Using publicly available databases, researchers found that growing interest in cannabis in the UK over the past decade has coincided with rising psychosis cases and prevalence — though correlation does not prove causation.
Read this if you are interested in the cannabis-psychosis debate and novel approaches to tracking population health trends.
10+ yearsDuration of parallel rising trends in UK cannabis interest and psychosis cases
What the researchers found
By cross-referencing Google Trends data, Our World in Data, and other public databases, the researchers found that interest in cannabis in the UK has grown steadily over 10+ years, with a parallel overlapping rise in both psychosis cases and prevalence. The authors argue that publicly available data tools could help answer whether cannabis use changes at the population level are associated with schizophrenia rates, a question that has been difficult to answer with traditional epidemiological methods.
Why it matters
The cannabis-psychosis link has been studied for decades, but population-level evidence has been equivocal. This study proposes a novel, low-cost method for tracking these trends using freely available public data, which could supplement traditional epidemiological approaches.
The numbers in context
- Over 10 years of growing cannabis interest in the UK (Google Trends)
- Parallel increase in psychosis cases and prevalence
- Prospective studies since 1987 have suggested increased psychosis risk with cannabis use
- High-potency varieties confer the greatest risk
How the study worked
Ecological study using publicly available databases (Google Trends, Our World in Data) to track temporal trends in cannabis interest and schizophrenia rates in the UK. Cross-referenced trends to look for parallel patterns.
Who was studied
Population-level data from the United Kingdom
What this study cannot tell us
Ecological fallacy is a major concern — population-level correlations cannot establish individual-level causation. Google Trends reflects search interest, not actual cannabis use. Many confounders could explain parallel trends. The method cannot account for changes in diagnostic practices or healthcare access.
How to read the evidence
Ecological study using publicly available proxy data. Innovative approach but highly susceptible to ecological fallacy and confounding.
When this study was published
Published in 2023. Uses the latest available public data tools for epidemiological inquiry.
The bigger picture
The dose-response relationship between cannabis and psychosis at the individual level is well-established from prospective studies. Whether this translates to population-level increases in schizophrenia cases as cannabis use rises remains one of the most important unanswered questions in cannabis epidemiology.
Questions still open
- Do countries with different cannabis legalization timelines show different schizophrenia trends?
- Can Google Trends data be validated against actual cannabis use survey data?
- Are changes in schizophrenia diagnostic practices confounding the apparent increase?
Read the original research
Can we interrogate public databases to fill critical gaps in mental health epidemiology? Testing the association between cannabis and psychosis in the UK as an example.
Epidemiology and psychiatric sciences, 32, e40
Citation
Di Gennaro, Gianfranco; Colizzi, Marco. (2023). Can we interrogate public databases to fill critical gaps in mental health epidemiology? Testing the association between cannabis and psychosis in the UK as an example.. Epidemiology and psychiatric sciences, 32, e40. https://doi.org/10.1017/S2045796023000537