Something Not Unlike Research

Like a cricket match in hell

Kevin Outterson has a great post today on how industries that market junk food to kids are opposing the US federal government's draft guidance on a voluntary program restricting the advertising of unhealthy foods to children. If you are concerned about childhood obesity, read it. My title is from his description of the industry's strategy in this fight:

But lobbyists also play long ball, games that last years, even decades. Like a cricket match in hell.

Specifically,

For more than a decade, the Washington Legal Foundation has filed legal cases and briefs advancing the “commercial speech” theory of the First Amendment, an idea that was unanimously rejected by the Supreme Court in 1942. But today, commercial speech is a profound de-regulatory tool, striking down regulation of tobacco ads, pharmaceutical data mining, liquor stores, and other products. Millions of dollars, invested strategically over many years, is now bearing fruit as judges use “commercial speech” to strike down public health rules.

And now they come for childhood obesity.

So I was curious, and took a quick look at two legal briefs written by well-known law professors, arguing the case for the junk food manufacturers, or the "Sensible Food Coalition," as they style themselves. Several arguments caught my eye.

First, they have a reasonable argument, that there should be no regulation unless there is good evidence that advertising unhealthy food to children contributes to childhood obesity. The authors, however, seem to think that there is conclusive evidence that advertising does not have such an effect. I'm skeptical, and hope to look at the data they cite soon.

What was most striking to me, however, was the preposterously strident civil libertarian tone of these arguments. Federal guidelines on marketing junk food to children are, apparently, the top of a slippery slope to "manipulating citizen behavior through the selective suppression of speech advocating lawful action." Each notes the obvious counter example, which is that the pornography industry is not allowed to market its products to children. I had always thought that such limits on pornographers were a good thing, but Professors Redish and Sullivan offer no reason I can find why we shouldn't similarly free the pornographers from the oppression of tyranny.

 

Canadians need to control health care costs too

  My good friend Nils Finne raised a question in the comments to this post about how well Canadian are doing at controlling health care costs. The answer is: not well. The chart is made from data presented by Towers Watson. The vertical axis is the growth in national medical spending minus the general inflation rate for the country (or average of countries). Canadian cost growth has been greater than US or an advanced country average over the last five years.

Canada is, of course, increasing health care costs from a lower base level of spending compared to the US (but not compared to other advanced countries). As someone who has spent a career inside the US health system, things look lean here. Nevertheless, if these data are correct, these are truly dangerous rates of growth in cost. 

eHealth technology enhances active self care, but is not a substitute for it

post by Bill Gardner

Here is a cool new eHealth technology: a bone cast with wireless telemetry of muscular activity. It is

threaded with electromyographic (EMG) sensors that track muscle activity around a fracture site. Created by Pedro Nakazato Andrade, a recent graduate of the Copenhagen Institute of Interaction Design, the cast can relay data wirelessly to physicians and physical therapists. It shows how much mobility a patient has in an affected area and how physical therapy and daily activity are affecting the area, allowing clinicians to adjust treatment accordingly.

The key thing this device does is measure "how physical therapy and daily activity are affecting the area." It is not, itself, a device with therapeutic activity. More to the point, it does nothing if you are not doing your physical therapy -- which is most of us. eHealth will fundamentally change health care only in the context of interventions that successfully engage patients in self-care. 

I've overstated the point somewhat. My motivation to adhere to my physical therapy regimin will likely increase if that I know (or at least believe) that that regimin will work. So if the smart cast makes therapy more effective, I'll be more likely to do it.

The problem is that eHealth technology is often presented as if it were a pill: you put it on and you get better. Rather, you put it on, and your self-care becomes more effective. And getting patients to effectively and consistently care for themselves is the hardest problem in chronic care. 

Jimmy Connors was once asked, what are the most effective exercises to improve your tennis?Connors: "The ones that you will actually do."

Canadian Patient-Oriented Research

post by Bill Gardner

Although the Canadian health system differs greatly from the US system, it faces many similar problems:

Canada faces a dual challenge in the research-to-practice continuum, often referred to as the two "Death Valleys" in the health care landscape. Valley 1 refers to the decreased capacity to translate the results of discoveries generated by basic biomedical research in the laboratory to the bedside or careside as well as to successfully commercialize health discoveries. This negatively impacts Canada's clinical research and knowledge base and its international competitiveness. Valley 2 refers to the limited capacity to synthesize, disseminate and integrate research results more broadly into clinical practice and health care decision-making. These two valleys must be bridged if Canada is to bring evidence to bear to enhance health outcomes and ensure a sustainable health care system.

In response, the Federal Health Minister, Leona Aglukkaq, has issued a new Strategy for Patient-Oriented Research. Among the key objectives:

• To strengthen organizational, regulatory and financial support for clinical studies in Canada and enhance patient and clinician engagement in these studies; and
• To improve processes for the early identification of best practices, expedite their development and harmonization into guidelines for patient care and support their adoption by clinicians, caregivers and patients.”

This similar to the strategy for comparative effectiveness research (CER) promoted by the US Patient-Centered Outcomes Research Institute (PCORI). Many conservatives oppose the PCORI, which was established as part of President Obama's Affordable Care Act. I've never understood why, but I urge them to reconsider their views on PCORI and CER. The Canadian strategy, after all, is the policy of the Conservative government of Prime Minister Stephen Harper.

A post idea culture?

post by Bill Gardner

Neal Gabler in The New York Times:

Ideas just aren’t what they used to be. Once upon a time, they could ignite fires of debate, stimulate other thoughts, incite revolutions and fundamentally change the ways we look at and think about the world... It is no secret, especially here in America, that we live in a post-Enlightenment age in which rationality, science, evidence, logical argument and debate have lost the battle in many sectors, and perhaps even in society generally, to superstition, faith, opinion and orthodoxy. While we continue to make giant technological advances, we may be the first generation to have turned back the epochal clock — to have gone backward intellectually from advanced modes of thinking into old modes of belief.

This just seems wrong. Here are some enormous ideas that are transforming our culture:

• Physics. E.g., the multiple universe ideas canvassed in Brian Greene's most recent book.
• Climate science. Lovelock, the global warming scientists. There is a bigger practical question than the fate of the climate?
• Neuroscience, and in particular the neurobiology of emotion.
• Artificial Intelligence. The internet, and Google, and so on are a series of experiments that escaped from a lab.
• Statistics. The widespread availability of analytical tools, skills, and data means that many people are acquiring an empirical mindset through learning by doing.
• Social evolution. Dawkins, of course, but also the many more recent theorists who are advancing evolutionary explanations of social cooperation.
• Moral and political philosophy. Parfit says we are just starting to understand how these work in a post-theistic world, and he is right.
• Applied economics. Health economics, the economics of poverty, growth, and development.

You will have noticed that I am citing disciplines instead of authors. There is a long term shift toward a scientific and engineering culture. This means that ideas are being attacked in a disciplined, serious, and cumulative way by specialists. These specialists, however, are connected to broader communities of discussion; in part through traditional media, and increasingly by the social software that Gabler disdains. Through that diffusion, that are catalyzing the kinds of intellectual shifts that structure and shape the way in which collectively we interpret our experience and our world.

 

Carroll and Fried on Physician Rationing

post by Paul Kelleher

I am preparing syllabi for two upcoming courses in which rationing and, specifically, physician rationing will be discussed. I won't rehearse it here (since you should go read it), but this post by Aaron Carroll will be required reading in each class.

In perusing my files for other readings, I came across an old paper by Charles Fried, a long-time Harvard Law School professor and Reagan's Solicit0r General. Fried's paper closes with this memorable quote that echoes Carroll's sentiments:

The doctor in his dealings with his patient must not act like a bureaucrat, policy maker, or legislator. But policy makers, voters, and legislators must think in different terms. It is monstrous if an individual doctor thinks like a budget officer when he cares for his patient in need; but it is chaotic and incoherent if budget officers and voters making general policy think like doctors at the bedside. 

Much more on these themes in due course.

ADDENDUM: I should note that Carroll does not suggest that it would be "monstrous" if individual doctors engaged in rationing (though he may believe that; I don't know). His agreement with Fried is that someone other than doctors should be tasked with saying "No." 

Are people ready for Personal Health Records?

post by Bill Gardner

A Personal Health Records (PHR) is a record of your medical and health data that you access and maintain yourself. In theory, PHRs should empower patients to better understand and manage their health care. Google Health, however, was unable to build up a sufficient user base to make it viable even for a very well funded and patient company. This may be one reason why:

This week, I talked with Dan Russell, a search anthropologist at Google, about the time he spends with random people studying how they search for stuff. One statistic blew my mind. 90 percent of people in their studies don't know how to use CTRL/Command + F to find a word in a document or web page! I probably use that trick 20 times per day and yet the vast majority of people don't use it at all. "90 percent of the US Internet population does not know that. This is on a sample size of thousands..." 

It may be some time before PHRs can make a significant contribution to population health.

 

Taylor's Second Law of Health Policy

post by Paul Kelleher

Friend-of-the-blog Don Taylor recently released his e-book Balancing the Budget is a Progressive Priority on Amazon. Five of the book's twelve chapters are extremely rich discussions of health reform which alone justify paying the book's $4.50 pricetag.

Taylor begins the first chapter on health reform thus:

When teaching my introductory health policy class at Duke University, I begin the first day by describing two "laws" that I believe govern all health care systems:

Everyone dies

Before that, the healthy subsidize the sick

The first point is a fact. The second implies a choice (or many choices). The essence of health policy is working out how the healthy will subsidize the sick.

I am interested in the moral and conceptual issues connected to Taylor's Second Law and its corollary that health policy is fundamentally about how the healthy will subsidize the sick.

Since he's a social scientist, one natural interpretation of Taylor's Second Law is that it is merely an account of what in fact happens in rich nations. But Taylor seems to mean more. In giving the historical example of healthy hunter/gatherers bringing food back to the sick, Taylor likens this to "Modern health policy financing" which "is just another example, a modern extension of this basic principle of being human." This suggests to me that Taylor's Second Law is in part a normative proposition--it describes how things should be, not just how they are or have been in societies bygone: people should take care of other people, and in particular, healthy people should take care of sick people. Anything less would be inhumane (if not inhuman).

However, if we are now in the realm of normative health policy, then I believe the Second Law is too slender a reed to be of much use. Actually, a better metaphor is that it is too thin a thread to be of much use. For Taylor argues that his Second Law

is not unique to health care, and in one sense it is just the practical definition of insurance...There have to be those who pay premiums and who do not make claims for the insurance market to remain solvent and continue to operate.

I have already explained in a previous post why moral lessons for social insurance policies cannot be teased out of the features of free market insurance contracts. While Taylor is right that individual insurance policies involve cross-subsidization from the healthy to the sick, the reasons for entering into them are purely self-interested. Those reasons cannot, on their own, support the kind and degree of cross-subsidization found in Britain's NHS or in what Fuchs and Enthoven refer to as the "quasi-social insurance" that most Americans enjoy in the form of community rated plans available through their employer. And yet Taylor claims that the common thread running through the NHS, U.S. employer-based insurance schemes, and individual insurance contracts is his Second Law. Unfortunately, there is simply no cogent interpretation of that thread that could transform it into a principle that vindicates the social insurance state.

So Taylor's Second Law needs to be supplemented by other normative principles that could justify forcing some to pay for the care of others. Here's one that Taylor doesn't formulate explicitly: the rich should subsidize the poor and/or sick. I won't defend that proposition here, except to observe that if we discoved Bill Gates had a serious chronic disease, we would not conclude that he should no longer have to pay taxes to support Medicaid. But then this would be a case of the sick subsidizing the sick, a departure from the Second Law. 

If we had an argument that the rich should subsidize the poor and/or sick, then this could do a great deal of work in normative health policy. For example, instead of being an argumentatively inert common thread running through real world examples, the principle that the rich should subsidize the poor/sick could actually provide a basis for critiquing certain familiar health policies. Here is Mark Pauly doing just that with respect to the quasi-social character of employer-based insurance:

Formally, community rating can be thought of as financing a subsidy to insurance for higher risks (there being potential social value to the subsidy) through an excise tax on insurance sold to lower risks. The economics just reviewed tells us to avoid specific excise taxes in favor of more general taxes. The alternative subsidy/financing model outlined above does just that; there is no reason to think that higher income taxes would appreciably affect the insurance purchasing plans of anyone, regardless of risk level. From the viewpoint of economic efficiency, subsidizing high risks is often desirable, but paying for that subsidy by taxing low risks never is. Nor is there an obvious equity reason why low risks should pay. The reason for the political appeal of this policy is that the tax on low risks is not counted as a tax, does not show up on any budget, and is opaque for all, even to low risks who are likely to blame the insurers, not the regulators, for their high premiums. (pdf, p. 26)

Even if one disagrees with Pauly's preference for less government involvement than the Affordable Care Act will exert, his rationale for critiquing employer-based insurance could be embraced by proponents of single-payer. For many single-payer advocates would like to see it financed by progressive taxation, just as much of Medicare is. But that would amount to the rich(er) subsidizing the poor(er). Since Taylor in fact endorses progressive financing of many policies he favors, perhaps this is reason enough for us to speak of Taylor's Third Law of Health Policy?

Finally and most importantly: read the book.

Search Friction and Adverse Selection

Today Austin Frakt introduces many of us to the notion of "search friction", i.e. "inefficiencies in one’s ability to wisely choose, to be a savvy shopper." Search frictions can occur when there are many plans on offer whose complexity makes comparisons between them an "apples to aardvarks" enterprise. When there are multiple plans on offer, search frictions can be a source of market power for insurers, since they enable firms to charge above competitive prices--i.e. the prices that would prevail under perfect information and understanding.

In response to Austin's post, Harold Pollack tweeted: "Search friction can be helpful to reduce adverse selection. Easy mobility & full information w/o risk adjustment is disaster." Harold's point, I take it, is that without search frictions--i.e. when individuals do have full information and thus can be savvy shoppers--shoppers will enroll in plans that in fact work best for them. So sick individuals will enroll in plans that will take care of them, thus driving up the premiums. This gives healthy people reason to avoid those plans, leaving the enrolled population sicker and causing premium rates to rise. As Harold notes, a risk adjudgment mechanism is needed in this scenario to avoid a death spiral that'll make the plan too expensive for anyone to afford.

All this reminds me of a great anecdote from David Cutler's book Your Money or Your Life (p.98):

In response to these selection concerns plans prefer to prevent high-risk people from enrolling...They can spend large resources on well-baby care (to attract healthy women) but cut back on care for the mentally ill. They can raise the cost of sharing on syringes and insulin to make it more difficult for diabetics to be in the plan. (In Germany, one plan wrote a letter to its diabetic members pointing out that another plan was far superior on diabetes care; the plan that was noted to be better complained.)

 

 

Physician Profit and the Rule of Rescue

post by Paul Kelleher

Austin Frakt recently blogged about a really cool study on how payment systems affect physicians' providing behavior:

In a controlled setting, the researchers asked medical students to choose the quantity of medical care to provide to hypothetical patients enrolled in either fee-for-service (FFS) or capitated insurance plans. Under the former, physicians are paid for each additional unit of care. Under the latter, they receive a lump-sum payment independent of units of care provided. In the experiment, quantity of care is an amount indexed by the integers 0 (no services) to 10 (the most services). In advance of making the quantity selection, the physician has full information about how the quantity selected will affect payment, costs, and profit (or income) and how it will benefit the patient.

Although this is an experimental context, the participants' decisions involved real money: the participants get to keep the profit associated with their chosen allocation, while the benefit that is afforded to each hypothetical patient is given to a real charity. The authors of the study remark that "This feature of our experimental design implements an incentive for the physician to take the patient benefit into account."

The authors present their summary results in the graph below. Red circles indicate the average choice made in the fee-for-service scenario, black triangles indicate choices under the capitation scenario, and the black dots indicate the level of services--regardless of payment system--that are optimal for the given abstract patient:

Austin says he is not surprised that how physicians get paid and how much they get paid affects their behavior. And in general, I am disposed to believe that too. (There is a good deal of supporting evidence for that view; here is a short brief by Jacobson and Newhouse that offers a useful real world example of the general phenomenon.)  Still, I find these particular results shocking. After all, this is a highly stylized scenario in which participants are told explicitly that (for example) the optimal level of services for patient 1E is 5, and yet medical student participants who must chose allocations under the fee-for-service decision scenario wind up on average giving that patient level-8 services. In doing so, the participants know that level-5 services yield a clinician profit of 5.10 Taler (the experimental currency) while level-8 services yield a profit of 11.60.

Austin notes that in the real world, doctors are never so sure about which level of services is optimal for a patient. And I think he is right that this uncertainty may well lead doctors to rationalize levels of services that also happen to be in their finanical interest. If this is right, then we might expect doctors to be even more profit-driven in the real world where uncertainty is the norm. But this of course doesn't explain why providers were so brazenly profit-driven in the experimental setting, where everyone involved knew that level-5 services were optimal for patient 1E.

One thing that would explain the behavior observed in the experimental setting (and something Austin mentions) is the fact that the patients are abstract (i.e. not identifiable) and that the monetized benefit afforded to the patient go to a medical charity. This hypothesis is supported by a recent study showing that many physicians prioritize gravely ill identifiable patients who cannot meaningfully be helped over statistical ones who can be helped much more. 

In "Rule of Rescue or the Good of the Many?" Kohn et al. presented clinicians and nurses with vignettes designed to elicit their preferences for allocating their last ICU bed to an identifiable cancer patient with little chance of survival or to a deceased or dying patient for whom aggressive management could help others through organ donation. Respondents were randomly assigned vignettes differing with respect to the aggregate number of life-years to be gained by organ recipients (either 5 or 30) and with respect to whether the donor was dying or dead according to a neurological or circulatory criterion of death.

The authors report that "more than one-third of ICU clinicians forewent substantial social benefits [to organ recipients] so as to devote resources to an individual patient unlikely to benefit from them." They also note that clinicians' decisions were not much affected by the magnitude of social benefit (i.e. whether organ recipients received a total of 5 or 30 years of extra life) or by the method of determining that the donor was dead. Finally, qualitative analyses revealed that the most common reason for giving the last ICU bed to the identified patient "was that clinicians perceived strong obligations to indentifiable living patients."

Kohn et al. conclude that the so-called "rule of rescue" may present a "substantial and persistent barrier to the efficient allocation of scarce resources and may impede efforts to improve triage or constrain costs." And this is something health system reformers will want to explore and study further. Moreover, the "rule of rescue" phenomonon may offer a good explanation of why the medical students displayed their specific degree of profit-seeking behavior in the highly stylized experimental context of the first study. It may turn out that doctors' natural interest in their own bottom line can be sufficiently contained when they're dealing (1) with an identifiable patient, and (2) with reliable data about optimal quality that simply does not yet exist in the real world.

(Edited slightly for clarity.)