Health Affairs, 23, no. 1 (2004): 107-112 doi: 10.1377/hlthaff.23.1.107 © 2004 by Project HOPE	New Online Disparities in Health:   Obama aide on legislation   Smedley call to action   Shrink regulations first   A Real Policy Debate   Top 10 HA Articles Of '07

This Article Abstract Figures Only 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 ISI Web of Science 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 HighWire Citing Articles via ISI Web of Science (15) Citing Articles via Google Scholar Google Scholar Articles by Reinhardt, U. E. Search for Related Content PubMed PubMed Citation Articles by Reinhardt, U. E. Related Collections Pharmaceuticals Research And Technology

Medicare Challenge

PERSPECTIVE

An Information Infrastructure For The Pharmaceutical Market

Abstract

 
The primary contribution of Joseph Newhouse’s paper is to make the point that it is impossible to devise a drug pricing policy that satisfies either the conditions for economic efficiency or prevailing standards of "fairness." Not specifically mentioned by Newhouse is a rule for the socially efficient outlays on selling and marketing by the pharmaceutical industry. In fact, it is not clear that the current allocation of the industry’s revenue dollars to marketing and R&D is efficient from society’s point of view. This paper explores this question and proposes an independent, publicly funded information infrastructure to study and disseminate results on pharmaceutical cost-effectiveness.

The sobering insight gained from Newhouse’s paper is that it is impossible to devise a drug pricing policy that satisfies the conditions for economic efficiency at any moment in time and over time, let alone prevailing standards of fairness. In that important insight lies the paper’s major contribution. Nations have coped with this dilemma by muddling through more or less elegantly with their policies on drug pricing, leaving policy analysts to explore whether and how existing policies might be improved and whether, in general, governments are more or less elegant than private "markets" in muddling through this economic swamp.¹ In the end Newhouse opts for private markets, albeit without enthusiasm, presumably because he knows that even the relatively unregulated U.S. market for drug products is a far cry from the textbook model of efficient markets taught in Econ 101.

The Stringent Requirements Of Economic Efficiency

Top Editor's Notes The Stringent Requirements Of... The Information Infrastructure... The Case For Federal... Concluding Observations Notes

 
To see why this is so, it is worth summarizing the stringent conditions for efficiency mentioned throughout Newhouse’s paper. Briefly, these conditions are as follows: (1) The level of output of each drug in any given year should be such that the last unit produced should yield society as much benefit as those resources would have yielded society, if they had been used to produce some other goods or services. (2) The production of drugs should be organized and managed to minimize the value of the real resources used to produce a given volume of a given drug, where this "value," once again, is measured by the benefits from other goods and services that are sacrificed by using the real resources in drug production rather than elsewhere.

(3) The pricing and distribution of drugs among members of society should be such that no unserved person would have been willing to pay as much or more for an additional unit of a drug than the incremental cost of producing that unit. Here, willingness to pay would signal the monetary value placed on that unit. The incremental costs are only those that would be avoided if that incremental unit were not produced. By definition, they exclude whatever fixed costs the accountant allocated to the drug to calculate its "fully allocated unit cost" and are therefore quite low in the pharmaceutical industry.

(4) Year after year, the last dollar spent on drug research and development (R&D) should yield society as much benefit as it would have yielded if it had been spent to produce other goods or services.

Newhouse does not explicitly mention a rule for the socially efficient outlays on selling and marketing by the drug industry. As is suggested by Exhibit 1, these outlays may well exceed spending on R&D, which in 2002 absorbed a weighted average of only 14 percent of total revenue in thirteen of the largest research-based pharmaceutical companies (weighted by the firms’ total revenue), while selling and general administration absorbed close to 33 percent. Although these allocations vary somewhat across firms and for each firm from year to year, the ratios are consistent with other sources on these allocations.²

View larger version (15K): [in this window] [in a new window]   EXHIBIT 1 Allocation Of Revenue Dollars In Thirteen Large Research-Based Pharmaceutical Companies, 2002

Newhouse rightly argues that each firm’s marketing outlays are likely to be optimal from that firm’s perspective, in the sense that its profits are maximized, other things being equal. But private optimality at the level of the individual firm does not necessarily meet the condition of social optimality set forth above, because part of the marketing by drug makers is likely to be a self-canceling, zero-sum game over given market turf. It is thus pertinent to ask whether the current allocation of the industry’s revenue dollars to marketing and R&D is efficient from society’s point of view.

The Information Infrastructure For Pharmaceutical Therapy

Top Editor's Notes The Stringent Requirements Of... The Information Infrastructure... The Case For Federal... Concluding Observations Notes

 
The words "benefit" or "value" run through every one of the above conditions for efficiency. The implication is that reliable information on the value of rival drugs is the sine qua non of an efficiently functioning market. Can it be argued that today’s U.S. market for pharmaceuticals meets this stringent condition? Two facts stand out in this regard.

First, the products distributed through this market can bestow enormous benefits on patients. On the other hand, as the Institute of Medicine recently pointed out, if improperly used these products can also cause serious harm to patients. Overuse, underuse, and misuse appear to be as common in drug therapy as in other forms of medical treatment.⁴

Second, given the importance of drugs in the armamentarium of modern medicine and their potential to do either good or harm, the information about drugs on which physicians and patients now base their treatment decisions lacks the scientific rigor warranted by so important a component of medical practice. For the most part, that information appears to emanate from the supply side of the market and is transmitted to physicians and patients as part of the industry’s marketing efforts.⁵

The potential for bias. One of the more astute observers of the U.S. health system, Michael Millenson, recently detailed the deficiencies of this information infrastructure in a well-documented paper.⁶ Noting that the compensation of the detail men and women who bring information about drugs to practicing physicians is "tied closely to moving revenue targets," Millenson observes that "in other industries, financial incentives, such as those long common in the pharmaceutical industry, have corrupted the information-providing process," and he cites the financial sector, whose provision of information to customers, too, has been seriously compromised in recent years.

It would not make sense, and probably would be unconstitutional, to forbid drug manufacturers from sponsoring research on the cost-effectiveness of their products, communicating the findings directly to physicians and patients, or from communicating with their customers in any way, subject only to the regulatory strictures on overtly misleading information. At the same time, a good case can be made for developing for the drug market a completely independent, unbiased information infrastructure, based on the idea that the information needed to make this market work most efficiently is inherently a public good that should be publicly financed.

Information as a public good. A public good has two important properties: It must be what economists call "nonexcludable," and it must be "nonrival." The sense of national security produced by national defense and the knowledge produced by research are classic examples of a public good in that narrow sense.

The property of "nonexcludability" means that it is technically difficult—and sometimes impossible—to prevent people from using an existing public good free of charge. The term "nonrival" means that one person’s use of a public good does not in any way detract from any other person’s ability to use the same good in the same quantity. That property implies that an additional use of a public good entails zero incremental costs. Economic efficiency therefore requires that such use should be free of charge to all potential and actual users.⁷ The information that facilitates the proper functioning of a health care market—such as that for drugs—does have those two properties.

Private underprovision of public goods. As a consequence of the two important attributes of public goods, the private sector typically does not produce these goods in socially efficient quantities. The private costs of producing the information can easily exceed the private benefit to its producer, even if the potential social benefits of the information far exceed the cost of its production. The non-excludability of the public good prevents its producer from capturing as revenue the benefits freeloaders derive from the good.

This could explain, for example, why the private health insurance industry has never actually given the insured the information infrastructure that had been hailed as early as twenty years ago as the hallmark of the model of managed competition coupled with managed care.⁸ To illustrate, several years ago the national Blue Cross Blue Shield Association founded RxIntelligence, a research arm charged with providing the association’s members with information on the relative cost-effectiveness of rival drugs. Although that organization has issued three or four research reports on commonly used drugs, it has remained in a fledgling state.⁹

The Case For Federal Funding

Top Editor's Notes The Stringent Requirements Of... The Information Infrastructure... The Case For Federal... Concluding Observations Notes

 
Every student in economics is taught that upon the nonrivalness and nonexcludability of public goods rests the case for the public financing of the production of such goods, which does not imply, however, that they need to be produced in publicly owned facilities. In the case of the information infrastructure outlined above, a good case can be made for federal financing of that infrastructure, on the grounds that the information can benefit every American user of drug products. If Congress is intent on using private market channels to administer its proposed new drug benefit for Medicare, an unbiased information infrastructure will be essential for cost-effective drug therapy and for the prudent use of taxpayers’ money.

To facilitate the production of this information infrastructure, the federal government might levy on the nation’s annual outlays on prescription drugs (currently about $200 billion) a small surcharge—perhaps one-half percentage point or less—whose proceeds would be explicitly earmarked for establishing permanent endowments for several independent PharmacoEconomic Research Institutes (PERIs).¹⁰ PERIs would be structured to function as independent not-for-profit foundations whose annual operating budgets would not depend on research grants from either the pharmaceutical industry (whose sponsorship would make the PERIs’ research suspect) or private insurers (whose sponsorship would be suspect as well, on the suspicion that insurers might be interested mainly in cost savings, with inadequate weight given to quality).

The modus operandi of the PERIs. The fiscally independent PERIs would fund both intra- and extramural research on the cost-effectiveness of new and existing drugs, perhaps in order of their relative share of total drug spending. Because the product of this research would be a pure public good, it should be freely available to all segments of society.

PERIs’ research would be strictly peer-reviewed and of a quality sufficient to warrant publication in prestigious scientific journals. Furthermore, all of the steps leading to research findings—from the retrieval of raw data to their transformation and eventual statistical and economic evaluation—would be fully transparent to any interested party. Drug manufacturers or researchers under contract to them, for example, would have full access to all of the raw data used in a particular study and be fully apprised of the methodology used.

Doctoral candidates might seek to make a name for themselves by critically examining the PERIs’ research. Alternatively, they might wish to make their career by working for the PERIs, which could easily attain the academic status of a first-rate university. One could imagine, for example, that distinguished academicians might wish to spend a few sabbatical years with equally accomplished colleagues at a PERI.

Finally, the PERIs’ research findings would not be formally binding upon any decision-maker in the health sector. They would serve merely as one reliable database on which physicians, patients, and insurers could draw to formulate their decisions concerning drug therapy and demand-side pricing.

A public agency for pharmacoeconomics. An alternative to administratively and financially independent PERIs would be a federal agency located in the executive branch, such as the Agency for Healthcare Research and Quality (AHRQ), subject to annual appropriations by the Congress and to routine congressional oversight. Although this alternative might be attractive to Congress and seems to make sense at first blush, it probably would not serve society well. First, the government, too, would be accused by the drug industry, physicians, and patients of putting cost savings above quality in its research efforts. It would only be a matter of time before the agency would be decried as an instrument of tacit rationing and would therefore be neutered. More importantly, as noted, inclusion of the PERIs in the executive branch would make them dependent upon annual appropriations by Congress. As AHRQ’s disturbing history and continued precarious existence has shown, the approach would make the PERIs vulnerable to lobbying by interest groups, because one or a few members of Congress could easily imperil the PERIs’ existence through the appropriations process.¹¹

Indeed, even the Securities and Exchange Commission (SEC), which is to stand guard over the ethical conduct of the nation’s financial markets, apparently has not been immune from untoward micromanagement by Congress. As Stephen Labaton reported in the New York Times, "When Arthur Levitt Jr., the chairman of the Securities and Exchange Commission, tried to impose tough conflict-of-interest rules on the accounting industry two years ago, he was hit by a barrage of high-powered lobbying, including calls from 10 or 11 senators. The senators...warned that if he did not relent on the new regulations, the agency’s appropriations could be cut, he said...‘I have never been subjected to a more intensive and venal lobbying campaign,’ said Mr. Levitt."¹²

Any organization charged with exploring the cost-effectiveness of health care would likely suffer the same fate were it lodged in the executive branch and subject to annual congressional appropriations. If the United States ever were serious about cost-effectiveness in health care, that research should be entrusted to a semipublic institution with the independence and the stature of the Federal Reserve.

Concluding Observations

Top Editor's Notes The Stringent Requirements Of... The Information Infrastructure... The Case For Federal... Concluding Observations Notes

 
After a brief respite during the 1990s, when U.S. health spending grew at only about the same pace as its overall gross domestic product (GDP), the growth of health spending has resumed its former pace—roughly three percentage points above the GDP growth rate. Total health spending is now projected by the government’s actuaries to absorb about 17.7 percent of GDP by 2012.¹³ If the more rapid recent growth trends prevail for the next decade, closer to 20 percent of GDP might be absorbed by health care in 2012, just as the baby-boom generation begins to retire.

In the face of these daunting growth trends and pressed by other claims on their GDP, Americans will sooner or later have to inquire whether all of their spending on health care is cost-effective. (To be cost-effective, a treatment regimen should not cost more per added quality-adjusted life year [QALY] saved with the therapy than would the best medical practice would indicate.) Although spending on prescription drugs amounts to only about 11 percent of total health spending now and is projected to absorb only about 14 percent in 2012, it is unlikely to escape the search for greater cost-effectiveness.

A publicly financed information infrastructure for drugs would be a useful platform for this search. Its main virtue would be twofold. First, it would be free from the taint of bias that now, rightly or wrongly, afflicts sponsored research in pharmacoeconomics, whether the sponsor be on the supply side or the demand side of the market. Second, the infrastructure would make this unbiased information freely available to all. In other words, its dissemination method would be inherently efficient.

The purpose of this information infrastructure would be not be to strengthen solely the demand side of the market for pharmaceuticals. In many instances, it might bolster the supply side as well. Recent research by a number of independent economists has suggested that spending on technological innovation in health care, in general, and on pharmaceutical innovation, in particular, has been among the country’s best investments in the past half-century.¹⁴ There is, therefore, no reason to suppose a priori that an information infrastructure of the sort proposed here would reduce society’s financial support for drug industry R&D. The hope is merely that it would raise the benefit-cost ratio of that research to yet higher levels than it has attained so far.

Editor's Notes

Top Editor's Notes The Stringent Requirements Of... The Information Infrastructure... The Case For Federal... Concluding Observations Notes

 
Uwe Reinhardt is the James Madison Professor of Political Economy at Princeton University.

Notes

Top Editor's Notes The Stringent Requirements Of... The Information Infrastructure... The Case For Federal... Concluding Observations Notes

 

1. See J. Hutton et al., "The Pharmaceutical Industry and Reform: Lessons from Europe," Health Affairs (Summer 1994): 98–111.
2. In 1998, for example, marketing, selling, and administration for the world’s eight largest research-based drug companies amounted to about 37 percent of sales revenue, and R&D to only about 13 percent. See Deutsche Bank Alex. Brown, Pharmaceutical Industry Outlook: Sobering Up on Drugs (New York: Alex. Brown, 11 October 1999), Figure 40. Similar figures for 1999 can be found in Deutsche Bank AG, Global Pharmaceuticals: The Six-Month Checkup (New York: Deutsche Bank AG, 6 September 2000). See also M.L. Millenson, "Getting Doctors to Say Yes to Drugs: The Cost and Quality Impact of Drug Company Marketing to Physicians," Exhibit 6, BCBSHealthIssues.com/relatives/20705.pdf (3 November 2003).
3. See S.F. Jencks et al., "Quality of Medical Care Delivered to Medicare Beneficiaries," Journal of the American Medical Association (4 October 2000): 1670–1678.
4. L.T. Kohn, J.M. Corrigan, and M.S. Donaldson, eds., To Err Is Human: Building a Safer Health System (Washington: National Academies Press, 2000), chap. 2. See also A.J. Wood, "The Safety of New Medicines: The Importance of Asking Questions," Journal of the American Medical Association (12 May 1999): 1753–1754.
5. Oklahoma’s Medicaid program last year found widespread prescriptions of drugs for uses that have not been approved by the Food and Drug Administration (FDA). Apparently, the prescribing physician’s "authority" in such cases resides in the Drugdex Information Service, which is published by a private company that, according to the report, receives "substantial revenue from the big drug companies that benefit from Drugdex’s listings." Although Drugdex might well try to base its listings on evidence in the scientific literature, FDA Commissioner Mark McClellan was quoted as stating that this "is not the kind of definitive evidence we would like to see." D. Armstrong, "How Drug Directory Helps Raise Tab for Medicaid and Insurers," Wall Street Journal, 22 October 2003.
6. Millenson, "Getting Doctors to Say Yes."
7. To be sure, legal strictures and the coercive power of government to enforce such strictures can be used to convert a pure public good—such as information—into a "proprietary good," as is routinely done when intellectual property is protected through patents. But that protection ipso facto leads to a violation of the condition of static efficiency, which requires that prices be set equal to incremental costs. The latter remain zero even for a patent-protected public good.
8. The few instances in which the insured were provided usable information on the prices and quality of health care services have remained the exception, rather than the rule.
9. Alain Enthoven, Stanford University, personal communication, 23 October 2003. Enthoven served on the Board of RxIntelligence. In his communication, Enthoven explicitly mentions the problem of financing the production of the pure public good produced by RxIntelligence with private funds.
10. I proposed such a policy for the pharmaceutical industry in U.E. Reinhardt, "Perspectives on the Pharmaceutical Industry," Health Affairs (Sep/Oct 2001): 136–149, although I do not claim originality for this general idea. The proposal to establish a federally funded but fully independent technology assessment board for health care can be found, for example, in P.M. Ellwood, A.C. Enthoven, and L. Etheredge, "The Jackson Hole Initiatives for a Twenty-first Century American Health Care System," Health Economics 1, no. 1 (1992): 149–168.[Medline]
11. B.H. Gray, M.K. Gusmano, and S.R. Collins, "AHCPR and the Changing Politics of Health Services Research," 25 June 2003, www.healthaffairs.org/WebExclusives/Gray_Web_Excl_062503.htm (3 November 2003).
12. S. Labaton, "Auditing Firms Exercise Power in Washington," New York Times, 19 January 2002.
13. S. Heffler et al., "Health Spending Projections for 2002–2012," Exhibit 1, 7 February 2003, www.healthaffairs.org/WebExclusives/Heffler_Web_Excl_020703.htm (3 November 2003).
14. See, for example, the series of essays in K.M. Murphy and R.H. Topel, Measuring the Gains from Medical Research: An Economic Approach (Chicago: University of Chicago Press, 2003); and in the September/October 2001 issue of Health Affairs, titled "The Value of Innovation."

                      What's this?

This article has been cited by other articles:

				E. J. Emanuel, V. R. Fuchs, and A. M. Garber Essential Elements of a Technology and Outcomes Assessment Initiative JAMA, September 19, 2007; 298(11): 1323 - 1325. [Full Text] [PDF]

				G. R. Wilensky Developing A Center For Comparative Effectiveness Information Health Aff., November 1, 2006; 25(6): w572 - w585. [Abstract] [Full Text] [PDF]

				R. A. Rosenheck Outcomes, Costs, and Policy Caution: A Commentary on the Cost Utility of the Latest Antipsychotic Drugs in Schizophrenia Study (CUtLASS 1). Arch Gen Psychiatry, October 1, 2006; 63(10): 1074 - 1076. [Full Text] [PDF]

				S. R. Tunis and S. D. Pearson Coverage Options For Promising Technologies: Medicare's 'Coverage With Evidence Development' Health Aff., September 1, 2006; 25(5): 1218 - 1230. [Abstract] [Full Text] [PDF]

				U. E. Reinhardt, P. S. Hussey, and G. F. Anderson U.S. Health Care Spending In An International Context Health Aff., May 1, 2004; 23(3): 10 - 25. [Abstract] [Full Text] [PDF]