The largest estimated difference in a meta-analysis is not necessarily its best-established finding. Potvin and colleagues’ 2018 analysis associated methamphetamine use disorder with a large social-cognition difference, but that estimate rested on only three studies. Its size, uncertainty and limited replication should travel together.
Forty-four studies did not assess every domain
The analysis included 1,592 people with methamphetamine use disorder and 1,820 healthy controls across 44 studies published before May 1, 2017. Participants were not randomly assigned to a chronic drug-use history. The synthesis therefore concerns differences between observationally defined groups, rather than an experimental estimate of what methamphetamine caused.
The researchers organized tasks into 12 cognitive domains. Social cognition included abilities such as identifying emotions or understanding others’ mental states. It was distinguished from working memory, executive function and a reward- or impulse-related domain. A person’s performance in one domain does not automatically establish performance in the others.
The largest effects came from smaller evidence subsets
Social cognition had a standardized mean difference, Cohen’s d, of 1.117, with a 95% confidence interval of 0.810–1.423, based on three studies. The reward/impulsivity domain had d = 0.926 (0.716–1.135), based on eight studies. Positive values meant worse average performance in the methamphetamine-use-disorder groups.
By comparison, global cognition combined 27 studies and had d = 0.462 (0.329–0.595). The sample of 44 studies cannot be used as the denominator for the social-cognition estimate. Nor can a standardized mean difference be read as the percentage of people who are impaired.
The authors explicitly called for replication of the social-cognition result. A comparatively narrow interval in the included studies does not remove uncertainty about selection, publication or applicability to other populations.
Results were not uniformly negative across cognition
Visual learning did not show a statistically significant group difference: five studies produced d = 0.275, with a 95% confidence interval from −0.107 to 0.657 and p = 0.158. That is a null significance result, not proof of identical ability or evidence that visual learning is unaffected in every individual.
Other domains generally had smaller estimates than social cognition and reward/impulsivity. The profile should therefore not be compressed into “global severe impairment.” Averages across different tests describe a research pattern; they are not a comprehensive clinical assessment of each participant.
What about publication bias and confounding?
The global-cognition publication-bias test was significant, with Egger’s t = 5.1 and p = 0.00003. The authors recognized that bias could inflate reported effects. The analysis also depended on the original studies’ matching, exposure descriptions and management of other relevant differences.
The review excluded studies involving HIV-confounded groups and cognitive testing during imaging sessions under its stated selection rules. That helps define the included evidence but narrows the populations and settings to which its estimates directly apply. It is not a complete summary of cognition in every person who has used methamphetamine.
A null abstinence association is not a recovery forecast
Study-level meta-regressions did not find significant associations between overall deficit magnitude and age, sex composition, education or abstinence duration. The abstinence analysis involved 14 studies. It compared study characteristics; it did not repeatedly assess the same individuals as they recovered.
Accordingly, it cannot establish that longer abstinence never helps, that all deficits are permanent or that a person reaches normal function after a specified interval. The paper likewise did not establish the social or occupational consequences of the observed effect sizes.
How should this evidence inform an expert opinion?
The synthesis supports taking social cognition seriously as a research and assessment domain. It does not justify inferring that a person misunderstood another’s intentions, lacked a particular mental capacity or acted impulsively because methamphetamine was detected. Those individual conclusions require direct evidence beyond an observational group association.
Source
Potvin S, Pelletier J, Grot S, Hébert C, Barr AM, Lecomte T. Cognitive deficits in individuals with methamphetamine use disorder: a meta-analysis. Addictive Behaviors. 2018;80:154–160. doi:10.1016/j.addbeh.2018.01.021. Historical meta-analysis of human group-comparison studies.
Leave a Reply