What went wrong with U.S. health care cost control?

by on December 14, 2011 at 5:27 am in Medicine | Permalink

So if the real problem with U.S. health spending is that the U.S. diverged from its peer countries for a decade-long stretch, solving that problem isn’t quite as simple as mimicking the institutions and strategies of our peer countries, whether it’s Canada’s single-payer system or the hybrid models of France or Germany. Our peer countries are facing the same challenges we are, albeit with slightly more breathing room.

This raises the question of what exactly changed in the 1980s. Daeho Kim, a graduate student at Brown University, offers a provocative hypothesis in a new working paper. As Kim explains, a 1983 Medicare reform created the prospective payment system, or PPS, which offered fixed reimbursements for the use of a medical technology. If a physician decides to use bypass surgery as a cardiac treatment, she won’t be paid on the basis of what it cost her to perform the surgery. Instead, she’ll be paid the national average cost. This way, there is a strong incentive to beat the national average cost of performing bypass surgeries, thus lowering, in theory, systemwide costs.

But something quite different seems to have happened. A big part of the story is that providers can choose from a number of different cardiac treatments, some of which are more expensive than others. PPS encouraged them to focus on the treatments where the marginal cost — the cost of providing one more treatment, in this case — fell below the average cost, even if there are more cost-effective treatments available. Kim suggests that PPS may account for one of the most distinctive aspects of the U.S. health system — our extraordinarily overreliance on costly treatments. If Kim is right, it is the failure of bureaucratic price-setting, not the failure of market competition, that may have supercharged health inflation in the 1980s and beyond.

That is from Reiham Salam.  I will read through the Kim paper carefully and perhaps report back on it.  This is an important topic.

Pavel December 14, 2011 at 7:57 am

Looking at the graph, something did happen in 1983: The second derivative became the same in USA as in comparable countries so I would say that 1983 reform was effective

k December 14, 2011 at 8:38 am

I can’t see the point for the US during the 1984-1990 period, if you notice it’s just a straight light from 78-84 to 90-95.

Rahul December 14, 2011 at 8:57 am

Something’s wrong with that plot. All the other raw per-capita expenditure plots seem to be getting steeper with time. i.e. the first derivative (spending growth) ought to be rising not reducing.

e.g.
http://sanluisvalleyhealth.org/wp-content/uploads/2011/07/health-care-expenditures-per-capita-historical.jpg?9d7bd4

What gives? I don’t get it.

Cliff December 14, 2011 at 9:01 am

Won’t a level growth rate result in a climbing first derivative?

NAME REDACTED December 14, 2011 at 9:28 am

Dollar growth versus percent growth.

Jeremy H. December 14, 2011 at 10:02 am

The data source in unclear on the chart you linked to, since OECD data shows US spending in 2008 at $7720 (not $3500) and e.g., Canada at $4024 (not ~$2900). [Source: http://www.oecd.org/document/16/0,3343,en_2649_34631_2085200_1_1_1_1,00.html

In any event, the confusion likely comes from plotting a linear rather than log scale. If you plot the per capita expenditures from the OECD on a log scale, you will indeed see that the growth rate has slowed down, though not as much as other countries.

Rahul December 14, 2011 at 8:42 am

>>> that may have supercharged health inflation in the 1980s and beyond. <<<

Isn't health inflation actually lower after 1980s than before? If growth in costs reduced from 12.5% to 5% how is that “supercharged”.

I am a bit puzzled by the graph and the “worsening” hypothesis; whatever the healthcare spending situation is; it looks like it has improved post 1983. I know that can’t be true but the plot puzzles me. Maybe one of the shortcomings of this kind of plot is it only shows changes; the important thing to know would be what was the baseline spending of uSA versus peers at the point the time series starts.

Jeremy H. December 14, 2011 at 11:24 am

*All* consumer price inflation is lower since the early 1980s.

And thanks for posting the good graphs below!

mpowell December 14, 2011 at 2:52 pm

If you are telling me that these are not the inflation adjusted numbers, this graph is completely worthless, since inflation dropped significantly from the early 80s to the late 90s.

Jeremy H. December 15, 2011 at 12:43 pm

Yes, I am almost positive that these are NOT inflation adjusted. But the graph is not completely worthless, since the comparison with other countries is the relevant point (even if this graph makes it poor).

Rahul December 15, 2011 at 1:51 pm

Here’s how the graph ought to look (I might be wrong!) after inflation adjustment and some smoothing:

http://bit.ly/CPI_Adjusted_Healthcare_Spending

The spike in spending of the early 80′s is merely an artifact of inflation. The real spending growth has been steadily slowing down since the 60′s.

Here’s the various healthcare plots together:

http://bit.ly/Healthcare_Plots

Bill December 14, 2011 at 9:17 am

As others have pointed out, a “rate of growth graph” fails to note the starting point in real expenditures, and is only a derrivative.

So, we should be “happy” that the rate of growth off of a high base is not accelerating? Real expenditures are still growing at a higher rate than the rate of inflation.

Andrew' December 14, 2011 at 11:29 am

Simply this, they aren’t controlling costs either.

Maxxxx December 14, 2011 at 8:08 am

American politics likes to reduce problems to a single factor but I think here just a lot of factors that play a role here. Just a few I have noticed:
- Drug advertising to consumers
- Excessive malpractice lawsuits
- No wide-spread availability of effectiveness statistics for treatments. Quality numbers of doctors aren’t available either. How is the consumer who is supposed to be market participant to choose a treatment or a doctor?
- Doctors and hospitals seem to make up prices for treatments based on the time of day, the weather or whatever. Prices should be published. Why do insurances know exact prices for treatments but self-insured people or people with HSA can’t even get a straight price quote?
- Big one: Employer-based health insurance. This makes no sense at all. There should be one country-wide pool for all insured people. It makes no sense that people get disqualified for preexisting conditions but as soon they get hired by a certain employer they get insured. This also hurts mobility of workers and makes the economy stagnant. For example it’s difficult to start a business after 40 because you will have trouble finding health insurance.
- Too many separate public programs. Why do children, seniors, veterans and poor people need separate programs? They should all participate in the insurance market like everybody else with probably some subsidies.

One could probably come up with many more. I am sure if there were efforts to address some of them the health system wouild be more affordbale and would ala

DKF December 14, 2011 at 12:05 pm

I believe many of the items you bring up first good points, but are actually symptoms as opposed to root causes. For me, the big themes to the high cost structure, in order are: (1) 3rd Party Payer, (2) Medicare Fee-for-Service Structure, (3) Tax Code [enables (1), because it encourages employer coverage] and (4) Excessive Tort Judgments.
I believe that’s the real tragedy of PPACA…it reinforces (1), applies a central rationing board [IPAB] in an attempt to regulate (2), addresses (3) only tangentially [delayed "luxury tax" on certain high-value plans], and only applies window-dressing to (4) [pilot programs].

mpowell December 14, 2011 at 2:58 pm

I would be shocked if anyone who actually had any significant hospital experience thought that 3rd party payer was a bad idea. Do you have any idea how complicated the delivery of medical care is? Or how time-sensitive it can be? Expecting consumer to make health care purchasing decisions by comparing cost and quality of a variety of options is insane given the vast array of possible arrangements few of which even a significantly above average consumer is likely to be able to effectively process. It’s not like buying a new car.

DKF December 14, 2011 at 3:49 pm

Pretty good idea of how complicated the delivery of medical care is, though my experience is mostly secondhand. Aware that it can be time-sensitive; but we’re not talking about emergency care, but rather about the vast majority of costs attributable to routine procedures and the management of chronic conditions.
And btw, I should have specified “first dollar 3rd party payer”. Most individuals obviously need insurance to cover catastrophic events. But the idea that 3rd party, first-dollar coverage encourages overutilization and waste is not exactly controversial–and individuals make health care decisions every day–overtly and through lifestyle decisions. They also choose options in the health-care plans they get through employer coverage or their own purchases. Yes, we have “unschooled” individuals making all sorts of healthcare decisions, every day…”insane” isn’t it?

mulp December 15, 2011 at 12:07 am

So, you are arguing the reason health care costs are so much lower in the other OECD nations like Japan, Canada, France, Germany, Israel, et al is the government is not involved and patients pay all medical costs out of pocket while the US is based entirely on third party payments???

Or are you arguing that when Europeans and Asians have children in the US, they suddenly become very different health wise and require twice the wasted spending in the US than they required in Europe and Asia??

Other than the universal coverage of everyone and greater government oversight in the other OECD nations, why does health care in the US far exceed those other nations as a share of a larger GDP while the aggragate health outcomes in the US are measurably worse? (Maybe you believe most other nations have longer lifespans because requests to die are placed on a two year waiting list while in the US you are quickly ushered to your death??)

All your reasons for the high cost in the US must be tested against all the other OECD nations and found predictive in most, and you easily have dozens of nations to test your hypothesis with.

zbicyclist December 14, 2011 at 5:34 pm

I’m guessing you don’t remember doctors’ and dentists’ offices in the 1950s, which usually had common prices posted. And, upon entry to the hospital for my emergency appendectomy in 1964, my mother had to make a decision on a 4 person room for one specific price, or a semiprivate room for another specific price.

It’s not that hard to provide prices, since all this stuff ends up with insurance codes anyway. And just because the system would be hard to make perfect doesn’t mean it can’t be made pretty good.

mulp December 15, 2011 at 12:15 am

The reason air travel is so much cheaper (relatively speaking) today than before 1980 is they began posting all the airfares on the wall in 1980?

(I believe airfares were posted “on the wall” until the price tables became too complicated for all the thousands of combinations of origins and destinations, and only in the 70s were fares changed often in the official CAB price schedule, as competition was allowed by rubber stamping most fare proposals if matching a competitor fare.)

JWatts December 15, 2011 at 4:02 am

“The reason air travel is so much cheaper (relatively speaking) today than before 1980 is they began posting all the airfares on the wall in 1980?”

Well since the mid 90′s at least, yes cheaper posting on the ‘wall’ is a significant factor in price decline. Airfares are cheaper because it’s trivial to check the cost of a flight on the internet. Some part of the price drop is just removing commissions from booking agents but the price drops are larger than that savings alone.

mulp December 15, 2011 at 5:38 pm

But the “price on the wall” is constantly changing and is different for different people. The price I see sitting shoulder to shoulder with you for the same trip is probably going to be different, just like the price you pay to to see my doctor will be different from the price I pay.

The websites incorporate such things as browser cookies tracking searches to figure out what price to post for you based on the success of offering lower prices only for the most determined low price shoppers, knowing you will thing the price they give you is THE “price on the wall” when it is merely your custom prices designed to take as much from your pocket as possible.

Dave Anthony December 16, 2011 at 12:06 pm

mulp you seem to have a very either/or approach to this problem. The explanation for why our costs in the United States are higher doesn’t mean that the costs in Europe are lower because of the opposite. They have chosen to manage a lot of their costs through a central bureaucracy.

Just because they have lowered costs through that method DOES not mean that the ONLY way to lower costs is to follow their example.

I’m not sure why you brought up the airline industry in this — prices decreased dramatically after 1980 because the airline industry was deregulated during the Carter adminsitration, leading to increased competition from new entrants to the market (whereas previously the big 4 or 5 airliners had a steady oligopoly for decades).

Ryan December 14, 2011 at 8:18 am

I must again reiterate that these kinds of comparisons are misleading because of the nature of the healthcare industry. US profits subsidize profits abroad.

What I mean is, if a company can broach the US market, they can then make enough money in the US system to enter foreign markets. If they don’t make it in the US market, it is doubtful that entering other markets will prove to be economically worthwhile.

For this reason, I say over and over again that the US healthcare system is subsidizing the more “experimental” healthcare systems of the rest of the world. If we bring the US system in line with what exists elsewhere, the whole healthcare industry will either collapse or become fully socialized. There is no other option.

NAME REDACTED December 14, 2011 at 9:28 am

+1

Bill December 14, 2011 at 10:35 am

Oh, please. If we want to subsidize an industry, we should do it directly, rather than indirectly. A CEO can profit any of the rent you give him, whereas if you want to contract with a university, or even a firm, to perform research, you can negotiate deliverables and maybe even get a payment back for what has been developed.

Secondly, your argument works in reverse: a foreign firm, which developed the product abroad, can free ride on a non-competitive healthcare system and ship the money back to England or Switzerland.

Ryan December 14, 2011 at 1:07 pm

No, Bill. Every company only has one R&D department. Pharma companies don’t engage in nation-specific R&D unless the regulatory regime demands that they “demonstrate comparative effectiveness locally.” Long story short, Pfizer develops a product in one location (or purchases a patent) and then applies for regulatory approval and formulary listing on a country-by-country basis. They start in the largest market (USA), and then proceed outward.

A product/service may only be worth $10 million in the Canadian market, whereas it may be worth $100 million in the US market. Given that the cost of preparing regulatory submissions is roughly the same in all countries, this makes a HUGE DIFFERENCE when it comes to the company’s strategy. Why undertake all that effort in the Canadian marketplace for $10 million (a drop in the pharma bucket), when that money would be better spent in the e.g. Japanese marketplace where it might be worth $50 million?

Rahul December 14, 2011 at 1:18 pm

Isn’t what you are describing simply “price discrimination”? And isn’t the libertarian viewpoint on price discrimination that it tends to be welfare maximizing?

I guess the operative question is whose welfare; humanity or Americans?

Ryan December 14, 2011 at 2:37 pm

Your statement implies that regulatory burden is a “price,” which is patently absurd.

Rahul December 14, 2011 at 2:56 pm

@Ryan

Companies sell medicines cheaper in (say) Asia, Africa etc. not because it is altruistic but simply because it is a profit maximizing strategy and makes commercial sense.

BTW, if the cheaper prices in other “wealthy” countries are objectionable one practical option would be allowing import of drugs sourced from other (western?) countries. Arbitrage should kill price variation. The “drug safety” issues will be raised by the pharma lobby but are quite a red herring. One can require the importer to perform random quality tests or proof of provenance and thus ensure quality compliance.

Ryan December 14, 2011 at 3:04 pm

Rahul, you might be getting me wrong. I’m not faulting the companies, I’m faulting the regimes.

Richard December 15, 2011 at 5:23 pm

Employing arbitrage in Rahul’s way would cause medicines to converge on a single price – one that the rich world can pay. Much like how oil prices are the same across the world (minus taxes and subsidies by national governments), you would make it impossible for a poor African child to obtain AIDS medication.

Rahul December 16, 2011 at 7:43 am

@Richard

Fine. Just allow imports from “western” nations.

Bill December 14, 2011 at 2:12 pm

Ryan:

If we are stupid and do not want to act like a wise consumer of resources your statement is true: “A product/service may only be worth $10 million in the Canadian market, whereas it may be worth $100 million in the US market.”

It’s not a “market” if the consumer, through intermediaries who have no interest in cost minimization, make the purchase and simply pass on the cost to someone else (e.g., the government) which is barred from negotiating with the supplier.

Ryan December 14, 2011 at 2:36 pm

If you recognize that, then I have no idea why you’re disagreeing with me. The government is the intermediary in non-US countries. We are turning the government into the US intermediary. This is my whole point.

TallDave December 14, 2011 at 2:24 pm

Exactly. Nature actually had a study on this — it’s why the U.S. gets new treatments on average 18 months earlier than Europe. Those treatments are developed for the U.S. market, and they get the leftovers at the marginal cost. It’s like a trillion-dollar welfare program for the rest of the OECD at the U.S.’s expense.

Unfortunately, since they outvote us in the WTO, we can’t prevent them from stealing from us this way, and the only way to avoid it would be to cut off our nose to spite our face — shut down R&D by adopting similar policies ourselves.

Ryan December 14, 2011 at 2:44 pm

Yeah. My view is that it’s in the US’s best interest to just let everyone leech off of our free market. It may be annoying, but it’s preferable to a collapse of the health care industry. New treatments are getting progressively more expensive to develop; if we play around with this too much, we won’t get any new treatments at all.

Anyone with a chronic or potentially terminal illness knows how scary that is.

mjw149 December 14, 2011 at 3:36 pm

Pfizer made an $8bil profit this quarter. So we’re not subsidizing research, we’re subsidizing their executives. That’s not effective as a research strategy. Should we look back at history and see how ‘unproductive’ less profitable drug companies were in prior decades? Should we consider that the people doing the actual work in these industries aren’t being paid more over time and are still motivated and educated like prior decades when there were smaller profits? Is it too obvious that profits aren’t the relevant metric of a healthcare system, or is it only obvious to less greedy people?

TallDave December 14, 2011 at 11:01 am

Yep, other countries only cover the marginal costs of treatment. R&D is mostly paid for here.

Once again saving European lives at our own expense thanks to their fascination with leftism. It’s a recurring theme.

NAME REDACTED December 14, 2011 at 11:14 am

Yep, spot on Dave.

Bill December 14, 2011 at 12:01 pm

Dave, It is not a fascination with leftism. Europe is the beneficiary, and you would not call that rightist.
Rather, it is the right wing in this country that blocks hard bargaining with medical device and drug companies.
What is your position with the government, as the payor, negotiating with the drug companies and device manufacturers directly.

DKF December 14, 2011 at 12:25 pm

You appear to have a great deal of faith in the government’s ability to bargain for greater benefits at lower costs…my experience doesn’t bear that out (e.g. defense, education). Government is not the payer, only the conduit…a better plan is the “premium support” concepts under consideration, so that individuals have skin in the game to negotiate on their own behalf…millions of individual decisions, better than one central controller with weak incentives and conflicted interests.

FooFighter December 14, 2011 at 1:28 pm

The problem with that is that the individual has no bargaining power compared to the collective.

ie it is super easy to price discriminate when dealing with individuals, eating up consumer surplus and further exacerbating the problem of high-paying Americans ‘subsidizing’ low paying individuals.

Bill December 14, 2011 at 2:14 pm

DKF, If you believe that the government doesn’t have the capacity to bargain for greater benefits at lower costs…why prohibit it from doing so. It has more control as a purchaser of weapons systems than as a purchaser of aspirin.

Bill December 14, 2011 at 2:16 pm

DKF, And, by the way, the government does and is permitted to purchase medicine and devices–but only for the VA.

DKF December 14, 2011 at 3:26 pm

@FooFighter…the market *is* the collective (as opposed to the govt)…but we don’t even need to go there…just look at medical procedures that individuals buy (and pay cash for) compared to medical procedures in which the govt is involved. 3 example? Cosmetic Surgery, Certain dental procedures (braces, cosmetic, etc) and Lasik. Transparency, innovation and lower prices prevail…what if all healthcare was that way?
@Bill…I don’t have anything against medicare trying to lower (my) costs…but to a large extent, it already does that through the cost-adjustment mechanism of covered procedures, and the process by which it decides which procedures to cover. That hasn’t worked very well, and I’m skeptical that some sort of new “bargaining” activity at, say, the device level will work any better. It doesn’t address the roots of the problem: 3rd party payer, fee-for-service schedule.

mjw149 December 14, 2011 at 3:41 pm

Yet Medicare Part D has worse prices with a market mechanism that Medicaid and other Medicare. So the gov’t is quite effective, read the report they issued this year. You have too much faith in greed and what are obviously broken market mechanisms. If capitalism worked so well, why do we have laws? Why does nearly every nation provide public health care at all? Why did America have such a better economy with high taxes and big gov’t in previous decades?

Individuals have no negotiating power with healthcare and never will. Insurers are incented to provide worse coverage. That’s why an expert controlled gov’t system has usually provided better results. The market forces are completely unbalanced in healthcare and probably always will be.

DKF December 14, 2011 at 5:26 pm

@mjw149…are you trying to set a record for straw men? The market mechanisms *are* broken…not through the presence of self-interested individuals, but by the presence of govt >50% of the health care dollar. Why have laws? We have laws to enforce contracts and ensure individual rights (things that free markets don’t do). If you wondering about the past apparent health of the economy compared to now…might I suggest a look at “The Great Stagnation”?
As to the Medicare Part D “study”, the ability of Medicaid to shift costs onto Medicare and private insurers through the rebate process is hardly evidence of the superiority of gov’t mandate over a well-functioning market. Rather, it’s evidence of a market that is not functioning well. But, if you want to push medicaid rebates through part D, might want to take a look at this: http://bit.ly/qjN06B.

FooFighter December 14, 2011 at 6:27 pm

You just named three things that are purely elective. Of course they’ll have lower costs – the demand curve is super elastic.

TallDave December 14, 2011 at 2:18 pm

Ah, the eternal confusion of leftism. Why do you think companies are going to do R&D if gov’t mandates that no one pay for it?

mpowell December 14, 2011 at 3:04 pm

There are a lot of concerns as to how much, if any benefit, accrue to American healthcare consumers from all this private R&D. The basic approach is, take a known drug, spin it to get a different set of side effects, bribe and influence doctors to prescribe it, make money off your new patented drug. The assumption that the cost to the American public bears any reasonable relationship with their healthcare benefit is questionable at best. There are alternative models, but I’m betting you work from the assumption that they must give worse results.

mjw149 December 14, 2011 at 3:44 pm

Universities make most of the discoveries, pharm companies will always have incentives to commercialize new medicine and treatments. The pharmaceuticals aren’t spending enough on R&D, they’re doing billion dollar stock buybacks. And those new medicines and treatments are useless if they aren’t affordable for most people, which the pharmaceuticals are incentivized to prevent. Which makes their private R&D counterproductive. Public R&D would be better. Those researchers and scientists aren’t making the profits and largely aren’t motivated by large amounts of money. R&D isn’t more effective because executives at Pfizer get millions in compensation, it’s worse.

TallDave December 15, 2011 at 3:02 pm

pharm companies will always have incentives to commercialize new medicine and treatments.

Not if you explicitly eliminate them.

Public R&D would be better.

Gov’ts do not productize well.

TallDave December 15, 2011 at 3:03 pm

There are a lot of concerns as to how much, if any benefit, accrue to American healthcare consumers from all this private R&D.

Yes, there are bad incentives in the system, but that doesn’t argue for effectively ending medical innovation.

Dan Dostal December 14, 2011 at 4:53 pm

Leftism? Don’t you mean liberalism? Shouldn’t libertarians prefer the ease which other nations can gain access to American produced drugs? Is this an issue with generics competing with producers? Isn’t that another point for libertarians? Your statement comes off as reactionary nationalism, no sensible economics.

DKF December 14, 2011 at 11:47 am

Yes, I am constantly repeating this also. The foreign systems that some hold as a model for the US cannot be evaluated as if they are independent of actions (and expenditures) which arise from the US system.

NAME REDACTED December 14, 2011 at 8:18 am

Pavel has a good counteragument. What would be the response to him?

Scott Lange December 14, 2011 at 9:04 am

I’m always suspicious when someone proposes a counter-intuitive theory and supports it with a strangely put together chart. What are the ten other countries in this chart? Why on earth are the growth rates averaged across these weird six-year intervals? What were the growth rates before 1972 (the commentary says they “diverged,” but since the graph suddenly stops at 1972 when they come together, we can’t see whether 1972 was an aberration)?

Douglas Knight December 14, 2011 at 9:43 am

Yes, bucketing is evil. If the data is noisy, smooth it over every 6-year interval, not arbitrary 6-year buckets.

Rahul December 14, 2011 at 12:24 pm

Here’s 6 year smoothing applied to the annual spending growth data:

http://bit.ly/6_year_smoothed_trend_of_spending_growth

Douglas Knight December 14, 2011 at 4:13 pm

Thanks!

JWatts December 15, 2011 at 4:10 am

+1

Andrew December 14, 2011 at 10:53 am

Agree. I think this chart is strange. And surely the starting levels matter?

Very suspicious.

albert magnus December 14, 2011 at 9:18 am

My theory is that the periods of high cost growth correspond to the adoption of MRI, Computer Tomography and Conformal Radiotherapy. In other words, it corresponds to the rise of expensive computers and experts that need to run them.

anon December 14, 2011 at 10:01 am

“Medicine Needs Frugal Innovation”
http://www.technologyreview.com/business/39216/?p1=BI

From the article:

“The up-front cost of the pocket ultrasound device is about $7,700, but there is no extra cost for an unlimited number of readings.

That makes these small devices a formidable challenge to business as usual in American health care. Each year in the United States more than 20 million echocardiograms (ultrasounds of the heart) are performed, and so are a similar number of abdominal and fetal ultrasound examinations. Each of these diagnostic procedures is done in a dedicated laboratory setting, either in the hospital or in a doctor’s office, with expensive equipment—and a combined professional and technical charge of $1,000 to $2,000. The math is straightforward. If a pocket ultrasound device were incorporated into routine physical exams the same way we use a stethoscope, several billion dollars in unnecessary charges would be saved each year.”

mulp December 15, 2011 at 5:52 pm

But those high costs for tech translated into high costs only in the US. In the rest of the OECD, the pricing policies applied by either the government or the legally sanctioned industry price boards drove down the price of the high tech, and at the same time took the market share from the US and gave it to Japanese, German, et al innovators.

The costs of CT scans in Japan fell to 5% of their initial costs by simple price theory: if decision makers are billing for scans more and more, the price paid is too high, so cut the price. This led in to the virtuous cycle of Japanese scanner designers cutting both capital and operating costs to restore high profits to their bottom lines by increasing profits at the providers, leading to further price cuts for CT scans. Japan has more scanners per capita than the US because in Japan they buy cheaper scanners and do many more scans for very low prices because the scanners are easier to operate, faster, and fit in regular office spaces.

The US medical system is based too often on buying the Lamborghini while Japan buys the Accord and Germany the Beetle.

Andrew' December 14, 2011 at 9:18 am

Do people just get bored making straight-forward charts?

Rahul December 14, 2011 at 9:43 am

Agreed.

For whoever might care, try this (better) plot of raw data:

http://bit.ly/healthcare_plot

The plot has Raw $ spending (PPP adjusted unfortunately), country names are specified, no data massaging, per year data without arbitrary 6-year aggregation.

So, now we can tell that circa. 1980 usa was at $1000 and the rest of the countries (Germany, Canada, Australia, uk and France) in the $500-$1000 range. Today America’s at $6000 while the rest of them are $2000-$3000.

Conclude what you will, but at least based on legible data.

The Engineer December 14, 2011 at 9:23 am

Could another answer be that the US actually had economic growth in the 1980s, and Europe did not? As we have seen during the Great Recession, there is nothing better than slow economic growth to kill inflation in just about every sector.

mulp December 15, 2011 at 5:58 pm

Health insurance premiums were going up for employers at 8-10% annually for the decade before Obama took office. Was inflation all that high?

And my insurance premium fell 10% this year because of the medical loss ratio rule requiring something like 80% of premiums to go to health, not overhead, marketing, and profit.

spencer December 14, 2011 at 9:36 am

I suspect that the break in trend around 1980 would be due to the overall shift from accelerating inflation and wage growth to a disinflationary. macro environment.

If you look at a long run chart of nominal US medical spending there is a clear break around 1980 in trend that coincides exactly with the overall inflation rate.

Maybe the main reason medical spending has grown faster than NGDP in recent years has been the slow growth in nominal GDP, not particularly strong growth in medical spending.

charlie December 14, 2011 at 9:48 am

what percent of Europeans wear contact lenses?

How many British people have good teeth?

Does your dog have an insurance plan?

75% of your medical expenses are in the last six months of life.

Douglas Knight December 14, 2011 at 10:00 am
Rahul December 14, 2011 at 11:10 am

Based on your (and Jeremey H above) OECD data I plotted the per-capita spending time series for usa, uk, Canada and Germany from 1960-2010

http://bit.ly/Health_spending_per_capita

This gives a much better insight IMHO than the arbitrarily bucketed growth plot.

Jeremy H. December 14, 2011 at 11:34 am

Thanks for posting this!

Another point to keep in mind: health care spending tends to rise (as % of GDP) as GDP/capita increases. GDP/capita in 2008 in the UK, Canada, and Germany was roughly equal to US GDP/capita in 1993, 1996, and 1985 (Maddison data). So another relevant comparison is their spending (as % of GDP) in 2008 versus US spending in those earlier years. US spending was still higher (except in Germany, roughly equal), but not nearly as much as today (OECD Data). Life expectancy in the US was also shorter in those years, so the comparison is imperfect but still useful.

Rahul December 14, 2011 at 11:54 am

Here’s another plot of the annual percent growth of per capita American healthcare spending from 1960 to 2010. None of that arbitrary 6-year data binning.

http://bit.ly/Annual_Spending_Growth_1960to2010

The data is richer than the MR plot seemed to show. Sure there’s a downward trend over the decades but I’m skeptical how influential that 1983 Medicare event is. The downward trend has started pre-1983. Interestingly there seems to be a huge year-to-year variation. Not sure why. The data’s pretty noisy (why?).

Jeremy H. December 14, 2011 at 12:06 pm

The downward trend that started after 1980 is a *general* price trend, not just health care spending. Those “annual spending” OECD are not inflation adjusted. That’s why the cross-country comparison is more relevant. None of what I said proves the 1983 Medicare point, but I think you are missing the relevance of the graph Tyler posted (as poor as the graph is).

http://3.bp.blogspot.com/_nSTO-vZpSgc/THP9Qhj3yJI/AAAAAAAAJNQ/nLkieSjLNtc/s1600/CPI+1960-Present.png

Rahul December 14, 2011 at 1:00 pm

I see what you mean. Nominal spending numbers need some CPI adjusting; I’m just too lazy to do that. :)

Jeremy H. December 14, 2011 at 4:35 pm

Right! Plus you need a different “CPI” for each country, Worth doing at some point, but probably not this deep in a blog comment thread ;-)

Rahul December 14, 2011 at 4:59 pm

At some point though doesn’t this all become very circular? Don’t medical goods and services constitute a fairly large fraction of the basket used for the CPI computation?

Jeremy H. December 15, 2011 at 9:41 am

No. Medical care is currently just 6.6% of the CPI. http://www.bls.gov/news.release/cpi.t01.htm

mulp December 15, 2011 at 6:08 pm

“Right! Plus you need a different “CPI” for each country,”

OECD data is in PPP which mitigates the difference in CPI, PPP being purchasing power parity, which if in dollars converts all prices to nominal dollars without regard for exchange rates.

OECD also does percent of GDP which eliminates both inflation and differing purchasing power. The US with a higher GDP per capita spends way more on health care as a share of GDP than all the other nations, without anyone demonstrating the health outcomes are better in the US for the huge increased sacrifice on health care spending.

Rahul December 16, 2011 at 4:10 am

When OECD translates foreign currency to $ does it use the PPP of today or of appropriate corresponding historic year?

steve December 14, 2011 at 10:29 am

We have known for a long time about this divergence in the 1980s. If you wish to posit the PPS as cause, you need to figure out why the divergence went away in the 90s.

I think the more plausible explanation is the proliferation of new technology that came on board in that era. Imaging got much better and OR technology got much better. We did a lot more because we could. However, the European countries and Canada had ways built into their system to control costs. To control the proliferation of devices. The US did not, so our spending increased much faster.

There were also predictions in the 80s about the numbers and kinds of physicians we would need. IIRC, there was a prediction that more specialists would be needed, pushing med students toward the more expensive specialties.

Steve

Silas Barta December 14, 2011 at 10:47 am

Ah, when all else fails, use a logarithmic scale or go up one derivative!

TallDave December 14, 2011 at 10:57 am

PPS encouraged them to focus on the treatments where the marginal cost — the cost of providing one more treatment, in this case — fell below the average cost, even if there are more cost-effective treatments available

Great, great point. A classic case of missing the forest for the trees. And it created all kinds of similar incentives on the medical treatment side as well — why develop cheaper versions of existing treatments when the market doesn’t differentiate between them?

KLO December 14, 2011 at 11:37 am

Also, why take large risks developing significantly better treatments when marginal improvements to existing treatments will command top dollar? This seems to me to be the biggest issue. We have bought hundreds of billions of dollars worth of statins to prevent first heart attacks when the data show that they do not reduce all cause mortality. The benefits of statins, even when the side effects are discounted, are marginal to non-existent for most people currently taking them. The statins problem is writ large across treatment for virtually all conditions. Our poor understanding of underlying conditions in combination with our to “do something” causes us to pay enormous sums for ineffective or marginally effective treatments.

TallDave December 14, 2011 at 2:26 pm

Yep, terrible incentives are everywhere.

mulp December 15, 2011 at 6:11 pm

But statins are cheap to make. Except for Lipitor. But now the $14B annual Lipitor tax is getting slashed.

Ed December 14, 2011 at 12:26 pm

Maxx makes an excellent point. I have conservative friends who say that if you get rid of employer provided healthcare you fix the system and I’ve come to agree, though unlike them I still favor keeping the fancy socialist european government provided stuff around at least to ease the transition.

I wonder what society would like like if all compensation for labor had to be paid in cash, within two weeks of the service being rendered, excepting perks directly related to being able to do the job or mandated by occupational health and safety regulations.

Andrew' December 14, 2011 at 12:33 pm

Since insurance is fundamentally pooling, I’d suspect that would directionally shrink the size of corporations.

Maxxxx December 14, 2011 at 1:37 pm

Wouldn’t this be a nice side effect?

Marcos December 14, 2011 at 1:39 pm

With the bad consumer laws of the US, forget it. You are better with employer provided anything.

lxm December 14, 2011 at 2:48 pm

I think it would be great if the PPS was found to be the cause of increased medical costs in America because then there would be something that could be fixed. But it’s a medicare rule and doesn’t apply to all medical costs. And there are so many other obvious issues such as increased administrative costs in doctor’s offices due to multiple insurance company requirements and the overhead in multiple insurance companies rather than single payer.

The billing for medical services is just bizarre. I just got a flu shot. Would have cost me $25 at a local grocery store. My doctor charged $80. The insurance company paid $2.85 and I paid a $10 co-payment.

And the delivery for medical services is pretty bad too. Have you ever had to wait? And wait. And wait. My wife and I were at the doctor for a review after a surgery. We were in his offices for two hours and we saw him for two minutes and listened to him mostly just admire his work. I do not know what the pressures are on physicians but the patient’s demands seem very low on the totem poll.

And you can’t comparison shop either because the medical profession protects doctors.

God help you if you don’t have insurance.

Jeremy H. December 14, 2011 at 4:38 pm

If you paid $10 and insurance paid $2.85, where did the other $67.15 come from? Sounds like the insurance company is doing a good job of containing costs in this case!

And if you think two hours is a long time to wait, well you know the rest…

Careless December 15, 2011 at 2:48 am

the $80 was a made-up number the doctor uses so the insurance has a large amount it can bargain down from. Go in without insurance and they won’t charge you that (if you make sure to tell them you’re not covered)

Jeremy H. December 15, 2011 at 9:42 am

While the billing may be “bizarre,” your anecdote does not demonstrate that insurance is driving higher costs.

lxm December 15, 2011 at 4:22 pm

Never said insurance is driving costs higher. What I said is that billing and medical charges are bizarre no matter who pays.

I also implied that doctors face inadequate competition.

And God help you with if you have no medical insurance and have to compete in the market with out a big hammer.

Steven J. Davidson, MD, MBA December 14, 2011 at 5:33 pm

Eli Ginsberg, an economist at Columbia who focused on healthcare, has pointed out that mean physician incomes rose by ~50% between 1965 and 1975 during the first decade of medicare and medicaid. As a teenager at the beginning of that period and a newly graduated from medical school intern at the end of that period I clearly recall the change in lifestyle that income begat for many physicians of my acquaintance and relation. When PPS was introduced and I was in practice, I remember the apprehension we hospital based practitioners experienced and what a non-event it turned out to be. I clearly recall that 1984 budgets were generous because our hospital did so well under PPS. Anecdotal to be sure, but I assert that the experience of 1965-1975 (and beyond) “primed” physicians and health care delivery organizations generally to learn the rules so as maximize gains. Medicare/Medicaid created an environment where payment for all care was expected and expected to stay even with inflation or better; PPS, a new program was the next opportunity and this time–a nearly a generation (18 years) later–health care delivery organizations and physicians were prepared to maximize their opportunity.

mulp December 15, 2011 at 7:35 pm

The Medicare cost control circa 1997 limited the PPS to inflation plus 1% but the Republican controlled Congress has repeatedly decided that inflation plus 1% was draconian price control and deferred that rule repeatedly so the current prices are 27% higher than inflation plus 1%.

PPS was a shift in direction away from managed care by physicians who opposed HMOs. But after getting Congress to force HMOs to bear the burdens of insurers and operate as insurers instead of prepaid integrated care organizations, insurers came under pressure to control their premium price increases now that they had eliminated the HMOs that limited their prices, so insurers began insurance bureaucrat managed health care dictating to doctors. By the end of the 90s, doctors, patients, hospitals, and insurers so hated that method of cost controls, the insurers just started hiking premiums 8-10% a year, 1998 onward.

Kevin December 16, 2011 at 5:37 pm

Reminds me of this article from The New Yorker a few years back: http://www.newyorker.com/reporting/2009/06/01/090601fa_fact_gawande

Comments on this entry are closed.

Previous post:

Next post: