• What makes the US health care system so expensive – Red Herrings

    If you haven’t read the introduction, go back and read it now.  That introductory post also includes links to all the posts in this series on what makes our health care system so expensive.  Each of these pieces is going to discuss one of the components of unexpected spending that accounts for why our system is so expensive.

    Remember, these posts are going to follow a common theme.  I am going to highlight how the United States is spending more than you’d expect given our wealth.  Much of this comes from the McKinsey & Company study, Accounting for the cost of health care in the United States.

    Red Herrings are things that divert our attention away from things of true significance.  In the case of health care spending, I’m using red herrings to describe the things that people sometimes believe are the real reasons for blame, when they either just aren’t or not nearly as much as thought.

    1) People in the United States are older.

    One time when I was a guest on a radio show, a caller claimed that in other countries, they rationed care for the elderly.  Therefore, old people died.  This meant that there were more old people in the United States.  Since the elderly consume more care, in general, that accounts for our extra spending.  Done!


    As you can see, the United States has the lowest percentage of its population age 65 and above.  This argument is completely wrong.  In fact, it gets worse:

    The United States also has the largest percentage of children.  They are the cheapest group to care for.  If you need further convincing, here is a chart showing the median age of these countries (the US is that thick red line at the bottom):

    So – if anything – the age of the United States population is set up to make our health care system cost less.  Try again.

    2) People in the United States have unhealthy habits.

    Surely you’ve heard this one.  We make unhealthy choices.  We bring it on ourselves.  How?  First of all, we smoke less:

    And we drink less:

    So, those reasons are out.  In fact, those details should make our system cheaper!  Yes, we do eat too much:

    But does that alone account for our increased costs?

    3) We are inherently more unhealthy.

    Since I’ve covered this elsewhere in detail, I’m not going to do it again here.  I encourage you to go read that post.  But, in summary, analyses have been conducted looking at the relative prevalences of disease states (including obesity) in the United States and other countries.  And, perhaps, the prevalence of diseases (including obesity) in the United States could account for about $25 billion in extra costs.  But it’s not the demon some claim it is.  It’s a red herring.

    4) Malpractice.

    I’m already bracing for the screaming.  You should be ready for this, if you’re a reader of this blog.  Yes, the malpractice system costs money.  Yes, defensive medicine exists.  But no, malpractice is not the real reason for the high cost of care in the United States, and no, tort reform won’t fix it.

    How much does the malpractice system cost?  The most recent estimate published in Health Affairs found that medical liability system costs are about $55.6 billion in 2008 dollars, or about 2.4% of all US health care spending. Most of this, or about $47 billion, is due to defensive medicine.  So yes, that is theoretically care that should be reduced.  But we have no idea how much of it is actually not beneficial.  It’s likely that some good comes from that care.  How much?  Blaming the massive amount of overspending we’re seeing on this relatively small amount is not going to help.

    Moreover, pushing this as the real cause of high costs is misleading because there’s little reason to believe that tort reform will do any good.  A recent study showed that tort reform which led to a 10% reduction in malpractice premiums might translate into a health care spending reduction of 0.1%.  That’s not going to make any difference.  This is confirmed by what we’ve seen tort reform do in Texas.  And, it’s confirmed by what we’ve seen tort reform do in California.

    Malpractice isn’t the root cause of our cost problem, and tort reform isn’t the solution.  I wish it were that easy.

    So, to recap, our population’s age doesn’t account for any costs at all.  Both disease prevalence and malpractice do, but even then, they aren’t significant factors.  Let’s add them to the chart:

    • Thanks for this great series!

      At the risk of seeming to complain, the introductory post needs to be updated with links to this post and the previous post.

    • Aaron
      I would appreciate your take on this one. I know you know the literature on this subject as well as anyone, and have definitely heard the quoted figure of $150B from the citation below.

      Given the costs of our system, the $25B you reference above and in other posts, just seems low, and +/- biologically plausible based on associated disease prevalence (DM), and the toll obesity has on societal productivity, etc.

      This is not a challenge so much as a query for a possible explanation on the disparity. Why the difference in figures, and do you feel certain of the lower estimate of $25B?

      I have my doubts, again, just does not feel right on its face.



    • Brad,

      You have to remember that I’m not presenting total costs, but the amount that we’re spending ABOVE what we should be. When I say $25 billion, it’s not that we’re spending $25 billion total. It’s $25 billion more than can be blamed on prevalence alone compared to other countries. We may be spending many billions on obesity – but they have a lot of obesity, too.

      Plus, some things are less prevalent in the US, saving us money.

    • um, smart guy, what if we’re unhealthy because we drink less?

      seriously though, I’m not giving up the ship on malpractice despite, like yourself, having seen all the data. There’s just an overwhelming amount of anecdotal information from literally every physician I know describing tests they order in the interest of defensive medicine and calming the patient/family down. I actually think we just don’t have good data on this. To really get good data on unnecessary procedures/tests, the only way to build a numerator that means anything is through exhaustive chart reviews and even that might not be enough.

      If I had the means and the wherewithal I think I’d design a study using the privacy protections of the Patient Safety and Quality Improvement Act and have physicians self-identify procedures they ordered that they felt were unnecessary.

    • ThomasEN,

      Even if you’re right, research shows that tort reform doesn’t affect medical spending. Docs keep on doing what they are doing. You need real system change to alter their behavior.

    • got it, thanks aaron, dont know whty i didnt pick that up in the first place.

    • oh agreed. Tort Reform – at least any version I’ve seen – is the wrong solution to an actual problem.

    • Are you sure that the population under 19 years of age is cheaper than, say, 20 to 50 year olds? There’s immunizations, which are the lion’s share of a pediatric office’s expenses, and premature infants, with expensive NICU stays (for those that a poorer country might not even count in their birth rate (!)). I suspect that kids have more well checks than the average middle age patient, as well…

    • Ohio passed tort reform also, in 2003. It hasn’t done anything for costs.

    • Nowhere in all the literature I have seen about the recent epidemic of obesity, does it mention that the official definition of obesity was changed to include more people.

      Same for the epidemic of diabetes — they changed the definition of a diabetic to include more people.

      I do not have numbers, I just know that it happened. So how does this affect the cost appraisal?

    • And EC – the same applies to Hypertension – we have changed the threshold level to make sure we include more people in those having high blood pressure

      What is most annoying is that all of these adjustments are based on inference – and at best inferences that are drawn from bad or poorly designed experiments/studies.

      In the end what is great about all this is that we create “bad people” who we can blame for our problems ie “Health care would cost a whole let less for me if there were fewer “bad people” out there smoking, drinking and/or eating too much”

      This is very convenient – but totally bogus…

    • In the alcohol graph, why is “liters per capita” measured in percents? Is this liters per capita per year?

    • I am trying to understand what the lawn green two thirds of the pie chart means… If that 2/3 represents overspending on items not otherwise accounted for in the pie chart, then does it not imply overspending caused by the structure of the health care system itself?

      Could this waste be as a result of duplicate spending from competing interests? While each of 5 to 10 insurers may not have administration costs that are that much higher than in other countries, for example, running so many organizations at the same time could mean spending a lot more on administration than a country with only 1 central administration. Similarly, having competing hospitals, ambulance services etc. could mean more money going to each simply to operate. It would most certainly be interesting to know to what degree the amount of redundant health service delivery affects that pie chart.

      Perhaps outlining the relative percentages of people employed in each area of health delivery by country would show such a difference were it to exist.

    • Elephants Child and Lonely Libertarian,

      Please, look around you, there has been a sharp increase of people in the US with Type 2 Diabetes. A majority of people diagnosed with Type 2 Diabetes are overweight (many obese). Many cases of Type 2 diabetes can be prevented – through education and awareness. This is an epidemic that needs to be taken into consideration when discussing health care costs. In my opinion, there needs to be an incentive to be healthy and therefore preventing medical costs. How is this going to be done? I seriously wish that I had an answer….

      Aaron – I would be interested in hearing your opinion on Accountable Care Organizations that are starting to become popular across the country?

    • The idea that U.S. citizens eat too much is an oversimplification of the problem of obesity here. The high obesity rate is largely due to medical practice and the food industry in this country. I mentioned in another comment that this health care system is guided by industry and not by human health. Most of the people we view as overeaters are actually suffering from an undiagnosed or poorly treated endocrine disorders or hormonal imbalance. It would take a whole series of my own to explain, but I’ll try to whittle it down to a single brief example scenario.

      A woman with a family history of breast cancer follows doctor’s orders and has frequent mammograms but is never offered a shield to protect her thyroid. The problem is exacerbated by the fact that she is constantly exposed to goitrogenic compounds, xenoestrogens and poor nutritional quality in the processed foods she eats. Over time, she notices that she is gaining weight, losing hair, having joint pains and feeling fatigued and depressed. She brings these complaints up to her doctor, who tells her she is growing older and prescribes Celexa to cope with her depression. In reality she’s hypothyroid but her doctor missed this significant factor because he relied on a standard TSH test that tends to miss at least half of patients suffering from hypothyroidism. Celexa is contraindicated for her condition and leads to a more serious state of hypothyroidism. She gains more and more weight and becomes less active because her joints ache and she suffers from fatigue. She becomes diabetic, contracts heart disease and eventually requires a quadruple bypass.

      This is but one of countless similar examples. Medical doctors here are trained to prescribe medications or procedures that treat symptoms and they are poorly prepared for seeking root causes of health problems. Doctors are regularly visited by drug reps who use all means of persuasion to convince them to prescribe medications their patients don’t need. The system is expensive because it is motivated by profit.

    • Just a couple comments – the magnitude of the obesity disparity is notably larger than the magnitudes of the differences in other “unhealthy” metrics like alcohol and tobacco consumption. Observationally, obesity has greater health impacts than moderate alcohol consumption (a statement I believe is backed up by evidence that I can’t quote off the top of my head) and perhaps has impacts on the order of those of tobacco use. Not that this undermines the premise of this series – that our system’s expense is multfactorial – but perhaps worth exploring further.

      On the subject of tort reform, I’ll offer two contradictory comments. Medicine is an inherently conservative field, with incentives that often overlap with those created by the threat of litigation. This is to say that nobody wants to miss health harming conditions, whether its because of some lawyer or because it is our job not to miss these conditions, This contention would undermine the benefit of tort reform, although the benefits to health care workers mental health would most likely be substantial.

      A possibility that hasn’t been raised in this comment thread, and that has actually been contradicted through assumption in one of the above statements, is that this extra medical attention might actually be harmful, not just wasteful. Our current chest pain paradigm is a prime example. National guidelines dictate that patients presenting with chest pain of a potentially cardiac etiology necessitate a “provocative test” within 72 hours – which practically means a stress test or potentially a cardiac catheterization. When applied broadly enough to ensure that we do not exceed the “acceptable miss rate” of ~2% of heart attacks, you end up with an exceptionally low risk group of patients, meaning that the false positives – a phenomenon that is unfortunately common with stress tests, and all the inherent expenses and risks of further work up – become significant in relation to the true positives, all the while maintaining a unexpectedly high rate of false negatives – aka misses or people that enter the health care system, are worked up, and still have a heart attack or sudden death . In short, we apply a poor test to a low risk patient group and end up with a high ratio of iatrogenic harm to benefit, without adding much benefit to a simpler workup such as having a normal / unchanged EKG, two negative troponins +_ maximal medical therapy. Chest pain is one of the more outstanding examples, but it may emblematic of a larger philosophical flaw in our medical system – one generated not by our fear of lawyers but our own and society’s expectation that we can’t miss serious disease at even near zero rates.