Attention Research Update

March 2001

"Helping parents, professionals and educators stay informed about new research on ADHD"

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

In this issue...

New Outcome Reports From the MTA Study

Medication Treatment for ADHD: The Importance of Ongoing Monitoring


Most subscribers to Attention Research Update are probably familiar with the initial results of the NIMH collaborative multi-site, multi-modal treatment study of children with ADHD (the "MTA Study", for short).  This study is the largest and most comprehensive treatment study of ADHD ever conducted, involving 579 children between the ages of 7 and 9, each diagnosed with the combined subtype of ADHD (i.e. these children had both inattentive and hyperactive-impulsive symptoms). The study took place at 6 different sites around the country.

Children participating in the MTA study were randomly assigned to one of 4 different treatment conditions: combined treatment - a combination of carefully administered medication treatment and intensive behavioral intervention; medication management only; behavioral treatment only; and community care (i.e. these children received treatment as usual in their communities.)  Fourteen months after treatment began, the children were assessed on a variety of different outcome measures covering a variety of domains of functioning, including primary ADHD symptoms, oppositional behavior, parent-child relations, peer relations, self-esteem, anxiety, symptoms of emotional distress, and academic achievement.

As is typical of a study of this size, the initial results were complex, but can be reasonably summarized. First, children in all 4 groups were doing better at the end of the study than they were when treatment began.  Second, on some outcomes, combined treatment and medication management alone were superior to behavioral treatment or community care.  Finally, no statistically significant differences were found between combined treatment and medication management on any of the 19 individual outcomes examined (although there was some indication that children receiving combined treatment fared somewhat better).  For a comprehensive review of this initial set of findings, including a careful description of the different treatments provided in this study, click here.  Reviewing this extensive summary of the initial MTA publication will be helpful in considering the information that follows.

The absence of significant differences between the combined and medication management treatments has been widely interpreted to indicate that behavioral interventions do not provide any incremental benefit to well conducted medication treatment.  The MTA researchers themselves, however, never made this conclusion. In fact, two papers published in the February 2001 issue of the Journal of the American Academy of Child and Adolescent Psychiatry suggest that the combination of medication and treatment is preferable in many cases to medication alone.

In the first paper -- "Multimodal treatment of ADHD in the MTA: An alternative outcome analysis" (Conners et al., JAACAP, 40, 159-167) -- the authors take a different approach to examining treatment outcome than that used in the initial publication.  Rather than examine each outcome measure separately -- which was done initially to determine whether response to the 4 different treatments varied for specific outcomes -- the authors of this paper created a single composite outcome measure by averaging children's scores on the individual measures.  This composite measure can be thought of as an indicator of how each child was doing in general, across multiple domains of functioning.  Although this approach eliminates the possibility of comparing treatment outcomes in individual domains, comparing the composite outcome scores for children in the 4 groups enables one to obtain a more global perspective on the impact of the different treatments.

The second "re-analysis" of the treatment outcome data -- "Clinical relevance of the primary findings of the MTA: Success rates based on severity of ADHD and ODD symptoms at the end of treatment" (Swanson et al., JAACAP, 40, 168-179) -- takes a slightly different approach.  Rather than creating a composite outcome score that reflects how children were doing in multiple domains of functioning, the authors focus on parent and teacher ratings of core ADHD symptoms and symptoms of ODD (Oppositional Defiant Disorder).  And, the primary question examined is the degree to which each treatment resulted in children displaying levels of ADHD and ODD symptoms similar to what is typical for children without ADHD.  When this was true, treatment was considered to be successful.  This approach to examining the data (i.e. the percentage of children showing non-deviant levels of symptoms at the end of treatment) is especially instructive.


The results from analyses using the broad composite outcome described above are informative, and modify (somewhat) conclusions drawn from the initial study results.  The authors report that, when this composite was used to measure outcome, children receiving combined treatment did significantly better than children in any other group.  They did much better than the children who received community care or behavioral treatment alone, and modestly better than children whose treatment was restricted to careful medication management.  This latter result differs from previously published findings in which researchers did not find statistically significant advantages for combined treatment relative to medication management for individual outcomes.  When medication management alone was compared to behavioral treatment alone, medication treatment demonstrated a modest superiority. .

Results from the second paper help put these results in a somewhat clearer perspective.  Recall that in this paper, the authors focused on the percentage of children in each group who had average parent and teacher ratings of ADHD and ODD symptoms at the end of treatment -- i.e. symptom ratings were similar to those of children without ADHD.  Results from this analysis are shown below.

Combined     Medication      Behavioral        Community Care

68%                  56%                  34%                  25%

As the numbers indicate, over two-thirds of the children receiving combined treatment had "normalized" scores after 14 months, compared to only 1 in 4 treated in the community.  Normalized outcomes were more likely when treatment included the careful medication component (i.e. combined or medication) rather than intensive behavioral interventions alone.  Finally, combined treatment alone was modestly superior to medication management. Specifically, these data suggest that if children in the medication group had also received the MTA behavioral interventions, a greater number would have been in the "normal" range at the 14-month outcome assessment.

While these results demonstrate the dramatic improvements in core ADHD symptoms that are provided by effective treatment, it is also important to point out that such improvement is not always the case.  Even when state-of-the-art medication and behavioral interventions were combined, about one-third of children continued to show elevated levels of ADHD/ODD symptoms 14 months later, relative to non-ADHD peers.  Among those receiving the most carefully conducted medication treatment available, over 40% continued to show elevated levels of core ADHD symptoms.  This does not mean, of course, that these children were not benefiting from the treatment. It does indicate, however, that many continued to experience difficulties despite receiving the best possible care currently available.


In general, results from these two papers are consistent with the initial set of published findings.  As reported in the initial outcome paper, children in all 4 groups showed significant improvement. Children receiving careful medication treatment were doing somewhat better than children whose treatment was limited to intensive behavioral interventions.  What is evident here that was not initially reported, however, is that adding behavioral interventions to careful medication management yields significantly better outcomes when a composite outcome measure is used (i.e. study 1 above) or when one considers the likelihood of normalized scores on core ADHD/ODD symptoms (i.e. study 2).  Thus, the benefits of "multi-modal" treatment for ADHD are more clearly supported by these results.  (Note: It is important to remember that participants in the MTA study were restricted to those with the combined subtype of ADHD and included no children with inattentive symptoms only.  Thus, these results do not inform us about the efficacy of the different treatments for children with the inattentive subtype.)

In translating these findings to the issues faced by individual parents and clinicians, several things are noteworthy.  First, parents need to be vigilant about trying to obtain treatment for their child that is as close as possible to treatments used in the MTA study.  In regards to medication, this means a careful initial trial is necessary to determine the optimum dosage and medication for their child, followed by systematic monitoring to determine how their child is doing and make adjustments as indicated. The excellent results obtained by children treated with medication -- either alone or in combination with behavioral interventions -- points to the importance of this careful approach.

The results also indicate that the addition of well-designed and carefully implemented behavioral interventions could reasonably be expected to provide some modest additional benefit. One caveat to mention here, however, is that the behavioral interventions used in the MTA study would be difficult to duplicate in most communities.  Thus, it remains unclear whether the intensity of behavioral interventions that are more routinely available in this age of managed care would be similarly effective.

Finally, it should be noted that the treatments tested in the MTA study were limited to medication and comprehensive behavioral treatment.  As noted above, although these treatments were clearly shown to be helpful, many participants continued to experience important difficulty despite receiving state-of-the art care using these approaches.   This highlights the need for continued efforts to develop other types of interventions. Parents should be aware that promising results have been reported for a number of alternative treatments including dietary interventions and neurofeedback. Thus, should traditional approaches to treating ADHD (i.e. medication and/or behavioral therapy) prove to be insufficient for a particular child, there are other options that may prove fruitful.


Many experts agree that there are at least 2 important problems with how medication treatment for ADHD is provided to many children today.  First, when medication treatment is initiated, there is frequently no systematic procedure used to determine the optimum dosage for each individual child. Rather than collecting systematic ratings of children's functioning from parents and teachers on a range of different doses, physicians typically start a child on the lowest possible dose, obtain only anecdotal feedback on the child's behavior, and elect to maintain the first dosage that seems to be effective.  In many cases, this is unlikely to be the dosage that would provide the greatest improvement in a child's functioning.

A second important problem is, even when the maintenance dosage selected is appropriate, there is often little ongoing effort made to systematically monitor how a child is doing over time.  As a result, adjustments to medication -- or to any other type of treatment the child is receiving -- are not made, and symptoms that were being managed effectively at one time begin to significantly interfere once again with a child's functioning.  The extent to which ongoing monitoring of medication treatment is required is highlighted by another paper to come out of the MTA study (Vitello, B. et al., 2001. Methylphenidate dosage for children with ADHD over time under controlled conditions: Lessons from the MTA. J, 4ournal of the American Academy of Child and Adolescent Psychiatry, 40, 188-196.)

In the MTA study, 289 of the 7- to 9-year-old participants with ADHD were assigned to receive medication treatment -- either alone or in combination with intensive behavioral interventions.   To determine the best medication and dosage for these children, an initial titration trial was conducted in which children's functioning at school and home was compared when receiving different doses of medication (all children began with methylphenidate, the generic form of Ritalin, and a full range of doses was tested for each child) or a placebo.  By comparing parent and teacher ratings on each dosage to the placebo, the researchers sought to determine the optimum dosage for each child.

When no clear benefits were obtained for any dosage of methylphenidate tested, or if adverse side effects were apparent, a similar trial was conducted with another type of stimulant.  Using this strategy, a clearly optimal medication regime was identified for approximately 90% of the participants.  For the vast majority of children (79%), good results were obtained on at least one of the methylphenidate doses tested.  For another 11% satisfactory results were provided by the second stimulant used (dextroamphetamine, the generic version of Dexedrine.)  For other children, the response to placebo was so robust that they were not continued on medication after the titration procedure.

After medication treatment was implemented, careful monitoring was conducted of it's ongoing effectiveness over the next 13 months.  Each month, information was obtained from each child's parents and teacher about key ADHD symptoms, functioning at home and school, and possible side effects. If these reports indicated adequate control of symptoms and no side effects, the child continued on the current medication regime.  If reports indicated that symptoms had emerged that were causing impairment, or if possible side effects were reported, the medication regime was changed.  This could involve either increasing the dosage to obtain better symptom management or lowering the dosage in an effort to eliminate adverse effects.  Overall, this careful monitoring was intended to insure that no child remained on less than optimal treatment. This was done even though an extremely careful procedure had been used initially to determine the best dosage for each child. (Note: More detailed information on the medication treatment used in the MTA study can be found here.)

This procedure enabled the authors to examine several important issues related to medication treatment for ADHD.   First, how does the optimal medication and dosage identified by a careful initial titration trial compare to what is required over the course of treatment to maintain optimal management of ADHD symptoms?  Second, how soon into treatment do medication adjustments typically need to be made and how frequently do these tend to occur?  Finally, does gender or the presence of other conditions (i.e. Oppositional Defiant Disorder, Conduct Disorder, or anxiety disorders) impact the need for medication changes during ongoing treatment?


How does initial medication and dosage compare to what is required in ongoing treatment?

Of those children for whom an optimal medication treatment regime was identified by the initial titration procedure, only 17% continued on the same medication and dosage throughout the entire 13-month maintenance period.  The remaining children all experienced at least one change in drug or dosage during this period.

Of the children for whom methylphenidate was the medication on which maintenance began, 12% needed to be switched to a different drug during maintenance in order to promote optimum symptom management.   For the children who remained on methylphenidate, at the end of the maintenance period, 29% were on the same dose, 41% were on a higher dose, and 18% were on a lower dose.  Overall, daily dosage required increased from an average of 30.5 mg/day at the beginning of maintenance to an average of 34.4 mg/day by the end.  (Note:  In the MTA study, children received 3 doses per day so these amounts were divided across the 3 doses with the third dose being half the amount of the first 2.)  As you might expect, those who began on a low dosage were likely to have it increased. Doses for those starting on high doses (i.e. 35 mg/day) tended to decline.

Eleven children who had started on medication were no longer on medication at the end of the study, presumably because side effects had emerged.  Of the 32 placebo responders who did not begin the maintenance period on any med, all but 4 required medication at some point during maintenance.

How soon into treatment do medication adjustments typically need to be made and how frequently do these tend to occur?

Three months into the maintenance period, 56% of the children had already had their medication or dosage changed.  The average amount of time to the first dose change was between 4 and 5 months.  Across the entire maintenance period, the average number of changes required for each child was just over 2, but some children required as many as 10 medication adjustments.  Of the total medication changes made, 62% involved increasing the dosage of the current medication, 31% involved decreasing dosage, and only 7% involved changing types of medication.

Does gender or the presence of other conditions (i.e. oppositional defiant disorder (ODD), conduct disorder (CD), or anxiety disorders) impact the need for medication changes during ongoing treatment?

About 20% of the children in the MTA study were girls.  On average, girls remained on doses that were approximately 20% lower on a mg/kg basis than boys.  The time required to the first medication change, or the number of changes required over the maintenance period did not differ between boys and girls.  In addition, the presence of other disorders in addition to ADHD was not related to dosage at either end of titration or maintenance, the time to first change, or the number of changes required.


The results of this study make it clear that, even when extreme care is taken to determine the optimal medication treatment regime for a child with ADHD, changes in that regime are likely necessary to maximize the ongoing management of symptoms.  If one simply continues to maintain a child on the initial regime that seems best, it is very unlikely that the child's symptoms will continue to be managed as effectively as possible.

In my opinion, the importance of these results cannot be overstated. Without careful ongoing monitoring, and adjustments to treatment made when indicated, most children with ADHD are simply not going to do as well as they otherwise could.  Although the focus of this study was on the monitoring and adjustments required to medication treatment, it is important to emphasize that careful monitoring is essential regardless of what type of treatment, or combination of treatments, a child is receiving.  One simply cannot assume that an initially positive response to any treatment will be maintained consistently over time.  Instead, it is necessary to carefully track how a child is doing, and make modifications to existing treatment(s) when it becomes evident that symptoms are no longer being managed as effectively as they need to be.  Although some children will continue to have problems regardless of the modifications made (see article above), the likelihood of maximizing a child's ongoing success is certainly increased when this approach is followed.

(c) 2001 David Rabiner, Ph.D.

Information presented in Attention Research Update is for informational purposes only, and is not a substitute for professional medical advice.