An attention score that looks better with lisdexamfetamine than with immediate-release mixed amphetamine salts does not, by itself, establish superiority or a longer clinically meaningful duration. Martin and colleagues’ 2014 pilot deliberately used descriptive analyses. Its estimates are informative, but they should retain that exploratory status.
Eighteen adults received three treatments
The randomized, double-blind, three-period crossover study enrolled 18 adults with ADHD, with a mean age of 30.8 years. Participants received oral lisdexamfetamine 50 mg/day, mixed amphetamine salts immediate release 20 mg/day and placebo, each for seven days in randomized order. The study selected people with a history of successful amphetamine-based treatment, recruited from one investigator’s practice.
All 18 entered the pharmacodynamic and safety analyses; one withdrew because of adverse events. Some treatment-specific assessments therefore involved 17 people. This was not an experiment in an unselected population beginning stimulant treatment, nor a healthy-volunteer trial of nonmedical cognitive enhancement.
What did “power of attention” measure?
The primary outcome summed simple reaction time, choice reaction time and digit-vigilance speed from a computerized battery. The composite was expressed in milliseconds, with a lower score representing faster combined performance. It was not an IQ score, a blood drug concentration or a direct measure of workplace effectiveness.
Assessments extended to 16 hours after dosing. At day seven, the largest placebo-adjusted reduction occurred at five hours: −150.0 ms for lisdexamfetamine (95% CI −235.41 to −64.50) and −79.8 ms for immediate-release mixed salts (−165.72 to 6.21). Those are separate active-versus-placebo estimates.
Why the confidence intervals do not prove an active-drug difference
The paper explicitly stated that it did not conduct inferential statistical comparisons among treatments. The pilot was not sufficiently powered for those comparisons. A confidence interval excluding zero for one placebo contrast, while another includes zero, does not establish a statistically significant difference between the active treatments.
The time course was also less uniform than a simple “16-hour benefit” label suggests. At 12 hours, the lisdexamfetamine estimate was −26.4 ms with a 95% interval from −90.69 to 37.91. The 16-hour estimate was −65.9 ms (−129.25 to −2.63). Preserving the intervening estimates avoids implying an uninterrupted, precisely established duration.
Other outcomes did not show the same pattern
Self-reported ADHD symptoms on the Conners’ Adult ADHD Rating Scales short form were similar under active treatment and placebo. The detailed results described little change in tertiary attention-related measures, including continuity of attention and response variability. A favorable speed composite should not be expanded into improvement across the whole battery.
The investigators discussed possible reasons for the symptom-scale result, including short treatment periods, timing and measurement sensitivity. These were explanations to investigate, rather than proof that the symptom scale was wrong or that participants failed to recognize a benefit.
What were the safety observations?
Treatment-emergent adverse events occurred in 12/18 participants during lisdexamfetamine, 9/17 during mixed salts and 8/17 during placebo. Most were mild or moderate; no serious treatment-emergent events were reported. The withdrawal involved nausea and vomiting outside the study’s defined treatment-emergent window, so it should not be recategorized as a treatment-emergent withdrawal.
Small samples and short exposure cannot establish comparative safety. The study was manufacturer funded and excluded important medical and psychiatric comorbidities. Its strongest use is to inform how repeated attention testing might characterize pharmacodynamic profiles and guide larger trials. It does not settle an individual treatment choice, prove a driving benefit or demonstrate one formulation’s superiority.
Source
Martin PT, Corcoran M, Zhang P, Katic A. Randomized, double-blind, placebo-controlled, crossover study of the effects of lisdexamfetamine dimesylate and mixed amphetamine salts on cognition throughout the day in adults with attention-deficit/hyperactivity disorder. Clinical Drug Investigation. 2014;34:147–157. doi:10.1007/s40261-013-0156-z. Historical exploratory human experiment.
Leave a Reply