• What makes the US health care system so expensive – Introduction

    This post is cited as a “must read” in the 14 October 2010 edition of Health Wonk Review.


    After I wrote this post, you all went crazy.  While few of you were arguing with my main point (that obesity isn’t to blame for our high health care costs), you all wanted to know what makes our health care system so expensive.  You all want to know where the money is going.

    The quick answer I give my students is that it is multifactorial.  But I know that’s flippant and that this is an opportunity to do better.  So I am going to.  You deserve it.

    I will be endeavoring this week (and further) to give you better answers.  I’m going to draw heavily on the McKinsey & Company paper I have referenced in the past, supplementing that when appropriate.  I’m also basing much of this on an important point made by Uwe Reinhardt and colleagues in their paper U.S. Health Care Spending In An International Context: Why is U.S. spending so high, and can we afford it?:

    No single factor explains the levels or rates of increase in health spending among industrialized countries. However, ability to pay, as measured by GDP per capita, has repeatedly been shown to be one of the most important factors. About 90 percent of the observed cross-national variation in health spending across the OECD countries in 2001 can be explained simply by GDP per capita.

    In other words, the vast majority of health care spending can be explained by wealth.  As countries become wealthier, they spend more on health care, and they do so in a predictable fashion.  At least, most of them do.

    Let’s start here:

    What you are seeing is a chart comparing health care spending to GDP.  No one is making the argument that we should not be spending more on health care than other countries.  We are richer, and it is appropriate that we therefore spend more.  We would expect to get better outcomes* for that extra spending, which would also be appropriate.

    But as you can see, almost every other country lines up on an almost straight line.  There is a very stable relationship between how much money a country has in terms of GDP per capita, and how much it spends on health care.  As countries become wealthier (moving to the right on the x-axis), they spend more on health care (moving up the y-axis).  These countries aren’t working in concert.  Nor are their systems the same (regardless what you’ve heard).  And yet, they all seem to have found the sweet spot.  They all seem to have found the same ratio of wealth to spending.

    This makes sense in its own way.  At some point, you can spend more money and not get the same rate of return.  At some point, you exhaust your resources and you’re wasting money.  It is not inconceivable that most systems find an equilibrium.  That is likely along the line.

    There is one exception – the United States.  We are way above that line.  Moreover, it’s debatable that we are achieving better outcomes than these other countries.  That would suggest that the money we are spending above the line might be, for lack of a better word, waste. Regardless, that money is more than you might expect, based on every other country in the world.  If we explore how that extra money is being spent, we can start to think about where spending might be cut.

    This is the concept that I will follow through this series.  I am going to talk about the money we spend that is more than you’d expect given our wealth.  It is the money that we spend above the line.  It’s the money that we likely could focus on first as something potential to cut.  It’s the money that is most likely to be the least cost-effective.

    Make no mistake, though.  These decisions won’t be popular.  They won’t be easy.  That orange line, the spending above what you’d expect for our wealth, though, accounted for $643 billion in 2006.  It’s more now.  It’s huge.

    If you were expecting a bad guy here, you’re going to be disappointed.  There is no one thing to blame.  There is no one fix we can make.  It really is multifactorial.  But it’s my hope that when we’re done here, you will have a better understanding of why our system costs more, regardless of whether you agree with that spending or not.


    This series will include one post each weekday over the next two weeks. The link in each item below will become live once the post is up.

    1. Introduction — September 20, 2010 (this post)
    2. Inpatient Care — September 21, 2010
    3. Outpatient Care — September 22, 2010
    4. Drugs — September 23, 2010
    5. Administration and Insurance — September 24, 2010
    6. Investment in Health — September 27, 2010
    7. Health Care Workers — September 28, 2010
    8. Areas of Underspending — September 29, 2010
    9. Red Herrings – September 30, 2010
    10. Conclusion — October 1, 2010
    11. Responses to Comments
    12. Responses to Comments, ctd.

    *You will see that I describe our lack of better outcomes a number of times.  While this series focuses on costs, I think a future series will be dedicated to outcomes.

    Comments closed
    • I’ll try to keep reading the series, but given that so much of the wealth in the U.S. is concentrated in an increasingly smaller and smaller group, it doesn’t quite make sense to me that overall wealth tracks with increased health care spending.

      Hopefully this is something you’ll address in your series.

      • Surgeons get very little pay compared to hospitals. The ER will use a multiplier ox 6x on every invoice, in the case that patient doesn’t pay. The financial of non-payment will be declared as tax deduction in hospital state and federal tax filing. Despite declaration of a complete loss, there is still profit for hospital due to invoices that are announced at 6x above cost of production. The 6x factor is actually the 1/x of the going tax rate if anybody is interested. The borderline of multimillion dollar tax fraud. The 16% relation to the GDP is very much due to the way hospitals with ER are allowed to file their taxes.. Doctors and surgeons are good, but nurses are under qualified, under paid, and overworked. Nurses will make all the mistakes the highly trained doctors won’t do. Doctors jump between patients leaving nurses without supervision. The lack of quality is compensated by paying malpractice. In USA you pay $300 for doctor to see you for about 3 minutes. In other countries corrective procedures fixes a poor surgery outcome, but in USA you know that if a hospital messed it up once, they will surely mess up the 2nd time too., If you chose a hospital with poor standards, your only hope of recovery is to take a wand of money over to another more reputable hospital. Additionally since hospital profits are way higher than the surgeons compensation, the willingness for surgeon or doctor to make good does not help at all, because the majority of costs are from the hospital, to bolster insurance company share holders and hospital share holders. Half of the airplanes you see in the sky are owned by Well Point, CIGNA, AETNA, Humana, United Healthcare. Electronic medical records is truly counterproductive to the consumer. The electronic health records is automatically also the list of pre-existing conditions that no American wish they have when they need to buy insurance. The percentage of Americans that report “perfect health” much higher than in other countries. Can anybody figure out why that might be?

        • Of course the people claim to be healthy in order to get insurance. 1 question though the Hospital only writes down a loss on the patients that do not pay, correct? What is the percentage of people that don’t pay? and how does that percentage relate to other countries.

        • Your article focuses on hospitals. However you list,; Well Point, CIGNA, AETNA, Humana, United Healthcare, all insurance companies as the recipients the profits of the borderline ‘tax fraud’.

          Instinctively the bad guys are insurance companies. They add noting to the transaction. Hospitals as an aggregator of services do, medical professionals have an obvious role but the insurance company simply is the beneficiary of bad government policy. I understand that an instinctive intuition is not data but I see my hunch reinforced in your entry.

      • About 2nd surgery needed to correct a poor outcome of 1st surgery: Please imagine the deterrent of the 6x invoice above cost has on the likely hood of receiving the 2nd surgery. Insurance company will immediately bail out of responsibility and blame surgeon and doctor, and asking them to carry 100% of the cost of 2nd surgery. So doctors need expensive insurance both to cover hospital use costs, and malpractice lawsuits from patient. The villain is neither the doctor nor the patient. There is seriously a need for a “Blue book pricing list”. Since USA has a very large proportion of government employees compared to private sector employees, the demand for a “Blue Book price list” has to come from the government itself. When congress and senate and president agree that the 50 billion spent annually to render health benefits for government employees can be used as bargaining tool, then prices could go down also for employees in the private sector. Federal Employees Health Benefits (FEHB) annually 50B.

    • What if the expense grow exponentially? That would make a sense no?

      Do you have data for higher per capita countries (even if they are small) like Singapore, Luxembourg, Liechtenstein…

    • Which section are malpractice covered under? Administration and Insurance? Red Herrings?

    • Can you also distinguish countries funding is driven by government (i.e. Canada) vs private (i.e. US) health funding. You might mention where benefits for public(i.e. “economy of scale” for the government sponsored) vs private (i.e. only the wealthy can afford some services and medications/products)

    • I would be interested in seeing more countries plotted on the graph. Is there a lack of data for Japan, England, China, Saudi Arabia, ect?

      I would also be very interested if you could adjust for age. How does the average population age affect spending?

      Thank you for attempting to explain this. I am very interested in reading your series.

    • Wait a minute. I understand the concept that those with more money might spend more money. But when rich people spend more money, it is generally because they want to buy better things. You don’t simply buy a Porsche because you have more money. Yes that is neccesary condition, but not a sufficient condition. You only spend more money on a Porsche because it is better in some quantifiable way than a Civic (better performance, looks better, etc.). That’s the problem. I’ve never seen any study that shows that our health outcomes are better than those in comparable countries.

      I know you’re going to address outcomes later, but really I think this should be stated at the outset.

    • @Carmen – I’m not sure how that would change the equation. Especially since a graph charting wealth and health care spending on an individual basis would likely not have the same relationship. Sick people cost more money…

      @brabrabra – It’s not a totally linear relationship. It’s a fitter curve. And it’s not exponential…

      @Eric B – Patience! :) Send me an email, and I’ll point you to data on this…

      @Bernie – Lots of countries (like China) don’t share data. Not sure why England and Japan not there, but those are LOW spending counties.

      @Shonin – I don’t disagree, but you can only put so much in a post. Outcomes are coming…

    • Thank you for this series. It is great.

    • A few points:

      Interestingly most everything else is cheaper in the USA.
      Interestingly we have always spent more.
      Other countries constrain supply AND SPENDING, in the USA we only constrain supply.

    • So where are all the other countries? UK? Italy? Japan? Is South Korea somehow more relevant than these?

      Or maybe the line was drawn first and the countries that came close to it were then selected. Show us the graph with all countries on it, and I might take this seriously.

    • Fraud and Corruption pursue wealth and insane inefficiency. I would say that if fraud/corruption were eliminated, our health care costs would be cut in half, at least.

    • When doctors bill an insurance company, the company generally pays only a percentage of that bill. If the doctor bills out $1000, the insurance company won’t bat an eye at sending a $200 check. The doctor can hire a lawyer, or just bill more next time. Which sounds easier?

      This artificially inflates the cost of care pretty rapidly.

    • Good post, and a timely one.

      The McKinsey study, prepared by our good friend Bob Kocher and his colleagues, is a very good start in trying to understand this issue.

      The current study, however, was preceded by a McKinsey study in 1996, in which they tried to decomposed differentials in health spending between that of the US, the UK and Germany (then West Germany), based on 1990 data. It was a very sophisticated and expensive study, using tracer analysis for several standard diseases that literlaly assembled informnation of physical inputs into the treatment of the episodes in the three countries. These inputs could then be priced with prices in purchasing power parity (PPP) dollars. Based on these tracer studies, the team then worked the numbers up for the nation as a whole as best they could. Alan Garber of Standford, Kenneth Arrow and several other distinguished health economists were consultants to the project.

      As I recall it, according the the study, and after age adjusting, in PPP dollars Americans in 1990 spent about $1,000 (PPP) per capita more on health care than did Germans.

      In fact, however, Americans were estimated to receive $390 less care than did Germans, but spent $360 more per capita on administration and $737 per capita more due to higher prices. There was a category “other” accounting for the remainder of $250 per capita spent more by Americans.

      Unfortunately and somewhat curiously, I don’t recall that the current McKinsey study focuses on prices separately as an expanation for the spending differential at all. In a Health Affairs paper entitled “It’s the prices, Stupid!” a number of us specifically fingered higher prices as a main culprit.

      And why the higher prices? Because unlike the other nations in the OECD, which allocate considerable barganing power to the payment side of the health system (perhaps too much), the US has fragmented its payment side so that each individual payer has little economic leverage over the more concentrated supply side in bargaining over prices. Divide et impera is the battle cry of the supply side.

      Taking the matter close to Austin’s home, how much bargaining power does insurer Pilgrim Health actually have vis a vis the giant Health Partners in Boston?

      The whole thrust of US health policy is to constrain spending by lowering utilization — e.g., through ever higher cost sharing by patients at point of service — even though Americans are already relatively low users of health care. We should instead pay more attention to prices.

    • The reason that healthcare is so much more expensive in the U. S. than in other OECD countries is easy to explain if you consider it historically. The recent history of healthcare pricing can be thought of as consisting of three phases.

      The first phase followed the enactment of Medicare/Medicaid. For the next five or six years healthcare providers’ incomes grew rapidly due to increased utilization.

      In the early to mid-1970s that phase ended–the increased utilization phase was essentially complete but providers had become accustomed to sharply increasing revenues and Congress wasn’t saying “No”. During this phase providers increased their prices for the same services.

      In the late 1970s and early 1980s Congress, alarmed at the unexpectedly large size of Medicare/Medicaid expenses, began to economize. Thereafter the high price of healthcare can largely be explained solely on othe basis of annual increases “to cover inflation” at the rate of 4-5% or so per year. By the power of compound interest over time that has risen so that Americans are spending three times what other OECD countries are for healthcare.

    • One observation. I would agree that part of the price of health care is correlates with national income, the US figure does not fall on the line determined by the rest of the data. The US is ~28% above that line.There’s a number of other factors here. So we’re still paying too much and by most measures we arent getting a better product.

    • One big factor is that there is no real healthcare market – meaning that the consumer does not know what he is paying for healthcare – the cost of healthcare is not transparent – it is based on formulas pulled from the RUC (read about the RBRVS – talk about command and control!) and then handed to Medicare and then funneled down to the consumer.

    • No disrespect meant, but an economist’s lens will explain the complexity in a particular way which necessarily excludes deconstructing our whole notion of “health care.” For example, what if we were to ask ourselves what IS caring for one’s health and for our society’s health? What if we looked at the issue historically and cross-culturally? For example, by contemporary European human-rights standards, our system is a violation of basic human-rights. In Germany, for example, there exists hardly a culture of prescription drugs for all the psycho-illneses we seem to have. How can that be? Are Americans are more depressed and anxious? I doubt it. Might the difference have something to do with industry and profit motives? That our entire health care system is underpinned by for-profit industries? If you haven’t seen Sicko, you should. Interestingly, it didn’t get much press because healthcare industry lobbies made sure it didn’t. Remember all the booing and hissing at the Canadian system that circled the air waves; that was just about the time that Sicko was playing in theaters.

      I’m not trying to reduce the complexity, but to reframe the discourse a bit. I observe that what’s expensive is an outcome of the industry we call health care, not the notion of health care itself. Indeed if you looked at the results and conclusions of studies used to put products and services on the market, you might be surprised by what you find. Cholesterol research is one. If you looked at alternative health care methods, and age old, indigenous, wisdoms of caring for one’s health, you’d see another angle too. In short, more cost-effective ways exist for caring for one’s health than our society on the whole, is capable of perceiving and systematizing.

      If I could make one oversimplified point it would be this: On the one hand, caring for one’s health is an individual’s responsibility in the United States, because we are a society who does not believe in the concept of “the state” caring for its citizens in the way Europeans do. On the other, to not consider our societal and cultural context within which the individual exists, and how it undermines basic human health, (in part by not caring for it) is to be blind to the fundamental inhumanity of individualism.

      Citations available

    • Looking forward to reading the rest of this…
      Can you also please include something addressing the possibility that the US, with its huge post-graduate system might be underwriting much of the R&D costs that can then be leveraged by other countries?
      Would looking at some form of cash flow in each segment help? E.g. Pharma companies spend $X MM with a return of $Y MM which further fuels spending on one or more areas?

    • One thing I’d like to know is actual operating costs for technician level services. Take a half-million dollar MRI machine or a $300k x-ray machine (or however much they cost), amortize over, say, 5 years, add in operating costs (office rental, operating technician, etc) and add in a reasonable profit; divide that by operating non-stop during normal business hours.

      What do you get? (And compare that to industry standard cost of said service)

    • RE: Carmen Phillips
      “I’ll try to keep reading the series, but given that so much of the wealth in the U.S. is concentrated in an increasingly smaller and smaller group, it doesn’t quite make sense to me that overall wealth tracks with increased health care spending.

      Hopefully this is something you’ll address in your series.”.

      While this is a valid point it isn’t all that different to other countries. I’m pretty sure that almost all of the countries concerned here probably have the largest portion of their wealth contained within the smallest portion of their populace. Every country has their incredibly wealthy and their poverty stricken. While possibly more pronounced in the USA it isn’t unique.

    • “We are way above that line. Moreover, it’s debatable that we are achieving better outcomes than these other countries.”

      Can you quantify this? I have always understood that the United States has the best healthcare system in the world. Much of the rest of the world owes the quality of its healthcare to what we have done (and paid for) here in the United States. And people from the rest of the world still come here for health care unavailable (quantity and quality) in their own countries.

    • Tell me how you explain this. I had a heart attack in ’08. Went into the hospital on Friday night, was released on Monday. After a while, I received a statement of charges from the hospital for $114,000, which almost sent me back in! Later, the insurance company negotiated a final payment of just over $10,000, which the hospital accepted. A year later, I received another bill for $2,000 from the hospital with an explanation that an “audit” found additional un-billed charges. I had to pay this myself or suffer all the nasty collection processes they employ. How can $114,000 be whittled down to $10,000? This tells me the original charge was fluff anyway!

    • As a Registered Nurse I would like to comment on you blog by stating My opinion of why our healthcare cost is so inflated. First, I would like to refer to the following article Thorpe, K. (2007) Reframing the debate over health care reform: The role of system performance on affordability. Health Affairs, 26 (6) 1560-1562. From this articles Thorpe (2007) states that there are several factors that account for a 40 present rise in spending on health care. Those factors include: a rising share of spending flowing through insurance, medical malpractice, rising real income, and demographics.
      From you blog I agree that the increase in wealth and spending on health care for the most part is a waste. If we are the riches nation, why is it so difficult to use our funds wisely? Obesity is a major factor in the rise of health care. Obesity leads to multiple co-morbidities; such as, heart disease, stroke, peripheral artery disease, and diabetes. All of these disease processes significantly raise health care cost, by increasing number of hospital admissions and they are difficult to manage. I feel that health care reform, with its promise to increase access to health care and accountability on physicians for better performance may decrease spending. I know this will only occur if our government, lobbiest, and special interest groups can come to an agreement.
      Thank you, kvess

    • Anyone who thinks the relationship between income and spending is even approximated by the McKinsey regression should read this:

      • @Chris Conover – Why does it make sense to add all fifty US states and then to give each the same weight as the individual countries? Since the data from the states are highly correlated (more so than data across nations), that effectively gives the US massive influence over the shape of the fit curve, forcing the US to look less of an outlier. Of course the methodology drives the conclusion, but the methodology has to stand on its own as sensible. That at the link you provide doesn’t. (Or if it does, I’m not seeing how.)

        • By the same logic, how in the world could it make sense to compare Iceland (population 319,000) to the U.S. (population 307 m.)? Indeed, Iceland is smaller than every single state in the U.S. Likewise, 25 U.S. states are larger in population than Ireland (4.5 million).

          In truth, the countries included in the McKinsey chart are a quite curious amalgam: two of the OECD’s most populous countries are missing (UK and Italy, both with populations of 60 million), yet South Korea (population 48 million) is included. Portugal and the Czech Republic both are included, with populations of 10.5 million, but Australia, with double that population, is left out. Apart from Iceland, there are 5 other countries with populations under 10 million included by McKinsey even as it ignores 6 other countries with populations larger (in some cases, MUCH larger) than that.

          The point is that if it’s legitimate to include countries as tiny as Iceland, there’s no a priori basis for leaving out any of the U.S. states. And even if the analysis were restricted to countries of some minimum population size–say, 10 million–seven U.S. states meet that threshold. If the minimum were set at 5 million (reflecting McKinsey’s inclusion of countries such as Austria, Switzerland, Denmark and Finland–all with populations between 5-10 million), an additional 15 states would qualify.

          As for correlations in spending, it is true that a national program such as Medicare helps drive spending across all states. But we also know that even within Medicare, reimbursements per enrollee vary across hospital market areas by a factor of 3, so there is nothing “inevitable” about any particular level of health spending. This point is further illustrated by the 68% differential in per capita health spending (by state of residence) between Utah and Massachusetts. Indeed, the spending of states such as Utah, Arizona and Idaho is quite comparable to spending in countries such as Norway, Switzerland, and Luxemborg (countries notably absent from the McKinsey compilation) despite their having very different forms of health care financing and delivery.

          • The population has no relevance to the issue. It’s the cost to the individual that matters. The USA is world renounced as the most expensive place to get sick. Even travelers have from around the world have to pay extra for insurance if they want to visit the USA.

    • Can you please address prices and the lack of transparency for me, the purchaser? I wonder what would happen if a Dr. posted the prices for an office visit?

    • A well done research by Aaron Carroll. I couldn’t be more impressed.! I will be studying this more, the 12 segments. Thus far I have only gone through the first six.

    • I am denied medical help because of pre-existing conditions: really? is that
      right? I am with a non-profit, Medi-share, not a ‘regular medical insurance company”:

      I wrote this today to them:

      “What we need to know (and admit) about insurance companies.. even Medishare.. is that neither is acting in the best interest of the person in need. Pre-existing
      conditions make its insured unable to get help. So, then.. someone like
      me gets to either go thru massive struggle and pain, or go into debt and lose
      their home over medical debts to get ‘well’… when there is a pre-existing condition.

      What about developing a plan that takes care of His children, in spite of pre or
      not? I am in foreclosure, no income, disabled
      by a health condition that even OHP and DHS won’t recognize…

      Lisa…your heart HAS to be ‘there’… and it must frustrate you to work for MediShare.
      It is not doing what Jesus would do, and you know it. It falls so short.
      It fails in so many ways (yes, I do see the value when it has it) I know
      your heart, as it is mine.. as we are His.

      What would you do, to create such an entity to take care of His own…… if you could
      develop such an entity? I’d love to talk more.. but know you are only an
      employee and could get fired if found out..

      Keep listening to His promptings… Don’t put your faith in your job, but in His will
      for you and the voice of the Holy spirit.

      After all.. this is just a temporary setting for us.. yes? Because of the
      shortfalls of Medishare and the ‘system’.. I am going to do great things in His
      name.. and engage in massive change at top levels, and help people like me to get help
      that they deserve. I so love this opportunity! happy, I am….

      Blessings to you and your family.


    • The delta between the U S and the curve on the chart may have another factor, which many would call waste.

      All of the other nations have one, central government

      The U S has one central government, plus 50 smaller, local governments

      all of whom have an oar into the health care costs of the nation

    • I look forward to reading the rest of the series ( I wasn’t on the web back in 2010!) I’ve a question/suggestion, the WHY? why do Americans seem to ‘need more health-care’..I believe there is a link with the multi national food corporations.. With the only options produced for general consumption being filled with either gmo’s, high fructose corn syrup, artificial sweeteners, preservatives, unhealthier fats (than trans fats) well as in the feed given to animals, filled with anti-biotics, other chemicals..all mutate our bodies..creating disease…,, I wonder if this is addressed in your series..or maybe there’s one already out there? :-) Either way THANK YOU for writing and posting this series.. Happy Day to you and yours! :-)

    • If you have ever worked in any form of medical billing, you know that the biggest reason for increased healthcare costs is the lack of truth in billing. From the convoluted mire that causes double billing to up charges on anything and everything in order to maximize reimbursement, it is without a doubt the biggest organized crime in America today. The consumer protection laws we need to fix this are already in the books, all we need is for some of our Congress people to grow a spine and fight for enforcement.

    • That’s not “waste” it is “profit”. This is a stunningly elementary conclusion.

      “Which of these things is not like the other” logic might be your first recourse, but as someone who works for a large medical center, some familiarity with healthcare economics would provide a clue as well.

      This is not inspiring confidence.

    • I have seen this study quoted over and over that the US has a much higher health care cost per capita than other countries. Does this study count the estimated 23%-26% of our US population who are not citizens, yet still receive health care in the US? Maybe the study is simply using the wrong denominator when calculating this figure, and using the correct denominator would yield different results more in line with other countries.

    • Correction, the correct figure is 4%-6% instead of 23%-26%.


    • Look…I may be out of line here, but ,I’d like someone to
      take note of the fact that many of our physicians have no bedside
      manners. I have experienced extremely negligent doctors, and merely
      did not have the confidence to speak up when they should have
      ordered a test, a skin scraping sample, a bloodtest, etc… In
      exchange for my insecure behavior, I was forced to visit 5
      different physicians in my search for relief from an unexplainable
      rash that covered half of my body this search lasted^6 months…I
      was prescribed 6 medications…and gained 20 lbs…,the cost of my
      office visits and misdiagnosis, in addtion, were completely

    • How can there be any comparison between healthcare systems in the United States and other countries? We are about the only developed country that does not provide a universal/single-payer plan for its citizens. The whole point here is profit for the insurers. There’s no basis for comparison. It’s like comparing apples and oranges.