Regulatory and Liability Issues
Developing and registering new products are generally lengthy and complicated processes (Abraham and Reed 2002; Baylor and McVittie 2003; FDA 2004) that are regulated both at the national level and, in some circumstances, at the international level. The role of the regulatory system extends beyond the launch of a new product to manufacturing and compliance standards and to post marketing surveillance for clinical effects and potential untoward outcomes. For products that are intended to be deployed in global markets, manufacturers have to comply with regulatory requirements in the country of origin as well as the requirements of each country where the product may be marketed. One exception is the mutual recognition systems used currently by European Union countries (Pignatti, Boone, and Moulon 2004). The situation may be different for products intended for use only in developing countries; however, for legal and liability reasons, manufacturers in developed countries have refrained from working with two different sets of regulatory requirements.
The best example for illustrating this process is the FDA (2004). Over the years, FDA regulations have developed into a clear pathway. The process is initiated through an application by the manufacturer and a step-by-step approach toward licensing. The agency gets involved in every phase of the development process and approves in advance the experimental design, assays, and endpoints for clinical trials. After it has collected all the information, the agency examines the materials submitted and reaches a decision. The FDA process extends through regulating and approving marketing materials and post licensing collection of efficacy data and information about possible side effects.
The FDA approval process differs somewhat for pharmaceutical products and vaccines. One of the main differences is the obligation of vaccine manufacturers to prepare materials for use in phase 3 trials in the final and approved production facility. This requirement means that the firm must invest in completing the manufacturing plant well ahead of launching a specific product, a process that can take three to six years. The regulatory process for vaccines also dictates batch release for every batch ready for deployment in the marketplace. This part of the regulatory process, although it ensures quality control, adds to costs and to the timeline.
In 1996, the European Union adopted a centralized procedure for applications and approvals through the European Medicines Evaluation Agency and through a mutual recognition process (Pignatti, Boone, and Moulon 2004). In many ways, the procedure parallels the FDA process, with several differences reflecting the fact that the European Union consists of many countries, each with a country-based process that remains as an alternative or an addition to the community wide process. The International Conference on Harmonization of Technical Requirements for Regulation of Pharmaceuticals for Human Use was established to achieve coordination of the process of drug development between industry, Japan, the United States, and the European Union (Abraham and Reed 2002; Ohno 2002). The conference's activities have improved understanding of the regulatory process and reduced duplication.
In contrast, the absence of a unified or harmonized approach to product registration and approval at the global level adds multiple layers of complexity. National systems consist of complex processes with differing thresholds and interpretations and with changing requirements in addition to differing Global Manufacturing Program standards and enforcement. A number of recent attempts have been made to resolve the issue. First among these is the World Health Organization's effort to expand its prequalification system, to develop technical standards earlier in the approval process, and to expand the availability of reference reagents for international calibration (Milstien and Belgharbi 2004). These efforts aim at injecting a higher level of quality control and transparency into the global regulatory system. The effort may have the potential to provide a global process that transcends national borders. Such a process should provide a simplified, systematic, and disciplined system that would reduce costs and speed up market access for new products.
The issue of liability in relation to harm to individuals receiving pharmaceutical products has been extremely significant in U.S. product development. It is entirely appropriate for those developing new products to be sued if they are negligent in their research or product development, but in some cases pharmaceutical firms have been sued for side effects of drugs that may have been unforeseeable or may not even have been the result of the product. This type of liability can be a barrier to product development. Although perhaps a less serious concern since the 1993 Daubert v. Merrell Dow Pharmaceuticals lawsuit in the United States, a case that has been interpreted to restrict the presentation to juries of evidence determined not to be "scientific," the issue is still significant. It may also be part of the reason the U.S. vaccine industry has shrunk significantly, and it has certainly affected the direction of investment, pushing it away, for example, from products such as vaccines that are used in one or a few doses in healthy people toward products used repetitively by those who already have a chronic disease (Institute of Medicine 2004). It, thus, provides pressure directly contrary to public health priorities, which emphasize prevention and, therefore, the use of vaccines.
Whether or how this trend in the United States will affect the developing world is unclear. Europe has moved toward a liability system somewhat similar to that of the United States, but many developing nations may not have such a tort liability system. Even if they do not have such a system, groups participating in pharmaceutical development might be sued in the United States for harm occurring in the developing world. Doctrines exist that restrict such suits, but firms may fear that these doctrines are insufficiently effective. Hence, recognizing the potential costs of protecting against liability and, at the same time, ensuring that products are designed and manufactured to the highest standards will be important.
Notes
1. The phase transition probabilities in DiMasi, Hansen, and Grabowski (2003) and the overall success probability of 0.215, conditional on entering human trials, are estimated from a larger sample of 538 investigational compounds first tested in humans between 1983 and 1994.
2. The size of trial required to estimate statistical significance depends on the magnitude of the drug effect; the extent of stratification within the total sample by patient age, condition, and so on; the required statistical confidence; and other factors.
3. DiMasi, Hansen, and Grabowski's (2003) data for average cost and average number of patients are based on actual, retrospective cost data, whereas the GATB estimates are prospective estimates (best guesses) based on prior clinical trials for tuberculosis drugs in the United States and a survey of clinical trial experts to determine administrative and data management costs.
4. The MMV estimate of in-kind contributions does not include the value of basic research conducted by universities and foundations from which it obtains its lead compounds. Similarly, commercial firms also benefit from such basic research and it is omitted from the DiMasi, Hansen, and Grabowski (2003) estimates, so comparisons are not necessarily biased by this exclusion.
5. Tropical diseases include parasitic diseases (malaria, African trypanosomiasis, Chagas disease, schistosomiasis, leishmaniasis, lymphatic filariasis, onchocerciasis, and intestinal nematode infections); leprosy; dengue fever; Japanese encephalitis; trachoma; and infectious diarrheal diseases.
6. This section is based in part on Barton (2004).
