Natural Solutions Radio header image


by David Rabiner, Ph.D.
Senior Research Scientist
Duke University

The MTA Study is the largest treatment study of ADHD every conducted.  Participants were 579 8-12-year-old children diagnosed with the combined type of ADHD who were randomly assigned to receive 1 of 4 different treatments: intensive medication management alone (MM); intensive behavior therapy alone (BT), the combination of medication management and behavior therapy (Comb), or community care (CC; children assigned to community care received whatever community-based treatments their parents selected). A thorough description of these different interventions is available at

Initial results from this landmark study were published in 1999 and suggested that children who received careful medication management (i.e. those in the MM and Comb conditions) did better than those who did not (i.e. those in the BT and CC conditions), and differed only slightly from each other. This was not true for all the different treatment outcomes considered individually, but was evident when outcomes were combined into a single overall indicator of treatment response.

And, when the investigators defined an "excellent response" as one where the child had parent and teacher ratings of core ADHD symptoms plus oppositional behavior that fell in the average range, the following percentage of children in each group were found to be excellent responders:

Combined - 68%; MM - 56%; BR - 34%; CC - 25%

These figures illustrate that those in the combined group were most likely to be "normalized", and that there is also a clear break between children who received MM alone compared to those getting intensive behavioral treatment or standard community care.  (For a complete description of this study, go to )

As informative as these results are for the field, an important issue not examined in these earlier papers is why some children were excellent responders while others were not.  For example, although 68% of children in the combined condition had an excellent response, nearly 1/3 did not. Similarly, 56% of children receiving careful medication management alone had an excellent response but nearly 1/2 did not.

Why did some children do well in each of the treatment conditions studied in MTA  - even in the CC group, 25% were excellent responders - while others had less positive responses?  Is it possible to identify pre-treatment characteristics of children and/or their parents that differentiate which child is likely to derive greater and lesser benefit from the different interventions studied?  This would be quite helpful, as it could provide more precise knowledge about whether a particular child is likely to derive greater benefit from one treatment(s) vs. another.

Identifying child and parent characteristics that influenced children's response to treatment in the MTA study was the topic of a recently published paper titled "Which treatment for whom for ADHD? Moderators of treatment response in the MTA" (Owens et al., 2003. Journal of Consulting and Clinical Psychology, 71, 540-552).  Six child characteristics - i.e. gender, prior medication use, the presence of co-occurring oppositional behavior/aggression, IQ, and initial severity of ADHD symptoms - and 3 family characteristics - i.e. maternal education, amount of depressive symptoms in the child's primary caregiver, and whether or not the family received public assistance - were examined to learn whether they moderated (i.e. influenced) the likelihood that the child would experience an excellent treatment response.  Excellent response was defined as the child having parent and teacher ratings of ADHD and Oppositional Defiant Disorder (ODD) symptoms that were not considered problematic 14 months after treatment began (i.e. the average rating for these symptoms on a 4-point scale averaged 1 or below.  This rating corresponded to reports that the symptoms were present "just a little", as opposed to "pretty much" or "very much").  


As the first step in their analysis, the authors examined whether the type of treatment children received (i.e. MM, Comb, BT, or CC) influenced the likelihood of an excellent response.  In considering this, the two treatments that included careful medication management (i.e. MM and Comb) were compared to the treatments that did not (i.e. BT and CC).  Consistent with results that have been previously reported, 62% of children in the MM or Comb treatment had an excellent response compared to only 30% who received BT or CC.  Thus, when careful medication management was part of a child's treatment, reductions in ADHD/ODD symptoms to the normal range was twice as likely.

(Note: Many children treated in the community received medication as part of their treatment.  Medication treatment in the community was less effective than medication treatment through the study, however, perhaps because the careful procedures used in the study are not typically practiced by community physicians.  For a discussion of how to maximize the benefits of medication treatment, visit ).

In subsequent steps, the authors considered whether any of the 6 child and 3 family variables noted above were related to the probability of an excellent response.  For the 30% of children in the BT or CC groups who had been excellent responders, no such factors were found. Thus, none of the child or family variables that were examined predicted which children treated with BT or CC would do especially well in treatment.

For children in the MM/Comb groups, however, several variables influenced the probability of an excellent response.  The most important variable was parental depression.  When the primary caregiver scored below the cut-off for "mild depression" on a self-report measure of depressive symptoms, nearly 70% of children were excellent responders to MM or Comb treatment.  In contrast, when parents' depressive symptoms were above this threshold, an excellent response was obtained in only 45% of cases.

Not surprisingly, the likelihood of an excellent response was also related to the initial severity of children's symptoms.  Regardless of parent's level of depression, higher initial levels of ADHD/ODD symptoms reduced the likelihood that children in the MM/Comb groups would have an excellent response by nearly 50%.  Thus, even when children's treatment is consistent with current "best practice" guidelines, the majority of children whose initial ADHD and ODD symptoms are severe will continue to display experience ongoing difficulty from their symptoms.

The final variable to moderate treatment outcome was the child's IQ, but this was only true within a particular group - those whose primary caregiver had high levels of depressive symptoms and whose pre-treatment level of ADHD/ODD symptoms was high.  Within this high-risk group, only 10% of children with below average IQ's had an excellent response compared to 48% of children with IQ's that were average or above.  


Among the important implications of these results are the following:

For reducing core ADHD symptoms, treatments that include careful medication management are likely to be most effective.

As discussed above, children who received careful medication treatment - whether alone or in combination with intensive behavior therapy - were over twice as likely as other children to be excellent responders.

This does not mean, of course, that medication should be the treatment of choice for every child with ADHD, or that behavioral treatment is not also effective.  It does indicate, however, that at this point in our knowledge of how to treat ADHD, medication is generally the most effective intervention for alleviating core ADHD symptoms.  Whether it is also superior in alleviating other difficulties that often accompany ADHD - e.g. academic problems, peer difficulties, parent-child relationship problems - was not directly examined in this study and remains to be determined.  

How depressed a parent is at the time their child begins treatment may be an important factor in treatment success.

This finding highlights that aspects of the child's family environment can be important to treatment success.  This is a far cry from blaming parents for their child's disorder - as still occurs far too often - but does suggest that parent/family factors can moderate how children with ADHD develop over time.  It also suggests that alleviating parental depression may not only be helpful to parents, but might also increase the likelihood of an excellent treatment outcome for children with ADHD.  Screening parents for depression when a child begins ADHD treatment may thus have important clinical utility.

Even the best treatment currently available will fail to normalize the symptoms of many children with severe ADHD.

As discussed above, children with severe symptoms before treatment began were less likely to be "excellent responders".  Thus, even though they may have experienced important benefits from treatment, they continued to display ADHD and ODD symptoms that were above the normal range.  This finding highlights what a difficult condition ADHD can be to treat effectively, and the importance of providing ongoing support and assistance to children with ADHD and their families.

Although effective treatments for ADHD are available, there is a pressing need to develop new ways for effectively treating this disorder.

As results from this study make clear, even those interventions for ADHD that currently enjoy the greatest research support - i.e. stimulant medication treatment and behavior therapy - fail to normalize core symptoms in a large percentage of diagnosed children.  There is thus a pressing need to examine promising alternative interventions that can be helpful for children who do not obtained the desired response from these approaches.

As has been reviewed in prior issues of Attention Research Update, several promising interventions for ADHD have appeared in the literature.  Among those interventions for which encouraging results have been obtained are neurofeedback and the Interactive Metronome.  Additional research is needed, however, before the effectiveness of these interventions for ADHD is conclusively established.

As such research is published, I will certainly include it in the newsletter.

Sponsors' Message from Play Attention:


We understand the struggle that some children have paying attention and the frustration of caring for a bright child whose attention problems keep them from learning up to their potential. That's why we developed Play Attention.

Play Attention is used in hundreds of school systems, learning centers, homes, and psychologists' offices to help children develop their attention and learning skills. Developed by an educator and successfully tested in classrooms nationwide, Play Attention is an interactive computerized learning system that  incorporates advanced NASA feedback technology and the latest brain research to increase attention and reduce impulsive behaviors.  It is fun, easy to use, and can play an important role in helping your child experience greater success at school.

Please click here to learn more about Play Attention and to receive a free interactive CD that will show you how Play Attention can help your child succeed

Thanks again for your ongoing interest in the newsletter. I hope you enjoyed the above article and found it to be useful to you.

The next issue of Attention Research Update will be sent out to you in approximately 3 weeks.


David Rabiner, Ph.D.
Senior Research Scientist
Duke University

(c) 2003 David Rabiner, Ph.D.

Copyright Issues?