Can Diet Really Improve Autism? Interview with Jennifer Harrison Elder, PhD

Can Diet Really Improve Autism?CCPR: Dr. Elder, you have written a number of papers on dietary interventions in both autism and ADHD. How did you get interested in this?

Dr. Elder: I have been in the autism field since the early ‘80s, and I learned a long time ago to listen to parents. I kept noticing that it was becoming more and more common that parents either had their children with autism on a trial of a gluten-free, casein-free diet, or they were seriously considering doing so. This type of diet was not well-studied in any formal, randomized, clinical trials and there was no empirical evidence of its effectiveness. I have noticed over the years, because we don’t have a cure for autism, people are desperate to try to find something that works. And sometimes they report that they see effects when they really aren’t that noticeable to others who can be more objective.

CCPR: Why do you think that is?

Dr. Elder: Some people have called this the “parent placebo effect.” I don’t have a child with autism, and I can only imagine that if I did I would be trying everything possible, and I would really want to see positive effects, and any small positive effect would be huge to me. I think in some cases, not all, some objectivity may be limited.

CCPR: In one of your papers you referred to a study where parents were unaware if their children were or were not being given sugar. Those who thought their child was having sugar answered differently from those who thought their child was having sugar substitute (Feingold BF. Why your child is hyperactive. New York: Random House;1975).

Dr. Elder: Right. I mentioned that as an example of the fact that we have to be concerned that there is a possible placebo effect. In that study parents were told one time that the child was getting a drink that did not contain sugar, and they were told another time that they were getting a drink that did contain sugar. When they thought the child was drinking the drink with sugar, they saw more hyperactivity, but actually it was just the opposite according to clinical measurement. It was all about the power of suggestion.

CCPR: And you tried to design a study that would take this into consideration.

Dr. Elder: My research team and I set up a 12-week double blind, randomized, placebo-controlled trial of a gluten-free, casein-free diet for children with autism. For half of the 12 weeks (6 weeks) they got the gluten-free casein-free diet, and for the other half they got a diet that contained gluten and casein. It was a small trial. It was very difficult to get the children to comply with the diet. We provided all their foods and they were instructed not to have anything else unless it was approved ahead of time by the dietitian. Parents were great but they admitted that sticking to the restricted diet over time was very hard (Elder JH et al, J Autism Rel Disord 2006;36(3):4l3-420).

CCPR: And what did you learn?

Dr. Elder: Our findings from comparing the two groups were not significant when we analyzed them, but we had some families who said they saw effects that we didn’t measure with the instruments we used. They saw enough effects, in fact, that some decided to keep the children on the diet. So we thought that was interesting. Since that time, at least one other randomized clinical trial had almost the exact same results that we did. So the take away messages were, first, we need to have more rigorous trials with larger samples. And, I am beginning to think more and more that there also may be a subset of children who are responders that we have yet to conclusively identify.

CCPR: Do you think that would explain the various reports of people having seemingly miraculous differences?

Dr. Elder: Yes I think so. It’s possible that we just haven’t done a good job of identifying that subset. For example, these children may have allergies to certain food or they may have unusual GI problems. I quite often hear anecdotal reports from parents who say what unusual bowel movements their children have. Sometimes it almost sounds like an irritable bowel syndrome where it goes from constipation to diarrhea and back again. And I have even considered doing a follow-up trial with a diet where we only include children who have these kinds of reports and see if that would help us to narrow it down.

CCPR: So as clinicians, how should we approach the question of diet with our families?

Dr. Elder: I wrote an overview article about a mother that I met at a conference. I asked her about the gluten-free, casein-free diet and she said that she put her child on it. And I said, “Well we don’t have the empirical evidence to prove that works.” And she said, “You know I am a very well-educated woman, I am a very good consumer, and I am very well read, but if I wait for science, my child will be grown. If there is something I can do that can possibly help my child that is not going to be dangerous or risky and I can afford it, I am going do it.” And you know I can’t much argue with that.

CCPR: So there’s no reason to discourage a parent from trying the diet, even in the absence of scientific evidence.

Dr. Elder: I think what we have to do as clinicians is make sure that whoever we are talking to is well informed. It is very common for parents to come with a lot of information, but I always go back to: what sites did you get it from? Who are you quoting? I often refer people to Autism Speaks because I think they do a very good job filtering out a lot of information that is just not credible and only reference what is credible. And then if, for instance, they are interested in implementing the diet, I will ask them: Have you thought about what is going to be involved? Have you thought about your child being maintained on the diet and nothing else? If you’ve got other children in your home, have you thought about locking cabinets and the refrigerator? Do you have somebody who can keep records? Have you thought about the cost in getting the food? And then they need to make sure that if they do it, that they do it long enough. One criticism of our study was that it didn’t show results because we didn’t do it long enough. So I recommend that families do a trial of at least three months.

CCPR: And what about outcome measures? How do they know whether it worked, especially in view of the placebo effect?

Dr. Elder: Parents can do assessments such as the CARS, but I would encourage them to partner up with an objective person: a professional who can assess the children behaviorally.

Because we don’t have a cure for autism, people are desperate to try to find something that works.
~ Jennifer Harrison Elder, PhD

CCPR: Any final thoughts, Dr. Elder?

Dr. Elder: I would just like to say that I think any good clinician should listen to parents because parents live with the child 24/7. We have an obligation not only to listen to them, but to help them make informed decisions about treatments, because there are always more that are being developed. The lag time between when these treatments become popular and when we actually have the science to back them up or to refute them is something that must also be taken into account. I call it a feedback loop. It is very important to shorten that by listening to parents. Also, we as researchers need to be researching what is important to them and get that information back to them as quickly as we can.

CCPR: Thank you very much, Dr. Elder.

Can Diet Really Improve Autism? Interview with Jennifer Harrison Elder, PhD

This article originally appeared in:

The Carlat Psychiatry Report
Click on the image to learn more or subscribe today!

This article was published in print November 2013 in Volume:Issue 4:6.


APA Reference
Harrison, J. (2016). Can Diet Really Improve Autism? Interview with Jennifer Harrison Elder, PhD. Psych Central. Retrieved on June 27, 2019, from


Scientifically Reviewed
Last updated: 29 Feb 2016
Last reviewed: By John M. Grohol, Psy.D. on 29 Feb 2016
Published on All rights reserved.