An executive-function responder can mean someone crossed a specified questionnaire threshold, rather than someone demonstrated normal performance on every executive task. Brown and colleagues’ analysis of extended-release mixed amphetamine salts is useful because it makes the response definitions explicit. The definitions support a particular interpretation of reported improvement, not a universal restoration claim.
These were additional analyses of two existing trials
The paper appeared online in 2020 and in a 2022 journal issue. It reanalyzed two randomized, placebo-controlled adult ADHD trials of oral SHP465 mixed amphetamine salts. One used dose optimization over seven weeks, with a 12.5–75 mg range; the other lasted six weeks and assigned 25, 50 or 75 mg doses with titration. The intent-to-treat populations were 268 and 405, respectively.
The original primary endpoint was an ADHD symptom rating. The Brown Attention-Deficit Disorder Scale, or BADDS, was secondary, and the responder analyses reported here were post hoc. These publications do not create two additional independent trials beyond the original cohorts. For a different endpoint from those cohorts, see how ADHD remission definitions change trial comparisons.
A questionnaire about everyday difficulties
BADDS has 40 items covering reported problems with organization, attention, effort, affective interference and working memory or recall. Frequency ratings generate a total from 0 to 120; higher scores indicate more difficulty. Although administered by a clinician or trained person, the measure is a self-report scale.
The analysis used two definitions: an endpoint total below 50, interpreted as no clinically impaired executive function by that threshold; and an endpoint score below the treatment group’s pretreatment 90% confidence-interval range, used as an index of improvement. Crossing a fixed severity threshold and improving relative to a baseline reference are different achievements. Neither is an individualized reliable-change calculation established by this paper.
The available BADDS denominators were smaller
In the dose-optimization study, 54/129 active-treatment participants, or 41.9%, had scores below 50 at endpoint, compared with 24/125 placebo participants, or 19.2%. The advertised intent-to-treat counts were 136 and 132. The responder figure therefore should not be described as 41.9% of all randomized active-treatment participants.
In the fixed-dose study, 152/293 participants across active doses, or 51.9%, crossed the threshold, compared with 16/96, or 16.7%, on placebo. Its intent-to-treat counts were 302 and 103. Endpoint meant the last nonmissing postbaseline assessment, which could differ from completing the full scheduled treatment period.
Improvement did not require crossing below 50
Using the baseline-range definition, the reported improvement percentages were 57.4% versus 29.6% in the dose-optimization study and 70.6% versus 32.3% in the fixed-dose study. These higher percentages do not contradict the fixed-threshold results: a person can improve while retaining substantial reported difficulty.
Conversely, a score below 50 is a classification based on the scale, not proof that all relevant abilities are normal. The study did not establish restored employment performance, safe driving, or normal results on every laboratory executive-function measure. The authors discussed rating scales and neuropsychological tests as different approaches to evaluating executive function.
How much confidence should the responder analyses carry?
The comparisons favored active treatment, but the studies were not powered for these post hoc analyses. Reported p values were nominal and descriptive. The samples were primarily White and had combined-presentation ADHD, and clinically important comorbidities were excluded.
The results support improvement in measured self-reported executive difficulties in selected trial participants. They do not justify replacing a comprehensive individual functional assessment with one response percentage, or treating a questionnaire cutoff as a toxicological impairment threshold.
Source
Brown TE, Chen J, Robertson B. Improved executive function in adults diagnosed with attention-deficit/hyperactivity disorder as measured by the Brown Attention-Deficit Disorder Scale following treatment with SHP465 mixed amphetamine salts extended-release: post hoc analyses from 2 randomized, placebo-controlled studies. Journal of Attention Disorders. 2022;26(2):256–266. doi:10.1177/1087054720961819. Historical secondary analysis; online publication 2020.
Leave a Reply