A small open-label medication study can justify a larger experiment without establishing that a treatment works. A subsequent randomized trial of modafinil for methamphetamine dependence illustrates why that distinction matters: favorable-looking subgroup patterns did not overturn the trial’s negative overall result.
What did the investigators actually compare?
Heinzerling and colleagues reported a historical, randomized, double-blind, placebo-controlled human trial in 2010. Of 207 people entering screening, 71 were randomized: 34 to modafinil and 37 to placebo. Both groups also received access to weekly cognitive behavioral therapy and contingency management, which rewarded methamphetamine-negative urine specimens. The medication comparison therefore tested modafinil’s added effect within that treatment platform.
Oral modafinil started at 200 mg daily for three days, increased to 400 mg each morning, and returned to 200 mg for the final three days of the 12-week medication phase. Participants attended three times weekly for urine testing. Significant medical and psychiatric exclusions and exclusions for several other substance dependencies narrowed the population studied.
What were the main results?
The investigators found no statistically significant difference in aggregate urine outcomes in the planned full-sample analyses. In their longitudinal model, the odds ratio for providing a methamphetamine-free urine screen with modafinil versus placebo was 0.78, with a 95% confidence interval of 0.39–1.56 and p = 0.49. That interval permits appreciably different effects in either direction; it does not establish equivalence.
Completion was 41% with modafinil and 35% with placebo, p = 0.60. Mean retention was 57 versus 47 days, p = 0.16. There was also no significant medication-group effect on craving or depressive symptoms. Improvements over time within treatment cannot, by themselves, be assigned to modafinil when the comparison group also changes.
Why did counseling attendance attract attention?
Exploratory analyses divided participants by baseline use frequency and by attendance at zero to two versus at least three cognitive behavioral therapy sessions. Some numerical patterns favored modafinil among people with low attendance or higher baseline use. The reported aggregate urine and retention comparisons did not establish significant benefit in those small subgroups.
More fundamentally, attendance was behavior occurring during the trial, rather than a randomized intervention. People who attended more sessions could differ in motivation, stability, medication-taking behavior or other factors. The association between attendance and better urine outcomes does not isolate a causal counseling effect or validate a rule that medication should be reserved for people unlikely to attend.
How complete were the measurements?
Participants missed an average of seven of 36 possible urine screens in the modafinil group and five in the placebo group. Pill-count adherence was available only when medication packages were returned: three modafinil participants and eight placebo participants returned none. A high adherence percentage among those with count data is consequently not verified exposure for everyone randomized.
Four modafinil participants required dose reduction for anxiety or sleep complaints; medication discontinuations included an elevated-heart-rate event in that arm. The authors described treatment as generally well tolerated, but this small, selected study cannot establish safety across broader clinical populations.
What does this historical evidence support?
It supports reporting a negative overall trial with uncertain subgroup observations. It does not establish modafinil as an effective methamphetamine treatment, identify a validated responder subgroup or address driving impairment. A urine endpoint measures the study’s drug-use outcome, not an individual’s functional state at a particular time.
Source
Heinzerling KG et al. Randomized, double-blind, placebo-controlled trial of modafinil for the treatment of methamphetamine dependence. Drug and Alcohol Dependence. 2010;109:20–29. doi:10.1016/j.drugalcdep.2009.11.023. Historical controlled human evidence.
Leave a Reply