News of the WeekEpidemiology

Vaccine-Autism Link Dealt Blow

See allHide authors and affiliations

Science  12 Sep 2003:
Vol. 301, Issue 5639, pp. 1454-1455
DOI: 10.1126/science.301.5639.1454a

This article has a correction. Please see:

Two new studies cast further doubt on the theory that a mercury-based preservative in vaccines causes autism. Called thimerosal, the preservative has already been phased out in many industrialized countries but is still used in the developing world. The new findings “provide additional, extremely reassuring data,” says William Schaffner of Vanderbilt University School of Medicine in Nashville, Tennessee.

Thimerosal first attracted attention in the United States in 1999, when the Food and Drug Administration realized that toddlers, who are typically injected with several vaccines simultaneously, might be receiving higher doses of mercury than allowed by one federal standard. As a precaution, vaccinemakers began to phase out thimerosal that year.

Not long after, parent advocate groups from Safe Minds proposed that mercury, from the preservative and other sources, might be a factor in the rising incidence of autism, which often appears at about the same time that 2-year-olds get a round of shots. Many scientists were skeptical, given the minute amount of mercury and the different symptoms of mercury poisoning and autism. But in 2001 an Institute of Medicine panel concluded that there wasn't enough evidence to rule out (or accept) the link.

Now the first big epidemiological studies weigh in. One comes from Denmark, which eliminated thimerosal from childhood vaccines in 1992. A team led by Kreesten Madsen of the Danish Epidemiology Science Centre in Aarhus reasoned that if thimerosal were a major cause of autism, incidence of new cases should drop once it was removed. In the September issue of the journal Pediatrics, they report that, instead of declining, the incidence continued to skyrocket after 1992. Like many epidemiologists, Madsen says the rising incidence could be a result of increased awareness and broader definitions of the disease. In any case, Madsen says, because incidence didn't even slacken, thimerosal is not a major cause of autism.


New cases of autism continued to rise in Denmark after a vaccine preservative was removed.


But Mark Blaxill of Safe Minds argues that the study is “distorted and misleading.” He notes that in 1995, the Danish health registry began tracking a new category of patient, called autism outpatients. This and other factors, he says, are artifacts that confound the interpretation. Madsen responds that an unpublished analysis without outpatients showed the same increasing trend. A similar pattern emerges from health statistics from Sweden, where total mercury in childhood vaccines began to decline in the late 1980s, as reported in the August issue of the American Journal of Preventive Medicine.

Epidemiologist Craig Newschaffer of Johns Hopkins University in Baltimore says that these ecological studies have inherent limitations, because they look at populations rather than studying individual diagnoses and exposures. But more in-depth studies are expensive, he adds, and with scientific skepticism about a possible link between thimerosal and autism mounting, may be difficult to fund. Although the current studies are unlikely to end the controversy in the United States, where many lawsuits have been filed, the new findings are reassuring to the World Health Organization, which continues to recommend the use of small amounts of thimerosal to keep down the costs of essential vaccines.


Navigate This Article