Trial quality, bias may diminish value of methylphenidate for ADHD

A recent clinical review in JAMA reported that evidence supporting the use of methylphenidate among adolescents for treatment of attention-deficit/hyperactivity disorder is of very low quality and should be reconsidered for future therapy guidelines.

“Methylphenidate was associated with improved teacher-reported symptoms of ADHD and general behavior and parent-reported quality of life,” Ole J. Storebø, PhD, of the child and adolescent psychiatric department at Region Zealand Psychiatry in Denmark, and colleagues wrote. “However, given the risk of bias in the included trials, and the low quality of evidence, the magnitude of improvement is uncertain.”

The researchers conducted a Cochrane review of all randomized controlled trials of methylphenidate, consisting of 38 parallel-group trials and 147 cross-over trials conducted between 1981 and 2014. There were 12,245 children aged 3 to 21 years (mean age, 9.7 years) included in these studies. Placebos were used in 175 of the trials; the other 10 had no intervention.

Results of the meta-analysis showed that methylphenidate improved teacher-rated ADHD symptoms among adolescents. Other outcomes included improvements in general behavior and parent-reported quality of life. The researchers also found that there was an increase in nonsevere adverse outcomes, including trouble sleeping and lack of appetite.

Storebø and colleagues, however, applied the Grading of Recommendations Assessment, Development and Evaluation system and determined that all of this evidence was of very low quality, due to a high risk for bias, imprecision and heterogeneity. The researchers determined each study’s integrity by identifying seven sources of bias, including how each was randomized, how allocation was concealed, how participants and personnel were masked, how outcome assessments were masked, completion of data, selective reporting and vested interest bias.

“Findings from our review suggest that the recommendations in these guidelines should be re-evaluated because of the very low quality of evidence in the included trials,” Storebø and colleagues wrote. “Randomized nocebo tablet (active placebo)-controlled, high-quality clinical trials with longer follow-up are necessary.”

In a related editorial, Philip Shaw, MD, PhD, of the National Human Genome Research Institute, wrote that while Storebø and colleagues’ review was conducted with commendable attention to detail, their conclusion may be overzealous. Shaw said many studies within the review were deemed lacking based on the presence of any form of bias; however, he noted that all bias is not equally damaging. Shaw recommended a more nuanced approach to determining and eliminating studies based on the severity of their bias.

“Sometimes in medicine, the best available data are imperfect,” Shaw wrote. “Such imperfections do not render the data unusable; rather, the limitations can be weighed by physicians and other health care professionals, and by families as they decide how best to help a child struggling with ADHD. Psychostimulants improve ADHD symptoms and quality of life. This meta-analysis highlights the complexities in quantifying this benefit.”

Shaw agreed with Storebø and colleagues’ recommendation for better research into the use of psychostimulants among adolescents.

“It is preferable to focus on the development of interventions that retain the benefits of methylphenidate without the associated adverse events,” Shaw wrote. “The authors’ principal recommendation is to conduct larger, longer trials that minimize bias and to report all results in a comprehensive and accessible manner. Few would disagree.”

Expert opinion mixed

According to further research published simultaneously in JAMA, current evidence supports the use of extended-release methylphenidate and other stimulant medications, such as amphetamine formulations, atomoxetine and guanfacine, among adolescents.

“Evidence suggests that both pharmacological and psychosocial treatments for adolescents with ADHD are associated with improvements in ADHD symptoms and related academic or organizational, behavioral or emotional, and functional impairments,” Eugenia Chan, MD, MPH, of the division of developmental medicine at Boston Children’s Hospital, and colleagues wrote. “Results of these adolescent-only studies are generally consistent with those conducted in other age groups, although the effective dose in adolescents may be higher than for school-aged children.”

Chan and colleagues conducted a systematic review of 16 randomized clinical trials and one meta-analysis conducted from 1999 through Jan. 31. Studies included randomized trials of ADHD treatments among 2,668 adolescents aged 12 to 18 years. The researchers evaluated the quality of psychosocial treatment and pharmacological treatment for each study.

The evidence of efficacy was strongest for methylphenidate, amphetamine and atomoxetine treatments, the investigators wrote. They also stated that evidence was not as strong for the efficacy of guanfacine or clonidine, despite FDA approval for their use in adolescents.

Chan and colleagues agreed that future research should address the deficit in data regarding treatment effectiveness and adverse outcomes among adolescents.

“Future ADHD research must include study participants who are in the adolescent age range,” they wrote. “Additional treatment studies in adolescents, including combined pharmacological and psychosocial treatments, are needed.” – by David Costill

References:

Chan E, et al. JAMA. 2016;doi:10.1001/jama.2016.5453.
Shaw P. JAMA. 2016;doi:10.1001/jama.2016.3427.
Storebø OJ, et al. JAMA. 2016;doi:10.1001/jama.2016.3611.

Disclosures: Storebø and Shaw report no relevant financial disclosures; Chan reports receiving peer reviewer royalties from UpToDate for ADHD-related content. See the full study for a list of all other authors’ relevant financial disclosures.