argues that there is no such thing as a perfect drug and the FDA is right to focus on scientific strategies for better determining which patients are most at risk for adverse reactions from medicines that are otherwise highly safe and effective.
For the Center for Science in the Public Interest, the perfect medication would quell all interest in fat, sugar, alcohol, coffee, milk, meat, cheese and fruit juice. Eat nothing tasty, and society won't need a selection of medications that, the center says, all do the same thing.
The Food and Drug Administration lacks the luxury of making such outlandish but oftquoted statements. As the federal agency that vets the effectiveness and safety of pharmaceutical products, the FDA knows what every patient knows: One pill does not fit all.
Vioxx, for example, worked wonderfully well for millions of patients worldwide but during its four years on the market maymayhave caused or exacerbated fatal coronary problems in several hundred patients.
That experience prompted a report on FDA policies and procedures on drug safety by the congressionally chartered but privately financed Institute of Medicine. Released last fall, the report concludes generally that the FDA too quickly approves drugs and too slowly yanks them off the market because it is too close to the pharmaceutical industry and its scientists.
The FDA's critics, however, and the private foundations that fund their scientific research can also be biased, usually toward government control of medicine, period.
On occasion the FDA has erred in quick approvals, and in quick rejections. From the institute's report, it has culled some sensible changes: It will extend safety studies to 18 months after approval. It will research why some people react adversely to drugs that greatly benefit most patients, then develop a model to predict and therefore avoid adverse reactions. That's the only way so far to let patients who greatly benefit from medications have them, withhold them from patients who don't and find the latter a safe alternative.