Health Care and Furnace Repair

Filed Under (Health Care) by Nolan Miller on May 6, 2010

I recently had a problem with my furnace.  Water was dripping from a joint in the PVC exhaust pipe.  So, I called the heating company.  The repair man came out and told me that it looked like the original installer had had a problem with that joint, so they smeared a bunch of glue over it to seal it.  He said that he could do the same, and it might work for a while.  But, if the joint opened up, there was the possibility that exhaust/carbon monoxide would enter the house.  The other option was to rebuild the exhaust pipe, which would cost $150 more but would eliminate the added risk of killing me and my family.  I weighed the options and chose to rebuild the pipe.


I thought about this story when reading this recent article from the New York Times about efforts to teach doctors about the costs of various medical treatments.  In particular, I wondered what the right role of a doctor is in deciding, based on cost, whether I should have a particular treatment or not.  Suppose, for the moment, that medical care were not insured.  You could imagine a conversation very much like the one that took place between me and the furnace repair man.  “We could put you on drugs, and that might help.  Or, we could do surgery, which would fix the problem but cost $5000 more.”  I could then weigh the costs and benefits and make a decision.  Actually, now that I think of it, the doctor doesn’t even really need to know the cost of the treatments.  Dr. Jones could explain the health effects of the various treatments, and then Smith from the business office could come in and explain the cost of things.


If people are given this kind of information and make decisions based on it, then we will tend to choose expensive treatment when we think the benefits warrant the extra cost.  While this may not bring the cost of care down to European levels, it will at least improve efficiency – we’ll tend to spend money on services we think are more valuable.  But, this is really a point about giving patients information on costs, not doctors.


So, what is the role of doctors in medical cost-benefit analysis?  I’m not sure.  But, it seems like there are at least two key differences between health care and furnace repair.  First, there’s insurance.  If I’m paying $0 out-of-pocket regardless of the treatment, I’ll tend to choose treatments that have higher cost.  So, informing the doctor of the costs and benefits of treatment might be important here.  But, this would only be the case if the doctor then makes different choices than he or she would without the information.  My sense is that doctor’s often view themselves as advocating for patients against insurance companies.  So, the doctor, just like the patient, may say that the patient doesn’t pay anything more for the high-cost treatment than the low-cost treatment, so why not go with the better one?  It is unclear whether just giving the docs information, without any financial consequences for choosing high-cost treatment, will make a big difference.


A second reason why health care is different than furnace repair is that people in need of health care often do not have the time or capacity to make well-reasoned decisions.  If I am acutely ill, doctors may have to make choices about how to treat me without consulting me about how I feel regarding various cost-benefit trade-offs.  In such a case, it is also unclear whether the doctor would get anything useful out of knowing the cost of various treatments, except maybe in the case of treatments that are dominated both on cost and benefit grounds.  As one surgeon put it “I get them out of the operating room alive.”


In the end, while it is interesting that medical schools are beginning to invest more time and energy into teaching doctors about the cost of care, it is unclear whether and how that will translate into changes in the overall cost of medical care in this country.

A case study on the high cost of health care.

Filed Under (Health Care) by Nolan Miller on Apr 1, 2010


There was a really interesting piece this week in the New York Times about a recent FDA decision to allow AstraZeneca to market its drug Crestor to a new class of people – those who are basically healthy.  The story, I think, yields some insight into why our system is so costly and why it is not so easy to curb our spending.

As background, Crestor is one of the newest and most powerful statins, a class of drugs that had traditionally been used to treat high cholesterol.  Side effects associated with statins are generally low.  However, the risk of side effects increases with the strength of the drug, and Crestor is one of the most potent ones around.  There have even been claims that the risk of side effects with Crestor is significantly higher than other statins.

Now, here’s the interesting part.  How did the FDA come to approve this use of Crestor?  The approval is based on the results of a study that showed Crestor led to a a significant reduction in heart attack and other risks from mostly healthy men over 50 or women over 60 with a risk factor such as high blood pressure.  These people did not have high cholesterol, the condition for which Crestor was originally developed.  The study was funded by AstraZeneca and conducted by the inventor of the CRP test, Dr. Paul Ridker of Harvard Medical School, both of whom stand to gain financially from the study’s results, AstraZeneca through increased sales of Crestor  (which commands a high price due to its on-patent status) and Ridker through increased use of the CRP test, for which he receives royalties.  Now, I won’t speculate as to the motivations of AstraZeneca, since it is a huge corporation not a person, but it seems like Dr. Ridker is a sincere believer in the power of the CRP test to more accurately predict heart attack risk, and so his motivations for promoting greater use of the CRP may be entirely altruistic.

So far, so good.  This is the way the capitalist system is supposed to work.  AstraZeneca created Crestor in order to make a profit.  In order to increase its profit it went out and determined that Crestor could provide health benefits to a new class of people.  If these people can be convinced to buy Crestor, AstraZeneca will make more profit and the people will reduce their heart attack risk.  In an ordinary market, the mere fact that people were willing to buy Crestor for this purpose would be de facto proof that the benefits are worth the cost.  Crestor is willing to offer the pills at a price, around $3.50 per day, and people are willing to buy at that price.  So, they must value the benefit from the drug at least that much.  End of story.

But, health markets are not ordinary markets, for at least four reasons.  First, the government regulates the market, with the FDA endorsing the use of drugs for certain purposes.  FDA approval is based on clinical effectiveness, not cost effectiveness.  So, the mere fact that the FDA says that a drug can be used for a certain purpose does not mean that it is a cost effective measure.  There may be other, cheaper ways of achieving the same results.  FDA approval does not say anything about that.  Second, drugs are often covered by insurance, and insurers often decide whether to cover some or all of the cost of a drug based on FDA approval.  And, if a drug is covered by insurance, much of the cost of the drug is rolled into insurance premiums.  So, the out-of-pocket cost of an individual who decides to buy a drug may be much lower than the total cost of the drug charged to the insurer.  Third, decisions as to whether a patient should take a drug are made in cooperation with physicians.  Often, the patient plays little role in deciding which medication to take.  Physicians, on the other hand, are more concerned with health than cost, and they may recognize that the drug they are prescribing has a low cost to patients due to insurance.  If the pill has a benefit and little cost to the patient, they why not prescribe it?  Sure, it is costly to insurance companies, but many physicians have, at best, an adversarial relationship with insurers, anyway.  I don’t think they would bat an eye at costing the insurer a few bucks in order to improve their patients’ health.  Finally, health care decisions are incredibly complex.  Information on the benefits of treatments is hard to come by, and cost-benefit information is even harder.  Faced with FDA approval of a drug for a certain purpose and their physician’s advice that they should take it, many people will simply go along with the decision rather than try to overrule the experts.

All-in-all, these four factors on the “demand side” for drugs, and for health care more generally, point to why we tend to overuse them, at least from a cost-benefit perspective.  Drug companies produce drugs and put them out there in an attempt to make a profit.  There’s nothing wrong with that.  However, due to diffusion of decision-making responsibility, lack of information and muted price signals, people are not doing the right price-benefit calculation on the demand side.  The result is overuse.

In the case of Crestor, the Times article estimates that “500 people would need to be treated with Crestor for a year to avoid one usually survivable heart attack.  Stoke numbers were similar.”  Treating those 500 people for one year would cost about $640,000 (at $3.50) per pill.  Since these are usually-survivable heart attacks, that $640,000 might not even prevent any deaths.

Repeat this for every drug and every treatment, and that’s how you get to where we are today.  So, how can we encourage people to make better decisions?  As always, there answer here is not as clear.  One way would be to ask people to pay more of the cost of drugs and other health care out of pocket.  Faced with the full cost of their decisions, people will make better cost-benefit trade-offs.  Of course, doing this will erode some of the benefit of insurance in reducing risk due to health care costs, which may be undesirable.  (On the other hand, it can be argued that medication costs are predictable and, as such, are not really good subjects of insurance, anyway.)

While it is true in principle that increasing out-of-pocket costs can encourage people to make better decisions, the evidence here suggests that, due to the complexity of health decisions, this may not be the case in practice.  For example, it has been shown that, faced with an increase in drug copayments, people tend to cut back on all drugs – both essential and non-essential ones – by the same amount.  So, if we are going to ask people to do this sort of cost-benefit analysis, we’re going to have to give them better information about their choices.

Another suggestion has been to help educate doctors about cost-benefit tradeoffs.  Today, most of the information about the effectiveness of drugs that doctors receive comes from the drug companies themselves.  Of course, these companies have no interest in point out alternative treatments that are equally effective and less costly, and we shouldn’t expect them to.  Education of this sort would likely help, but there is currently no such infrastructure, and it is unclear where such information would come from.  The federal government has shown increasing interest in cost-effectiveness research in recent years, and the development and provision of this information (if it can be done in a fair, trustworthy way) might help.  Of course, the fact that the government will be actively working to undercut a firm’s product might undesirably dull firms’ incentives to innovate, so the implementation here would be tricky.

Both of these suggestions point to a critical need that must be addressed if we are going to bring down spending on health care.  Doctors and patients need better information on the benefit and cost of various treatments in order to make good decisions.  Although it wouldn’t solve the entire problem, I think that it is a necessary condition for improvement.  Some have argued that the right thing to do is for the government to make decisions regarding costs and benefits and directly regulate which treatments should be used in which cases.  Such one-size-fits all solutions, however, overlook the wide variability in patients’ conditions and individual taste regarding treatment.  Another approach would be for the government to invest in developing cost-effectiveness information and making that information available to physicians and patients in as useful a way as possible, allowing them to make better decisions.  To my mind, the government providing information to markets to help them work better is a much better intervention than directly regulating particular modes of treatment.

Adverse Selection — California Style

Filed Under (Health Care) by Nolan Miller on Feb 9, 2010

News from the West Coast today that Anthem Blue Cross, one of the largest private insurers in California, is raising the prices for the 800,000 or so people it sells individual health insurance policies by up to 39%.  The Obama administration is not happy, to say the least.  HHS Secretary Kathleen Sebelius fired off an angry letter to Anthem and its parent company, WellPoint, demanding an explanation.  Of course, this also comes at a time when the Obama administration is struggling to make the case that health insurance reform is urgently needed, so this also provides a perfect example for them.  The letter is kind of cool, since I have never seen an angry letter from a Cabinet Secretary before.  The text is here.

What I find more interesting as an economist, however, is WellPoint’s response.  They haven’t replied formally to the letter yet, but in a statement WellPoint’s spokesman said the following:

“As medical costs increase across our member population, premium increases to the entire membership pool result. Unfortunately, in the weak economy many people who do not have health conditions are foregoing buying insurance. This leaves fewer people, often with significantly greater medical needs, in the insured pool. We regret the impact this has on our members.”

So, where’s the economics lesson here?  In a competitive market, health insurance prices are driven by the cost of caring for the average person in the insurance pool.  That means that healthy people usually pay more than their actual cost of care and sick people pay less.  Although healthy people pay more than their average health expenditures in any year, they’re still willing to buy insurance because it provides them with, well, insurance.  In the event that they have a car accident or other unexpected, large expenditure on health care, they’re protected against the financial consequences.  This works fine as long as the premium (driven by the average cost of care) doesn’t get too high above what the healthy people are willing to pay for insurance against relatively rare events.

Now, enter the recession.  People are losing their jobs, wages for the employed are stagnating, and people are losing money on housing and financial investments.  In light of these challenges, some healthy people are looking at their health insurance premiums, their income, and the likely cost of going without insurance, and deciding not to buy health insurance.

This is a perfectly rational response to increasing premiums and decreasing incomes.  However, it results in the remaining people in the insurance people being, on average, sicker.  This means that the average cost of caring for the insurance pool will be higher, which will necessitate higher premiums.

Unfortunately (and interestingly if you study this stuff), this leads to the potential for what is known as an “adverse selection death spiral.”  The idea is that once premiums rise, the healthiest people who are still buying insurance may decide to drop out of the pool.  Since the remaining pool is even less healthy on average, premiums will once again need to rise to cover their higher medical needs.  And then the cycle starts over again.  In extreme cases, the premium just keeps going up until nobody is willing to buy insurance.

So, what next?  Well, the adverse selection story holds in competitive markets.  But, you can already see Secretary Sebelius telegraphing the administration’s punches.  They will argue that the price increases are not due to competitive pressure and an increasingly unhealthy insurance pool but rather a greedy, for-profit insurer trying to take advantage of people when they’re down.  For their part, WellPoint/Anthem will argue that this just shows why health reform is needed, but health reform of a fundamentally different sort than Obama has proposed.

My prediction is that we’re headed for a highly charged series of Congressional hearings that boil down to an attempt to drive home to voters that something needs to be done.  Really went out on a limb, there, didn’t I?

Time for Health Care Credit Cards?

Filed Under (Health Care) by Jeffrey Brown on Feb 8, 2010

Yes, I know that credit cards have gotten a bad name given the fact that so many families have gotten themselves deep into personal debt by not managing them effectively.  But a couple of interesting ideas have come across my desk recently that have me thinking that a “health care credit card” is an idea worth considering as part of an intelligent health care reform.

As background, my colleague Nolan Miller made a post last week about “The Health Reform House of Cards” in which he highlighted the difficulties in tackling reform.  He walks through the steps that one could take in order to put together a reform that actually works, highlighting how one reform element must be accompanied by other reform elements if it is all going to work.

As luck would have it, I recently received a proposal from a good friend of the College of Business, Jerry Carson.  In his proposal, which is much more detailed than what I will provide here, he outlines a system that appears to have many of the elements that Nolan calls for – and his approach has a role for health care credit cards.  I probably will not do Jerry’s proposal justice, but here are a few of the highlights …

First, require that everyone purchase a catastrophic loss health insurance policy.  In Jerry’s proposal, anyone could sell this insurance, including private insurers or the government.  Premiums would be based on the risk pool of the entire nation, which, in Jerry’s words is “the only true economical form of insurance – the broadest possible base with high deductibles.”  Policy terms would be the same without consideration of pre-existing conditions, and premiums would be tax deductible.  if insurers want to provide policies that wrap-around this catastrophic coverage by providing benefits that are more generous, they may do so, but such policies would not be tax deductible.  This would help limit the “Cadillac plan” problem that leads to inefficient over-utilization of health care services.  It also limits the tax burden arising from the “tax subsidy” of these plans.

Second, Jerry suggests allowing individuals to use a “health care credit card” that can be used at point-of-service.  These cards are with full recourse to the individual.  Private lenders can issue the cards, and again, if someone cannot get one then the government can issue one.  In the case that the government issues one, all services would be reviewed by a federal agency, and unpaid costs would be deducted from federal/state benefits.  As Jerry puts it, it would be “socialized health care without the free lunch.”  

Jerry is certainly not alone in trying to come up with creative ways to solve our health care dilemma by providing incentives for individuals to care about the cost of care (a focus on incentives and individual responsibility) while still protecting them from catastrophic losses (a focus on social insurance).  A few months ago, the esteemed economist Martin Feldstein (former President of the NBER and former CEA Chair for President Reagan) wrote an op-ed in the Washington Post that called for us to replace the current tax subsidy approach with a health care voucher system.  Interestingly, he also calls for “the government to issue a health-care credit card to every family along with the insurance voucher.”  

Health care reform is an enormously complex topic, and as Nolan’s blog and Feldstein’s op-ed both suggest, the solution requires a number of inter-locking pieces to work.  Both Jerry Carson’s and Marty Feldstein’s creative ideas may be the kind of innovation we need to get out of the current health care reform quagmire.

The Health Reform House of Cards

Filed Under (Health Care) by Nolan Miller on Feb 4, 2010

[Note:  I originally wrote this last week and scheduled it to post today.  Yesterday I read Uwe Reinhardt's post on the NYTimes Economix blog that expresses much the same sentiment with more real-world input and fewer of the blow-by-blow details.  If I hadn't written my post already, I might have just linked to his.  But, since the work is already done, I'll leave mine here.  If you are at all interested in health, I highly recommend reading everything Uwe writes at Economix.  Even when you don't agree on the conclusions, he's right on the facts and identifying the key issues.]


In the wake of President Obama’s falling approval ratings and last month’s stunning upset in the Massachusetts Senate race, I hear a lot of people saying that this is the inevitable result of Democratic hubris, and in particular of Obama and the Democrats trying to do too much on health care too quickly, especially when the country has not yet bounced back from the current, severe recession, is (still) fighting wars in Iraq and Afghanistan, and faces an ongoing a real terrorist threat.

I’m willing to concede the points in the last paragraph.  Maybe this wasn’t the time to try health care.  Arguably, though, given the Senate’s anti-filibuster rules, the chance may not come again for a long time.  I’m also not a huge fan of the bill because I don’t think it went far enough to contain cost growth, which I (and others) have said is the real threat to the system in the long run.  But, I think that those who argue that Obama and the Dems should have taken a more incremental approach to expanding access to health insurance also miss the mark.  Here’s why.

Suppose that you want to extend access to the health care system to the roughly 46 million uninsured people in the U.S. (or just the 80 percent of them who are U.S. citizens).  The least obtrusive way to do this in the context of the existing U.S. health care system would be to either expand Medicaid to cover wealthier people or to bolster the dysfunctional individual (non-employment-based) insurance market.  Now, Medicaid does a good job of providing basic care to poor people, especially children and their families, but because of its low reimbursement rates is probably not a great way to extend care more broadly.  And, any broad expansion of this government-run program would surely meet with strong opposition.  So, this leaves us with bolstering the individual market.

The individual market is the insurance market for people who don’t get insurance through their employers.  To put it bluntly, this market doesn’t work very well.  Only about 5% of the U.S. non-elderly population gets its health insurance through this market.  There are several reasons for this.  First, coverage is expensive because those buying insurance through the individual market do not have access to the economies of scale and bargaining power of employment-based coverage.  Second, those who seek to buy insurance on this market tend to be sicker than the population as a whole.  Insurers respond to this by charging higher prices based on medical history (“risk rating”), excluding pre-existing conditions, charging higher rates for coverage, putting annual or lifetime limits on benefits, or simply refusing to cover the sick.  And, in many states they can cancel an existing policy with little or no justification.

All this translates into private, individual coverage being very expensive at the same time those who might buy such policies are relatively poor.  This leaves us in a situation where the sick can’t get insurance and the well do not want insurance because it is too expensive.  As a result, there are few people for whom buying insurance on the individual market is an attractive option.

So, how do you fix it?  The first step would be to prevent insurers from engaging in the types of practices mentioned in the previous paragraph.  So, let’s require that insurers charge everyone the same price for insurance and must enroll anyone who is willing to buy insurance.  Let’s also prevent excluding pre-existing conditions and lifetime limits on benefits.  In other words, we outlaw “abusive insurance practices.”

Fine.  Now anyone who wants to buy insurance can buy it at any time.  How does a reasonable person react to this?  Well, if you are currently sick, you buy insurance.  But, suppose you aren’t sick.  A reasonable calculation would be to not buy insurance while you are well and take advantage of the prohibition on denying coverage in order to buy insurance only if you become sick.  The result of this will be that only sick people will have insurance and that, without healthy people in the risk pool to balance them out, premiums will have to be high.  So, even though everyone will be able to buy insurance, it will still be expensive to buy, and many insurers will find that it is simply not worth the trouble of insuring an exclusively sick population.

So, how do you bring down the cost of insurance?  The way to do this is to bring the healthy people into the risk pool.  After all, even though they don’t buy insurance while well, they are already benefiting from the system which guarantees them access to health care if they get sick.  In exchange for this guarantee, let’s force them to buy insurance early.  So, we mandate that individuals buy insurance.

The mandate puts healthy people into the risk pool, so now we have a mix of sick and healthy buying insurance from the individual market.  This should bring average prices down.  How might insurers react?  Well, all else equal, the insurer does better if it attracts a relatively healthy pool to its policies.  While the regulations above prevent insurers from explicitly excluding sick people, they can try to design policies that are implicitly more attractive to the healthy than the sick.  For example, they may exclude mental health benefits, refuse to cover weight-loss surgery, put their offices on the third floor of no-elevator buildings, etc.  But, these “dumping” practices are wasteful.  So, let’s prevent insurance companies from engaging in these practices by standardizing benefits.  This will have the additional benefit of making it easier for consumers to shop for plans since it will be easier to compare apples to apples.  And, if consumers become more sensitive to quality and price differences between plans, this will encourage plans to improve quality and lower price, which is an added bonus.

In fact, let’s push that idea further.  In order to encourage insurers to compete more vigorously, let’s set up insurance markets, or “exchanges,” where people can easily shop for plans.  This will also help people choose a health plan, an extremely complicated decision.

Finally, we need to face up to the fact that the uninsured are primarily poor.  This means that many of them will not be able to afford coverage no matter how well the individual insurance market works.  If we want to them to have access to private insurance, we’re going to have to help them pay for it.  So, let’s subsidize poor people to buy insurance through the private market.  We can do this in two ways.  We can either directly subsidize purchases of individual insurance policies or force employers to expand their insurance offerings by requiring employers to offer insurance or pay a penalty if they don’t.  But, if we’re going to subsidize individual insurance purchases, the money is going to have to come from somewhere.  So, we’ll have to increase taxes in one way or another.  (Aside: Clearly, the best way to do this is through a tax on tanning salons, as proposed in the Senate Bill.  The only question I have is why we didn’t come up with such a brilliant idea sooner.)

And there you have it.  If you want to expand coverage and you want to use the private market to do it, you quickly find yourself with a very big piece of legislation.  It’s a house of cards, and without any of the pieces it will quickly fall apart.  (Another aside: many people feel the penalties paid by individuals and employers who choose not to buy/offer insurance are insufficient in the current legislation, so the house of cards may be destined to tumble, anyway.  See this op-ed by Martin Feldstein.)

While going all the way may be too far, it is unclear whether there would have been a way to ease into reform.  Unfortunately, the one part of the bill that can be chopped out without jeopardizing the short-run goal of covering more people is the one that we really need to address to ensure the long-run viability of the system — the cost reduction part.  Even if we expand coverage in the short run, without addressing cost and especially the rate of cost growth, we’ll be right back in the position of insurance coverage being unaffordable for an ever-increasing segment of the population in a matter of years.

Tuesday was a good day for health insurance stocks.

Filed Under (Finance, Health Care) by Nolan Miller on Jan 20, 2010

Following up on a post from last month on the public option and health insurer stock prices, here’s a chart of the price of CIGNA, Aetna, WellPoint, Coventry Health, Humana and United Health (along with the DJIA and Nasdaq Composite Index) from Friday until now.   Due to the Martin Luther King Holiday there was no trading on Monday, which means that Tuesday morning was the market’s first chance to react to the series of reports over the weekend that Democrat Martha Coakley’s bid to take over Ted Kennedy’s seat in the Massachusetts special election on Tuesday appeared to be in trouble.  In fact, in the Tuesday special election Coakley was beaten by Republican Scott Brown who campaigned on a vow to oppose health care reform.  Due to the fact that Brown’s victory will result in only 59 Senators in the Democratic camp, preventing them from defeating Republican filibuster tactics, the media has speculated that Brown’s victory will spell the end of health reform.


(Source: Google Finance.  Click here to make your own graph.  The original idea came from the Huffington Post, here.)

The impact on stock prices is striking.  See the two flat curves hovering around zero?  They are the DJIA and Nasdaq Composite Index.  The other lines (the ones that jumped up Tuesday morning) are the major insurers CIGNA, Aetna, WellPoint, Coventry Health, Humana and United Health.  As to what this means for the prospects of health reform (and what health reform means for insurers), I think the graph speaks for itself.

By the way, Coakley, Brown and their families kept robo-calling me over the weekend despite the fact that I don’t live in Massachusetts anymore.  I hope they figure that out before this seat comes up again in 2012!

The Public Option and Stock Prices

Filed Under (Health Care) by Nolan Miller on Dec 24, 2009

I assume that everyone has better things to do than read this blog over the holidays.  But, cyber-space is forever, so I thought I’d post anyway.  Here’s an interesting piece from the Huffington Post (sent to me by Dan Karney — thanks, Dan!) that traces the stock prices of health insurers since Joe Lieberman laid siege to the Senate bill on October 27, threatening to support a filibuster unless the public option were removed.

As a health economist, I’m pretty unsure about what the public option will do.  But, the market seems pretty clear in its expectations.  Basically, the stock of major health insurers such as Coventry, CIGNA, Aetna, WellPoint, UnitedHealth and Humana went up anywhere from 13 to 31 percent between October 27 and December 18, relative to a 2.3 percent increase in the DJIA.  Now, this isn’t all due to the Lieberman threat and the possible removal of the public option.  Some of this may be due to increased skepticism about the likely passage of any bill at all or other changes in the legislation that were introduced during this time.  But, it certainly appears that developments over the last month and a half have been interpreted as good for insurance companies.

Here’s the link.

Happy holidays to all.

Atul Gawande on How the Senate Bill Would Contain the Cost of Health Care

Filed Under (Health Care) by Nolan Miller on Dec 17, 2009

Atul Gawande just wrote (yet another) great piece on health care reform for the New Yorker.  Rather than spout off about it, I’m just going to copy a particularly interesting paragraph and give you the link.  Enjoy.

“There are, in human affairs, two kinds of problems: those which are amenable to a technical solution and those which are not. Universal health-care coverage belongs to the first category: you can pick one of several possible solutions, pass a bill, and (allowing for some tinkering around the edges) it will happen. Problems of the second kind, by contrast, are never solved, exactly; they are managed. Reforming the agricultural system so that it serves the country’s needs has been a process, involving millions of farmers pursuing their individual interests. This could not happen by fiat. There was no one-time fix. The same goes for reforming the health-care system so that it serves the country’s needs. No nation has escaped the cost problem: the expenditure curves have outpaced inflation around the world. Nobody has found a master switch that you can flip to make the problem go away. If we want to start solving it, we first need to recognize that there is no technical solution.”

Read more:

Will reducing obesity lower health care costs?

Filed Under (Health Care) by Nolan Miller on Dec 11, 2009

A number of the comments on my recent postings have focused on the idea that we could reduce health care costs in the U.S. if we lived healthier.  In particular, if we reduced obesity.  Sounds plausible.  So, I thought I’d check it out.

First, obesity in the U.S. is high and growing.  The obesity rate in the U.S. was about 25% in 2006.  One study estimated that, based on current trends, all adults in the U.S. will be overweight or obese.  Clearly this is not going to happen.  But, obesity is a clear and growing problem.

Second, what does obesity do to health care spending?  A recent study by Eric Finkelstein and coauthors in the journal Health Affairs estimated that the medical costs of obesity in 2008 could have been as high as $147 billion dollars, which is around 10 percent of all medical spending.  That’s a lot.  The average obese person incurs $1429 annually in additional medical costs when compared to a typical non-obese person.  If we could reduce this cost from our medical bill, this could represent a significant savings.  It wouldn’t bring our per-capita health expenditures in line with those of Europe, but it would help.

Or would it?

The Finkelstein paper above answers a particular question.  If you compare the annual health expenditure for an obese person and a non-obese person, how much higher are the obese person’s medical expenditures?  This answer provides an important part of the answer to the overall question of whether reducing obesity will reduce health care costs, but it isn’t the whole story?  Why, because reducing obesity will result in people living longer.  And, the longer you live, the more years you are around to incur health care costs.

Now, moderate obesity reduces life expectancy by 3 years and severe obesity reduces life expectancy by around 10 years.  So, let’s say that eliminating all obesity increased life expectancy by 5 years.  These years probably come when the beneficiary is on Medicare, and average annual Medicare spending is about $8000, so this would add $8000*5 = $40,000 on average to a person’s lifetime health care cost.

So, eliminating obesity reduces average health expenditure by $1429 per year but adds about $40,000 in health expenditure at the end of life.  So, it would basically take about 28 years of annual savings to offset the additional spending at the end of life.  Of course, there are many factors that must be taken into account, like the time-value of money,  changing technology, and the fact that when people live longer they also draw non-health benefits like Social Security for longer.  But, once you factor in that the obese people will live longer and make greater use of the medical system during that time, it is no longer so clear that reducing obesity will reduce overall health care costs.  In fact, here’s a study that looks at this trade-off (between lower costs and longer life) and concludes that while “obesity prevention may be an important and cost-effective way of improving public health, … it is not a cure for increasing health expenditures.”

Now, you may think that what I’ve said is absolutely nuts.  So, let me take a moment to defend myself.  Here are statements I agree with:  (1) reducing obesity would improve health.  (2) if a person loses weight and goes from being obese to not, they will on average reduce their annual health care expenditures.  (3) if a person loses weight and goes from being obese to not, they will on average increase their life expectancy.  (4) the U.S. should be making greater effort to reduce obesity, especially among the young.  But, based on the available evidence it seems that these goals, while desirable, will not result in lowering the overall cost of health care in the U.S.

The Tomato/To-MAH-to of the CBO report on the effect of the Senate bill on premiums.

Filed Under (Health Care) by Nolan Miller on Dec 2, 2009

The CBO released a report this week that attempts to estimate the effect of the new Senate health care bill on insurance premiums.  The report is potentially important because supporters of the bill argue that it will lower health insurance premiums, while opponents argue that it will increase them.  In fact, not only do the two sides differ on what the bill will do, they differ on what the CBO says the bill will do.  From the New York Times:

“’The C.B.O. has rendered a fundamental judgment that this will reduce the deficit and reduce people’s premium costs,’ said Rahm Emanuel, the White House chief of staff,”

and yet …

“’The analysis by the Congressional Budget Office confirms our worst fears,’ Mr. Grassley [R, Iowa] said. ‘Millions of people who are expecting lower costs as a result of health reform will end up paying more in the form of higher premiums. For large and small employers that have been struggling for years with skyrocketing health insurance premiums, C.B.O. concludes this bill will do little, if anything, to provide relief.’”

So, I went and check out the report, which you can download here.  It’s actually quite an interesting read.

As background, what is the Senate plan supposed to do?  Among other things, the bill prohibits insurers from charging individuals higher prices based on their health status.  This is a laudable policy goal.  But, it must battle against a phenomenon called adverse selection.  Adverse selection refers to the fact that if you offer an insurance policy for, say, $5000, the only people who will be wiling to buy it are those who expect to have relatively high insurance costs.  So, someone who thinks they will only incur $500 dollars worth of medical expenses will be less likely to buy the policy than someone who expects $20,000 worth of expenses.  Currently, insurers address this problem by charging a low price to the person who expects $500 in expenses and a high price to the person who expects $20,000.  However, the new bill prevents this practice.  Suppose the insurer responds by charging a single premium of $3000.  This will attract all of the people who expect high costs and few of the people who expect low costs.  The result is that the average cost of those who buy insurance may be much higher than the average cost in the population as a whole.

If the insurer can’t make money charging $3000, it may raise its price.  But, doing so only discourages healthy people from buying insurance even more.  The result is that the insurer may find that it is able to offer a policy only at a very high price, or maybe not at all.  So, mandating uniform prices may not, by itself, be a sustainable policy.

The Senate bill addresses this by also mandating that most people buy insurance.  Since most people without insurance do not have the option of buying it through their employers, this means that they will have to buy insurance from the non-group (individual) markets.  Some of the new purchasers of insurance will be those who are sick and could not buy insurance in the private market before.  Others will be those who are healthy and chose not to buy insurance before.  Whether the net effect increases or decreases premiums will depend on whether the new entrants to the insurance pool tend to fall more in the former camp or the latter.

A third goal of the Senate bill is to encourage competition in the private insurance market.  Through the creation of insurance exchanges that facilitate shopping for good plans and possibly the addition of a government option, the bill seeks to encourage private insurers to offer high-quality plans at low prices in order to attract new customers.

A final goal of the bill is to ensure that all insurance products offer a certain, minimum level of coverage.  In effect, the bill requires many policies currently being offered in the individual market to cover more services than they do currently.  In particular,  individual-market plans that participate in the insurance exchange will be required to offer benefits that cover at least 60 percent of typical health expenses, and they would have to cover additional services that are not typically covered by individual policies today, including “maternity care, prescription drugs, and mental health and substance abuse treatment.”  Since these services are not covered by many plans now, this requirement will increase premiums.

So, to summarize:  the new bill should be expected to change premiums by (i) changing the services that are covered by a typical individual-market policy, (ii) encouraging competition, and (iii) changing the group of people who choose insurance.  These three mechanisms operate fairly independently. In other words, covering more services would increase premiums even holding the level of competition and pool of risks constant; increasing competition would lower premiums even holding services and the risk pool constant; and changing the risk pool would affect premiums even holding the services offered and level of competition constant.  Further, since the large group market is already fairly competitive, has no risk rating, and takes all employees, we expect the effects to be concentrated in the individual market.

The CBO’s findings are summarized in the table below, which I copied from the report.


The first column of numbers represents the effect of the Senate bill on per-person premiums in the nongroup market.  Covering additional services is expected to increase premiums by 27 to 30 percent.  This is not surprising and should not be controversial.  If you want to cover more services, then it is going to cost more.  People are paying more, but they’re also getting more.  The next two rows in the first column show that the bill is expected to have beneficial effect on prices due to increased competition, lowering premiums by 7 to 10 percent, and increasing the size of the pool of those buying insurance will reduce premiums by another 7 to 10 percent.  So, the overall increase in premiums in the non-group market is estimated to be somewhere in the range of 10 to 13 percent.  The second and third columns show that the bill is expected to have negligible effects on the small group and large group markets.

So, who’s right, the Democrats or the Republicans?  Well, there’s a grain of truth in what both sides are saying.  The bill will increase premiums for those in the nongroup market.  But, this increase is due exclusively to giving them more generous policies than they currently purchase.  Holding fixed the level of benefits, the bill lowers the cost of insurance, which means the bill is achieving its goals.  So, it seems to me the debate should be centered on which services we want to mandate.   If it is worthwhile to cover services like maternity care and substance abuse treatment, then it might make sense to mandate that all policies cover such services and to incur the extra cost of covering them.  (Although we might also ask, if the market wants these services to be covered, why does the market not currently cover them?) The current debate, however, is focused rather superficially on whether the bill increases or decreases premiums, which misses the point that the product is changing as well.

Incidentally, let me mention the Democrats’ other point, which is that even though the bill increases premiums in the nongroup market overall, many people in this market will be eligible for government subsidies to help offset the additional cost.  So, while total premium will go up, total out of pocket cost (premium net of the subsidy) will go down for the majority of people in the nongroup market.  However, this point is a bit disingenuous as well, since these subsidies are coming from taxpayers somewhere.  So, while these individuals may be paying less for their premiums, the cost of the whole system is certainly increasing.