The original randomized controlled trials of gluten- and casein-free diets may have been complicated by parental expectation bias.
My video Are Autism Diet Benefits Just a Placebo Effect? begins with graphs depicting the findings from the first randomized, controlled trial to put a gluten- and casein-free diet to the test for children with autism. Significant improvements were reported in attention problems, and there was less aloofness, fewer needs for routines and rituals, and improved response to learning after a year on the diet. In contrast, there were no significant differences reported before and after the year-long experimental period in the control group. Similarly, there were improvements in social and emotional factors in the gluten- and casein-free diet group, with kids having significantly fewer problems with relationships with their peers, less anxiety, more empathy, and more acceptance of physical contact. And, again, no significant changes were reported in the control group. There were significantly fewer communication problems in the diet group, too, with more facial expression, better eye contact, and improved responsiveness, and fewer incidences of “language peculiarities,” such as meaningless word repetition, whereas, once again, there were no significant changes in the control group. Finally, in terms of cognitive factors and movement, after a year on the diet, there was significant improvement in the kids’ ability to judge dangerous situations, expanded personal interests, and lower likelihood of being inordinately restless or passive.
The study, however, was problematic in that the researchers relied mostly on parental report. They asked parents a series of questions before and after the year-long trial to see if they detected any differences in their children, which you can see at 1:20 in my video.
Is that really a problem? Who better knows the day-to-day functioning of children than their own parents? Yes, the researchers could have had an impartial observer assess each child before and after the experimental period, blind to whether they were in the diet group or the control group, but those assessments would just be snapshots in time. Who better than the parents themselves to know what was going on with their children? Well, there really is a problem with parental assessment: The problem is the placebo effect. “After implementing an intervention for up to a year, parents…had invested a great deal of time and effort into maintaining the diet and may have been looking for improvements in their children’s behavior.” Wheat and dairy are in so many products that eliminating them from the diet can be a big shift for most families. So, while the families in the control group did nothing special that year and reported no significant changes before and after the experimental period, the families in the diet group had put in all this work, so when they were asked if their kids appeared better, their opinions may have been impacted by their expectations of benefit. In other words, “placebo effects may have been at play.”
Are parents really that gullible? Well, “the power of suggestion on the part of parents can be very strong in situations affecting their children’s behavior.” For example, there was a famous study in which all the children were given a drink with artificial sweetener, but half of the parents were told the drink was sweetened with sugar. The parents who thought their children had received the sugar-sweetened drink rated their own kids’ behavior as significantly worse.
So, in these sorts of autism studies “where placebo effects exist, it is possible that parents are looking for positive changes in behavior and ignore or explain away negative ones.” Ideally, we need double-blind studies, where kids are given foods made to look and taste the same, but one food contains gluten and casein while the other is gluten- and casein-free. The kids don’t know which is which, and neither do the parents. Even the researchers don’t know which is which until the end of the study when they break the code. “In this way, the behaviors recorded after the food challenges could not be impacted by preconceived ideas or biases.”
Well, why wasn’t the study double-blind then? “With regard to design,” the researchers conceded, “it might be argued that a double blind, cross over study might have been ideal. With all children on diet, gluten and casein could have been administered, for example in capsules during specific altering periods. Parents and caretakers would have been blind to who was on the diet and who was on ordinary nutrition,” secretly getting gluten and casein unbeknownst to them. Then, they could have eliminated the placebo effect and eliminated that expectation bias. So, why didn’t they?
The researchers were so convinced that gluten and casein were harmful that—“from an ethical viewpoint”—they just couldn’t bring themselves to give these kids gluten or casein. The kids in the diet group seemed to be doing so much better, and the researchers had seen cases in which the children appeared to relapse when those proteins were reintroduced back into the diet, so they just couldn’t bring themselves to slip any to the kids on the sly.
I understand that, but if they really were so certain that gluten and casein were bad, by designing a less-than-ideal study, they were then potentially dooming scores of other children by failing to provide the strongest possible evidence. Thankfully, four years later, other researchers stepped in and published the first double-blind clinical trial on diet and autism, which I cover in detail in my video Double-Blind Trial of Diet for Autism.
This article discusses the fourth video in a six-part series on the role of gluten- and dairy-free diets in the treatment of autism. If you missed the first three, check out:
The last two videos in this autism series are Double-Blind Clinical Trial of Diet for Autism and Pros and Cons of Gluten-Free, Casein-Free Diets for Autism.
Keep abreast of all of my videos on autism here.
For videos about gluten-free diets in general, see:
Michael Greger, M.D.
PS: If you haven’t yet, you can subscribe to my free videos here and watch my live presentations: