Health Affairs, 22, no. 1 (2003): 125-137
doi: 10.1377/hlthaff.22.1.125
© 2003 by Project HOPE
 
New Online
 * Getting Health Reform Done
 * After the State of the Union
 * Incremental Reform
 * E-Health in Developing World
 * Most-Read Articles in 2009
This Article
* Abstract Freely available
* Reprint (PDF)
* Submit a response to this article
* Alert me when this article is cited
* Alert me when Comments are posted
* Alert me if a correction is posted
Services
* E-mail this article to a friend
* Similar articles in this journal
* Alert me to new issues of the journal
* Add to My Personal Archive
* Download to Citation Manager
*Reprints & Permissions
Citing Articles
* Citing Articles via HighWire
* Citing Articles via Web of Science (18)
* Citing Articles via Google Scholar
Google Scholar
* Articles by Daniels, N.
* Articles by Sabin, J. E.
* Search for Related Content
PubMed
* Articles by Daniels, N.
* Articles by Sabin, J. E.
Related Collections
* Ethical Issues
* Pharmaceuticals

Pharmacy Benefits

An Ethical Template For Pharmacy Benefits

Norman Daniels, J. Russell Teagarden and James E. Sabin

PROLOGUE: "Does everybody need the purple pill?" the director of a major employee benefit program lamented at a recent national health policy conference, following a dramatic upsurge in prescriptions to treat indigestion. In this paper Norman Daniels, Russell Teagarden, and James Sabin propose a decision guide, or ethical template, to help patients, clinicians, and the public learn how to share medical resources fairly. They argue that if people affected by benefits decisions understand the rationales behind the decisions, they are more likely to be able to accept limits. By framing decisions in this way, purchasers and consumers can make decisions that are informed, rational, equitable, and defensible.

Whether lifestyle drugs, experimental drugs, or life-saving maintenance medications for chronic conditions, retail pharmacy outlays topped $150 billion in the United States in 2001, an increase of 17 percent over the previous year. Tight budgets and a tight economy make it imperative that some hard decisions be made about pharmacy benefits. The ethical template offers a framework that the public (consumers), benefit providers (purchasers), and suppliers (pharmaceutical companies) can use to present, examine, and defend ethically relevant decisions on pharmacy benefits. The authors argue that using the ethical template will help to educate the public about how to set limits that are both fair and reasonable. They warn that failure to enlist public support and understanding can lead to escalating distrust, litigation, and costs.

Professor of ethics and population health at the Harvard School of Public Health, Daniels has written widely on the philosophy of science, ethics, political and social philosophy, and medical ethics. Teagarden is vice-president, clinical practices and therapeutics, for Medco Health Solutions and also holds academic appointments at Rutgers College of Pharmacy, Ohio Northern University College of Pharmacy, and the Philadelphia College of Pharmacy and Sciences at the University of the Sciences in Philadelphia. Sabin is a clinical professor of psychiatry at Harvard Medical School, director of the Harvard Pilgrim Health Care Ethics Program, and cofounder of Harvard’s Center for Ethics and Population Health.


   Abstract
 
We propose an ethical template for pharmacy benefits and a fair process for using it. The template delineates four levels of decisions about pharmacy coverage, connecting ethically acceptable types of rationales for limits with decisions made at each level. It provides a framework for organizing ethically relevant reasons for coverage (or the tiered copayments). The process for using the template assures accountability for the reasonableness of benefit decisions. It requires transparency and relevance of rationales for limit setting and revisability of decisions, including through fair procedures for appeals. The template and the process facilitate broader public learning about fair limit setting.


Pharmacy benefits have become the problem child of health care coverage. Rising demand, increased use, and surging costs terrified purchasers even when the economy was good. Limiting access to some drugs, whether through restrictive coverage policies or tiered copayments, has often met with resentment and distrust from patients and clinicians. Placing this problem child in a properly equipped classroom, with appropriate educational tools and processes, can school us all in how to develop, understand, and accept fair limits.

Most public encounters with limit setting in health care, both in the United States and abroad, have been episodic and involved big-ticket items. These often have featured denials of "last chance" treatments, creating life-and-death dramas. A year ago in Boston, for example, Belynda Dunn, an HIV-positive patient, was denied a liver transplant by a managed care plan on the grounds that it was "experimental" for patients whose immune systems might be compromised. That plan then contributed to a charity set up to "rescue" Dunn and others like her from the consequences of reasonable limit setting.

From episodes like this one, the public is likely to draw confused lessons about the necessity and reasonableness of limits to care. It would be easy to conclude from the Dunn case that there is always a way around denials if one advocates vigorously enough, can point an accusing finger at the "evildoer" who is denying care, and can attract public sympathy. The public furor following denial of funding for Child B in England, like similar cases in Norway, Australia, and New Zealand, shows the great difficulty of accepting limits even when public officials replace private insurers as the "evildoers" in question.1

In contrast, pharmacy benefit management lacks such high drama. This mundane dullness may provide a distinct advantage in helping society to move along a learning curve about setting limits fairly. We are more prepared to learn about limits when we are not in extremis.

In addition, pharmaceuticals potentially touch everyone, especially the chronically ill, for whom they are often the main treatments. What is generally seen as the problem—growing patient demand for higher-cost drugs, fed by direct-to-consumer advertising—actually offers an opportunity to teach large numbers of us about the need to find ways of setting limits that we can accept as reasonable and fair. Pharmacy benefits may provide a last chance to come to grips with the broader issue of health cost management. If this is not accomplished with proper public learning, it is an area that could bog down the health system in incessant conflict, escalating distrust, litigation, and rising costs.

   A Tool For Learning, A Process For Using It
 Top
 A Tool For Learning,...
 The Ethical Template
 Using The Template
 Social Learning About...
 NOTES
 
We take up this challenge by viewing the management and implementation of pharmacy benefits as a school in which patients, clinicians, and the general public can learn how to share valuable medical resources fairly. We equip that classroom with a tool for learning—specifically, an ethical template for pharmacy benefits—and we describe a process for using that template. The template delineates four levels of decisions about pharmacy coverage and connects ethically acceptable types of rationales for limits with decisions made at each level. The template is useful for evaluating the basis for tiered copayments and not just explicit exclusions from coverage. In short, the template provides a framework for organizing reasons for coverage or noncoverage that people can accept as relevant to meeting health care needs fairly under resource constraints.

Publicity, relevance, and revisability. The process for using the template should assure the transparency or publicity and the relevance of the rationales underlying limit-setting decisions. People affected by decisions will accept limits only if they see the rationales for those limits as relevant to meeting drug needs fairly under reasonable resource constraints. The ethical template makes it easier to understand and evaluate rationales; in this sense, it gives us a graspable framework for delineating relevant reasons. (It goes without saying that we presuppose an honest, nondeceptive representation of rationales throughout.) The process also allows for the revisability of decisions in light of new evidence and arguments, including the learning that comes from a fair appeals process.

Together these three elements assure what we elsewhere have termed "accountability for reasonableness."2 However, acceptance of the reasonableness of limits also depends on social learning about fair limit setting. A public that has no grasp of the need for fair limits is unlikely to accept even reasonable limits. In that spirit, the ethical template is an important tool for societal learning.

Corroboration. The ethical template thus embodies the type of approach to decision making that others have called for.3 From a consumer and market perspective, it carries out the informative and educative role regarding the value and cost of drug therapy called for by J.D. Kleinke in this journal.4 We also agree with Uwe Reinhardt’s recent comment that "whatever means employers and government ultimately adopt to shift more of the rising cost of prescription drugs onto patients, one can expect much rancor over the practice—and possibly much litigation—unless the underlying formularies or therapeutic groupings can be explained to physicians, patients, and juries with appeal to scientifically sound cost-benefit analyses."5 The ethical template provides a framework for understanding what kinds of explanations are required and how they need to be structured.

   The Ethical Template
 Top
 A Tool For Learning,...
 The Ethical Template
 Using The Template
 Social Learning About...
 NOTES
 
The ethical template maps decisions about drug coverage limitations onto generic rationales (Exhibit 1Go). We call this mapping a "generic template" because it serves as a pattern or guide for constructing and evaluating pharmacy benefits. Unlike true templates, however, the benefits patterned on it are not replicas or duplicates of the template. Rather, the pattern is a guide to the overall structure but not to its fine details. As the details of specific coverage decisions and the rationales underlying them are worked out, the fine structure of various reasonable benefit plans emerge from the same basic pattern or template. When a plan is fully specified, the generic template is converted into a detailed blueprint that shows the pattern of values that governs the actual plan.


View this table:
[in this window]
[in a new window]
 
EXHIBIT 1 Overview Of Elements Of The Ethical Template For Pharmacy Benefits

 
A hierarchy of coverage decisions. The template we describe reflects a hierarchy of four levels of decision making about drug coverage: (1) the categories and subcategories of drugs to be covered; (2) the selection of drugs within categories; (3) the indications or uses for which selected drugs may be used; and (4) further use limitations on the amount (quantity/unit time or dosage form) or source (retail or mail order) of drugs that may be covered.6 Many people see these decisions not as strict limits on what drugs they can obtain, but as different tiers of copayments or as limits on prescription size or the source they must use.

Level 1. To illustrate these levels of coverage decisions, consider how a hypothetical pharmacy benefit might view Imitrex (sumatriptan), a common treatment for migraines. Since migraines are a major form of pathology for many people, it would be important to include in the benefit the category (or subcategory) of triptans, a large family of drugs that greatly reduces the seriousness of migraine attacks. Ideally, the identification of needs and the specification of drug categories at Level 1 would be based on evidence about the disease profile of the covered population and evidence about effective ways of managing these diseases.

Level 2. Several of the triptans are generally interchangeable, having similar effectiveness and side effects for most people. Because of discounts and rebates, pharmacy benefit managers (or other administrators and purchasers) can usually secure some of these drugs at lower costs than others can. The lower cost is potentially an advantage for all stakeholders in the plan, depending on how the savings are distributed. Since, in our hypothetical example, Imitrex offers such a cost advantage, it will be selected for coverage over others not offering such an advantage. Alternatively, Imitrex may be available with little or no copay, while the drugs with less cost advantage are available, if at all, only with higher copays.

Levels 3 and 4. Imitrex has proven efficacy, however, only for treatment, not prevention, of migraines. Accordingly, at the drug indication level (Level 3), Imitrex is covered only for treating migraines. Finally, since there is evidence that most people with migraine problems do not have more than a few episodes a month, a benefit plan can avoid wasteful spending if it limits coverage to a quantity sufficient to treat only a few episodes (Level 4), provided, as we emphasize later, there is an easy way for patients who need more treatments to get them.7

A pattern of ethical concerns and rationales. One important feature revealed by the template is that different types of ethical issues are raised at different levels in the hierarchy. At the drug category and drug indication levels (1 and 3), the ethical issues focus on meeting important health needs. At the drug selection and drug use levels (2 and 4), the central ethical issue is how to gain a cost advantage without placing undue burden or risk on certain patients—either of noncoverage itself or of too much hassle obtaining an exception to the policy.

The generic rationales that emerge at the different levels reflect this pattern. We focus here on three such generic rationales, two at Levels 1 and 3 and one at Levels 2 and 4. These rationales are not intended to be exhaustive, although they are central. Each is widely accepted at a general or abstract level, but all raise controversy in their application, which is an opportunity for social learning.

Generic rationales and priority principles. Treatment versus enhancement. Each of the generic rationales involves what may be thought of as a rationing principle.8 More euphemistically, we can think of each rationale as appealing to a priority that reflects a collective goal in providing the benefit. One such priority, evident at Levels 1 and 3, is to give weight to providing drug treatments (including some forms of prevention) for major medical problems and to give little or no weight to interventions that involve the enhancement or improved control of otherwise normal conditions. This priority principle is widely invoked both in the United States and abroad and has an important influence in shaping medical insurance coverage.9 For example, consider a typical drug indication (Level 3) coverage limit: The topical tretinoin Retin-A is to be covered for treatment of acne but not for control of wrinkles that result from aging. The relevant generic rationale for this restriction would then appeal to the priority principle that values treatment of pathology more than it does cosmetic enhancements.

This priority principle is not always uncontroversial in its application. Consider the Level 1 exclusion made by many health plans that do not cover oral contraceptives (occasionally this will be a Level 3 exclusion, allowing their use for medical purposes but not for family planning). Even if preventing normal pregnancy cannot count as treatment of pathology, many would argue that support for reproductive planning is itself an important social goal, and we ought to share the burdens of financing it through drug coverage. Indeed, vasectomy and tubal ligation are generally covered even when they are aimed at reproductive planning. Making rationales for limits explicit means that there is greater pressure to develop consistent, justifiable patterns of coverage based on a coherent set of priorities.

This point is given special relevance by the outcry about gender bias that arose several years ago when Viagra (sildenafil citrate) was covered by many health plans that denied coverage for oral contraceptives. If both were "lifestyle" drugs and not simply treatments of illness, then why one and not the other? Another controversial exclusion by some plans concerns costly treatments for infertility. Although infertility often involves clear pathology, so that treatment would ordinarily be covered, the exclusion is justified in the minds of some by the (unpersuasive) claim that treating infertility is just a lifestyle choice. Similarly, confusion often surrounds the rationales underlying Level 1 decisions to exclude coverage for smoking deterrents and treatments or weight-loss agents. A real debate and deliberation about coverage policy at Level 1 would be furthered by the practice of making rationales for limits explicit.

Proven versus unproven effectiveness. A second priority principle that is prominent in decisions at Levels 1 and 3 is to provide coverage for interventions known to be effective and to exclude those that have been shown to be ineffective or whose effectiveness is not yet proven. The restriction on Imitrex for the treatment of migraines depends on this principle, for it is effective for treatment but unproven for prevention. This priority principle can sometimes be controversial in its application, as when an unproven use of a drug may offer some patients a last chance at treatment. Here, too, making the rationale explicit promotes clearer grounds for debate and deliberation.

Cost-worthiness. A third rationing principle requires that we aim for more rather than less cost-effective modes of treatment. The justification for this priority is that more needs are met by following this principle than not, given resource limits. Although this principle could also be invoked at Level 1 or 3, we have found few examples in the United States of that happening (but see the discussion of Viagra in the next section).10 At the drug selection and drug use levels (2 and 4), however, some version of the third rationing or priority principle plays a dominant role, since cost advantage is a crucial driver of these decisions.

Constraints on seeking cost advantage. Nevertheless, the appeal to cost advantage should not operate alone. Further ethical considerations should qualify or constrain this priority principle. Specifically, the cost advantage to the purchaser or collective in these cases should not impose major risks or burdens on people affected by the limitations.

To illustrate, suppose that two drugs are considered to be generally interchangeable. Using the less expensive one, then, will not impose a major burden on most patients. If there are reasons that particular patients will not do well on the selected drug, there should be quick and easy ways for those patients to secure a more appropriate drug. We therefore have two important constraints on appeals to cost advantage: Not only must the selected drug be generally interchangeable with those that are excluded (or require a larger copay), but there must be a quick and efficient way to remedy the situation for patients for whom the interchangeability does not hold. The generic rationale for limits at Level 2 (and 4) would refer not only to the principle requiring cost-worthy treatment but also to the evidence about general interchangeability. There should also be publicly accessible evidence about the effectiveness of the process of obtaining exceptions for patients who truly need the drugs that are not selected.

One further ethical constraint on cost advantage has a bearing on complaints made by some purchasers and consumers against pharmacy benefit managers (PBMs). The complaint is that the cost advantages they obtain have not worked to the collective advantage of purchasers or enrollees and sometimes have disadvantaged them. The further ethical constraint identified in the template is the requirement that the benefits that derive from seeking cost advantage be distributed in a way that serves the shared interests of the various stakeholders—employers, enrollees, and PBMs. Minimally, this constraint means that there should be an accounting of how the savings from limits at Levels 2 and 4 are used to sustain the benefit, to stretch it further, or to reduce costs to the various stakeholders. In short, the cost advantage must be explained so all can see whether and how collective cost-lowering mechanisms, including discounts or rebates, work to help meet health care needs fairly—that is, how they meet the relevance requirement.

Because drug coverage limitations at Level 4 (that is, quantity of drug per unit time, or dose format, or drug procurement channel) also depend on claims of cost advantage, a similar structure for generic rationales applies at this level as at the drug selection level (Level 2). For example, the evidentiary basis for claiming that most patients need only so many migraine treatments per month should be supplied in a rationale for such a limit. In addition, a clear process that is quick and efficient for patients who need more treatment in a month should be specified. There must be a reduction in the risk to them when the collective scheme takes advantage of the cost advantage of paying for fewer treatments that may go unused. Finally, the savings that result must be used in a way that serves the varied but shared interests of different stakeholders in the plan.

Disagreement and deliberation. Developing and using the template does not resolve all of the disputes that will arise. Rather, it focuses those disputes on the underlying rationales for decisions. The template thus leaves room for the kind of variation among principles and rationales in diverse health plans. Whatever the rationales, however, accountability for reasonableness requires that they be public, be viewed as relevant by those affected by them, and be revisable in light of further evidence or argument. The template facilitates but does not substitute for scrutiny by various stakeholders of the adequacy or appropriateness of rationales.

   Using The Template
 Top
 A Tool For Learning,...
 The Ethical Template
 Using The Template
 Social Learning About...
 NOTES
 
We have developed the ethical template for two practical purposes: as a tool for design and management of pharmacy benefit plans, and as an educational tool or heuristic device to foster public deliberation and societal learning about setting limits fairly. The best way to show these purposes would be for us to report on the template’s actual use in practice and its demonstrated results. However, we cannot do that yet since the template is too new. An alternative would be for us to describe just how the template might be used in a hypothetical case. Since that would leave the template open to the objection that it is part of an unrealistic, even utopian, project, we adopt here a third approach. We describe some examples of actual practice where key elements of our approach are used, and we then suggest how the outcomes would be even better if the full approach were deployed. This approach provides an antidote to the charge that what we propose is not realistic.

Viagra coverage. In 1998 Harvard Pilgrim Health Care, a not-for-profit insurer, had to decide about coverage for Viagra, which had received Food and Drug Administration (FDA) approval in February of that year. The team charged with recommending a policy decision elected to consult with the Harvard Pilgrim Ethics Advisory Group (EAG). The EAG includes consumers, physicians, employers, and staff. It deliberates about cases brought for consultation and offers nonbinding advice about the ethical dimensions of the situation. We note that the EAG’s involvement of a broad group of stakeholders facilitates accountability for reasonable decisions by increasing the likelihood that rationales that are adopted will be seen as relevant and acceptable to those who are affected by them.

At the meeting, several participants applied terms such as "enhancement" and "lifestyle drug" to Viagra and questioned whether it should be covered. At this point, a female EAG member from a purchaser group spoke with great force: "I can’t believe that at the end of the twentieth century—after all that Freud taught us—that we are talking as if sexuality were an optional matter of ‘lifestyle.’ Sex is an essential aspect of our being. To reduce the ability to have intercourse to the status of a ‘life-enhancing’ activity is to belittle a patient’s valid desire to restore a basic function."

These comments led the EAG to see Viagra as an innovation offering important value. As a result, it answered the Level 1 question about whether the drug should be eligible for coverage positively. But it advised Harvard Pilgrim’s management that Viagra’s use should be compared with the value provided by alternative uses of the funds. It advised that if the organization made a decision to limit coverage (a Level 4 decision), it should explain the cost factor to members of the health plan and the public (exactly what the template would propose).

The fact that the EAG followed the steps suggested by the template suggests that the kind of reasoning the template asks for is not pie in the sky. Had the template actually been available, however, it would have reassured EAG participants that their reasoning fits well with the overall requirements of Level 1 decisions. The template leaves explicit room for disagreements about how to interpret basic concepts, such as treatment and enhancement. Its structure, however, assures that use of the template will foster coherence among decisions over time. The one-time consultation in this case would have been connected to a structure that embodies broader experience concerning many similar decisions.

In the aftermath of the EAG discussion, Harvard Pilgrim decided to cover four Viagra tablets a month at the discretion of the prescribing physician, thus imposing a Level 4 restriction. In announcing the policy, the medical director tried to educate members and the public about limits:

"I don’t want to be in the position of saying how often people should have intercourse. We are not telling physicians they can write prescriptions for only four pills....What we are saying is that we will pay for only four a month...."As to why Harvard Pilgrim settled on four pills as opposed to, say, eight, Dorsey [the medical director] admits there is no scientific logic to that number. "It demonstrates a commitment to using the resource pool to help make it possible for people to use the medication...but it also reflects the fact that Viagra is too expensive for us to pay for unlimited use"11

This narrative suggests that the clarity about the rationale for limits on drug coverage and lucid public explanation called for by the template promote improved decision making about a pharmacy benefit and foster public understanding of the need to make difficult Level 1 and Level 4 choices and to accept limits as disappointing but fair. Had the template itself been available, it would have added assurance that the type of decisions made in this case were not idiosyncratic or ad hoc but fit a pattern that had coherence and justification.

Selective serotonin reuptake inhibitors (SSRIs). Our second example is based on an article in this journal that describes how Kaiser Permanente, a prepaid group practice, manages coverage for SSRIs.12 SSRIs are more expensive than the tricyclic antidepressants that predominated when Prozac first came onto the market. Because SSRIs do not produce a higher cure rate than the older medications do, many coverage programs initially tried to exclude them as a category, except when tricyclics did not work (Level 1). Over time, however, from a combination of consumer demand, physician acceptance, favorable side-effect profiles, reduced risk of death from overdose, and studies that suggested cost-effectiveness, SSRIs have become the most widely used antidepressants.13 Not surprisingly, the discussion of how Kaiser Permanente handles SSRIs within its drug benefit simply assumes the Level 1 decision to include them as a category of antidepressant.

As called for by Level 2 of the template, the physician-governed Kaiser Permanente practices used the research pharmacists in its Drug Information Service to compare the four competing SSRIs (Celexa, Paxil, Prozac, and Zoloft) for clinical effectiveness. They concluded that none of the four was a clear "winner."14 Given comparable efficacy—or, more explicitly, general interchangeability—for treating depression, cost advantage became the appropriate determinant for choosing a preferred drug. At this point, Kaiser, which serves eight million members, used its bargaining power to seek discounts from the manufacturers.

While the article does not tell us which medication(s) were chosen or what members are told about the rationale, it reports in detail on how the organization interacts with physician prescribers. Drug education coordinators (trained clinical pharmacists) present the clinical and economic rationale behind the formulary choices. Had the template been available, the specific case here could have been shown to be consistent with a more general approach, and the template might have allowed comparisons to further cases.

Since no general rule can apply reasonably to all members of a large population, Kaiser builds in a simple, hassle-free approach to the exceptions process. If a Kaiser member is unresponsive to, intolerant of, or allergic to the formulary choice, the physician prescribes a nonformulary drug and indicates the reason on a special prescription form. This ease of applying the process is the only way to undercut the most common quality-oriented criticism of this kind of limit: namely, that it exposes patients who do not fit the general rule to medically burdensome outcomes.

The Kaiser approach to choosing a preferred SSRI, explaining the rationale to prescribing physicians and making clinically based exceptions, is precisely what the template calls for, again confirming that it does not make unrealistic demands. Had the template been available to the physician prescribers, it could have been used in their discussions with patients, a point we clarify in the next vignette.

A clinician encounter about drug dosage. A clinically trivial example illustrates how clarity about the rationale for limits can support education in the doctor-patient relationship. Some years back, one of us (Sabin) spoke as follows with an elderly street-smart man who was benefiting from Paxil: "You are taking 10 mg of Paxil each day. If I prescribe 20 mg and ask you to break the pill in half, it will cost the pharmacy less than if I prescribe the 10 mg pill size. The pharmacy can use the savings for other medications. Is it fair for me to ask you to break the pill in half?" After a few moments of thought, the patient emphatically agreed it was fair: "Doc, if anyone fights you on that one, they’re just out to break your [expletive]!"

This exchange represents the essence of Level 4 coverage decisions. The physician proposed a way of prescribing that saved money for the pharmacy budget in a prepaid program (Level 4). Savings would benefit other patients. The rationale was made clear to the patient, who endorsed it in his own vivid language.

Had the template been available, the clinician here would not only have been able to show the good common sense of the dosage regimen but would also have been able to demonstrate that it was part of a broader strategy to use resources wisely within the plan. The template encourages stakeholders to exchange pertinent information, and if the patient expressed concern about whether pill splitting was safe, a well-prepared clinician would be able to cite pertinent literature.15 More to the point, the physician could become part of a more general educational strategy if proper technology made tools available to clinicians. We have in mind something not beyond the reach of technological feasibility today—say, handheld computers with access to plan rules and rationales organized in the way that the template suggests. Armed in this way, physicians (and pharmicists) could educate themselves and could extend that schooling to their patients.

   Social Learning About Pharmaceutical Value For Money
 Top
 A Tool For Learning,...
 The Ethical Template
 Using The Template
 Social Learning About...
 NOTES
 
Although not a panacea for the many forces in medicine that stand in the way of the kind of deliberation we call for, the template can facilitate some positive steps already in practice. We conclude, however, by pointing to some features of the U.S. situation that make it harder to learn about the true value, or true value for money, of pharmaceuticals. In most countries, purchasers of drug benefits are the populations covered by them, through either national health systems or social insurance schemes. The interests of societal agents who manage the insurance are similar to the interests of the insured populations.16

In the United States, however, employers—the largest purchasers of private insurance—have various motivations and interests that may diverge from the health interests of their employees and their insured dependents. Kleinke argues, for example, that the high turnover rates in health plans—some 22 percent annually, he claims—mean that employers and health plans may not have an economic interest in the long-term savings that many drug regimens offer.17 Further, he cites Bryan Luce’s point that "the separation of pharmacy budgets from other entities...dictates that simple acquisition costs often determine coverage decisions," where these acquisition costs may have no real relationship to the value of the drugs—or to their value for money spent—to the covered population.18 Neither incentives nor regulation properly aligns the interests of purchasers with the interests of the populations for which they buy insurance. This misalignment reflects how the United States structures insurance pools, not the intrinsic ethical orientation of employers that purchase pharmacy coverage.

In countries where the interests of purchasers and managers are more closely aligned with those of the public they insure, social learning about limits to coverage is easier. Were an ethical template to be used in that context, there would likely be more agreement about how to achieve societal value for money through drug coverage. What counts as a "relevant rationale" is less likely to be controversial in such countries than in the United States.

We see the ethical template as a way to address the U.S. mismatch between the interests of purchasers and covered populations. If it is used as we have suggested, the template should facilitate discussion about the relationship between acquisition costs and more scientifically based calculations of cost-effectiveness (or cost-value) of certain drugs.19 With transparency, with a focus on relevant reasons that stakeholders can agree to, and with the opportunity to revise decisions, the discussion will generate understanding about the proper basis for limits to care. The social learning curve that results will then not only inform drug coverage decisions but will also contribute to broader democratic understanding of how to improve the regulation and design of the insurance system as a whole.

Another key context in which the template and its use are important is in the education of regulators and legislators who aim to improve the ways in which the insurance system, including pharmacy benefits, meets public needs. Rather than encouraging piecemeal mandating and regulatory intrusions in specific decisions, the template and the lessons it teaches put the focus on improving the transparency and elements of fair process that surround such decisions. This result would work to the advantage of all parties.

With the exception of Oregon’s effort a decade ago, the United States has resisted any effort to publicly discuss limit setting to help the population learn about setting limits fairly. We have not had the national commissions set up in a number of other countries that have universal coverage to propose principles for setting limits or priorities. Whatever their limitations, these commissions at least begin a public effort at education about the need for limits and a fair way of establishing them.20 Although cost containment is widely talked about in the United States, the pretense is always that this effort concerns only the elimination of waste or unnecessary services and never anything more. This pretense is maintained at the federal and state levels, as well as in the language of private insurance plans. Far from reassuring the public, the pretense builds suspicion and distrust.

Our policy proposal for public education is to encourage private and public corporations and agencies here and abroad that make decisions about pharmacy benefits to use the ethical template in the context of procedures that are accountable for reasonableness.21 Perhaps, as many observers claim, Americans are "different" and lack the concern about solidarity or sharing that people in other countries have. Without a serious effort at education, using proper tools, the claim is untested and possibly a self-fulfilling prophecy.

   Editor's Notes
 
This work was supported in part by grants from the Merck Company Foundation and the Open Society Institute.

   NOTES
 Top
 A Tool For Learning,...
 The Ethical Template
 Using The Template
 Social Learning About...
 NOTES
 

  1. C. Ham and S. Pickard, Tragic Choices in Health Care: The Case of Child B (London: King’s Fund, 1998).
  2. N. Daniels and J. Sabin, "The Ethics of Accountability in Managed Care Reform," Health Affairs (Sep/Oct 1998): 50–64; N. Daniels, "Accountability for Reasonableness," British Medical Journal 321, no. 7272 (2000): 1300–1301; and [Free Full Text]N. Daniels and J.E. Sabin, Setting Limits Fairly: Can We Learn to Share Medical Resources? (New York: Oxford University Press, 2002).
  3. S.L. Burton et al., "The Ethics of Pharmaceutical Benefit Management," Health Affairs (Sep/Oct 2001): 150–163. For an empirical study of drug rationales in a Canadian setting, see D.K. Martin, J.L. Pater, and P.A. Singer, "Priority-Setting Decisions for New Cancer Drugs: A Qualitative Study," Lancet 358, no. 9294 (2001): 1676–1681.[Medline]
  4. J.D. Kleinke, "Just What the HMO Ordered: The Paradox of Increasing Drug Costs," Health Affairs (Mar/Apr 2000): 78–91.
  5. U.E. Reinhardt, "Perspectives on the Pharmaceutical Industry," Health Affairs (Sep/Oct 2001): 146. Even scientifically sound cost-effectiveness or cost-benefit analyses have distributive implications that require further ethical deliberation. See L.B. Russell et al., "The Role of Cost-Effectiveness Analysis in Health and Medicine," Journal of the American Medical Association 276, no. 14 (1996): 1172–1177; and [Abstract/Free Full Text]N. Daniels, "Distributive Justice and the Use of Summary Measures of Population Health Status," in Summarizing Population Health: Directions for the Development and Application of Population Metrics, ed. M.J. Field and M.R. Gold (Washington: National Academy Press, 1998), 58–71.
  6. J.R. Teagarden, N. Daniels, and J.E. Sabin, "The Need for Ethical Frameworks in Prescription Drug Benefit Allocation Policy," Journal of the American Pharmaceutical Association (Jan/Feb 2003).
  7. W.F. Steward, A. Shecter, and R.B. Lipton, "Migraine Heterogeneity: Disability, Pain, Intensity, and Attack Frequency and Duration," Neurology 44 (1994, Suppl. 4): S24–S39.[Free Full Text]
  8. Where rationing is carried out under budget limits that require comparisons of drugs across different groups of patients, rather than for different drugs for the same diagnoses, other principles will be invoked, such as "give priority to those who are sickest." See Daniels and Sabin, Setting Limits Fairly, chap. 3.
  9. In this case, the priority principle converges with concerns to avoid moral hazard. Pathological conditions are actuarial; we can assign probabilities to them based on experience. Conditions we simply would prefer to change are not actuarial.
  10. In the United Kingdom, by contrast, there is currently controversy about whether to cover (at Level 1) several drugs on grounds of cost-worthiness. These are all new and expensive drug treatments of limited effectiveness for neurological conditions. Tony Hope, professor of medical ethics at the University of Oxford and director of Ethox, personal communication, 9 April 2002.
  11. A. Pham, "Harvard Pilgrim Puts Limit on Viagra," Boston Globe, 12 June 1998.
  12. S. Levine et al., "Kaiser Permanente’s Prescription Drug Benefit," Health Affairs (Mar/Apr 2000): 185–190.
  13. G.E. Simon et al., "Initial Antidepressant Choice in Primary Care: Effectiveness and Cost of Fluoxetene versus Tricyclic Antidepressants," Journal of the American Medical Association 275, no. 24 (1996): 1897–1902.[Abstract/Free Full Text]
  14. Levine et al., "Kaiser Permanente’s Prescription Drug Benefit."
  15. J.S. Cohen, "Tablet Splitting: Imperfect Perhaps, but Better than Excessive Dosing," Journal of the American Pharmaceutical Association 42, no. 2 (2002): 160–162.
  16. Even in such settings, however, decisions to save money made within a pharmaceutical budgetary "silo" may increase costs in other "silos" that are budgeted separately.
  17. Kleinke, "Just What the HMO Ordered."
  18. Ibid., 84, citing B. Luce, "Pharmacoeconomics and Managed Care: Methodologic and Policy Issues," Medical Decision Making 18, no. 2 (1998): S6.
  19. E. Nord, Cost-Value Analysis in Health Care: Making Sense out of QALYs (Cambridge: Cambridge University Press, 1999).
  20. S. Holm, "Developments in the Nordic Countries—Goodbye to Simple Solutions," in The Global Challenge of Health Care Rationing, ed. A. Coulter and C. Ham (Philadelphia: Open University Press, 2000), 29–37.
  21. C. Ham and G. Robert, eds., Reasonable Rationing (Buckingham, England: University Press, forthcoming).


Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati    What's this?


This article has been cited by other articles:


Home page
J Oncol PractHome page
G. P. Browman, B. Manns, N. Hagen, C. R. Chambers, A. Simon, and S. Sinclair
6-STEPPPs: A Modular Tool to Facilitate Clinician Participation in Fair Decisions for Funding New Cancer Drugs
J. Oncol. Pract, January 1, 2008; 4(1): 2 - 7.
[Abstract] [Full Text] [PDF]


Home page
J Oncol PractHome page
J. E. Calfee
Can Quality-Adjusted Life Year Avoidance Help in Oncology Drug Reimbursement Decisions?
J. Oncol. Pract, January 1, 2008; 4(1): 8 - 8.
[Full Text] [PDF]


Home page
Am J Health Syst PharmHome page
J. R. Teagarden and M. K. Wynia
Ensuring fairness in coverage decisions: Applying the American Medical Association Ethical Force Program's consensus report to managed care pharmacy.
Am. J. Health Syst. Pharm., September 15, 2006; 63(18): 1749 - 1754.
[Full Text] [PDF]


Home page
Health Aff (Millwood)Home page
S. G. Morgan, M. McMahon, C. Mitton, E. Roughead, R. Kirk, P. Kanavos, and D. Menon
Centralized drug review processes in australia, Canada, new zealand, and the United kingdom.
Health Aff., March 1, 2006; 25(2): 337 - 347.
[Abstract] [Full Text] [PDF]


Home page
Clin TrialsHome page
N. Daniels and J. Sabin
Comment
Clinical Trials, February 1, 2004; 1(1): 116 - 117.
[PDF]