If you've read the press coverage prompted by the federal government's crackdown on ephedra, you may be under the impression that the herbal stimulant is good for nothing but killing people. But the RAND Corporation study the Food and Drug Administration cites to justify warning labels actually shows that ephedra works and that it's remarkably safe when compared to many over-the-counter and prescription drugs.

Based on its evaluation of controlled trials, RAND reports that dieters who use ephedra achieve "a weight loss of approximately two pounds per month greater than that with placebo, for up to four to six months." (Longer-term studies have not been done.) More-limited research involving "small samples of fit individuals" indicates "a modest effect of ephedrine plus caffeine on very short-term athletic performance."

As for the risks, the news media keep citing an estimate of "at least 100 deaths" (or, in the case of The New York Times, "more than 100"), with no time frame specified. The RAND study looked at every report of an "adverse event" in the FDA's files and in the medical literature that was available as of September 30—a total of about 1,600 cases. The reports included only two deaths that were documented and investigated to rule out other causes. RAND also identified "four myocardial infarctions, nine cerebrovascular accidents, one seizure, and three psychiatric cases" that fell into the same category of "sentinel events."

Even in such cases, the FDA concedes, the reports "may indicate a safety problem but do not prove that ephedra caused the adverse event....Other unmeasured factors may have contributed, and such serious adverse events are likely to happen (albeit at very low rates) among the millions of users of ephedra anyway."

Consider a few more facts that put these "sentinel events" into perspective:

1. The reports cover a period of several years (exactly how many isn't clear).

2. The industry estimates that 12 million to 17 million Americans consume something like 3 billion doses of ephedra products each year.

3. In 1999 alone, the government's Drug Abuse Warning Network reported 641 deaths linked to diphenhydramine (Benadryl), 477 deaths linked to the antidepressant Elavil, 427 deaths inked to acetaminophen (Tylenol), 305 deaths linked to Prozac, and 104 deaths linked to aspirin.

FDA Administrator Mark McClellan concedes that "serious adverse events from ephedra appear to be infrequent" and that "it may be possible to use ephedra safely, in small doses, for some purposes." He is nevertheless compiling a record to show that ephedra "presents a significant or unreasonable risk of injury" so he can ban it.

The rush to condemn ephedra is apparent in the response to last month's death of Baltimore Orioles pitcher Steve Bechler. Bechler collapsed during a workout in Ft. Lauderdale and died of heatstroke. The Washington Times notes that he was overweight and had "an enlarged heart," "borderline hypertension," and "an abnormal liver." Yet it says "medical examiners...have identified dietary supplements containing ephedra"—which the paper describes as an "amphetaminelike stimulant"—"as a key factor in the pitcher's death." That's a bit different from the account in The New York Times, which reports that "a Florida medical examiner said the supplement might have contributed to his death." (Emphasis added.)

Bureaucrats are just as susceptible to scary anecdotes as journalists are. "I would not take [ephedra]," says Secretary of Health and Human Services Tommy Thompson. "I do not know why anyone would take these products. Why take the risk?"

I have a feeling that's a question the government will not let consumers answer for themselves.