Despite research advances, it remains unclear if long-term, regular cannabis use harms cognition once intoxication has passed. Our meta-analysis aimed to investigate the association between cognitive functioning and long-term (mean ≥2 years), regular (mean ≥4 days/week), recreational cannabis use in adults during abstinence (mean ≥12 hr). We searched PubMed, PsycINFO, CINAHL, Scopus, and Dissertations and Theses International for English-language articles from the date each database began until May 22, 2019. We identified study inclusion by completing abstract and full text screening using predetermined criteria and Preferred Reporting Items for Systematic Review and Meta-Analysis guidelines. We classified cognitive performance into 6 cognitive domains (attention, executive function, learning and memory, decision making, information processing, and working memory), and included a global measure. Effect sizes were calculated for each domain using univariate meta-analyses. There were 30 studies with a total 849 participants who used cannabis (M = 30.7-years-old, SD = 5.5-years-old) and 764 control participants (M = 30.3-years-old, SD = 5.9-years-old). Cannabis was associated with significant but small-magnitude deficits in executive function, learning and memory, and global cognition, while decision making had moderate deficits. There were small-magnitude and nonsignificant group differences for information processing, working memory, and attention. Cannabis use duration, age of onset, and prolonged abstinence (≥25 days) did not influence outcomes, except group differences in executive function were nonsignificant in analyses of prolonged abstinence. Our results suggest that long-term, regular cannabis use is associated with small to moderate deficits in some cognitive domains.

No abstract

Objective Opioids, often prescribed for chronic non-cancer pain, may adversely affect cognition. Research has not been synthesised in recent years, during which time academic interest has increased. This study presents meta-analyses on cognitive performance in people taking opioids for CNCP. Methods We ran systematic literature searches in EMBASE, Medline, and PsycINFO. Eligible studies included people taking opioids for CNCP and an opioid-free group (i.e., case-control) or session (e.g., pre-post), and objective cognitive assessments. Using random-effects meta-analyses, we computed pooled effect sizes for differential task performance for each study design across five domains (motor performance, attention, working memory, executive functions, memory). Results Seventeen studies were included. Case-control studies covered 3 control types (healthy, CNCP, taper-off). Pre-post studies were grouped into 5 follow-ups (4–6 and 6–9 weeks; 3, 6, and 12 months). Effect sizes ranged from 0.02–0.62. Cases showed small magnitude impairments in attention and memory compared with healthy controls. Although limited by small sample sizes, there was no clear evidence of impairment in cases compared with opioid-free controls with CNCP. Cases showed some cognitive improvements from opioid-free baseline to follow-up. Effects were strongest for attention and working memory, and were apparent from 4 weeks to 6 months follow-up. Other effects were small and non-significant. Conclusions Opioid therapy for CNCP did not worsen cognitive performance and improved it for some domains. People who take opioids for CNCP may evidence deficits in attention and memory, but this is unlikely to translate to global impairment and likely relates to pain more so than opioids.