Withdrawal outcomes need not move together. A controlled d-amphetamine withdrawal pilot in people who used methamphetamine found a relative increase in craving after medication discontinuation, while Stroop cognitive scores showed no significant effects.
Thompson and colleagues enrolled 34 treatment-seeking adults in a four-week residential trial. Everyone initially received d-amphetamine, increasing to 30 mg twice daily. During weeks two and three, participants were assigned under randomized, double-blind conditions either to continue d-amphetamine or to receive placebo. Everyone received placebo in week four.
The study standardized exposure to prescribed d-amphetamine before examining discontinuation. It was not an experiment abruptly stopping a controlled methamphetamine dose, and the two drugs should not be interchanged in describing the protocol.
The placebo-switch group showed increased methamphetamine craving during weeks two and three relative to those continuing d-amphetamine. The study also followed withdrawal ratings, mood and vital signs. The Stroop Color and Word Test did not show significant effects.
That cognitive null finding does not negate the craving result, nor does increased craving establish a broad cognitive deficit. The measures address different experiences and functions. It would be inaccurate to compress them into one statement that withdrawal either affected everything or affected nothing.
The investigators did not collect urine samples for illicit-substance testing because study medication would complicate interpretation of amphetamine detection. This is an important protocol limitation when considering how drug use was monitored. It also illustrates why identifying the analyte and the medication context matters.
The sample was small, groups differed in age despite randomization, and only eight participants per group completed the study. The residential setting and standardized medication exposure differ from uncontrolled community use. The trial primarily supports the feasibility of a research model, rather than a universal treatment protocol or a personal withdrawal timetable.
These results do not establish driving impairment or a concentration threshold. For why measured exposure and behavioral effects remain separate questions, see our discussion of differing behavior with similar blood results.