As the son of a doctor and the husband of another, I'm always on the lookout for irresponsible health journalism. And, boy, did I find some this weekend.

Yesterday's New York Post featured an "EXCLUSIVE" headlined "My girl died as 'guinea pig' for Gardasil." The article, by Susan Edelman, tells the story of Jessica Ericzon, a 17-year-old who died of unknown causes two days after receiving a dose of the vaccine, which prevents the strains of human papillomavirus most commonly responsible for cervical cancer.

An average reader would come away from this tear-jerking (and genuinely tragic) account with the impression that Gardasil is a dangerous, poorly-understood drug that only remains on the market because the Food and Drug Administration is under the thumb of Big Pharma, or perhaps just has it in for adolescent girls. If that reader were a young woman, or the mother of a teenage girl, she might be frightened away from a vaccine widely considered safe and proven to ward disease, infertility and even early death.

But Edelman's reporting, while heavy on scare-mongering, is suspiciously light on the critical context a layperson needs to evaluate the evidence. Edelman notes that the Vaccine Adverse Events Reporting System, operated by the FDA and the Centers for Disease Control, "has collected 8,000 reports of problems after Gardasil shots, including paralysis, seizures and miscarriages. Seventeen other deaths [in addition to Ericzon's] have been reported since Merck & Co. introduced it in 2006."

That sounds truly awful, doesn't it? But it's less shocking if you understand how VAERS works. Anyone at all can report an "adverse event"; the database includes complaints so vague, the patient can't even be identified. Physicians are asked to report any and all possible side effects, whether they think them related to a vaccination or not. In other words, if you get a measles booster today, eat some bad oysters next Sunday and go in Monday morning complaining of food poisoning, your doctor is supposed to report your vomiting and chills to VAERS, even if he knows it was probably the oysters.

So to understand what to make of those 8,000 reports of serious problems, you need to know how large the universe of Gardasil-injected patients is. The FDA and CDC don't know how many women have received Gardasil in the two years it's been available, but they do know roughly how many doses have been distributed: more than 11 million in 2007 alone. (Not that you would know that from the Post story.)

Anytime you're tracking a group of millions of people over a period of years, some of them are going to have freaky, terrifying medical problems, and some of those freaky problems are going to arise, as a matter of chance, shortly after a vaccination. The proper question is whether those problems are occurring at a greater frequency among women who've received Gardasil than among women who haven't.

And every time anyone has looked for such an association, they've come up blank. (Actually, not every time: Researchers have determined that girls who got Gardasil shots were more likely to complain of sore arms immediately after the injection. Go figure.) Take those claims of paralysis, reported to VAERS as Guillain-Barré syndrome. CDC looked into them and determined, "Among 9 to 26 year-olds, the number of reports of GBS received by VAERS are within the range that could be expected to occur by chance alone after a vaccination." Or take the reports of Gardasil-induced miscarriages. Fox News duly noted, "Clinical trials show the vaccine's miscarriage rate is consistent with that of women in the general population who were given placebos." And so on, to the point that the chairman of the American Academy of Pediatrics' Committee on Infectious Diseases recently endorsed making Gardasil part of a standard cocktail of vaccinations, along with shots for meningitis, tetanus and diphtheria. "If we look at the number of doses given vs. the reports, it's very clear that there are significant benefits that far outweigh potential risks at this time," he told the Dallas Morning News.

So when a woman tells the story, in the comments section of a CBS News story about the Gardasil scare, of how her 14-year-old daughter had a seizure 16 days after getting immunized, and then demands, "Do healthy girls just start having seizures?", the correct answer is: Yes, some of them do, unfortunately.

There's a political component to all this. Long before it received FDA approval, Gardasil was the target of furious lobbying by the religious right, which opposes it on the grounds that making sex safer encourages promiscuity.

I'm not suggesting that the conservative bent of the New York Post influenced Edelman's reporting, although I have yet to hear back from her* on just why she chose to treat Lisa Ericzon's claim with so little skepticism. I'm also not suggesting that the FDA doesn't drop the ball occasionally and miss a drug risk, as it did with Vioxx. But when journalists are misrepresenting medical science and misleading readers about the relative risks of vaccination versus exposure to a disfiguring and deadly virus, something is very wrong.

*UPDATE, 3:32 p.m.: Edelman responds. For starters, she notes the paper ran a lengthy story on the issue with more context earlier this month.

"I am not worried that my reporting will scare people from getting a vaccine," she writes, "but as noted in my previous story, a top scientist from Dartmouth Medical School who helped conduct clinical trials on Gardasil has been sharply critical of the commercial push to give Gardasil to girls at age 11 or 12.... I did not write the headline, but it's an accurate one based on Jessica's mother's feelings. The millions of girls and young women getting Gardasil are indeed part of a vast public-health experiment. Gardasil was tested in only about 5,000 females in clinical trials.

"As far as the FDA considering the drug safe, that's what they always say in the beginning. I covered deaths, blood clots, strokes and other serious events in young women who took the Ortho Evra contraceptive patch, but it took the FDA about two years to warn publicly that the patch was twice as risky as the pill. In the beginning, FDA scientists and spokespeople insisted, backing up Johnson & Johnson, that the patch was no more dangerous than the pill.

"I was not influenced at all by any 'conservative bent' of the Post. Rather, I am proud that the Post would allow me to write a story that challenges the FDA and a new, barely tested and widely marketed pharmaceutical.

"I'm a parent who wants to know all I can about the risks of drugs or vaccines. I have a 13-year-old daughter -- and based on the adverse events reported with Gardasil and the CDC's ongoing review of these reports, I'm in no rush to give her Gardasil."

Fair enough. I still think the story -- and its predecessor, for that matter -- sounds the alarm more loudly than the evidence warrants. In fact, stories like Edelman's may make it more difficult to get at the truth, as rumors of Gardasil's danger spread and more and more parents whose children have been vaccinated report any mysterious health symptoms at all to VAERS. As that happens, you can be sure that papers like the Post will interpret the surge in reported symptoms as proof of its toxicity.