Methamphetamine increased sexual desire in a controlled human study, but it did not significantly change the group’s average hypothetical condom-use decisions. Desire, decision-making and actual behavior are different endpoints.
Berry and colleagues studied 19 recreational stimulant users in a within-subject, placebo-controlled experiment. Participants received oral placebo, 20 mg methamphetamine and 40 mg methamphetamine. Assessments included sexual-desire ratings, a multidimensional desire-and-arousal inventory, a Sexual Delay Discounting Task and monetary delay discounting.
The sexual decision task asked about hypothetical willingness to use a condom under differing delay and partner conditions. It did not observe actual sexual encounters or measure subsequent infection outcomes.
Methamphetamine produced dose- and time-related increases in a single-item desire rating and dose-related increases on the multidimensional inventory. However, neither average condom-use likelihood nor monetary discounting differed significantly by drug dose.
Changes in desire were associated with changes in condom-use likelihood in some partner conditions. That association is not equivalent to a significant average drug effect on the task. Selecting only the correlation would omit the study’s central null comparisons.
The study does not prove that methamphetamine invariably produces risky sexual behavior. It also does not show an absence of risk: a small sample and hypothetical choices may fail to capture effects arising in real situations.
Individual responses varied, and partner context mattered. Group averages cannot determine a particular person’s intentions, choices, capacity or behavior during a separate event.
The direct experimental evidence supports increased reported desire under the tested doses and conditions. Broader statements about actual behavior require additional evidence. The distinction is particularly important when using scientific literature in an individual forensic assessment.
No biological specimen concentration from this task can be substituted for an assessment of a person’s functional state, and the experiment did not establish such a threshold.
Related reading: What a controlled driving study can establish.