Provide feedbackProvide feedback

« Back

MTA Cooperative Group uses 8-year MTA follow-up data to assess long-term outcomes in children with ADHD

Monday, October 29 2007 | Comments
Evidence Grade 0 What's This?
Initial data from the Multimodal Treatment Study of Children with ADHD (MTA) provided information about the efficacy of different treatments and treatment combinations in pediatric attention-deficit/hyperactivity disorder. Now, with 8-year follow-up data available, the MTA offers insight into the long-term outcomes of ADHD, including prognosis, the impact of medication on a child's physiology, and the risk of substance use and delinquency, according to members of the MTA Cooperative Group.

Beginning in 1993, 579 children (aged 7 to 10 years; 80% boys) with ADHD were randomized to 1 of 4 treatment groups: medication management (primarily methylphenidate), behavioral therapy, combination treatment (medication and behavioral therapy), or community care as usual, with treatment in each group lasting 14 months.

At the end of the randomized treatment phase, there were significant between-group differences that suggested greater efficacy with medication management or combination therapy than with behavioral therapy or community care, according to Dr. Eugene Arnold, a faculty member in the department of psychiatry at Ohio State University Medical Center. These differences persisted at 24 months (10 months after the end of randomized treatment), but they were smaller. By 36 months, the between-group differences had largely dissipated.
From 36 months to 8 years, initial improvement was generally maintained, Dr. Arnold noted, but the children did not reach the level of function observed in a local normative comparison group (LNCG). The LNCG was recruited at the 2-year follow-up assessment and initially included 289 children without diagnosed ADHD.

At 8 years, 30% of the children in the MTA cohort met diagnostic criteria for ADHD (compared with 100% at baseline). This decline, Dr. Arnold suggested, in part reflects the general improvement of the group but also reveals the lack of developmental criteria for ADHD.

To more closely evaluate clinical outcomes in the MTA cohort, Dr. Arnold and colleagues used growth mixture modeling to evaluate the 36-month MTA data. In so doing, they identified 3 classes of patients that differed according to their symptom course from baseline to 3 years. Class 1 (approximately one-third of the sample) exhibited gradual improvement with time. Class 2 (approximately 50% of the sample) exhibited an excellent response to treatment at 14 months and maintained this response at 3 years; this group contained a preponderance of children initially randomized to combination treatment and medication management. Class 3 (approximately 14% of the sample) had an immediately good response to treatment and was clinically similar to class 2 at 14 months but exhibited a decline after the end of randomized treatment.

Dr. Arnold noted that these 3 groups also seemed to follow distinct clinical paths from 36 months to 8 years: Class 1 continued to improve with time, class 3 regressed and then started to recover, and class 2 continued to have the best outcome. However, none of the groups achieved a level of function comparable to that observed among the children in the LNCG.

The 8-year MTA data have also provided insight into the effects of long-term medication use on the physiology of children with ADHD, according to James Swanson, a professor of pediatrics at the University of California, Irvine. Specifically, among 275 children with complete data for physiological outcomes, the 8-year follow-up shows minimal (if any) effects of consistent medication use on blood pressure and heart rate. Data also suggest that consistent medication use is associated with an initial decline in z-weight and z-body mass index followed by a rebound of both of these variables, but the ultimate long-term effects on weight and BMI are still a bit uncertain.

With respect to height, the data described in prior reports from the 14-month assessment (based on subgroups randomized to treatment) and the 24-month assessment (based on naturalistic subgroups established at that point) indicated initial suppression of height velocity with medication. However, in the subgroups that emerged based on choices about medication during the entire 8-year follow-up, differences between the consistently medicated group and the never-medicated group were present at baseline; these differences persisted at 96 months (representing an overall difference of approximately 4.5 cm).

Lastly, results of the long-term naturalistic follow-up of the MTA cohort have the potential to inform parents, patients, and physicians about the risk of substance use and delinquency associated with ADHD and its treatments, according to Brooke Molina, a faculty member of the department of psychiatry at the University of Pittsburgh.

At the 3-year, 6-year, and 8-year assessments, Molina noted, the children in the MTA group had significantly higher rates of both substance use and delinquency relative to the children in the LNCG.

In analyses that accounted for medication use in the previous year, data did not reveal either a protective or predisposing effect of stimulants on substance use. Interestingly, though, the 36-month data suggested that behavioral therapy might have a protective effect against stimulant use. Specifically, the children originally randomized to 14 months of either behavioral therapy or combination therapy reported significantly less substance use at 3 years (8.1% and 8.5%, respectively) relative to the children randomized to medication only (17.1%) or community care (14%). These between-group differences were not apparent at the 8-year assessment.

Dr. Arnold noted that the MTA study is ongoing. Ten-year follow-up data are being processed, and the 12-year follow-up assessment is under way.

By Courtneay Parsons

Print  |  E-mail


Be the first to write a comment for this article!

You must be logged in to post a comment.