In the year 2000, a study titled The effect of vitamin-mineral supplementation on the intelligence of American schoolchildren: a randomized, double-blind placebo-controlled trial was published detailing the results of providing vitamin and mineral supplements to school children in Phoenix, Arizona. Read the abstract of the study. [Note: I believe there is a misprint in the abstract and that the number "81" cited in Result #4 was lifted from an earlier 1992 study by the same author, a study that yielded much the same conclusion, incidentally. It should probably have been "101" instead.]
Goal of the Study
The study goal was to determine what effect, if any, vitamin supplementation would have on the nonverbal IQ of school children, as measured by the Wechsler Intelligence Scale for Children-Revised (WISC-R). Twelve previous studies had shown a statistically significant positive effect, but each suffered from design flaws, so there was a need for this effort.
Design of the Study
245 schoolchildren aged 6 to 12 years in two Phoenix AZ elementary schools received daily vitamin-mineral supplementation at 50% of the U.S. daily recommended allowance (RDA) for 3 months with approximately half of them getting only a placebo instead of the multi-vitamin.
1. The group of 125 children who received the multi-vitamin (the active group) gained a statistically significant 2.5 IQ points on the nonverbal IQ test, versus the 120 children given the placebo tablet (the placebo group.)
2. This result was comparable to the average 3.2 IQ point gain found in the earlier studies.
3. A significantly higher proportion of children in the active group gained 15 or more IQ points when compared to the placebo group. That is, the active group had a much larger number of students who made individual gains that were quite large.
4. The students were sorted into matched pairs (one on active, one on placebo) where each member of the pair had a similar outcome in the study. After the pairing, as the abstract states: “The modest 2.5 IQ point net gain for the entire sample can be explained by the remaining 24 children who took active tablets, and had a 16 point higher net gain in IQ than the remaining 19 placebo controls.”
The Authors’ Conclusion
The authors concluded, with justification, that increasing the nutrition of the students resulted in a quite significant increase in nonverbal IQ scores for 24 of the children in the active group. Their second conclusion was that the rest of the students in the active group, who saw no significant increase in IQ scores relative to their placebo counterparts, were already sufficiently nourished and therefore did not benefit from the nutritional supplement.
A Different Model Yields a Somewhat Different Conclusion
First, note that 24 of 125 is about 19%. That is, 19% of the active group’s students experienced very significant gains in nonverbal IQ in just three months. This is a huge gain over such a short time period. If nutrition, or malnutrition, were at fault, one would expect to see a more graduated type of response, with those who were just modestly malnourished benefiting only a bit, but still benefiting, while those who were borderline well-nourished would also see some benefit. Instead, it’s like an on/off switch was switched on in almost 20% of the active group.
Could a Dyslexia Gene (or Genes) be the Switch?
I’ve maintained consistently on this site that dyslexia is an inherited disorder that is developmental in nature and that affects about ten to fifteen percent of the school population. Furthermore, the nature of the delayed development seems quite variable and appears to be modifiable by various therapies, including vision therapy, listening therapy, exercise therapy, and now even by nutrition.
Another Possible Conclusion
Using this new model, it’s entirely possible that a larger portion of the active group was malnourished than the 19% assumed by the authors of the study (based on those who responded.) However, if members of the malnourished group were not carrying the genetic predisposition to dyslexia, they did not benefit from the active treatment because they were not already developmentally delayed. Instead, only the dyslexic group, i.e., those carrying the gene(s) for dyslexia, were positively influenced by some nutrient, or possibly several nutrients, in the multi-vitamin administered to the active group.
If you have access to the entire study you will note that it fails to provide sufficient detail to expand on what is written in the abstract regarding the out-sized performance of the responders. However, a chart on the vitamin deficiency status of the participants indicates that both responders and non-responders were deficient in various vitamins or minerals with about twice as many responders deficient. So, why didn’t the non-responders who were vitamin-mineral deficient respond to treatment? Possibly because they didn’t carry the gene(s) for dyslexia? I don’t believe this hypothesis can be ruled out at any rate, though the authors might have sufficient data to do. For instance, it would be interesting to know if the 24 students who showed such dramatic improvement in nonverbal IQ demonstrated dyslexic tendencies in the form of earlier speech delays, difficulty learning to read, etc., prior to the participating in the study.
Verbal Versus Nonverbal IQ Scores
Note that the study measured nonverbal IQ only. The nonverbal IQ is also referred to as the “performance IQ” and is generally considered to demonstrate strengths or weaknesses in visual tasks and processing speed, as opposed to language/auditory tasks, which are measured by the verbal IQ portion of an IQ test. One way of diagnosing a learning disability is to base it on an unusually large discrepancy between the verbal and nonverbal portions of the IQ test, regardless of which score is higher. That is, a very low nonverbal versus verbal score can still indicate dyslexia.
The study showing that nonverbal scores jumped by a very significant amount in a subset of the active group, and not at all in the rest of the group, could mean that the nutrient supplementation had a positive effect on the visual skills and/or processing speed of that subset, and that subset only. How much of a stretch is it to assume that the nutrients filled a developmental need, thereby generating out-sized performance gains in just three short months of the study? And if this were the case, which nutrients were primarily responsible for filling the need? Of course, all of this is speculation.
The Logical Conclusion, Regardless
Regardless whose model is correct here, one obvious conclusion to draw is that nutrition matters, and could matter a great deal. I’ve discussed the modern-day prevalence of both vitamin D3 and Omega-3 deficiencies, and here is yet another study (supporting, don’t forget, the conclusions of twelve previous studies) that diet is somehow important to mental abilities. It would be so “magic-bullet” to suppose that the active component in the fish oil study contained enough Vitamin D3 (it probably didn’t) and that the modest amount of Vitamin D3 in the multi-vitamin in this study (most likely a minimal 200 IU or so) were both sufficient to overcome a Vitamin D3 deficiency.
It's more likely, however, since our modern diets tend to have replaced Omega-3's with Omega-6's, that the improvements noted in both this study and the fish oil study discussed on the previous page were related directly to the Omega-3 supplementation. It would be interesting to see the results of a similar well-designed study where the supplementation was of vitamin D3 instead.