Medicare Millionaires Emerge in Data on Doctor Payments | Hacker News Reader

Medicare Millionaires Emerge in Data on Doctor Payments

bloomberg.com

61 pointsadventured51 comments

Does any one have a link to the actual data being discussed? They are big numbers but it doesn't necessarily seem to be (except in a handful of cases) outrageous. The numbers are just annual billings, not annual take-home pay, right?

For example, I just had a surgical procedure last week to have an otologist repair my eardrum. The breakdown in cost is:

   Surgeon charge:   7,043
   Facility charge: 10,200
   Anesthesiology:   1,000

Now, the surgeon is very good at what he does and I feel he certainly deserves that $7000. His surgical schedule has him doing procedures twice a week, probably some more complicated than mine, and I'm sure he easily brings in several million dollars to his practice, some non-trivial amount of which comes from Medicare. Other specialties (like oncology and ophthalmology) are obviously going to make up a bigger share of the amount paid by Medicare because they are disproportionately treating older patients.

Until we know more, I think it's silly to be outraged about this.

Here you go: 1.7GB uncompressed tab-delimited text, as a 414MB zipped download

http://www.cms.gov/Research-Statistics-Data-and-Systems/Stat...

(link goes to the overview download page with context, not directly to the ZIP)

Here's a good overview of the limitations of the data:

http://www.npr.org/blogs/thetwo-way/2014/04/09/300857727/dos...

The numbers being discussed are the physician fees, not the facilities or supplies charges. This is the cash going directly to the docs to compensate them for their time.

They still need to pay malpractice insurance etc and more than one person is often billing to that same number so it's not really direct to doctor pay.

True! It's not a straight salary and docs do have expenses they need to pay out of that money.

I was just trying to call out that the cost of supplies and facility fees are paid by Medicare separately.

This does not include cost of injectable or infused drugs. These are not paid separately by Medicare.

Sure they are! If you're an in-patient then drugs are paid for via the DRG payment (all costs associated with treating the patient). If you're an out-patient, they are reimbursed through ASP.

The physician fee is separate from these payments.

I'm skeptical for these ophthalmologists to be billing those numbers and NOT include the average sales price (ASP) of their injectable drugs.

For those looking in, ASP is how medicare reimburses outpatient drugs. The physician practice buys the drugs up front, then bills Medicare the "average sales price" + 6%. The issue at hand is that Lucentis runs $2120 per injection. If you have a decent or large sized macular degeneration patient population, you'll be doing a lot of these injections. However, the number the practice collects is $320. Out of that number, you have to pay overhead / office staff / supplies etc.

I just downloaded one of the excel sheets and I only see procedures on the list, no drugs, but your suspicion is warranted.

The other thing to remember is that if it did include ASP reimbursement, ophthalmologists wouldn't be at the top, oncologists would. If a cancer doc saw 200 HER2 positive breast cancer patients in one year (16/month), treating them with Herceptin alone would cost $14,000,000. Most docs I've spoken to see 100-200 cancer patients per month (of course not all get treated with expensive drugs).

Yes, but you have to remember that these are Medicare data. The prevalence of 65+ patients getting Her2+ breast cancer AND getting treatment with Herceptin is probably lower than the number of 65+ patients getting ARMD and going on treatment with Lucentis.

Unfortunately, I don't have a good source. The above is merely my speculation.

And keep in mind that facilities fees often go to highly profitable outpatient facilities owned by the doctors.

That's an investment made by physicians who rightfully should share in the profit sharing of their business.

It's also unethical for a doctor to refer patients to a facility he or she owns. This is a huge source of Medicare and Medicaid fraud.

This is actually illegal. See http://en.wikipedia.org/wiki/Stark_Law

I've seen physicians disclose their relationship with said facility, which may or may not be used as a casual referral. I think the communication of the conflict of interest is the important issue here.

If that physician did just 3 of those procedures per day, every day, 5 days a week for a year: $7,043 x 3 x 5 x 52 = $5,493,540

Except that OP said he worked twice a week, so even if he did 3 a day (probably more like 2) and assume two weeks cavaction a year, then 7043250 = 704300.

Figure malpractice insurance at $50, and he's netting $650k. Figure 10% for billing and business expenses, accountant, etc, and he's netting $600k/year.

Is that still unreasonable for a highly specialized surgeon?

$7,043 x 2 x 2 x 50 = $1,408,600

But anyway I don't think any of the numbers are unreasonable. These are ballpark estimates, and I'm pretty sure my tympanoplasty was a relatively boring procedure for him, compared to stuff like cochlear implants and other basal skull surgery that he does sometimes. My main point was that I don't think it's unexpected at all for some surgeons to bill several millions to Medicare in a year. They can easily clear several millions already with a very light schedule and, depending on the specialty, some or most of that may be from Medicare patients. Honestly I think there is no story here.

I am curious about that one opthalmologist who made $21m in a single year from Medicare alone. That's top hedge fund manager or Fortune 500 CEO territory...

Is it possible that this is a businessman with many clinics collecting payments in his name, or is that really one individual who somehow billed this much?

While that is a very large number, its the gross Medicare payments to the ophthalmologist's practice via doctor/surgeon charges, not his salary. He personally only gets a fraction of that. Granted its a multi-million dollar fraction, but still a fraction.

The way Medicare for these high-reimbursement specialties (Ophthalmology, Radiation Oncology, and Medical Oncology) is that the Doctor's practice will pre-purchase the medications/injections/treatments before administering to patients and then Medicare will reimburse the practice. These doctors are certainly not taking home anywhere near a fraction of that number.

See: http://oig.hhs.gov/oei/reports/oei-03-10-00360.pdf

The executive summary discusses how two injections used for treatment of Age-Related Macular Degeneration are purchased.

That's called "buy and bill" and that is paid separately from the physician fees discussed in the article.

If you did include the cost of drugs, you'd be correct that physician salary is only a small fraction of the cost of treating a patient. Since some cancer drugs cost over $10,000/month, some clinics spend 90%+ of their revenue just on drugs.

I'm pretty sure they are including buy-and-bill numbers here. The #1 doctor, Salomon Melgen, is a retinologist who is currently being investigated by CMS for suspected Medicare fraud. The numbers in this article match the numbers from the CMS data dump.

See http://www.breitbart.com/Big-Government/2013/02/14/Medical-E...

Turns out you were correct! I wouldn't have imagined that Lucentis would have such a impact on the numbers.

My guess, and this is only a guess, is that the physician is focusing on routine procedures (cataract surgery) and has optimized the procedure greatly.

Someone else preps the patient, doc walks in, spends 15 minutes doing the surgery and then moves to the next patient. Nurses or other medical professionals care for the patient after that.

Unfortunately I can't find it anymore, but there was an article a while back about physicians performing an amazing number of surgeries and if you calculated it out (based on the time estimate by Medicare) they were working 200 hrs per week. In the end, the physician had just chosen to specialize in one procedure and optimized everything about his clinic around that.

The FBI is asking the same questions of this doctor:

http://www.nytimes.com/2014/04/10/business/doctor-with-big-m...

That's kind of answered by some of the other doctors in the article:

"Michael McGinnis, a New Jersey pathologist who was the third highest paid in the Medicare data, said that payments made to multiple doctors may be recorded under just one in the Medicare data released, making that person look like they were receiving a suspect amount of money. Provider Codes

"McGinnis’s provider code was used for about 27 doctors at Plus Diagnostics in Union, New Jersey, where he is the medical director, he said in a telephone interview today.

"'I don’t really work directly at the facility, I’m doing administrative work,' McGinnis said, 'I’m not offended by it, but it’ll need to be interpreted correctly so the wrong ideas and wrong statements won’t be made.'

"Franklin Cockerill, listed as the fourth-highest paid physician listed within the Medicare data, is in a similar situation, said Bryan Anderson, a spokesman for the Mayo Clinic where Cockerill is employed as chair of the Departmenet of Laboratory Medicine and Pathology.

"Cockerill’s 'name is listed as the billing physician for claims submitted for payment under the clinical lab fee schedule,' Anderson said. While he clinic’s labs performed more than 23 million lab tests in 2013, Cockerill is a salaried physician who doesn’t receive Medicare payments."

Course, that just raises the question of why Medicare allows for A to bill medicare for a procedure under B's name. I'd think there'd be a way to have separate code for doctors in the same practice.

This isn't the first article I've seen on this and they all seem to imply that something unscrupulous is going on because some doctors make so much from medicare. In some cases there probably is but is data on total payments received alone enough to tell people anything meaningful?

I would think you would also need to know what their percentage of elderly patients were vs. private insurance patients. You would also need some measure of their operating costs. Absent that kind of information this seems only useful as a political tool to plant the idea in people's minds that hey these doctors make a lot of money and some of it might be fraudulent.

does the AMA realize that if we follow their objections a to a logical conclusion that the American Public and Gov will ask to release more data not less?

Some context, Medicare rates and fess are not adjusted on a timely basis to adjust for changes in medical procedures and technology ..procedures become less time consuming to complete and things become less costly to have completed..

But that is only one context of many that has to be exposed and we need more data to do that

This is why the overhaul of Medicare data via the billing system is so vital to improving the benefit the government gets per dollar it spends on Medicare.

A very complex CS and Engineering problem that no statup has adequately tackled yet

Medicare and medicaid have traditionally underpaid regular commercial insurance with the result that more and more doctors have dropped it over the years.

So it makes sense that payments would concentrate around certain doctors that are still accepting it because more people on medicare will go to them simply because they don't have any other choice, especially in areas that have high concentrations of the elderly, like Florida.

Great point. It should be stipulated (perhaps speculatively) that doctors have left Medicare to go with private insurance to make more. Thus some of the ones who still accept Medicare do so since it is still worth their time and effort (i.e. still making the big bucks).

I'm not a data scientist but I'm guessing this data is somewhat normally distributed with a long tail on the right side. I would also venture to guess that if we looked at individual patient spending we'd see some patients spending millions per year. (Though w/ HIPPA maybe anonymized data isn't even available.) Both patients/doctors that are outliers should draw additional attention as the chances of fraud are increased, but the distribution is probably somewhat normal.

This is a really cool tool:

http://projects.wsj.com/medicarebilling/

Does anyone here know much about Ophthalmology?

What is a "Ranibizumab injection"? It's making bank!

The top guy does 37,075 of these are year, at $320 a pop.

That's about 100 a day....

Ranibizumab is the generic name of Lucentis, an injection into the eye that is used to treat age-related-macular-degeneration. The way Medicare works is that Lucentis is pre-purchased by the ophthalmologist and then reimbursed by Medicare. The doctor is taking home only a fraction of that reimbursement, as the majority of that reimbursement is going directly to the manufacturer.

I posted this elsewhere, but the way Medicare for these high-reimbursement specialties (Ophthalmology, Radiation Oncology, and Medical Oncology) is that the Doctor's practice will pre-purchase the medications/injections/treatments before administering to patients and then Medicare will reimburse the practice. These doctors are certainly not taking home anywhere near a fraction of that number.

See: http://oig.hhs.gov/oei/reports/oei-03-10-00360.pdf

Ok. That makes sense. I'll put away my pitchfork. But a follow up question. Why are some offices doing so many of these? 100/day. That's like a Ranibizumab assembly line.

One fairly likely possibility is that he is billing for other doctors:

http://www.melgeneyecenter.com/index.html

He's actually being investigated for fraud. So if we take him out of the picture, the remaining numbers make sense.

The article states that many of the offices getting the largest reimbursements have all of their billings going through one doc. So a large-ish ophthalmology clinic with fifteen eye docs looks like one doc according to the data.

The standard treatment protocol is monthly injections on an indefinite schedule; given the prevalence of age-related macular degeneration, it's easy to see how a clinic in a high-retirement area could profitably optimize to do hundreds of these. Lasik clinics operate on a similar principle. (I'll spare you my argument that medicine is basically the last outpost of the Marxian labor theory of value, but suffice it to say that when capital investments in an industry don't improve profit, then Taylorist labor optimizations will step in to fill the void.)

riahi might be able to answer this -- any idea what the cost is of ranibizumab these days? A single injection used to run about $2K, but I don't know what the price is now.

Yes. If you download the April ASP data from CMS [1], you can search for "Ranibizumab injection". CMS has calculated a $397.014 payment per 0.1 mg. Each dose injection is 0.5 mg, so 5 * 397.014 = $1985.07. If you add 6% to that, you get $2104.17.

Your memory was correct.

1: http://www.cms.gov/Medicare/Medicare-Fee-for-Service-Part-B-...

http://en.wikipedia.org/wiki/Bevacizumab#Uses_in_eye_disease

Probably why there is a lot of controversy surrounding this and the price of this drug.

We're still only talking about 12% of total Medicare spend - so there's no doubt some fat in there, but you're not going to significantly affect the overall system working on a 12% slice.

Do you never change your salary unless it's by more than 12%?

Are you ignoring the simple math that you can't change the Medicare picture on the back of a 12% allocation?

Changing pictures isn't math, dude.

"Changing picture" = "Make a significant difference in the current financial health or structure of Medicare". In other words, let's say you can change that number to lower it by 25%. Now you've affected 3% of Medicare's cost - hardly the main problem.

How is it that you're incapable of recognizing that 3% is larger than 0%?

I wrote a neat visualization of medicare spending last year. It doesn't go down to the physician level, but I think it is a cool way of viewing the data:

http://stonefinch.com/Projects/Medicare

It'll take a while to load, there is a 19MB file that gets loaded into your browser. Chrome suggested.

I have a question. Seems like the practice of billing for a single physician is problematic in understanding potential fraud/inefficiencies. Is there a reason why it can't be mandated that the physician responsible for care not be the one who is on record for payment?

This is one problem with the current healthcare system. Doctors get paid based on the amount of stuff they do to you, and not based on whether you get better or not.

The problem with sysops is they get paid for the number of CVEs they respond to, and not based on whether CVEs stop being issued.

Reply on news.ycombinator.com