Thursday, July 19, 2007

Skeptic's Circle

A 4th grade class will be visiting the Museam of Skepticism.
You can too.

Monday, July 16, 2007

I’m Gonna Buy Me, One of Ben’s T-Shirts

This morning, I was reading the daily web news and generally minding my own business when I came across this little gem on that bastion of internet truth, MSN. I recognized this diet as it was was popular in the 70's and 80’s for hyperactivity and because I had the opportunity to be on it as a child.

This was perhaps a little surprising considering how much research shows the Feingold Diet doesn’t help ameliorate ADHD or hyperactivity and that most serious science in the field doesn’t give the Feingold Diet the time of day
anymore. But that can happen when the last piece of serious science on the subject was published over ten years ago.

Krummel et al., (1996) are pretty direct when they say:

“Numerous double-blind studies of the Feingold hypothesis have led to the rejection of the idea that this elimination diet has any benefit beyond the normal placebo effect. Although sugar is widely believed by the public to cause hyperactive behavior, this has not been scientifically substantiated. Twelve double-blind, placebo-controlled studies of sugar challenges failed to provide any evidence that sugar ingestion leads to untoward behavior in children with Attention-Deficit Hyperactivity Disorder or in normal children. Likewise, none of the studies testing candy or chocolate found any negative effect of these foods on behavior. For children with behavioral problems, diet-oriented treatment does not appear to be appropriate.”

Now, the role of Nutritionist is one I would like to have a lot of respect for; after all, who is against nutrition? However, it turns out that just about anyone can call themselves a
Nutritionist. It also seems that my general lack of respect for the title of “Nutritionist” isn’t limited to just me.

The Feingold Diet, for the uninitiated, was named after pediatrician and allergist Ben Feingold. It is a so called, elimination diet, where certain foods or additives are removed from a child’s diet. This leads to a theoretical improvement in a child’s behavior. Among the usual suspects are various food additives and dyes e.g. (red #40).

The Feingold Diet has never officially included sugar in its banned substances list. However, Feingold practitioners may advise the removal of sugar (of one sort or another) in addition to the complete Feingold Diet, if the diet by itself is not enough to reduce behavior problems. Dr. Feingold himself mentioned that cane sugar was suspect.

The diet does have some empirical support (Conners et al., 1976; and Rowe & Rowe, 1994). It looked fairly legitimate at one point in time. It also had its champions. At least one name should be familiar to those of us in the autism world. The late Bernard Rimland, who was better known in his latter years for his alternative theories; encouraging chelation for kids with autism; was an outspoken defender of the Feingold Diet.

Dr. Rimland raised a number of interesting
criticisms including the fact that the disproving studies couldn’t possibly have studied all 3,000 additives prohibited in the diet. Of Dr. Rimland didn’t explain how anyone could have accomplished this feat. He also mentions that the darn kids were sneaking “illicit food” during the study.

In a very interesting passage he writes:

“Who needs artificially colored and flavored food anyway? For millennia the human body - and mind - has evolved and thrived on real food. It is prudent to feed our children and ourselves real food, not the denatured, "refined," additive-laden artificial foods that emanate from factories. What is the cost to us, to our country, and to our civilization of allowing ourselves to be seduced into consuming the gaudy colors and deceptive flavors that are used to make non-nutritious food appear desirable?”

Anyone reading the above passage shoudl quickly note the logical fallacies present. Such fallacies inlcude the argument by rhetorical question. Argument from tradition. The fallacy of the false dilemma. Argument from adverse consequences. And another false dilemma. So, Dr. Rimland racks of 5 fallacies in 4 sentences.

But it isn’t just Dr. Rimland. There seems to be a fundamental logic problem in the whole nutritionist cum Feingold culture.I remember the first time I realized there might be something funny with Feingold Diet was in High School, when I read a pro-Feingold book that stated that even if what the Feingold Diet treated was not really an allergy, it was okay to call it such, because the problem was analogous to an allergy.

However, that book was written in the late 70’s and it was already 20+ years old when I read it. Different time….right? Different standard of proof.

In the early days the Feingold supporters had a very scientific looking way to help parents and caregivers navigate through which foods were a problem and which were not. The simply would add or take away a food item and over the course of a week see what would happen. And just like that, post hoc, ergo propter hoc, the answer would be made plain.

This represented a can do attitude on the part of Feingold and company. It involved the parents, it put the power and control in their hands. In today’s alternative medical circles some people would call them “health care freedom fighters” or some such title.

However, we have reason to be a bit cautious here. Hoover & Millch (1994) found that after giving supposedly sugar sensitive children artificial sweetener and telling their parents it was sugar, the parents rated the children’s behavior significantly worse than controls. This result seems to indicate that even the people who know a child the best, can still fall victim to the confirmation bias and other self-trickery. Post hoc logic, even when dressed up to look like science, still isn’t science.

Keeping the above fact in mind look
here. It is a bit disappointing to see the same pseudo-science being pushed 30 years after the fact to justify a diet that is largely negated by the research, and that hasn’t been seriously researched since 1994.

At one time the Feingold diet seemed like a plausible treatment. That day has come and gone, because the majority of the research didn't back it up. But the Feingold advocates are still promoting their theory which has seen little or no change since the last bit of science on the issue. Once again, the observation that the more science changes, the less woo does, is proved to be true.


Conners CK, Goyette CH, Southwick DA, Lees JM, Andrulonis PA. (August 1976). "Food additives and hyperkinesis: a controlled double-blind experiment.". Pediatrics 58(2): 154-66.

Hoover, D. W., Milich, R. (1994). Effects of sugar ingestion expectancies on mother-child interactions. Journal of Abnormal Psychology. 22(4), 501-15.

Krummel D. A., Seligson, F. H., Guthrie, H. A. (1996). Critical Reviews in Food Science and Nutrition. 36(1), 31-47.

Rowe KS, Rowe KJ (1994). "Synthetic food coloring and behavior: A dose response effect in a double-blind, placebo-controlled, repeated-measures study". Journal of Pediatrics 125: 691–698.

Friday, July 13, 2007

CDDS Autism Zombie

(click to make larger)

I have already written extensively about the California Department of Disability Services autism data. Its getting to hard to imagine that there is something to still say.

But at least 4 times a year, I find myself writing something about the latest CDDS report. This is because the CDDS issues a quarterly report and a group of individuals who advocate a vaccine etiology of autism analyze the data in manner that inevitably supports their theory. This happens every quarter. This happens no matter how many times we criticize the use of the CDDS data because it is not supposed to be used as epidemiologic data, as the CDDS themselves clearly state. Just like Ouija Boards and astrology, there are always a few who resurrect this undead creature, and send it growling off towards public consumption.

What is more, this zombie seems impervious to attack. If the data indicate a continued rise, this is proof that vaccines cause autism and even though much of the thimerosal has been removed, there is still some being used in the flu vaccines and this is the likely culprit. But, if the numbers drop, this is also proof that the vaccines cause autism, because the sudden drop can be correlated to the not quite complete removal of thimerosal.

I should say that some take a more reasonable approach. They accept the criticism that these data are not meant to be used for epidemiology. However they then argue that it is the best that can be had. And that if one sees trends here, even if the data are flawed, this is still indicative of an actual effect. This argument is actually somewhat widespread.

But the argument is still flawed. If one refers to the graph, it should be clear that the natural statistical variation allows for an occasional decrease; even one extending for a few quarters before an uptake is noted. That is the nature of statistics; one is going to see an occasional slight downswing. Moreover, the fact that the CDDS data are so regular does not make them reliable or valid.

The CDDS uses a form version from 1986. It has the following categories.

0 = 'None''
1 = 'Full Syndrome''
2 = 'Residual State''
9 = 'Suspected, not Diagnosed

The form counts the 1’s and 2’s. These categories are not part of DSM-IV-R. They are older categories based on earlier ideas. Thus, it can not be correctly said that CDDS offers services only to Autistic Disorder as defined by the DSM-IV. Full syndrome includes both PDD-NOS and Autistic Disorder.

An Autistic Spectrum Disorders Best Practice Guidelines For Screening, Diagnosis and Assessment statewide standards was held in 2002. A revision to this practice that conforms to the DSM-IV and ICD-9 will not be implemented until 2008.

These data are not good measures of Autistic Disorder, as they have been advertised by some persons, almost unceasingly. Moreover, there are no reliability checks conducted to make sure the State diagnosticians are conforming to recommended practices. This makes them an unreliable and invalid, source of autism data. For these reasons any attempt to use these data for purposes of epidemiology is ill-founded.

Sunday, July 08, 2007

MMR Shenanigans

For those of you who may have missed the recent hoopla. A study by Simon Baron-Cohen’s lab has been leaked to the British press. It describes a prevalence of 1 per 58; using a tool that lacks evidence showing that it is appropriate for use in the general public. It seems that a research assistant in the study who now works for Andrew Wakefield had attempted to connect the 1 per 58 number to the MMR.

I have learned via Bad Science that a connection to the MMR is not the study authors’ conclusion. Seems like there is some funny reporting going on.

Saturday, July 07, 2007

7 Myths of Science

The following are the 7 most common myths about science and research that I have seen. I have attempted to refute them in brief, even while recognizing that short answers to such widely held ideas are almost never satisfactory. As always, your comments are welcome.

1 Anything can be proven with statistics; therefore such-and-such a study which did/did-not show an expected result is bogus because the authors could have shown anything with their stats.

Reply: A knowledgeable statistician can arrange a situation so that it appears as if a negative result is actually a positive. However, this is one of the reasons for peer review in scientific journals. It is one thing to get statistical manipulation past the general public; it is another to get it past one’s peers who have also been trained in depth in statistics. In fact one of the major focuses in statistics and research courses is detecting statistical and graphical shenanigans.

2 Science changes its mind all the time, so why should I accept what the research shows on such-and-such an issue.

Reply: In science particularly in broad explanation based theories; paradigm shifts do occur. Numerous examples of this in physics or medicine should familiar to everyone. However, smaller bits of descriptive science can be more robust. For example, our knowledge of how genetics influences health is likely to alter significantly in the indeterminate future (maybe even cause a paradigm shift), but this fact will not likely invalidate the studies showing penicillin to be effective for such-and-such a strain of bacteria. This is sometimes true even in broader applications. Newtonian physics may have been displaced in the broad sense by relativity, but as many a sore handed college student can attest, differential and integral calculus is still with us and still has utility.

3 Science can not explain the entire human experience.

Reply: Science can only deal with the objective, or what two or more people can detect in the same way, at the same time. However, this fact can be twisted into justifying certain beliefs. Because science has this basic requirement it can not effectively assess a great deal of the human experience including thoughts, feelings, dreams, and perceptions. However, it is a logical error to assume that because science has this gap, we can or should believe in the issues science can not assess.

4 Science can not answer metaphysical questions.

Reply: That really depends on the specific metaphysical question. If there is measurable outcome and a detectable initial cause, then science can indeed assess the question. For example we can not measure divine intervention, but we could assess the effect of prayer on the recovery of heart surgery patients.

5 Law courts convict or release people, on far less and weaker evidence than what passes in science.

Reply: The systems are considerably fundamentally different. In courts (in many places) the people making the decision are limited in the type of answer they can give. They also have a distinct time limit in which they must reach a decision. They also are by no means experts in the necessary elements of the issues at hand. They are also restricted in the type of evidence they can hear.

Compare this to science where the people are experts in the specific areas of question, they have unlimited time, the nature of the arguments they are permitted to hear is not limited, and the type of answer they can give after assessing the evidence is not nearly as limited.

6 Science at times seems to violate common sense, therefore such-and-such a scientific result is false.

Reply: Common sense, when invoked merely means one person’s perception of what most people believe. That perception may be incorrect. Also, no matter how many people hold the same view in common, doesn’t mean the view is factual.

7 Science is a religion and well known scientists are the high priests of the religion.

Reply: This line of thinking is so underdeveloped, as to scarcely be worth the time. I offer a refutation only because it seems to be so popular. The word “religion” means something specific, the methods and logic used in science, do not fall under the umbrella of religion. Well known scientists are well known because of their contributions to science and/or the clarity of their writing. If ever they cease to be clear in their quality of thought, they may no longer be quite so esteemed e.g. (Linus Pauling).

Friday, July 06, 2007

Study abstract:

“Autism is a developmental disability characterized by severe, pervasive deficits in social interaction, communication and range of interests and activities. The neurobiologic basis of autism is well accepted, although the specific etiology is unknown. It has been theorized that autism may result from a combination of predisposing genes and environmental factors. While autism has a known association with environmental factors such as rubella and valproic acid exposure in utero, other proposed environmental mechanisms such as mercury toxicity or other heavy metal exposure have minimal research support. Despite this fact, interventions including oral and topical chelation therapy are being used to treat autism following evaluation of hair, blood, or urine samples for heavy metal toxicity.
In this study, hair samples were obtained from 15 children with autism between the ages of 2 and 6 years and 16 controls in the same age range who are the siblings of the subjects. The hair samples were obtained according to lab. specifications and submitted in a blind fashion to Doctor's Data Lab. for measurement of mercury levels. Data from the two groups were then compared using T-test. No significant differences were found between mercury levels for the two groups. This study raises questions about the theory that mercury toxicity causes autism and points to the difficulty in quantifying chronic mercury exposure through currently available laboratory measures.”


The research was supported in part by a grant from the University of Louisville Pediatric Research Foundation.


, G. P., Hersh, J. H., Allard, A., & Sears, L. A. (2007). A controlled study of mercury levels in hair samples of children with autism as compared to their typically developing siblings.

Thursday, July 05, 2007

Skeptics Circle

Check out the good posts on autism and other assorted critical thinking.
Although I admit I have a favorite.

Skeptics + Ninjas = Awesome