Lancet Psychiatry
Optimal doses for ADHD meds mapped
Clinical Takeaway: Don't settle for subtherapeutic doses when symptoms remain uncontrolled, but don't push past licensed maximums expecting more benefit either. Group-level efficacy generally plateaus at or near approved doses while side effects keep climbing.
Most ADHD guidelines say little about how to dose the medications they recommend, leaving clinicians to navigate titration largely on their own. Real-world data show many children stay on starting doses that may be too low to work, while others get pushed above licensed limits with unclear benefit. This analysis aimed to map where the dose-benefit curve actually levels off and where harms start to outweigh gains.
In children and adolescents, methylphenidate efficacy peaked around 45 mg per day, amphetamines around 25 mg per day, and guanfacine around 4 mg per day, with no clear added benefit above those doses. In adults, amphetamine efficacy plateaued around 50 mg per day, while methylphenidate showed continued gains across the dose range studied, though with diminishing returns approaching 50 mg per day. Atomoxetine and modafinil did not show clear dose-response patterns, partly because of sparse data.
Discontinuation due to side effects climbed steadily with amphetamine doses above roughly 25 mg per day in children and 50 mg per day in adults, and above 50 mg per day for methylphenidate in adults. At 90 mg per day of methylphenidate in adults, the estimated dropout risk from adverse events reached 10%, compared with about 7% at the 60 mg per day licensed maximum, while the efficacy gain was modest. Guanfacine dropout risk in children rose with doses up to about 4 mg per day, the point where efficacy also peaks.
Researchers analyzed 113 randomized controlled trials including 25,154 participants, 68 trials in children and adolescents and 45 in adults, using a Bayesian dose-effect network meta-analysis.
"Our findings suggest that clinicians should avoid using doses that are too low to be effective. If symptoms are not well controlled, the dosage may need to be increased. We also found no evidence that going beyond the licensed maximum doses improves average effectiveness, and higher doses are usually linked to more side effects. However, our results derive from group averages. Specific individuals with ADHD may benefit from and tolerate well unlicensed doses," said first author Mikail Nourredine, PhD, of the University of Lyon.
Source: Nourredine M. Lancet Psychiatry. 2026 May 14. Pharmacological interventions for ADHD: a systematic review and dose-effect network meta-analysis