, due to multiple factors, including spending more on fraud prevention and generally higher doctor reimbursement rates.

            The Medical Loss Ratio definition in the ACA considers fraud prevention spending to not count as money spent on treatment, meaning that it improves your MLR to allow fraud (since money spent on uncaught fraud *does* end up counting as money spent on treatment, since you don’t know its fraud.)

Please to post comments

Comments are closed.