Impairments in cognition are frequently associated with acute cannabis consumption; on the other hand, controversies persist regarding the residual cognitive impairments of cannabis, with some estimates highlighting significant or mild cognitive impairment. One of the main limitations of the available research syntheses is that little attention has been paid to individuals with cannabis use disorder. Thus, our main objectives are to determine the amplitude of the cognitive deficits associated with cannabis use disorder, and to identify the cognitive domains the most and least impaired.

Studies with a patient group with a cannabis use disorder diagnosis and data from at least one validated neurocognitive test were selected. After manual extraction, data were pooled in a multivariate meta-analysis and effect size estimates were calculated for 13 cognitive domains. Meta-regression analyses on potential moderators were performed.

There were small-to-moderate impairments in 10 out of the 13 cognitive domains. Deficits in verbal learning/memory, speed of processing and working memory were more prominent (d = 0.4/0.5) whereas verbal fluency and attention were the least affected. No association was observed between the potential moderators and global cognition.

This meta-analysis shows that cannabis use disorder is associated with moderate deficits in verbal learning/memory, speed of processing and working memory. Despite the limitation of the studies in the field, our results should serve as a reminder that the residual cognitive impairments associated with cannabis should not be under-estimated prematurely.