However, based on the non-causality-proving research we do have, we can make certain inferences about risk that are clinically appropriate, but then we need to leave this debate behind and move on to practical solutions that mitigate risk in both therapeutic and non-therapeutic uses of cannabis.

In other words, we have to move past prohibition vs. legalization and develop nuanced policy that supports individuals. 

Cannabis and Psychosis: What We Can (and Cannot) Say About Causality 

Debate about whether cannabis "causes" psychosis often gets framed as all‑or‑nothing, but the science points to something more nuanced: cannabis probably increases psychosis risk for some vulnerable people while not being a necessary or sufficient cause on its own. 

Why the Evidence Will Always Be Imperfect 

Most of the data we have come from observational studies: case-control samples, survey samples, or retrospective cohort studies. These designs allow us to ask whether people who use cannabis are more likely to later develop psychotic symptoms or disorders than people who do not, and whether heavier or earlier use is associated with more risk. 

What they cannot do is fully eliminate all confounding or prove causality. 

Several challenges recur: 

Despite these limitations, the field has gradually moved toward a "converging evidence" approach: we look for consistent temporal ordering, dose-response relationships, robustness to adjustment for confounders, and alignment with plausible neurobiological mechanisms. 

What Longitudinal Studies Show 

Large longitudinal studies generally find that cannabis use is associated with an increased likelihood of later psychotic outcomes, even after adjusting for many potential confounders. (1) 

Key points that appear repeatedly include: 

Nonetheless, several robust studies have shown NO association between cannabis use and psychosis. (6,7)

From a causal‑inference standpoint, this pattern suggests that cannabis can act as a contributory factor in some individuals. However, the magnitude of risk, if real, is modest at the population level, and residual confounding can never be fully excluded. (1,5)

What Twin Studies Add (and Why They Matter) 

Twin studies are particularly valuable because they eliminate (or attempt to) the genetic component of risk.

Studies where one identical twin uses cannabis but the other twin does not, repeatedly show no increase in psychosis in the cannabis using twin vs. the non-using twin8,9. Both twins might develop psychosis, but the cannabis-using twin doesn't develop psychosis more often than the other. 

This suggests that the risk of psychosis is genetically controlled, at least in part, but that adding cannabis does not trigger that psychosis. 

A Component in a Multifactorial Causal Web 

A useful conceptualization is that cannabis is a component cause in a multifactorial causal web. It is: 

Genetic polymorphisms (for example in dopamine‑related or cannabinoid‑related pathways), family history of psychosis, early trauma, and developmental stage (e.g., adolescence) may all modulate whether or not cannabis exposure contributes to the brain's vulnerability. This helps explain why risk is concentrated in particular subgroups rather than uniformly distributed across all users. (1,3,12) 

Getting Practical: Whose Risk Is Most Concerning? 

The evidence is most consistent for elevated relative risk in: 

Relative Risk vs Absolute Risk 

Even where relative risk doubles or quadruples in high-use groups, the absolute risk of developing a chronic psychotic disorder remains low at the population level. (2,5,10)

For public communication, this distinction can support a balanced message: "risk is real but concentrated and still uncommon." 

Moving Beyond the "Does It Cause Psychosis?" Argument 

Since we will never be able to answer this question ethically, a more scientifically honest and clinically useful framing might be: 

Practical Implications for Clinicians: Risk Mitigation 

For clinicians, the key task is not to resolve the causality debate but to systematically identify vulnerable patients and suggest ways to reduce modifiable risk. 

For patients at elevated risk, prioritize clear, individualized guidance: 

Adolescents and emerging adults: 

Patients with current or past psychosis, or strong family history: 

Patients using cannabis therapeutically: 

Integrate risk‑mitigation into ongoing care: 

By embedding this structured, risk‑focused approach into routine practice, clinicians can meaningfully reduce psychosis risk associated with cannabis use, even in the absence of definitive experimental proof. 

Declaration of generative AI and AI-assisted technologies in the manuscript preparation process 

During the preparation of this work the author(s) used Perplexity.ai in order to expedite formatting.  After using this tool/service, the author(s) reviewed and edited the content as needed and take(s) full responsibility for the content of the published article. 

Declaration of funding sources and conflict of interest 

No funding was received for this article and the author has no conflicts to report.  

References 

[1] Moore, T. H., Zammit, S., Lingford-Hughes, A., Barnes, T. R., Jones, P. B., Burke, M., & Lewis, G. (2007). Cannabis use and risk of psychotic or affective mental health outcomes: a systematic review. The Lancet, 370(9584), 319-328. https://doi.org/10.1016/S0140-6736(07)61162-3