1. Association of cannabis and tobacco use with incident anxiety, mood and psychotic disorders: a systematic review

Thursday, 24 November, 2022 - 13:20 to 14:50
Knowledge market 1 (K1)


Cannabis and tobacco use have been reported to be associated with increased risk of various psychiatric outcomes, but findings from observational studies have been mixed. This study aimed to systematically review longitudinal studies reporting the association between tobacco use, cannabis use and incident mental illness.

Following a pre-registered protocol, CINAHL, Embase, MEDLINE and PsycINFO were searched from database inception to January 2022. Longitudinal, observational studies were included if examining new-onset mental health conditions as outcomes. Risk of bias was assessed using a modified Newcastle Ottawa Scale, and E-values were used to evaluate sensitivity to unmeasured confounding.

A total of 16,764 records were identified through electronic database searches and 77 studies were retained for inclusion, representing nearly 2 million participants. Overall, approximately half the studies reported that baseline tobacco/cannabis use was associated with onset of mood disorders, with the remainder reporting null results. Few studies reported a positive association between tobacco/cannabis use and onset of anxiety disorders. There were more consistent positive associations between tobacco/cannabis use and later psychotic disorders. E-values suggested several estimates were sensitive to unmeasured confounding.

Literature regarding the association between cannabis use, tobacco use and subsequent mental ill health is inconsistent. There is insufficient research on the longitudinal association between cannabis-tobacco co-use and mental illness. Studies were biased to key limitations (e.g. unmeasured confounding) and there was some variability in exposure and outcome definitions, limiting interpretations. Further research is needed to comprehensively understand these relationships, particularly triangulating findings with more robust, design-based approaches to causal inference.


Presentation files

24 A6 1320 Chloe Burke.pdf1.25 MBDownload



Part of session