A post hoc comparison favored lisdexamfetamine over extended-release mixed amphetamine salts when improvement was restricted to the highest clinician-rating category. The parent trial’s primary outcome did not show a difference between the active treatments.
López and colleagues analyzed a randomized, double-blind crossover classroom trial. Fifty-two children aged 6–12 were randomized and 50 completed it. The primary outcome assessed classroom deportment; a secondary clinician scale rated global improvement.
The reanalysis separated a rating of very much improved from every other category. This narrower cutoff asked a different question from the common combination of much improved and very much improved.
The combined favorable categories occurred in 74% under lisdexamfetamine and 72% under mixed amphetamine salts. The highest category alone occurred in 32% and 16%, respectively. The paired comparison for that strict category yielded p = 0.0386.
That post hoc finding should not replace the primary endpoint result or be presented as broad superiority across symptom and performance measures. A cutoff chosen after inspecting category distributions can emphasize a difference that was not the trial’s primary question.
The trial was small and not prospectively powered to establish active-treatment differences of this kind. Global clinical impressions and repeated classroom observations also differ in scope and assessment timing.
The industry-funded analysis offers a hypothesis about depth of rated improvement. It does not prove that one formulation is always better, establish a dose-equivalence rule, or identify an individual impairment threshold. A sound comparison states both the narrow positive result and the parent trial’s primary finding.
See what a formulation-level classroom comparison establishes.
Physician perception of clinical improvement in children with attention-deficit/hyperactivity disorder: a post hoc comparison of lisdexamfetamine dimesylate and mixed amphetamine salts extended release in a crossover analog classroom study. doi:10.2147/NDT.S17002.