Dialysis, Mergers, MASSIVE FRAUD, and Competition Policy

It's a bit hard to categorize this one, folks

Until 2011, when Medicare overhauled its billing system, the business model of the major chains offering dialysis was literally just fraud. Whenever they took over a new clinic by acquisition, costs sky-rocketed while mortality increased. Much of the increase in costs was due to changes in medication prescription practices which were in some cases arguably wasteful, and in other cases unambiguously fraudulent.

Some preludial facts, drawing upon Eliason, Heebsh, McDevitt, and Roberts (2020). Dialysis is what you do when someone’s kidneys are failing. Essentially, you go into the clinic, get some tubes stuck into you, and have a machine do what your kidneys would have done. Generally this is through cycling blood through a filter, although 10% of patients use fluids pumped into the abdomen to draw out the toxins. In particular, we need to clear urea, which is a toxic by-product of normal biological processes. Most people have to go in three times a week for four hours. This is generally agreed to be an absolute pain-in-the-ass, and in any case, the mortality rate is not great. A kidney transplant is preferred, but there is an enormous shortage of donors. I strongly encourage you to do the right thing, and register here to donate. (I tried, but was rejected – apparently, I have pre-diabetes, and may well need a donor myself in the future!)

We spend an enormous amount of money on this. 1% of the federal budget is spent on kidney dialysis, and this is in spite of Medicare actually being extraordinarily stingy. Everyone who is diagnosed with end stage renal disease becomes automatically a part of Medicare, regardless of age. People on private insurance retain their coverage for 30 months after diagnosis, and it is they who account for basically all the profits of the dialysis companies – they pay twice as much.

The simple expenses of needing to pay for the machine and someone to run it are not all there is to dialysis. Many – most – patients are also anemic, for which they are treated with Epogen (EPO), a drug that stimulates the production of erythropoietin, and some sort of intravenous iron substitute, which comes with two brand names, Venofer and Ferrlecit, which are essentially identical. We spent simply gobs of money on Epogen – at $430 per patient, Medicare spent $2 billion a year on it in the 2000s, which was the most of any drug.

There are two main chains in dialysis care, Davita and Fresenius. During the 2000s they expanded greatly, and control 60% of the market for dialysis. When they took over clinics, a few big changes happened to prescribing practices. First, prescription rates of EPO doubled. Second, the prescription rate of Venofer greatly increased versus Ferrlecit. They also reduced the quality of medical staff, and made patients less likely to enter a transplant waitlist (on the order of a 10% decrease).

The increase in Epogen worsened health. The same team of authors have another paper exploring the mortality effects of the 2011 change which undid the excessive prescribing of the drug. Using the fact that people at higher altitudes produce many more red blood cells as an instrumental variable, they find that cutting EPO prescribing actually improved mortality.

The change from Venofer to Ferrlecit was not explained by the difference in payment rates, which were similar per milliliter, but by undiluted fraud. You see, Venofer came in 100mg vials, while Ferrlecit came in 62.5mg vials. Each time you opened up a vial, you had to discard whatever you didn’t use on a patient, but you could charge for the whole content of the vials. DaVita paid $450 million to settle a whistleblower lawsuit about the practice – judging from the complaint, DaVita was doing approximately everything fraudulent under the sun, including deliberately creating dosing charts to maximize wastage.