American Journal of Law & Medicine

The patient life: can consumers direct health care?

Then said Evangelist, "If this be thy condition, why standest thou still?" He answered, "Because I know not whither to go."

John Bunyan

The Pilgrim's Progress


The ultimate aim of health care policy is good care at good prices. Managed care failed to achieve this goal through influencing providers, so health policy has turned to the only market-based option left: treating patients like consumers. Health insurance and tax policy now pressure patients to spend their own money when they select health plans, providers, and treatments. Expecting patients to choose what they need at the price they want, consumerists believe that market competition will constrain costs while optimizing quality. This classic form of consumerism is today's health policy watchword.

This article evaluates consumerism and the regulatory mechanism of which it is essentially an example--legally mandated disclosure of information. We do so by assessing the crucial assumptions about human nature on which consumerism and mandated disclosure depend. Consumerism operates in a variety of contexts in a variety of ways with a variety of aims. To assess so protean a thing, we ask what a patient's life would really be like in a consumerist world. The literature abounds in theories about how medical consumers should behave. We look for empirical evidence about how real people actually buy health plans, choose providers, and select treatments.

We conclude that consumerism, and thus mandated disclosure generally, are unlikely to accomplish the goals imagined for them. Consumerism's prerequisites are too many and too demanding. First, consumers must have choices that include the coverage, care-takers, and care they want. Second, reliable information about those choices must be available. Third, information must be put before consumers, especially by doctors. Fourth, consumers must receive the information. Fifth, the information must be complete and comprehensible enough for consumers to use it. Sixth, consumers must understand what they are told. Seventh, consumers must be willing to analyze the information. Eighth, consumers must actually analyze the information and do so well enough to make good choices.

Our review of the empirical evidence concludes that these prerequisites cannot be met reliably most of the time. At every stage people encounter" daunting hurdles. Like so many other" dreams of controlling costs and giving patients control, consumerism is doomed to disappoint. This does not mean that consumerist tools should never be used. It means they should not be used unadvisedly or lightly, but discreetly, advisedly, soberly, and in the fear of error.


   Who is in charge of the clattering train? 
   The axles creak and the couplings strain, 
   And the pace is hot and the points are near', 
   And sleep hath deadened the driver's ear, 
   And the signals flash through the night in vain, 
   For Death is in charge of the clattering train. 


The United States spends too much for the health care it receives. Yearly, the problem worsens. (1) More of the GDP (16%) goes to health care than in any comparable country without buying appreciably superior health. Medical goods and services cost more than in similar countries. (2) Their prices rise faster than inflation. And perhaps "more than 25 percent of all health care spending is either entirely unnecessary or is of only questionable or marginal benefit." (3) Who is in charge of this clattering train?

Everyone wants the train subdued. Governments worry because they pay for so many people's care. Employers worry because the more they pay for insurance the less they can pay in wages. Employees worry because they defray ever larger shares of medical costs. We all should worry because while insurance appears to insulate us, ultimately we pay--as taxpayers, as employees, as consumers--for what we receive. And even while we worry about costs, we worry also about quality. Controlling the train means reducing costs while improving quality.


Lawmakers, employers, health plans, and hospitals have long tried to stop the clattering train. But costs rise irrepressibly because irrepressible forces push them. Technology, for example, can be marvelous, but it can also be costly. (4) And specialization combines with technology to require expensively bureaucratized medicine.

The financial structure of medical care also accelerates costs. Doctors and hospitals commonly charge fees for services. The more services, the more fees; the higher the fees, the higher the income. Doctors decide what patients need. Doctors are professionals, and "commercial" practices are undignified and displease professional associations, so historically, "[p]rice competition was disavowed as unbefitting a learned profession and as inappropriate for a vital service." (5) Thus "medical professionalism created a spiral of expanding capacity, technology, utilization, and cost. Fee-for-service payment to physicians and cost-based reimbursement to hospitals rewarded extensive and complex care. Physicians recommended more visits, more tests, and more procedures." (6) So even early in the twentieth century, costs climbed. (7)

The clattering train was once slowed by want of fuel. When shallow-pocketed individuals paid, costs could only go so high. Then came the deep pockets. Employers (stimulated by tax incentives) insured employees. Government (stimulated by social duty) insured the old and the poor. Insurers paid "usual and customary" rates--essentially, what doctors and hospitals charged. Thus fueled, the train sped up. Robinson puts it neatly:

   Physicians and hospitals were motivated to provide ever more 
   and better services, since higher costs generated higher revenues. 
   Patients were motivated to demand more and better services 
   since the costs were shifted onto insurers and thence to 
   employers and taxpayers. Employers were motivated to expand 
   the breadth and generosity of insured benefits to capture the tax 
   subsidy. Medicare and Medicaid viewed their covered services as 
   a statutory entitlement for beneficiaries ... and were eager to 
   accommodate physician interests so as to mitigate the traditional 
   hostility of organized medicine. Everyone was saying yes and no 
   one was saying no to the expansion of utilization, specialization, 
   and expenditure. (8) 

So the train rattled along briskly. But perhaps costs were justified by the quality of care? Alas, doctors were presumptively the only qualified judges, and the "profession systematically refused to discuss or divulge quality concerns to nonphysicians." (9) However, unsettling hints abound. For example, hospitalization and surgery rates have "varied widely across geographic regions, with the best predictor being the number of hospital beds and practicing physicians; the prevalence of disease was not consistently associated with the frequency and intensity of care." (10) And "[r]esearch studies, authoritative entities such as the Institute of Medicine (IOM), and the insurers' own claims data reveal widespread overuse, underuse, and misuse of services, compared with evidence-based norms and clinical guidelines." (11)

Employers and governments particularly have labored to slow the train. Their best success has been managed care. It wants for patients what consumers shop for--low prices and high quality. It assumes that providers drive the clattering train and that they can be persuaded to slow, steer, and stop it. Managed care stems costs by limiting patients' choice of providers and treatments and by inducing providers to economize. Inducements have included "oversight and report-card keeping on physicians' prescribing practices, rules about the use of primary care physicians as gatekeepers or their elimination altogether ... evidence-based medicine ... and requirements that doctors gain permission for expensive procedures." (12)

Managed care has had successes in controlling costs, but it has been savaged by patients who felt they were losing control of their medical care and by doctors who felt they were losing control of their work. (13) So while managed care's success has been limited, managed care may not have failed; it was from its mother's womb untimely ripped.

Managed care seeks good care at good prices by influencing providers. Many of its practices survive in some form, but the top-down, supply-side principle has been pummeled. That left a bottom-up, demand-side principle: Why not treat patients like consumers? Won't they select the care they need at the price they want? Won't the market spur competition that expands consumers' choices while constraining costs? This "consumerism" is today's watchword. (14) As two writers said fifteen years ago, "[I]f we want to solve the nation's health care crisis, we must apply the same commonsense principles to medical care that we apply to other goods and services." (15) As the former HHS Secretary said last year, "We have a better option, to provide beneficiaries with reliable information about the cost and quality of their care. When given that kind of information, we know that consumers will make decisions that drive costs down and the quality up." (16)


Consumers search for goods and services of high quality at low prices. Crucial among services is medical care, but few people shop for it like consumers, if only because of insurance. (17) This is pleasant, but the dismal science warns us that when consumers want the best and ignore price, costs soar like the lark ascending. And when providers bill for each service, prices soar, well, like the hawk ascending.

Of course, patients have been buying medical services for years. Managed care actually accelerated the movement toward consumerism and competition, if only by hastening the organization of doctors and hospitals into groups that competed for patients. Thus Professor Sage could write years ago, "The last two decades have witnessed an extraordinary shift toward reliance on competitive forces to reshape the American health care system, which previously had been characterized by professional decisionmaking, consumer deference, and price-insensitive insurance payment." (18)

In any case, patients now make more purchasing decisions, and more consumerist proposals are aborning. These proposals vary, (19) but they use similar methods, make similar assumptions, and raise similar questions. We therefore evaluate consumerism through an ideal type. In this ideal type, consumers make three principal purchases. First, they choose their health care plan. Like good consumers, they decide which plan best suits their wants and purses. As consumers buy plans, competition for their custom generates splendid plans at sensible prices.

Second, consumerism expects patients to choose doctors, hospitals, and other providers considering both quality and cost. Consumerist patients ask: Which providers offer inexpensive services? Which providers work with patients to control expenditures? Which providers treat patients decently? Which providers are competent--or truly excellent? Here, again, consumers already make some such choices--they may choose doctors, for instance. (20)

Third, consumerist patients buy tests and treatments. (21) In its strong form, consumerism has people spend their "own" money, not "insurance" money, to buy treatments, and hence patients choose thriftily. Few people can afford all the treatment they might need, so consumerism assumes insurance would pay "catastrophic" costs (that is, consumers get insurance with high deductibles).

Most people can't afford high deductibles either; they teeter too close to the economic brink to find thousands of dollars in their pockets. (22) So the tax code now lets employers create tax-sheltered "health-savings accounts" in which unspent funds accumulate from year to year. To qualify for the tax shelter, deductibles must range (currently) from $1,150 for individuals to $11,600 for families. (23) Typically, employers put much of the deductible into the employees' account. Nevertheless, consumers spend their own money in several senses. First, they contribute to the account. Second, the employer's contribution is effectively part of the employee's salary. Third, consumers can use unspent money that accumulates (tax-free) for most health care expenses and can distribute it in their estates.

Parts of the consumerist agenda are already operating. Many insurers have adopted consumerist devices. "Deductibles are increasing, copayments for physician office visits and prescription drugs are increasing, and health plans are increasingly more likely to provide incentives for beneficiaries to use generic drugs and/or mail order pharmacy services, and other forms of tiered benefits...." (24) Already, "40% of primary care physicians are providing care to CDHP enrollees," and this "may be an underestimate because it does not account for those physicians who are unaware of their patients' CDHP coverage." (25) In addition, "It]his policy approach has been strongly adopted in the Medicare program, where older adults typically face choices among 50 or more prescription drug plans as well managed care options." (26)

So surely consumerism's time is come. Not only are its three parts old hat; both sides of the ideological spectrum can like it. The right can like using markets to tame prices and ensure quality. Consumerism simultaneously fits neatly with the standard principle of the leftish "bioethics" side of health law: Offering patients control over their medical care is the fons et origo, the alpha and omega of bioethical law. (27) That law works mainly through mandated disclosure--by requiring patients be given good enough information to make their own decisions. (28) The paramount example is informed consent, but it is only one of many programs that--like consumerism--aspire to "empower" patients by giving them information and thus the power to choose what they want. (29)


This article undertakes three enterprises. First, consumerism is the latest of many battered attempts to control costs while optimizing care. Will it work? Will it quiet the clattering train? Our second enterprise is evaluating the regulatory mechanism of which consumerism is an example--legally mandated disclosure of information so that one party to a transaction can make good decisions. Our third enterprise is assessing the crucial assumptions about human nature on which consumerism and mandated disclosure depend.

The consumerist agenda provokes deja vu all over again. Many plans have been laid to tame costs, improve quality, and expand control. Taming costs fails with doleful regularity. Improving quality is horribly hard. And the principal legal attempts to give patients more control routinely disappoint. (30) When plausible policies repeatedly fail to achieve excellent goals despite exceptional efforts, there are usually high, and sometimes insuperable, barriers to reaching the goals. Is it thus with consumerism?

Consumerism operates in a variety of contexts in a variety of ways with a variety of aims; its success thus depends on a variety of things and must be variously measured. All this variety is inevitable, if only because consumerism cannot simply be ordained. Rather, the federal government has used tax policy, its influence as an insurer, and its power as an employer to promote consumerism and all that may still be too little. Nor have commentators agreed on what consumerism is and how it should work. So "consumerism" in health care is various both in action and in books, and it has been defended and attacked in various ways.

To assess this Proteus, (31) we scrutinize consumerism's distinctive and crucial feature: purchasers choosing well. We use a fresh technique--we systematically ask what prerequisites must be met for consumerism to succeed and whether they will be met. Our answers rest on empirical evidence about how people buy health plans, choose providers, and select treatments. When there is no such evidence, we consult empirical research on comparable decisions.

More specifically, we ask what a consumer's life would really be like in a consumerist world. What would the human experience of consumerism actually be? Would people behave as required? The literature abounds in theories about how these medical consumers should behave. We look for evidence.

This brings us to our article's second enterprise. One of the law's favorite devices is to require a more knowledgeable (and thus presumptively stronger) party to provide information to a less knowledgeable (and thus weaker) party. (32) Securities laws are an ambitious example. Courts and commentators debate what disclosures should be required in boilerplate contracts. Product liability law invites manufacturers to disclose dangers their goods pose. Credit card companies must reveal their terms to card users. Truth-in-lending acts are supposed to make borrowers good judges of lending terms. Miranda warnings give suspects information to use in dealing with police. And there are hundreds of other examples.

Mandated disclosure and consumerism operate similarly. Their success depends on how well information can be assembled, disclosed, received, understood, analyzed, and used. This is why research on mandated disclosure helps us understand consumerism, and why our technique for studying consumerism helps us evaluate mandated disclosure as a regulatory method.

So to our third enterprise. We conclude that consumerism, and thus mandated disclosure generally, are unlikely to accomplish the goals imagined for them. Consumerism's prerequisites are too many and too exigent. Consumers make many purchases with aplomb. Most things consumers acquire they buy regularly and judge easily. Those products are sold in national markets in which competition works its Smithian magic. But health plans, providers, and treatments are chosen sporadically, are hard to assess even after the purchase, and are often sold in small and stodgy markets. So health care choices are harshly more demanding than most other purchases, as Kenneth Arrow famously said long ago. (33)

Consumerism and mandated disclosure, then, rest on suppositions about how people respond to incentives to buy unfamiliar products wisely. They rest, that is, on assumptions about human nature. Those assumptions are generally unexamined and generally misleading. They fundamentally misunderstand human nature. First, they misunderstand how people make decisions and over-estimate their willingness and ability to do so. Second, they misunderstand how people want to lead their lives; they forget that people have better things to do than becoming model consumers.

Our review of how people in a consumerist world would choose health plans, providers, and treatments, then, concludes that at every stage people would encounter daunting hurdles. Like so many dreams of controlling costs and giving patients control, consumerism is doomed to disappoint. This does not mean that consumerist tools should never be used. It means they should not be used unadvisedly or lightly, but discreetly, advisedly, soberly, and in the fear of error.


   Between the idea 
   And the reality 
   Between the motion 
   And the act 
   Falls the Shadow 

T. S. Eliot

The Hollow Men

We now begin our project of evaluating consumerism by systematically reviewing its prerequisites. The first prerequisite is that consumers need alternatives adequate to the variety of purchasers and circumstances for each purchase: plans, providers, and treatments. Will this happen? Sometimes somewhat, but often little.


Some prominent consumerists--Havighurst is a distinguished example propose that costs may be moderated and patients may better match their expenditures to their preferences if they can purchase health insurance the way they purchase other services. The consumerist agenda will thus wither if consumers lack adequate choices. Yet many consumers have no such options and faint prospect of getting them.

Many people "have no real or perceived choice in their employment, health insurance, or health plan." (34) Even larger employers in larger cities may offer employees only a few choices, and few employers offer many. For example, "in 1999 approximately half of individuals covered by employer-sponsored plans had no choice or limited choice of health plans." (35) And "Robert Berenson ... contends that 'only a few large and prominent employers are redesigning employer contribution formulas to encourage value purchasing, contracting directly with provider-based delivery systems, or using quality data ... in any meaningful way to select plans.'" Medium-sized employers "'do not even seem to understand the difference between the price of particular provider services and their aggregate cost ..., much less to be able to separate fact from fiction on quality measures.'" (36)

Even consumers offered several plans may have scanty choices. This insurance comes in prix fixe packages; you can't choose & la carte. But you will rarely be offered the meal you want. Buying insurance is not (to change the metaphor) like buying a car, where you can take a showroom model or design your own farrago of options. Insurers cannot afford to let people buy only coverage they know they will need. Nor must competition inspire variety; on the contrary, competitors may converge around packages likeliest to attract buyers.

Employers might even dislike long a la carte menus. Employers don't want under-insured and hence under-treated, bankrupt, and bitter employees. The more options consumers have, the likelier they are to choose improvidently, and pressure to economize drives employees toward underinsurance. Under-insurance matters, especially since the distance between lavish and stingy is wide. Under-insurance can wreck financial and physical health. Costs of illness (particularly medical bills and loss of income) contribute to more than half of personal bankruptcies. (37) And even shaving back coverage can hurt. One longitudinal study, for example, "demonstrate[d] significantly worse health outcomes among individuals who reported restricting their use of prescription medications because of cost.... [C]ost-related medication restriction was associated with almost twice the odds of experiencing a significant decline in overall health over 2 years of follow up." (38)

In short, health plans do not now meet consumerism's first prerequisite adequate choice. Nor can they without restructuring health care financing in large, unlikely, and possibly unwise ways.


Consumerism promises to fix the clattering train by pressuring patients to "take a more active interest than they hitherto have had in the cost-effectiveness of their care." (39) Patients spending their own money should buy treatments cheaply and eschew wasteful care. But will patients have scope to economize on treatments? Not if no savings are practically possible or if reforms have already eliminated excess expenditures. Nor need patients economize if they will exceed the year's deductible. So, how much opportunity and incentive to save will patients have?

This question has two parts. First, are there substantial unexploited opportunities for economies? After all, "health care plans today already contain both substantial cost sharing and managed care measures that are likely to reduce spending." (40) Second, will consumerism induce patients to exploit any such opportunities?

Consumerism's easiest case is where equally effective treatments differ in cost. The classic example is generic drugs. Since by hypothesis the drugs are identical, the cheap treatment is so plainly preferable that doctors hardly need offer patients a choice. In these areas it's not clear how much room for improvement is left and whether referring decisions to patients helps. After all, managed care has campaigned for years to get doctors to substitute generics for brand-name drugs, and health plans increasingly use formularies to get doctors to prefer cheap to expensive drugs. (41) These efforts and a changing medical culture have helped. For example, as early as ten years ago "[o]nly 8% of physicians preferred brand-name medications over generic drugs regardless of cost." (42) Similarly, patients who use emergency rooms or specialists for routine problems notoriously waste resources, but health plans have already attacked this problem by increasing copayments for ERs and by requiring referrals to specialists.

Another easy case for consumerism is where patients decide whether a genuinely elective treatment is worth paying for. This category's borders are hazy, but some decisions, like cosmetic surgery, fit it neatly enough. But how much can consumerism save here, since patients already pay for most such treatments?

Consumerists' remaining core case is an optimal treatment and a cheaper sub-optimal treatment. At one extreme, the efficacy gap is so small and the cost gap so great that consumer choice is hardly needed to reach the economical result. At the other extreme, the efficacy gap is so great and the ailment so severe that the cheaper treatment seems hardly better than no treatment. Are there enough cases between these extremes to permit substantial savings?

Two cases may exemplify this category. The patient has an upper respiratory tract infection and acute sinusitis. A cheaper antibiotic works 70% of the time, a costlier one 80% and about a day faster. The consumerist patient weighs "the increased cost and potential side effects against the possibly quicker return to health." A "day of lost work may be more costly to her, in money or in other ways, than the additional cost of the more expensive antibiotic." (43) Another example: In Pegram, (44) the plan delayed a test that could have revealed the patient's appendicitis. Delay was probably cost-effective for the patient population, but an individual patient might want to pay extra for peace of mind or to avoid coping with a burst appendix. (45)

Tests are a common example in this category. They can be expensive (e.g., MRIs), and any test might be unnecessary (if the diagnosis is clear, tests are pointless). Furthermore, while tests are expensive, their costs usually fall within a high deductible, so that the patient might bear the cost. For example, some causes of back pain can be detected through an MRI, but those are unusual and not urgent problems. So it makes economic sense to treat the likelier causes and use an MRI only if that fails. But an anxious patient might think the test worth the cost.

In yet another consumerist category, patients need a treatment but can't afford it. This is a category many doctors know well and the category likeliest to expand with thorough-going consumerism. Preventing needed care is presumably not consumerism's goal. Who wants more arbitrary, inequitable, and inefficient rationing? Faced with such cases, doctors often become social workers, scrambling to help patients jury-rig treatments. This helps patients, but it misuses doctors' skills.

In sum, for consumerism to subdue costs, patients will need chances to economize. We have now asked how much choice will be available and how much scope for savings patients will have. The short answer is some of each, but not a lot of either.


As to Holmes, I observed that he sat frequently for half an hour on end, with knitted brows and an abstracted air, but he swept the matter away with a wave of his hand when I mentioned it. "Data! data! data!" he cried impatiently. "I can't make bricks without clay."

Arthur Conan Doyle

The Adventure of the Copper Beeches

Now suppose away the problems we've discussed. Consumers have enough choices of plans, providers, and treatments to suit their wants and their wallets. They next need information. They need to know what's on offer, its quality, and its cost. Does this information exist? Will it be forthcoming? From reliable sources? In brief: sometimes possibly, but not easily often.


Nobody can be told what nobody knows. Much information consumers need for choosing plans or providers cannot practically be compiled, would not willingly be compiled, or has not been worth compiling. Consumers get much less information than they need. Reinhardt is blunt: "Only rarely, in a few locations, do American patients have access to even a rudimentary version of the information infrastructure on which the theory of competitive market and the theory of managed care rest." (46) Nor is managed care's experience inspiring. Organizations had every incentive to explain themselves. Yet they offered "patients a bewildering array of acronyms and concepts that even specialists sometimes have trouble characterizing" while doing "little to explain the concept" of managed care. (47)

Might patients' demands induce providers to assemble information? Some, surely. But much information is hard to compile in a practical form. Take one indispensable datum--what doctors and hospitals charge. Prices are often unavailable in any form consumers could use to compare plans, providers, and treatments. They are often unavailable in any form anyone can readily use. (48)

Hospitals' fees, for instance, are listed in a "charge master"--"a confidential list of charges made by the hospital for all its goods and services. …

Log in to your account to read this article – and millions more.