Are prescription drugs safe? Not necessarily

The Food and Drug Administration approved them, your doctor prescribes them, and you see them advertised on TV-so your medications must be safe, right? Then why is Consumers Union bothering to publish alerts about drug hazards?

More than half of all prescription drugs cause adverse effects—some serious or fatal—that aren't detected until after the FDA approves them, sometimes many years later. Such delayed detection contributes to the high number of drug-related injuries in the U.S. In 2008 alone, the FDA received more than 100,000 reports of serious injuries related to adverse drug events, an increase of about 25 percent over the previous year, according to the Institute for Safe Medication Practices.

Some of the delay is inevitable: It's simply not possible to detect every risk before doctors start prescribing a drug. Doing so would require clinical trials that would be prohibitively large, long, and costly.

In addition, the U.S. system for identifying drug risks before approval is flawed—and the one for spotting them afterward is even worse. The 2007 FDA Amendments Act promised many changes in the drug-safety system, but whether they will be sufficient, even if the agency uses its new powers effectively, remains to be seen.

Here's a closer look at why you face these unexpected dangers, and what you can do to protect yourself.

Holes in the preapproval process

Clinical trials submitted to the Food and Drug Administration before approval have several limitations, some of them unavoidable. They can't adequately assess safety because they're:

• Too small. They typically involve only about 500 to 3000 volunteers, enough to spot only the more-common adverse effects. Rarer ones may emerge only after millions of people have taken the drug.
• Too short. The trials may last for just a few months, but some adverse effects develop only after patients take a drug for many years.
• Too unrealistic. Most trials are conducted under scrupulously controlled conditions, with carefully selected patients in order to demonstrate that the drug actually works. Indeed, randomly-controlled trials are the gold-standard for doing this. Yet adverse events are not the focus of the studies and therefore may not be detected.

In the real world, patients may receive lax care that exposes them to greater risks; moreover, doctors are allowed to prescribe approved drugs for unapproved purposes, sometimes for patients at considerably higher risk than the trial volunteers. The trials also tend to include few women and minority-group members.

And they often use clinical lab results—for example blood cholesterol levels—to determine how well a drug works rather that measuring how often it helps a person avoid having a heart attack or if it helps them to actually live longer. Such measures may represent a misleading picture of a drug's risks and benefits.

In addition, the FDA may approve some new drugs more quickly that others. The time devoted to preapproval review, particularly for drugs deemed "priority," has dropped substantially since 1992. Congress passed a law designed to speed up the FDA approval process for some drugs to get them to the market faster. But doing so may have come at a significant cost. A 2003 survey of FDA reviewers indicated that they generally felt rushed and, in some cases, pressured by their supervisors to approve medications. Some studies—though not all—have found that the faster the review, the greater the chance of hazards emerging after a drug's approval.

Bigger holes after approval

Because all hazards can't be reliably detected before approval, some are discovered afterward. Unfortunately, the system for detecting risks after approval has been the weakest link in the drug-safety system, for several reasons:

• Incidents aren't reported. Patients, physicians and other health professionals are encouraged to inform the FDA and drug manufacturers about any new adverse drug effects. But it's voluntary effort, and estimates suggest that only a small fraction of such incidents—about 1 percent to 10 percent—are actually reported.
• Studies after approval aren't done. Clinical trials published after a drug's approval can reveal previously unknown hazards. The FDA sometimes requests such studies, but companies have usually delayed or avoided them. For the first time, the 2007 law allowed the FDA to require, rather than merely request, safety trials after a drug's approval and to impose penalties for delays, but whether that is sufficient is not yet clear.
• Studies aren't published. Drug companies sponsor many studies that are never published in medical journals, often because of unfavorable results. One notorious example: Published trials sponsored by Merck, maker of rofecoxib (Vioxx), concluded that the drug was "well tolerated." But the company's internal analyses of the data, which wasn't given to the FDA or made public until much later, showed a significant increase in the risk of heart attacks and death for people who took rofecoxib.
• Surveillance isn't adequate. Drug hazards can be detected by searching large databases of patient records or insurance claims, from Medicare or the Veterans Administration, for example. But the U.S. currently has no coordinated, computerized system for regularly scanning and analyzing such data. The 2007 law called for the creation of such a system, but it's not scheduled to be completed until 2012.
• FDA hasn't been effective. The Institute of Medicine, a private group that advises the government, issued a report in 2007 that sharply criticized FDA's drug-safety efforts. Indeed, the agency's safety division was just one-tenth the size of the approval division. The bigger group, less focused on safety, dominated the final decisions about warnings and restrictions even after a drug's approval. The 2007 law mandated an increase in the safety group's size and status, as well as an improvement in the cooperation between the divisions. It's too soon to determine the results from that effort.

Misleading drug ads

Despite all those problems, pharmaceutical companies typically start advertising a drug within a year of approval, long before anyone knows whether its benefits outweigh its risks. And the companies may minimize the known risks, often by running distracting or reassuring images while mentioning adverse effects.

The FDA has lacked the resources and authority to regulate drug ads effectively. Months can go by before it catches up with misrepresentations, exaggerated claims, or inaccuracies. While the 2007 law authorized more FDA staffing to scrutinize ads and more power to regulate them, companies still don't need the agency's go-ahead before running their promotions in most cases, though some do so voluntarily.

What you can do

Even if the FDA reforms are fully implemented, taking prescription drugs, especially new ones, still requires caution. To protect yourself, take these steps:

• Don't overreact. In general, if you hear that a drug you're taking may have a newly discovered risk, don't stop abruptly. Ask your doctor whether the benefits outweigh the risks for you and whether other options can achieve adequate results.
• Consider older options. If you're thinking about taking a drug that's been around for only a few years, ask your doctor whether there's an older, equally effective one (it will probably be less expensive, too). Consult a nonbiased source, such as this Web site from Consumer Reports Health or the National Library of Medicine's Medline Plus, to learn more about the safety and effectiveness of the various options.
• Resist the ads. Don't let direct-to-consumer advertising move you to pressure your doctor to prescribe a drug unless other options haven't helped and there's strong evidence that you need it.
• Watch for alerts. Check this site for reports about newly discovered drug risks. They're created by Consumers Union and an independent academic group, Research on Adverse Drug Events and Reports (RADAR), devoted to detecting hazards of approved medications. Check the FDA site as well: http://www.fda.gov/Safety/MedWatch/default.htm.
• Do your part. Report any serious drug reactions to your doctor. In addition, you or your doctor can report them to the FDA by telephone (1-800-FDA-1088) or by completing a form on the agency's Web site, at http://www.fda.gov/medwatch/report/consumer/consumer.htm.