The Trouble With Medicine

Robin Hanson has a fondness for bold, outlandish claims so I think his argument that half of all medical spending is waste seems overstated. Nevertheless, the general shape of his point is something I certainly agree with -- an awful lot of medical spending is wasteful, and spending money on medical care isn't a particularly effective way of improving the health of the population. I'm taken with this opening metaphor:

Car inspections and repairs take a small fraction of our total spending on cars, gas, roads, and parking. But imagine that we were so terrified of accidents due to faulty cars that we spent most of our automotive budget having our cars inspected and adjusted every week by Ph.D. car experts. Obsessed by the fear of not finding a defect that might cause an accident, imagine we made sure inspections were heavily regulated and subsidized by government. To feed this obsession, imagine we skimped on spending to make safer roads, cars, and driving patterns, and our constant disassembling and reassembling of cars introduced nearly as many defects as it eliminated.

The crux of the matter, is that while there certainly are specific areas of medicine where you see a lot of efficacy (Hanson names "immunization, infant care, and emergency care") there are also a lot of areas where you don't see it. Moreover, while the current American policy environment is stingy about health care in some regards, Medicare is a very generous program and the tax treatment of employer-sponsored health care is, de facto, a giant and totally untargeted subsidy for the consumption of medical care. Obviously, though, health care policy has a "health" dimension but also a social justice dimension so even though a libertarian like Hanson and a liberal like myself can agree about this basic point this still leads in various directions.

For one example, see the centrist authoritarianism of Philip Longman's "TheHealth of Nations" for one approach to the phenomenon Hanson's pointing to. Alternatively, there's the wonky social democracy of Brad DeLong's impractical scheme or some versions of Jason Furman's progressive cost sharing proposal.

Be that as it may, I think Hanson's observation that "humans long ago evolved a tendency to use medicine to 'show that we care,' rather than just to get healthy" partially explains why things like the UK's National Health Service generate so much bang for the buck. In effect, a highly centralized state run health care system is able to put a cap on how much demonstrative caring can be done through the health care system. Nobody's going to say to his or her spouse, "well, sure we could afford the procedure, but it doesn't really stand up to cost-benefit analysis compared to spending the money on organic produce for the kids" but if bureaucrats stand in your way well, then, that's hardly your fault.

Comments

"Nevertheless, the general shape of his point is something I certainly agree with -- an awful lot of medical spending is wasteful, and spending money on medical care isn't a particularly effective way of improving the health of the population."

Matt,

You and I are both too young to remember, but there used to be plagues affecting the US that were obvious and scary to the population, such as a very high death rate from heart attack for men in their 50s or polio or breast cancer. I think we are now in the post-Ralph Nader health care era. There is much more regulation, most specialties promulgate evidence-based guidelines, and primary and preventive care are routine in populations that benefit from them.

The over-spending in medicine comes from 3 sources: 1) aggressive attempts to prolong life in its last year 2) expensive imaging to confirm diagnoses 3) the lack of generic medication use and 4) the incredibly high administrative costs of running private health insurance. It is nearly impossible to fix problems 1 or 4 without a political mandate, fix #2 without an evidence-based funding limit, or fix #3 without limiting medication payments. In other words, although we have been successful, reigning in the increase in spending basically requires a political will that neither doctors, who make money and want to be sure they are doing the right thing, or patients, who want something/anything done, are interested in pursuing.

Thanks,

Josh

Posted by Josh B | September 12, 2007 8:37 AM

Off point slightly, but...
How many other countries have their airways flooded with Pharma commercials touting new vaccines we never knew we needed. Or the horrors of "restless legs" and erectile dysfunction? I mean, we are barraged with how diseases and syndromes are lurking in our bodies and our only salvation is "consult your doctor".

Just following doctor's orders!

Posted by magisterludi | September 12, 2007 8:39 AM

"The crux of the matter, is that while there certainly are specific areas of medicine where you see a lot of efficacy (Hanson names "immunization, infant care, and emergency care") there are also a lot of areas where you don't see it"

Name a few.

Sk

Posted by Sk | September 12, 2007 8:47 AM

Ths analogy is flawed.

80 percent of medical costs treat 20 percent of the population. We don't spend our money on unnecessary tests and checkups (inspections and adjustments) we spend it on sick people getting treatment (replacing the transmission after it blows at 100,000 miles).

He's setting up a false choice here by suggesting that most healthcare is unnecessary treatment of healthy people, so if we just make it harder to get in the door, costs will be contained, to no bad effect.

That's not true. Most unnecessary care is actually provided to sick people, and it's not all unnecessary, just some of it, and it's buried deep and tangled up in the necessary care.

You can ferret out and contain these costs and improve care, but a blunt force approach ain't gonna work, and can actually cost you more money.

Also, it's not compassion that motivates overtreatment, it's much more a lack of focus on palliative care, and an avoidance of dealing with death. People want a cure because they can't live with the pain. Families want you to keep trying because then they don't have to face the fact that their loved one ain't gonna make it. Terry Schaivo? Denial is powerful.

Even if people WANT to refuse medical treatment, we really still lack the institutional procedures for them to communicate those wishes in advance. There's a woman who has chosen to tattoo "do not resuscitate" on her chest, and I saw a bunch of experts saying that she'd likely get treatment anyway.

I think it's ludicrous to suggest that there's easy free money to be had in restricting unnecessary care for healthy people.But it's also ludicrous to take the contrarian point, that there's only money in restrcting necessary care to sick people.

There is, in fact, a lot of unnecessary care being provided to sick people, and a serious focus on best practices, palliative care, and end of life care could save some big money, while giving people the healthcare they want and need. It's not easy free money, it's HARD free money. But it's there.

Posted by anon | September 12, 2007 8:54 AM

Watching economists and policy wonks comment on healthcare is like reading up to 1960 in that recent Time magazine list of 'Worst Cars'. The wackiness is obvious if you think for a minute about the ideas that emergency rooms are a really high value-for-money experience, or that Medicare is a 'very generous program'.

To use Hanson's own automotive analogy, is it really smarter to spend your money pulling a wrecked car out of the ditch than it would be to change out brake pads regularly? If Medicare has the financial clout to buy brake pads and service at half the price anyone else pays, and decides regular changes are cheaper than fixing wrecked cars, does that make it the last of the big spenders?

Perhaps we should expect that when young affluent writers wander into the weeds on healthcare, what we're going to get is word salad.

Medicine as we know it is a social construct in which the patient is exempted from some social obligations in exchange for assuming and fulfilling the role of the 'patient'- a word usually defined as meaning 'long suffering'. However, this construct, as we know it, does not go back 'centuries'- it springs from the Flexner Report of 1919, the template used by the AMA and state medical associations to persuade state legislatures to pass laws creating medical monopolies.

Wouldn't it be great if our smartaleck wordsmiths had to spend two years working as a nursing aide before they were allowed to comment on healthcare? Opinions would still differ, but at least then they would have the memories of some actual patients peering over their shoulders as they wrote.

Posted by serial catowner | September 12, 2007 8:57 AM

Re serial cat's point: even if we change the payment scheme (e.g. to single payer), unless we also change the model and incentives for primary care workers nothing significant will change in the entire system. I was very lucky to have a very good primary care physician for the last 3 years, but he just left for a salaried teaching job (leading to a shot at a deanship). I suspect the reasons were lack of money (I can't figure out how his practice survived) and lack of respect from the entire rest of the medical world. We will absolutely need a restructuring that changes this or the whole effort will be for naught.

Cranky

Posted by Cranky Observer | September 12, 2007 9:14 AM

but if bureaucrats stand in your way well, then, that's hardly your fault

We have bureaucrats here, too, MY. They're just private. (They're the ones that are spot in the middle of most medical horror stories.) And yet, there's still a bang-for-the-buck differential.

Posted by SomeCallMeTim | September 12, 2007 9:23 AM

I very much recommend Hanson's posdcat on Econtalk. It is interesting and thought provoking.

http://www.econtalk.org/archives/2007/05/hanson_on_healt.html

Hanson might have libertarian tendencies.

However if you accept his various arguments regarding our irrationnality towards medical care, the case for an English style socialized medical system becomes quite strong, as you correctly point out.

In particular, hanson argues that:
- people don't want to think about death and therefore don't really try to gather the relevant information about the effects of medical procedures or the performance of individual caregivers,
- when they are given the information, they don't act on it,
- people don't like to make choices about medical care especially economic tradeoffs...
- people like to be overinsured, etc...

When you put everything together, a clear conclusion is that people would be collectively and individually quite happy to have some bureaucrats make the decisions for them and that it would lead overall to more rational decisions and a more cost-effective system.

In the long run however, one could be worried about the effects of such a system on innovation in medical care.

Posted by fabrice | September 12, 2007 9:41 AM

We have bureaucrats here, too, MY. They're just private.

Right, they're private and so they don't have the right incentives.

Posted by Matthew Yglesias | September 12, 2007 9:43 AM

Matt, I normally think your insights are pretty good and am not inclined to use profanity, but may I ask what fucking planet you're living on? Can you think of one single instance of a real person sending a family member to the doctor primarily for "demonstrative caring"? Or to put it in your own condescending example language "I love you so much, dear, go have a nice medical procedure!"

This is another in a long line of arguments from market-worshippers that depend on the idea that people really want to go to the doctor, and have to be prevented from doing so, because going to the doctor too much is what makes our healthcare expensive. Going to the doctor is time-consuming, tedious, often difficult to fit into a work schedule, frequently uncomfortable, and that's before you factor in fighting with your insurance to pay for some of it. Nobody likes going to the doctor. This is just a retread of the "moral hazard" pro-HSA myth, using large-scale aggregate statistics to make unsupported claims about specific behavior.

Previous commenters have outlined most of the actual reasons for the cost of our healthcare system, to which I'll add one more -- the perverse incentive in our decentralized system not to encourage preventive care, even though it is well-established to produce better outcomes and lower costs, because the benefits are likely to accrue to a different insurer than the one bearing the costs. That's the real moral hazard in our system.

And if you think that only national health services have bureaucrats standing in your way, clearly you've never dealt with insurance company pre-authorizations. That's one instance where I'll give some credit to the free-marketers -- a government bureaucrat whose job it is to contain costs has nothing on a corporate bureaucrat whose corporate performance depends on denying them.

Posted by Redshift | September 12, 2007 10:11 AM

There's an old saw about a businessman who says, "At least half the money I spend on advertising is a complete waste of money. The trouble is that I don't know which half."

The same problem exists in medicine. Even if half the money is wasted, we often don't know what was necessary and what wasn't until we've cured or killed the patient. And sometimes not even then.

And I'm not sure that's a fixable problem.

Posted by MikeT | September 12, 2007 10:49 AM

Matt, the comments you are getting here suggest that most people from all political sides have not accepted the point on which you and I most clearly agree: the very low health-value of added medicine. David Cutler, who acts surprised that I think this fact hadn't gotten enough attention, should take note.

Posted by Robin Hanson | September 12, 2007 11:14 AM

I'm not so sure it's overstated.

When you figure in 1000% patent markups on drugs, the crowding out of perfectly efficacious treatments for even the most basic ailments by the most expensive high-tech equipment, the inflated salaries of the professional cartels (including the unnecessary crowding out of GPs by specialists), and the high insurance overhead, it may actually be understated. Most of the real medical care the average person needs could probably be done by a GP or even a PA, on retainer (a la the "old lodge practice" system that the professional cartels wiped out), using generic medicine, in a neighborhood cooperative clinic for a membership fee of a few bucks a month.

Posted by Kevin Carson | September 12, 2007 11:33 AM

Once you start pumping billions of dollars into the health care system through direct payments and subsidization of private insurance, the system gets distorted and re-oriented around those payments. So much so that for the most part it makes no sense to even talk about the health care system in this country as being driven by motivations to make individuals healthy. The key relationships and motivations in the system are those between providers, insurance companies and the government. Bureaucratic "inefficiency" is not a bug it is indeed a feature. Excessive tests, prescriptions, and overweighting of care on the dying are not bugs, they are features. I am not saying it was all designed that way on purpose from the beginning, but the system has evolved this way so that now all these things as features. Patients indeed just play a role.

Posted by Doug | September 12, 2007 11:37 AM

Wouldn't it be interesting to ask a chronically ill pundit, one who receives a lot of medical care, what they think of over-spending on health care?

Posted by matt | September 12, 2007 11:55 AM

Wow, here's some low fruit:

Can you think of one single instance of a real person sending a family member to the doctor primarily for "demonstrative caring"?

"Little Billy has the sniffles(an earache, a toothache, cries a lot), Doctor. Can you give him some antibiotics(or other pill)?"

Posted by kenga | September 12, 2007 12:08 PM

I applaud what serial catowner wrote:
"Perhaps we should expect that when young affluent writers wander into the weeds on healthcare, what we're going to get is word salad."

I propose that only people with chronic illnesses and uninsured people over the age of 50 be allowed to write on the topic of medical economics. These young, well-off, self-assured (apparently all male) cost-benefit analysts should, I suggest, study Pentagon economics, where more money is wasted with a quite deleterious effect on health worldwide.

Josh, my health insurance company makes my primary care physician's life miserable with endless paperwork for prescribing a nongeneric drug when its "experts" suppose there's an equivalent generic, even when I've been taking that generic (OTC actually) for years and it's given me side effects that had to be treated with other drugs. No, I'm not in an HMO. How many people these days are wasting money this way?

anon, some hospitals are very aggressive about pushing patients to complete "living wills." As a friend said, "I expected another box to check off labeled 'please euthanize me.'"

I wonder if those Dartmouth researchers wondered how many elderly people felt "chilled to the bone" in Minnesota winters and moved to Miami to try to warm up?

And just what is this "added medicine" of "low value" that people should be doing without?

Posted by Joyful Alternative | September 12, 2007 12:08 PM

Brilliant ju jitsu except for the bit about organic produce. There is no evidence that organic produce is any healthier than non organic produce. The existence of the market with no government subsidies at all shows the limit of the libertarian argument. People do silly things even when they are not subsidized.

Also, as you note above, much of the cost of health care is do to underutilization of cheap things like statins (I take mine every day and I bet you don't)

Posted by Robert Waldmann | September 12, 2007 1:59 PM

Re: Also, it's not compassion that motivates overtreatment, it's much more a lack of focus on palliative care, and an avoidance of dealing with death.

Even palliative care is hugely expensive. Two years before he died my father fell desperatelty ill and spent five weeks hospitalized. Hios doctor predicted he would not pull through, and nothing extraordinary was done for him: No ICU, no surgeries, just basic supportive care and meds. The hospital bill for that was over 60K (and this was in 1989).

Re: Even if people WANT to refuse medical treatment, we really still lack the institutional procedures for them to communicate those wishes in advance.

Um, living wills, advanced medical directives, medical power of attorney etc? When my father did come to the end of his life his oft-expressed wishes never to hooked up to "machines" was honored without question. He even had an orange plastic bracelet on his wrist, symbolizng "Do Not Ressucitate". And when he breathed his last, they didn't. How common is it really that such orders are ignored? Maybe in religious hospitals steeped in "culture of life" stuff, but otherwise I think it would be quite rare. Terri Schiavio, let us recall, left no such orders, and was incapable of giving them later. But in every hospital death I've known about, the dying person was not ressucitated, or life support was unhooked in hopeless cases, when his/her wishes were made known in advance.

Posted by Jonf | September 12, 2007 3:06 PM

"Little Billy has the sniffles(an earache, a toothache, cries a lot), Doctor. Can you give him some antibiotics(or other pill)?"

You proved your point with an invented scenario?

How often does your above exchange happen, and how much does it cost the rest of us?

Please avoid replying with "a whole bunch, shuh!"

Posted by MasonMcD | September 12, 2007 4:06 PM

An old department store owner used to say that half his advertising budget was wasted, but he just couldn't figure out ahead of time which half. S

Same with medical care. I had a doctor tell me not to worry about that big lump in my armpit. It didn't justify getting a scan. (Those things are expensive, you know). So, I found another doctor who rushed me into a CAT scan and, sure, enough, it was lymphatic cancer.

Posted by Steve Sailer | September 12, 2007 4:32 PM

Ha ha, Steve Sailor reminded me of my own big lump. It's an undoubtedly (I hope) benign lump that has grown over the years to a remarkable size. I just don't want to take the time to have it removed.

Anyway, a few years ago I was pretty darn sick, they couldn't figure out what caused it, and boy, did they look suspiciously at that lump. And that's all they did, because I kept telling them it had been there for years.

So, God knows how many hundreds of thousands of dollars later, I left the hospital, lump intact. They had quite economically never done the $100 procedure to remove it and end the guessing.

From what I have seen, it is very rare for a Medicare patient to undergo an unneeded procedure.

Posted by serial catowner | September 12, 2007 6:34 PM

I'm a pundit and cancer survivor, thanks. What this thread overlooks is the information vacuum in which people make medical decisions. Fire up the data and a lot of this will take care of itself.

The problem is, people may not like what the data say, or may want treatments the data say they should ditch in favor of cheaper, equally effective alternatives. Remember the HMO backlash? It'll come again as electronic health records get linked into something that lets us analyze outcomes intelligently for the first time. Your doc who likes to do bypasses -- expensive bypasses -- is going to fight for his right to party. In Utah, to take one example, the leading hospital chain has all but barred elective deliveries before 39 (I thinkj it was) weeks, because statistically they lead to high complication rates. Good news economically? Sure? Good for babies? Yup. But not so good for a Mom who wants to exercise her own choice, and whose doctor may be confident that there will be no complications. Look for lots of arguments like this -- but this is where the costs are. Clinton's health plan makes a stab at some of how the government will advise on which procedures are wasteful while trying to avoid mandates (HMO-like, plus the worst of the English system) that will just make people mad. Watch that space.

Bush, for his faults, actually did a fair amount of good in building the information infrastructure of a functioning health market. Don't throw all the virutes of the first really functioning market in health care out with the HSA bathwater --- the Dem pres candidates all know about health IT and how it will change the health market, and all support it. It's also the case, if you beleve Rand's numbers, that savings from health IT will be enough to extend the solvency of the Medicare trust fund indefinitely.

The immediate policy question is whether Medicare will lead in acting on this data as it improves, to stop paying for bad and unnecessary care. This is where the real money is -- not nonsense about drug advertising, or the notion that all will be well if only Joe Biden decides off the top of his head what Cardizem should cost. Please. Last week's announcement that Medicare will stop paying for care necessitated by medical errors is a start. But why do Medicare recipients cost twice as much in Miami as Minneapolis or Portland?

Taint private insurance's fault. There's no difference in drug-advertising rules. It's that payers, led by Medicare but not limited to it, don't crack on providers whose protocols are wasteful. Why? Because no one wants to take their Congressmen's phone calls. And because the data, such as it is, is often limited or subject to debate. That's what EMRs will ultimately fix, especially as they are linked and allow for a dramatic acceleration of research. As data drives more ecisions about which care to use, health cost explosions will begin to fix. (May already be happening; see moderating health inflation of last two to three years, which coincides with a near-doubling of the proportion of providers who use EMRs).

This saving will also help access. An aide to a Dem Gov. told me in 2006 -- and the Gov largely agreed the next day -- that the first thing to do with the savings from health IT is to funnel the first few percent into SCHIP. It was an elementary matter to cover every child in that state if the cost savings were as large as they expected, the aide said. And that's without a single bureaucrat governing care of consenting adults, or months-long queues for elective procedures.

Be gentle with me. My diagnosis anniversary (#3) is today.

Posted by Tim Mullaney | September 13, 2007 8:29 AM

Shouldn't writing about an issue require some familiarity with basic facts? Where on earth did Hanson (and MY) get an idea that emergency care has "a lot of efficacy"? This is precisely backwards. Unless they can point to some reputable research (i.e., peer reviewed analysis, as opposed to a spin-tank press release), the old saw about "an ounce of prevention" still holds true.

Smallpox immunization cost money, but incomparably less than treating everyone in an epidemic in emergency rooms.

So not only Johnny's running nose should be looked at, but the most efficient way to spend healthcare money is to have Johnny looked at very regularly. Then maybe you won't have to resort to heroic (and hideously expensive) measures trying to save Johnny's life when he gets very sick.

Posted by AO | September 13, 2007 9:41 AM

AO-
I read that "emergency care" as "trauma care", which seems to make more sense. Broken legs and amputated fingers, and maybe sudden 103 degree fevers. Not, say, complications from long-untreated tuberculosis.

Posted by jack lecou | September 13, 2007 7:24 PM