![](https://ezraklein.typepad.com/bina_.gif")
« A Malignant System | Main | What Union Members Understand »
November 20, 2006
The Problem With Health Spending In A Single Graph
Ever thought to yourself, "Man, I know a small number of patients account for an inordinately high percentage of total spending, but I sure wish I had a graphical representation of the breakdown?" Uh, me neither. But I'm glad Kaiser created one anyway:
You want to slow health spending? You have to slow it on these patients. HSAs and their brethren like to pretend that by forcing caution on when you get a test for strep throat, we can significantly effect health costs. Not so. HSAs have a spending cap, and once it's broken, all care is covered. They do nothing but disincentivize basic care, which doesn't cost much anyway.
Instead, you're going to need to attack these costs at the top. You may need to ration. You'll definitely need to cut down on wasteful care. You'll definitely need to erase the payment structure that encourages doctors to prescribe the most intensive treatments. You'll probably need to give the government bargaining power so we begin paying less per unit of care, as all other countries do. More here.
November 20, 2006 in Consumer-Directed Health Care, Health Care | Permalink
Comments
Wow. I see Two Americas.
Posted by: Neil the Ethical Werewolf | Nov 20, 2006 1:26:51 PM
They do nothing but disincentivize basic care, which doesn't cost much anyway.
And which can help prevent more expensive care being needed later on ... as Poor Richard said, "an ounce of prevention is worth a pound of cure." For those who are wont to compare health insurance to auto insurance (and point out thus that health insurance covering health maintainance is really not "insurance" in the strict since, which indeed it is not), if preventive auto maintainance would so greatly lower the expected cost of auto-insurance claims, you could bet auto insurers would be providing auto-maintainance coverage too!
Posted by: DAS | Nov 20, 2006 1:42:32 PM
Don't worry, once Rangel gets them all drafted and medically discharged they will be eligible for VA health care.
BTW, my neighbor next door does not share your view of the VA system. He has been using it since he lost part of his leg in Vietnam and is in Walter Reed right now recovering from back surgery that was delayed for months.
Posted by: Guy Montag | Nov 20, 2006 1:53:14 PM
Shorter Montag: Irrelevant Non Seqitur! Anecdote is Data! Ha! Pwned!
Posted by: paperwight | Nov 20, 2006 2:09:07 PM
Huh... my ultra-conservative colleague is a retired Green Beret colonel, 100% VA disabled (parachute didn't open), and he loves his healthcare, although he has learned to not talk about his $5 prescriptions and low-cost family coverage in front of the rest of us too much, since small-business rates aren't exactly cheap. It took a couple of us a good part of the morning to explain to him why our parents and adult kids can't be covered on our insurance, because he simply hasn't dealt with for-profit healthcare for most of his life.
Might as well throw another anecdote out there, I suppose.
Posted by: latts | Nov 20, 2006 2:16:30 PM
What percentage of those top users are illegal immigrants?
Posted by: American Hawk | Nov 20, 2006 2:23:21 PM
American Hawk,
Aw, now come on, they probably did not mean to break the law. Illegal is so harsh. Let's call them accidental visitors.
Posted by: Guy Montag | Nov 20, 2006 2:32:14 PM
Sorry American Hawk, I know you weren't actually expecting an answer to your question...but here it is. Immigrant underutilize care, they don't overutilize it. Additionally, for reasons doctors don't quite understand, Hispanics are much healthier than whites of similar demographic characteristics. Experts call it an the Hispanic epidemiological paradox.
Your move.
Posted by: Ezra | Nov 20, 2006 2:36:51 PM
Ezra, American Hawk really meant all of the illegal IRISH immigrants, with their drinking patterns and fatty starchy foods. What about them? Huh? Huh?
Posted by: paperwight | Nov 20, 2006 2:48:17 PM
This Hispanic epidemiological paradox is pretty neat. I wonder if people of mixed ancestry are resistant to more diseases. That'd bode well for any children I'm likely to have in the future.
Posted by: Neil the Ethical Werewolf | Nov 20, 2006 2:58:16 PM
The 80/20 axiom for health care spending is more accurately a 50/5. While the rate of random vs. chronic illness is important, the VA model makes a lot of sense for those who we know are ill and that we know are going to stay ill.
I think the upside to HSAs is that, in lieu of universal coverage, it is probably the most appealing way to coax younger (and healthier) workers back into the pool. One of the most disturbing trends in health care is that offers of coverage have remained fairly constant, but the decline in employer-sponsored care is primarily due to take up rates by employees. For younger, healthier workers this may not actually be a bad decision as they are taking it in lieu of wages and it is simply being passed along to someone else (another reason ESI needs to die).
Posted by: Alex | Nov 20, 2006 3:04:57 PM
Ezra-- That link you gave only has *averages*. My question was, at the top 10% or so of healthcare users, what are the demographics like? For instance, 90% of illegal immigrants could be near the bottom of healthcare usage, but the other 10% could be spread throughout the top 10-15%. The problem with averages is that it obscures patterns like that. Is there any actual information available on the breakdown on any of these top groups?
I'll put it in another way that might make progressives happier: So, are these top 10% rich people who can afford needless care that probably isn't doing anything, or average joes bankrupting themselves? Any data on that?
I seriously want to know.
Posted by: American Hawk | Nov 20, 2006 3:23:49 PM
I wonder if people of mixed ancestry are resistant to more diseases.
No way, 'cause I get sick way to much for that to be true.
I'm English and German.
Posted by: Stephen | Nov 20, 2006 3:27:29 PM
Shorter American Hawk: Is there any plausible way at all I can blame this on the Messican Hordes? Any way at all? Please?
Posted by: paperwight | Nov 20, 2006 3:30:43 PM
...I wanna get in on this epidemiological paradox.
Posted by: Adrian | Nov 20, 2006 3:35:09 PM
Oh yeah, that take up rate thing has a lot to do with whether they elect family coverage. And on the immigrant spending deal, the highest cost of care associated with immigrant groups is, not surprisingly, pregnancy. If the kid is born here, then he/ she is an American citizen (at least under one reading of the Constitution). I'd guess they don't measure the spending in the study by whether the parent was legal/illegal. If they further allocate the cost of the pregnancy to the American-born child, then it would be further skewed. If anyone has access to the study, I'd be interested to know about the methodology.
Posted by: Alex | Nov 20, 2006 3:50:31 PM
Hawk -- read more closely. It is averages. But it's total, too. Given the trends explained above, if immigrants make up 10% of the population and account for 8% of the total spending, then they are under-represented in that top group.
Posted by: Ezra | Nov 20, 2006 4:41:35 PM
Ezra, American Hawk really meant all of the illegal IRISH immigrants, with their drinking patterns and fatty starchy foods. What about them? Huh? Huh? - paperwight
Why when you mentioned this, I immediately thought of the Simpsons episode where Bart and Lisa revealed that "snake whacking day" was originally an excuse to beat up the Irish, and once this is revealed, a stereotypically Irish person appears and confirms this?
Posted by: DAS | Nov 20, 2006 4:59:49 PM
I took many a good lump meself, but it was all in good fun.
Posted by: Alex | Nov 20, 2006 5:20:02 PM
Shorter Hawk: Look! Brown people gonna git'chur health care!
Posted by: NBarnes | Nov 20, 2006 6:20:40 PM
So long as private care is still available for those who want fork up the cash for it, I'm game for all of this.
Posted by: DRR | Nov 20, 2006 7:56:20 PM
Neil the ethical etc-
If you see "two Americas," you're reading the graph wrong. The "bottom 50%" category doesn't mean the bottom 50% in terms of income. It means the bottom 50% percent in terms of medical usage. For example, I personally am in the top 10% of income in the US, but I haven't seen a doctor in 3 years and I've never spent a night in a hospital. So I'm in the bottom 50% for medical usage.
The point is that most health costs are spent on the small fraction of the population that is really sick. It makes no sense to try to cap costs on people who aren't sick, because those people aren't over-utilizing care. What you have to do is to figure out more efficient ways of caring for people who are sick.
I'm afraid that this does mean rationing. For example, my father, who died at age 82, was healthy, living at home and caring for himself, until to eight weeks before he died. He had some prescription drug expenses, but no serious illnesses. Then he had a stroke, and spent eight weeks in intensive care, all paid for by insurance. For the first four weeks, the care was warranted, but then he had a heart attack and in a few days it was clear that he was not going to survive. Even so, he had four weeks of round-the-clock intensive care, during which he was unconscious and on life support. He eventually died when his kidneys failed- but if we had wanted, we could have put him dialysis for more weeks of heroic care. At least two weeks of what he did get - the last two weeks - was of no use to him or us. The cost of treatment like that is what you see in the column on the far left. No reasonable society would even offer that kind of medical care, much less pay for it.
Posted by: JR | Nov 20, 2006 8:18:27 PM
Rationing is a loaded term, because it is not limited to end-of-life care.
But it's difficult to talk about dying cheaply without sounding incredibly insensitive. and don't forget the Schiavo lunatics. not using extraordinary measures to get every last second of life is apparently so grave a sin as to require the intervention of the Senate Majority Leader.
Posted by: Francis | Nov 20, 2006 8:56:12 PM
of course there was the Kerry health plan proposal for government risk pools for catastrophic cases
http://personalinsure.about.com/cs/whatsnews/a/aa120103a.htm
Never got a single coherent analysis except big guns from the right.
Posted by: marc | Nov 20, 2006 10:44:09 PM
Good post, but the graph alone does not, I think, destroy the argument for HSAs, at least as I understand it. Nobody credible argues that HSAs are some sort of panacea. They are but one means for reducing utilization at the margin, and they may be among the least costly. Other methods, such as by extending the use of bureaucratic fiat and queuing into more treatment areas, are going to be less popular even than HSAs. Price controls -- and that is what "government bargaining" really is -- will definitely reduce innovation in healthcare technology over the medium and long term. Because of development and regulatory lags we won't see the impact right away, so a system with price controls on drugs and devices would appear to work in its early years. Eventually, though, the innovation will slow down. My wife has MS and is on drugs that probably would not have been developed if the United States government acted as a monopsonistic buyer. That result is painful to contemplate.
Posted by: TigerHawk | Nov 20, 2006 11:44:30 PM
The comments to this entry are closed.
![](https://ezraklein.typepad.com/ezraklein.jpg")