Resident Fellow Scott Gottlieb

Almost 13 years after the drug Bexxar was first used in cancer patients, the Food and Drug Administration cleared it for marketing in June 2003 to treat a particularly deadly form of non-Hodgkin's lymphoma. Bexxar represents a leading edge of cancer innovation, attaching a radioactive payload to a protein that is designed to hone in on cancer cells and unload its toxic cargo. The drug isn't a certain cure but has clearly prolonged many lives in its four years on the market, and might have already benefited twice as many patients if it didn't spend an equal number of years awaiting FDA approval.

Bexxar got hung up inside the agency, in part, over questions about its safety. The FDA spent years debating what restrictions to place on Bexxar's prescription in order to "manage" these uncertainties, settling on some measures that ultimately slowed the drug's adoption. Not all of the safety concerns were borne out once Bexxar was approved and widely used, either. While hindsight is easy, the drug's protracted review should be a cautionary tale to Congress as it contemplates the most significant new drug approval regulations in decades. While ensuring safety is paramount, drug delays and restrictions on access also carry human costs.

The new drug safety legislation, which is attached to a larger bill that renews the FDA's principal funding stream called the Prescription Drug User Fee Act, will change the way drugs are used by patients through provisions that give the FDA more control of medicines after they are approved. One central measure would put the FDA squarely in the role of regulating medical decisions in order to "manage" drug risks--by giving the agency unprecedented new authority to control the way drugs are distributed by pharmacies and prescribed by physicians. It's a watershed measure, one that will grant the FDA some of the same responsibility for regulating medical practice that has been traditionally left to the states and professional medical bodies.

The provision, referred to as risk management or "REMS" in the Senate legislation, gives the FDA legal power to restrict the doctors who can prescribe drugs, and even which patients can purchase them. These risk management plans are being offered by some in Congress as a solution to gaps in our knowledge about safety-related issues with new drugs.

There's no question that our system for monitoring drugs after approval for both new benefits as well as side effects is woefully inadequate--and some drug companies themselves have historically been reluctant to do the kinds of aggressive investigations needed to surface safety problems. Anecdotes of industry foot-dragging have cast a cloud over the entire safety debate. But Congress, and the FDA, have ample opportunity to plug information gaps by dealing with these issues directly, rather than through regulatory measures that simply impede access. The logic of "risk management" seems to be that drugs become safety problems only if doctors choose to prescribe them.

Some of the elements used in REMS include restricted distribution of newly approved drugs to only certain pharmacies or doctors, required education programs for prescribers and patients, mandatory blood or urine monitoring for drugs that can cause organ damage or birth defects, and limits on the initial launch of some drugs to only certain "sentinel" sites. The legislation's unspoken belief is that specialist doctors are more attentive to drug warnings and should be preferred prescribers for "risky" medicines.

These measures may sound like practical ideas, but there are problems when federal authorities impose rigid solutions on a diverse healthcare system ill-equipped to deal with one-sized approaches to care, and on patients and doctors that require latitude to exercise individual preference in their treatment plans. Some REMS include agreements to restrict sales and marketing in order to limit which doctors actually prescribe the medication, or to limit what can be put on a drug's label. Drug companies are expected to enforce these requirements on doctors and pharmacies through flimsy contracts that carry little legal recourse beyond the threat of being cut out of the drug supply chain. The FDA will eventually find that it is folly to endeavor to regulate medical practice. Trying to do so is going to set the agency up for failure, although a lot of obstacles and uncertainty will be created along the way. Ultimately, the agency cannot regulate away the fact that some doctors and patients deliberately disagree with the agency's advice or mistakenly ignore important warnings.

While the agency has used "voluntary" risk management measures in the past for a limited number of approvals, it has mostly adopted these restrictions with caution, recognizing the intrusions they cause into patient care. Any forbearance is swept away by the new legislation, which provides economic and political incentive to seek wider use of these kinds of restrictions. Right now, the only legislative mechanism the House bill allows for the FDA is to waive any risk management requirements for drugs that pose no risk to anybody, a faux standard that's impossible for even simple medicines like blood pressure pills to meet.

The intrusions enabled by REMS will be especially burdensome for patients who live in rural or inner-city settings, where busy clinics don't have the resources to comply with the restrictions, and patients often don't have easy access to the kinds of specialist providers ordained by many REMS to prescribe certain drugs. Such was the case with one recently approved diabetes drug that worried the FDA because of its propensity to induce low blood sugar when improperly combined with too much insulin. To deal with this risk, the FDA required that doctors have nurse diabetes educators on their staff to counsel patients before doctors could prescribe the medicine. Few inner-city or rural providers can afford such a luxury as specialized support staff, ensuring that the drug is less likely to be used in those settings.

It's especially difficult to see how direct regulation on physician prescribing is really going to help improve drug safety, especially since the kinds of problems we are increasingly concerning ourselves with are hard to uncover or even mitigate through reasonable measures.

A decade ago, we were mostly focused on uncovering the remote risk that certain drugs could, in rare cases, cause immediate, unexpected and potentially life-threatening problems such as liver damage or disturbed heart rhythms. Surfacing these problems was made easier by the fact that the side effects occurred in close proximity to administration of the medicine, so it could be deduced when the drug was at fault.

Now we're focused on unearthing cases where prolonged use of some drugs may prompt small elevations in the risk of otherwise naturally occurring and even common events, like heart attacks. This was the issue that caused the withdrawal of the painkiller Vioxx and the premature flap around the diabetes drug Avandia.

Our heightened vigilance will surely improve the information people have to make informed choices. But finding and proving these kinds of rare and latent problems will often take many years and entail large and rigorous clinical testing. Even then, deciding if such remote kinds of risk outweigh benefits requires an often subjective calculus that is best left for patients and doctors to decide individually.

All of these ideas about "managing" risks by regulating medical practice assume that safety is a binary event--that products are completely unsafe and ineffective until the day they are approved for marketing and then made fully safe the day after. It assumes that products with certain questions about their safety will always have those doubts surrounding their use, and that static regulatory measures on how they are used therefore need to be erected.

The reality is that information about risks and benefits of new products is an evolving continuum. Binary approaches to regulation based on snapshots of risk encourage policy makers to erect overly conservative barriers to market access and restrictive approaches to manage how products are used. While this may provide public confidence for some, it will certainly lead to reduced and delayed access to potentially life-saving medicines for many.

Congress would do well to strip out the efforts to "manage" risks and thus regulate medical practice, and focus instead on a section already in the bill to improve measures for surfacing safety-related information through advances in post-market monitoring. The provision, called "RAMS" in the legislation, creates tools for actively collecting "signals" about potential safety problems and would help assemble the growing troves of electronic medical data available to health plans into large registries the FDA can use to better analyze its safety suspicions. This in turn should provide opportunities to streamline the development side of the equation, and allow products to come to market earlier because of confidence that rare but significant safety signals can be quickly identified and addressed post-approval.

When it comes to trying to manage drug risks and encourage safe prescribing, the FDA has plenty of reason to worry about how drugs get used. Many practitioners have a hard time explaining remote risks to their patients, and many patients a harder time understanding them. But the FDA can't rationalize the diversity of peoples' medical decision-making or the fact that some doctors, and patients, will simply choose to reject the agency's medical practice advice or its computations when benefits justify certain risks.

Nor will any amount of regulation on how medicines are used defuse the risk that comes from medical care, or the uncertainty that inevitably lingers around new treatments. When it comes to medical practice, the agency's job isn't to find reasons to keep effective drugs from patients by actively managing prescribing decisions, but to surface obtainable information about risks and benefits so that patients can make informed choices for themselves.

Scott Gottlieb, M.D., is a resident fellow at AEI.