To meta-analyze or not to meta-analyze? A combined meta-analysis of N-of-1 trial data with RCT data on amphetamines and methylphenidate for pediatric ADHD

Salima Punja, Christopher H. Schmid, Lisa Hartling, Liana Urichuk, Catherine Jane Nikles, Sunita Vohra

To assess how the inclusion of N-of-1 trial data into randomized controlled trial (RCT) meta-analyses impacts the magnitude and precision of yielded treatment effects, using amphetamines and methylphenidate for pediatric ADHD as a model. We combined the N-of-1 and RCT data generated from previously conducted systematic reviews using parent and teacher ratings of hyperactivity/impulsivity as the outcome. Data was combined using standardized mean differences assuming a random effects model. The amphetamine and methylphenidate evidence were synthesized separately. We found that the inclusion of N-of-1 trial data in the meta-analysis impacted both magnitude and precision. The addition of the N-of-1 trial data narrowed the confidence intervals in 3 of the 4 comparisons as compared to the treatment effect yielded by RCT-only data. Furthermore, the addition of N-of-1 trials changed the overall treatment effects yielded by the RCT-only meta-analyses from statistically non-significant to statistically significant in one of the four outcomes. If the overall goal of a meta-analysis is to synthesize all available evidence on a given topic, then N-of-1 trials should be included. This study shows it is possible combine N-of-1 trial data with RCT data as well as the potential merits of this approach.