A Trend Will Continue Until it Can't

Reading an op-ed piece, that wasn't bad by any means, in the Washington Post inspired a desire to ask for some sense in discussing health care costs.







From Forbes


Scary graph and all, but does anyone really think this is sensible? Continue this graph out and some time in the 2100s health care spending will exceed 100% of GDP.

Rather obviously, this doesn't make any sense. The current trend won't continue because it can't, whether or not we have health care reform.

This isn't to say that reform isn't necessary, that we don't need more doctors, or that we shouldn't be worried about the problems. However, the sharp trendlines and high cost increases are just noise, they simply aren't possible. If Medicare spending gets cuts few doctors will be able to cease seeing Medicare patients, for many doctors those over 65 make up the majority of their patients. You can't make any money by not seeing anyone. In the end, what we are discussing with health care reform is who is going to be paying for forced changes, not whether or not current trends will continue.

I think this is critical to keep in mind when discussing health care. There is no solution where everybody wins, current trends, and thus current power arrangements, cost structures, and winners and losers, cannot be the same because the current trend is unsustainable. What we are debating is who wins and who loses. This is why the debate is so fierce, no one wants to be the loser. This is also why everyone should be very skeptical of claims made and aware of framing, everyone has some skin in this game so expect dishonest presentations from just about everyone involved in the health care debate.

Examining Walker's Claims Regarding the ACA in Wisconsin

Scott Walker made some claims about how the ACA will impact Wisconsin that I found rather surprising. So I did the sensible thing, I read the report he supposedly got the data from.

First of all:

100,000 people will be dropped by their employer-sponsored health insurance;

According to the cited report, "there will be 10,000 fewer Wisconsin residents covered via ESI due to the effects of the ACA. This represents less than a 1% decline in ESI enrollment."  (p 14) Walker seems to be referring to he figure on p. 14 which shows firms dropping coverage, individuals voluntarily leaving, and new firms gaining coverage. To assess the ACA obviously the net effects are the relevant figure, isolating one number is cherry picking. It should also be noted that it is likely that many of those voluntarily leaving their employer sponsored coverage for the exchange feel their options have been improved by the reform. No one has seriously claimed that there won't be winners and losers from this policy change (or any other), isolating the downside when the net effects are clearer and more intuitive is a poor way to present data.

Observations on Public Opposition to Health Care Refrom

I've noticed something of a disconnect among liberal commentators on why many Americans are opposed to health care reform. Living in a more conservative area, and having a lot of contact with individuals of all income levels, I've heard a fair amount about why people don't like Obamacare.

On an anecdotal level, the main thing I'm hearing is that Obamacare does little more than create another bill to pay for people that are already having trouble making ends meet. Sure, they will have healthcare now, but probably on a high deductible plan. This doesn't help their pocket book issues any, if they have a major illness, they don't have the assets to get through any time out of work, not to mention the health care deductibles and coinsurance. The stark reality they're facing is that they were ruined before Obamacare if they got sick, and they'll be ruined after if they get sick (now, they'll at least receive better medical treatment, but it's still a bitter pill if they lose all their assets in the process).

What these people want help with is the costs of the kind of medical care that they can afford and have more contact with. They want cheaper regular check ups, cheaper minor procedures, like setting their kid's broken bones, and less regular medical expenses, like their monthly health insurance bill. Obamacare doesn't really deliver any of this, so they remain opposed. They see it as increasing their monthly expenses when they're already nearly broke, doing nothing to reduce the cost of the medical care they do seek out, and not being strong enough to protect their assets when they do get gravely ill.

There is a lot of truth to these criticisms. Obamacare does do a lot more to make sure doctors and hospitals get paid than it does to make costs lower for the average working class American (various sub-categories, like people with existing health conditions, are thrilled about the health care law, but for the general, healthy, lower and middle class there are much more mixed feelings). However, I believe there is a sequencing issue here. I don't see any way to have a properly functioning marketplace without certainty of payment. I'll write more about this later, right now the system is like a game of hot potato, everyone is trying to pass the bill along and not be the one stuck with it when the music stops. To get the changes that the average American wants, this aspect of the system needs to be eliminated. Obamacare is a significant step towards eliminating the systemic problem, though it necessarily leaves the need for later reforms to get at the pocketbook issues that concern the average guy.

Bundling and the Importance of Standardization

I have done more than a few posts discussing how concentrated medical expenditure in the US is.  Ezekial J. Emmanuel does a good job discussing one way to focus on this problem, bundling payments.

What I want to bring up in particular, is this:

The last big barrier to switching from fee-for-service is to get Medicare as well as the other insurers to change how they pay in unison. It is extraordinarily difficult for doctors and hospitals to change how they practice if they are partially paid via fee-for-service and partially through bundling — especially if the bundles differ between insurers. The incentives pull in opposite directions — fee-for-service to deliver more services for sick patients, bundled payments to be efficient and keep patients healthy. But if Medicare and private insurance agreed to coordinate a switch to bundled payments, doctors and hospitals could follow suit. This will not be easy, but it is absolutely necessary if we want to save real money and improve care.

This brings up one of the big issues for people that like to argue health care should be more decentralized.  The lack of standardization in payment systems is one of the big inefficiencies in our health care system.  Bundled payments aren't the kind of reform that is likely to happen spontaneously across the US, it takes someone with coercive powers to make all of the actors move together.  A decentralized system is highly unlikely to be able to institute the kinds of reforms that we need on a broad scale.  While individual institutions can pursue this, it only takes one powerful actor, such as a hospital group, to prevent this kind of reform in an area.  The reality of the economic system is that in complex markets like health care, a number of actors serve as veto holders.  To get them in line requires a centralized coordinating body.

While this example is specific to health care, it can also be applied more generally.  Some systems require greater specialization and competition, others require greater coordination and standardization.  Areas requiring standardization virtually always require a degree of coercion to be able to reach an efficient equilibrium.  Properly identifying which is which is a critical challenge for every society.  Insisting that all situations are one or the other is just plain unhelpful, both kinds of problems exist and require an evolutionary case by case response.  Ideologues, either those that insist that everything requires competition and unique solution or those that insist everything is reducible to basic principles to be determined by centralized planning (rare today but a significant historical viewpoint) both threaten one of the fundamental tasks of society and social evolution.

The CLASS Act and the Individual Mandate

The CLASS Act failure displays all of the main reasons that the individual mandate is necessary.  The problems with providing long term care and the difficulty of convincing currently healthy people to sign up for the plans are the exact same problems the individual mandate is needed to solve.  Long term care, like most conditions you need health insurance for, is simply beyond the ability of most people to pay for.  Under current legislation, private plans are often ineffective at dealing with these problems, too many people have insurance with payout limits or that might be terminated if they lose their job due to their illness.  Many conditions that afflict young people, such as certain cancers or mental illness such as schizophrenia, are chronic and require long term treatment.  How can anything but a mandate bring in enough money to pay for the long term costs of the costliest patients, those who were young and healthy but unexpectedly acquired a long term illness that will impact the rest of their lives?

Now, as a separate issue, we definitely need some form of long term care plans in the US.  Home health aids are generally far more efficient than nursing home care, altering Medicare and Medicaid rules to help people stay in their own homes would be a good first step forward (this isn't to ignore scandals like home health aide programs in NYC, even with these problems this approach is more cost-effective in aggregate).  More comprehensive long term care would probably be easier to get off the ground once existing programs are altered to support it, CLASS was getting ahead of itself.  It's also notable that most long term care plans offered in the private sector fail for the same reasons the government plan did.

Nice Little Story on Medicaid Fraud

For the record, most studies of Medicare/Medicaid fraud put the actual amounts at quite a low percentage, though raw numbers can look quite large considered that Federal Medicaid outlays were $204 billion in 2008 (not sure what the number would be counting state outlays).  However, one of the most frequently abused aspects of these programs is home health aides.  There was a bit of a scandal regarding NY City payments a few months back.

Back on topic, Fox News has a nice little story about a veteran who investigated and tracked down a multimillion Medicaid fraud.  Just a neat story, no wider implications.  Well, beyond noting that considering that there is a percent or two of Medicare/Medicaid fraud and abuse we obviously can't rely on individuals to report it all since stories like this are uncommon enough to be newsworthy.  Programmatic changes are needed.  Still nice to see an individual making this much of a difference though, even if there are no systemic implications.

Brief Clarification on my Earlier Post

After some brief consideration I decided my post yesterday on Brooks' health care column was unusually rambling.  I thought I would briefly lay out why I thought this, without wading into the more complicated argument.

To put it succinctly, Brooks seems to be basically repeating Hayek's argument in the "Road to Serfdom" about centralized planning vs. market decision making.  This is accurate as far as it goes.

There are two general problems I have with this thinking generally, which can be extended to health care.  The first problem is that Hayek's argument is essentially ahistorical.  In my opinion, the idea that the government just needs to get out of the way is starting the story of market formation in the penultimate chapter.  There's quite a bit of research on early markets, these show that marketization is a highly complex, negotiated process (for an example I recently read, Perdue's China Marches West has an excellent section on Chinese efforts to get private traders to play a greater role in supplying settlements in the steppe territories.  The Chinese have known for many centuries how much more efficient markets are, but are more honest about the difficulties of getting markets to work.  The process is anything but spontaneous.  Also, while not as explicitly, he details how other social factors compete with economic growth, such as the need to secure the northern provinces requiring non-market intervention to attain horses which the Ming could not obtain in sufficient numbers through private traders (though they still fell short with non-market expedients as well), partially due to under-monetization, a problem that wouldn't be resolved till the Qing despite strenuous efforts, and how the inability to attain these horses undermined efforts to contain nomadic raiding which weakened the northern economy (not that this stopped traders from obtaining high quality horses for sale elsewhere, the benefits were accrued by the merchant and the end purchaser but ultimately at a high cost to many others as the lack of horses for military response depopulated northern areas and weakened overall economic growth).  This is obvious with a more developed theory of the process of market exchange, markets are little more than a communication device that relies on a mutual understanding (more technically, a set of constitutive norms) of what information the price signal conveys and a mutual agreement to leave out extraneous information.  For most purposes, once a market has developed, intermediate institutions (such as the state, but also including ethnic groups and other social institutions depending on the market under discussion) can largely get out of the way and the market will continue to be self-sustaining.  At this stage, once the constitutive norms have been set, then the further extension of markets exhibits spontaneous collaboration.  But this is the late stage, after the heavy lifting has already been done.  It doesn't occur on its own in isolation.  This is easily seen by looking at primitive societies, absent state intervention trade and economic relations look nothing like modern markets.  It is only after a great deal of work has been done that market cooperation occurs.  Health care is a peculiar area where for a number of reasons the normal spontaneous transitive process of marketization did not occur properly as it did with most other sectors.  This requires an outside entity to step in and properly set the rules of the game, and to bring the players into line to equalize power disparities, so that a functioning market can be formed.  While this may happen on its own with time, the time we're talking about is a very long period of time, which does little for people living today.  I don't much care if left to its own the various parties involved will reach an optimally efficient level of organization in a 100 years or so, I can't wait that long (assuming this would happen at all, which I doubt).

The second issue is that for economic reasoning to work, economic assumptions of human behavior have to hold.  This is a major problem with health care.  Home economus is a pretty good description of human behavior as it applies to economic transactions.  It is not a good description of how humans form personal relationships, how we act when scared, our desires for personal security, or how identity is formed.  This is normally widely accepted, aside from some people with too much specialized education, most people realize that human relationships and political concerns are only partially connected to our economic actions and interests.  We have multiple identities that only have partial interaction with each other.

Health care being a big exception to this (other areas, such as land ownership, do display this to some degree, this is related to what I mentioned above about market formation, markets in goods that functioned similarly to modern markets significantly predated free markets in land for this very reason, and since health care displays even more of these properties if a free market in health care is the ultimate teleological development then we should expect that it will be an even longer, more drawn out process than markets in land and labor were and to display an even greater need for interventionist policies to get to that penultimate stage where the state can get out of the way).  Health care impacts on too many other areas of human interest that fall outside of neoclassical, utility maximizing assumptions.  It impacts identity, we here a lot about the basic dignity of human life with health care, we hear a lot less about this with labor or land markets and not at all with luxury car markets.  We hear of personal security, we never act as economic maximizers in regards to physical security if it involves someone with a gun, we shouldn't expect to do so with regards to cancer either.  Family plays a much larger role here than with most economic decisions, with the big expensive stuff the individual is not going to often be the best unit for analysis as it is for most economic transactions.  I could go on quite a bit, but the general point I am making is obvious, there are too many aspects of human life impinging on this sector for normal economic assumptions to hold.

This leads to why the state has to play a larger role.  In modern society, the state has assumed most of the roles that other social actors once played in responding to non-economic actions of the people making up the society.  There just isn't any other entity that can respond to anxieties about the impact on the family (such as an elderly person using up what would otherwise have gone to their children on medical care, often because they are no longer able to make a decision themselves and family members authorize a spare no expense approach individually to signal the value of family and family solidarity to others), buyers remorse regarding care received (people regretting authorizing aggressive treatment rather than hospice is extremely common, when emotionally overwrought it's simply a human reaction to over-respond and to regret it later, the state can pick up on this but market interactions by their nature are limited to the time of interaction between contracting parties), etc.  There is just too much evidence that the health care sector is not displaying the properties of modern markets to think that state withdrawal would make it function according to these assumptions better and too much input from aspects of human nature that do not conform to neoclassical assumptions for the frame of centralized vs. decentralized decision making to hold properly.  Applying neoclassical assumptions to the health care sector is simply inappropriate giving the social institutions we are currently living with.  At some point the state may be able to get out of the way, but that point is very, very far off yet.  What the comparative evidence keeps saying is we're just not there yet.

Winning the Argument by Arguing in the Abstract

David Brooks' column this morning annoyed me.  In it, he argued that there is a fundamental difference between Democrats and Republicans in that Democrats believe in centralized control and Republicans believe the world is complicated and imperfect necessitating a market based approach.

This is nonsense.  At this level of abstraction the Republican viewpoint is the only logical viewpoint.  Put this way, other than a handful of hard-left bozos out there absolutely no one believes this.  When the decentralized market based approach works, it always works better than central planning.

But this isn't really the frame from which the Democrats are arguing health policy, try as hard as Republican leaning pundits might want to move this onto an abstract philosophical plane where they know they will win.  The main Democrat critique is one of practical, empirical policymaking vs. this kind of soft theorizing.  The simple fact is that health care systems that have heavy government intervention, such as the rest of the developed world, score overall much better and at much lower cost.  We actually have a close experiment of attempted privatization of the health system in the Chinese reforms, they had to abandon the market based approach due to poor results, even relative to the sorry state of the system before the reforms.  They exhibited startlingly similar flaws to the US as well, such as overinvestment in high technology of marginal medical value.  From this perspective, it seems readily apparent than any positivist has to insist on the superiority of the state based schemes, whether or not this is explicable through theory.

After Some Delay, the Positive Aspects of that Post on Medicaid I Mentioned the Other Day

So, while I had a negative reaction to Scott Walker's thoughts on the existence (or lack thereof) of sovereignty in the subordinate units in a federal state, I do think he does a very good job of pointing out some of the major flaws in the Medicaid program.  In particular I thought this bit was very perceptive of the problems with Medicaid, I run into these all the time at my job:

"In recent years Washington has taken an obsolete program, which covers health care for low-income Americans, and made it worse through restrictive rule-making that defies common sense. It is biased toward caring for people in nursing homes rather than in their own homes and neighborhoods. It lacks the flexibility to help patients who require some nursing services, but not round-the-clock care."

He is also quite correct in pointing out that:

Time and again states like Wisconsin have blazed the path in Medicaid — from giving individuals greater control over their care to expanding the use of electronic medical records — while the federal bureaucracy has lagged behind. Just now Washington is discovering accountable care organizations (networks of doctors and hospitals that share responsibility for caring for patients and receive incentives to keep costs down) and “medical homes” (a model in which one primary-care doctor takes the main responsibility for a patient).

He is also quite correct in pointing out that (there are a few other things in here I agree with, these are the big things):

We need to modernize not only Medicaid’s benefits and service delivery, but also its financing. In good times, the open-ended federal Medicaid match encourages states to overspend. Amazingly, the program is now viewed by some states as a form of economic development because each state can at least double its money for each dollar spent. That matching feature penalizes efficiency and thrift, since a reduction of $1 in state spending also means forfeiting at least one federal dollar, often more.

I am in complete agreement that these issues are two major flaws in the current Medicaid system.  Medicaid's bias towards institutional care is a major issue and something that comes up a lot in the disability field.  Institutional care is vastly more expensive than home care and the more we can go towards it the better.  I also see cost shifting all the time, the current Medicaid model has different cost share schemes for different services so in many cases the more expensive option is cheaper for the state administering the program.  I think most of this is handled well enough in the op-ed that I don't have to go into more detail (though I may pick it up in later posts, if I ever get around to looking at New York's Medicaid reform many of these issues will come up in more detail).

More on New York and Medicaid

Following up on the last post, which got heavily side tracked about two minutes into writing it requiring a change of course, this post is on what I originally intended to discuss.

First home home care:

Care delivered at home by health aides trained to handle minor medical tasks grew by 90 percent per recipient from 2003 to 2009, to reach more than $15,500 a year. Personal services delivered by nonmedical attendants, such as assistance with dressing and bathing, grew by 38 percent, to reach almost $30,000 a year. Right now this care is mostly delivered on a fee-for-service basis with hours of care determined primarily by the providers who have an incentive to run up the bill with more and more services.

 If I was at work I'd have an excellent article to quote from on this, however I don't have it so this will be more general.  An issue I have with this is that there is a huge discrepancy between these kind of benefits through most of the state and these benefits in New York City.  The article I alluded to looked at 24 hour nursing care in the rest of the state compared to New York City, New York City had more than a magnitude difference in proportion of people receiving 24 hour nursing care relative to population compared to the rest of the state.

There is of course a good reason for part of this, care in facilities is higher in New York City than in the rest of the state so there is a larger group where the marginal cost of providing 24 hour care is still lower than the cost of care in a facility.  Still, this doesn't explain the entire discrepancy by a long shot.  So an examination of how these programs are run could easily find savings to be had by simply making the local bureaucracies adhere to in state best practices (though it should be noted that areas in the rest of the state would probably benefit by providing 24 hour care more frequently since this would often be cheaper than institutional care, savings can be had on both ends).

The Market and Health Insurance Sales Across State Lines

The New York Times this morning had a short piece on New York Medicaid cuts.  It's a decent overview, there are just a few bits I'd like to add my 2 cents to.  This post is a bit of a tangent, it deals with a concrete reason for why I oppose cross state sales of health insurance, given current institutions.  New York is the state with the most generous state health benefits so it provides an excellent example for how an integrated market forces democratic decision making up to the national level because decision making done at the state level imposes decisions made in one state on others, which is arbitrary and undemocratic from the point of view of the other 49 states.

To frame this, the high cost of long term care in New York, and that mental illness is a portion of this.

In 2009, the latest year with audited data, New York covered more than 300,000 patients — the elderly, chronically ill, patients with mental illness, substance abuse and other difficult problems — at a cost of $23.1 billion or almost half of the state’s total Medicaid spending that year.

And this is essentially why I'm opposed to cross state sales of health insurance.  New York has chosen to be the most generous state in the nation as far as medical benefits go. Whether or not any individual agrees with this, it is our democratically chosen decision.  As a result of this, mental health parity laws are a very big deal here, they mean that private insurers pick up some of the tab for those patients with mental illness rather than the state, without this those with expensive conditions would fall back on the state for benefits while private companies only paid for the cheap stuff (or at least for this one condition, though this could be extended to other areas of health care New York is generous in).  Many states do not have mental health parity laws and their private health plans offer little, if any, benefits to patients with mental illness, which results in throwing them back on state services. 

Letting insurance be sold across state lines would pose major problems to a state like New York.  Maintaining the level of services to the mentally ill that we currently provide would become far more difficult if people started opting for HSAs that were cheaper in states without mental health parity and then developing schizophrenia in their 20s or 30s (I'm choosing mental health for this topic because the population it impacts is the one most likely to opt for inexpensive health insurance with minimal coverage, which is problematic because it's also one of the most expensive long term conditions).  This isn't so costly for those states that give less benefits, for us, it would put additional strain on our mental health system, which is already under substantial strain with or without the budget crisis.

Some Lessons from a Needle Injection Pilot

Excellent article (and some must see accompanying photographs in the linked page) on a drug injection site project in Vancouver.  Articles on these places always interest me, the evidence is overwhelming that they bring multiple benefits and no one has been able to show significant downsides to them (assuming they are put up in already rough areas, I can see why people wouldn't want them in an area with little existing crime or illicit drug use).  Yet, in most places, the public and the government hates them.  While I haven't read anything to confirm this, there are two reasons for this that seem intuitive.  First, they make problems more visible that people prefer to ignore and have left to back alleys and dingy rooms rather than publicly funded clinics.  Second, they are very challenging ideologically to anyone that takes a tough on crime stance.   That these things work and have a growing body of literature to support this is tough when the tough on crime stance and incarceration enjoys little data to support its effectiveness.  If more data continues to come in showing these things work, especially if they affect drug usage rates as well (which this article doesn't go into), it brings a lot more into question than needle injection site.

A second observation from the article is that it illustrates a theme I keep coming back to with my health care posts.  That theme is the idea that the fractured nature of payments in the US is one of the key drivers of health care costs.  Since in Canada all health care is paid for by the government these problems don't exist.  AIDS patients are very expensive to treat and can have high externalities for the rest of society.  From the government's perspective almost the entire externality is captured, it bears the costs of treatments, the lost taxes due to illness, costs imposed by new infections, public order costs from drug use outside the clinics, political pressure from fears of a rising AIDS rate, etc.  In the US, no single actor feels the effects of all of these externalities.  A private company may bear the cost of treatment, but it doesn't see a decline in revenues nor does it feel public pressure by a rising AIDS rate and likely does not bear the cost of an additional infection, it certainly doesn't feel any responsibility for the public order issues of drug injection.  The same can be said for the US government, it doesn't necessarily bear the full cost for each infection since the private sector is responsible for some of the payments.  While AIDS and drug injections are an unusually clear case of this, it's illustrative of why I think only the government is positioned to fully bear both the costs and benefits of health issues and is the only economically efficient actor to bear these costs, due to the number of costs and benefits that cannot be internalized by other economic actors.

The article contains some striking information on the public health issues involved, one striking number given was this:

Even $50 million spent on drugs, he said, ultimately saves $300 million because roughly 400 people a year avoid infection. (The estimated lifetime cost of treating a Canadian with AIDS is $750,000.)

 A very powerful argument in favor of these treatments, if they can be backed up, but one that only partially applies in the US, where in many cases the government won't be baring the costs, assuming that the drug addicts will rarely receive retrovirals and the people getting treated are those infected when it spreads into the wider community (through many channels, including prostitution, but also other channels as well).

Another striking statistic specific to HIV infections:

A 1997 study in The Lancet found that in 29 cities worldwide with needle exchange, H.I.V. infection dropped 6 percent a year among drug injectors, while in 51 cities without, it rose by about 6 percent.

Too Sick to Fail (to be treated at massive taxpayer expense)

As you no doubt know by now a Virginia judge has ruled the provision of the health care reform law that requires that people buy insurance unconstitutional.  If this ruling stands, which it probably won't, this makes health care reform basically impossible without constitutional reform. 

It comes down to a problem that since the public isn't willing to simply let sick people die in the gutters and requires that emergency rooms treat patients regardless of ability to pay, everyone already has their health implicitly backed by the government.  Let's call it Too Sick to Fail (to be treated at massive taxpayer expense).  Of course, since this treatment is reserved for those that have become far too sick to be treated efficiently or, in many cases, to work to pay back even a fraction of their medical expenses, this becomes a very costly medical subsidy for those that choose to recklessly speculate on margin that their physical condition will stay sound enough that they can forgo spending on health care now to raise present consumption.  They can do this knowing that if things go too badly they can rely on the government, and those moved with sympathy by their sob stories, to provide a level of medical care sufficient to prevent truly disastrous consequences.

Given that this implicit government and community backing has proven time and time again to exist, it simply makes sense to acknowledge its existence and make those covered by the implicit government subsidy pay for this backing and face the costs of it themselves, either through obtaining health insurance or through a national taxpayer funded plan, rather than being able to gamble that they can enjoy more consumption today knowing that if they do get sick they'll be shielded from the worst consequences by the tax dollars paid equally by those that bought insurance, who do not rely to the same extent on health care provided at public cost.  This would have a further advantage in providing the means to pro-cyclically monitoring a person's health and perhaps catching a disease before it became both more costly to treat and more likely to prevent later recuperation of the costs through future income by the patient.

Or we can simply bury our heads in the sand and try to rule modern problems unconstitutional in an effort to avoid having to come to terms with the fact that our favored solutions can't fix modern problems.

Nursing Home Care: A Specific Failing of Our Health System

An excellent article in the NY Times today covers the topic of middle class families using spousal refusal to pay for nursing home care.  A few things stand out, first average care costs of over $100,000 a year.  Surely, there must be a market for more mid-range care, while some patients undoubtedly need this level I can't imagine that this need is universal.  There seems to be an opening for research on whether market failure is occurring, potentially because of programs like Medicaid and long term care insurance, of if there is something else going on, insurance costs leap to mind.

The second thing that jumps out at me is that this is a prime example of how our mixed system of health care creates market failures.  In this case, to qualify for Medicaid an individual can retain:

The federal government allows a healthy spouse to keep a house, a car, up to about $2,700 a month in income and up to about $110,000 in other resources. Anything above that must be spent on nursing care before Medicaid kicks in.

This system obviously discriminates heavily against savers.  Anyone that wanted to fully fund their retirement would be basically wiped out by costs of $100,000 a year or more.  If someone is unfortunate enough to have a loved one get a condition that requires long term care then trying to take responsibility and pay for it will rapidly deplete any savings they had, of course if they had been spending the money from the start they wouldn't have this problem.

I don't really see how this problem is solvable without universal coverage.  Conditions that result in this situation are rare enough, and private options such as long term care insurance uncertain enough (I've seen a number of articles on how these plans are constantly having problems and having to shut down, or make other changes), that it's unlikely someone could adequately prepare in advance.  With universal coverage this problem can be dealt with more effectively.  Either its paid for by universal premiums to insurance companies, which would then have a huge incentive to look at and press for reform at long term care facilities, or it is paid for universally through taxes.  As it is today, there is no good way to distribute costs, people that try to pay privately get wiped out and those that don't suffer these misfortunes resent having this paid for by the public purse rather than by the individual suffering the tragedy, who does have significant assets that could partially pay for the care.

At the end of the article a reform effort within the existing system is mentioned:

Lawyers within the New York State Bar Association who represent the elderly have proposed a compromise system. It would allow people to negotiate a payment plan with the state, in which they would pledge to pay a “fair share” of their spouses’ nursing home costs — perhaps half their assets — while keeping enough for their own living expenses, Mr. Krooks said. There are plans to try out a version of this system upstate.

 While better than what we have today, half of someone's assets seems high, especially when Medicare gives so much better coverage to people suffering from diseases requiring hospital care instead.  I continue to have the same thoughts on the subject I expressed earlier in my posts on medical care, we all run the risks of acquiring these conditions and very few of us possess the ability to pay for them, therefore the only way to pay for these conditions is through universal cost sharing.  Anything else will lead to perverse incentives, such as the byzantine bureaucratic process described in the article with suits, counter-suits, and settlement negotiations becoming an expected part of life for individuals who are likely undergoing some of the most traumatic times of their lives due to the conditions their loved ones are suffering from.

Another Example of the Dangers of Overzealous Budget Cutting

The first to fall in times of tight money are often the most innovative programs.  The NY Times has an excellent article from a writer with The Bay Citizen on a program named Caduceus which seeks to provide services to people who are severely mentally ill and homeless.  The program relies on donated time and since its population is unable to meet even very basic eligibility requirements, in some cases things as simple as filling out forms, has difficulty accessing funding.  It is however very cheap due to its structure.  Despite this, the program is unable to secure funding to maintain itself and will be wound down withdrawing services from these individuals.  More traditional service systems will supposedly fill this need, whether they can effectively remains to be seen.  It is likely these other programs will however cost more than one that relied on donated time.  Programs like this are always the first to go when the calls are for budget cuts, rather than for reform which tends to have better long run impacts on both programs and budgets.

Bad Labor Habits and Medical Errors

Just finished reading a NY Times piece on medical errors.  It was good and it's worth reading the whole thing.  A couple of quotes jumped out at me though.

Dr. Wachter said: “The study is telling us how hard improving safety really is. Process changes, like a new computer system or the use of a checklist, may help a bit, but if they are not embedded in a system in which the providers are engaged in safety efforts, educated about how to identify safety hazards and fix them, and have a culture of strong communication and teamwork, progress may be painfully slow.”

Earlier in the article however:

“A third of the errors in the intensive care unit disappear when residents work 16 hours or less,” Dr. Landrigan said, but noted that senior residents are still allowed to work longer.

Not having people work 16 hour days isn't hard, it's common sense.  Of course there will be a few situations where it's beneficial but this is an instance where I see a cult of hard work having significant negative effects in the real world.  I am certain it's not the only one.  We need to stop talking about the hard work that goes into a 16 hour day as a good thing and start condemning it as reckless behavior.  While the ill effects are clearest in medicine this is simply forcing people to do things we're not built to do.  We need to build systems that work with people as they actually are, not as we wish them to be and to stop treating reckless behavior as something that should be honored.  It reminds me of the bankers justifying their salaries through the long hours they work.  My first reaction was that perhaps the system wouldn't have crashed if the lot of them had been getting enough sleep and weren't making stupid errors due to fatigue.

So What Should We Do About Health Care?

The last few posts left it unclear exactly where I think we should go with health care.  The honest answer is that I don't think it matters that much as long as we get universal coverage.  Once that is in place I think the various actors in the debate will have their incentives lined up in a way that will over time lead to cost controls.  The problem is essentially structural in that it has to do with the power, and diversity, of the various groups involved far more than it has to do with the technical aspects of providing care.

That said, I think there are some clear areas where there are real possibilities for reform that would save costs.  The first of these is to debate whether we want to move to fairly comprehensive health care as the norm, or if we want more limited access.  In either model, catastrophic coverage must be given, and have it clear who is providing it (having the government as a backstop for insurers is the problem, this exists today) for the system to work.  Beyond this, we can have a reasonable debate about how much we are willing to pay to increase access.  Given the disparity between low and high cost patients it is also essential to understand that expanding access is unlikely to do much to increase costs, between Medicare and Medicaid we are already paying for the most expensive conditions.  But there is a real debate to be had over whether we should be paying for routine payments out of pocket, or through individually purchased health care, or if the advantages of early detection and consistent follow up for more cost sensitive patients are worth the additional costs to the system

The second major issue is what limits we wish to impose on care provided from universal insurers.  Right now, it seems pretty easy to continue to get treatment well after the possibilities of success are low.  Should people really have the option of having the government pay for 2nd and 3rd line cancer treatment or is it reasonable to limit the publicly provided funds to the treatment with the highest chance of success, to give one example.  At some point a line will have to be drawn, the question is where.  My take is that everyone deserves at least one shot at treatment, after that I don't think there is a public responsibility to continue to provide public treatment for the same condition (I'd also hedge this with numerous exceptions, such as cost benefit analysis.  If a second treatment isn't expensive no reason to stop it.  Also if its a chronic condition and there is a high long run payoff to successful treatment it should also be paid for; some form of expert panel to weigh these issues, with feedback from the public, is necessary.).  Though a responsibility for hospice remains.

The third question regards supplementary insurance.  This being America, I think the answer is easy.  Whatever the minimum level of insurance required we should always have the option of purchasing more.  This should not bar us from getting the public version as well to prevent perverse incentives.  This should be a less regulated market but if we're not going to cover 2nd and 3rd line cancer treatments, or cover routine doctors visits from the other side, we certainly should allow a market in supplementary benefits.  Those that favor high degrees of equality are not infrequently against this, I ran into it often enough in Canada, but here I think there is a very strong argument that whatever the government offers those that want a Cadillac plan should be able to buy it.  I certainly don't think this should be subsidized, however.

At the end of this, I will admit that I simply don't know enough about the system to make specific recommendations.  I do think there are compelling reasons to approach the problem from a more systemic approach of how health care needs to look in the long run to fulfill its purpose.  Without doing this as well as looking at specific reforms that are possible today I don't think its possible to carry the public along with reform proposals or to deal with the perverse incentives that got us to where we are today.  Its not just a matter of specific flaws in the current system, it's also about fundamental problems with the system itself.

The Essential Problem with Health Care

I tend to believe that the primary systemic issue driving US health care costs is the interaction between the very concentrated nature of health care needs and the very fractured system of paying for it.  This gives every individual payer a massive incentive to attempt to pass on those with high health care needs to another payer rather than to seek ways to reduce costs for those people with high needs.  I'm not saying that fixing this problem will do anything on its own to reduce health care costs.  What I am suggesting is that this problem creates incentives that prevent fixes from being made.  Unless these incentives can be fixed health care reform to lower costs will be impossible. 

The only tested way to fix this problem is universal health care.  Reform proposals that avoid this don't appear to me to address this structural issue.  The essential reason for this is that without universal health care uncertainty as to whether a procedure will be payed for enters the system.  If there was a reasonable certainty that those without coverage that need treatment wouldn't be treated than it would be reasonable to expect that a non-universal system could work.  But as long as needed treatment can still be expected to be received whether or not someone can pay it is reasonable to expect that there is no real possibility of a proper market ever functioning in a way that will control costs (though it can function to raise capital and expand the system, the only part that seems to be malfunctioning is cost control, which seems to me predictable as a result of needed treatments being covered anyway).

With the payment system how it is, any individual entity can best achieve cost control by maximizing its responsibility for the 50% of people with low costs and minimizing its coverage of the 5% responsible for almost half the costs.  This is simply a much bigger reduction in costs than trying to deliver care most efficiently for a given mix of patients would be.  This problem has led to a vast array of ad hoc laws seeking to plug various gaps in the system.  I know that in New York one passed over the past year that would prevent health insurers from canceling entire insurance policies to avoid covering individuals with very high costs.  Laws such as this exist in every state and massively increase the complexity of regulation (and in my opinion, the size of the state) and thus the inefficiency of our health care system.

Public opinion on these issues unsurprisingly displays a high degree of cognitive dissonance.  We get extremely upset over individual instances of insurance company abuse, such as rescinding entire policies to deny one person coverage, yet get very upset over the high cost of insurance due to covering people that actually cost significant amounts of money to treat, like the kid whose insurance coverage got rescinded by the insurer.  At the same time, we expect to be covered ourselves if something ever goes horribly wrong.  The costs we face day to day in health care are just so disconnected from the high costs of treating serious conditions that it is difficult to keep the message focused on paying for complex medical needs rather than how much it costs to get a bottle of penicillin.  Which provides those that profit from expensive health care plenty of room to muddy the debate by taking the focus off the high cost sections of the system to instead focus on the low cost sections we experience day to day that are the source of all their profits.

Concentration of Health Care Expenses in the US

It took me a little longer to get to this post than I expected, it's been a busy week.

I was looking for a report covering the issue in some detail that was not behind a pay wall.  The High Concentration of U.S. Health Care Expenditures by MW Stanton* fits the bill nicely.  I've seen this data in various places and it is all fairly similar, all the facts below will be taken from this report.  I believe the essential problem driving US health care problems, and really problems throughout the western world, is that health care costs are so concentrated that there is a serious disconnect between people's experiences and the actual costs of the system.  There is a basic distrust rooted in the fact that we don't see where the dollars are going to in our own lives.  This disparity comes clearly in the statistics.

• 5% of the population accounts for 49% of expenses (2002)
• 34% of those in the top 5% of spenders in 2002 where also in the top 5% for 2003 
• Those over 65 were 43% of the top 5% of spenders, this isn't a solely age related phenomenon
• The 15 most expensive conditions account for 44% of expenses
• The lower 50% of spenders account for only 3% of expenditures

While this data doesn't lead us to particulars about how to fix our current system it does tell us what some of the systemic issues are.  Mostly, it tells us that what people need insurance against is becoming part of the top 5% (or top 10% which isn't much better).  It also tells us that insurance needs to cover us for several years in the top part of the distribution. Further, for an individual insurer costs are always controlled most effectively by seeking to insure the lower 50% of spenders which still represent a very large pool.  For the insured however, we never know if we're going to be part of that 5% tomorrow, which is what we're trying to insure against.

It should also be noted that with costs so disparate there are differences in what each of us needs the health care system to do for us.  A system that is most efficient for paying the health care expenses of the 50% of the population that only spends 3% of the health care dollars isn't going to be the same as that which pools the resources of the other 95% to pay for the unlucky 5%.  For most of us, the most efficient system is likely no system at all, or health savings accounts or other mechanisms that can be used to pay the low day to day expenses.  However, if something goes wrong than we will quickly become very worried about how the system works for the upper 5%.

This does of course leave open the possibility that separate systems for basic and costly care could be constructed, perhaps with the government, or government regulated insurers, providing mandatory catastrophic coverage with an lightly regulated system for the rest of our health expenditures.  I haven't heard this proposed though so won't spend time on it.  I'll take up some more issues related to this in the next post.

*Stanton MW. The High Concentration of U.S. Health Care Expenditures. Research in Action, Issue 19. AHRQ Publication No. 06-0060, June 2006. Agency for Healthcare Research and Quality, Rockville, MD. http://www.ahrq.gov/research/ria19/expendria.htm

Health Care vs. the Health Care System

I think it can be conceptually useful when discussing the problems that we have with health care in the US to distinguish between health care and the health care system.

Health care has been around as long as recorded human history and likely far longer.  No particular system was required for health care for most of our history, the market or other naturally occurring institutions were sufficient for providing most primitive health services.  The small exceptions would be the simple hospice care provided by early hospitals or institutions such as leper colonies.  But for the most part, little was required beyond the skill of the individual physician and some participation from the local community.

A great deal of modern health care is similar.  Delivering primary care and many health services, such as hospice and a range of simple surgical procedures, probably could be delivered effectively outside of a broader health system.  There are of course advantages to embedding these services in a broader system, and drawbacks, but for this rather broad (and here ill-defined) range of services there is no need for some overall health care system.

However, modern medicine has also developed a number of interventions that are very resource intensive and require some kind of dedicated system to keep running.  I don't have the knowledge to give an exhaustive list of these interventions but it is difficult for me to see how medical issues such as emergency care, especially when emergency treatment leads to a need for intensive care, could possibly function without a system in place to make sure that the hospital can almost always recoup its expenses without necessarily knowing if a given patient will in fact be able to pay.  While this system doesn't (theoretically) necessarily require government intervention, in practice existing systems have always involved the state. 

I believe this distinction is necessary to make because of how it effects the political discussion.  Most of our interaction with the health care system is with the parts of the system that don't really require a system at all.  This gives a massive political advantage to those that want to keep the system as weak as possible.  It isn't hard to think of reforms that can improve access and there is a quite wide array of ways to make it easier for someone to see a doctor and to easily pay for something simple like antibiotics, setting a broken bone, child birth, or simple outpatient surgeries.  These areas can no doubt be improved by reforms such as improved access or could be easily paid for by greater use of health savings accounts or other free market oriented reforms.

The hard choices all lie in the areas where a health care system is required that most of us have little to no interaction with through most of our lives.  How do we build a system that can pay for treating autism?  How can we pay for cancer treatment or organ transplants?  How can our system create the right mix of specialists and general physicians when years of education can be a decade or more?  This is where we need to focus our attention if we want to build a working system.  People are afraid of the possibility that a medical condition may bankrupt them or that they can't provide a newborn child with the care it needs.  However, dealing with this is difficult because they are angry about the day to day costs of the medical system and the frustrations and delays they run into with the general health care needs that are part of their daily lives.  Solving this requires acknowledging that people's fear and anger are pulling in different directions and that all the hard choices have to be made regarding the systemic issues that are the source of their fears.  Going after the little stuff that causes anger just derails the conversation and prevents us from moving forward.

I'll expand on this idea a bit more tomorrow with some numbers to back it up.