There is disagreement among professionals involved in research and treatment of ADHD as to the usefulness of vitamin and mineral supplementation as an effective treatment strategy. In a recently published review of the literature on alternative treatments it was concluded that mineral supplementation (e.g. iron, magnesium) in ADHD children shown to be deficient in those minerals yielded promising results in several studies and warranted additional research using double-blind, placebo-controlled trials. However, no convincing data on the use of vitamin supplementation has been reported that I am aware of, and a recent review of alternative treatments for ADHD is very negative about this approach. It should be noted that this applies to mega-vitamin treatments in which children are given very large doses of certain vitamins, rather than simply bringing children up to currently recommended levels.

You may be curious about the impact of possible vitamin and mineral deficiencies on aggression and other forms of antisocial behavior. Although such behavior is not among the core symptoms of ADHD, and many children with ADHD do not display any significant antisocial characteristics, it is widely documented that such behaviors do occur at elevated rates among ADHD children. Might vitamin and mineral deficiencies stemming from poor nutritional habits play a role in the development of antisocial behavior among school-age children?

This question was addressed in a recent study published in the Journal of Alternative and Complementary Medicine (Schoenthaler, S.J., & Bier, I.D., Vol. 6, 2000.) Participants in this study included 468 children between the ages of 6 and 12 from two predominantly Hispanic elementary schools. This represented approximately 75% of the total population of those schools from which parental consent to participate was obtained.

The design of this study was simple and straightforward. Children were randomly assigned to either an experimental group or a control group. Care was taken to ensure that the groups were equated in terms of children's intellectual ability. Children in the experimental group received two chewable tablets each day for four months that provided daily vitamin-mineral supplementation at 50% of the U.S. recommended daily allowance. The supplement was designed to raise vitamin-mineral intake up to the levels currently recommended by the National Academy of Sciences, which is quite different from some popular mega-vitamin approaches. Children in the control group received two identical looking placebo tablets (these tablets provided no vitamin or mineral supplementation).

Throughout this time period, the antisocial behavior of study participants at school was carefully tracked. Children were disciplined for antisocial behavior including things like fighting, threatening others, property destruction, and refusing to complete their work. The numbers of such behaviors that occurred over the 4-month study period and were deemed by school staff to reflect serious rule violations were tallied. This enabled the researchers to compare the rates of antisocial acts among those children receiving the vitamin-mineral supplementation treatment with those children receiving only a placebo.

Rather than make this comparison among all the children participating, the authors restricted their analysis to 80 children (40 in each group) who had been formally disciplined for violating school rules during the 4 months prior to the start of the study.  This was done because they were interested in whether vitamin-mineral supplementation resulted in reduced antisocial behavior; and among those children who had not committed any prior antisocial acts according to school records, no such improvement would be possible. It is important to note that the children in the supplement and placebo groups did not differ on any pretreatment measures examined (such as IQ or their number of prior offenses).

RESULTS

Before examining the impact of the vitamin-mineral supplementation on children's rates of serious rules violations, the authors first examined whether children could detect whether they had received the supplement or the placebo. The majority of children--regardless of actual group assignment--guessed that they had received the supplements, and the rate of accurate guessing did not differ between the groups.

During the 4-month intervention period, the children who received the active vitamin-mineral supplementation had an average rate of serious rule violations at school of 1 per child. For children receiving the placebo, the average number of serious rules violations nearly doubled to 1.875 per child. This is a statistically significant difference. The reason for the higher average rate among the placebo group was due almost entirely to a relatively small number of children who committed multiple offenses. Thus, among children receiving the supplements, there was only a single child who committed more than one offense during the 4-month study period. Among the placebo group, in contrast, there were 9 students who committed multiple offenses.

SUMMARY AND IMPLICATIONS

This well designed study provides solid preliminary evidence that for some children, vitamin and mineral deficiencies may play a significant role in their behavioral difficulties. The authors suggest that this occurs because of the adverse affect that vitamin and mineral deficiencies can have on brain functioning. They also note that their findings replicate those that have been reported in several prior studies.

This cannot be concluded with certainty for several reasons, however. First, we do not know whether those children in the placebo group who committed multiple offenses were actually deficient in key vitamin or minerals to begin with because such an assessment was not conducted. Second, we do not know whether providing active supplements to the more serious offenders in the placebo group would have resulted in a reduction in their antisocial behavior. Such a finding would have been particularly compelling, although this was probably not possible to do because it would have extended the study beyond the school year.

It is also important to emphasize that because children in this study were not selected according to core ADHD symptoms, and ADHD symptoms were not one of the outcome measures reported, the implications of these results for children with ADHD are unclear.  Thus, this particular study provides no information on whether vitamin and mineral deficiencies are related to ADHD symptoms and whether supplementation might be helpful with these symptoms. One could interpret these results as suggesting that vitamin and mineral supplementation may be helpful in minimizing the amount of antisocial behavior engaged in by children with ADHD, although the caveats noted above would certainly apply and a careful replication of this work is necessary.

The most well known and widely used of the objective measures for diagnosing ADHD are called Continuous Performance Tests (CPT). In a typical CPT, an individual sits in front of a computer terminal and is required to press (or not press) certain keys depending on the stimulus that flashes on the screen. The test typically lasts between 14 and 20 minutes and is purposely designed to be repetitive and boring. Good performance requires the child to sustain attention to a rather uninteresting task and to refrain from responding impulsively. Both errors of omission (failing to press the designated key in response to the target stimulus flashing) and errors of commission (pressing the key to a non-target stimulus), along with several other variables (such as reaction time and reaction time variability) are computed, and a child's score can be compared to how children of the same age and gender typically perform. Several studies have shown that children with ADHD perform poorly on these tests relative to non-ADHD children, and many clinicians routinely incorporate the CPT into their ADHD evaluation procedures.

An important problem with many studies using the CPT is that children with ADHD have been compared directly to children without any psychiatric disorder (non-clinical controls) rather than to children with an alternative psychiatric diagnosis (clinical controls). This is a serious limitation. For a test like the CPT to be useful, it must not only differentiate children with ADHD from "normal" children, but must also discriminate between children who have ADHD and children with other psychiatric disorders (such as anxiety, oppositional behavior, or depression) and children with learning difficulties. In most instances where a child is being evaluated, clinicians do not simply have to decide whether a child has ADHD or not, but are involved in the more challenging task of determining the best explanation for the difficulties that a child is displaying. Thus, clinicians are generally required to determine whether a child's symptoms reflect ADHD or some other type of problem (differential diagnosis).

How useful is the CPT for this purpose? This question was addressed in a recent study published in the Journal of Abnormal Child Psychology (McKee, R.A. et al., Vol. 28, 2000). Participants in this study included 100 children between the ages of 6 and 11 who had been consecutively referred for assessment of potential ADHD to an outpatient child mental health clinic over a 2-year period. Consistent with the pattern of referrals to psychiatric clinics in general, most of the children (79%) were males. All children received a thorough evaluation from a multi-disciplinary team that included interviews with parents and the child, the collection of standardized behavior rating scales from parents and teachers, and behavioral observations of the child. In addition, as part of the diagnostic work up, children were given the Conners' CPT and several other tests designed to evaluate children for reading disability.

The Conners' CPT is one of several commercially available CPT programs and is perhaps the most widely used by clinicians. It differs from other CPTs in that it requires the child to respond by pressing a designated key when all stimuli except the predetermined target are flashed on the screen. Other CPTs require the child to respond only when the target is flashed. The importance of this distinction is that the Conners' CPT places a greater emphasis on the child's ability to inhibit themselves from responding when they are not supposed to do so. Because a deficit in being able to inhibit behavior has been proposed as the core deficit in ADHD (Phil: Link to article on Barkley's theory of ADHD) the Conners' CPT is believed by many to be more useful in evaluating children for ADHD than other available CPT programs.

Based on the diagnostic work-up, children were divided into 4 groups: those with ADHD alone (n=42); those with only a reading disability (RD) (n=14); those with ADHD and RD (n=14); and those with another psychiatric diagnosis (n=32). The latter group consisted primarily of children diagnosed with oppositional defiant disorder (ODD) or conduct disorder (CD), although a variety of other conditions were also represented. In arriving at the diagnosis of ADHD, children's results on the Conners' CPT were not used. This was because the researchers wanted to compare the CPT results of children in the four groups who were diagnosed by standard clinical procedures. If CPT results were used in assigning the original diagnoses, these comparisons would be confounded.

RESULTS

Although a number of different scores are computed for the Conners' CPT, the authors focused on the overall index in their analyses. The overall index provides a global summary of how the child did on the test. According to the manual which accompanies the Conners' CPT, an overall index score above 11 is considered a conservative cutoff for attention problems, and children who score above this are considered to have "failed" the CPT. Thus, if the Conners' CPT is useful in helping clinicians make differential diagnostic assessments, one would expect children with ADHD to have significantly higher scores than children in the other groups. However, this was not the case. Children with ADHD + RD performed worse on the Conners' CPT than children in the other groups, but those with ADHD alone, RD alone, and the psychiatric controls did not differ. In fact, the data indicate that children with RD are more likely to fail the Conners' CPT than children with ADHD.

The authors also examined what percentage of children diagnosed with ADHD were considered to have failed the CPT. This was the case for a little over 50% of the children. This means that among those with a carefully established ADHD diagnosis, the likelihood of them performing above the clinical cut-off on the Conners' CPT was no better than chance.

Note: Because a number of other important variables besides the overall index are computed on the Conners' CPT, the authors repeated their analyses using several other variables as well. The pattern of results obtained for these other measures did not differ substantially from those reported above.

SUMMARY AND IMPLICATIONS

The primary conclusion from this study is that the Conners' CPT has questionable value as a diagnostic instrument. Although the authors recognize that supporters of the Conners' CPT may criticize their study for oversimplifying their interpretation of the instrument, they correctly point out that the inability of the overall index, as well as several other measures derived from the test, to distinguish ADHD subjects from clinical controls speaks to the test's need for considerable refinement. Although these data do not refute the potential utility of using instruments like the CPT as one component of a diagnostic evaluation, they certainly underscore the need to avoid using CPT results as a primary basis for deciding whether or not a child has ADHD. In many cases, doing so is likely to result in diagnostic errors that result in the selection of ill-advised treatments.

In general, parents and clinicians are advised to be cautious in attributing too much significance to how a child performs on a CPT or on other so-called "objective" indicators of ADHD symptoms. The American Academy of Pediatrics and the American Academy of Child and Adolescent Psychiatry recently published guidelines for the evaluation of ADHD and neither advocates that any such tests be routinely incorporated into ADHD evaluations, although recent research suggests that QEEG procedures for diagnosing ADHD show promise. Thus, for the time being, the use of careful clinical interviews that incorporate information from multiple sources will remain the cornerstone of a comprehensive ADHD evaluation.