New GFCF Study
A new GF/CF study will be presented at IMFAR. Hyman et al (2010) took a challenge based approach, which is different from what has been seen so far. The authors enrolled 22 pre-school aged children and placed them all on a GF/CF diet. The children were given a snack once a week that either contained gluten/casein or placebo. Snacks that contained gluten and/or casein were designed to be indistinguishable from the snacks that did not. This was described by the authors as an “exercise in innovative cooking.”
Strengths:
Innovative design
Nice use of a challenge based methodology
Solid, double blind methodology
Smart controls over history and maturation threats
Weaknesses:
22 participants is a respectable number, but it still could have been higher.
Can’t answer for the diet’s efficacy for children with actual bowel conditions (Celiac etc)
Conclusions:
The autism community has been waiting a long time for a study like this. The best we have seen up till now has been the Elder (2006) study. The methodology is simply rock solid.
Predictions:
I do not predict that this study will have any effect on the advocates of the GF/CF diet. For some reason the GF/CF advocates have been more polite than some of their colleagues. They have typically responded to negative GF/CF studies by trying to find flaws in the study methodology, rather than silly accusations about “shills”.
They will do the same here, I think. My guess is that they will target the number of participants (which is silly in this case). They will also target the time-span (18 weeks) the children were on the GF/CF study in total and on the diet before the first gluten/casein snack was given (4 weeks). Such criticism will be irrelevant however, as there are no standards in this regard. I expect plenty of goal posts to be shifted. I am convinced at this point that being an academic GF/CF advocate means never having to pick an actual study length.
Strengths:
Innovative design
Nice use of a challenge based methodology
Solid, double blind methodology
Smart controls over history and maturation threats
Weaknesses:
22 participants is a respectable number, but it still could have been higher.
Can’t answer for the diet’s efficacy for children with actual bowel conditions (Celiac etc)
Conclusions:
The autism community has been waiting a long time for a study like this. The best we have seen up till now has been the Elder (2006) study. The methodology is simply rock solid.
Predictions:
I do not predict that this study will have any effect on the advocates of the GF/CF diet. For some reason the GF/CF advocates have been more polite than some of their colleagues. They have typically responded to negative GF/CF studies by trying to find flaws in the study methodology, rather than silly accusations about “shills”.
They will do the same here, I think. My guess is that they will target the number of participants (which is silly in this case). They will also target the time-span (18 weeks) the children were on the GF/CF study in total and on the diet before the first gluten/casein snack was given (4 weeks). Such criticism will be irrelevant however, as there are no standards in this regard. I expect plenty of goal posts to be shifted. I am convinced at this point that being an academic GF/CF advocate means never having to pick an actual study length.
posted by Interverbal at 10:32 AM
17 Comments:
The methodology of this study is not related to the previous work on the topic. Jonathan I thought I have explained enough how , at what deepness Elder was incomplete and not well designed- besides very partial in their conclussions..
You may find here the list of how and where the study is incomplete and it does not consider the status of the art of the biochemical/metabolic studies and others relevant to the outcome of the GFCF- and should be soy free. done in ASD
http://www.autismweb.com/forum/viewtopic.php?f=6&t=22394
http://www.autismweb.com/forum/viewtopic.php?f=6&t=22413
Without the proper control of confounders, no progress is going to be possible....
Such criticism will be irrelevant however, as there are no standards in this regard.
yes, there are previous work of Dr Reichelt pointing to a minimun of 28 weeks to clear the effects of gluten. The criticism is relevant and to the point- and core to the situation because only behaviors were considered- with all the problems in consideration of behaviors in autistic children- and no clinical metabolic/mitochondrial or biochemical studies were included.
Hi Maria,
Simply put; the authors asked the questions I wanted to see asked. They did so with real methodological controls. You want different questions asked. I have no opposition for any study that attempts to answer those questions, but I am not about abandon the questions I wanted asked or to criticize the current study or the Elder study because it failed to ask your questions.
You say: “there are previous work of Dr Reichelt pointing to a minimun of 28 weeks to clear the effects of gluten.”
What do you want to bet, that it would take me 5 minutes to hunt down half a dozen diet miracle anecdotes that happened in a few weeks or under? This is what people read Maria, this is what people act on, especially in the GF/CF diet.
OK, let´see
Are the question you want to be answered useful in practical terms for autistic children with dozens of abnormal results related to digestion or GI issues?
NO, simply.
A diet without the concomitant consideration of other issues going on is a failure in general,especially in so short terms of time,without careful testing up to date in terms of gluten/casein intolerance-allergy or sensitivity except in very few cases- such as celiac disease with behavioral presentation with symptomatology associated to an ASD diagnosis. The only aspect that this kind of studies address is that. Again this is known.
There are published work of Dr Reichelt on the topic, there are new research that points to the complexity of the answer related to the interventions- not anecdotic evidence.
If you have 20 or more problems going on why only the avoidance of one , in a very short time, without all the other confounders considered (that may affect results and outcomes considering only the behavioral aspect)are going to give any meaningful answer or clue?
You see validity to study one aspect by time, but this is not related to the actual experience of many-or the research on the topic and it has near 10 years old in approach.GFCF- soy free- is not a treatment of autism, is the treatment of part of the concomitant medical problems going on with autism- and for many subgroups is not the only one.
This may be nick-picking, but the final results are based on only 14 participants, not 22.
The authors also excluded 2 children who had possible (but apparently not confirmed) celiac disease. The also excluded one child who showed iron deficiency. I would argue that these three make up the part of the group that would most likely benefit from the diet.
Two of my children had (and still have) iron deficiency and had higher anti-gliadin IgG (although not tTG) and have benefited from the diet. They would have been excluded from this study.
The duration of the "strict" GFCF part is also very, very short. You may feel that the standard story is that you see "instant" results but I don't think that is the norm. At 4 weeks we were still removing food from the our children's diet (per their doctors orders). You shouldn't instantly change a child's diet and yet that seems to be exactly what happened here. And at 4 weeks we were starting to see some changes in behaviors but the majority of the benefit happened after that point.
If we had "challenged" them at that point I doubt we would have seen any obvious effects. However, three years later we do have issues when there are dietary infractions (bowel issues for a few days and stims go WAY up). The behaviors are so obviously different that the children's therapists have picked up on every infraction without us telling them.
The other thing that really bothers me is that they made no attempt to screen for children who might benefit from the diet. We already know that not every child will benefit, so it is just plain silly to continue to keep ignoring that fact.
Take for example the recent trials of Abilify for treating autism. If the drug had been given to every child with autism instead of just children with irritability, do you think it would have shown any effect? Of course not.
The same would hold true here. If you test the diet in children who do not have the underlying problems, you are not going to see the results.
I am not sure what questions you think this study answers, but it doesn't tell us anything new.
There is another recent study on the diet that is much longer (2 years), larger (72 children), has a control group, and screened for children who might benefit from the diet. This study showed a modest positive benefit from the diet.
http://www.ncbi.nlm.nih.gov/pubmed/20406576
Hi MJ
Well, such as I understand the goal was to test the GFCF diet in a group of autistc children without wheat intolerance by testing and no abnormal results or problems detected in the tests they did:
milk/wheat allergies, celiac disease, and anemia/iron status by RAST, TTG and CBC/ferritin respectively.
Now, there are autistic children who responds to the diet done longer without clear symptomatology of celiac disease or wheat/milk intolerance or abnormal results in iron status. BUT iron status is not the only parameter to test, especially in a short trial like this. Confounders like fungal or bacterial infections in gut- abnormal gut flora should also be considered carefully. Enzymatic status also should be considered- because digestive enzymes are abnormal in many autistic children. Problems like GERD should be also taken into account to know about impact of the GFCF diet. It is not clear if the diet was also soy free for example. No aminoacids testing was including before and after the diet, no other essential and toxic elements.
Under this analysis, this study does not say something new.
My son also would have been excluded of this study. He was clearly celiac by testing and IgA severely defficient- besides other abnormal results.
Maria,
Doesn't that make the entire study just a little silly? If you don't think there is a gluten intolerance or problems processing gluten or related problem, why would you try a gluten free diet? The same for milk and soy. If you exclude the group that is most likely to benefit, then what's the point?
To use my earlier example, if would be like testing abilify by excluding all of the irritable children.
I would have been very interested in seeing a case study on what happens when you put a child with autism AND celiac on a gluten free diet. Would the symptoms of autism improve?
I wish the full study were available so we could see what the researchers actually did rather than guessing from the abstract and press releases.
Hi MJ
There are published case reports of autism and celiac disease
Please look at
http://www.ncbi.nlm.nih.gov/pubmed/19564647
and
http://memo.cgu.edu.tw/cgmj/3204/320414.pdf
About your question
Doesn't that make the entire study just a little silly? If you don't think there is a gluten intolerance or problems processing gluten or related problem, why would you try a gluten free diet?
Well, perhaps they thought in a potential different (altered) immune answer to gluten/casein than celiac but affecting behaviors, an approach that is NOT so silly IMO. Gluten intolerance sometimes presents wiht only abnormal behaviors- without biomarkers in blood- the common antigliadins or endomysium- or abnormal biopsy. IF their goal was that, again, the design of the study is poor because they should have tested for aminoacids ( the product of protein metabolism) status previous and during diet to know about changes- and again the time is very short. Being proteins, they are metabolized to aminoacids therefore aminoacids testing may give clues. Besides aminoacids are linked to neurotramission therefore if they wanted to know about impact in behaviors, they should have tested for biochemical consequences of protein avoidance in autistic children in blood/urine to increase knowledge to begin with. This way, nothing can be concluded from the results.
Recent researhc
http://www.ncbi.nlm.nih.gov/pubmed/19455131
pediatric celiac disease is being more studied
http://www.ncbi.nlm.nih.gov/pubmed/19606042
new study
http://www.ncbi.nlm.nih.gov/pubmed/19771572
and
http://www.ncbi.nlm.nih.gov/pubmed/20437547
if you see my former post there is a link where I posted recent reserach in the topic of biomarkers of gluten metabolism problems.
Hi MJ
Please look at this manuscript from 2002
http://www.ncbi.nlm.nih.gov/pubmed/12298186
Coeliac disease should be taken into consideration in patients with psychiatric disorders, particularly if they are not responsive to psychopharmacological therapy, because withdrawal of gluten from the diet usually results in disappearance of symptoms. In recent years, an increased incidence of subclinical/silent coeliac disease has been reported. Psychiatric symptoms and psychological behavioral pathologies could be the only clinical manifestation of coeliac disease, but the epidemiological aspects need further investigation.
Hi MJ,
“This may be nick-picking, but the final results are based on only 14 participants, not 22.”
Noted
“The authors also excluded 2 children who had possible (but apparently not confirmed) celiac disease. The also excluded one child who showed iron deficiency. I would argue that these three make up the part of the group that would most likely benefit from the diet.”
That is possible. More research will be needed to see.
“You may feel that the standard story is that you see "instant" results but I don't think that is the norm.”
Fast success stories are not hard to come by. This is partially how the GFCF diet is advertised. Even if they are not the norm, they are common enough where they support my original point.
“The other thing that really bothers me is that they made no attempt to screen for children who might benefit from the diet.”
Yes, that has been the main bone in contention. You said it, Maria said it, and William’s dad said it on AW. What you all want to know is for whom (and under what special conditions) does the diet work.
Again, I am in support of research to address that question. But I am glad the current authors asked the questions they did.
I think up till now there has been room for disagreement on the efficacy of GF/CF diet in the general case. No longer….. The special cases remain open however.
“I am not sure what questions you think this study answers, but it doesn't tell us anything new.”
In the general case (with certain groups removed) does the diet work? As I said, up till now I think debate on this point was reasonable.
“There is another recent study on the diet that is much longer (2 years), larger (72 children), has a control group, and screened for children who might benefit from the diet. This study showed a modest positive benefit from the diet.”
And was less controlled. I have problem when results go up as control goes down.
Hi Maria,
I see nothing in your latest comment that is not addressed to my satisfaction in my first response. If you have a specific question for me, let me know.
Maria - Thanks for the pointers.
Interverbal, you said -
"Fast success stories are not hard to come by. This is partially how the GFCF diet is advertised."
Actually I would disagree with that statement. What you are stating is how the diet is perceived by a certain group of people. Nobody is "advertising" the diet but parents do talk amongst themselves and swap ideas for what has worked for them. The majority of this chatter is not on the internet so would not be visible. For every anecdote that you can find online, I would suggest that there are ten times that number that are done in private and out of the public eye. As for a doctor who recommends the diet, I have not run across very many (if any at all) who would tell you that there will be instant results from it.
I have run across these attitudes many times, so I took the time to write about our experiences with the diet here -
http://autismjabberwocky.blogspot.com/2009/08/why-we-tried-gfcf-diet.html
and the changes that we thought we saw here -
http://autismjabberwocky.blogspot.com/2009/08/our-results-with-gfcf-diet.html
"I think up till now there has been room for disagreement on the efficacy of GF/CF diet in the general case. No longer….. The special cases remain open however."
Let me ask you a few questions then.
Given the incredible diversity in severity, abilities, problems, and potential causes when it comes to autism, what is the "general case" of autism? Think of it this way, if you have a group of 14 children with autism, what is the chance that they will all have a specific problem in common? Even in my own family, the problems my youngest daughter have are very different than the ones her sisters have. Her sisters are identical twins, and even then they only share about some of their problems (the have the similar communication problems and some repetitive problems but their social skills and the rest of the repetitive problems are different).
If you have a large enough set of participants, you will be able to get around this problem. But I don't think it is possible for 14 children to properly represent the "general" case.
Second, you say that there is no model of a study like this and that 4 weeks is an acceptable length of time to test. That isn't strictly true, there is a model in using the GF part of the diet for celiacs. So ask yourself (or better yet, go research and talk to people with celiacs), if you did the exact same trial in children with celiacs, would you expect to see clinically significant changes for the entire group after just 4 weeks? And given the frequencies of the challenges, do you really think that the participants would be able to get past the baseline measurement? I have know people with celiacs and I know that even with being strictly GF, it took them far longer then 4 weeks to get rid of their symptoms. I don't know what the average length of time required, is but I do know that it is different for each person and longer than just 4 weeks.
If the theories about how the diet work are true, you would expect that the child would need some time after the food was stopped to clear it out of their body and then some further time for the damage caused to subside. After that point I would expect to start seeing any changes in the core symptoms of autism.
Hi MJ
I agree. I have talked in extense with parents of celiac children and with their doctors- specialists in celiac disease. Besides the diagnosis, the good ones are very concerned about thyroides fucntioning, Vitamins, minerals (calcium, magnesium, zinc, selenium, copper
) and aminoacids status, they test for IgA and IgG, IgE, metabolic acidosis, liver and renal functioning, and fungal and bacterial infections as routine in a celiac child.In childrem, the results of the diet may be of different intensity at short time, but they test for these in baseline and at 6 and 12 months, with re-testing once a year to monitor evolution. The results of the diet are very individual in nature and depends on the results. Some children in my country have a very complex presentation of celiac disease, with even epilepsy and cerebral calcifications.
Hi MJ,
“Actually I would disagree with that statement.”
http://www.gfcfdiet.com/Successstories.htm
Notice how many are fast acting. That took maybe 30 seconds to find.
“What you are stating is how the diet is perceived by a certain group of people. Nobody is "advertising" the diet but parents do talk amongst themselves and swap ideas for what has worked for them.”
The families hear these stories; they act on these “recommendations”.
“Given the incredible diversity in severity, abilities, problems, and potential causes when it comes to autism, what is the "general case" of autism?”
Unknown (by anyone). But for some people the majority of cases (even every case) of autism is causes by GI related issues, that can be treated in part/whole by the GFCF diet.
“If you have a large enough set of participants, you will be able to get around this problem. But I don't think it is possible for 14 children to properly represent the "general" case.”
Let’s say that students taking cell phones to school is the general case. If you walked into an average classroom and poll half the kids (14 kids), would you expect that this would be large enough to represent the general case? Because I do. I think what you are really after is a subset not the general case.
“Second, you say that there is no model of a study like this and that 4 weeks is an acceptable length of time to test. That isn't strictly true; there is a model in using the GF part of the diet for celiacs.”
So, because this is the case for celiac persons, we should apply these standards to non-celiac persons? I don’t follow the logic there, but alright. Let’s look at celiac improvements for a minute.
http://digestive.niddk.nih.gov/ddiseases/pubs/celiac/
“Improvement begins within days of starting the diet.”
http://www.glutenfreemylife.com/celiac/just-got-diagnosed-with-celiac-disease-how-long-will-it-take-to-feel-energetic-again-after-giving-up-gluten
“But it’ll probably be better in a few weeks.”
“I felt better almost instantly!”
“I started feeling better after two weeks”
But my favorite quote is from here:
http://www.celiac.org/index.php?option=com_content&view=article&id=11&Itemid=16
“CDF strongly recommends that you get correctly diagnosed with Celiac Disease by a qualified physician before modifying your diet! Do not self-diagnose.”
Actual standards are precious things and feel so very absent for anything to do with autism and the GFCF diet.
“The other thing that really bothers me is that they made no attempt to screen for children who might benefit from the diet.”
This isn't exactly the case. I asked Dr. Hyman about the scientific basis for their research while at the IMFAR press conference. She reminded us that when this research began in 2003, a prevailing hypothesis that underlied all the parental anecdote was a leaky gut leading to absorption of [mystery] peptides that could dramatically and quickly influence "autistic" behavior.
The hypothesis was based on wide popularity of anecdote, as explained at the outset of their study:
"[A gluten-free casein-free (GFCF) diet is one of the most popular interventions for autism. In our area, about 50% of families with preschool children with ASDs use this diet, despite the absence of evidence for its clinical effects or safety...]"
"[The hypothesis that dietary gluten and casein alter the behavior of a subset of children with autism will be tested by assessment of children's responses to a series of dietary challenges delivered as snacks--containing wheat, milk, both or neither--during their educational program. Both the treatment team and the family will be blind to the content of the snacks]"
They studied exactly what, at the time, parental anecdotes seemed to be claiming. Incidentally, I highly doubt this is the last we'll hear from the University of Rochester on the subject. Dr. Hyman seems committed to investigating many other possible roles of nutrition in autism.
Post a Comment
<< Home