Tuesday, October 17, 2006

This Just In: The News Moves Fast. Autism Research Doesn't.

Monitoring news coverage of autism research is like watching one of those live police car chases on California highways that TV stations used to feature regularly: it is an exercise full of information but lacking in insight. We can see that the cops are following a motorist. But why, exactly? Why won't the motorist pull over? What does it mean? And why is it on TV?

TV, it turns out, is at the center of the latest buzz in autism research, featured yesterday in the online magazine Slate. (See the article, "TV Might Really Cause Autism," by clicking here.) Gregg Easterbrook, a visiting fellow at The Brookings Institution and a journalist who pro football fans know as the Tuesday Morning Quarterback columnist on ESPN, came up with this hypothesis: it was in 1980 that the United States started to see a rise in autism cases, according to government reports, and it so happens that was the same year when children's television programs -- on cable channels, and on videos -- became omnipresent forces in very young children's lives. Could the two be related?

Lo and behold, economist Michael Waldman and his research team at Cornell University's Johnson School of Business, this week published a paper, "Does Television Cause Autism?" on the school's website. (Find it online here). Now the fact that an economist is concerned with rising autism rates, and contributing to the research by tackling this question, shows the widespread interest autism has gained as a subject of inquiry -- "one of the most important health care crises facing the United States," as the study says. But reading through the paper, and Easterbrook's summary of it, one learns these facts:

1.) When it rains, people watch more TV.
2.) In the areas of California, Oregon and Washington where it rains more, people watch more TV.
3.) In those areas of those three states where it rains more and people watch more TV, more children are reported to have autism.

The problem, of course, is that just because 1, 2 and 3 are true doesn't mean that watching TV leads to autism. Correlation, things happening at the same time, doesn't necessarily lead to causation. Like watching a police car chase on live TV, it's easy to see action. But meaning and understanding are elusive. (The Freakonomics blog has a more detailed critique of this study.)

The Cornell authors readily note their work is not conclusive: "Although our findings are consistent with our hypothesis, we do not believe our findings represent definitive evidence for our hypothesis. The only way to establish definitively whether or not early childhood television watching is a trigger for autism is to more directly test the hypothesis." One suggestion they give: "monitor the viewing habits of a large number of children from the ages of zero to three and see whether the children who are eventually diagnosed with autism on average watched more television before the age of three."

For parents, there are a few takeaways from the Slate article and Cornell study and the ongoing flow of autism research that comes out in the media:

* Do what we can to encourage more research. Momentum continues to build as the public becomes more aware of autism as an important challenge for society. There are economists, journalists and parents of all backgrounds and professions who are asking questions now, in addition to scientists, doctors and educators. Lobby for their continued efforts.

* Know the general themes of autism research. This is a growing list that includes genetics and heredity, the environment including pollutants, brain function and structure, general physiology, immune systems and now ... television.

* Understand the limits of media coverage. News stories can only tell so much. See who's writing it, and how many people and studies they quote (rule of thumb: more sources is better). And try to evaluate whether any study looks at alternative explanations for the findings it reports. If it doesn't, that's a potential weakness.


Anonymous said...

I'm sorry, but you have got to be kidding. There's a big difference between a causal relationship and a correlation. For example, There are more crimes when ice cream sales goes up, but ice cream doesn't cause crime. Crime and ice cream sales go up in hot weather.

TV causing autism? This is just WAY OUT THERE, IMHO.

Anonymous said...

The author of the post clearly makes the distinction between correlation and causation, cites the possibility that this is only correlation and links to a critique of the study.

Ironically, you do the same thing you accuse the author or doing. The fact that it's "way out there" in your mind has no bearing on its veracity. That's a correlation not a causation.