The Medical-Cannabis Scoping Review Did Not Produce a Pooled Cognitive Effect Size
The medical-cannabis review synthesized findings qualitatively because the evidence did not support statistical pooling. Its conclusions should not be presented as a single meta-analytic effect estimate.
The studies were not one standardized experiment
Patient populations, cannabis products, doses, tasks and testing times varied across the included trials. The reviewers explain that this heterogeneity prevented a meta-analysis. Differences in cognitive and motor test methods were a central obstacle rather than a minor reporting detail.
A qualitative pattern is a different form of synthesis
The authors could describe recurring observations, exceptions and gaps without calculating a combined standardized difference. Such a review remains useful. It does not, however, supply a pooled confidence interval, a universal percentage of impairment or an exact average recovery curve.
Do not invent precision to simplify the message
A statement that findings generally improved over time should retain the underlying variation. It cannot be converted into a numerical effect size merely by counting studies with significant results. Small samples and incompatible measurement schedules further constrain that approach. A careful citation identifies the paper as a scoping review and distinguishes its narrative synthesis from a formal pooled analysis. That allows readers to assess what supports the conclusion without giving disparate studies the appearance of one harmonized clinical trial.
Related analysis: the review’s broader findings and limitations.
Research cited
This analysis draws on the supplied review: Eadie and colleagues (2021). Duration of Neurocognitive Impairment With Medical Cannabis Use: A Scoping Review. Frontiers in Psychiatry.
