Autism, Vaccines, and the Problem of Selective Evidence
By Whole Health Web Admin
We all tend to pay attention to information sources that reinforce our existing beliefs. Conservatives watch Fox News, liberals read The New York Times. The same pattern appears in discussions surrounding autism and vaccines.
I have often heard people say, “They did the study—the Amish don’t get autism.” That kind of statement reflects a broader intellectual laziness that shows up in politics, education, and the media. On the other side of the debate, people who strongly trust the medical establishment often state with equal certainty that there is no connection between vaccines and autism. They feel just as strongly—and often with just as little direct evidence—as the first group.
When I looked for the often-cited “Amish autism study,” I couldn’t find one. What I did find was a series of investigative articles by Dan Olmsted, a journalist who explored autism prevalence in communities with lower vaccination rates. Olmsted reasoned that if vaccines were contributing to autism, one way to explore the question would be to examine relatively under-vaccinated populations.
In his reporting on Amish communities, Olmsted concluded that while the Amish do vaccinate, they tend to vaccinate less than the general population—and, according to his observations, appeared to have lower autism rates. These conclusions were observational and not the result of a formal epidemiologic study.
In one article, Olmsted described children treated by Homefirst Health Services in Chicago. Many of the families served by the clinic homeschool and tend not to vaccinate. Dr. Mayer Eisenstein, a physician at the clinic, was quoted as saying:
“We have a fairly large practice. We have about 30,000 to 35,000 children that we’ve taken care of over the years, and I don’t think we have a single case of autism in children delivered by us who never received vaccines.”
Dr. Eisenstein also acknowledged that these observations do not rise to the level of a controlled study, noting that it is possible—though unlikely—that families with autistic children may have moved or sought care elsewhere.
Olmsted’s work has been widely criticized. Some critics argue that he did not look thoroughly enough for autism diagnoses within Amish communities, particularly in Lancaster County, Pennsylvania. Dr. Kevin Strauss, a pediatrician at the Clinic for Special Children, has stated:
“The idea that the Amish do not vaccinate their children is untrue. We run a weekly vaccination clinic, and it’s very busy.”
Dr. Strauss also notes that while the Amish do vaccinate less than the general population, autism is not a single diagnosis but a description of behavioral patterns that may accompany conditions such as seizure disorders, intellectual disability, Fragile X syndrome, or Rett syndrome—conditions that are seen in Amish populations.
Critics of Olmsted point out that his work does not prove a causal link between vaccines and autism. At the same time, it is also true that existing research has not definitively proven the absence of any relationship. Much of the research cited to dismiss concerns focuses on short-term mercury measurements in blood, urine, or stool, or on comparisons between countries with different vaccine formulations—approaches that may not fully address long-term or developmental effects.
All major medical journals accept advertising from pharmaceutical companies. Whether financial influence affects research priorities, interpretation, or publication bias is a question worth asking in any scientific field.
Autism was extremely rare prior to the mid-20th century. In 1986, the U.S. Congress passed the National Childhood Vaccine Injury Act, which substantially limited the ability of individuals to sue vaccine manufacturers for injury. Following this legislation, the number of vaccines recommended for children increased dramatically. During the same period, reported cases of autism rose sharply. This temporal overlap does not, by itself, establish causation, but it is a central part of the controversy and cannot be ignored. From a scientific standpoint, it would be reasonable to examine long-term health outcomes in populations exposed to differing vaccine schedules, while also accounting for changes in environmental exposures, diagnostic criteria, and reporting practices.
Until such comparisons are performed in a rigorous and transparent way, the discussion remains unresolved. Asking difficult questions does not require adopting predetermined conclusions—but it does require openness to inquiry from all sides.
Editor’s Note:
This article is preserved as an archival discussion reflecting questions and debates that were actively explored in earlier decades. Whole Health Web does not present this article as definitive evidence or medical guidance. Vaccine policy and autism research continue to evolve, and readers are encouraged to consult current, peer-reviewed research and qualified healthcare professionals.
