Cannabis, schizophrenia & psychosis: The current evidence

1.0 Introduction

Cannabis is the most widely used illicit drug in the world. Various preparations are available including traditional hash resin and marijuana as well as more potent products such as sinsemilla. The latter product caused some alarm in the media although this has now probably been superseded by concern over the so-called “legal highs”, a range of substances, which can be purchased on-line, that includes stimulants as well as synthetic cannabinoids such as “spice”, containing HU-210, JWH-018 etc1. The main worry regarding the recreational use of cannabinoids (herbal or synthetic) is the risk to the users mental health, specifically the risk of psychotic illness2 .

It has long been known that, in some users, cannabis/THC intoxication can elicit an acute paranoid psychosis3. This is beyond doubt, appearing in the ancient Chinese and Indian medical texts, the writings of Baudelaire and Moreau from the nineteenth century and latterly, in formal laboratory-based or community based experimental studies4-8 There is also little doubt that cannabis use can worsen the course of a pre-existing chronic psychotic disorder, with a recurrence or worsening of positive symptoms (hallucinations and delusions) and further hospitalization being the usual outcome measures9-14. And it is now clear from numerous studies that schizophrenic patients who have a history of cannabis use go through their first psychotic breakdown at an earlier age, compared to those who did not use the drug15-23, (by 2.7 years on average24). This is not merely a statistical point because the earlier a psychosis emerges the worse the outcome25. Where there has been some controversy is around the issue of whether cannabis can actually cause schizophrenia in the first place26.

2.1 Cannabis use & schizophrenia

Epidemiological surveys conducted in numerous countries, since the late 1980s, have been consistent in showing an association between cannabis use and psychotic symptoms/schizophrenia27-37. But an association is not the same as causation. Alternative explanations, at least in theory, are reverse causality (i.e. that people with an existing psychosis are more likely to use cannabis, perhaps as self-medication) and confounding (e.g. cannabis is merely a marker for a 'true' causative agent, perhaps the use of another drug such as amphetamine)38.

Regarding the issue of reverse causality, in longitudinal follow-up studies, where the temporal relationship between cannabis-use and schizophrenia can be assessed with some confidence, the most common finding is that cannabis-use predates the onset of mental illness38 39. Regarding confounding, all of the longitudinal studies to date have made allowances, by incorporating factors such as amphetamine-use into the statistical model3839. The general finding is that the incorporation of other factors reduces the strength of the association between cannabis and schizophrenia, but that the association remains (and remains statistically significant)38. It is possible that residual confounding persists, and that an unknown factor 'drives' the relationship – but there are no suggestions as to what this unknown factor could be26.

2.2 Cannabis & Schizophrenia: The strength of the relationship.

The strength of the association between cannabis and schizophrenia is best framed in terms of an odds-ratio. Overall, taking an average of the recent studies, it has been found that cannabis-use approximately doubles the odds of developing schizophrenia40-42.Importantly however, there appears to be a dose-response relationship, in that the more extensive the use of cannabis the higher the risk. The Swedish conscript study showed this clearly. For those men that had taken cannabis on over 50 occasions, the odds of developing a schizophrenic illness increased from x2 (in those who endorsed ever having taken cannabis) to x728. A recent study from DiForti & colleagues also found a clear relationship between the frequency of cannabis use and the chances of developing a psychotic illness43.

2.3 Cannabis is a component risk factor for schizophrenia.

It has been pointed out repeatedly, that the use of cannabis is neither necessary nor sufficient for the development of schizophrenia. People can develop schizophrenia having never taken cannabis, and millions of people worldwide have used cannabis without developing a major mental illness. But the same can be said for cigarettes and lung cancer (where the risk is much higher). It is perhaps more useful to consider cannabis-use as a (component) risk-factor for major mental illness, in the same way that there are risk-factors for cardiovascular disease, such as high fat diets etc39.

2.4.1 Interactions between cannabis-use & other risk factors for schizophrenia

A major theme has been to clarify which additional factors interact with cannabis to confer risk. Some studies have investigated the interaction between cannabis and known environmental risk factors (such as being brought up in an urban versus a rural setting, or having a history of maltreatment in childhood)44. In general a supra-additive (synergistic) effect has been observed between cannabis-use and these other risk factors45-49.

2.4.2 Cannabis-use and susceptibility genes

It is suspected that particular genetic variants modulate the risk of cannabis for users in terms of psychosis outcomes. Polymorphic variation in dopamine components, and the enzyme AKT1 have attracted the most support. Initial work implicated an interaction between cannabis and a functional polymorphism in the gene for catechol-O-methyltransferase (COMT)50, and this interaction received support in laboratory-based and community-based experimental studies8 51. However two further epidemiological studies have been negative52 53.

More recently two independent groups have observed an interaction between cannabis and polymorphic variation at rs2494732 in the gene for the intracellular enzyme AKT1, which is an intermediate between neurotrophin receptors and mRNA translation within dendritic spines. Both studies found that, in conjunction with cannabis-use, CC carriers were twice as likely as TT carriers to develop a psychotic disorder54 55.

2.5 The age of cannabis onset

The risk of cannabis in terms of adverse mental health outcomes probably depends on the age when the subject begins to use cannabis. There appears to be a higher risk for use that emerges in early/mid adolescence compared to use that begins in adulthood29 56 57 An appealing explanation is that cannabis impacts upon the developing neural networks. Some animal studies, but not all58, support the idea that the sustained adverse consequences of CB1 agonists on cognition and social interaction arise if the drug is administered during maturation as opposed to adulthood59-61 Certainly, and perhaps to an even greater extent in higher primates, there is a massive re-organization of the nervous system in adolescence. In humans, the reorganization of synapses occurs in parallel with the development of abstract reasoning, and the emergence of social, philosophical, political and lifestyle attitudes62. The implication is that cannabis has the ability to disrupt the unfolding of the highest faculties of the nervous system, increasing the chances of future involvement with mental health services. The long-term impact of cannabis on IQ also appears to be age dependent. And again it is the adolescent period that appears to constitute a window of vulnerability63.

2.6 The type of cannabis

Recent epidemiological studies have begun to explore the nuances of the cannabis-schizophrenia relationship in more detail. One question in particular is whether sinsemilla [high THC: negligible CBD64] constitutes a higher risk (for mental health) than traditional cannabis products. This appears to be the case. A study from South London showed that patients in the midst of their first psychotic breakdown and healthy matched controls were equally likely to endorse ever having taken cannabis, but that patients were about x7 more likely than their peers to have used sinsemilla43. The higher the frequency of sinsemilla use, the higher the risk. Sinsemilla is largely devoid of cannabidiol (CBD), which can pharmacologically antagonize some of the effects of THC65, and may have anti-psychotic properties in its own right66, suggesting that the risk of sinsemilla might be partly accounted for by the absence of CBD rather than by elevated THC alone67. It has been argued that users of sinsemilla might self-titrate their intake of THC, much in the same way that spirit drinkers consume less volume of fluid than beer drinkers, however if the critical factor is the ratio of THC: CBD, then self-titration might be less important. Two epidemiological studies have shown that the relative absence of CBD in cannabis products is associated with more positive psychotic symptoms68 69. This is in agreement with laboratory-based studies in healthy controls in which doses of the two cannabinoids can be tightly controlled70 71.

2.7 Heavy use of cannabis & cannabis Addiction

The South London study also showed that there are a significant proportion of people who used cannabis (including sinsemilla) every day43. Many psychiatrists in clinical practice will encounter people who freely admit to having used cannabis, all-day, every-day, beginning immediately on waking. Some initial reports have suggested that sinsemilla is more addictive than traditional forms of cannabis, and similar to the case with psychosis, the relative absence of CBD may be a factor72. Studies are now beginning to explore whether CBD has efficacy against cannabis dependence.

It is now clear that cannabis dependence exists as a phenomenon73, and there is little doubt that a cannabis withdrawal syndrome exists, characterized by cravings, nervousness, insomnia, nightmares, irritability and abdominal pain74 75 Not surprisingly, people who experience the most severe withdrawals are much more likely to relapse and begin taking cannabis again74. People who are addicted to cannabis suffer poorer mental health generally compared to non-dependent users, with elevated rates of mood disorders as well as psychotic disorders76 77.

2.8 An acute cannabis-psychosis is a marker for the emergence of schizophrenia

People who experience an acute psychotic episode following cannabis, to the extent that treatment is needed, are at high risk of going on to develop a chronic psychotic disorder78.In a recent study from Finland (n=18,478), 46% [95% CI, 35-57%] of people who had been hospitalized for cannabis induced psychosis developed schizophrenia over the next 8 years, compared to 30% [95% CI, 14-46%] who had been hospitalized because of amphetamine induced psychosis79. This suggests that acute-psychotic experiences following cannabis are perhaps not as benign as was once believed.

3.0 Summary

There is little doubt that some people run into problems with cannabis, as is the same for any recreational substance. The worry in regard to cannabis is that many young people are putting their long-term mental health at risk by using the drug.

Early, heavy and dependent patterns of use clearly amplify any inherent risk of cannabis per-se. Sinsemilla appears to be more habit-forming than traditional forms of cannabis, and the evidence that it constitutes an elevated risk for psychosis is now fairly robust.There may well be SNPs that determine how risky cannabis is [in terms of psychosis outcome] for a particular individual. Currently variation in the gene for AKT1 is the most convincing, although COMT has not been discounted. Small-scale genetic studies can be revealing but carry the danger of false positives, and at the present time, there is no genetic test that can be used clinically to estimate the risk of cannabis-addiction or cannabis-psychosis for a particular individual.

But what we can say with some confidence is that the following patterns of cannabis-use put any individual at risk – particularly if the personality and mind are still maturing; 1.Everyday (dependent) use and 2.The use of high potency products. Finally, a history of acute cannabis-induced psychosis must be regarded as a red flag, warning against further use.

references on request.