A Dose of Our Own Medicine: Alternative ... - Wiley Online Library

A Dose of Our Own Medicine: Alternative Medicine, Conventional Medicine, and the Standards of Science E. Haavi Morreim

T

scientificstandardsas conventional medicine, they are headed for multifaceted disappointment. Standard medicine is not nearly so scientific as is usually assumed. Among other factors, there are far too many phenomena to study; limited research resources are often directed as much by political and commercial interests as by medical needs; actual practices do not reflect well the science that has been gathered; the most pristine science is often the least useful in the realworld care of ordinary patients. Additionally, it will be argued that medicine can sometimes be more “guilty” than CAM of wasting money, engendering false hopes, and even causing harm. In the end, even though conventional medicine can rightly take credit for remarkable successes in healing what ails humankind, much of actual clinical practice does not and never can measure up to the scientific standard to which critics of CAM would like to hold alternative modalities. Accordingly, it becomes unclear how to interpret the injunction to “hold” C A M modalities to full and rigorous scientific standards. Perhaps it means we should ban any modalities that have not been proven by solid science to be safe and effective. However, as discussed below, because so much of conventional clinical medicine is not scientifically validated, we would be banning large portions of it right alongside CAM. Reciprocally, insofar as CAM is to be held to “the same” scientific standards that clinical medicine actually meets, then CAM may enjoy a remarkably low threshold for acceptance. Perhaps instead the edict to hold CAM to scientific standards might simply mean that people should be discouraged from seeking scientifically unsubstantiated healing modalities. If this is all that is meant, it is not clear why the fuss is so vociferous. In a nation favoring free speech, any citizen is entitled to make such a recommendation. Moreover, it is not clear from whom the CAM critics think such recommendations should come. Conventional medical providers,

he discussion about complementaryand alternative medicine (CAM) is sometimes rather heated. “Quackery!” the cry goes. A large proportion “of unconventional practices entail theories that are patently unscientific.”’ “It is time for the scientific community to stop giving alternative medicine a free ride. There cannot be two kinds of medicine -conventionaland alternative.There is only medicine that has been adequately tested and medicine that has not, medicine that works and medicine that may or may not work.”2“I submit that if these treatments cannot withstand the test of empirical research, ...then we have wasted a lot of time and effort. The time has been wasted on all the people who have spent years learning falsehoods about acupuncture points and the principles of homeopathy. And the patients have wasted their time, money, and efforts receivingtreatments that were not what they were represented to be or were harrnf~l.”~ On a more conciliatory note, it is often proposed that some CAM modalities may be acceptable, but only those that stand up to the test of science. For instance, the White House Commission on Complementary and Alternative Medicine Policy, established in March 2000 by President Bill Clinton, concluded that “conventional and CAM systems of health and healing should be held to the same rigorous standards of good scien~e.”~ More precisely, “[tlhe same high standards of quality, rigor, and ethics must be met in both CAM and conventional research, research training, publication of results in scientific, medical, and public health journals, presentations at research conferences, and review of products and device^."^ In overview, this essay argues that, if critics of CAM expect to limit its influence by holding CAM to the same ]oumal ofhw, Medicine &Ethics, 31 (2003):222-235. 0 2003 by the American Society of Law, Medicine & Ethics.

222

TheJournal of Law, Medicine dr Ethics often the source of such counsel, are increasingly in competition with CAM providers, thus perhaps tainting their recommendation with conflictof interest. Indeed, courts have found that such conduct can, under at least some conditions, violate antitrust law.6 Perhapspublic health authorities might lend such a voice. With legislativeauthorization they could influence or directly regulate many aspects of health care. Indeed there may well be a legitimate, even increased, role for regulation of CAM.’ Yet here, too, we must hesitate before holding CAM modalities to a standard that conventional medicine could not itself meet. If not carefully undertaken, regulation can become a tool to suppress unwanted competition more than to promote public health. Sall another interpretation may be that the public should be protected from healing modalities that cause harm or that waste money. Yet here, too, conventional medicine might fare worse than CAM,as argued below. In perhaps the most realistic interpretation, the injunction might mean not that access to CAM should be harshly restricted, but rather that common funds, such as those in government or private health insurance plans, should only be spent on healing modalities that withstand scientificscrutiny. Yet even here, much of ordinary clinical medicine would be challenged while CAM modalities might achieve a level of acceptance its critics would find intolerable. As a result of these considerations, a relatively tolerant approach seems most appropriate. While it is not the purpose of this essay to explore the broad question of whether and to what extent society, or particular health plans, should fund access to CAM,one potential option for reconciling at least some of the tension will be explored -one that neither rejects nor unduly entrenches either kind of healing. An emergingapproach to health care financing, loosely dubbed “defined contribution” or “consumer-directed”health plans, may offer one reasonable response to the funding challenge -arguably the “bottom line” issue in many disputes about CAM. These plans commonly pair high-deductible catastrophic plans with personal spending accounts that cover the costs within the deductible. By thus placing a significant portion of routine, lower cost health care within the ambit of the spending account and thereby under the enrollee’s personal control, these plans can empower enrollees to direct much of their own health care. That control permits the enrollee to use both CAM and conventional modalities as he sees fit, even while encouraging judicious use of both and minimizing imposition on others, as the costs come from the enrollee’s own spending account.

scientific disciplines that have shown fundamentally what the human body is and how it functions -medicine is profoundly scientific. Moreover, research has led to a variety of stunning successes, such as organ transplants, joint replacements, and the curingor even eradication of infectiousdiseases that have historically killed many thousands of people. And yet ordinary clinical medicine the actual practice in which physicians examine individual patients, recommend tests, make diagnoses, and provide treatments -often has rather little resemblance to the tight data and gold-standard rigor of basic science. There are many reasons.

-

Insufficient science is available On the most obvious level, there is simply too much to do. The resources of science will probably never be sufficient to permit every phenomenon of human mental and physical function and well-being to be thoroughly studied. And as new technologiesare introduced at staggeringspeed, the task of evaluating their impact and best uses expands the job immeasurably. In addition, no individual physician can possibly keep up with and appropriately incorporate into his practice everything that is known, nor assimilate the rapid evolution in what becomes known. However, the ways in which clinical practice stray from science go far beyond this. The medical community has a long history of accepting new technologies, and new uses of existing technologies, with little science to connect theoretical foundations to such practical applications. For instance, pulmonary artery catheterization, widely used for three decades to monitor cardiopulmonary function in critically ill patients, is only now coming under scientificscrutiny as some recent reports suggest it may actually d o more harm than good.8 Similarly, angioplasty to open up clogged arteries in the heart was “performed in hundreds of thousands of patients prior to the first randomized clinical maJ demonstrating efficacy in 1992”9;likewise, although coronary artery bypass surgery was first performed in 1964, its efficacy was not scientifically evaluated until 1977. And in the case of bypass surgery, most of the patients who receive it are not like the ones in whom it was tested: “only 4 to 13 percent of the patients who now undergo this operation would meet the eligibility criteria for the randomized controlled trials that established its efficacy.”’O Moreover, this particular surgery is used significantlymore in the United States than in Canada and Europe, with no conclusive justification in terms of patients’ illness or infirmity.” It is also becoming evident that a significant amount of clinical care is based on studies that are methodologically inadequate. For instance, arthroscopic debridement or lavage for osteoarthritis of the knee has been widely practiced, based on theoretical promise and two methodologicallylimited studies.12 Recently, a gold-standard randomized, double-blind, placebo-controlled trial showed that this pro-

CLINICAL MEDICINE AND SCIENCE

Perhaps the most salient characteristicof conventional medicine is its scientific roots. Particularly at the level of basic science-the anatomy, physiology, biochemistry, and other

223

Volume 3 1 :2, Summer 2003

More recently, a promising, inexpensive approach to treat life-threateningsepsisz6has struggled to find funding for testing, largely because the major parties have financial stakes elsewhere. In thisinstance, studies that are promising but small indicate that long-term, low-dose steroids may reduce fatalities from sepsis by as much as 30 percent. However, while the steroids cost around $50, a major pharmaceutical company has recently launched an antisepsis drug costing $7,000 per dose.27On one hand, neither that pharmaceutical company nor any other has any financial desire to fund studies for a generic steroid that brings no profit. On the other hand, government agencies like the National Institutes of Health generally fund basic research, not clinical drugtesting. Hence, “a Catch-22: Because money is unavailable, only small studies are possible. Because they are small, they are viewed as less than convincing, allowing skepticism to persist -and money to remain unavailable. The drugs that draw the industry’s heavy research and promotional money are the branded ones, which are also far more expensive.”28 Such economic factors are hardly strangers to the conduct of scientific research. Even where basic science is supported by noncommercial resources, much of the science that translates biology into medical drugs, devices, and procedures is the product of commercial enterprise. This is not to say that such research is necessarily tainted or invalid. Drug and device manufacturers, after all, have produced some of medicine’s most stunning successes. Rather, it is to say that the direction of the research and the methodology of study are inevitably shaped by the goals of those who sponsor the research.29And therapeutic projects that have no sponsors tend to languish. In another step down the clinical line, the “outcomes studies“ that try to track the actual medical effects of these new technologies in clinical use are sometimes fraught with even greater problems of poor methodology, conflict of interest, and clinical ~nreliability.~~

cedure is no better than a sham surgery in which no surgical invasion of the knee took place.I3 Despite the limited evidence supporting the procedure, it has been performed on more than 650,000 people per year.I4 Similarly, after many years of thinking that hormone replacement therapy (HRT) for postmenopausal women brought a wide variety of benefits -based on observational studies in some cases funded by pharmaceutical companies -only recently has a long-term, double-blind, placebo-controlled trial been completed, suggesting that HRT may, on balance, do more harm than good.IS The surprising result has prompted many observers to wonder how many more of the well-accepted recommendations issued by conventional medicine are likewise flawed.I6While observational studies and other kinds of nonrandomized, noncontrolled studies are often the only or best evidence that can be gathered, episodes such as this prompt significant doubts. As one commentator suggested, “‘[ilt is mind-bogglingto give 20 million American women a drug with major side effects without definitiveproof that it in another well-known example, many physicians provided high-dose chemotherapywith autologousbone marrow transplant (HDUABMT) for breast cancer in the 1990s, despite the fact that there was essentially no credible evidence to support it, and some analyses had already indicated that the treatment provided no benefit over standardchemotherapy and in certain categories actually diminished patients’ p r o g nosis.I8 Nevertheless, by the end of the decade, well over 42,000 women had undergone the proced~re.’~ Demand for it became so intense that many insurers began to cover it in response to 1itigation;O and many legislaturesmandated coverage?’ Completinghigh-qualityresearch became exceedingly difficult because, once so many women had gained direct access to the treatment through their insurers, relatively few were willing to accept a trial that might randomize them to conventionaltreatment.u Not until 1999did studiesshow that the procedure was no better than standard chemotherapyu In some cases a dearth of science undergirding clinical practices may be the product of nonscientific forces. A research proposal must usually be either commercially or politically attractive if it is to be funded.” Gavras and colleagues observed this phenomenon in conjunction with clonidine, a very effective and inexpensive drug for congestive heart failure:

Clinical practice often does not reflea available science Even aside from such scientific potholes, clinical practice often does not reflect the available science. As noted by the Institute of Medicine: In the current health care system, scientific knowledge about best care is not applied systematically or expeditiously to clinical practice. An average of about 17 years is required for new knowledge generated by randomized controlled trials to be incorporated into practice, and even then application is highly uneven.... The extreme variability in practice in clinical areas in which there is strong scientific evidence and a high degree of expert consensus about best practices indicates that current dissemination efforts fail to reach many clini-

despite multiple efforts, there has been no interest so far from anyone to fund large controlled studies to demonstrate the long-term impact of these alterations on patient outcome. The clinically observed benefits of clonidine in heart failure are known to a small circle of academiccardiologists, but the primary care physicians who treat 90% of patients with chronic heart failure are totally unaware of themu

224

The Journal of Law, Medicine &Ethics

rately interpreted. When the tests were reread by a group of expert cardiologists, one-quarter of patients determined by the original reading to have the most severe disease did not have it. Six percent of persons who were told that their test results were not severely abnormal actually had severely abnormal results. One-third of persons whose bypass surgery was considered necessary or appropriate based on the original interpretation of the angiography results underwent surgery that was of uncertain benefit or inappropriatebased on the gold-standard review?’

cians and patients, and that there are insufficient tools and incentives to promote rapid adoption of best practice^.^' That variabilityhas been widely documented by researchers like John Wennberg, whose Dartmouth Atlas surveying nationwide health care practices indicates that the amount of care provided to patients depends far more on the supply of facilities than on the actual needs of patients.32In the federal Medicare program, for instance, “supply-sensitive”services such as specialist consultations, hospitalization, and physician visits have been shown to vary widely, in ways unrelated to patient needs, and with no discernible impact on health outcomesor life expectancy. The discrepanciesallegedly generate excess spending up to $40 billion per year.33 Many commentators have observed the loose connection between what is scientifically known and what physicians actually practice, as they note pervasive overuse, underuse, and misuse of common medical interventions.” Readers are referred to other sources for this broad literature,3sbut a few examples here will highlight the situation. Overuse can be seen in the proliferation of antibiotics, which have been used with such unnecessary frequency and potency that resistant organismsare increasinglya problem.j6 Similarly, fetal ultrasound has been shown to be of little use in routine pregnan~y,~’ yet to this day it is commonly used, regardless of maternal risk status. Underuse is likewise not difficult to find, even regarding simple, inexpensive measuresthat can have a significanthealth impact. For instance, during a myocardial infarction (heart attack), thrombolytic (clot-busting)agents can dramatically improve survival rates. Yet these drugs are seriously underused..*8After a myocardial infarction, aspirin and betablocker drugs can significantly reduce the likelihood of a second episode. Yet an average of only 37 percent of physicians actually prescribe these drugs for their post-myocardial infarction patients.39Similarly, many physicians fail to prescribe standard antiasthma medications,4°to order standard diabetes care:’ to prescribe diuretics for hypertension, to recognize and treat depression in the outpatient or to wash their hands between patient visits.4J Misuse is also a significant problem. Peptic ulcer disease, for instance, is often caused by a bacterium, for which antibiotics are the treatment of choice. Nevertheless, one study showed that “physicianscontinued to use traditional and ineffective [drugs] as their preferred approach for 72 percent of all patients.”+’ Another example concerns coronary angiography, an invasive test used to diagnose heart disease and determine appropriate treatment.

Problems with physicians’ diagnostic accuracy parallel similar problems in treatment choices. A review of autopsies at a major New Orleans medical center found that, of the 250 tumors found at autopsy, 111were undiagnosed or misdiagnosed. In 57 percent of these patients, the underlying cause of death was directly related to the undiagnosed or misdiagnosed malignancy? Another study revealed deficiencies in standard techniques of physical examination, specifically, the ability of trainees in internal medicine and family practice to recognize important sounds when listening to patients’ hearts. Overall, senior residents recognized only 20 percent of what they heard, no better than medical students whose skills were also evaluated, by implication raising questions about the skills of physicians in their practice after ~esidency.~’ At least some of this lack of connection between available science and actual clinical practice comes from economic and market forces. For instance, Cox-2 arthritis were introduced into the market on the premise that, although they did not relieve pain or reduce inflammation any better than conventional nonsteroidal antiinflammatory drugs (NSAIDs), they were less prone to cause gastrointestinalp r o b lems. Thus they have never been touted to offer extra value for acute injuries or for patients without significant risk for gastrointestinal complications. Nevertheless, one study showed that, within 9 months of these drugs’ introduction to the market, Cox-2s accounted for 40 percent of prescriptions in one health plan, even though only 14 percent of its patients had arthritis, and even though these drugs are vastly more expensive than conventional NSAIDS.~~ This shift took place well before these drugs began aggressive direa-to-consumer marketing. Recently, seriousquestionshave been raised about whether Cox-2 drugs even offer superior protection against gastrointestinal ~ide-effects.~~ More broadly, evidence indicates that it is not uncommon for physicians unwittingly to base treatment and drug selection decisions more on the basis of drug company promotion than on the medical literat~~re.~’ In one study, for instance, physicians who were hosted at luxury resorts to learn about expensive new drugs roughly trebled their use of those drugs upon returning home.52

Analysis of a random sample of angiographies performed in one state showed that only half of the tests were done competently enough to be accu-

225


Gap between medical science and clinical practice is unlikely to fade In reply to all this it might be argued that such problems are simply happenstance, that medical practice can be improved and made more scientific. To be sure, great efforts are now under way and many of them will improve the quality and scientific basis of health care. However, there will always remain a profound disconnect between medical science and clinical medicine. Science is the study of the general, while clinical medicine is the study of the particular. For a host of reasons, ranging from resource constraints to human complexity to the ethical limits on acceptable research, it will forever be impossible to do medical science in a way that makes ordinary medical care a straightforward matter of applying science to generic patients. For one thing, over the past few decades “an explosion has occurred in the proliferation and supply of drugs, the availability of technological tests and bedside procedures, and the array of high-tech diagnostic methods and invasive therapeutic maneuver^."^^ The changes are so enormous, pervasive, and rapid that it is essentially impossible for physicians -or anyone else, for that matter -to keep abreast of the developments. In that vein, some areas at the forefront of medicine change so rapidly that it is difficult to achieve scientifically worthwhile results. By the time enough research subjects are enrolled to provide statistical significance in a clinical trial, physicians’ experience and with it their use of the technology have evolved so far that the data are already outdated. This phenomenon was observed in the early use of ECMO (extra-corporealmembrane oxygenation)technol~gy;~for instance, as well as a host of other cumng-edge interventions. On a more commonplace level, it must also be understood that the more scientifically pristine a study is, the less it actually applies to ordinary people. Science’s gold standard, the randomized, double-blind, controlled trial, is particularly problematic. To test strictly for the effects of the intervention under investigation,study design must use narrow eligibility criteria. Usually enrolleesmust suffer exclusively from the particular condition being studied, with a minimum of other diseases and medications, lest research results be confounded by ancillary factors.55Once the study is complete, however, it is applied in clinical practice to all those complex patients who would never have been eligible to be test subjects in the study.S6One commentator noted: as many as 60% to 80% of patients with heart

failure have been excluded from clinical trials of angiotensin-convertingenzyme inhibitor therapy owing to comorbid and associated conditions that tend to obscure the efficacy of the drug in improving functional capacity or prognosis. This standard practice in research has a rational basis. However, the clinician must treat 100%of the patients

with heart failure, not just the 20% to 40% who are free of comorbidities and associated conditions. Moreover, the clinical effectiveness of drug therapy is often limited by the very comorbid and associated conditions for which patients were excluded from drug trials.57 The clinical implicationsof this have been noted in the case of heart failure. Data regarding this condition have been significantly skewed because “[olnly young subjects with minimal comorbidity had the low mortality rates that were typically seen in contemporary clinical trials of heart failure. For most community-dwelling subjects with heart failure who were more likely to be older women with significantcomorbidities, their prognosis remained poor.ns8Accordingly, it is possible “that the evidence-basedpractice that currently exists for heart failure may only be appropriate for the limited segment of the population with heart failure included in the trials.”s9 In sum, the more perfectly scientific and highly controlled a study is, the less its enrolled subjects resemble the ordinary souls, with their multiple problems, for whom ordinary physicians care. No clinical rules can be detailed enough to accommodate patients’ vast variability, because biological idiosyncrasies, complex conditions, and atypical presentationsoften do not fit into preestablishedcategories.a An important implication of this misfit between scientific studies’ narrowly identified participants and the broad spectrum of people who later receive a given medical intervention is that sometimes even well-researched new drugs and procedures must be quickly withdrawn from the market, as they suddenly produce undesirable results and side-effects that were not seen during the research period. Between September 1997 and September 1998, five drugs approved by the Food and Drug Administration (FDA) were removed from the market because of unexpected side-effects or interactions with other drugs. In one case, by the time the drug mibefradil (Posicor) was removed, it “was known to interact with 26 different drugs, a number and diversity that could not, in practical terms, be addressed by standard labeling instructions or additional public warnings.”61 From the foregoing, one can only conclude that the actual clinical practice of medicine cannot realistically, faithfully claim to be the scientificenterprise that is presupposed when CAM modalities are criticized as being “unscientific.” Medicine has strong roots in science, and the scientific method has produced many wonderful diagnostic and therapeutic tools. But the actual process of daily clinical care often is, and in many ways must inevitably be, rather far removed from those scientific roots. OTHER CRITICISMS OF CAM

Aside from the charge that CAM modalities are unscientific, critics also argue that CAM can be harmful, that it can create 226

The]ournal of Law, Medicine &Ethics rors in the outpatient setting provide a substantial even if unquantified addition to the total damages?’ It would appear that, if we are concerned about direct harms to patients, conventional medicine poses the greater danger. However, critics’ chief complaint is not so much that alternatives cause direct harm, but that patients opting for these may forego the (presumedly more effective) approaches offered by standard medicine. In some cases this is surely true. Diabetes Type I is more effectively treated with insulin than with prayer. Yet in other cases the claim is dubious, as where the supposedly superior medical intervention is in fact unproven -not entirely uncommon, per the discussion above. More importantly, in most instances there is little reason why patients should be forced to choose on an eithedor basis. With limited exceptions, such as herbs that might cause adverse interactions with conventional pharmaceuticals, there is usually little reason to force patients to forego standard medicine in order to pursue alternatives. The harm may thus be more in the forced choice than in the alternatives themselves. As another genre of harm, alternative modalities are often criticized on the ground they create false hope. h a number of instances this is probably true. Yet the same is true of conventional medicine, particularly where an intervention is not scientifically documented or where available science may not fit the individual patient. Indeed, medicine has produced a number of ineffectivetreatments offered as last-ditch hopes for patients. HDGABMT for breast cancer, as discussed above, was largely the product of a bit of theory and physicians’ desire to do something- anything- to help patients for whom they had little to offer. They peddled false hope, at great personal and financial cost.

false hope, and that people should not be wasting their money on unproven healing methods. However, just as above, conventional medicine is as, or more, subject to the same weaknesses.

Doing harm To be sure, many CAM modalities have not been well-studied, and their potential for harm has not been adequately evaluated. Some herbs, for instance, have been associated with liver damage, and other harmful connections could of course be identified in the Nevertheless, most CAM modalities tend to be fairly benign. Homeopathy, acupuncture, massage therapy,therapeutic touch, hypnosis, reflexology, and aromatherapy, for instance, do not use high-risk, invasive interventions of the sort common in conventional medicine. Spiritual approaches such as prayer, transcendental meditation, guided imagery, and support groups do not involve physical invasion at all, though psychological harms could occur in some instances. Chiropracticis sometimes thought to pose risk,yet available data indicate the risk seems limited. The more serious reported complications are the cauda equina syndrome from lumbar manipulation and cerebrovascular artery dissection from cervical manipulation. The apparent rarity of these accidental events has made it difficult to assess the magnitude of the complication risk. No serious complication has been noted in more than 73 controlled clinical trials or in any prospectively evaluated case series to date. Seriouscomplications from lumbar spinal manipulations are extremely rare, estimated to be 1 case per 100 million manipulations. For cervical manipulation,the risk for a cerebrovascular accident has been calculated by various authors to range from 1 in 400,000 to between 3 and 6 per 10 million manipulations. The figures have been primarily based on retrospectively collected single case reports and unsubstantiated practitioner surveys.6’

Wasting money Finally, CAM modalities are often accused of wasting patients’ moneyY Yet once again, conventional medicine appears guilty of vastly greater waste. Admittedly, cumulative CAM expenditures are not insignificant. Eisenberg and colleagues found that in 1990 Americans spent nearly $14 billion on such alternatives,threequarters of that sum out of pocket.69There were more visits to unconventional providers than to all primary care physicians combined, and the people most commonly using alternatives were relatively affluent and well-educated.” By the time Eisenberg and colleagues revisited the question in 1997, there had been a 47 percent increase in the number of visits to alternative practitioners, and a 45 percent increase in spending, to an estimated $27 billion.” Nevertheless, typical CAM interventions, such as over-the-counter herbs, massage, prayer, and even chiropractic, are inexpensive, relative to mainstream medicine. Moreover, standard medicine has been known to waste vast amounts of money - enough to make CAM pale by

On the other side, conventional medicine carries wellknown, sometimesenormous risks, whether from side-effects, complications, or outright errors. According to a 1999 report by the Institute of Medicine (IOM), mistakesin hospitals alone kill somewhere between 44,000 and 98,000 people per year.@Using the IOMs low-end figure, “medical errors rank as the eighth leading cause of death, killing more Americans than motor vehicle accidents, breast cancer, or AIDS,” at an annual cost of $17 billion to $29 billion.6sIn a similar vein, more than 100,000persons die from adverse drug reactions in U.S. hospitals each year, “making these reactions between the fourth and sixth leading cause of death.”“ Er227

Volume 31:2, Summer 2003 comparison, even if every CAM dime were considered a waste. Arthroscopic debridement or lavage for osteoanhritis of the knee, noted just above to be better than sham surgery, has been performed more that t 650,000times per year at an average cost of $5,000 each -wasting $3.25billion per year.72As opined by the study’s authors: “the billions of dollars spent on such procedures annually might be put to better use.”73In the coronary angiography study discussed above, which found that only half the tests were done competently and many were misread, the authors observed that “[nlearly 1.3 million coronary angiogaphies were performed in 1998 nationally. If the results of this study held nationally, nearly 650,000 tests would be difficult to interpret accurately; at $12,450 per test, that is more than $8 billion in wasted expen~e.”’~ Over the course of a decade, more than 42,000women received the later-disproven HDC/ABMT for breast cancer, at an estimated cost of $3.4 billi~n.’~ This figure does not count the enormous sums spent on lawsuits that often required health plans to pay for the treatment and in some cases levied millions of dollars in punitive damages against plans.76Nor does it include the many familieswho incurred overwhelming personal debts” to pay for the treatment because doctors said it represented the patient’s best or only hope. Medical history is replete with examples of treatments that became well-established without a good scientific basis and which, when finally tested, turned out to be worthless or outright harmful. Examples include high-dose oxygen for premature infants (later recognized as medically unnecessary and a major source of blindness) and internal mammary artery ligation for heart disease.’” Other once-popular, later-disproven, money-wastingtreatments include glomectomies for asthma, photodynamictherapy and organicsolvents for herpes simplex infection, and gastric freezing for ulcers.79 To these we can add the costs of overuse and misuse alluded to above. Physicians’ clinical practices have been documented to vary widely,sOin ways not justified by differences in patients’ conditions and, while we cannot be certain “which rate is right”81for various surgeries,diagnostic interventions, and drug prescriptions, it is safe to suppose that at least some of this variation represents unnecessary care and thereby additional wasted money. As noted above, such unjustified variations allegedly create excess spending up to $40billion per year.“z

held to a standard, of proof that medicine cannot possibly meet. The successes of alternatives cannot be dismissed as mere “anecdotes” and “testimonials,” if from conventional providers we accept case reports, case series, observational data, and physicians’ statements that “my patients have done welVpoorly with this treatment” or “in my clinical experience X has been true.” When alternativeproviders emphasize individualizingcare, they should not be called “unscientific” if, at the same time, physicians who also individualize their interventions are approvingly seen to practice the “art” of medicine. A CAM modality should not be singled out as “peddling false hope” if, in a comparable situation, physicians might paint overly optimistic pictures and offer treatments they know will not likely help, on the ground that “we can’t take away the patient’s hope.” However, more concrete responses are also available. One of the most contentious aspects of the debate about CAM is the question of whether CAM should be covered by third-party insurers. Neither insurers nor those who pay the premiums want to spend limited funds on worthless care. This maxim should apply to conventional as well as to alternative modalities and, indeed, a significant focus of managed care during the 1990s aimed to make health care expenditures more rational and evidence-based.Managed care tactics may have been badly ineptP but the goal was worthwhile. It is not the purpose of this article to survey all the possible answers to the question of which kinds of care to fund at what levels. However, one recent turn in the evolution of health care financing offers a particularly promising avenue for permitting people to seek a variety of healing modalities, both CAM and conventional, in ways that minimize inappropriate impositions on common resources. It is the shift Although from “defined benefits” to “defined contrib~tion.”~~ not a panacea, nor the only viable approach, it is an idea that exemplifies the kind of creativity that will be needed if we are to permit people to exercise reasonable choices over their own health care, while limiting the ways in which they can impose on others’ resources. Under the traditional defined benefits approach, an institutional purchaser such as an employer determines what range of services it will cover, then seeks or creates a plan that will cover those services for an acceptable price. Unfortunately, these defined benefit systems have become increasingly difficult to sustain. A steady stream of emerging technologies requires an equally steady stream of decisions about which ones will be covered by the plan, leading inevitably to an equally steady stream of disappointments over those that are not covered. In the early twenty-first century, health care inflation is again rising while cost-cutting tools have become largely impotent, thus necessitatingeven sharper cuts in coverage.ss As a result, some employers are shifting to a defined contribution approach in which they determine at the outset how much they will spend for health care, then provide op-

3 . 1 )

RESOLVINGTHE TENSIONS The upshot is clear enough. Any attempt to throw out or discredit CAM on grounds of scientific inadequacy is sure to toss out large portions of conventional medicine alongside. To “hold” both to “the same” standards appears to bode far worse for medicine than for CAM. Hence, one important step toward resolving the tension is to avoid double standards in which CAM modalities are

228

The]ournal of Law, Medicine 6 Ethics 1994, “85%of Americans spen[t] less than $3000 a year on medical care, and 73% ha[d] less than $500 a year in claims.”y0Similarly, as of 1996, 1 percent of patients consumed 27 percent of total health expenditures, while the top 10 percent of patients consumed nearly 70 percent, and the top 30 percent consumed 90 percent. This picture has not changed significantlyover several decade^.^' Thus constructed, consumer-directed health care has many potential advantage^.^^ At least within the range of the deductible, enrollees have freedom to seek health care as they see fit, often with few or no financial barriers because most, if not all, of the money they need is already in their personal spending accounts. Many of the most noxious features of managed care are obviated. Instead of being shuffled from one doctor to the next every time their employers switch to a different health plan, or every ’time their health plan shuffles its provider network, enrollees can choose -and keep - the providers they want. They pay the doctor directly from their own funds, and this financial accountability brings the power of the purse. Similarly, there is no need for gatekeepers and utilization managers, at least within the range of the deductible, because the employee himself has reason to be prudent about how much of his own spending account he uses for what purposes. As a further and very important consequence, the physician-patient relationship can be placed on an ethicallysounder footing. With n o more need for odious external micromanagement, or for financial incentives that pay physicians to withhold care, physicians and patients can make their own decisions, based on what is best for the patient. In this setting, the physician who says “you don’t need the highcost drug advertised on T.K” can be regarded as an ally trying to help the patient avoid wasting his own money, not as an enemy trying to cut costs for an insurance company. Reciprocally, the advice that “you really do need this test” can be taken seriously, because the funds in the personal spending account cannot be spent on anything but health care. Additionally, to ensure that employees do not forego important preventive or disease management care, some employers provide separate, first-dollar coverage for these kinds of services. Administrativesavingscould be substantial,as bills within the deductible range are paid directly by patients, often by swiping a health-fund debit card. Physicians and other providers are paid immediately, with no, need to hire endless armies of clerical workers to file, seek approval, and re-file for the simplest reimbursements. At the,same time, the impetus toward fraud is markedly diminished. Even a provider who might be willing to defraud a large insurer may be far more reluctant to steal from his own patient - a patient who suddenly has good reason to examine his bills more carefully. As a further advantage, once the great majority of people and their low-end expenses are decided directly by doctors and patients who have direct incentive to be both medically

tions from whch beneficiariescan choose.86The oldest form of defined contribution is essentially a voucher system in which the employer identifies a set of standard health plans and provides enough funds to cover the least expensive. Employees can purchase costlier plans by covering the difference out of pocket. Probably the most familiar example is the Federal Employees Health Benefits Program. Newer versions of defined contribution, commonly implemented as “consumer-directed healthcare benefit^,"^' are considerably more interesting when it comes to the question of funding CAM.These plans feature financialaccounts from which the beneficiary draws part of the funds to purchase high-deductiblecatastrophic coverage, then keeps the remainder of the money in a personal spending account on which he can draw for health expenses within the deductible range. The spending account might be sufficient to cover the entire deductible, or might require the employee to ante up some limited portion of expenses out of pocket. As an example, the deductiblemight be $3,000 and the account might be $2,000. The enrollee would then have essentially firstdollar coverage up to the $2,000, then would ante the next $1,000 out of pocket, after which the catastrophic plan takes over. This is only one example, of course, because exact financial arrangements can differ widely. The variations on this theme are endless, particularly as Internet-based health plans create important new options.s8 Commonly, these Lnternet health plans provide or help members to find catastrophic coverage, then help them to make informed decisions about the routine care that falls within the deductible. Such plans, for instance, may provide lists of avajlable physicians in various specialties, along with their fees, education, credentials, and perhaps consumer-satisfaction ratings. In some cases the plan might indicate what payment it will provide for a given service, enabling the enrolleeto choose whatever provider he wants and pay out of pocket any difference between the plan’s allowance and the provider’s fee. Some plans permit the enrollee to establish a special account to cover routine servicesand preventive care, or to set up subaccounts to cover specified medical procedures such as carpal tunnel repair or knee arthroscopy. Throughout, the Internet-based plans also help enrolleeskeep track of expenditures and plan for future needs. In an important development, the Internal Revenue Service has ruled that money left over in these personal spending accounts at the end of one year can now be rolled over into the next year, thus permitting people to accumulate funds for larger expenses in later years.89It also means that, even if someone faces the potential for sigruficantout-of-pocketcosts in the early years of a defined contribution plan -as where the deductible is $3,000 but the spending account is only $2,000 -over time he may be able to build an account that reaches or exceeds that deductible. Indeed, most people could expect to build up such padding. In any given year, the vast majority have relatively limited health care expenses. As of 229

Volume 3 1:2, Summer 2003

and fiscally prudent, health plans can be free to focus on major illness and injury. These account for the majority of health care expenditures, even if only a minority of patients. In this focused area, plans can use their technology assessment resources to p r o m o t e the kind of scientific, evidence-based health care that conventional medicine has long claimed to deliver but, per the discussion above, too often does not achieve. In sum, consumer-directed plans could return control over health care to the great majority of people because few people enter the realm of catastrophic coverage. To be sure, some out-of-pocket expenditures could be required, even under a consumer-directed health plan. Yet even here, defined contribution plans can limit the impact. As noted above, the roll-over provision will permit many people to build up the funds in their spending accounts over time. It is already a given that consumers will pay substantially more for their health The question is only how the money will be taken, and who will control it. In old-style defined benefit plans, the consumer will shoulder higher co-insurance, higher deductibles, and higher co-pays for care that is largely chosen, or at least heavily restricted, by his employer or health plan. Under defined contribution, he at least has the opportunity to choose much of that care himself. The implications of consumer-directed health plans for CAM should by now be obvious. Given that most CAM modalities are relatively inexpensive, most if not all decisions about CAM would be made by patients themselves, spending from their own personal accounts within their deductible. The person who prefers massage over drugs for his fibromyalgia; or the arthritis patient who prefers acupuncture, can make his choice without haranguing an insurer or paying the entire tab out of pocket. The person who wants the expensive drug advertised on T.1can have it, but with a personal incentive to consider the generic version, as the extra cost comes from his own funds. In the process, each can explore a greater range of healing modalities that may, for whatever reason, suit his own needs better than the narrower range of options available under traditional health plans.

-

CONCLUSION In many ways, the controversy about CAM is much ado about nothing. In actual clinical practice, conventional .medicine cannot always claim to be more scientific, less harmful, or less wasteful than its CAM alternatives. Nor is it necessarily more effective. As noted by Eisenbcrg and colleagues, most CAM is used for problems such as chronic illness or chronic pain, for which conventional medicine sometimes has rather little to offer.Y4 In a shift to consumer-directed health plans, CAM modalities could be more easily available to those who want them, without intruding on common resources. At the same

time, insurers n o longer preoccupied with the huge realm of lower level health care costs would finally be free to focus on the issues that most urgently need attention the relatively few patients with catastrophic illnesses and injuries and bring to them the systematic, evidence-based healing that scientific medicine can and should provide.

-

REFERENCES 1. E.W. Campion, “Why Unconventional Medicine?,” N. Engl. J. Med., 328 (1993):282-83, at 282. 2. M. Angel1 and J.I? Kassirer, “Alternative Medicine The Risks of Untested and Unregulated Remedies,” N. Engl. J. Med., 339 (1998): 839-41, at 841. 3. L.J. Schneiderman, “AlternativeMedicine or Alternatives to Medicine? A Physician’s Perspective,” Cambridge Quarterly of Healthcare Ethics, 9 (2000): 83-97, at 91. 4. White House Commission on Complementary and Alternative Medicine Policy, Final Report (March 2002):at Executive . Summary, available at 5 . Id. 6. Wilk v. American Medical Ass’n, 895 F.2d 352 (7th Cir. 1990), cert. denied, 498 US. 982 (1990). Other conventional providers are beginning to provide some forms of CAM themselves, under the heading of “integrative medicine.” See R. Snyderman and A.T. Weil, “Integrative Medicine: Bringing Medicine Back to its Roots,” Archives oflnternal Medicine, 162 (2002): 395-97. 7. D.M. Marcus and A.I? Grollman, “Botanical Medicines -The Need for New Regulations,” N.Engl.]. Med., 347 (2002): 2073-76; PA.G.M. DeSmet, “Herbal Remedies,” N. Engl.J. Med., 347 (2002): 2046-56; D.M. Eisenberg et al., “Credentialing Complementary and Alternative Medical Providers,” Annals of Internal Medicine, 137 (2002):965-73. 8. J.B. Hall, “Use of the Pulmonary Artery Catheter in Critically Ill Patients: Was Invention the Mother of Necessity?,”JAMA, 283 (2000): 2577-78; J. Rapoport et al., “Patient Characteristics and ICU Organizational Factors that Influence Frequency of Pulmonary Artery catheterization,” JAMA, 283 (2000): 255967; G.R. Bernard et al., “Pulmonary Artery Catheterization and Clinical Outcomes: NHLBI and FDA Workshop Report,”JAMA, 283 (2000): 2568-72; J.D. Sandham et al., “A Randomized, Controlled Trial of the Use of Pulmonary-Artery Catheters in High-Risk Surgical Patients,” N. Engl. J. Med., 348 (2003):5-14; RE. Parsons, “Progress in Research on Pulmonary-Artery Catheters,” N. Engl. J. Med., 348 (2003):66-68. 9. J.E. Dalen, ‘“Conventional’and ‘Unconventional’Medicine,” Archives of Internal Medicine, 158 (1998): 2179-81, at 2180. In another example, as of 1988, a national conference on antithrombotic therapy (anticlotting treatments used to prevent stroke, pulmonary embolism, and the like) evaluated the scientific foundation for various recommendationson which physicians based treatment. The American College of Chest Physicians found that only 24 percent of those recommendations were based on appropriately scientific studies, while 55 percent were based on uncontrolled clinical observations. Ten years later, 44 percent of the recommendations were science-based, though this was largely because of Food and Drug Administration requirements for the testing of new drugs. Id. at 2179. 10. A.C. Gellins, N. Rosenberg, and A.J. Moskowitz, “Capturing the Unexpected Benefits of Medical Research,” N. Engl. J. Med., 339 (1998):693-97, at 694. See also A.R. Feinstein, “Clinical Judgment Revisited: The Distraction of Quantitative

-

230

The Journal of Law, Medicine 6 Ethics Methods,” Annals of Internal Medicine, 1230 (1994): 799-805. 11. R.A. Lange and L.D. Hillis, “Use and Overuse of Angiography and Revascularization for Acute Coronary Syndromes,” N. Engl. J. Med., 338 (1998): 1838-39. ”Although the patients enrolled in the United States were more likely than their Canadian counterparts to undergo coronary angiography (68 percent vs. 35 percent, respectively) and subsequent revascularization (31 percent vs. 12 percent), the incidence of reinfarction and death during more than three years of follow-up was similar.. .. [Elxcessive use is probably related to the more widespread availability of facilities and trained personnel in the US.” Id. at 1839. See also E.S. Fisher and H.G. Welch, “Avoiding the Unintended Consequences of Growth in Medical Care: How Might More Be Worse?,”JAMA,281 (1999): 446-53. 12. D.T. Felson and J. Buckwalter, “Debridement and Lavage for Osteoarthritis of the Knee,” N.Engl.J. Med., 347 (2002): 132-3 3. 13. J.B. Moseley et al., “A Controlled Trial of Arthroscopic Surgery for Osteoarthritis of the Knee,” N. Engl. J. Med., 337 (2002): 81-88. 14. Id. at81. 75. G . Kolata and M. Petersen, “Hormone Replacement Study a Shock to the Medical System,” New York Xmes, July 10, 2002. 16. G. Kolata, “In Public Health, Definitive Data Can Be Elusive,” New York Times,April 23,2002. See also D. Grady, “Scientists Question Hormone Therapies for Menopause Ills,” New York Ximes, April 18, 2002; Kolata and Petersen, supra note 15; D. Grady, “A 60-Year-Old Woman Trying to Discontinue Hormone Replacement Therapy,” JAMA,287 (2002): 2130-37. 17. Kolata, supra note 16 (quoting Dr. Deborah Grady, a professor of medicine and epidemiology at the University of California in San Francisco). 18. See ECRl (formerly the Emergency Care Research lnstitute), High-Dose Chemotherapy with Bone M a m w Transplantfor Metastatic Breast Cancer (1996): at Brief Summary, available at . See also High-Dose Chemotherapy with Peripheral Stem Cell/Autologous Rescue, Treutment for Breast Cancer, in 2 Hayes Directory of New Medical Technologies’ Status (updated May 1997), available through ,Hayes-Med Database. 19. M.M. Mello and T.A. Brennan, “The Controversy over High-Dose Chemotherapy with Autologous Bone Marrow Transplant for Breast Cancer,” Healthwuirs, 20, no. 5 (2001):101-18, at 110. 20. Wl?Peters and M.C. Rogers, “Variation in Approval by lnsurance Companies of Coverage for Autologous Bone Marrow Transplantation for Breast Cancer,” N. Engl. /. Med., 330 (1994): 473-77. 21. “ln 1998, Congress enacted the ‘Women’s Health and Cancer Rights Act of 1998,’ amending ERISA. The Act requires all group health plans and health insurance issuers offering coverage for mastectomies to provide reimbursement for reconstructive surgery that is associated with a mastectomy” (citing 29 U.S.C. 5 1185(b) (Supp. 1999)).S. Hoffman, “A Proposal for Federal Legislation to Address Health lnsurance Coverage for Experimental and Investigational Treatments,” Oregon Law Review, 78, no. 1 (1999):203-74, at 252. See also T.F. Theodos, “The Patients’ Bill of Rights: Women’s Rights under Managed Care and ERISA Preemption,” AmericanJournal of Law t9Medicine, 26 (2000): 89-108. 22. As noted in one commentary: problems have arisen in efforts to recruit women for trials designed to assess the benefit of HDC/ABMT in the treatment of metastatic breast cancer. Because of the

availability of HDC/ABMT outside of clinical trials, many women with metastatic breast cancer have not been willing to accept the chance of being randomized to a control group in a trial designed to evaluate the effectiveness of HDUABMT. As a result, it has taken much longer than expected to obtain an adequate number of pamcipants in these studies to resolve the uncertainty over the value of this technology. E.P Steinberg, S. Tunis, and D. Shapiro, “Insurance Coverage for Experimental Technologies,” Health Affairs, 14, no. 4 (1995): 143-58, at 1SO. See also G. Kolata, “Women Rejecting Trials for Testing a Cancer Therapy,” New York Times,February 15,1995, at C8; G. Kolata and K. Eichenwald, “Hope for Sale: Business Thrives on Unproven Care, Leaving Science Behind,” New York Emes, October 3,1999, at Al. 23. Of five studies released in 1999, four indicated that highdose chemotherapy with bone marrow transplant was no better for breast cancer than conventional chemotherapy. A fifth study, done in South Africa, suggested some benefit. However, several months later, as scientists looked at this study more closely in an effort to replicate its results, the principal investigator admitted to having falsified some of the data “out of a foolish desire to make the presentation more acceptable” to the scientific meeting sponsored by the American Society of Clinical Oncology. R.B. Weiss et al., “High-Dose Chemotherapy for High-Risk Primary Breast Cancer: An On-Site Review of the Bezwoda Study,” Lancet, 355 (2000): 999-1003, at 1003. See also K.H. Antman, D.F. Heitjan, and G.N. Hortobagyi, “High-Dose Chemotherapy for Breast Cancer,” JAMA,282 (1999): 1701-03; ‘NJ. Gradishar, “High-Dose Chemotherapy and Breast Cancer,” IAMA, 282 (1999): 1378-80; PA. Rowlings et al., “Factors Correlated with Progression-Free Survival After High-Dose Chemotherapy and Hematopoietic Stem Cell Transplantation for Metastatic Breast Cancer,” IAMA, 282 (1999): 1335-43; R. Horton, “After Bezwoda,” Lancet, 355 (2000): 94243; J. Bergh, “Where Next with Stem-Cell-Supported High-Dose Therapy for Breast Cancer?,” Lancet, 355 (2000): 944-45. Another study completed even more recently comes to the same conclusion, namely, that bone marrow transplant offers no advantage over conventional chemotherapy. See E.A. Stadtmauer et al., “Conventional-Dose Chemotherapy Compared with HighDose Chemotherapy Plus Autologous Hematopoietic Stem-Cell Transplantation for Metastatic Breast Cancer,” N. Engl. J. Med., 342 (2000): 1069-76; M.E. Lippman, “High-Dose Chemotherapy Plus Autologous Bone Marrow Transplantation for Metastatic Breast Cancer,” N. Engl. /. Med., 342 (2000): 11 1920. In all, some 30,000 women had received the treatment, at a cost estimated around $3 billion. Associated Press, “Breast Cancer Procedure Bogus,” Memphis Commercial Appeal, March 11, 2000, As. It might also be noted that, during the heyday of using autologous bone marrow transplant for breast cancer, many hospitals and physicians made enormous sums of money from the treatment. See Kolata and Eichenwald, supra note 22. 24. The manufacturer of a costly new drug for arthritis is hardly likely, for instance, to d o a scientific comparison between its products and copper bracelets. Even a remote possibility of finding such an inexpensive remedy to be effective is enough to discourage such a study from being undertaken. And because copper is so inexpensive, the bracelet manufactures can’t make enough profit to justify the expense of the research - particularly since they are not required to do any science as long as they make no health claims. Indeed, science is not merely unlikely in such scenarios. History has shown that sometimes even when high-quality scientific trials have been done, their results may not

23 1


see the light of day if they are unfavorable to the study’s sponsor. See D. Rennie, “Thyroid Storm,”]AMA, 277 (1997):1238-43. 25. 1. Gavras, A.J. Manolis, and H. Gavras, “The Economics of Therapeutic Advances: The Paradigm of Sympathetic Suppression in Chronic Heart Failure,” Archives oflnternalMedicine, 159 (1999):2634-36, at 2635. Others have likewise observed that inexpensive, effective therapies tend to be ignored. See J.S. Goodwin and J.M. Goodwin, “The Tomato Effect: Rejection of Highly Efficacious Therapies,”]AMA, 25 1 (1984):2387-90;J.S. Goodwin and J.M. Goodwin, “Failure to Recognize Efficacious Therapy: The History of Aspirin Treatment for Rheumatoid Arthritis,” Perspectives in Biology and Medicinc, 25 (1981):78-92; F.A. Lederle, % A.! Applegate, and R.H. Grimm, Jr., “Reserpine and the Medical Marketplace,” Archives of Internal Medicine, 153 (1993):705-06. 26. Sepsis is a devastating blood infection in which the body’s normal inflammatory response to infection goes out of control, threatening major organ systems. 27. Interestingly, although the drug is the first and only drug approved for the treatment of severe sepsis, several commentators have indicated that they “believe that there is not sufficient evidence at present for it to become the standard of care,” because the performance of the drug was not consistent throughout the trial, leading to a mid-trial change in the protocol. H.S. Warren et al., “Risks and Benefits of Activated Protein C Treatment for Severe Sepsis,” N. Engl. 1. Med., 347 (2002):1027-30, at 1030.See also B.J. Mannis et al., “An Economic Evaluation of Activated Protein C Treatment for Severe Sepsis,” N. Engl. I. Med., 347 (2002):993-1000;J.l? Siegel, “Assessing the Use of Activated Protein C in the Treatment of Severe Sepsis,” N. Engl. J. Med., 347 (2002):1030-34. 28. T.M. Burton, “Why Cheap Drugs that Appear to Halt Fatal Sepsis Go Unused,” Wall Street]ournal, May 17,2002,at A l , A7, at A7. 29. For a considerably more detailed discussion of these points, see E.H. Morreim, Holding Health Cure Accountable: Law and the New Medical Marketplace (New Kork: Oxford University Press, 2001): at Chapter 5. 30. “The HTA [health technology assessment] activities of many for-profit and not-for-profit research organizations are based almost exclusively on funding from the private sector (e.g., pharmaceutical, biotechnology, and medical device manufacturers). These assessments are commissioned generally to fulfill the needs of the funding organizations .... The most likely causes of wide variance in results are the use of evidence of inferior quality or incomplete scope, applying inadequate resources, or allowing self-interest to affect the outcome in favor of the health plan.” S. Perry and M. Thamer, “Medical Innovation and the Critical Role of Health Technology Assessment,” JAMA, 282 (1999):186972, at 1870-71. Another type of industry-driven outcomes research is postmarketing studies of FDA-approved products. Typically, this research is undertaken to demonstrate which pharmaceuticals or other products are most cost-effective. Yet such research has no official requirements for scientific rigor, and studies sometimes are more of a marketing device than a bona fide research effort. The only requirement is that the results of such studies not be used in official marketing. A.L. Hillman et al., “Avoiding Bias in the Conduct and Reporting of Cost-Effectiveness Research Sponsored by Pharmaceutical Companies,” N. Engl. /. Med., 324 (1991):1362-65. See also A.F. Nelson, E.S. Quiter, and L.I. Solberg, “The State of Research Within Managed Care Plans: 1997 Survey,” Health Affairs, 17, no. 1 (1998):128-38; M. Friedberg et al., “Evaluation of Conflict of Interest in Economic Analysis of New Drugs Used in Oncology,”]AMA, 282 (1999):

1453-57; S. Krirnsky, “Conflict of Interest and Cost-Effectweness Analysis,”]AMA, 282 (1999):1474-75. 3 1, Committee on Quality of Health Care in American, Institute of Medicine, Crossing the Quulity Chasnr: A New Health System for the 21st Century (Washington, D.C.: National Academy Press, 2001):at Executive Summary, at 14. 32. G. Kolata, “Research Suggests More Health Care May Not Be Better,” New York Times, July 21,2002. 33. J.E. Wennberg, E.S. Fisher, and J.S. Skinner, ”Geography and the Debate Over Medicare Reform,” Health Affairs, at < h ttp :// ww w. hea I t h a f f a i rs. o r g/We b Exc I usi ves/ We n n b e r gWeb-Exc1-021302.htm> (posted February 13, 2002). 34. “Underuse is the failure to provide a health care service when it would have produced a favorable outcome for a patient.... Overuse occurs when a health care service is provided under circumstances in which its potential for harm exceeds the possible benefit.. .. Misuse occurs when an appropriate service has been selected but a preventable complication occurs and the patient does not receive the full potential benefit of the service.” M.R. Chassin and R.W. Calvin, “National Roundtable on Health Care Quality. The Urgent Need to Improve Health Care Quality,”]AMA, 280 (1998):1000-05, at 1002. See also T. Bodenheimer, “The American Health Care System: The Movement for Improved Quality in Health Care,” N. Engl.]. Med., 340 (1999):488-92;M.A. Schuster, E.A. McGlynn, and R.H. Brook, ”How Good Is the Quality of Health Care in the United States?,” Milbank Quarterly, 76,no. 4 (1998):517-63; M.R. Chassin, “Is Health Care Ready for Six Sigma Quality?,” Milbank Quarterly, 76,no. 4 (1998):565-91. 35. For a broad survey with numerous references, see Morreim, supra note 29. 36. J. Avorn and D.H. Solomon, “Cultural and Economic Factors That (Mis)shape Antibiotic Use: The Nonpharmacologic Basis of Therapeutics,” Annals of Intenial Medicine, 133 (2000): 128-35; H.S. Gold and R.C. Moellering, “Antimicrobial-Drug Resistance,” N.Engl.]. Med., 335 (1996):1445-53,at 144546; R. Gonzales et al., “Decreasing Antibiotic Use in Ambulatory Practice,”]AMA, 281 (1999):1512-19; N. Joshi and D. Milfred, “The Use and Misuse of New Antibiotics,” Archives of Internal Medicine, 155 (1995):569-77;D.N. McKay, “Treatment of Acute Bronchitis in Adults Without Underlying Lung Disease,”]ournal of General Internal Medicine, 11 (1996):557-62; WJ. Hueston, “Antibiotics: Neither Cost Effective nor ‘Cough’ Effective,”]ournu1 of Family Practice, 44 (1997):261-65; G.L. Fraser et al., “Antibiotic Optimization: An Evaluation of Patient Safety and Economic Outcomes,” Archives oflntemal Medicine, 157 (1997): 1689-94;A. Nyquist et al., “Antibiotic Prescribing for Children with Colds, Upper Respiratory Tract Infections, and Bronchitis,” ]AMA, 279 (1998):875-77; B. Schwartz, A.G. Mainous, and S.M. Marcy, “Why Do Physicians Prescribe Antibiotics for Children with Upper Respiratory Tract Infections?,” IAMA, 279 (1998):881-82;Chassin and Calvin, supra note 34;R. Gonzales, J.F. Steiner, and M.A. Sande, “Antibiotic Prescribing for Adults with Colds, Upper Respiratory Tract Infections, and Bronchitis by Ambulatory Care Physicians,” IAMA, 278 (1997):901-04; L. Culpepper and J. Sisk, “The Development of Practice Guidelines: A Case Study of Otitis Media with Effusion,” in EH. Boyle, ed., Getting Doctors to Listen: Ethics and Outcomes Data in Context (Washington, D.C.: Georgetown University Press, 1998):at 77-85;S.F. Dowell et al., “Principles of Judicious Use of Antimicrobial Agents for Pediatric Upper Respiratory Tract Infections,” Pediatrics, 101,suppl. (1998):163-65;S.F. Dowell et al, “Otitis Media - Principles of Judicious Use of Antimicrobial Agents,” Pediatrics, 101,suppl. (1998):165-71;A.D. Fix, G.T. Strickland, and J. Grant, “Tick Bites and Lyme Disease in an Endemic Set-

232

The Journal of Law, Medicine t 9Ethics ting,” JAMA, 279 (1998): 206-10. 37. B.G. Ewigman et al., “Radius Study Group. Effect of Prenatal Ultrasound Screening on Perinatnl Outcome,” N. Engl. /. Med., 329 (1993): 821-27; R.L Berkowia, “Should Every Pregnant Woman Undergo Ultrasonography?,” N. Engl. J. Med., 329 (1993): 874-75. 38. H.M. Krumholz et al., “Thronibolytic Therapy for Eligible Elderly Patients with Acute Myocardial Infarction,” JAMA, 277 (1997): 1683-88; C.L. Pashos et al., “Trends in the Use of Drug Therapies in Patients with Acute Myocardial Infarction: 1988 to 1992,”Journalof the American College of Cardiology, 23 (1994): 1023-30; D.C. Franceset al., “Outcome Following Acute Myocardial Infarction,” Archives of InternalMedicine, 159 (1999): 1429-36. 39. T.J. Wang and R.S. Stafford, “National Patterns and Predictors of R-Blocker Use in Patients with Coronary Artery Disease,” ArchivesofInternalMedicine, 158 (1998): 1901-06. See also L.N. Newcomer, “Physician, Measure Thyself,” Health Affairs, 17, no. 4 (1998): 32-35; T.M. Burton, “An H M O Checks Up on Its Doctors’ Care and Is Disturbed Itself,” Wall Street Journal,July 8, 1998, a t A l , A8; R. Winslow, “Studies Show Doctors Underprescribe Beta Blockers for Heart Attack Patients,” Wall Street Journal, August 19, 1998, at B5; Frances et al., supra note 38; R.M. Califf and C.M. O’Connor, “R-blocker Therapy for Heart Failure: The Evidence Is in, Now the Work Bepins,”JAMA, 283 (2000): 1335-37, at 1336. In another study, “less than 50% of cardiologists’ patients were taking R-blockers.” M.T. Donohoe, “Comparing Generalist and Specialty Care: Discrepancies, Deficiencies, and Excesses,” Archives of Internal Medicine, 158 (1998): 1596-608, at 1597. Likewise, it is well-known that patients with congestive heart failure can benefit greatly from angiotensin-converting enzyme (ACE) inhibitor drugs. Yet in one study of patients with congestive heart failure, “only three quarters of eligible patients were taking an [ACE] inhibitor, and only 60% of those were at doses known to be efficacious.” Id. 40. A.I? Legorreta et al., “Compliance with National Asthma Management Guidelines and Specialty Care,” Archives of Internal Medicine, 158 (1998): 457-64; T.V. Hartert e t al., “Inadequate Outpatient Medical Therapy for Patient with Asthma Admitted to Two Urban Hospitals,” AmericanJournal of Medicine, 100 (1996): 386-94 ; D.P. Havranek et al., “Process and Outcome of Output Management of Heart Failure: A Comparison of Cardiologists and Primary Care Providers,” American Journal of Managed Cure, 2, suppl. (1996): S6412, at S10. See also Donohoe, supra note 39, at 1599. 41. J.P Weiner et al., “Variation in Office-Based Quality: A Claims-Based Profile of Care Provided to Medicare Patients with Diabetes,” JAMA,273 (1995): 1503-08; L.L. Leape, “Translating Medical Science into Medical Practice: Do We Need a National Medical Standards Board?,”]AMA, 273 (1995): 1534-37; H.M. Harris, “Disease Management: New Wine in New Bottles?,” AnnalsofInternalMedicine, 124 (1 996): 838-42; Newcomer, supra note 39; R.S. Epstein and L.M. Sherwood, “From Outcomes Research to Disease Management: A Guide for the Perplexed,” Annals of Internal Medicine, 124 (1 996): 832-37; Donohoe, supra note 39, at 1600; Burton, supra note 39. 42. Donohoe, supra note 39. 43. At the beginning of the study, “handwashing compliance before and after defined events was 9% and 22% for health care workers in the medical ICU and 3% and 13% for health care workers in the cardiac surgery ICU, respectively. After the educatiodfeedback intervention program, handwashing compliance changed little. ...” Of note, after an alcohol-based waterless handwashing antiseptic was made easily available by each bed,

233

handwashing compliance improved to 48 percent. Still, that is less than half the desirable level. WE. Bischoff et al., “Handwashing Compliance by Health Care Workers: The Impact of Introducing an Accessible, Alcohol-Based Hand Antiseptic,” Archives of Internal Medicine, 160 (2000): 1017-21, at 1017. 44. L.N. Newcomer, “Medicare Pharmacy Coverage: Ensuring Safety Before Funding,” Health Affairs, 19, no. 2 (2000): 59-62, at 60 (citations omitted). 45. E.A. McGlynn and R.H. Brook, “Keeping Quality on the Policy Agenda,” Health Aflaim, 20, no. 3 (2001): 82-90, at 83. 46. E.C. Burton, D.A. Troxclair, and WE Newman, “Autopsy Diagnoses of Malignant Neoplasms: How Often Are Clinical Diagnoses Incorrect?,”]AMA, 280 (1998): 124548 (finding a 44% discordance between clinical and autopsy diagnoses of malignant neoplasms). See also G.D. Lundberg, “Low-Tech Autopsies in the Era of High-Tech Medicine,” IAMA, 280 (1998): 1273-74. Another study comparing autopsy findings to premortem diagnoses found that in a third of the cases, the autopsy detected an unexpected pathological diagnosis directly contributing to the patient’s death. Additional unexpected diagnoses were found in nearly 80 percent of the cases. S. Durning and L. Cation, “The Educational Value of Autopsy in a Residency Training Program,” Archives of Internal Medicine, 160 (2000): 997-99. See also T.C. Mort and N.S. Yeston, “The Relationship of Pre Mortem Diagnoses and Post Mortem Findings in a Surgical Intensive Care Unit,” Critical Cure Medicine, 27 (1999): 299-303 (finding a 41 percent rate of discrepancies between pre- and post-mortem diagnoses in a surgical intensive care unit; of these errors, 85 percent were undiagnosed infectious processes). 47. “[AIII residents we tested had great difficulty in identifying 12 commonly encountered and imponant events. Residents were incorrect 4 of 5 times, improved little with year of training, and were not more accurate than a group of medical students. Indeed, trainees in both residencies were less accurate than students [for certain kinds of heart sounds] .... me found minimal gains, if any, as a result of residency training. Deficiencies of this type will probably persist even after residents enter practice. Indeed, increasing evidence in the literature seems to suggest that errors in physical diagnosis are commonly encountered among generalists. These errors may even lead to greater utilization of resources and a higher cost of care.” S. Mangione and L.Z. Nieman, “Cardiac Auscultatory Skills of Internal Medicine and Family Practice Trainees: A Comparison of Diagnostic Proficiency,” JAMA,278 (1997): 717-22, at 721. 48. The chief brand names are Celebrex and Vioxx. 49. Newcomer, supra note 44,at 60. “A month’s prescription of either Celebrex or Vioxx costs about $80, many times the cost of generic pain relievers.” M. Peterson, “Study Finding Celebrex Safer Was Flawed, Journal Says,” New York Times, June 1,2002. 50. €? Judi, A.WS. Rutjes, and EA. Dieppe, “Are Selective COX 2 Inhibitors Superior to Traditional Non-Steroidal AntiInflammatory Drugs?,” BMJ, 324 (2002): 1287-88; K.L. Francis et a]., “Celecoxib versus Diclofenac and Omeprazole in Reducing the Risk of Recurrent Ulcer Bleeding in Patients with Arthritis,” N. Engl. J. Med., 347 (2002): 2104-10; D.Y. Graham, “NSAIDs, Helicobacter pylori, and Pandora’s Box,” N. Engl. J. Med., 347 (2002): 2162-64. 51. J. Avorn, M. Chen, and R. Hartley, “Scientific versus Commercial Sources of Influence on the Prescribing Behavior of Physicians,” American]oumal of Medicine, 73 (1982):4-8; R.K. Schwartz, S.B. Soumerai, and J. Avorn, “Physician Motivations for Nonscientific Drug Prescribing,” Social Science and Medicine, 28, no. 6 (1989): 577-82. 52. J.l? Orlowski and L. Wateska, “The Effect of Pharmaceu-


tical Firm Enticements on Physician Prescribing Patterns,” Chest, 102, no. 1 (1992): 270-73. 53. A.R. Feinstein, “System, Supervision, Standards, and the ‘Epidemic’ of Negligent Medical Errors,” Archives of Internal Medicine, 157 (1997): 1285-89, at 1286. Feinstein goes on to note: “The combination of shortened durations of time for both the patient in [the] hospital and the house officer on a service has reduced the house officer’s sense of continuity of care within the hospital ... and has increased the difficulties of maintaining rigorous patterns of supervision and discussion.” Id. at 1287. 54. J.D. Lantos and J. Frader, “Extracorporeal Membrane Oxygenation and the Ethics of Clinical Research in Pediatrics,” N. Engl. ]. Med., 323 (1990):409-13. 55. In scientific research, evaluation of diseases and treatments “generally requires a priori hypotheses, randomization (to eliminate selection bias and confounding), homogeneous patients at high risk for the outcome, experienced investigators who follow a protocol, a comparative measure such as a placebo (if ethical), and intensive follow-up to ensure compliance. Under these circumstances, if a treatment proves to be better than a placebo (or a comparative measure), one can be reassured that the treatment can work. However, questions may remain about the ability of the treatment to work adequately in a broader range of patients and in usual practice settings in which both patients and providers face natural barriers to care.” Epstein and Sherwood, supra note 41, at 833. See also A.R. Feinstein and R.I. Horwitz, “Problems in the ‘Evidence’ of ‘Evidence-Based Medicine,”’ American Journal of Medicine, 103 (1997): 529-35; K.B. Wells and R. Sturm, “Care for Depression in a Changing Environment,” Health Affairs,14, no. 3 (1995): 78-89, at 80. “Clinical trials are not real life. To assess efficacy in as unconfounded a manner as possible, trials sometimes exclude certain patients (e.g., the elderly, the very young, those too sick, or those taking certain other medications). Any special vulnerability to adverse events in those groups will be missed.” M.A. Friedman et al., “The Safety of Newly Approved Medicines: Do Recent Market Removals Mean There Is a Problem?,”]AMA, 281 (1999): 1728-34, at 1733. 56. Epstein and Sherwood, supra note 41. “The failure to account for the effects of comorbid and associated conditions on the clinical outcome of chronic diseases is reflected in the common disjunction that occurs between the efficacy of an intervention, such as a drug used in a clinical trial, and the lack of effectiveness of the same drug used in clinical practice.“ R.B. DeBusk et al., “Chronic Disease Management,” Archives of Internal Medicine, 159 (1999):2739-42, at 2740. Regarding coronary bypass surgery, Gellins and colleagues observed that “only 4 to 13 percent of the patients who now undergo this operation would meet the eligibility criteria for the randomized controlled trials that established its efficacy.” Gellins, Rosenberg, and Moskowitz, supra note 10, at 694. See also Feinstein, supra note 10. 57. DeBusk et al, supra note 56, at 2740. 58. I? Jong et al., “Prognosis and Determinants of Survival in Patients Newly Hospitalized for Heart Failure,” Archives of Internal Medicine, 162 (2002): 1689-94, at 1692. Mainly these research subjects were white male populations with a mean age of about 60 and few comorbidities. 59. Id. at 1692. 60. Leape, supra note 41; S.H. Woolf, “Practice Guidelines: A New Reality in Medicine. Ill. Impact on Patient Care,” Archives oflnternalMedicine, 153 (1993):2646-55, at 2646; E.J. Power, “Identifying Health Technologies That Work,” JAMA, 274 (1995): 205; Feinstein, supra note 10. 61. Friedman et al., supra note 55, a t 1729.

62. T.L. Delbanco, “Bitter Herbs: Mainstream, Magic, and Menace,” Annals of Internal Medicine, 121 (1 994): 803-04; D.W Gordon et al., “Chaparral Ingestion: The Broadening Spectrum of Liver Injury Caused by Herbal Medications,”]AMA, 273 (1995): 489-90; R.S. Koff, “Herbal Hepatotoxicity: Revisiting a Dangerous Alternative,”]AMA, 273 (1995): 502. 63. WC. Meeker and S. Haldeman, “Chiropractic: A Profession at the Crossroads of Mainstream and Alternative Medicine,” Annals of lntemal Medicine, 136 (2002):216-27, at 222. But see W.S. Smith et al., “Spinal Manipulative Therapy Is an Independent Risk Factor for Vertebral Artery Dissection,” Neurology, 60 (2003): 1424-28. 64. L. Kohn, J. Corrigan, and M. Donaldson, eds., Committee on Quality of Health Care in America, Institute of Medicine, To E n Is Human: Building a Safer Health System (Washington, D.C.: National Academy Press, 1999): at 1, 26. 65. Quality Interagency Coordination Task Force, Doing What Counts for Patient Safety: Federal Actions to Reduce Medical Errors and Their Impact, Report to the President (February 2000): at 29 (citing L. Kohn, J. Corrigan, and M. Donaldson, eds., supra note 64). See also L.L. Leape, “Foreword: Preventing Medical Accidents: Is ‘Systems Analysis’ the Answer?,” American Journal of Law 6 Medicine, 27, nos. 2 and 3 (2001): 14548; T.J.Krizek, “Surgical Error: Ethical Issues of Adverse Events,” Archives of Surgery, 135 (2000): 1359-66. 66. J. Lazarou, B.H. Pomeranz, and PN.Corey, “Incidence of Adverse Drug Reactions in Hospitalized Patients. A MetaAnalysis of Prospective Studies,”]AMA, 279 (1998): 1200-05, at 1203. 67. L. Landro, “The Informed Patient: Deadly Errors Dog Procedures at Doctors’Offices and Clinics,” Wall Street Journal, August 29, 2002. 68. “[Platients have wasted their time, money, and efforts receiving treatments that were not what they were represented to be or were harmful.” Schneiderman, supra note 3, at 91. 69. D.M. Eisenberg et al., “Unconventional Medicine in the United States,” N.Engl. J. Med., 328 (1993): 246-52, at 250. 70. Id. at 24849. 71. Modalities showing the greatest increase in use were herbal medicine, massage, megavitamins, self-help groups, folk remedies, energy healing, and homeopathy. D.M. Eisenberg et al., “Trends in Alternative Medicine Use in the United States, 1990-1997,” JAMA, 280 (1998): 1569-75. 72. Moseley et al., supra note 13, at 81. 73. Id. at 87. See also Felson and Buckwalter, supra note 12; S . Horng and F.C. Miller, “Is Placebo Surgery Unethical?,” N. Engl.]. Med., 347 (2002): 137-39. 74. McGlynn and Brook, supra note 45, at 83. 75. Mello and Brennan,supra note 19, at 110. “Although the cost has decreased somewhat in the past several years, it remains in the neighborhood of $80,000.” Id. See also Kolata and Eichenwald, supra note 22. 76. Peters and Rogers, supra note 20. See also J.H. Ferguson, M. Dubinsky, and PJ. Kirsch, “Court-Ordered Reimbursement for Unproven Medical Technology: Circumventing Technology Assessment,”J4MA, 269 (1993): 21 16-21. 77. Hoffman, supra note 21. 78. Another example is cerebral revascularization featuring superficial anastomosis of temporal to middle cerebral artery. See EJ. Cohen, “The Placebo Is Not Dead: Three Historical Vignettes,” IRE, 20, nos. 2-3 (1998): 6-8. 79. F.A. Tausk, ”Alternative Medicine: Is It All in Your Mind?,“ Archives of Dermatology, 134 (1998): 1422-25, at 1423 (citing A.H. Roberts et al., “The Power of Nonspecific Effects in Healing: Implications for Psychosocial and Biological Treatments,” Clini-

234

The journal of Law, Medicine &Ethics cal Psychology Review, 13 (1993): 375-91). 80. J.E. Wennberg, ed., The Dartmouth Atlas ofHealth Care in the United States (Chicago: American Hospital Publishing, 1996); J.E. Wennberg, “Understanding Geographic Variations in Health Care Delivery,” N. Engl. J. Med., 340 (1999): 52-53. 8 1. J.E. Wennberg, “Which Rate Is Right?,” N. Engl. J. Med., 314 (1986): 310-11, at 310. 82. Wennberg, Fisher, and Skinner. supra note 33. 83. E.H. Morreim, “Defined Contribution: From Managed Care to Patient-Managed Care,” CatoJournal, 22, no. 1 (2002):

definedcontribution.pdf>; J. Robinson, “The End of Managed Care,” JAMA, 285 (2001): 2622-28; M. Freudenheim, “H.M.O. Costs Spur Employers to Shift Plans,” New York Tmes, September 6, 2000; Parrish, supra note 86; R. Winslow and C. Gentry, “Health-Benefits Trend: Give Workers Money, Let Them Buy a Plan,” Wall Street lournal, February 8, 2000, at Al, A12; G . Shaw, “The Defined Contribution Solution,” Managed Healthcure News,February 2001, at 1. 88. See, e.g., HealthMarket, at ; Lumenos, at ; Vivius, at ; Definity Health, at ; myhealthbank, at ; HealthAllies, at cwww.healthallies.com>. 89. Health reimbursement arrangements (HRAs)can be rolled over from one year to the next. This guidance was released on July 15,2002, and applies to Section 105 and 106 of the Internal Revenue Code. See Rev. Rul. 2002-41,2002-28 I.R.B., available at ~http://www.unclefed.com/Tax-Bulls/2002/rrO2-4l.pdf> (“Holding: Employer-provided coverage and medical care expense reimbursements made under t h e reimbursement arrangement that allows unused amounts to be carried forward ... are excludable from gross income under §§ 106 and

103-20. 84. Portions of this discussion rely on an earlier publication.

See id. 8 5 . J. Gabel et al., “Job-Based Health Benefits in 2002: Some Important Trends,” Health Affairs, 21, no. 5 (2002): 143-51. 86. R. Winslow and L. McCinley, “Back on the Front Burner,” Wall Street Journal, February 21, 2001, at R3; M. Parrish, “A New Day Dawns ... When Patients Buy Their Own Health Care,” Medical Economics, 78, no. 5 (2001): 95-1 11; D. Blumenthal, “Controlling Health Care Expenditures,” N. Engl. J. Med., 344 (2001): 766-69; J.V Jacobi and N. Huberfeld, “Quality Control, Enterprise Liability, and Disintermediation in Managed Care,” Journalof Law, Medicine &Ethics, 29, nos. 3 and 4 (2001): 30522, at 309-12; R Calvin and A. Milstein, “Large Employers’ New Strateges in Health Care,” N. Engl. J. Med., 347 (2002): 93942. 87. See Wye River Group on Healthcare et al., An Employer’s Guide to Consumer-DirectedHealthcare Benefits(2001),available at . See also K.E. Martin, Shifting Responsibilities: Models of Defined Contribution (Washington, D.C.: Academy for Health Services Research and Health Policy, 2002), available at < http://hcfo.net/pdf/

105 ....”). 90. Editorial, “Consumer-First Health Care,” Wall Streetlournal, July 21, 1994, at A12. 91. M.L. Berk and A.C. Monheit, “The Concentration of Health Care Expenditures, Revisited,” Health Aflairs, 20, no. 2 (2001): 9-18, at 12. 92. For further discussion, see Morreim, supra note 83. 93. Gabel et al., supra note 85. 94. Eisenberg et al., supra note 69.

235