David Epstein Discovers Hansonian Medicine

He writes,

In 2012, Brown had coauthored a paper that examined every randomized clinical trial that compared stent implantation with more conservative forms of treatment, and he found that stents for stable patients prevent zero heart attacks and extend the lives of patients a grand total of not at all. In general, Brown says, “nobody that’s not having a heart attack needs a stent.” (Brown added that stents may improve chest pain in some patients, albeit fleetingly.) Nonetheless, hundreds of thousands of stable patients receive stents annually, and one in 50 will suffer a serious complication or die as a result of the implantation procedure.

Almost twenty years ago, Robin Hanson set out to explain some puzzling facts about health care. Most notably, “users are losers” (my phrase). That is, if you take similar populations with different levels of health care spending, outcomes tend to be the same. Note that the term “similar” precludes the explanation that the people undergoing more procedures were sicker to begin with.

Robin reasoned as follows:

1. We know that some medical procedures are effective and improve outcomes.

2. However, on average, similar populations that undergo more procedures do not have better outcomes.

3. Therefore, there must be some fairly common medical procedures that worsen outcomes (at least on average), and these tend to balance out the helpful procedures.

For many years, I have been referring to these procedures with adverse outcomes on average as Hansonian Medicine. My own book on health care policy took this issue quite seriously. I preferred to talk about procedures with high costs and low benefits, and Hansonian Medicine as strictly defined means procedures with negative benefits.

Why do we have Hansonian medicine? Epstein cites doctors who are not as well informed as they should be, the threat of lawsuits, and the opportunity to make money doing procedures.

Hanson himself had an intriguing theory, which is that you undergo unnecessary procedures because your friends and family want you to “do something” when you are not well. They show that they care by encouraging you to visit a doctor and to undergo procedures. I think that this is right, particularly in end-stage procedures, where it is often the relatives rather than the patient who demand futile care.

All this raises the question of what to do about procedures with high costs and low (or negative) benefits. The centralized solution is to come up with a way to tell doctors not to undertake these procedures. One challenge with that is there may be specific instances where a doctor knows that a procedure will work, even though in many other cases it does not. Another challenge is that friends and family were not be persuaded by a centralized agency (aka “death panel”).

The approach that I advocated was to reduce third-party payments and let the patients sort things out for themselves as best they can.

22 thoughts on “David Epstein Discovers Hansonian Medicine

  1. I agree that cost should be related to benefit but it takes centralized statistics and decision making to make those determinations. It could be made part of the approval process, but it wouldn’t be easy or popular and would involve common standards. Procedures where costs exceed benefits should not pursued beyond the development stage unless they can be better targeted. Generally they still have to be pursued enough to gather that information, and a lot of that cost is learning. The worse outcomes are likely finding out what doesn’t work. This is a large proportion of how other countries can spend so much less than us. Directing spending to where it is most valuable saves lots of money.

  2. I couldn’t agree more. Many problems in life have complex causes, but in my opinion, the overwhelming cause of cost escalation in healthcare is the fact that most care is funded with “other people’s money”. I worked in the industry for over a decade, first as a consultant and later as an investment banking. The amount of money sloshing around in this sector is amazing, and I’ve always thought that if one is ambitious and wants a reasonable probability of accruing over a $100 million in net worth, healthcare is the best place to be (better than tech, finance and real estate). Unfortunately, the special interests in this sector are so entrenched that it is difficult to legislate solutions that would reverse the cost trend. One change that I think would go along way is to remove the tax advantages of running one’s healthcare coverage through employer insurance. As you know, such expenditures are tax deductible, but if a person pays for healthcare themselves, the expense is usually not deductible. Health savings accounts are overly complex and have many limitations, so the HSA solution is a weak second-best solution. I know because my wife and I had an unusual healthcare issue, and we were unable to qualify for an HSA. One solution would be to make all healthcare expenses-party payor, or if the patient pays directly. For example, medical payments could be 100% deductible, or some other percentage so that the total effect on federal revenues would be tax neutral. My guess is that the tax neutral number would imply a blended percentage of about 75% (i.e., if you spend $100, $75 of the expense is deductible). HC expenditures routed through one’s insurance would have the exact same deductible percentage.

    I could tell you many stories from my banking days of how reimbursement for specific treatments translated directly into big profits for the companies, lobbying on the part of companies, how reimbursement drove volume for specific treatments, and similar distortions in the market…..but I will spare you the agony.

    • Sorry, I had a massive typo in my Reply. 2/3rds of the way into the first paragraph, I meant to say “One solution would be to make all healthcare expenses subject to the same tax treatment, regardless of whether the funds originated from a third-party payor, or if the patient pays directly.”

  3. Last year, an 86 y.o. man of my acquaintance was diagnosed with pancreatic cancer. He was subjected to drastic surgery to remove the tumor and then agonizing chemotherapy. Who knows how much all that treatment cost. All to no avail. He died Thanksgiving weekend.

    I assume that the man’s devoted son insisted on trying every possible means to extend his life. And was not paying for it.

    • Pancreatic cancer? I suspect he was a data point. A lot of it is haggling over who pays for the experiment.

      • I don’t think they were giving him any experimental or innovative treatments. The surgery he had has been around since the 30s.

        • One of my beliefs is that the distinction between experimental and standard treatments is pretty fuzzy.

          Also, nobody has been stopping the NIH from doing cost-effectiveness research.

          Now that they didn’t do it, the idea that we can start making decisions based off research that hasn’t been done is dubious to me.

  4. “The approach that I advocated was to reduce third-party payments and let the patients sort things out for themselves as best they can.”

    How will the patient know if the procedure will be beneficial? Anyway, you present a false dichotomy of sorts, before thoroughly looking at the problem. Why do private insurers pay for care they know doesn’t work? Aren’t they profit oriented? People may want lots of things, or their families may want stuff, but why do insurers go ahead and pay? Mind you there is no government rule or regulation forcing them to pay for stuff that does not work. We are doing tons of back surgeries and knee arthroscopies that have been shown to not change outcomes. No rule forcing anyone to do them. Figure that out and you can answer a lot of your questions.

    Query- How much would you reduce third party payments? Since insurers don’t want to do this now, would you force them to do so? If that is accomplished, you do then realize that only wealthier people get to actually make a decision about what they want to do? Is that a bug or a feature?

    Steve

    • I think you’re missing part of the story. If the patient was paying directly, then it would be up to the patient to decide how much and how intensive of a procedure to receive. This would feed back into the pricing of procedures themselves. If you look at sectors of medical care that are dominated by, or at least have a significant component of true private pay (e.g., dentistry, cosmetic surgery, fertility, etc.) you can see this type of self rationing play out. Specifically, what happens is that the purchasing actions of the private payors tend to drive the overall pricing equation, and the third-party payors follow. With the vast majority of procedures where third-party payors dominate (e.g., dermatology is a prime example because of the heavy Medicare patient component), the third-party payment dynamic tends to drive the pricing, and the other payors (including the tiny sliver of true private pay) follow along. Because there is poor self rationing with these groups, the pricing can go haywire

      • There is no way something like major surgery is going to be affordable to most people out of pocket. You need to have third-party payors for things like that, one way or another.

        • There are No True Scotsman’s arguments going on here. If they are “major surgeries” that people can’t afford, then everyone can’t have them.

          So, that is what catastrophic insurance is for, and health savings accounts for the rest.

          • Insurance, whether voluntary or governmentally coerced, is another method of bringing in a third-party payor. I don’t see what your disagreement is.

        • Of course. What you do is have the high cost, low probability of occurance procedures be covered by true “insurance” with the other stuff paid for out of pocket. What you have today is not insurance in the pure sense of the word

          • I agree, of course. And I actually would have no objection to something structured like Obamacare as long as it was limited to catastrophic care, rather than extending coverage to kids’ tummy aches and forcing the whole country to pay for gender reassignment medicine.

  5. You don’t need a separate class of procedure to make the argument. The same procedure can produce the patient outcome effect you’re mentioning because of complications inherent in any intervention. A procedure that, on average saves more lives than not in the appropriate context (e.g., stents in heart attacks), will cause more problems than it prevents in another conext (e.g., cardiac stents in patients with toe pain, to use a ridiculous example). All you need is indication creep before the data has caught up to the practice. The problem is that once a procedure catches on, it’s hard to stop it unless there’s solid data against it, or payers stop paying.

  6. If a lot of late-state medicine is Hansonian and futile, and there is no practical way to resist the political temptation of third-party payments from the state, then maybe the most enlightened approach really is the plausibly-deniable rationing of long waiting lists, with the alibi of a labor shortage of medical personnel.

    We are told that the NHS and Veterans Administration have these very long waiting lists for questionable interventions like stents. Maybe that not only saved money, but in terms of overall patient outcomes, was all for the best?

    Actually, didn’t we just have a VA ‘scandal’ regarding long wait times, and we are now told that those wait times have been substantially reduced.

    This seems to me like something of a ‘natural experiment’ some researcher should be studying regarding how much life-outcome-improvement bang we got for all those extra bucks. The Hansonian estimate might actually be negative.

  7. A good example of Hansonian medicine is bariatric (weight loss) surgery. Operations like sleeve gastrectomy or gastric bypass give good results to some patients but many patients experience considerable morbidity. Revision surgeries and weight regain are very common, all paid for by third parties. Many clinicians, even those outside surgery, are aware of the problems from patients they’ve seen but are reluctant to criticize their colleagues. A lot of the reason for Hansonian medicine is from physicians who want to make money and don’t have an interest in containing costs or calling out docs in other specialties.

  8. There’s probably a big problem with this idea, but how about Medicat insurance that would cover a) serious existing conditions and b) end of life care and c) chronic conditions. All of that would be covered by the reinsurer of last resort, the USG. Then the normal insurance companies could cover ordinary insurance. Medicat would be funded by eliminating the tax exemption on health care.

    • The same way they do for other types of insurance, such as liability, auto, life etc. It’s not that complicated

  9. Nice article — thanks for the link. And timely too — I have 90-year-old relative whose doctor is talking about a stent for her. But she’s from an era where questioning the doctor’s advice is just not done, so sadly, so if he recommends a stent, she’ll probably end up getting it.

Comments are closed.