CDC Has Lost Control of the Autism Argument
David Kirby
Posted April 4, 2008 | 12:55 PM (EST)

On Wednesday, CNN's Larry King hosted Jenny McCarthy, myself, and several others to discuss the growing evidence of a link between childhood vaccines and autism. The CDC refused to send someone to appear on the show. Instead, on Thursday, the agency issued a statement meant to reassure the American public that all vaccines are safe for all kids.

But the CDC statement only served to show how out of touch the administration of George (really? gas costs 4 dollars?) Bush really is.

A recent government decision to award nine-year-old Hannah Poling taxpayer dollars for her multiple vaccine-induced autism, has left parents anxious and alarmed, especially when their own kid has a pending appointment to receive 5 or more vaccines in one sitting (Hannah had 9 at once).

So the CDC now issues a written statement meant to soothe jittery parents, by saying that "the recommended vaccine schedule is flexible." Such decisions, the friendly announcement said, "are best made in consultation with the child's doctor, and parents shouldn't be reluctant to have such discussions."

Of course parents shouldn't be reluctant to have this discussion, but they are. I get nasty emails from some pediatricians, and the number-one complaint I get from them is that, because of people like me, they must now "waste" (their word, not mine) precious billing hours talking to layperson parents about vaccine science.

These doctors' hostility is palpable, (and they hopefully represent a minority of pediatricians). And while I cannot imagine ever consulting anyone of such temperment for medical care in the first place, many parents are simply cowed into silence. For them, the CDC suggestion to ask vaccine safety questions at a well-baby visit is laughable, if not risible.

Interestingly, after years of being told that autism is purely genetic, and not some environmentally triggered epidemic, parents now learn that the CDC has begun a massive investigation, called the SEED study, to look at "genetic, environmental and hormonal factors, as well as selected mercury exposures," that cause autism

This is encouraging news, though one can assume that thimerosal is not among those mercury exposures that have been "selected" for study.

But the real problem here is the track record and credibility of the CDC to continute conducting any vaccine safety studies at all, a view that is shared by some on Capitol Hill.

This is not helped when the CDC continues to issue statements such as:

"Top scientists -- with the open-mindedness that characterizes good science -- planned and conducted the highest-quality, large-scale (vaccine safety) studies. No links to autism have been found."

Now, I have no doubt that CDC officials believe these truly were the "highest quality" studies available. But but many scientists, including some who authored the studies, disagree. They say many of the studies were flawed and/or inconclusive.


The flagship study was a four-year analysis by the CDC of a large US database called the Vaccine Safety Datalink (VSD). This study, published in the journal Pediatrics, was authored by Dr. Thomas Verstraeten, a visiting researcher from Belgium.

The final, published version of the study found no evidence of a link between thimerosal in vaccines and autism, though earlier analyses -- discovered through the Freedom of Information Act -- showed remarkable correlations.

At the time, the CDC called it one of the highest quality studies of its kind ever conducted.

But in 2006, Congress asked the National Institutes of Health to convene a special panel to investigate the quality and usefulness of the VSD database -- and by extension, the Verstraeten study itself -- as a means of investigating such a link.

The panel determined that there were "several serious problems" with the database and the study, including many "weaknesses" and "limitations" that could render certain analyses "uninformative and potentially misleading. "

The NIH Panel was "concerned" about how autism diagnoses were made and recorded by HMOs who take part in the database, and questioned if the HMOs had adequate services for autism families, who might seek care elsewhere. Panelists said these and other problems likely led to an "under-ascertainment" of autism cases in the HMOs.

The panel also cited many problems with the Verstraeten study design. It warned that a "large proportion, around 25%, of births were excluded from the analysis." Panelists wrote that these same children "may represent a susceptible population whose removal from the analysis might unintentionally reduce the ability to detect an effect of thimerosal."

Other "serious problems" were the facts that there was no consideration of pre-natal thimerosal exposures from immune globulin, or "other vaccinations given during pregnancy," (i.e., flu shot), and no accounting for, "the cumulative exposure to organic mercurials through diet or other environmental sources. "

The NIH panel determined that these problems, "reduce the usefulness" of the VSD to prove or disprove a link between thimerosal and autism.

Shortly thereafter, panel chair Dr. Irva Hertz-Picciotto, Professor of Public Health at U.C. Davis School of Medicine, told Dan Olmsted (formerly) of UPI that the VSD study "was not the last word... things need to be looked at again, perhaps with different methodology."

And Verstraeten himself said the study proved nothing. In a letter to Pediatrics, he wrote that, "We found no evidence against an association, as a negative study would. On the contrary, additional study is recommended, which is the conclusion to which a neutral study must come."


Two studies conducted in Denmark are always referred to by the CDC and others when trying to defend the injection of organic mercury into the systems of newborn babies and infants. These are among the best of the "highest quality" studies, we are told, that show no link between vaccines and autism.

The main Denmark study reported that the removal of mercury from vaccines was followed by a sharp increase in reported autism cases. But the authors admitted that much of this increase was possibly due to a major change in the way Denmark counted its autism cases during the study period (switching from inpatient diagnosed cases only, or about 13% of the total, to cases diagnosed in inpatient AND outpatient settings, or 100% of the total).

The CDC touts the high quality of the study, even though the authors cautioned in the study itself that "methodological limitations" -- such as the exponential expansion of patients (due to counting both inpatient AND outpatient cases) -- "may have spuriously increased the apparent number of autism cases."

Adding insult to understatement, Dr. Hertz-Piccotto said that, as bad as the VSD study was, the Denmark papers were even worse. "Some studies are stronger than others," she said. "The Verstraeten study was an improvement on other studies, including the two in Denmark, both of which had serious weaknesses in their designs."


In 2004, a committee of the Institute of Medicine reviewed the US, Denmark and a few other similar studies (whose methodology have been questioned in other venues including my book, "Evidence of Harm").

The IOM committee relied almost exclusively on large population studies (epidemiology), and virtually ignored the growing body of evidence emerging from clinical, animal and test tube studies from the fields of toxicology, immunology and other scientific disciplines.

(Interestingly, the US Federal Court system has determined that epidemiology alone is "insufficient" when trying to disprove a link between exposures and outcomes in an individual).

The committee concluded that the evidence did not support a vaccine link to autism. But it added that, "We cannot rule out, based on the epidemiology, the possibility that vaccines contribute to autism in some small subset," something that parents have been saying all along.

Finally, in 2005, another panel of the IOM criticized the CDC for a "lack of transparency" in its vaccine safety programs, particularly the VSD database. The IOM panel report noted that a CDC official had testified that some of the original datasets in the Verstraeten study, "had not been archived in a standard manner," and, "may not allow all the re-analyses that one might want to do, or in fact may not be available at all."

This same IOM committee, citing the Federal Information Quality Act, which makes it a felony to intentionally lose or destroy any publicly funded records, recommended that vaccine officials at the CDC's National Immunization Program "seek legal advice."

And despite all of this (and more) the CDC wants us to believe that these studies represent the "highest quality" analyses available.

The leadership of CDC Director Julie Gerberding has been replete with crushing morale depletion, embarrassing media coverage, congressional investigations into Katrina trailers, and a general sense of an agency in decline.

Sadly, the words "CDC" and "highest quality" can rarely be used in the same sentence anymore.