The CDC's Thomas Verstraeten found an 11.35-fold increased autism risk from thimerosal exposure in his initial 1999 analysis, but through five successive iterations of data manipulation between 1999-2003, this alarming finding was systematically reduced until the published study showed "no consistent significant associations." Internal CDC documents reveal officials were "stunned" by the initial findings and held a secret meeting at Simpsonwood to discuss managing the implications before Verstraeten left CDC to work for vaccine manufacturer GlaxoSmithKline.
The first unmanipulated analysis of the Vaccine Safety Datalink in November-December 1999, later termed "Generation Zero" by critics, revealed startling associations between thimerosal exposure and neurodevelopmental disorders. According to data obtained through Freedom of Information Act requests by the advocacy group SafeMinds, Verstraeten's initial analysis found relative risks of 7.62 to 11.35 for autism in children receiving the highest thimerosal exposures in their first month of life compared to those with no exposure. The analysis also showed an 8.0-fold increased risk for ADHD, a 2.1-fold risk for speech delays, and 5.0-fold risk for sleep disorders.
These findings prompted Verstraeten to write an internal CDC email on December 17, 1999, with the subject line "It just won't go away," referring to the persistent statistical associations between thimerosal and autism. In the email, obtained through FOIA requests, Verstraeten noted that "all the harm is done in the first month" of exposure. This Generation Zero analysis was never published or presented publicly, and its existence only became known through FOIA disclosures years later.
On June 7-8, 2000, the CDC convened a secret meeting at the Simpsonwood Methodist retreat center in Georgia to discuss Verstraeten's findings. The meeting brought together 52 attendees including CDC and FDA officials, WHO representatives, and vaccine manufacturers including GlaxoSmithKline, Merck, Wyeth, and Aventis Pasteur. All data was marked "embargoed" with strict instructions against photocopying or removing documents.
The 259-page transcript, later obtained through FOIA by SafeMinds, reveals participants' alarm at the findings. Verstraeten himself stated he was "actually stunned by what I saw" regarding the statistical associations. Dr. Robert Chen from the CDC noted they had been "privileged so far that given the sensitivity of information, we have been able to manage to keep it out of, let's say, less responsible hands."
Perhaps most tellingly, participants expressed personal concerns about vaccines for their own families. One attendee stated he didn't want his grandson receiving thimerosal-containing vaccines, while discussions focused heavily on managing legal liability and public perception rather than addressing the potential public health crisis. Dr. John Clements from WHO went so far as to say the study "should not have been done at all" and warned the results "will be taken by others and will be used in ways beyond the control of this group."
Between 1999 and 2003, the Verstraeten analysis underwent five distinct iterations that systematically reduced the statistical associations:
Generation Zero (November-December 1999): The original analysis with broad inclusion criteria and true zero-exposure controls showing 11.35x autism risk.
Generation 1 (February-June 2000): Presented at Simpsonwood, this version added exclusions and protocol restrictions, reducing relative risks by over 75% but still showing significant associations.
Generations 2-4 (2000-2003): Through progressive modifications, these iterations eliminated most significant associations by:
Final Published Version (November 2003): Published in Pediatrics, this version concluded there were "no consistent significant associations between thimerosal-containing vaccines and neurodevelopmental outcomes."
The investigation revealed specific techniques used to weaken the associations across iterations. After seeing the disturbing Generation Zero results, CDC researchers changed entrance criteria to require all study children to have received "at least two polio vaccines in first year of life." SafeMinds compared this to "trying to study lung cancer in two-pack vs. three-pack smokers without including any non-smokers."
The researchers also manipulated age cutoffs, with internal CDC emails asking "What happens if you do this?" regarding including only children 18 months or older. Since most autism diagnoses occur in the 2nd-3rd year of life, excluding younger children systematically biased the results toward the null hypothesis. The final study stratified data across three different HMO populations with conflicting results, diluting any clear signal.
In July 2001, while the study was still being analyzed and two years before publication, Thomas Verstraeten left the CDC to join GlaxoSmithKline in Belgium as a vaccine safety specialist. He continued working on the paper as an "adviser" despite his new employer being a major vaccine manufacturer. This revolving door pattern was common - Dr. Julie Gerberding, CDC Director during this period, later became president of Merck's vaccine division.
The broader financial context reveals extensive entanglements between public health agencies and industry. The CDC Foundation has received millions from vaccine manufacturers since 2000, while the CDC itself owns 56 vaccine patents and distributes $4.6 billion annually in vaccines. HHS employees can collect up to $150,000 annually in royalties for vaccine-related products.
A month after the Simpsonwood meeting, Verstraeten sent another revealing email to toxicologist Dr. Philippe Grandjean expressing frustration with the focus on damage control rather than science. He wrote: "I do not wish to be the advocate of the anti-vaccine lobby and sound like being convinced that thimerosal is or was harmful, but at least I feel we should use sound scientific argumentation and not let our standards be dictated by our desire to disprove an unpleasant theory."
Additional CDC whistleblower Dr. William Thompson came forward in 2014 with documents showing the agency had systematically buried evidence linking vaccines to autism in other studies. Thompson stated that "senior people just do completely unethical, vile things and no one holds them accountable" and described meetings where researchers destroyed documents in "a big garbage can."
Today, the Verstraeten study is frequently cited as evidence that vaccines don't cause autism, with the published conclusion of "no consistent significant associations" taken at face value. The medical consensus rejects any thimerosal-autism link, and thimerosal was removed from most childhood vaccines by 2001 as a precautionary measure.
However, the documented history of this study's evolution raises profound questions about research integrity when findings threaten powerful interests. The systematic weakening of associations through methodological changes, the secret meetings focused on liability rather than public health, and the lead researcher's move to industry during the study all suggest that the final published results may not reflect the true initial findings.
While determining whether these changes represented appropriate scientific refinement versus data manipulation would require access to the complete datasets - which remain largely unavailable to independent researchers - the pattern documented through FOIA releases presents a troubling picture of how vaccine safety research was conducted during this critical period. The original Generation Zero finding of an 11.35-fold increased autism risk, if accurate, would have represented one of the most significant public health discoveries of our time. Instead, it was buried through a process that prioritized protecting the vaccine program over transparently investigating potential risks.