The conservative legal group Judicial Watch last week issued a news release announcing that it had uncovered three deaths linked to the human papilloma virus vaccine (HPV), which has been the subject of furious debate around the country as states decide whether to require 6th grade girls to be vaccinated before admitting them to school. The release, in which Judicial Watch claimed that its perusal of adverse event reports from the vaccine also turned up cases of autoimmune disease and fetal damage in vaccinated pregnant women, produced a minor media splash, with articles in The Wall Street Journal and The Washington Times. The story didn't get play in most other outlets, probably for this simple reason: It's bullshit.
Like the social conservatives whose ideological affinity presumably drew Judicial Watch into the issue, I believe that making HPV vaccination mandatory is a mistake, at least at this time. Though it looks like a good vaccine, the HPV shot is very expensive and needs to be used more widely before we can be confident that it's truly safe and an effective way to prevent cervical cancer. In a strange way, this episode illustrates one of the reasons that the time is not ripe for compulsory use. But Judicial Watch got it completely wrong in its analysis of 1,637 consecutive reports from the Vaccine Adverse Events Reporting System (VAERS).
VAERS, a database of post-vaccination medical events that is analyzed by doctors at the FDA and the CDC, was set up in the late 1980s to serve as an early warning system for vaccine safety issues. It has turned up a few problems, most notably the 1998 association of intussusception, a painful and occasionally fatal infantile bowel problem, with Rotashield vaccine, which was promptly withdrawn by Wyeth.
The VAERS database is dominated by common, low-significance events like transient blackouts, vomiting, rashes and the like, which accompany the use of almost any vaccine or drug. Gardasil, Merck's HPV vaccine, is produced in yeast, which produces a fair number of allergic reactions.
There are also more serious events that may or may not have anything to do with the vaccine. The CDC/FDA analysts are looking for unusual patterns, especially those that make biological sense. For example, Rotashield was a live oral vaccine against a GI-tract infection, rotavirus. It made sense that it might cause intusussception, so the government scientists were on the lookout. But anyone can report anything to VAERS -- a patient, a doctor, a drug company. And they do.
For example, the guardians of an 18-year-old girl reported to VAERS that she had suffered a psychotic episode a week after getting Gardasil. Among other symptoms, the girl went around telling people that the vaccine was "poison coursing through my body." This doesn't sound like an actual vaccine reaction, or at least not one you can blame on the vaccine.
More to the point, the three deaths reported to VAERS after HPV were heart attacks that followed administration of the vaccine by three hours, a week and two weeks, respectively. It's a good thing that they were reported to VAERS. We need to know if people are dying after vaccination. But when people die following a vaccination, it isn't necessarily the vaccination that caused it, a fact that generations of vaccine critics have failed to clearly see as they bungle their interpretations of VAERS data.
When the CDC, responding to the three death reports, interviewed doctors and examined the charts, they found that one, a case of myocarditis, was caused by influenza virus. The other two were blood clots in adolescents taking birth control pills. Birth control pills are known to raise the risk of blood clots. "These deaths are tragic but appear to have causes unrelated to vaccination," says Dr. John Iskander, co-director of the Immunization Safety Office at the CDC.
Judicial Watch also claimed to have found 18 cases of birth defects and miscarriages among the reports. I counted 11 pregnancy-related adverse events, including six miscarriages that occurred between a few hours and four months after the shot, two unrelated abortions, two abortions following abnormal ultrasounds, and another report of an abnormal ultrasound with no indication of what was done about it (a lot of the VAERS reports are quite sketchy and confusing).
Eleven problem pregnancies is certainly fewer than would expect in any group of 3 million teenagers and women (30 percent of first-trimester pregnancies miscarry, and 3 percent of babies have at least minor birth defects). And since even live vaccines like varicella are known not to cause birth defects, Iskander points out, it seems unlikely that an inactivated subunit vaccine like HPV would do so.
For my money, the potentially most serious data point in the VAERS reports is the 10 or 15 instances of Guillain Barre Syndrome-like muscle paralysis or weakness. There have been some credible, if rare, reports linking GBS and vaccines against flu, tetanus and meningitis, and the latter is often administered to teen-agers concurrently with Gardasil. On the other hand, there may be no connection to the vaccines. About one or two in 100,000 adolescents every year experience Guillain Barre Syndrome -- an autoimmune disorder -- so if 3 million vaccinations have been given, you'd expect this many cases to occur by chance.
One other thing that's notable as you go through the reports (I went through all 1,637 of them) is the number of women who fainted. Apparently, fainting isn't at all unusual with vaccination or doctors' visits in general. This is especially true when people line up for shots at school or in the military. One person faints and then it spreads. This is more dangerous than it sounds, though -- unattended fainters can suffer brain damage when they hit their heads on hard surfaces. For this reason pediatricians are supposed to keep teen-aged and adult patients in the office for 15 minutes before they walk or get into their cars. The HPV database includes a handful of faints that led to head injury.