American politics
Democracy in America
Health care reform
Still diagnosing the problem
- Mar 13th 2010, 16:08 by M.S.
AUSTIN FRAKT has asked his readers to suggest responses to my post on just how big administrative costs in the health-care system driven by the fractured private health-insurance system are. Whew! That's a mouthful. Anyway, a couple of his readers have suggested possible responses to the Himmelstein-Woolhandler paper that argued that administrative costs in America suck up 31% of health-care spending. But they're not terribly definitive.
The first is a 1992 paper by Patricia Danzon in Health Affairs. While it mentions some of Mr Himmelstein's and Ms Woolhandler's earlier work, it obviously doesn't address the 2003 paper itself. The main thrust of Ms Danzon's argument seems to be that head-to-head comparisons between the American and Canadian systems are flawed because of "hidden costs" in the Canadian system, chiefly waiting lists and the deadweight loss of funding health insurance through taxes rather than premiums. This may be true, but it makes the argument even more complex by bringing in yet more debatable issues; I was hoping for someone who simply addressed the issue of how to calculate administrative costs in the American system.
The second is a response to Mr Himmelstein and Ms Woolhandler by Henry Aaron, the Brookings Institute economist, in the same 2003 issue of the New England Journal of Medicine in which they published their study. Mr Aaron begins by writing that he, too, thinks the fragmented American system entails absurdly inflated administrative costs:
Like many other observers, I look at the U.S. health care system and see an administrative monstrosity, a truly bizarre mélange of thousands of payers with payment systems that differ for no socially beneficial reason, as well as staggeringly complex public systems with mindboggling administered prices and other rules expressing distinctions that can only be regarded as weird.
However, Mr Aaron thinks the Himmelstein-Woolhandler approach grossly oversimplifies the task of estimating administrative costs. To start with, he thinks they overstate administrative costs by $50 billion (as of 1999) by using the gross difference between American and Canadian per capita administrative spending, rather than expressing the difference as a percentage of each country's health-care spending. That would still yield an estimate that America could save a whopping 14.3% of its health-care costs on reduced administration alone by switching to a Canadian-style single-payer system, if the rest of the calculation is correct. But Mr Aaron thinks it isn't; he thinks that complications like disparities in wages between clinical and administrative staff, institutional differences, and so on make it almost impossible to compare the two systems. Finally, he says,
The most important question is what these differences should tell policy makers. I believe the answer is, “Not much.”...The U.S. health care administration, weird though it may be, exists for fundamental reasons, including a pervasive popular distrust of centralized authority, a federalist governmental structure, insistence on individual choice (even when, as it appears to me, choice sometimes yields no demonstrable benefit), the continuing and unabated power of large economic interests, and the virtual impossibility (during normal times in a democracy whose Constitution potentiates the power of dissenting minorities) of radically restructuring the nation’s largest industry — an industry as big as the entire economy of France.
This seems to me like a very strange thing to say. Obviously different countries have different health-insurance systems for deep-rooted historical reasons, but to use that as a reason to refuse to make any cross-country comparisons is a rather self-defeating thing for a social scientist to do. I was really hoping that Mr Aaron might have simply suggested some better ways to compare administrative costs between private and single-payer health-insurance systems. As for the declaration of defeat in the face of path dependency, it appears at the moment that Mr Aaron was too pessimistic. Health-care reform now depends on a few wavering votes in the House. It may not make it. But reforming America's health-care industry doesn't seem to be "virtually impossible". It's just very, very hard. Given the long-term budget picture, it actually seems pretty much inevitable.
About Democracy in America
In this blog, our correspondents share their thoughts and opinions on America's kinetic brand of politics and the policy it produces.
Advertisement
Products & events
Stay informed today and every day
Subscribe to The Economist's free e-mail newsletters and alerts.
Subscribe to The Economist's latest article postings on Twitter
See a selection of The Economist's articles, events, topical videos and debates on Facebook.
Advertisement
The Economist welcomes your thoughts
We are making continuous improvements to The Economist website and are interested in your thoughts.
Readers' comments
The Economist welcomes your views. Please stay on topic and be respectful of other readers. Review our comments policy.
Sort:
Nice to have such a quick follow-up on this, even if it's not entirely satisfying.
One question along these lines: does the proposed health care bill require insurance firms to cover all doctor-ordered treatments (as, for example, the Dutch system does)? Anecdotes from my Father Doctor and Aunt Nurse suggest that denied payments are a huge issue for providers. Some providers suspect insurance firms reject most claims once or twice (for uncrossed t's and other still murkier reasons) simply because they know that hospitals — already drowning in paperwork — generally stop following up on claims after a couple attempts. Which, of course, forces them to make up the difference by charging everyone else more. Anyone know if this is a widespread issue, and if anyone's tried to quantify the cost of it?
I don't know how widespread a problem it is, Uncle, but I do know that the practice isn't limited to hospitals. My parents battled their insurance provider on a near-continual basis throughout most of my childhood, with just about every submitted claim rejected. They would then appeal, and sometimes appeal again after a second rejection, and finally force the provider to, you know, provide. At least most of the time. Fortunately, they were in a financial position to keep up those battles and cover the remainders, which is nice because otherwise I might not still have a sister.
What "fractured" the private health care system, and from what whole was it initially "fractured"? I think we may be leaning a little heavily on this single loaded (and not very accurate anyway) term lately.
Uncle Alfred, I wrote my college thesis in Economics on the topic of (in)efficiencyin hospital service funding. As of '95, I can tell you I looked hard for information on that and couldn't find it, although then it took days and weeks of library crawling and government-calling to find what five minutes of Googling would now yield with a pornographic lagniappe. But I can tell you a lot of data more obvious than that wasn't kept and might still not be kept.
But that might be a good example of how inefficiency builds on inefficiency. One thing that would be interesting to investigate (for a young, collegiate Doug Pascover and not the present lazy, balding and tired one)- is there a significant difference between the paperwork in-network agents are asked to provide insurers versus out-of-network. There should be some difference because in-network providers shouldn't have to certify themselves as bonafide healthcare enterprises and out-of-network probably should, but if insurers use paperwork as a means of avoiding payment, you might find a bigger difference than would be expected.
We all know the current system is grossly inefficient; exactly how bad it is scarcely matters. For thirty years now, I've heard doctors of my acquaintance say they would prefer nationalized health care to the mess they have to deal with. Until we have a single payer that eliminates fee-for-service, there will be no significant cost control. It needn't be run by the government; I'd rather see a private non-profit. Uncle Sam can give money to individuals based on age or need.
The insurance industry will have to die. Fee-for-service, too. The only question is how desperate the fiscal situation will have to get before it happens.
I can't think of another industry where everyone offers the particular solution while not being in that industry in any operational terms. If "Hammering" Henry Aaron or anyone else has so many good ideas on health insurance administration, why doesn't he start being an administrator rather than a wonk?
There is way way too much amateur knowledge (especially in DC) about a variety of industries, with healthcare being the current rage for politicians and bloggers. I keep reading how wasteful and low value add so many parts of the industry are so then...go start your own healthcare company, insurance company, pharmaceutical. What is stopping you?
JBP
Regarding administrative costs, the two systems are very different. Most of the admin costs in Canada are borne by either government or hospitals; in the US, admin costs are split among doctors' offices, insurance companies, hospitals, and governments (at all levels). Even the per capita costs may be calculated differently. The US, for example, includes long-term elder care in health care costs. And the US has the burden of the medical tort system - 2%, plus an undetermined amount for defensive medicine - which isn't present in Canada. It would be interesting to compare medical records practices in Canada and the US to see just how much is being done in the US to protect against lawsuits.
John Powers is right-on when he notes that many policy experts don't know how the system really works - or doesn't work. We're accepting input from people who've never worked in the system, don't understand it, and are often out to make a point. It surely needs improving. But as the man says, we're "still diagnosing the problem."
Interestingly, the Himmelstein-Woolhandler group seem to make a habit of challenging assumptions; their latest paper concludes that hospital computing improves neither cost or quality (Amer Jour Med, 2010). But that's really off topic.
CharlesVW: "We're accepting input from people who've never worked in the system"
That sounds like a pretty good idea to me, in basically all circumstances and for basically all industries.
"I can't think of another industry where everyone offers the particular solution while not being in that industry..."
Climate science? ;)
"go start your own healthcare company, insurance company, pharmaceutical. What is stopping you? "
Err, capital?
You ask this question pretty frequently and get the same response consistently. It's not like I can go out and get a bank loan to start a health insurance company. Or mix up some chemicals in my kitchen and bootstrap a pharmaceutical company.
But the underlying question with respect to health insurance is whether additional companies help or hinder.
If administration costs due to a bazillion different forms, varying payment amounts, etc. are huge, then even if my local credit union would finance my health insurance venture, the system would just be made worse.
Some industries benefit from competition (say, auto manufacture) while others don't so much (say, sewage sanitation). The underlying question is which category does health insurance belong to?
My gut sense -- as well as that of many Americans -- is that a single governmental payer with multiple competing private providers would be the most efficient. A gut sense that comes from observing the real-world experience of other countries with similar systems, BTW.
But hey, let's explore the underlying data. Except that it's evidently really hard to obtain...
@ Doug Pascover
There is a higher administrative cost for out of network in terms of contracting. Treatment for the patient has to be negotiated for each patient with the medical provider. For in network, there is already a contract in place for all patients.
@ bampbs
One way to eliminate fee for service is vertical integration. Kaiser Permanente is not only a insurance company but owns and operates it's own medical facilities. So doctor, nurse, and other salaries are paid out of fairly stable insurance premium revenue.
Vertical integration would also get rid of one of the biggest administrative costs, medical insurance claims.
"If administration costs due to a bazillion different forms, varying payment amounts, etc. are huge, then even if my local credit union would finance my health insurance venture, the system would just be made worse."
This can remedied if the medical industry would agree to implement common standards. The IEEE group does this very successfully for the computer industry and are starting to host conferences with an eye towards doing the same with medical.
But it doesn't address the issue of getting everyone to use the standards. Proprietary hardware and software are natural barriers of entry for competitors. The customer lock in they create guarantee years if not decades of profits.
The government might be able to help the process along. They can have programs like Medicare mandate the use of those standards. That would give the standards entry into every facet of the medical field.
sparkleby wrote:
CharlesVW: "We're accepting input from people who've never worked in the system"
That sounds like a pretty good idea to me, in basically all circumstances and for basically all industries.
-Except for climate science, of course ;)
If healthcare is so profitable or efficiency gains so easy to capture...then capital will certainly follow.
It is like a bunch of vegetarians telling me how T-Bones Steaks do not taste good, because per their theory the taste of the rib area of the steer is not as flavorful as tofu.
Go to Smith and Wollensky yourself and you will have a more informed opinion"
JBP
JBP