The Problem with "Studies".
Today I came across a study indicating that diet plays no role in autism. You can see the article here. Of course we don't see the actual data. All we see is a sensational headline that sounds like a definitive conclusion. Even so, it is easy enough to pick this study apart. One, it was a small study with only 14 children, which does not yield much power when doing statistical analysis. Two, there is no indication that these children had any prior dietary problems. I've said repeatedly that not all autism is affected by diet. A better study would have used children who were thought to be diet-responders versus those who are not. Third, the time period of the study (18 weeks) was much too short. Most who are familiar with the diet know that at least six months is needed as a fair trial. Also, no thought was given to observe the effects of dietary infraction at longer time points. Many parents report a delayed behavioral response to an infraction of 48 hours or longer. I would have the same criticisms of this study had the conclusion been reversed. This study is an example of the disconnect between traditional medical research and the reality of a developmental disorder. If researchers were to listen to families who deal with autism 24/7 perhaps more realistic and helpful outcomes would occur. -Devin