Why is health care so expensive?

Health care is currently taking up $2.8 TRILLION dollars of our economy, or roughly 16% of GDP. In 1960, fifty years ago, health care was taking up 5.2% of national GDP. So health care has gotten way expensive over the past fifty years. Is that good, or bad? Let's take a look...

Disease	Outcome 1960	Outcome 2009	Cost of treatment
Leukemia	Go home to die	33% survival rate after 5 years	$1,000,000
Liver failure	Go home to die	Transplant, almost 100% success rate	$400,000
Heart failure	Go home to die	Heart transplant	$300,000
Heart valve failure	Go home to die	New heart valve	$130,000
Kidney failure	Go home to die(*)	Dialysis or transplant	$130,000

And it goes on and on. Pages and pages of diseases where my mom's 1960 edition of the Merck Manual says "send patient home to die with palliative care" now have actual treatments. Most of these treatments were very expensive to develop, and many address rare but deadly diseases like leukemia where amortizing the costs over large numbers cannot be done, unlike, say, myopia, which afflicts 30% of Americans and thus allows the costs of developing LASIK to be amortized over millions of people so people can pay for it out-of-pocket.

So to a certain extent, it's good that health care spending has increased since 1960. We now have treatments for far more diseases than if we'd stayed at a 1960 level of health care spending. People blast the health insurance industry and Medicare for causing health care spending to skyrocket, and to a certain extent that's true -- without the pooling effect and resulting distortion of the free market created by health insurance and Medicare, we could never have increased health care spending to this extent, because sick people are consuming 16% of national GDP, yet produce nowhere near 16% of national GDP -- they simply could not afford this level of healthcare if forced to pay entirely out-of-pocket. But: The only reason we have these treatments in the first place is because of the pooling effect created by having health insurance and Medicare -- otherwise nobody could afford to buy them, and they would never have been developed because the few millionaires able to pay for these procedures out-of-pocket would be too few to justify developing them.

Okay, so now we have the #1 cause of the increase in health care spending: new and expensive medical procedures and drugs. The problem is that there is no inherent upper bounds here, and we're already creaking under the burden of 16% health care spending. Still, we do not want to cut off innovation by eliminating the ability of drug companies and researchers to create new medical procedures to cure as-yet-uncured diseases, so we don't want to just blindly say "No more spending, we cap health care spending at its current rate forever." We have to do that -- we go bankrupt as a nation otherwise -- but we have to be smart about it. We can cap spending at the current rate, but if we make the system more efficient we can get more medical innovation without compromising care.

The problem is, how do we do that? Well, here's how NOT to do it:

1. Tort reform: According to the CBO, tort insurance premiums for the last year they had full data for (2004) was $6.7 billion dollars. This is less than 0.3% of healthcare spending. And the CBO also casts a skeptical look on the claim that "defensive medicine" causes extra health care spending, noting that all such claims are based on one medical procedure in a hospital setting and there's no evidence that this happens in other settings, and noting that strict tort limits have had minimal effect upon healthcare spending between states with them and states that don't have them, finding for example that a tort limit makes only 1.4% difference at most in total health care spending costs. Indeed, California has had strict tort limits on malpractice claims since 1975, yet is still one of the most expensive states to receive medical treatment in. Frankly, you'd get as much savings by regulating doctors' purchases of stethoscopes and tongue depressors as you'd get with tort reform -- it's that trivial an expense for the healthcare system as a whole.
2. Make all healthcare out-of-pocket free market: Problem is, 16% of the nation's GDP is tied up in healthcare, but sick people don't make 16% of the nation's GDP. This would cause a significant decrease in health care spending by making any treatment more advanced than those available in 1960 too expensive for individuals other than millionaires to purchase, and would kill all medical advancement since there'd be no way to amortize the costs of development of new expensive medical treatments.
3. Prevention. We have ample evidence that prevention makes no real difference in overall healthcare spending -- fat people for example die earlier (at average of age 70), meaning they never get the expensive diseases that health people get in their 80's. We have studies on this -- see this one on obesity, and this one on smoking, for example, which prove that fatties and smokers actually save us money by croaking sooner. That said, we should decidedly fund prevention, but because it makes us healthier -- not because it saves any money.
4. Eliminate all private health insurance. Private health insurance accounts for only 35% of all healthcare spending in the USA. Of that, only 20% is overhead and waste (private insurers are currently running an 80% payout rate, i.e. 80c of every premium dollar goes towards healthcare). So we could save a maximum of 5% of total US healthcare costs by eliminating private health insurance, or reduce total spending from 17% of GDP to 16.15% of GDP. Which is useful, but this isn't going to solve the problem altogether, continuing inflation of provider costs will wipe out that savings almost immediately. Note that eliminating private health insurance isn't absolutely necessary to reduce provider administrative costs -- see below.
5. Create a Medical Costs Board to set reimbursement rates for providers and premium rates for health insurers, without a must-issue/cannot-cancel mandate on insurers: Providers and insurers would simply game the system. Insurers would kick sick people out of their insurance pools and insure only healthy people, and providers would simply shift costs and game the system by prescribing more treatments to get more money. The net result of this is 40% uninsured like in Taiwan in 1994 -- the least-healthy 40% would simply get gamed right out of the system.
6. Create a must-issue/cannot cancel mandate on insurers to avoid the above problem, without a must-buy mandate on individuals: Individuals simply wouldn't purchase insurance until they got sick, resulting in an insurance company death spiral as only sick people bought insurance -- and remember, sick people don't make 16% of national GDP.

So how do we control costs? Looking at what other countries have done, here's the deal:

1. All people must be in the system, via mandates on both individuals and on insurance pools (whether public or private is irrelevant for this item). Otherwise insurers or providers will respond to cost controls by kicking people out of the system, much as, e.g., state Medicaid boards are responding to cost problems by reducing Medicaid eligibility. Universal coverage is a prerequisite to *any* effective control of costs, otherwise the system responds by kicking people out.
2. Reduce administrative overhead. It is estimated that as much as 20% of the cost of the current system is administrative overhead caused by our multi-payer system. We could eliminate 97% of these costs by going to single-payer Medicare For All, but even if we retain the multi-payer system administrative overhead for providers can be reduced significantly. Doctors and hospitals should have *one* set of forms and procedures to follow for all claims, perhaps by the Federal Government setting up central claims-handling dispatch centers similar to the way they set up central check processing centers to dispatch checks quickly to their issuing banks. This center should also handle pre-approvals in a consistent manner via maintaining a map of diseases to benefits to approved treatments and providing ombudsmen to handle appeals in a consistent and timely manner. This would let doctors go back to being doctors, instead of insurance company lawyers.
3. Reduce incentives for over-treatment. Doctors should not be allowed to send patients to hospitals and diagnostic testing centers that they are part-owners of, doctors and hospitals should not be allowed to make a profit on medications and diagnostic testing that they prescribe (which gives incentive to over-medicate and over-test), and doctors' fees for procedures should be capped at a maximum per-patient level so that anything above that level, he gets only his expenses for the procedure -- no additional profit.
4. A ban on consumer-oriented drug advertising. We have plentiful evidence now that drug advertising leads to overtreatment. We should go back to the way it was when I was a child, when drug advertising was aimed at doctors and consisted of a brochure and a box with samples of the new drug.
5. A Medical Payments Board negotiates pricing and maximum reimbursement rates with all hospitals, drug makers, and providers (or their industry groups thereof) in order to allow for some margin but nobody gets rich off the illnesses of others. Medicine should be a calling, not a way to become a millionaire. Go into banking if you want to be a millionaire.
6. Increased incentives for people to go into primary medicine so that we have more primary care providers, without the crippling debt that is currently keeping people from going into primary care or causing them to overprescribe and game the system for the money to pay their outrageous medical school debt.

The experience of Taiwan, which covered 40% of their population with $0 increase in medical costs when they went single-payer, shows that we can squeeze at least 40% increased efficiency out of the current U.S. system without affecting the quality of care. That will take us at least until the end of the next decade before healthcare treatment innovation costs eat up that savings and we have to start imposing hard caps and start making hard decisions that will affect the pace of medical innovation or even cause that dreaded R-word (rationing) to rear its head. Most of these changes would be much much easier with a single-payer system -- otherwise we end up adding another layer between providers and insurers to bring providers' administrative costs down to a reasonable level -- but as a minimal first step, we need to bring everybody into the system. HR3200, the current AMA-endorsed proposal in the House of Representatives, is not going to solve this costs problem, but it does take a significant first step -- it brings everybody into the system and gives everybody, not just sick people, a reason to care about the next step of bringing costs under control by increasing the efficiency of the system.

-- Badtux the Healthcare Penguin

* The first kidney dialysis happened in March 1960, but it was an experimental treatment available in only a few locations.
** My thanks to those I've had Twiscussions with on Twitter for helping me dig up a lot of the data and formulate a lot of the conclusions mentioned above