Mental health

Super-strength 'skunk' cannabis linked to psychosis

" 'Skunk-like cannabis' increases risk of psychosis, study suggests," BBC News reports after a new study found high-potency strains of "skunk" cannabis – infamous for both its strength and its pungent smell – could be linked to one in four cases of new-onset psychosis. Psychosis is a mental health condition characterised by symptoms such as hallucinations and delusions.

The study compared cannabis use patterns among 410 people from south London who attended hospital with a first episode of psychosis, and 370 people from the general population without the condition.

It found the daily use of cannabis was associated with an increased risk of psychosis, and use of high-potency cannabis was associated with a greater increase in risk.

The researchers used their figures to estimate that 24% of new psychosis cases in the study population could be attributed to the use of skunk. But it's important to note this figure would not apply to populations where skunk use is less common than in the south London population the study looked at.

The figure also assumes that skunk definitely directly causes psychosis, which this study cannot prove by itself. However, there is growing concern that this could be the case, and other studies also support an association between cannabis use and psychosis.

If you are troubled by symptoms such as paranoia, depression and anxiety, and are concerned about your cannabis use, ask your GP for advice.

Where did the story come from?

The study was carried out by researchers from King's College London and Mount Sinai School of Medicine in the US.

It was funded by the UK National Institute of Health Research, the South London and Maudsley NHS Foundation Trust, the Institute of Psychiatry at King's College London, the Psychiatry Research Trust, the Maudsley Charity Research Fund, and the European Community's Seventh Framework Programme.

The study was published in the peer-reviewed medical journal, Lancet Psychiatry. It has been published on an open-access basis, so it is available online to download (PDF, 439kb) for free.

The Daily Telegraph is just of one of many media sources to make the mistake of extrapolating the "24% of first-episode psychosis caused by high potency cannabis" figure to the whole of Britain.

In fact, this figure is based on the high levels seen in south London in this study. It is also an estimate based on the assumption that skunk definitely causes psychosis, and that no other confounding factors are having an effect.

What kind of research was this?

This was a case-control study looking at the effect of the frequent use of high-potency cannabis (such as skunk) on the risk of a first episode of psychosis. High-potency cannabis contains more ∆-9-tetrahydrocannabinol (THC) than lower-potency cannabis.

Prospective observational studies have found a link between cannabis use and an increased risk of developing psychosis. However, exactly what patterns or aspects of cannabis use are associated with the greatest risk are not as clear. This study wanted to look specifically at whether the relationship was influenced by the potency of the cannabis smoked.

Psychosis is a general term covering symptoms of disordered thought patterns, delusions and hallucinations (including auditory and visual). It can be a feature of mental health illness (such as schizophrenia or severe depression), but an episode can also be triggered by other things, such as substance use or illness.

Observational studies are the only ethical way to study the effects of cannabis use in humans. Case-control studies are a good way to study outcomes that are not very common, such as psychosis.

Researchers need to make sure the groups being compared are similar in characteristics to mitigate potential confounding factors. This makes it more likely that the factor of interest is contributing to the difference in outcome, and not the confounders.

One limitation of case-control studies is they usually collect information on the exposure (cannabis use in this case) retrospectively, which means asking people what they did in the past. People may not be able to remember accurately, or may remember their habits differently if they think they contributed to their psychosis.

What did the research involve?

The researchers recruited adults with their first episode of psychosis (cases) and a control group of people who did not have psychosis. They got both groups to report details about their past use of cannabis, including high-potency cannabis. They then compared the cases and controls to see if they differed in their cannabis use.

The researchers asked all adults (18 to 65 years old) presenting at an inpatient psychiatry unit with first-episode psychosis to participate. They excluded those with an identifiable medical reason for their psychosis (organic psychosis), such as psychosis related to a brain tumour or serious head injury.

They advertised for volunteers from the same geographical area to act as the control group, and excluded anyone who had a current or previously diagnosed psychotic disorder.

The participants completed a modified cannabis experience questionnaire, which asked about whether they used cannabis, their age at first use, lifetime frequency of use, and the type used (skunk-type or hash-type).

The researchers used this information to rate each individual's lifetime exposure to cannabis into one of seven categories:

  • no use (score 0)
  • hash less than once a week every week (score 1)
  • hash at weekends (score 2)
  • hash every day (score 3)
  • skunk less than once a week (score 4)
  • skunk at weekends (score 5)
  • skunk every day (score 6)

The researchers then analysed whether the patterns of cannabis use and exposure differed between cases and controls. This analysis took into account potential confounders, such as sociodemographic factors, and participants' use of tobacco, alcohol, and other recreational drugs.

When an association was found, the researchers used the figures to estimate what proportion of first-episode psychosis is attributable to cannabis use – the population attributable fraction (PAF).

The PAF is a commonly used measure in public health, and is used to indicate how much of a disease could be prevented if a given risk factor is removed.

It is based on the assumption that the factor being assessed (cannabis use in this case) is directly affecting the risk of the outcome.

The PAF also overestimates the potential impact of individual risk factors, as it does not account for the interaction of multiple risk factors. It is affected by how common a risk factor is, and even a risk factor with a small effect can have a large PAF if it is a common risk factor.

What were the basic results?

The researchers recruited 461 out of the 606 (76.1%) adults with first-episode psychosis who presented at their unit. People who refused to participate were more likely to be of black Caribbean and black African ethnicity than those who agreed to take part.

The analyses included 410 cases and 370 controls who provided data on cannabis use. Compared with the controls, the cases were:

  • younger (average age 27 versus 30)
  • more likely to be male (66% versus 56%)
  • more likely to be black Caribbean (33% versus 20%) or black African (24% versus 10%)
  • more likely to have fewer educational qualifications or have been in higher education (for example, 15% with no qualifications versus 2%)
  • less likely to have ever been employed (88% versus 95%)

In their analyses, the researchers found that overall, having used cannabis was not associated with an increased risk of first-episode psychosis compared with never using cannabis.

When split by the type of cannabis used, there was not an increase in risk of first-episode psychosis in those who most used hash-like cannabis, compared with those who never used cannabis (odds ratio [OR] 0.83, 95% confidence interval [CI] 0.52 to 1.77).

However, the odds of first-episode psychosis in those who most used skunk-like cannabis were more than twice as high as those who never used cannabis (OR 2.91, 95% CI 1.52 to 3.60).

Daily skunk use was associated with more than five times the odds of first-episode psychosis (OR 5.40, 95% CI 2.80 to 11.30).

The researchers also found:

  • those who started using cannabis before the age of 15 were at an increased odds of first-episode psychosis – this increase was only of borderline statistical significance (OR 1.55, 95% CI 1.00 to 1.39); starting using cannabis later in life was not associated with first-episode psychosis risk
  • those who used cannabis every day had about three times the odds of first-episode psychosis compared with those who never used cannabis (OR 3.04, 95% CI 1.91 to 7.76) – less frequent use was not associated with first-episode psychosis risk

Based on their findings, the researchers calculated that:

  • 24% of first-episode psychosis could be attributed to the use of skunk-like cannabis – the figure was high because cannabis use was high among the cases in the study (53% of cases compared with 19% of controls)
  • 19.3% of first-episode psychosis could be attributed to daily use of cannabis
  • 16.0% of first-episode psychosis could be attributed to daily use of skunk-like cannabis

How did the researchers interpret the results?

The researchers concluded that the "use of high-potency cannabis (skunk) confers an increased risk of psychosis compared with traditional low-potency cannabis (hash)".

They say the ready availability of high-potency cannabis in their study population "might have resulted in a greater proportion of first-onset psychosis cases being attributed to cannabis use than in previous studies".


The current study has found the use of high-potency (skunk-like) cannabis is associated with an increased risk of first-episode psychosis than lower potency (hash-like) cannabis. The researchers calculated that in their population, 24% of cases of first-episode psychosis could be attributed to skunk-like cannabis use.

Still, it's important to note the figure of 24% of new psychosis cases being associated with skunk use is dependent on both the strength of the association between skunk use and psychosis, and how common skunk use is in the population. In this study, just over half of the people with first-episode psychosis used it.

The results would not apply to other populations where skunk use is less common. Overall, the proportion of cases and controls who had used any type of cannabis in their lifetime was also quite high (around two-thirds of each group).

This figure also does not take into account the potential for multiple risk factors interacting, so may overestimate the impact of a single risk factor.

There are a number of points to note:

  • the study is not able to prove that the patterns of cannabis use directly caused psychosis – it can only show associations
  • the researchers did not have information on the amount of cannabis usage in terms of joints or grams, and this may influence the associations seen
  • the researchers did not directly compare high- versus low-potency cannabis to see if the difference in risk between them was statistically significant

As with all studies of this type, it is difficult to know to what extent other factors could be confounding the association. That is, whether cannabis use causes psychosis directly, or whether other factors – such as personal characteristics, health and lifestyle – could increase the likelihood both of a person choosing to use cannabis and developing psychosis.

This study has attempted to control for this by adjusting for various factors, such as sociodemographics, tobacco, alcohol, and the use of other recreational drug use. However, it is difficult to ensure the influence of all other factors has been removed.

While this type of study by itself cannot prove that high-potency cannabis use directly causes psychosis, other studies also support an association. Given the possibility that cannabis could be increasing the risk of serious mental health issues and it is an illegal drug in the UK, it would seem sensible to avoid it all together.

NHS Attribution