A New Era in Drug Regulation?

The Food and Drug Administration has a difficult job. With a small staff and budget, the agency is assigned to regulate about one-fifth of the nation's economy. And it must do so with several policy considerations in mind, considerations that often cannot be neatly reconciled. The FDA is expected to protect the public health by ensuring that drugs and devices are safe and effective, but it must not be so cautious as to unnecessarily delay patients' access to novel interventions. It is expected to ensure that new products are carefully evaluated, but it must not unduly impede manufacturers' legitimate interests in delivering their products to market.

Though the FDA has extensive power, Congress sets the boundaries of FDA authority. When members of Congress believe the agency's policy concerns are out of balance, they enact legislation designed to rectify the problem. High-profile drug tragedies, corporate lobbying, and patient advocacy are just a few of the forces that can trigger congressional action.

For most of the twentieth century, both Congress and the FDA focused on protecting patients from dangerous drugs. This philosophy emerged after drug disasters such as those involving elixer sulfanilamide and thalidomide revealed the harm that could come from inadequate government controls. But the 1980s brought increasing dissatisfaction with the status quo. Business and patient groups claimed that the FDA's product approval process had become too rigid and time-consuming. New drugs and devices desperately needed by HIV and other seriously ill patients were being kept from the market by narrow-minded bureaucrats.

In an effort to promote faster access to new products, Congress passed new legislation, including the Prescription Drug User Fee Act and FDA Modernization Act.¹ By late 2004, however, another policy shift seemed imminent. New disclosures about risks posed by arthritis and antidepressant drugs triggered worries about the adequacy of FDA review. After two decades of emphasizing the need for speedy decisions and broad availability of new products, members of Congress were loudly proclaiming the value of rigorous testing and oversight. They called for legislative and regulatory action to promote drug safety and effectiveness.

Reform Proposals

Reformers endorse two major changes. One is to expand access to research information through establishing a clinical trials registry. This congressional proposal builds on the International Committee of Medical Journal Editors' decision to start demanding registration for the studies they publish.² The Fair Access to Clinical Trials Act would require sponsors to register trials before subjects were enrolled and report results upon trial completion. The bill makes registration a condition for Institutional Review Board approval of human trials. Registration would also be a prerequisite for receiving federal research funding. The bill describes information that sponsors would be required to report and authorizes audits by federal officials. Violations could result in large fines.³

Supporters say a mandatory registry is needed for many reasons. Without such a registry, information about investigational and clinical interventions remains incomplete. The scientific literature presents only a partial picture of research findings. This is because investigators and professional journals are prone to "publication bias," preferring to publish studies showing benefits from novel interventions. Sponsors may also suppress or manipulate trial results so that investigational drugs appear better than they are. Trials finding new approaches less effective or linked to serious side effects may never see the light of day.

Because a fraction of trial findings are presented to the wider world, physicians and patients are denied an accurate picture of medical options. Moreover, the dangers and relative ineffectiveness of new products may go unrecognized until they materialize in the general patient population. The situation has economic costs, as well. Limited resources can be wasted rediscovering information produced in earlier unreleased studies. Because new drugs tend to be more expensive, publication bias can also elevate demand for high-priced products. Unpublished trials conflict with human subjects protections, too, for they expose subjects to risks and burdens without adequate justification. To be consistent with the principle of beneficence, human research must contribute to social knowledge. But when results go undisclosed, no one can learn from them.⁴

Proposals for trial registries are not...