An economist's linking of weather, TV viewing, and autism sparks a firestorm
An economist's linking of weather, TV viewing, and autism sparks a firestorm
The mother of a severely autistic teenager blogged in the Huffington Post that she wished he would "drown professionally, under a pile of angry e-mails." A writer for the Adventures in Autism website called his research "indefensible lunacy and junk science." When Cornell economist Michael Waldman released a study in 2006 showing that television viewing in early childhood may be a trigger for autism, he was besieged with irate messages and postings from parents and doctors. After years of research showing that nothing in the family environment could cause autism, Waldman was blaming parents—though indirectly—for planting their children in front of the TV.
For Waldman, an expert in applied microeconomics and director of Cornell's Institute for the Advancement of Economics, the reaction was not entirely unexpected. As the father of a son who had suffered from an autism spectrum disorder, he knew the pain of the diagnosis and the struggle to find an effective treatment. "I'd been reading what both the medical community and the parent groups were saying," he says. "There's a basic belief that they have strong evidence that television or something in the family environment can't be the trigger—if there is a trigger. The reality is that there is no strong evidence. Very few studies have looked at it."
The diagnosis for his son, then nearly three, came just after his baby sister was born, when he started watching more television. Though none of his doctors mentioned it as a possible cause, Waldman sharply reduced the boy's screen time, among other interventions, because he had read about the negative effects it can have on child development. Within a year, his son had improved and eventually recovered fully, leading Waldman to wonder whether television had induced the disorder. Finding no studies linking TV and autism, he began contacting medical researchers. When none showed any interest in investigating the relationship, he decided to delve into the issue himself, assembling a team that included Cornell health economist Sean Nicholson and Nodir Adilov, PhD '05, one of Waldman's former graduate students, now an economist at Indiana University-Purdue University.
Autism diagnoses have skyrocketed over the past thirty years, rising from a rate of one in 2,500 children to one in 150, according to the most recent estimate from the Centers for Disease Control and Prevention. While research has focused primarily on genetic causes and abnormalities in brain structure, Waldman wanted to explore an environmental factor—television—by studying how children spend their time indoors. Using the Bureau of Labor Statistics' American Time Use Survey, he and his colleagues concluded that the amount of TV young children watch increases when it rains or snows. They then analyzed county-level autism rates in four states: Oregon and Washington, where the Cascade Mountains divide each state into wet and dry halves, and California and Pennsylvania, where they wanted to study the effects of cable subscription rates on autism.
The researchers found that in the western counties of Oregon and Washington, which have four times as much precipitation as those in the east, the autism rate was 2.2 times higher. And in California and Pennsylvania, they attributed 17 percent of the increase in the autism rate for children born between 1972 and 1989 to the growth of cable subscriptions and the subsequent rise in television viewing. They concluded that although the results don't definitely prove that early childhood viewing is a trigger, "we believe our results provide sufficient support for the possibility that until further research can be conducted, it might be prudent to act as if it were." Their findings, presented at a National Bureau of Economic Research conference in 2006, sparked charges that economists were not qualified to study autism. "It's pure speculation," said Mikhail Mirer, a pediatric neurologist who has a caseload of 100 autistic children at the Good Samaritan Hospital Medical Center on Long Island. "I could say that selling lollipops at gas stations is a cause."
Unfazed, Waldman and his colleagues published their research last November in the Archives of Pediatrics & Adolescent Medicine. Developed from their earlier work, the article reported that because autism rates were higher in the counties that had more precipitation, one of several environmental factors associated with indoor activities could be triggering the disorder in genetically vulnerable children. Among the possibilities: television, vitamin D deficiency, household chemicals, air pollution, or chemicals in the precipitation itself. The researchers backed off their emphasis on television as the primary trigger, they say, because it was easier to have the results published if they were less controversial. "Medical journals tend to be cautious, I think in a good way," Nicholson explains. "We can with more confidence say there is a relationship between precipitation and autism than we can say there is a relationship between TV and autism."
In a commentary published in the same issue, University of Washington epidemiologist Noel Weiss wrote that other factors may explain the higher incidences of autism. For example, the western, rainier areas of Washington and Oregon are more urbanized and wealthier than the arid, rural eastern counties—raising the possibility that the increase could be tied to socioeconomic status. Weiss also questioned whether the study would cause panic when the conclusions remain tentative. "What if someone is going to decide to move because they are worried that their child is going to become autistic?" Weiss asks. "Or not move, but just worry about it?"
In response, Waldman says he and his colleagues made it clear that the study was preliminary and should not be taken as a directive to parents. Their research, he adds, used controls that eliminated factors other than precipitation. For example, the study compared how autism rates rose and fell as precipitation varied over time within each county, and again it found that children exposed to more rainfall in their first three years had higher rates of autism. "The way we view the paper is: 'Here is a clue.' Maybe it will turn out to be important. Maybe it won't," he says. "But the relationship is certainly strong."
— Sherrie Negrea