This Article Abstract Reprint (PDF) Submit a response to this article Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services E-mail this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to Citation Manager Reprints & Permissions Citing Articles Citing Articles via ISI Web of Science (7) Citing Articles via Google Scholar Google Scholar Articles by Klein, R. Articles by Sturm, H. Search for Related Content PubMed PubMed Citation Articles by Klein, R. Articles by Sturm, H.

Health Tracking

MARKETWATCH

Viagra: A Success Story For Rationing?

Abstract

 
The 1998 launch of Viagra prompted widespread fears about the budgetary consequences for insurers and governments, all the more so since Viagra was only the first of a new wave of so-called lifestyle drugs. The fears have turned out to be greatly exaggerated. This paper analyzes the rationing strategies adopted in four countries (United States, Britain, Germany, and Sweden), relates them to the characteristics of different types of health care systems, and identifies the conditions necessary for successful cost containment. The case of Viagra, it concludes, holds out two general lessons: first, allow exceptions to total bans on reimbursement; second, involve the medical profession in the decision-making process.

These worries were all the more acute because Viagra was, if not the first, certainly the highest-profile example of a new generation of drugs—so-called lifestyle drugs—that raised these kinds of issues. If consumers could define their own medical necessity—for, say, drugs to reduce their weight—then, it was argued, the floodgates of drug spending would open, with dire consequences for the finances of insurers and health care systems if they chose to reimburse such prescriptions. Viagra thus posed a more general challenge and opened up a wider debate. If the distinction between drugs prescribed by doctors to deal with medical necessities and those demanded by consumers to enhance their lifestyles was often blurred—since the same drugs could serve either purpose—then how could their use be controlled or rationed?² And how, in any case, should medical necessity be defined, and by whom?

These questions cut across health care systems. The case of Viagra therefore offers an opportunity to compare how various countries reacted to the same specific, concrete challenge: whether or not to make Viagra a reimbursable drug by including it in the standard benefit package. In what follows, we analyze the policy responses—that is, rationing strategies—adopted in a variety of countries, drawing on the material publicly available either in print or on the Web, supplemented by some telephone interviews. Our aim in this is, first, to draw out a taxonomy of rationing strategies and, second, to relate those strategies to the characteristics of national health care systems. Accordingly, we have been selective rather than comprehensive in our choice of countries; we chose them to provide a sufficiently wide range of policy responses and types of health care systems. In all cases, we report on the immediate reaction to the introduction of Viagra and subsequent adaptations. This field is still evolving, however, so some of our information may have been overtaken by events since the completion of this study at the end of 2001.

Background. Before turning to the policy responses, however, it is worth noting some of the relevant background information about Viagra available to decisionmakers. First, it is an effective form of treatment for ED. Soon after the launch of the drug, twenty-one randomized controlled trials concluded that about 75–80 percent of men show a statistically significant improvement after taking Viagra.³ This eliminated the option of arguing that Viagra is an ineffective drug. Second, although ED is associated with a variety of diseases (and consequential surgical or pharmaceutical interventions), the most important correlation is with age. So the condition is not one that is self-inflicted—that is, the result of personal behavior. It cannot therefore be blamed on the patient. Third, the evidence suggests that Viagra is cost-effective when compared with other forms of treatment for ED.⁴ Attempts to push the analysis further and calculate costs per quality-adjusted life year (QALY) gained run into methodological problems, and any results must be treated with caution.⁵ In any case, the relative cost-effectiveness or cost-utility of using Viagra for the treatment of ED is only one consideration—and not the most important one—when considering the economics of making it a reimbursable drug. Much more important are the assumptions made about the likely increase in the demand for treatment of ED that is likely to follow such a decision. For policymakers everywhere the crucial consideration was how best to avoid an upsurge in the total volume of demand. Fourth, although in theory Viagra is a prescription-only drug, in practice it can be obtained quite easily over the Web.⁶ Whatever reimbursement policies are adopted, it is therefore in effect an over-the-counter (or, strictly speaking, over-the-Web) drug, largely outside the control of the medical profession.

Typology Of National Rationing Strategies

Top Typology Of National Rationing... Do The Characteristics Of... Implications For The Future NOTES

 
In this section we set out the various strategies for rationing Viagra adopted in the countries we studied. However, before doing so, we need to put the specific case of Viagra into the wider context of health care rationing more generally, to see whether it conforms to a standard pattern or has any special features.⁷

Forms of rationing. Rationing—decisions to deliver less than the optimum amount of effective health care as a result of setting priorities among competing demands on the system—pervades across all health care systems, regardless of spending levels. It takes many forms, of which the explicit denial of a service is the most dramatic but not necessarily the most important. Other forms of rationing are exclusion (sections of the population not covered), dilution (fewer tests ordered, fewer nurses on the ward), deterrence (making access to care difficult), and delay (waiting lists). But not only do the forms of rationing differ. So, too, does the decision-making mode involved.

Thus, decisions can be either made centrally or diffused among the professional service deliverers. Similarly, they can be made either explicitly (setting out the criteria for allocating resources to individual patients) or implicitly (fixing global budgets that force choice between competing demands on resources at the point of delivery). Generally speaking, diffused and implicit rationing by professionals has been the dominant mode cross-nationally, a strategy that diffuses not only responsibility but also blame. Presenting decisions about whom to treat and in what way as reflecting professional judgments and scientific evidence, rather than budgetary limitations, is clearly in the interests of politicians and insurance managers. It also may be a rational approach, given uncertainty about which medical intervention works for whom.⁸

Various attempts have been made to devise limited menus of entitlements with explicit exclusions; Oregon’s Medicaid waiver is the best-known example. But these have invariably run into trouble.⁹ Not only has there been menu creep (a combination of consumer pressure and professional ingenuity in reclassifying conditions has meant that the menu of services tends to be elastic), but also attempts to exclude specific interventions immediately raise the objection that almost every procedure or drug can be medically necessary for someone. Even cosmetic surgery, a standard item in most exclusion lists, may be crucial for someone contemplating a future career as a ballet dancer, for example. So explicit exclusion policies quickly develop holes as exceptions are allowed, as the case of Viagra illustrates.

In many respects, the case of Viagra follows the standard rationing pattern. When the drug was first launched worldwide, the overwhelming, although not entirely unanimous, response of decisionmakers was to exclude it from the reimbursable health care menu. Subsequently, however, policies have been modified to accommodate arguments of medical necessity. Total bans in practice turned out to be leaking colanders. However, it was mainly at this stage that differences in rationing modes emerged between countries. For the sake of simplicity, we present these differences as four models derived from the experience of specific countries. These, we must stress, are very much "ideal-type models"; that is, in practice there are overlaps between countries and modes, if only at the edges. However, they provide a useful analytic framework for analyzing policy responses across nations.

Diffusion by inaction. As so often in comparative health policy studies, the United States emerges as an outlier, unique unto itself. A nonsystem made a nondecision about Viagra. Absent a national decision, even U.S. federal programs adopted divergent positions. The Department of Veterans Affairs (VA) refused to add Viagra to its formulary on the grounds that the costs of providing the drug would add 20 percent to its pharmaceutical budget (although the ban was not complete; an escape clause allowed doctors to argue for its prescription as exceptions).¹⁰ In contrast, the Medicaid program automatically included Viagra for the treatment of ED following its approval by the Food and Drug Administration (FDA), as required by legislation, although the agency feared clinical and financial abuse.¹¹ Of course, the financial implications of this were relatively modest compared with those faced by the VA health system, given that only about 10 percent of Medicaid beneficiaries are adult males. In any case, the decision was variously implemented by the states. Some resisted outright (among them, New York, Wisconsin, and Nevada).¹² Others followed the recommendations of the Centers for Medicare and Medicaid Services (CMS) designed to minimize misuse and rationed the amount prescribed: from four pills per month (for example, in Alabama and Florida) to ten (in Utah).¹³

Health insurers and plans showed a similarly mixed picture. A very few plans included Viagra in their formulary from the start; one such was Tufts, which put it in its highest copayment category.¹⁴ The great majority resisted. "Simply put, having sexual relations is not a medical necessity," one Aetna official argued to the New York Department of Insurance. However, under the challenge of both court rulings and state regulators, many of the insurers were forced to abandon or modify the blanket exclusion of Viagra.¹⁵ Overall, then, the consequence is that access to reimbursable Viagra prescriptions for American men—the conditions under which it is prescribed, the number of pills deemed appropriate, and the level of copayments—depends on where they live and with whom they are insured. In this respect, of course, Viagra does not represent so much a deviant case as an illustration of the U.S. health care condition.

Juridification. Although Germany’s health care system could not be more different from that of the United States, there is one shared characteristic: The courts have played a major role in shaping decisions. Germany’s system is based on social insurance—that is, a network of sickness funds—and it has a corporatist style of governance. Within the broad framework set by the federal government, policy decisions are negotiated by the representatives of the medical profession and the sickness funds—the Bundesausschuss der Ärzte und Krankenkassen. It was this body that decided that Viagra should not be included in the standard package of reimbursable drugs. However, the decision was appealed. The Federal Social Court decided that the Bundesausschuss did not have the constitutional right to issue an unconditional ban on any drug.¹⁶ This left matters in limbo, and the court has yet to give a more detailed ruling about the specific issues raised by the case of Viagra and other "lifestyle" drugs. At first eager to secure such a ruling, the insurers have stopped pressing for a decision, fearing that the Federal Social Court would take its cue from the lower courts, which have consistently ruled in favor of patients appealing against refusals to reimburse Viagra.¹⁷

In a series of cases, the lower courts have decided in favor of reimbursing the cost of Viagra prescriptions wholly or partially. Among successful arguments have been that patients should be reimbursed when ED is the consequence of medical intervention or condition (for example, a bladder cancer operation, dialysis and kidney transplantation, diabetes, or multiple sclerosis) and when ED causes depression and psychosocial problems. In one case, the court sought to draw a distinction—central to the debate about lifestyle drugs—between using Viagra to enhance potency and prescribing it for the restitution of normal bodily function. Only in the latter case, the court determined, should Viagra be reimbursable (although normal may not be simple to define). "Intact erectile function is part of the image of a healthy man, including the elderly," the Hanover Social Court ruled.¹⁸

These individual, case-by-case decisions have not been generalized into any kind of applicable guidelines. Rationing in Germany continues to take the form of scattergun juridical decisions. Indeed, muddling through is in the interests of the insurers; if the Federal Social Court were to generalize the generosity of the lower courts, the result would be much more expenditure. For the time being, the original ruling of the Bundesausschuss therefore determines the policy of insurers—that is, no reimbursement, absent a specific court decision. For the longer term, it is worth noting that sickness funds and physicians share a common interest in limiting demands on their collective drug budgets: If individual physicians are overly generous in prescribing Viagra or any other lifestyle drugs, they not only limit the resources available to their colleagues but can be held personally responsible for the cost. Whether this shared interest in self-restraint will survive if the government implements its decision to remove the cap on the drug budget is another matter.

Centralization–politicization. In contrast to both the United States and Germany, policy in Britain for rationing Viagra in the National Health Service (NHS) was centrally determined by government ministers. Given the highly centralized nature of the NHS, this might at first appear to be a highly predictable outcome—an illustration of path dependency. In fact, this would be a misleading conclusion. The paradox of the NHS is that rationing has always been implicit. Traditionally, ministers have set budgets but have allowed the medical profession to translate financial constraints into clinical decisions—a highly effective blame-diffusion strategy.¹⁹ The oddity of the decision about Viagra was thus that it represented not so much the logic of the NHS as a new departure.

It was a reluctant departure. The first instinct of ministers was to depoliticize the issue by asking for expert advice.²⁰ But the Government’s Standing Medical Advisory Committee refused to oblige. It concluded that there was no medical reason for refusing to make Viagra available by prescription in the NHS—"in common with many treatments available under the NHS this improves quality of life, but does not save or prolong it"—but that it was for ministers to make the final decision in light of the "availability of resources." The decision of the secretary of state for health was that since "impotence is in itself neither life threatening, nor does it cause physical pain," and since Viagra threatened to increase the cost of treating impotence tenfold, general practitioners (GPs) would be restricted in their ability to issue NHS prescriptions for Viagra. Availability would be limited to groups of men whose disabilities were linked to specific medical conditions: for example, those treated for prostate cancer or kidney failure and those suffering from Parkinson’s disease and multiple sclerosis (MS). The official ration, furthermore, was to be one tablet a week. Exceptional cases not falling into the official categories would be referred to hospital specialists.

The logic of this decision was far from self-evident, as the leader of Britain’s GPs was quick to point out: Its only justification appeared to be that it promised to constrain demand and spending.²¹ Also, in apparently limiting the NHS’s treatment responsibilities to dealing with conditions that either threatened life or caused physical pain, the secretary of state appeared to be expounding a new restrictive, unsustainable doctrine. However, subsequent correspondence in the British Medical Journal suggested general support among doctors for rationing Viagra: "Nobody needs an erection at public expense" was the heading of one letter.²² Furthermore, British GPs have a shared interest with government in controlling demands. The creation of Primary Care Trusts, with responsibility for purchasing health care for given populations, has given them responsibility for controlling their own (capped) drug budgets.

Bureaucratization. Sweden is an interesting, because exceptional, case of a policy reversal. Although in many respects a first cousin to Britain’s NHS—inasmuch as it is funded through taxes—Sweden’s health care system is a far more decentralized one. County councils are responsible for running health care services and, since January 1998, for pharmaceutical budgets. However, decisions about drugs remain firmly national. As in Britain, policy is driven by the assumption that the same package of health care services should be available regardless of where people live. The result has been tension between the budget holders (the county councils) and the central decisionmakers. At the time of Viagra’s launch on the market, the rule was that any pharmaceutical product accepted as a prescription drug in Sweden automatically had to be included in the drug benefit package. Accordingly, Viagra was included.

However, conscious of the financial implications of automatically endorsing all new products and under pressure from the county councils, the Swedish government subsequently appointed a commission of inquiry. Its report, published in 2000, recommended that drugs be divided into two categories.²³ The first, involving treatment for disease and injury, would continue to be part of the standard package. The second, which included not only Viagra but also drugs for the treatment of obesity, smoking cessation, and hair loss, would be available only in exceptional circumstances. Detailed criteria were to be defined by a governmental committee, whose report was overdue at the time of this writing, to replace present procedures.

At present, decisions are made case by case by the Ministry of Health, in consultation with the Medical Products Agency (MPA), the Läkemedelsverket, which is the regulatory agency for medical products. In effect, there is bureaucratic rationing. Applications have to be made by the individual patients concerned, with support from their doctors. In making the determinations, the criterion appears to be different from that used in Britain (and other countries). The emphasis is on the consequences of ED, not the cause or associated morbidities. Treatment is sanctioned in those exceptional cases where ED aggravates an existing condition. In practice, this means psychiatric conditions. The system appears to have been effective in containing demand and expenditure. By the end of 2001 there had been roughly 3,000 applications, of which fewer than 10 percent had been approved.²⁴ Given the low success rate, it is perhaps not surprising that the number of applications has been diminishing over time. A further deterrent may well be the lack of privacy: Under the Swedish system of open government, applications are in the public domain.

Rationing by expertise. There is an emergent fifth model of rationing, relevant to the introduction of lifestyle drugs more generally, that overlaps with those already discussed but is worth noting. This is rationing by expertise. Since 1999 Britain has had the National Institute for Clinical Evidence (NICE), an agency charged with reviewing the evidence about new health technologies and producing guidelines about their use in the NHS. Had NICE been in existence in 1998, ministers would no doubt have referred the case of Viagra to it with a profound sense of relief. And, as noted above in the case of Sweden, bureaucratic rationing is seen as a temporary expedient until effective guidelines can be devised. In both instances, the hope is that rationing decisions can be depoliticized by invoking the expertise of a neutral, authoritative agency or committee. The experience of NICE so far suggests that this may be an overly optimistic view.²⁵ Many of NICE’s decisions have proved controversial, and some have been modified following lobbying by the pharmaceutical industry or consumer groups representing patients with specific diseases. Although it is relatively easy to determine which interventions are effective, deciding on priorities within constrained budgets is a different matter. It is far from clear that the expertise of agencies such as NICE carries legitimacy in determining this much larger question.

Do The Characteristics Of Health Care Systems Matter?

Top Typology Of National Rationing... Do The Characteristics Of... Implications For The Future NOTES

 
Can policymakers choose à la carte from the menu of rationing strategies outlined in the previous section? Or are their options contingent on the characteristics of specific health care systems? In the case of the four countries so far considered, different systems are matched with different rationing strategies. But if we are to draw any general conclusions from this finding, we have to test it by asking whether similar systems yield similar rationing strategies.

The United States and Britain are, in their contrasting ways, unique systems. No other country is as chaotic as the former or as centralized as the latter. But Sweden and Germany exemplify larger classes of systems. Sweden is an example of the "Nordic model" of health care: universal, tax-funded, but decentralized. Germany is an example of a social insurance–based system—with a plurality of insurers and providers and with a corporatist style of governance. In both there is a group of similar countries. Accordingly, we compare the rationing strategies of other countries within each group. In this exercise we adopt a "black swan" approach.²⁶ If it turns out that each group is consistent in adopting the same strategies, then there is a strong case for assuming that system characteristics influence (and perhaps determine) rationing strategies. If there is a deviant case (or black swan) within a group, however, any relationship must be more complicated.

The Nordic model. Here Sweden started as a deviant case when it automatically included Viagra in the standard benefit package but has since moved closer to practice in other Scandinavian countries. Finland has a three-tier system of refunding drug costs, with varying criteria and copayments.²⁷ In the top category, refunds are automatic. In the bottom category, "significant and expensive" drugs are reimbursed only if there are "sufficient therapeutic indications." Decisions about the classification of new drugs are made by the Council of State, which also sets out the conditions under which prescriptions may be eligible for a refund. Viagra, like certain drugs to treat MS and obesity, falls into the bottom category. It can be reimbursed only if ED is caused by "serious disease," such as total prostatectomy or vertebral trauma. Unlike in Sweden, psychological indications are not included. Patients have to apply for reimbursement to the Social Insurance Institution, with the support of their doctor. In Denmark, similarly, Viagra is not automatically reimbursed.²⁸ Decisions are made one at a time by the Danish Medicines Agency. Once again, the criteria favor ED consequent on or associated with medical interventions. Norway, too, controls the reimbursement of Viagra strictly, a policy introduced to avoid the cost explosion that took place in Sweden before its change of policy. Patients have to apply for reimbursement to a national insurance scheme, where officials then decide on the individual cases based on agreed-upon criteria.

In the case of the Nordic countries, there is therefore no "black swan." However, some swans have gray feathers. While there may be convergence on the bureaucratic model of rationing Viagra, there are variations in both criteria and procedures. Moreover, it cannot necessarily be concluded that convergence reflects only the shared characteristics of the health care systems. Two other, more general explanations could account for this phenomenon. The first is policy learning. The Scandinavian countries may have learned from each other’s experience (a point that applies strongly to Sweden and Norway). The second is that convergence may have nothing to do with the characteristics of the health care systems but may reflect a shared Nordic political and institutional culture.

Corporatist social insurance. Here we have only two cases to compare with Germany, those of Austria and the Netherlands. Many other countries have health care systems based on the social insurance principle (France, for example), but only Austria and the Netherlands share Germany’s corporatist model of governance. The similarities in the style of health care governance between Germany and Austria are particularly striking.²⁹ It is the insurers (Versicherungsträger), not the government, in both countries that determine the basket of reimbursable drugs. And in the case of Viagra, the medical superintendents of the Austrian insurers decided that treatment for ED would be limited to defined conditions—again, a familiar list, including spinal cord lesions, pelvic surgery, and so forth. However, in contrast to Germany, the courts have not intervened. This may be because of a difference in political culture, or, more plausibly, because the Austrian insurers were more flexible than their German counterparts were. Instead of imposing a total ban on Viagra reimbursement, they allowed some exceptions from the start, thus making their policies more acceptable and a legal challenge less likely. So Austria is the most "pure" example of corporatist rationing—that is, government delegating the task to insurers and providers.

The Netherlands, however, provides a black swan. Here the minister of health decided to exclude Viagra from the standard package.³⁰ Following the standard Dutch practice of carrying out medical and economic evaluations, the insurers’ College voor Zorgverzekeringen had recommended that Viagra should be reimbursed for the usual medical conditions and in strictly limited doses.³¹ However, the minister of health, Else Borst, overruled the recommendation. As in Britain, this was a political decision—not, as in Germany and Austria, the product of a corporatist-style consensus-engineering exercise involving insurers and the medical profession. So, in this group, there appears to be a deviant case. However, it may be a deviant case not because it is a black swan but because it should never have been put into this group in the first place. The Netherlands has always presented difficulties to political taxonomists, and its labeling as a corporatist state is not universally accepted.³²

Overall, then, the relationship between systems’ characteristics and modes of rationing remains an open question. Some policy decisions are indeed preempted by systems characteristics: A central government decision of the kind found in Britain and the Netherlands is unimaginable in the United States. But beyond that, our evidence shows that the relationship between system characteristics and rationing strategies is not direct—and that if there is a relationship at all, it is a complex one, mediated by other factors.

Implications For The Future

Top Typology Of National Rationing... Do The Characteristics Of... Implications For The Future NOTES

 
International norm. So far our analysis has concentrated on analyzing differences in both the rationing strategies adopted and the characteristics of health care systems. But this is to risk overlooking something far more important: that all of the health care systems analyzed have succeeded, in their various ways, in rigorously rationing the availability of Viagra as part of the standard package of reimbursable or free health care. Contrary to what might have been expected from the general experience of rationing reviewed above, governments or insurers have decided explicitly either to exclude Viagra from the basic benefit package or to make its availability contingent on specific medical conditions. This conclusion would hold if our analysis were extended to cover other advanced, postindustrial countries, such as Italy and Switzerland. Successful rationing is the international norm, thus making nonsense of apocalyptic speculations that Viagra would cause financial havoc.

Arbitrary distinctions? Is Viagra a one-off case of successful rationing, or does it point to more general conclusions? How far is Viagra representative of the wider class of lifestyle drugs and interventions? In answering these questions, the difficulty is that the whole concept of lifestyle drugs is problematic. The distinction between medically necessary and lifestyle interventions is, as has been forcefully argued, largely arbitrary.³³ If the aim of medicine is to improve the quality of life—to allow men and women to function to their maximum potential—then it is not self-evident that improving sexual performance is any different from improving the ability to carry out the activities of daily living. And in the latter case, it is accepted that medicine will intervene, often expensively, as in the repair or replacement of joints. If, further, psychological distress is put on a par with physical pain—as in practice it is—then the dividing line between medically necessary and lifestyle interventions becomes further blurred. For example, should psychotherapy be put into the lifestyle category? The problem is compounded when we consider drugs or procedures that enhance people’s ability to conform to the social norms of their society, ranging from having children (in vitro fertilization) or not having them (contraception) to having bodies of an acceptable shape and appearance (cosmetic surgery, treatment for obesity). In short, the lifestyle category turns out to be an overelastic hold-all. It covers a heterogeneous lot of drugs and interventions whose inclusion in the standard benefit package can be argued on grounds of promoting normal physical, psychological, or social functioning and for which notions of what is normal may well be contestable, vary over time, and differ between countries.

There is, however, one common element amid all this heterogeneity: that necessity is defined not by the doctor but by the consumer, not according to technical medical criteria but in light of social and cultural norms. Needs are equated with demands. A working, non-pejorative definition of lifestyle drugs or interventions might therefore be those for which the patient rather than the doctor not only diagnoses the condition but can also demand a specific remedy. It is in this respect that Viagra can be seen as representative of a wider class. To return to the starting point of this paper, the reason why the launch of Viagra prompted so much alarm among policymakers was precisely that need appeared to be determined subjectively, bypassing the filter of medical necessity. The spectre of moral hazard haunted policymakers everywhere: How could abuse—and the consequent cost explosion—be prevented if a drug for a self-reported condition were made reimbursable? To the degree that other drugs or interventions raise the same question, and however different they may be in other respects, the story of Viagra has general relevance.

Little public sympathy. We can concede straight away that in some respects the case of Viagra is indeed special, if only because there remain considerable inhibitions and prejudices about treatments involving sexual performance and potency. Sufferers from ED are unlikely to take to the streets carrying protest banners. Impotence is more likely to be suffered in private than paraded in public. Further, there is no concentrated constituency to campaign for a more generous policy. In contrast to homogeneous, organized pressure groups acting for patients with conditions such as MS, those suffering from ED are a scattered, heterogeneous lot without any organizational base. This limits the scope for a campaign designed to apply political pressure on governments and insurers. Moreover, any such campaign would be unlikely to enlist much public sympathy. Arguments about rationing Viagra prompt more jokes than indignation. Insofar as ED is correlated with age, it is often seen as somehow "natural" and inevitable. Private grief in such cases is not seen as calling for collective action—an argument that, however, is not applied to other degenerative conditions of old age for which treatment is automatically included in the basic package of health care benefits everywhere. Overall, there is a widespread view that treatment of ED should rank low in any system of priorities. As a leading British political commentator put it: "A nation which spends taxpayers’ money on better erections, while leaving old ladies to soil themselves and starve in under-staffed wards, is sick indeed."³⁴

Ease of purchase. The case of Viagra has another feature that, while not unique to it, serves to distinguish it. As already noted, it can be bought relatively easily and cheaply on the open market despite being classified as a prescription drug. If exit into the market is relatively cheap, if over-the-Web drugs are available, then it is unlikely that much voice will be raised in protest against rationing by price or that there will be serious worries about equity. Perhaps the public perception is that willingness to pay is a good measure of subjectively defined need. And, as far as equity is concerned, in the case of Viagra it can be argued that money has always bought ways of boosting sexual performance, from call girls to rhinoceros horns. No new inequity is therefore involved.

A policy blueprint. To the extent that other new drugs or interventions (whether or not labeled "lifestyle") share some or all of these characteristics, so policy outcomes are likely to mirror the story of Viagra. If the patient group involved is heterogeneous and unorganized, if there is little public sympathy for the specific condition involved, if demands can be met in the market place, then policymakers should be able to adopt rigorously restrictive policies without much difficulty. The converse, of course, also follows: If there is an organized constituency, if public sympathy can be evoked, and if heavy expense is involved, then policymakers are likely to encounter strong resistance when trying to restrict reimbursement for new drugs or interventions (whether or not labeled "lifestyle"). However, our analysis also suggests two more general conclusions, less contingent on the specific character of the innovation in question.

First, a common thread runs through the rationing strategies of different countries: All of the systems in our sample have allowed exceptions from a general ban on refunding, although some have done so only after regulatory or judicial rulings (as in Germany). Furthermore, the exceptions tend to follow a common pattern: Except in Sweden, reimbursement of Viagra is contingent on previous medical conditions or interventions. If there is any ethical logic in this, it appears to be a compensatory one: Somehow the men in this group are perceived to deserve special treatment as victims of unmerited, disproportionate misfortune. However, the real logic is surely economic and political. On the one hand, the criteria represent a sorting mechanism that is both reasonably objective and financially restrictive, distinguishing between need that can be defined by the medical profession and by patients’ demands. The formula provides a tool for the exclusion of pure lifestyle drugs—that is, those where the patient both diagnoses the condition and can demand a specific remedy. On the other hand, the strategy leaves scope for medical discretion by leaving some judgments to doctors. It is therefore more respectful of medical autonomy than an outright ban would be. While an outright ban challenges the medical profession to devise ways of gaming the system, allowing exceptions invites the cooperation of the profession, particularly if doctors have been involved in devising the criteria.

Second, the rationing strategies adopted have, by and large, obtained at least the passive support of the medical profession. There have been criticisms of the criteria adopted but no sustained campaign of opposition. Further, doctors working in health care systems with capped budgets, as in Britain and Germany as well as in some U.S. managed care plans, have an interest in restraining demand. To the extent that such capped budgets become the norm, so governments may find the medical profession a powerful ally in resisting any kind of open-ended commitment to lifestyle drugs as they come onto the market. Indeed, such drugs can be seen as representing as much of a threat to the medical profession as to budgets, to the extent that they undermine physicians’ monopoly of judgment about what is medically necessary—and, more generally, raise doubts as to what that hallowed phrase actually means.

While Viagra does have specific features that have made its rationing socially acceptable and politically feasible, the case history suggests that the same set of rationing strategies can be used successfully as other new, much-promoted drugs come onto the market. Two conditions seem necessary. First, rationing is an instance where the leaky bucket may be preferable to a water-tight one: Factoring in exceptions, based on some reasonably objective criteria, helps to make rationing strategy acceptable. Second, the acquiescence of the medical profession is essential, and including the profession in the design of rationing strategies is one way of achieving this. If these conditions are met, the new generation of drugs are unlikely to break the bank.

Editor's Notes

 
Rudolf Klein is visiting professor at the London School of Hygiene and the London School of Economics. Heidrun Sturm is a health care researcher at the Department of Clinical Pharmacology of the University of Groningen, the Netherlands.

This study was funded by the Milbank Memorial Fund. The dividends of the support given were long in coming, and the authors’ thanks go to Dan Fox for his patience.

NOTES

Top Typology Of National Rationing... Do The Characteristics Of... Implications For The Future NOTES

 

1. A.E. Benet and A. Melman, "The Epidemiology of Erectile Dysfunction," Urology Clinics of North America (November 1995): 699–709.
2. A. Keith, "The Economics of Viagra," Health Affairs (Mar/Apr 2000): 147–157.
3. A. Burls et al., "Sildenafil," Report no. 12 (Department of Public Health and Epidemiology, University of Birmingham, September 1998).
4. E.A. Stolk et al., "Cost Utility Analysis of Sildenafil Compared with Papaverine-Phentolamine Injections," British Medical Journal (29 April 2000): 1168–1173.
5. N. Freemantle, "Valuing the Effects of Sildenafil in Erectile Dysfunction" (Editorial), British Medical Journal (29 April 2000): 1156–1157.
6. A Lycos search for "Penispill" on 12 June 1998 produced seven Web sites on how to get prescriptions or how to order Viagra via phone or on the Internet.
7. R. Klein, S. Redmayne, and P. Day, Managing Scarcity (Buckingham: Open University Press, 1996).
8. D. Mechanic, "Muddling through Elegantly: Finding the Proper Balance in Rationing," Health Affairs (Sep/Oct 1997): 83–92.
9. On Oregon, see J. Oberlander, T. Marmor, and L. Jacobs, "Rationing Medical Care: Rhetoric and Reality in the Oregon Health Plan," Canadian Medical Association Journal (29 May 2000): 1583–1587. For another example, see D. Chinitz et al., "Israel’s Basic Basket of Health Services: The Importance of Being Explicitly Implicit," in The Global Challenge of Health Care Rationing, ed. A. Coulter and C. Ham (Buckingham: Open University Press, 2000), 44–52.
10. Department of Veterans Affairs, "VA Reaches Decision on Viagra," Press Release (in Mealey’s Impotency Drug Watch, 23 July 1998).
11. Centers for Medicare and Medicaid Services, Drug Policy: Medicaid Coverage of Viagra, www.hcfa.gov/medicaid/drpolicy.htm (5 January 2000).
12. "Managed Care Monitor—Viagra: Two HMOs, Two States Say ‘No’ to Coverage," American Health Line (6 July 1998).
13. "USA Today: States Draw Line for Viagra" Mealey’s Impotency Drug Watch (20 August 1998).
14. Tufts Health Plan, "Pharmacy Information," www.tufts-healthplan.com/members/pharmacy-3tier.html (19 October 1999).
15. "Viagra Coverage," Mealey’s Insurance Law Weekly (1 March 1999).
16. "BSG-Urteil zur erektilen Dysfunktion," Deutsches Ärzteblatt, 15 October 1999, C-1895.
17. "Erneut Novellierung der Arzneimittelrichtlinien?" (Revised supplemental medical directive), Ärztezeitung, 9 November 2001).
18. Court decision Az: S2 KR 485/99.
19. H.J. Aaron and W.B. Schwartz, The Painful Prescription (Washington: Brookings Institution, 1984).
20. S. Dewar, "Viagra," in Health Care UK, 1999/2000, ed. J. Appleby and A. Harrison (London: King’s Fund, 1999), 139–151.
21. J. Chisholm, "Viagra: A Botched Test Case for Rationing," British Medical Journal (30 January 1999): 273–274.
22. "Rationing of Sildenafil" (Letters), British Medical Journal (12 June 1999): 1620–1621.
23. Staten offentlich utredening (SOU) 2000:86, del 3 (2000) (Report of Swedish Investigation Committee); and Vanja Gavellin, Socialdepartementet, personal communication, 3 December 2001.
24. Lotta Eriksson, Socialdepartementet, and Jane Ahlquist-Rastat, Läkemedelsverket/MPA, personal communication, 3 December 2001; and Läkemedelsverket, "Aktuellt/Observanda," 6 December 2001, www.mpa.se/observanda/obs01/dispens_hokostnadsskydd.shtml (6 August 2002).
25. For a searing attack on NICE, see R. Smith, "The Failings of NICE," British Medical Journal (2 December 2000): 1363–1364. For the reply by the chairman of NICE and comments from a variety of sources, see the letters column of the British Medical Journal (24 February 2001): 489–491. For a neutral survey of NICE’s work, see J. Raftery, "NICE: Faster Access to Modern Treatments? Analysis of Guidance on Health Technologies," British Medical Journal (1 December 2001): 1300–1303.
26. K. Popper, The Logic of Scientific Discovery (London: Hutchinson, 1959). However many white swans we count, Popper argues, we cannot with certainty say that all swans are white. But if we see one black swan, we can confidently say that "not all swans are white."
27. Rajaniemi Rajaniemi, Institute for Social Insurance, SII, Helsinki, personal communication, 31 March 2002.
28. For Denmark, Karen Kolenda, Department of Drug Economics, Danish Medicines Agency, personal communication, 1 February 2002. For Norway, John Anderson, Health Ministry, Oslo, personal communication, 4 February 2002.
29. Anna Buscics, Hauptverband der Sozial-versicherungsträger, personal communication, 29 November 2001.
30. E.A. Stolk, W.B.F. Brouwer, and J.J.V. Busschbach, "Vergoeding van Viagra stuit op waarden en normen" [Reimbursement of Viagra is based on values and norms], Medisch Contact (28 April 2000): 626–629.
31. College voor Zorgverzekeringen (CvZ) Doc. no. CUH00378 (23 March 2000), Beordeling wachtkamermiddel Sildenafil (Original letter from the CvZ to the minister of health).
32. G.H. Okma, Studies on Dutch Health Politics, Policies, and Law (Utrecht: Medical Faculty of the University of Utrecht, 1997).
33. See Keith, "The Economics of Viagra." Keith is former director of economic policy analysis at Pfizer.
34. A. Marr, "Viagra: A Hard Choice," Observer, 24 January 1999.

                      What's this?