Loading ...

MADRID — Rates of cannabis-induced psychosis (CIP) and cannabis use disorder (CUD) have risen across Scandinavia alongside increasing cannabis potency, accompanied by a parallel increase in schizophrenia and mortality, research showed.

“It doesn’t prove causality, but everything behaves exactly the way we would expect it to under the assumption of causality,” said Carsten Hjorthøj, PhD, associate professor at the Research Center for Mental Health, University of Copenhagen, in Copenhagen, Denmark.

Speaking at a recent symposium on CIP, Hjorthøj said high-potency cannabis is the most significant known preventable risk factor for developing schizophrenia, said Hjorthøj.

The findings were presented on April 7 at the European Psychiatric Association Congress 2025.

Men at Greater Risk

The recent study by Hjorthøj and colleagues showed that the association between CUD and schizophrenia is particularly high in young men compared with women.

“It could be that males use more [cannabis], it could also be that males have brains that are wired differently. We don’t know, and that is something for future research,” he said.

The nationwide Danish register-based cohort study included data on all individuals aged 16-49 years during the period from 1972 to 2021 (n = 6,907,859), including 60,563 (0.9%) who had received a CUD diagnosis.

During a follow-up of almost 130-million person-years, investigators found that overall, patients with CUD had an adjusted hazard ratio (aHR) for schizophrenia of 2.4 for men and 2.02 for women.

However, this risk was markedly higher in men younger than 20 years (aHR, 3.84 in men vs 1.81 in women), and those between the ages of 20 years and 25 years (aHR, 2.58 and 1.91, respectively).

The study also examined whether sex-specific proportion of schizophrenia cases attributable to CUD change over time — from 1972 through 2021. This was measured as the population attributable risk fraction, which, “under the assumption of causality, estimates how many cases of schizophrenia could have been prevented if nobody had used cannabis,” he explained.

Findings showed that in 1970’s approximately 1% of female schizophrenia was linked to cannabis vs approximately 5% in 2021. “It’s a statistically significant increase, but it’s not earth shattering,” he said.

However, among men, there was exponential growth during that period, with the proportion of schizophrenia cases associated with cannabis rising from approximately 1%-2% to between 20% and 25% by 2021.

“That’s 1 in 5 to 1 in 4 cases of schizophrenia that might have been prevented by people not using cannabis,” said Hjorthøj.

He noted that studies have shown rising schizophrenia rates since the 1970s closely parallel to the increase in high-potency cannabis use.

“It’s getting stronger everywhere, we’ve seen this internationally, whether you legalize cannabis or not. The average potency in Denmark is around 20%-30% THC, and 10% is where we say the cutoff for high potency is.”

“There’s an increased public perception that cannabis is not very dangerous, that it’s not something to worry about, but that’s in stark contrast to the research findings that we’re presenting here,” Hjorthøj added.

One of the Strongest Schizophrenia Risk Factors

Similar patterns have been observed in Norway, said Eline Borger Rognli, PhD, Norwegian Center for Addiction Research, University of Oslo in Oslo, Norway.

In a recent study, her team used data from a national patient registry to identify all cases of substance-induced psychosis (SIP) and tracked them over 6 years to assess the cumulative risk of transition to a schizophrenia spectrum disorder.

Of the 3187 incident cases of SIP, 453 were classified as alcohol-induced, 562 as cannabis-induced, 707 as amphetamine-induced, and 1235 as related to multiple substances.

At follow-up, 636 patients were diagnosed with schizophrenia spectrum disorder – a 6-year cumulative risk of 27.6% for all substances, with the highest being 36% for cannabis.

“That really makes cannabis-induced psychosis a significant risk factor for schizophrenia, one of the strongest risk factors we know of,” she said.

But a growing number of experts, including Rognli are beginning to question whether cannabis-induced psychosis is a risk factor for schizophrenia, or whether it is an early sign, opening new questions about how best to treat it.

Emerging research suggests that whether cannabis is identified as the trigger for psychosis, or whether it is comorbid with psychosis, the mortality risk is similar.

Preliminary, unpublished data from a a Swedish cohort study showed that first-episode psychosis (FEP) with comorbid CUD — diagnosed within 2 weeks before, or after, the psychotic episode — was associated with a similarly elevated mortality risk as CIP (7.8% and 5.23%, respectively).

By comparison, FEP without any substance comorbidity had a lower mortality rate of 3.38%, reported Solja Niemelä, MD, PhD, professor of addiction medicine at the University of Turku in Turku, Finland.

Mortality by Cause

Sub-analyses of mortality by cause — distinguishing external from natural causes, including suicide, accidental overdose, and other nonoverdose accidents — revealed a consistent pattern of higher risk in individuals with both CUD and psychosis compared with those with psychosis alone.

Findings from another Scandinavian-wide study support this finding.

A recent study by Heidi Taipale, PhD, Niuvanniemi Hospital in Kuopio, Finland and senior researcher at the Karolinska Institute in Stockholm, Sweden, that used patient registry data from Denmark, Sweden, and Norway, showed rates of CIP increased steadily between 2000 and 2016 — from 2.6 to 5.6 per 100,000 in Denmark, 0.8 to 2.7 in Sweden, and 1.8 to 3.0 in Norway — corresponding to relative increases of 115%, 238%, and 67%, respectively.

About 80% of cases were men, and the median age of onset decreased from 29 years to 24 years over the study period, she said.

Similarly, unpublished mortality data from the same study showed a sevenfold increased risk for death among individuals with CIP compared with control individuals: 8.6% vs 1.5% in Denmark, 5.2% vs 0.6% in Norway, and 7.1% vs 1.2% in Sweden.

Taipale reported that the corresponding all-cause mortality HRs were 6.6, 9.01, and 7.59, respectively. The risk for suicide was also markedly elevated, with HRs of 20.48, 17.88, and 12.84 — reflecting about a 15-fold increase over the 16-year study period.

She emphasized that these findings highlight the critical need for improved prevention and treatment strategies.

Hjorthøj, Rognli, and Niemelä reported no conflicts. Taipale reported receiving lecture fees from Gedeon Richter, Janssen, Lundbeck and Otsuka, and research collaboration with Janssen.

Kate Johnson is a Montreal-based journalist with more than 30 years of experience writing about all areas of medicine.