SIZE AND GROWTH OF THE VACCINE MARKET

Vaccines are a very small enterprise relative to the pharmaceutical industry overall: vaccine revenues constitute only about 1.5 percent of global pharmaceutical sales (Batson, 2001). Global sales of all vaccines combined are roughly equivalent to the individual sales of such familiar pharmaceutical products such as Lipitor, Prilosec, and Zocor (Marketletter, 2002). In just three decades, the number of firms supplying routine vaccines to the United States dwindled to 5 companies that today produce all of the routinely recommended childhood and adult vaccines.

U.S. vaccine sales are estimated to be about $1.5 billion per year, one-quarter of the global vaccine market (about $6 billion per year) (Mercer Management Consulting, 2002). Most of the vaccines sold in the U.S. market are produced by four large pharmaceutical companies: Aventis Pasteur, GlaxoSmithKline, Merck, and Wyeth. Two of these companies— Merck and Wyeth—are U.S.-based; the others are based in Europe. A fifth, smaller company based in the U.K., Powderject, supplies adult influenza vaccine to the U.S. Vaccines represent a small fraction of the business of the four large companies and increasingly must compete with the companies' pharmaceutical divisions for internal resources (Arnould and DeBrock, 2002).

Mercer Management Consulting (2002) estimates that the global market for vaccines (childhood and adult) has grown approximately 10 percent per year since 1992. Globally, a significant proportion of the growth during the decade of the 1990s was the result of the worldwide effort to eradicate polio. The remainder of the market grew at an annual rate of only about 1 percent (Mercer Management Consulting, 2002). In the United States, 72 percent of the growth in revenues in the early 1990s resulted from the introduction of new vaccine products and 10 percent from the increase in the measles–mumps–rubella (MMR) dosage (from one to two doses) from 1990 to 1995 (Mercer Management Consulting, 1995). More recently, the introduction of childhood pneumococcal vaccine in 2000 nearly doubled the U.S. vaccine market.

Pediatric vaccines constitute the majority of the vaccine market (about 70 percent). Traditional childhood vaccines, such as MMR, polio, and diphtheria–tetanus–acellular pertussis (DTaP)—which represent the core of the U.S. national immunization system—are viewed by the vaccine industry as low-margin commodities. Projections of strong vaccine industry growth, however, spurred by new developments in recombinant technologies and other advances, have stimulated renewed interest in vaccines. Much of this interest is directed toward new therapeutic and cancer vaccines and adult vaccines for targeted risk groups. Some have suggested the possibility of a $10 billion market by 2010 (Hirschler, 2002). But the ability to bring new vaccines to market still involves extraordinary technical and regulatory challenges. Maintaining producer interest and stable sources of supply of routine childhood vaccines remains a significant challenge (Arnould and DeBrock, 2002).

Large, multinational producers sell vaccines through a two-tiered pricing system. Prices in developed countries are high—current prices in western Europe and the United States are comparable—while a large volume of vaccines is sold to the developing world at significantly lower, essentially marginal-cost prices. High-income countries generate about 82 percent of vaccine revenues but represent only 12 percent of doses (Batson, 2001). This system serves the needs of both the multinational companies and the developing countries. The large volume of global sales permits the vaccine companies to exploit economies of scale in production while earning high returns on sales to developed countries. European multinationals typically produce hundreds of millions of doses, while American companies produce tens of millions of doses (Mercer Management Consulting, 1995). This disparity in volume has resulted in higher average production costs in the United States than in Europe. (See also the later section on cost structure.)

VACCINE PRODUCTION

A large number of vaccines are licensed in the United States by domestic firms and foreign suppliers, taking into account multiple combinations, as well as vaccines that are not routinely used (see Tables 5-1 and 5-2). Some manufacturers are more active than others. For example, Wyeth has 16 licenses for vaccines in the United States and Merck has 13, while seven manufacturers have only 1.

TABLE 5-1

Domestic Producers of Vaccines for the U.S. Market.

TABLE 5-2

Foreign Producers of Vaccines for the U.S. Market.

While many pharmaceuticals are manufactured with relatively standardized chemical engineering processes, vaccine manufacturing is less standardized and less predictable. It often involves the complex transformation of live biologic organisms into pure, active, safe, and stable immunization components. Highly sterile, temperature-controlled environments are needed at each manufacturing step, and many vaccines must be maintained within a narrow temperature range during storage and delivery—referred to as the cold chain. Vaccines approved by the Food and Drug Administration (FDA) are subject to high standards of safety and quality assurance, including rigorous and pervasive review procedures in which each individual batch of vaccine is licensed—a procedure not required for pharmaceuticals (Hay and Zammit, 2002).

In addition, once in production, each batch must be tested and approved prior to release. Vaccines require both a product license application (PLA) and an establishment license application (ELA), while new pharmaceutical products (“new chemical entities” or NCEs) require only the former. The ELA certifies that the facilities, equipment, and personnel involved in the manufacturing process meet FDA standards and Current Good Manufacturing Practices. Furthermore, to obtain a facility license for a vaccine, a company must first create full production capacity for that vaccine (see the discussion below) (Hay and Zammit, 2002).

COST STRUCTURE¹

The costs of vaccine production include research and development (R&D) costs; costs related to the regulatory approval process; ongoing regulatory costs; plant costs, including depreciation; marketing costs; variable costs for labor, production, equipment, and supplies; and liability costs (Arnould and DeBrock, 2002).

Although there are substantial differences between development costs for vaccines and pharmaceuticals, the latter provide a useful benchmark. It has been estimated that, between 1980 and 1984, R&D and the regulatory approval process generated an average of 11 years of negative cash flow for NCEs introduced in the U.S. pharmaceutical industry (Grabowski and Vernon, 1997). DiMasi et al. (1991) estimate the mean out-of-pocket cost for a successful NCE at $32 million in 1987 dollars; when discovery, clinical testing, and failure costs are included, this figure rises to $115 million, while the inclusion of time and interest costs results in an estimate of $231 million (more than $300 million in 1997 dollars) (Grabowski and Vernon, 1997). A more recent study by DiMasi indicates that the out-of-pocket cost of an NCE has escalated to $403–$802 million (2000 dollars) when the time lag between investment and market release is capitalized (DiMasi et al., 2003).

Total development costs of bringing a vaccine to market are roughly similar to those for drugs and can be higher (Grabowski and Vernon, 1997). As part of the initial approval process, the FDA requires that the vaccines used in Phase III clinical trials be produced in a facility that will be used for commercial production if the vaccine is approved. As a result, manufacturers must frequently invest more than $30 million in the production facility prior to product approval (Grabowski and Vernon, 1997). Vaccine development costs have also risen as a result of the increased time it takes to achieve licensure, as well as larger FDA-required Phase III clinical trials for many recent vaccines (see Box 5-1). The size of clinical trials depends on a number of variables (Foulkes and Ellenberg, 2002), including the rates of disease and anticipated adverse events. The average size of clinical trials has increased over time (as has been the case for drugs) to provide an adequate base for identifying rare adverse effects during vaccine development. One industry expert estimates that a new vaccine costs $700 million from initial research to commercial production (Clarke, 2002). In addition to the requirement for early facility investments, production facilities for vaccines tend to be more capital-intensive than those for pharmaceuticals. On the other hand, vaccines tend to have higher success rates than pharmaceuticals and may be characterized by faster development times (Grabowski and Vernon, 1997).

BOX 5-1

Vaccine Development and Approval. Vaccine development begins with basic research, which is usually conducted by universities, biotechnology firms, or pharmaceutical companies. Two government agencies—the National Institutes of Health (NIH) and (more...)

Once a vaccine has been approved, the production process involves high fixed costs relative to variable costs. Fixed production costs, exclusive of up-front R&D and sales labor, represent 60 percent of total production costs for vaccines (Mercer Management Consulting, 2002). These fixed costs are not affected by changes in production volume. They are associated primarily with quality assurance activities, administrative labor, depreciation, and other manufacturing overhead. Industry representatives have indicated that increased regulatory requirements have resulted in increased costs for quality assurance employees relative to production employees. Semivariable costs make up 25 percent of total costs, excluding R&D and sales labor. Semivariable costs are batch costs that are constant per batch regardless of the number of batches (Mercer Management Consulting, 2002). Specific examples of batch costs are test animals and labor for production and testing. The remaining, variable, costs account for only 15 percent of total costs; examples of such costs are vials, stoppers, labels, packaging, and in-source components.

The costs of producing licensed vaccines have increased over the last decade as a result of several factors: mandatory removal of the mercury-containing preservative thimerosal, increased burdens associated with regulatory enforcement, a variety of improvements in vaccines that have been incorporated into existing products, both voluntary and mandated upgrading of production facilities, and increased direct provider shipment costs under new CDC contract arrangements (Hay and Zammit, 2002). Modern vaccines are also subject to constant updating and improvement, such as new stabilizers and new production technologies, as a result of scientific advances. The MMR vaccine that is currently produced for the U.S. market is far different from the version produced in 1971, having been subject to an array of technical improvements (Arnould and DeBrock, 2002).

While the costs of producing vaccines have generally been increasing, the revenues from vaccine sales have remained relatively constant. The revenue potential of vaccines is limited by the small number of vaccinations usually required. Many prescription drugs are taken by patients for years; most vaccines are administered between one and four times over a lifetime. Furthermore, vaccine production costs do not necessarily decline over time. A key factor that contributes to higher production costs is the rigid batch inspection process, which makes it difficult for companies to achieve more efficiency through a learning curve and to enjoy cost reductions related to process improvements (Grabowski and Vernon, 1997). Pressures on revenues have resulted from CDC's ability to negotiate discounted federal contract prices, federal price caps on certain vaccines since 1993, the gradually increasing public share of vaccine purchases (at discounted prices), and the addition of price competition to the government contracting process. The principal exceptions to this revenue picture relate to two fairly new vaccines—varicella and pneumococcal conjugate— which are priced higher than earlier vaccines.

RESEARCH AND DEVELOPMENT

In 2000, the leading global vaccine companies spent about $750 million on R&D (Mercer Management Consulting, 2002). This figure is significantly smaller than the $26.4 billion allocated to pharmaceutical R&D worldwide (Arnould and DeBrock, 2002). The United States has been responsible for the discovery and development of two-thirds of the world's new vaccines in the last 20 years. The major contributors to vaccine research in the United States are companies conducting industrial research, government agencies (the National Institutes of Health [NIH] and the Department of Defense [DoD]), and the academic institutions they fund.

There were 285 vaccine R&D projects ongoing in 1996 (not including HIV vaccine efforts), of which 133 were in the clinical trials phase (Grabowski and Vernon, 1997). Mercer Management Consulting (2002) reports that this activity had increased by 2000 to nearly 350 R&D projects—188 in the pre–clinical trial phase and 158 in clinical trials. The rate of U.S. approval of vaccine licenses has also been increasing. Between 1997 and 1999, 17 new licenses were approved, compared with 8 licenses between 1990 and 1992 (Mercer Management Consulting, 2002). A recent IOM study identifies additional vaccines that are expected to be developed by 2010 (IOM, 2000b) (see Box 5-3).

BOX 5-3

Vaccine Supply: The Case of DTaP. The supply of the diphtheria-tetanus-acellular pertussis (DTaP) vaccine provides a good example of trends for the industry as a whole. Key trends include the following. The total number of firms producing DTaP for the (more...)

Industrial Research

The National Vaccine Advisory Committee (NVAC) estimates that vaccine sales financed 46 percent of the $1.4 billion spent on vaccine R&D in 1995 (CDC, 1997). Vaccine R&D is conducted by both large and small companies. Large companies spent an estimated 15 to 20 percent of their product sales—about $650 million—on R&D in 1995. Many small biotechnology firms, ranging in size from 36 employees (Antex Biologics, Inc.) to over 1,600 employees (Immunex Corporation), are also involved in vaccine research. Their total sales range as well, from $500,000 (AVAX Technologies, Inc.) to almost $1 billion (Immunex Corporation). In 1995, small companies invested $250 million in vaccine R&D (CDC, 1997).

Some biotechnology firms receive funding directly from the government to develop vaccines for the military, such as vaccines against diarrhea and gastroenteritis. Other firms are subsidiaries of larger pharmaceutical companies or may be partially owned by another firm. Many small vaccine start-up companies receive a significant portion of their funding through venture capital (Arnould and DeBrock, 2002).

Some firms focus solely on vaccine research, while others emphasize multiple approaches to a single type of disease. Major targets of current research include respiratory diseases, viral hepatitis, sexually transmitted diseases (STDs), herpes virus diseases, parasitic diseases, fungal infections, and cancer vaccines. A recent breakthrough in research on the human papilloma virus (HPV) holds the promise of eliminating cervical cancer (Schultz, 2003). Vaccines in the pipeline, including recombinant vaccines for HIV, herpes simplex, diabetes, and infertility (see Box 5-2), are increasingly complex (Mercer Management Consulting, 2002).

BOX 5-2

Vaccines Expected to Be Developed by 2010. Borrelia burgdorferi Chlamydia

One of the major areas of recent research is vaccines for STDs and vaccines that can be effective in children. Extensive effort has been focused on finding a vaccine for HIV to stop the worldwide spread of the virus. Scientists have learned a great deal about how the immune system works through this research. This knowledge has spurred research on cancer vaccines, and the market for such vaccines is projected to grow significantly through 2007.

R&D projects are frequently aimed at diseases for which vaccines are not yet available (see Table 5-3). But a substantial amount of research is also directed toward vaccines that would be improvements upon or combinations of existing licensed vaccines, as well as directly competing vaccines. Considerable research is also directed toward new methods for administering vaccines, such as the recently FDA–approved nasal spray form of influenza vaccine (FDA, 2003).

TABLE 5-3

Deaths from Selected Diseases Not Yet Preventable by Immunization.

Despite these signs of commercial interest, product development is increasingly costly relative to the market potential of vaccines. The absence of a market capable of supporting production costs and providing an adequate return on investment, for example, has hampered the development of vaccines for malaria and other diseases that affect primarily the developing world (Kremer, 2000b). In 2000, the Global Alliance for Vaccines and Immunization (GAVI) was founded in part to respond to the need for vaccines in developing countries. GAVI is also interested in improving technologies for administering vaccines in the difficult environments commonly found within developing countries.

INDUSTRY CONCENTRATION

The rate of concentration in the vaccine industry has increased over the last four decades. Between 1966 and 1977, half of all commercial vaccine manufacturers stopped producing vaccines; and the exodus continued into the 1980s and 1990s. Between 1967 and 1980, the number of manufacturers licensed to produce vaccines for the U.S. market dropped from 26 to 17 (Cohen, 2002). Of the nine producers leaving the U.S. market during this period, eight were domestic firms (Sing and Willian, 1996).

These declines have continued in recent years. By 1996, a total of eight firms and laboratories were producing recommended childhood vaccines for the U.S. market (Sing and Willian, 1996). In 2002, only four firms remained (GAO, 2002). Similar exits have occurred among manufacturers of adult vaccines. The most recent of these was Wyeth's discontinuation of its influenza and pneumococcal polysaccharide vaccines (Wyeth Pharmaceuticals, 2002).

Three major factors influenced structural changes in the U.S. vaccine market from the mid-1960s through the early 1980s: (1) new FDA regulations, starting in 1972, that required evaluation of all previously licensed biological products (rather than submit data for evaluation, many firms simply withdrew from the market and requested that FDA revoke their licenses “without prejudice”); (2) growing concerns about liability; and (3) poor returns on investments relative to pharmaceutical and other products in the corporate portfolio (IOM, 1993). More recent observers have cited underlying economic reasons—particularly selected aspects of the cost structure of vaccine production relative to the size of the U.S. market—for the small number of suppliers in the U.S. childhood vaccine market. An analysis by Grabowski and Vernon (1997) focuses on four such factors: regulatory costs, liability costs, R&D costs, and low risk-adjusted returns relative to pharmaceutical products. Other observers of the recent decline in the number of producers of U.S. childhood vaccines have cited similar factors in market departures, including new safety-related requirements (removal of the mercury-containing preservative thimerosal), regulatory compliance issues, and investment decisions based on the larger portfolio of parent companies (GAO, 2002; Orenstein, 2002a).

The diminishing number of vaccine manufacturers has reduced the number of vaccine products within the U.S. market. Of the 146 vaccines approved since 1933, 62 were subsequently withdrawn from the market (see Table 5-4). Some withdrawals represent replacement decisions, whereby an approved vaccine was replaced by a more effective or safer product. The diphtheria–tetanus–whole-cell pertussis (DTP) vaccine, for example, was replaced by the safer acellular pertussis version, DTaP. (Box 5-3 illustrates trends in the supply of DTaP.) Combination vaccines have also been introduced, replacing “single-indication” vaccines in vaccination schedules. Much current R&D and product testing is directed toward expansion of combination vaccines because they generally reduce the number of doses and the administration costs of vaccination, even though they may be more expensive (Ellis and Douglas, 1994; see also Chapter 3).

TABLE 5-4

Approved Vaccines Withdrawn from the U.S. Market.

What is more troubling is the effect of economic factors on the withdrawal of vaccine products that are viewed as unprofitable or yield low returns relative to the production of pharmaceutical products. In some cases, demand for older vaccines is not strong enough to warrant continued production.⁴ Another reason for exit arises when the costs of vaccine operation and regulatory compliance are too great to support more than one producer. Often the result can be one or two suppliers, or no supplier, of the entire specific vaccine segment, as is the case with the vaccine for indicators of Lyme disease.

The changing structure of the U.S. vaccine industry reflects an international trend. As noted earlier, five multinational producers dominate global vaccine sales, although many small foreign producers exist as well. As a result, five of the current recommended vaccines in the United States have only one producer, and the others have either two or three. Prior to the 1980s, vaccine markets had regional, not global, leaders. Pasteur-Merieux and SmithKline led the European market, Merck and Lederle-Praxis were major suppliers for the U.S. market, and three Japanese firms (Takeda, Eisai, and the Research Foundation of Osaka University) were the major suppliers for the Japanese market. By the early- to mid-1990s, global acquisitions, mergers, and joint ventures had reshaped the industry as a whole (Mowery and Mitchell, 1995).

The decline in the number of vaccine producers reflects the consolidation that is occurring within the pharmaceutical industry as a whole. But it is also driven by the decisions of pharmaceutical companies to drop vaccines from their product portfolios. A rash of such exits occurred in the 1980s as a result of growing concern about liability exposure. In response, Congress passed the National Childhood Vaccine Injury Act in 1986, a no-fault damage award system designed to compensate victims who experienced adverse consequences from vaccine products, as well as to protect companies from litigation that might disrupt the production of vaccines with social benefits. Recently, litigation over thimerosal, an ethyl mercury-containing vaccine preservative, has raised questions about the effectiveness of the vaccine injury award system. The use of novel legal theories designed to circumvent the National Childhood Vaccine Injury Act is a source of deep concern to the industry (GlaxoSmithKline, 2002; Merck, 2002; Pisano, 2002).

The rate of exit from the vaccine industry raises two chief concerns. The most immediate of these is the lack of backup capacity should a manufacturer experience production problems or other disruptions. When Wyeth opted out of the production of DTaP in 2000, for example, the suddenness of the firm's withdrawal left competing firms unable to fill the gap, resulting in a 2-year shortage of the vaccine. Production efforts by other companies to compensate for such a supply disruption can take well over a year (Pisano, 2002). Longer-term concerns include the potential for the exercise of market power by the remaining firms and the potential for the total loss of supply of a vaccine product.

Industry observers have consistently issued warnings about the threats posed by the sole-supplier situation and the potential for supply interruptions given the existence of single producers of many vaccines, including 10 of the 15 recommended childhood vaccines (DeBrock and Grabowski, 1985; Arnould and DeBrock, 1993). The number of producers continues to decline as the number of vaccines provided by a single supplier increases (see Table 5-5) (CDC, 2003g).

TABLE 5-5

Number of Producers of Selected Vaccines for the U.S. Market, 2003.

REGULATION

Safety review is the responsibility of the FDA's Center for Biologics Evaluation and Review (CBER). Careful monitoring for purity and quality is required for vaccines, which are produced from or use living cells and organisms, as well as complex growth materials taken from living sources. Subtle changes in materials, process, or environment alter the final vaccine product and can affect its safety or effectiveness. Each batch must be carefully tested for composition and potency through a batch release process. Unlike other drugs, vaccines are used on healthy people to prevent disease; and as a result, vaccine production is subject to higher standards of safety than is the case for pharmaceuticals (Crawford, 2002).

Vaccine manufacturers have stated that while the regulatory guidelines (Current Good Manufacturing Practices) governing the production of vaccines have not changed substantially over the last decade, significant shifts have occurred in the interpretation and intensity of enforcement of those guidelines by CBER (Merck, 2002; Pisano, 2002). In a presentation to the committee, one industry representative illustrated this situation using a metaphor from baseball: “It's not that the strike zone has changed; it's the way it's called.” According to the vaccine producers, the new approach to enforcement reflects limited knowledge of vaccine manufacturing processes and important differences between pharmaceutical and vaccine facilities. Even minor changes in packaging may require a complete product review and relicensing of a vaccine production facility.

The FDA can also change requirements suddenly. Thimerosal is a good example (Freed et al., 2002b). Although levels of ethyl mercury did not exceed accepted guidelines, the FDA in 1999 required the removal or reduction of thimerosal in all pediatric vaccines. This decision necessitated major changes in production and bottling processes, including the replacement of multidose vials with single-dose vials that have different fill-volume requirements. Estimates of the impact of this change on product losses range from 20 to 30 percent for certain vaccines (Aventis Pasteur, 2002).

PRICING

Substantial variations occur in vaccine pricing that reflect the volume of sales and the market power of the buyer. As a general rule, most individual clinicians, clinics, and hospitals pay list price or near list price to purchase vaccines. Large group purchasing organizations, state consortia, and health plans can negotiate discounts of up to 15 percent (Mercer Management Consulting, 2000).

The federal government has historically negotiated discounts of 40– 50 percent below list price, although the discounts have been lower for the most recent vaccines. Federal contract prices averaged 75 percent less than catalog prices in 1987 and 50 percent lower in 1997 (IOM, 2000a). In contrast, discounts for varicella and pneumococcal conjugate vaccines were 9 and 22 percent, respectively, in 2002 (Orenstein, 2002b). Box 5-4 presents details on the vaccine purchasing practices of the Veterans Administration and DoD.

BOX 5-4

Vaccine Purchasing by the Veterans Administration and the Department of Defense. Veterans Administration The VA purchases vaccines using the federal supply schedule (FSS) for pharmaceutical and biological products. The FSS was established as part of (more...)

The committee identified two significant concerns with regard to vaccine prices. First, as noted earlier, the prices of new vaccines are very high relative to those of older vaccines, and these higher prices present significant problems for federal and state health budgets, clinicians, and consumers. Immunization rates may decrease when costs become too burdensome to the payor. The second concern is that some vaccine prices are too low and do not encourage desirable levels of investment in R&D and production capacity. Federal contracts for several vaccines are subject to price caps that have held price increases to no more than the rate of inflation since 1994. The price cap has been cited in particular as the major obstacle to the negotiation of a government contract for the tetanus toxoid vaccine. Detailed historical prices are shown in Table 5-6; trends in vaccine prices in current (2002) dollars are illustrated in Figure 5-1.

FIGURE 5-1

Federal contract vaccine prices in current dollars. NOTE: PPV = pneumococcal polysaccharide vaccine.

Vaccine price trends fall into four distinct periods.⁵ Up to and during most of the 1970s, vaccine prices were comparatively low and stable. Prices even decreased slightly for many vaccines during the 1970s.

The second period began in 1982, when prices increased sharply and continued to rise until the early 1990s. For example, an increase of 2,847 percent in the price of DTP occurred during the 15-year period 1977–1991 (IOM, 1993).

The third period occurred during the 1990s, when prices tended to be stable and in some cases declined. However, new vaccines appeared on the market, in some cases (e.g., inactivated poliovirus [IPV] and DTaP) replacing older vaccines at higher prices. For example, the DTaP contract price of $11.01 in 1992 was almost double the DTP price of $5.99.

The fourth period, characterized by the introduction of new vaccines at dramatically higher prices, began in the mid-1990s with the release of the varicella vaccine at a federal contract price of $32.70 per dose. This new pricing model continued with the introduction in 2000 of the new pediatric pneumococcal vaccine at $43.50 per dose (the list price was $58.00). Prices of older vaccines, on the other hand, have remained stable, despite considerable new investment in upgrading of facilities and the removal of thimerosal. However, the recent stability of pricing for older vaccines may be disrupted by the development of new combinations of individual vaccines and of existing vaccine combinations. For example, GlaxoSmithKline recently received approval for its DTaP–IPV–Hepatitis B combination. The initial list price of this new vaccine is only a few dollars higher than the sum of the prices of the separate vaccines, but the long-term pricing strategy for such new combinations is not yet clear.

TABLE 5-6

Federal and Private Prices of Vaccines Per Dose, 1983–2002 (nominal prices, excise taxes excluded).

The increase in the public share of vaccine purchases also affects prices. Because of deep government discounts, the expanding public market for vaccines results in lower average prices for their manufacturers. In addition, when contracting for the purchase of vaccines, the Veterans Administration (VA) penalizes firms for increasing prices charged to its nongovernment customers. For any increase in price higher than the consumer price index, the VA reduces the allowable Federal Supply Schedule (FSS) price that can be paid under federal contract. This practice has resulted in FFS vaccine prices as low as $0.01 per dose.

SHORTAGES

Over the past 20 years, the nation has experienced two major periods of vaccine supply shortages. The first was related to the product liability crisis of the mid-1980s, when production of DTP vaccine was curtailed as a result of manufacturers leaving the U.S. market. The second period of supply shortages lasted from fall 2000 to summer 2002. During this period, the United States experienced nationwide shortages of five childhood vaccines that protect against eight of the eleven childhood diseases prevented through routine immunization (GAO, 2002) (see Table 5-7). The recent shortages affected most of the manufacturers of childhood vaccines, with three of the four experiencing supply problems during the period (Orenstein, 2002b).

TABLE 5-7

Vaccine Supply Status in 2001–2002.

Supply problems were especially severe for vaccines that are in continuous demand, such as those for tetanus and influenza. In March 2000, there were two major producers of tetanus vaccine in the United States— Aventis Pasteur and Wyeth-Ayerst—and no shortages. By early January 2001, Wyeth had ceased production of the vaccine, leaving Aventis as the only supplier.⁶ Aventis could not scale production up rapidly enough to meet demand and was forced to ration supply (Arnould and DeBrock, 2002).

Several of these shortages were severe enough that the Advisory Committee on Immunization Practices (ACIP) recommended suspension of booster doses for tetanus–diphtheria (Td), DTaP, and pneumococcal conjugate. Shortages were most severe for Td and pneumococcal conjugate, for which there was a 40 percent shortfall in doses shipped. Shipments of varicella decreased by 26 to 29 percent. The stockpile of MMR was drawn down by 700,000 doses, but shipments were still off by 15 percent (Orenstein, 2002a). There were delays in adult influenza vaccine in the two previous seasons, and severe shortages of tetanus toxoid affected the availability of doses for adult boosters and emergency use. By July 2002, these shortages, with the exception of pneumococcal conjugate, had ended; supplies of pneumococcal conjugate vaccine are expected to return to normal in 2003 (CDC, 2003h).

No reports of regional outbreaks of preventable infectious diseases occurred during this period of vaccine shortages. However, shortages place stress on the fragile public–private partnership that delivers vaccines to the public. Public compliance with the recommended schedule can be threatened by the lack of vaccines and sudden changes in the schedule resulting from shortages. CDC reports that, as a result of the tetanus vaccine shortage, 52 percent of states suspended school immunization laws (Orenstein, 2002a). Given the recent intensity of antivaccine rhetoric, school administrators find themselves in an uncomfortable role as enforcers of laws that they themselves may not adequately understand. In a recent poll of school nurses, the majority of respondents indicated their belief that children may be receiving too many vaccines (Lett, 2002). These trends may make it difficult to reinstate school laws that are suspended as a result of shortages.

It is too soon to determine whether the recent shortages were a one-time event or an early sign of a recurring pattern. An important structural risk factor in supply disruption—the limited number of suppliers—has not changed. With only four suppliers for all universal childhood vaccines and monopoly suppliers of four of those vaccines, the United States remains highly vulnerable to disruptions in manufacturer production.

Vaccine shortages appear to result from specific and apparently unrelated causes rather than a single overriding factor (GAO, 2002; NVAC, 2003) (see Table 5-8). Vaccines affected by the shortages are both new, such as pneumococcal conjugate, and long-standing, such as MMR; and shortages have affected both sole-supplier and multiple-supplier vaccines. Some explanations for the shortages that have been advanced by the industry include problems associated with removing thimerosal from the production process, compliance with increasingly stringent Current Good Manufacturing Practices, disruptions due to plant renovations, unanticipated high demand for new vaccines, and sudden withdrawals from the market by producers. The FDA licensure process may create a structural barrier to rapid adjustment of output to address sudden shortfalls in supplies. The agency's requirement for full-scale production capacity before licensure is granted may tend to fix minimal excess capacity at start-up. Combined with the stringent entry requirements and lead times for licensure, little flexibility to adjust production remains (Arnould and DeBrock, 2002). There is also evidence that other developed countries, while not experiencing the critical shortages of the United States, are characterized by capacity constraints that could lead to shortages (Mercer Management Consulting, 2002).

TABLE 5-8

Vaccine Shortages and Their Causes.

Some have sought a relationship between vaccine pricing and shortages (Orenstein, 2002c). As shown in Table 5-9, however, short-run correlations between vaccine prices and shortages are not apparent. Prices for vaccines with supply problems are generally higher than those for vaccines without such problems. A more meaningful relationship would involve profit margins, yet even this relationship may be confounded by other variables.

TABLE 5-9

Vaccines With and Without Supply Problems (2002).

STOCKPILES

The vaccine stockpile program consists of an inventory system of storage and rotation contracts negotiated with manufacturers. Initiated in 1983 to establish a 6-month strategic reserve of each universally recommended vaccine, the program was initially funded with Section 317 funds. By 1988, stockpiles had been developed for six important vaccines and combinations (DTP, tetanus toxoid [TT], Td, oral poliovirus [OPV], IPV, and MMR). The Omnibus Budget Reconciliation Act (OBRA) of 1993 allowed VFC federal entitlement funds to be used for stockpile purchases, but approval from the Office of Management and Budget (OMB) is required for this purpose. CDC began to target purchases toward vaccines with sole suppliers to minimize financial risk. Multiple withdrawals from the stockpiles occurred between 1984 and 2002, mainly as a result of temporary manufacturing problems. The most recent drawdown was 700,000 units of MMR in 2001 (see the discussion of shortages, above). Of ten vaccines that CDC has targeted for stockpiling, only three were stockpiled in 2002 (Lane, 2002).

Building up the stockpiles to full strength and possibly increasing their capacity could help alleviate the shortages discussed earlier (GAO, 2002). Rebuilding the stockpiles would require substantial investment and OMB clearance. GAO has also recommended legislation that could authorize the use of VFC stockpiles for non–VFC-eligible recipients in cases of national shortage. But even at full strength, the stockpile program provides only a temporary buffer in cases of serious supply disruption. Given the time required for licensing a new facility and ramping up production, the stockpiles would be inadequate in the face of a total manufacturer withdrawal. No government contingency plan exists for this prospect.

Stockpiles are also costly. Moreover CDC has been conservative about developing stockpiles to minimize financial risk from, for example, a change in vaccine recommendations that could render a stockpile useless. Examples of such changes include the switch from OPV to IPV, the elimination of thimerosal from certain vaccines, and the future replacement of individual and exisiting combination vaccines with new combinations.

CDC CONTRACTING

Each year, CDC negotiates a federal contract for the purchase of ACIP-recommended childhood vaccines. CDC does not directly purchase vaccines; state and local grantees are each given a vaccine budget for the purchase of vaccines at the negotiated contract prices. With that budget, states can purchase, store, and redistribute these vaccines from their own depots or through contracts with pharmaceutical distribution companies. Some states allow clinicians to choose among competing vaccine products. States can also purchase vaccines under the CDC contract for non-VFC vaccines for other federally authorized state programs. Of the 52 percent of vaccines purchased under the federal contact, 35 percent are for the VFC program, while the remaining 17 percent are purchased by states using both Section 317 funding (10 percent) and state funds (7 percent) (Orenstein, 2002b).

Several factors in addition to negotiating leverage determine the contract prices. For some vaccines (OPV, IPV, Haemophilus influenza type b, Hib, MMR, DTP, DTaP, Td, adult pneumococcal, and hepatitis B), there are statutory price caps that were imposed at the time VFC was enacted to prevent rapid escalation of prices. The price caps hold vaccines to their price on May 1, 1993, plus an annual inflation adjustment. DTaP and hepatitis B are no longer subject to the cap. Vaccines that were approved after the enactment of the VFC program have never been subject to a cap. These include hepatitis A, influenza, varicella, and pneumococcal conjugate (CDC, 2002m).

Vaccine companies do not always bid the maximum price of the cap. For example, Merck has always bid the maximum for MMR, while Aventis Pasteur has consistently bid below the cap for IPV, despite its monopoly on that product (CDC, 2002m).

CDC has also introduced competition into the contract design. The original “winner take all” contracts were initially replaced with a multiple-supplier contract that guaranteed the largest market share to the lowest bidder (all Section 317 and half of VFC purchases). In 1998, CDC introduced the current competitive approach, under which states can purchase from the supplier of their choice at the federal contract price. Manufacturers can attempt to increase their market share by lowering their price several times during the contract period.

Private-sector buyers purchase vaccines through both wholesale distributors and direct customer sales. Clinicians typically pay high prices to distributors, but they are able to make small purchases when needed and benefit from business relationships with local distributors (Mercer Management Consulting, 1995).

In contrast with childhood vaccines, the public sector purchases a very limited share of adult vaccines. For example, only about 2 percent of the 90 million doses of trivalent influenza vaccine sold in the United States in a single year is purchased through federal contracts (Johnson, 2002). The two U.S.-based manufacturers of influenza vaccine emphasize direct sales to end users instead of to distributors.⁷ The third manufacturer is based in the United Kingdom and relies on U.S. distributors. Also, bulk-purchase arrangements are common with adult vaccines. Many employers offer mass vaccination services in the workplace. One large mass vaccinator recently reported administering over 1 million doses during the 2001– 2002 influenza season. Premier, a group purchasing association representing about one-third of the hospital beds in the United States, contracts for several million doses of influenza vaccine for its members each year (CDC, 2002n).

BARRIERS TO A WELL-FUNCTIONING VACCINE SUPPLY SYSTEM

This chapter has identified a number of barriers to a well-functioning vaccine supply system. These barriers are reviewed in turn below.

FINDINGS

The amount that the nation spends on vaccines appears to be insignificant compared with that spent on other medical and social interventions that may have lesser social benefits. While federal and state governments must address the vaccine line item as an expense to be managed, a commitment of resources substantially higher than current levels may be justified to address persistent breakdowns in the vaccine system.

The relationship between financial returns to the vaccine industry and future investment in production capacity and R&D is a fundamental concern addressed by this study. While proprietary industry information was not available to the committee, a large body of indirect and secondary evidence suggests that high development and production costs and stable revenues have constrained investments in new products within the vaccine industry as a whole. While many new candidate vaccines are in early stages of development, the overall level of investment in vaccine products is too low to support the level of R&D that is desirable in light of the social benefits of immunization. The committee finds that

• The U.S. vaccine market is small relative to total expenditures on personal health services and pharmaceuticals. The entire global market for vaccines is roughly equivalent to the sales of certain individual blockbuster drugs.
• The supply of U.S. vaccines is becoming highly concentrated, resulting in limited backup capacity in the event of supply disruptions.
• Inadequate build-up of vaccine stockpiles has limited their remedial effect on recent shortages. The development of 6-month stockpiles would help avert short-term disruptions in supply but would not address more fundamental concerns, such as the continuing loss of suppliers from the industry.
• The risks and costs to manufacturers associated with vaccine production have increased. Key factors include regulation, removal of the preservative thimerosal, and an increase in vaccine injury lawsuits.
• FDA resources for vaccine regulation have not kept pace with the growth and complexity of vaccine products. FDA regulation has shifted from a focus on science to a focus on enforcement. This shift may increase the risks and costs associated with vaccine production without increasing safety.
• The pace of vaccine R&D, particularly in the discovery stage, is currently high, but commercial development is impeded by pricing and industry returns. Investment in production capacity for existing vaccines is especially problematic.
• FDA licensure requirements—including the increasing size of clinical trials, the requirement that companies build full production capacity before licensure, and the inadmissibility of clinical data from outside the United States for U.S. licensure—create substantial barriers to entry.
• The requirement for building full plant capacity in advance of approval may limit fixed capacity and increase the chances of shortages.
• Vaccine company investments in R&D on new vaccines are sensitive to prices and expected returns on investment. Ensuring socially desirable levels of R&D may necessitate prices that are substantially higher than current prices for most routine childhood vaccines.
• By using its bargaining power to achieve substantial discounts in federal contracts, CDC may substantially undervalue vaccines and reduce industry incentives for investment in both R&D and short-run production capacity.

Footnotes

1

Information on the costs and revenues associated with vaccine production is difficult to discern from the public record. The committee sought this information as part of its factfinding effort by commissioning background papers on the vaccine industry (Arnould and DeBrock, 2002; Fine, 2003; Lichtenberg, 2002), corresponding with the five companies that produce recommended vaccines for the U.S. market (Aventis Pasteur, GlaxoSmithKline, Merck, Powderject, and Wyeth), inviting testimony in committee meetings from vaccine representatives, and condusting private interviews with company officials. This process yielded a substantial amount of qualitative information in support of the committee's analysis of the relationships among costs, revenues, returns, and investment in research and development (R&D). But verifiable, quantitative information on costs, revenues, and profits is lacking; and this lack of information represents an important limitation of this study.

2

Patents may sometimes be awarded for different vaccine products that are close substitutes, as is sometimes the case with pharmaceutical products. In these cases, monopoly profits may not be obtained.

3

To encourage the marketing of drugs that may have been patented long before potential FDA approval, the Waxman Hatch Act guarantees a minimum of 5 years of exclusive marketing to the patent holder for any FDA-approved pharmaceutical.

4

The producer of a vaccine for Lyme disease, for example, pulled out of the market because of inadequate product demand.

5

This discussion is based on price data supplied by CDC (2002l; 2003d).

6

Brichacek (2001) indicates that the Wyeth pullout from manufacturing the tetanus vaccine may have been related to a June 2000 issue with the FDA concerning acellular pertussis vaccine.

7

One of the two domestic producers recently dropped out of the influenza vaccine market.