Thimerosal: No Smoking Gun
CDC: No Evidence of Child Brain Damage Due to Vaccine Preservative
WebMD News Archive
Sept. 26, 2007 - Thimerosal in vaccines has not affected children's brains
or behaviors, the CDC says.
Thimerosal is a mercury-based preservative once common in vaccines. It's now
found only in small amounts in some flu vaccines. However, thimerosal was once
used in most vaccines given to children.
Although the amounts were tiny, they added up. Because symptoms of autism
often appear suddenly around age 1 -- when kids have received a number of
vaccinations -- many parents became convinced that
thimerosal caused their children's autism. But the
Institute of Medicine has twice rejected this idea.
The new CDC study carefully avoided the still-controversial issue of whether
thimerosal is linked to autism. A separate CDC study of this issue is under
way, with a report expected in a year's time.
"In this study there is nothing you can draw on regarding any
relationship to autism," researcher William W. Thompson, PhD, of the CDC's
National Immunization Program, said at a news conference.
Instead, the study looked for 42 different neurological and psychological
problems in more than 1,000 7- to 10-year-old children exposed to various
amounts of thimerosal while still in the womb, from birth to 28 days, or up to
age 7 months.
"This was a large, very careful study to determine whether thimerosal
exposures in early childhood was associated with bad neuropsychological
outcomes," Anne Schuchat, MD, director of the CDC's National Center for
Immunization and Respiratory Diseases, said at the news conference. "It
came out with very reassuring results."
"Ninety-eight percent of the answers here are very reassuring," John
Iskander, MD, MPH, acting director of the CDC's Immunization Safety Office,
said at the news conference.
Thimerosal and Tics
What about the other 2% of the study findings? The study looked at 42
different outcomes, ranging from intelligence to fine motor coordination. The
researchers performed 378 statistical tests on the data. Each statistical test
had 5% odds of being a chance finding.
"By chance alone we would estimate about 5% of test results would be
significant," Schuchat said. "And in fact that is exactly what we
found: 19, or 5%, of the statistical tests did find significant results. They
were pretty evenly split between [linking] better and worse outcomes [to