Questions Arise Over Merck Pharma's Gardasil Vaccine after 15 Women Die

The drug Gardasil--a vaccine from Merck & Co. Inc. manufactured to inoculate women against HPV and cervical cancer--is possibly killing women instead.

According to the U.S. Center for Disease Control (CDC) website, the CDC's Vaccine Adverse Event Reporting System (VAERS) "is a national program of CDC and the Food and Drug Administration (FDA) that monitors the safety of vaccines after they are licensed." This department has received 7,802 complaints about adverse effects resulting from Gardasil.

Of these complaints, the CDC website maintains that only 7% were complaints of serious adverse side effects, defined "by the Code of Federal Regulations as adverse events involving hospitalization, death, permanent disability, and life-threatening illness."

Oh, well, never mind, only 7% . . . but wait a minute, 7% of 7,802 is 546! 546 women who were maybe hospitalized, permanently disabled, or killed as a result of this drug! That seems like kind of a lot, though according to the CDC website, 12 million women received this vaccine between 2006 and 2007. 546 out of 12 million might be statistically a very small percentage, but it still seems to me like too many, so I want to know more.

The CDC's website offers that it received 15 reports of death and 31 reports of something called Guillain-Barré Syndrome (GBS), "a severe neurological disorder [that] causes increasing weakness in the legs and arms that can be severe and require hospitalization" (Citation). Okay, that's 46 reports that we know about now; so what were those other 500 out of 546 serious adverse side effects complaints about? Hm, no mention of those anywhere on the CDC's site.

Merck's website says, "Merck has analyzed the adverse events reported for GARDASIL relating to the recent reports of death and paralysis, and based on the data available to Merck, believes that no safety issue related to the vaccine has been identified." Well, are they looking at the same totally meaningless and spotty data as I am on this CDC website? If they have a better report, why can't I see it? Upon further digging, I find what the VAERS website calls "reporting data." Ah, excellent!

So I open up this little gem, and guess what? There is absolutely no statistical integrity in this document whatsoever. Look at it! It's a single-columned mess! (UPDATE: Someone below alerted me to the fact that the commas were separating the columns, but after separating the columns out, my faith is not any more boosted in the data; see comments below.) This particular spreadsheet is all vaccine reports--not just reports for Gardasil--from 2007. Hey math folk out there; how would you like to get this document from your boss and be told to make sense of it? Imagine your boss handing you this, and saying, "Give me a report on the serious adverse side effects mentioned in this spreadsheet about Gardasil." Haha!

. . . except not so much haha, because presumably, that's what Merck did--some employee of Merck's company was handed the "reporting data" document by his/her boss and told, "Analyze this, and tell me if you think there's a safety issue with Gardasil." Apparently, the employees at Merck are so magical that s/he was able to do that, no problem, and now there's no reason to do any follow-up investigating, and we're all safe, and we should just shut up about it. I feel better already!!

So going back to the CDC website, they tell us, "VAERS received 15 reports of death following Gardasil vaccination in the U.S. Only 10 reports contained the level of information adequate for further analysis. After careful review of those reports we could not establish the causal relationship between vaccination and death. For the remaining 5 reports of death, we were unable to obtain any patient identifying information; therefore we could not confirm death outcomes."

O.K., "10 reports contained the level of information adequate for further analysis" and "the remaining 5 reports of death [. . .] were unable to [furnish] any patient identifying information; therefore [you] could not confirm death outcomes." WHAT?? So, in other words, "Five women died, and we couldn't locate 'em afterwards, so screw 'em." "The level of information adequate for further analysis" is apparently, what, a name and address?? The inability to "establish a causal relationship" is based on what now? Presumably, there are reports on those inquiries; where are they? Who performed these follow-up investigations? Where is their jumbled mishmash of totally useless data?

This whole thing stinks to high heavens. Follow this link if you or someone you know has experienced harm as a result of Gardasil.


Lesberita said...

This just furthers my distrust of the pharmaceutical companies...

Anonymous said...

Um, that data is perfectly decipherable, its a CSV file - commas seperate fields, the first line is a column header, the rest are actual data.

Excel or any other spreadsheet should be able to open it just fine.

Seriously, most people that work with any kind of statistics, or deal with serious spreadsheet stuff for work, can handle this, it wouldn't be a 'nightmare' for anyone.

As for them determining that it was 'safe', there ARE an awful lot of 'patient complained of side-effects, but symptoms spontaneously passed after a couple of days' type entries. I imagine that the 500 'complaints' that the CDC received that weren't in that 46 that were deemed 'serious', were in fact just reports that the patients made, that their doctors HAD to report further up to CDC - they take this kind of thing pretty seriously, and can't just ignore a patient that says 'well, i felt kinda weird for a few days, but it went away'.

I'm certainly not 'trusting' of the pharma industry, but to see conspiracy here when you didn't really put much effort into looking at the data, is a bit sensationalist.

Ma'amselle Lezident said...

I did open it, and I did a search within it to narrow down the results for "Gardasil", but as you surmised, I had not processed the commas. After doing this, I'm still not terribly convinced. There is no standardized method of input, such that, e.g., I also found search results for "Gardsil," "Gardisil," and "Guardasil". There's also stuff like 248 "penicillin"s versus 122 "pcn"s in a variety of contexts spread throughout the document.

This spreadsheet lacks uniform input (garbage in); that can only result in skewed and/or meaningless statistical reporting (garbage out). These variables--the data within these columns--are descriptive (aside from things like gender and state), meaning they have little to no statistical relevancy. I also have to point out that this data wasn't even useful enough to track down 5 (1/3rd) of the 15 women who died, so I have serious doubts about the care administered by the recordkeeper(s) behind this document.

DoctorB-MD said...

As a physician, I can't let your paranoid, conspiracy-theory-like ravings go by without responding. The VAERS reporting system was established so that ANYONE could report ANYTHING that he or she thought might be related to a vaccination. Many are done by physicians but many are not. It is a passive system (by law), meaning that it is dependent on the reporter taking the initiative to do it. It is also completely dependent on the reporter to type things correctly and to provide enough information to evaluate the report (i.e. fill in all the blanks). The data file that you opened is the way that raw data is when the public is allowed to self-report. Nothing is consistent, and things left blank make it impossible to analyze or follow-up on many reports. It is left unchanged intentionally, so that it is provided to the pharmaceutical companies without any corrections, which may in themselves be done incorrectly. It is up to the analyzer to then do his or her best to make it uniform and consistent and try to make sense of it. That is what statisticians do for a living and the companies employ hoards of them. However, your criticism is misplaced. The problems with the data are the fault of those who reported it and took no interest in spelling things correctly and providing all requested information. Then they can try to figure out whether the deaths and other reported serious events are more than would be expected by chance. Interestingly, during the trials of Gardasil, 14 died in the group who received the vaccine and 10 in the group who received a placebo shot. Thus there were more deaths in those who received the vaccine, but many who received no vaccine died too, and the difference was not statistically significant (meaning that chance alone can explain it). Also, bear in mind that it isn't just the drug companies who analyze the data, but also the CDC and the FDA, and the CDC is an independent agency with no stake either way in the results it finds.

Ma'amselle Lezident said...

I think you've precisely caught on to what I was trying to say. People were like, "Um, my mom/girlfriend/sister/daughter died as a result of your vaccine," and Merck looked at this junky garbage file and said, "Oh, no, we've analyzed this [piece of ****] document and determined there's nothing to fear."

And the CDC is by no means an independent agency. It is one of the major operating components of the Department of Health and Human Services, a cabinet department of the U.S. govt.

Ma'amselle Lezident said...

Also, for comparison's sake, consider this:

Zostavax, another intravenous vaccine from Merck, hit the market in 2006 around the same time as Gardasil.

One thing you can cull from this 2008 garbage document is the incidences of occurrence. Of 10,164 complaints total in 2008, 436 contain Zostavax (4%); by contrast, there are 1,818 complaints involving Gardasil (17%, or nearly 1 in 7).

Considering these are both intravenously administered vaccines from the same company that hit the market at the same time, and given that the CDC's report claimed that a 7% incidence of serious-adverse-effects complaint is below the average, then wouldn't one expect to see a relatively more equal number of Zostavax complaints to Gardasil complaints?

DoctorB-MD said...

First of all, you are totally mischaracterizing what Merck said. They said there were 10 cases with enough information provided to analyze them and that they can say with confidence that those deaths do not appear to be related to the vaccine. Neither you nor I knows what the cause of death was in those people, but it is usually fairly easy to rule out certain deaths as being even remotely related to the vaccine.

As an example, for years when statin anti-cholesterol drugs came out, there was a statistically significant association between taking them and dying a violent death from things like murder and auto accidents. The association was there, but no one believed it could be related and subsequent larger numbers in studies showed it not to be real.

Merck also said there were 5 deaths they could not analyze because there was not enough information. That is undoubtedly because the person who reported them did not provide contact info or enough information to even do a follow-up. There is no other possible response to that except that they could not analyze them.

And Merck's conclusion was that they could not find any association based on the information available at this time. What else can they do but wait for more information? This is how the information comes on any new vaccine or drug, once initial testing is complete.

As to the CDC, what I meant was is that it has no stake in the approval or disapproval of drugs and vaccines and therefore no preset agenda. Its only purpose is to foster the public health, including preventing harm from vaccines.

As to your comparison of Zostavax and Gardasil, I could not find any numbers on the number of Zostavax doses administered to date, but since it is given to those over 60 who are notoriously underimmunized, I am willing to bet that ten times as much Gardasil has been administered. Thus based on your numbers, the fact that the reports on Gardasil are not 10 times what they are for Zostavax actually makes the latter the concern, not the former.

Ma'amselle Lezident said...

Merck didn't say any of those things. The case analysis all came from the CDC website; Merck's statement only mentioned reading the report.

I, too, could not find any numbers on how many people were given Zostavax, but I could find that Zostavax underwent 38,000 clinical trials versus Gardasil's 25,000. I'm not implying anything from that necessarily.

Something else I found was a 9-page heavily blacked-out warning letter from the FDA (on FDA's website) sent to Merck in April 2008 expressing "a number of significant objectionable conditions relating to [Merck's] compliance with the CGMP [current good manufacturing practice]." Surprisingly, this came up under a Zostavax search. There is this little tidbit, though:

"You failed to exercise appropriate controls over computer or related systems to assure that changes in master production are instituted and input and output from the computer or related system of formulas are checked for accuracy and maintained [21 CFG 211.68(b)], in that there is no documentation to support software manufacturing change performed to the [blacked out] code used in the manufacture of Gardasil(R), lots [blacked out] and [blacked out]."

"You failed to establish test procedures or other laboratory control mechanisms designed to assure that drug products conform to appropriate standards of identity, strength, quality, and purity and to assure that any deviation from the written test
procedures or laboratory control mechanisms shall be recorded and justified (21 CFR 211.160(a) and (b)]."

DoctorB-MD said...

Here is an example of one report. It is easy to see why no further follow-up or analysis would be possible based on the report.

"Information has been received from a consumer concerning a daughter of a friend who on an unspecified date was vaccinated with Gardasil (Lot# not provided). The consumer reported that her """"friends daughter had the Gardasil vaccine and had a near death experience"""". It was not reported if the patient sought medical attention for the event. No further information is available."""

Maybe whoever reported this thought he or she was doing a service, but it was just wasted time and effort. Absolutely nothing could be drawn from it or any follow-up pursued.

Liora Pearlman said...

You went to the wrong place to see the VAERS data. Go to http://www.medalerts.org and get familiar with the abbreviations. Then you can do a great search and it comes up in report format.

Ma'amselle Lezident said...

Liora Pearlman, you rock! That was a great resource. In the spirit of full disclosure, I did a search of all the "died? yes" entries. Here's what I found:

1. 12-year-old, died of heart failure, 6 days later

2. 19-year-old, no previous illness, died of blood clots 14 days later, autopsy performed by wrong kind of doctor (uncovered by deputy in investigation)

3. age unknown, reported by company representative, patient died of blood clot 3 hours after getting vaccine

4. 15-year-old, influenza and vaccine (administered twice?), unclear when died

5. age unknown, reported by nurse practioner via friend, patient died of blood clot after two weeks

6. 11-year-old, died of cardiac arrest, reported anaphylactic response to vaccine, 3 days later

7. ages unknown, doctor reported he heard at a conference that two had died, information being sought

8. 22-year-old, no history, also given hormonal contraception same day as vaccine, died suddenly 2 days later, causes undetermined

9. 17-year-old, diabetic anemic smoker (yikes), died 50 days later

10. 18-year-old, received vaccine as well as vaccine against meningitis same day, died 149 days later of meningitis

11. 12-year-old, no medical history, died 21 days later in her sleep

12. 26-year-old, smoker with asthma, date of death unknown, died of blood clots

13. 17-year-old, also on Yasmin birth control but with no prior medical history, died 2 days later of sudden heart failure

14. 12-year-old, arrhythmia and seizure history, died of heart failure 56 days later

15. 20-year-old, autopsy turned up no foul play, toxicology report pending, died 4 days later