The NIMH Collaborative Multisite Multimodal Treatment Study of Children with
ADHD, known as the MTA Study for short, is the largest ADHD treatment study
ever conducted. In this landmark study, 579 children carefully diagnosed
with ADHD were randomly assigned to one of four conditions:
- carefully managed and monitored medication treatment;
- intensive behavior therapy;
- the combination of careful medication treatment and behavior therapy; and;
- routine community care.
The goal was to learn how state-of-the-art medication and behavioral treatments
compared to each other, to the combination of these approaches, and to the
care that children routinely received in community settings. Treatments
delivered by study investigators lasted 14 months after which children received
whatever treatment(s) their parents elected. Detailed descriptions
of the MTA treatments can be found at http://www.helpforadd.com/yr2000/janissue.htm
Outcome data at 14 months indicated that children in all groups showed improvement
relative to baseline. Although the results are complicated and difficult
to briefly summarize, children receiving intensive medication treatment -
either with or without behavior therapy - showed significantly greater improvement
in ADHD and oppositional defiant disorder (ODD) symptoms compared to children
receiving community care or intensive behavior therapy alone. There
was also some evidence that combined treatment had benefits beyond medication
alone on certain outcomes including anxiety symptoms, parent-child relations,
teacher-rated social skills, and reading achievement, but these were not
especially strong effects.
Approximately 10 months after this initial outcome evaluation, the initial
advantage of combined treatment and medication treatment relative to behavior
therapy alone and community care was still evident, but had dissipated by
Detailed reports of these initial outcome studies can be found at:
The next wave of outcome data was collected 36 months after the study began,
which was 22 months after MTA delivered treatments had ended. Although
all 4 groups continued to show improvement relative to baseline, there were
no longer significant group differences in ADHD/ODD symptoms or other aspects
of children's functioning. Thus, there was no indication that carefully
monitored medication treatment, behavior therapy, or their combination had
benefits that persisted significantly beyond when these treatments ended.
Across all groups, between 45 and 55% of children continued to meet diagnostic
criteria for ADHD. A detailed discussion of findings from this study
can be found at http://www.helpforadd.com/2007/august.htm
At this follow-up, researchers identified 3 classes of children who showed
distinctively different trajectories to their symptoms between baseline and
36 months, irrespective of which group they were originally assigned to.
Class 1 (34% of the sample) showed a gradual improvement over time.
Class 2 (52%) showed a significantly larger initial improvement that was
maintained over time and Class 3 (14%) had returned to pretreatment symptom
levels following an initially positive response. Children in Class
2 were those who had been functioning better at baseline and who came from
more economically advantaged families.
Recently, the MTA Study group published 8-year follow-up data [MTA Study
Group, 2009. The MTA at 8 years: Prospective follow-up of children treated
for combined-type ADHD in a multisite study. Journal of the American Academy of Child and
Adolescent Psychiatry, 48
, 484-500.] At this follow-up, data
was collected on 436 of the original 579 participants who now ranged in age
from 13 to 18. The key study goals included the following:
1. To learn whether any treatment group differences evident at the 14 and
24 month follow-up may have re-emerged now that participants were adolescents.
2. To learn whether the different trajectories identified at the prior follow-up
predicted outcomes in subsequent years, and,
3. To compare the functioning of MTA adolescents relative to their non-ADHD
peers on a variety of important outcomes. Twenty-one different outcomes
were examined, including core ADHD symptoms, ODD symptoms, antisocial behavior,
various indices of academic achievement, social functioning, depression,
anxiety, and other psychiatric problems. The comparison peers were
289 children recruited 24 months into the study from the same schools study
1. Did any benefits of MTA treatments re-emerge
The answer to this question is an unqualified no. Of the over 20 outcomes
examined not a single significant difference based on initial random assignment
was found. The general pattern was for any initial group difference
to converge over time and for children in all groups to maintain improvement
relative to baseline.
2. Was continued medication use associated
with better outcomes?
Past year medication use at the 8-year follow-up was not associated with
better outcomes in any area except for math achievement. Thus, there
was scant evidence that medication treatment - at least as provided in community
settings - was associated with better adolescent outcomes.
It is interesting to note that only 32.5% of participants had received medication
over 50% of days during the past year compared nearly twice this figure,
63.3%, at the initial 14-month follow-up. Thus, there was a steady
decline in medication use over time.
3. Did the classes of symptom trajectories
identified at 36 months predict outcomes in adolescence?
In general, yes. Children in Class 2 - those with the best initial
treatment response - continued to fare better over time than children in
classes 1 or 3. Thus, differences in children's initial response to
treatment - regardless of whether that treatment is medication or behavior
therapy - tended to predict longer term outcomes as well.
4. How were children with ADHD faring in
comparison to peers without ADHD?
Although children with ADHD had shown significant improvement since their
original diagnosis, they were still impaired in most areas relative to peers.
In fact, for 19 of the 21 specific outcomes tested, adolescents in the ADHD
group were doing significantly worse. This was true even though only
about 30% of MTA participants still met full diagnostic criteria for ADHD.
About 25% of MTA youth were displaying clinically significant antisocial
behavior (vs. just 5 % of comparison youth) with roughly 27% having been
arrested at least (compared to 14% of peers). MTA youth had significantly
lower grades, were about twice as likely to have repeated a grade (37% vs.
19%), and score about 0.5 standard deviations lower on academic achievement
- Summary and Implications
The key findings from this study are as follows:
1. Any initial benefits of intensively delivered medication treatment and
behavior therapy relative to routine community care do not persist significantly
beyond when those treatments were administered.
2. The trajectory of ADHD symptoms in childhood - and how children initially
respond to treatment - is a good predictor of adolescent outcomes.
Specifically, children whose ADHD symptoms show the best initial response
are likely to maintain better outcomes in many areas over time.
3. Despite overall gains relative to baseline for all study groups, and even
though only 30% of MTA youth continued to meet ADHD diagnostic criteria in
adolescence, they were still functioning significantly less well than comparison
peers. Thus, the adverse long-term outcome of childhood ADHD was clearly
What are the important clinical implications of these findings? Some
have suggested that because no long-term benefit of medication treatment
was found, there is little value in this treatment. This does not seem
like a reasonable conclusion, however, as benefits of carefully monitored
medication treatment were initially found but those benefits do not last
much beyond when this treatment is in place. It is possible that if
careful oversight of medication treatment had been maintained across all
8 years, continued advantages would have been found. However, as the
authors note, this is mere speculation and it may not even be possible to
maintain adherence to a carefully monitored medication treatment program
over this time frame.
It is also evident that even though core ADHD symptoms may remit over development
to the point that full diagnostic criteria for ADHD are no longer met, many
youth will continue to struggle in multiple areas. This highlights
the importance of carefully attending to a youth's overall functioning -
and not just ADHD symptoms - in determining the need for ongoing treatment,
and to make sure that treatments delivered specifically address the areas
where functioning is compromised.
Perhaps the most important implication - as the authors emphasize - is the
extent to which findings highlight the crucial need for developing treatments
that are "...efficacious, accessible, and lasting for high school-aged youths
with ADHD and their parents". They note that teenagers with ADHD are
not easy to treat, even though the numerous outcome differences between MTA
youth and comparison peers highlights the frequent need for ongoing treatment.
While medication may be one component of effective treatment, adherence to
medication treatment in adolescence is notoriously poor and other approaches
clearly need to be developed. The authors suggest that motivational
approaches to "...encourage adolescent participation in nonpharmacological
interventions...and that also address continued family and school involvement."
Clearly, it is time to move beyond the strong emphasis on studies of medication
treatment and behavior therapy for ADHD, as maintaining adherence to behavior
therapy in multiple settings is also exceedingly difficult. In fact, one
of the many very important contributions made by the MTA Study is highlighting
the limitations of medication and behavioral treatments for ADHD, as this
will hopefully expand serious scientific study of alternative treatment approaches.
Fortunately, as recently published studies of neurofeedback, working memory
training, and other approaches suggest, this is already beginning to happen.
Lets hope it continues.