THE BLOG
04/26/2007 09:11 pm ET Updated May 25, 2011

Don't Believe Everything You Read

A few months out of college I started working for UPI in Mexico City. It was a great place to start life as a journalist--fun, adventurous colleagues, all of us young, and we could write whatever we wanted. I mean really write what we wanted. If you had two sources to confirm your story, out it went on the telex. Once, after interviewing some senior Mexico City police officers, I reported that they had captured Carlos the Jackal. They were lying, of course (this was in 1981). But the story still ran, until the correction went out about 10 hours later. I wasn't fired.

I have the feeling that the rules haven't changed that much at UPI. Take a look at some of the reporting of Dan Olmsted, a reporter in Washington who has become a popular voice in the vaccines-cause-autism community of belief. Olmsted seems convinced that we are in the middle of an autism epidemic that began in the 1930s when Baltimore psychiatrist Leo Kanner noticed the first patients (in fact there are many, many case reports of autistic behavior that go back well into the 19th century and beyond). In a series called "Age of Autism," Olmsted has reported, among other interesting bits of information, that the Amish have no autism (actually, they do) possibly because they don't vaccinate (actually, they do). Recently he has turned up the fact that a few of Kanner's patients had dads who worked with dangerous chemicals, including ethyl mercury--the active ingredient of thimerosal, a preservative that was in several pediatric vaccines until a few years back.

In his latest column, which builds to a crescendo like a good horror movie soundtrack, Olmsted reports that the "big bang" of autism started in and around Beltsville, Md., a suburb of Washington where the U.S. Agriculture Department had a research center (anxious cello dissonance while the camera pans over an ominous red brick complex). One of Kanner's kids was the son of a senior researcher who experimented with fungicides, including ethyl mercury. Another was a chemist at the U.S. patent office. A third lived in Greenbelt--very close to Beltsville (also close, for that matter, to the woods where they filmed The Blair Witch Project).

In case this isn't convincing enough evidence that these men and their kids were the first "victims" of (mercury-induced) autism, Olmsted introduces a fourth adult autistic whose dad who was an engineer in White Oak, Md., "just a hop and a skip across I-95 from the Beltsville agriculture center" (pounding kettle drums). Finally, he writes that while talking to a mom on the West Coast she had a "shock of recognition" when she realized that she (though not her autistic son) once lived on a farm in Burtonsville, Md., not far--you got it--from the evil Beltsville. (violins screech while Janet Leigh gets stabbed in the shower).

"This suggests," Olmsted concludes, "a new and deeply disturbing truth about the Age of Autism: our fate is not in our genes, Dear Brutus, but in the chemicals that increasingly pollute our world and our children."

This is all lovely stuff. Like David Kirby's Evidence of Harm, which won an Investigative Reporters and Editors award for its gripping (and misleading) portrayal of a government and drug company conspiracy to hide the poisoning of kids, "Age of Autism" has real screenplay potential.

Do I have to spell out how ludicrous it is for a wire service reporter-playing-epidemiologist to expect us to believe any of this? Olmsted has zero actual information about these families' exposures to ethyl mercury. Not to mention that these children were only a few of the millions of Americans living in a country that was one big toxicological experiment, with tons of contamination billowing out of coal and steel plants, benzene in the water, lead in the paint and the gasoline, and mercury spooned out liberally in teething powders and slathered on wounds. Dan Olmsted seems like a nice guy, and he writes well. He's a dogged reporter, too, but in this case at least, more in the sense of a dog chasing its tail than a dog that hunts.

If you want to read a more reasonable hypothesis on the autism "epidemic," try this blog by Richard Roy Grinker, an anthropologist at The George Washington University, and Kristina Chew, a professor of classics at St. Peter's College. Like many of those involved in this controversy, both have children with autism.

Noting that the CDC's February estimate of 1 in 150 children has led many autism advocates to claim there is an "epidemic" of the disease (SafeMinds' Mark Blaxill ridicules the idea of 1 in 150 adult autistics, calling them the "hidden horde"), Grinker and Chew note that there are indeed autistics living among us:

"Some live at home with their aging parents or siblings. Some live in group homes, or in institutions. Some have jobs and live independently. Many have the diagnoses given to them when they were children, such as mental retardation, seizure disorder, or schizophrenia. Recently, one of us met a severely autistic 60 year old woman in eastern Tennessee, who we'll call Donna. Donna's internist diagnosed her with autism ten years ago, when she was 50. Her mother said that Donna's first label, in 1950, was "mentally retarded with emotional block and obsessive compulsive traits." Today, for the purposes of public assistance, she is classified as mentally retarded."


"Now, imagine another, more mildly autistic adult, who has a job with minimal social demands (filing medical records, perhaps), has poor eye contact, and some anxiety. Perhaps he is even married. After all, as they get older, many people with autism, like the well-known writer, Temple Grandin, make significant advances. First, he may not seek treatment or even think he has problems. Second, even if he did, he may not go to a psychiatrist, but rather to an internist who could treat his anxiety. Third, even if he did go to a psychiatrist, it is not likely the psychiatrist would diagnose an adult with autism, especially if the psychiatrist had no clinical data on his early childhood (autism is still a developmental disorder diagnosed in childhood, and the tools for diagnosing autism in adults are not nearly as valid or reliable as those for children).

"So, unless such mildly autistic adults had been diagnosed as children, and unless there were good baseline data on these people as children, we would have a hard time going back in time and trying to predict whether they would have qualified for an autism diagnosis. Even in the old days when we had rates for autism of 5 in 10,000 children, epidemiologists would have been hard pressed to find the corresponding adults, unless they knocked on people's doors looking for autism. And even then, most of these people would not consider themselves to be 'autistic.'

"As an analogy, consider Fetal Alcohol Syndrome (FAS), which occurs in 1 in 500 children. Because FAS only became a diagnosis after the mid-1970s, there are virtually no adults with this diagnosis over the age of 30. (And yet no one would suggest that pregnant women only started to drink alcohol in the 1970s). To locate an adult with FAS you would have to have evidence of fetal alcohol exposure, plus childhood onset of the symptoms of FAS."

They conclude that "the high rate of autism among America's children is not evidence of an epidemic. Rather it is evidence of how far we have come in understanding autism. After all these years, scientists are finally getting it right."

While I'm not convinced there has been no increase of autism in recent decades, and find it possible that some cases of autism have an environmental contribution, I find Grinker and Chew's hypothesis to be the natural starting point for debate.