The Evidence Analyzed

The study generating these headlines is a comprehensive national cohort analysis examining individuals aged 16 to 49 in Denmark from 1970 to the present. This research is indeed impressive and, as an epidemiologist, I find myself envious.

Researchers analyzed every new schizophrenia diagnosis over the last five decades alongside substance use disorder diagnoses. After adjusting for age and various potential confounding factors, such as family mental health history and country of origin, they found that those with cannabis use disorder exhibited a notably higher risk of developing schizophrenia.

The authors calculated a Population Attributable Risk Fraction (PARF) for cannabis use disorder in relation to schizophrenia. This metric indicates the proportion of schizophrenia cases linked to cannabis use based on observed risk and drug usage in the population. If this relationship were entirely causal (meaning cannabis use disorder directly results in schizophrenia in all related cases), eliminating cannabis use could significantly reduce schizophrenia rates.

The results indicated a 15% risk increase for men and 4% for women linked to cannabis use disorder. Thus, if cannabis use disorder were entirely eradicated among adults aged 16 to 49 in Denmark, we might expect schizophrenia cases to decrease by 15% in men and 4% in women.

You might wonder about the 30% figure highlighted in the media. This discrepancy arises because the researchers calculated the PARF annually, and for the specific demographic of men aged 21-25 in one year, the figure approached 30%. However, in most years, this was significantly lower, averaging around half that number, showcasing how the media's portrayal was misleading.

Even with these percentages, it’s essential to put them into context. The overall rate of schizophrenia among adults in Denmark during this timeframe was approximately 65 cases per 10,000 people. Therefore, if cannabis were completely eliminated, we might expect the rate to decrease to about 58 per 10,000.

At worst, assuming all associations found in the study are causal, the actual risk increase is relatively small, especially for women.

Exploring Correlation vs. Causation

It’s crucial to highlight that, despite the extensive scope of this study, establishing a causal link is challenging. For instance, the authors did not rule out reverse causation—the notion that schizophrenia could lead to increased cannabis use rather than the other way around. It’s well-documented that individuals with schizophrenia may resort to self-medication using non-prescription substances, raising the possibility that much of the observed correlation stems from those already unwell turning to cannabis.

Moreover, the authors used cannabis use disorder as a proxy for general cannabis consumption, which could introduce bias. Research indicates that individuals with schizophrenia who use cannabis are more likely to receive a substance use disorder diagnosis compared to those without mental health issues.

The study also did not adequately control for various potential confounding factors, such as income, which can influence the risk of both mental health diagnoses and cannabis use disorder.

Given these limitations, we cannot definitively ascertain how much of the 15% and 4% figures can be attributed to cannabis causing mental health issues versus mere correlation. It’s possible that the true population attributable risk of schizophrenia related to cannabis use is 0%, or it could be higher—we simply don’t know.

This is not the first investigation to highlight a significant correlation between cannabis and psychosis or schizophrenia. Decades of research have consistently demonstrated a connection between cannabis use and schizophrenia diagnoses.

However, the complexity of this research cannot be overstated. Most studies tend to compare cannabis users with non-users, making it difficult to address the inherent biases. Even studies that suggest a causal connection between cannabis use and long-term psychosis typically reveal a modest overall risk increase. For instance, a 2016 review indicated that individuals with psychosis who ceased cannabis use experienced a reduction in relapse risk, albeit of a relatively modest size.

While the evidence does support a reasonable causal link between long-term cannabis use and an elevated risk of mental health disorders, the risk increases are much smaller than sensationalized media reports imply. Additionally, the aforementioned 2016 review indicated that the heightened risk is not permanent; individuals who stopped using cannabis exhibited similar rates of mental health issues as those who never used it.

In summary, it appears that avoiding long-term cannabis use is likely to be beneficial for one’s health. However, this principle applies to many substances, and the risk increases identified in research are not nearly as dramatic as the headlines suggest. Smoking cannabis is not ideal for your health, and it is unlikely to be responsible for 30% of schizophrenia cases.