The Labour Government had downgraded cannabis from grade B to grade C on the scale of harmful drugs. This was not a very timely decision, in view of some research published in the British Medical Journal and The British Journal of Psychiatry recently. I am glad to say that before the last election, the Minister regraded cannabis back up to level B.
Some earlier (1987) Swedish research [1] using nearly the whole cohort of Swedish males (during national service) showed that there is a correlation between drug use in the teenage years and the later incidence of schizophrenia. As explained above, a correlation does not of itself imply a causal relationship. Instead, the Swedish work allows one to speak of an "association" between cannabis use in the teenage years and later schizophrenia. The study showed that the chance of being schizophrenic at a later date was 6 times as high in the group using cannabis than in the group not using it. The pro-cannabis lobby has criticised this and other studies, saying that the analysis does not ask whether there is a common cause of both the psychosis and the use of cannabis; for example, its use was illegal, thus causing a lot of worry. Another suggestion is that in the early stages of the disease, cannabis is perceived to be a palliative, and [seems to] reduce the symptoms, leading to more use by those with mental problems than by the general population. The new work [2] tends to discount the theory about illegality, since there was not the same association between schizophrenia and other even more illegal drugs. In the Swedish work, it was not shown that both cannabis use and later psychosis are not both caused by another circumstance, namely, a predisposition to schizophrenia. In [2], this predisposition was measured at age 11 using standardised criteria; this vulnerable group had a statistically significant increased association with later schizophrenia if they took cannabis, as compared with vulnerable children who did not take it; the remaining group also showed an increase, but it was less marked. The third result of [2] was that the chance that a cannabis user would develop schizophrenia at 26 was greater for those who started heavy use of cannabis at 15, as compared with a start at 18. The conclusion of the study was that policy should concentrate on trying to delay the start of the cannabis habit in all children, but especially in the vulnerable group.
In [3], a study is made of whether cannabis use is a cause of schizophrenia, which could then be regarded as an outcome. Here, the three most important items which must be present, if we are to say cannabis is a cause, are association, temporal priority, and direction. I can't do better than quote from [3]: "Association is the requirement that a cause and an outcome appear together. When the putative cause is present, the outcome rate [must be] higher than when the putative cause is absent. There is no requirement for the putative cause to be present in every case of the outcome, just that the rate of outcome is higher in those with it than without it." Thus, association is established if the two effects are positively correlated. More, "temporal priority is the fundamental property that the putative cause be present before the outcome." Finally, "Direction refers to the fact that changes in the putative cause will actually lead to a change in the outcome". Thus direction is the same criterion as I described above in the EPR experiment, if we are to say that there are influences faster than light in quantum mechanics: Alice's and Bob's spins are correlated (so they have an association) but the requirement of direction does not hold: changing Alice's spin value after she has measured it does not affect Bob's spin. By the way, the criterion of temporal priority also fails in the EPR experiment, since if Bob has already measured his spin, its value cannot be caused by Alice's observation, which comes later.
In [3] the authors do a meta-study of the literature on cannabis and schizophrenia with a view to finding out whether various suggestions are causes of the illness, using the above criteria. The hardest one to prove is direction. In [4] it is remarked that all three criteria have been demonstrated in some cases, and that cannabis intoxication is a cause of acute transient psychotic episodes in some people, and that it is a cause of recurrence of pre-existing psychotic symptoms. This was not just an association: direction was proved by giving randomly chosen patients the active chemicals in cannabis, and looking for symptoms (as assessed by the doctor using standard criteria, combined with regular reports of the patients themselves on a standard form) soon after. The results showed a larger incidence of psychosis in those patients who had been given the active chemicals, compared with patients who had not been given any. In a discussion in The Moral Maze, on BBC Radio Four, Melanie Phillips tried to make this point, but was out-gunned by Stephen Rose. He said that people who [unlike himself] have no expertise in neural science should not speak on the matter, and that "it is very difficult" to establish a causal relation. He did not say so, but he might have been referring to the fact that association (by itself) is not proof of causality. Unfortunately for his argument, this causal relationship was established in [4], since all three requirements were met. This conclusion is not vitiated by the possibility that use of cannabis caused changes in the brain, and that by stopping the use the patient suffered withdrawal symptoms which led to the observed psychosis. The outcome is the same.
According to [3], controversy remains only as to whether cannabis use actually causes schizophrenia in the long run, as opposed to temporary symptoms. The authors define "heavy use" of cannabis to be 50 or more joints per year, and they enquire whether heavy use is a cause of schizophrenia in the long term. They conclude that association and time priority were established in all the studies they quote. An association persists when other possible causes are factored out, such as use of other drugs, disturbed behaviour, low IQ, place of upbringing, cigarette smoking, poor social integration, sex, age, ethnic group, level of education, unemployment, and single marital status.
The idea that cannabis causes long term psychosis is supported but not proved by the relation between the size of dose [of cannabis] and the severity of the disease later in life. The causal hypothesis passes all tests of association and time priority when these other possible factors are adjusted for. This robustness is not true of some of these other possible causes (in particular, use of other drugs): they fail to satisfy association when cannabis is factored out. A history of previous psychotic symptoms is the exception; it is reasonable to assume that such previous history is a cause of some of the later illness. This group [the vulnerable group] is the most at risk of enhanced illness in the group that took cannabis: there, the association is stronger. They conclude that a doubling of overall risk is associated with heavy use, after factoring out the effect of previous psychotic symptoms. The authors predict an increase in schizophrenia in proportion to the increased use of cannabis in the USA and London over the last 10 years. I note also a claimed recent drop in cannabis use in Holland, so another test of association is that the illness should fall there, in line with this reduction, over the next ten or fifteen years. Neither of these predictions, if fulfilled, will be proof of direction: an increase in both cannabis use and in schizophrenia in London and the USA might have a common cause [eg an increase in a population susceptible to both]. Similarly, a decrease in schizophrenia in Holland ten years after the recent observed reduction in cannabis use might, if found, also have a common cause [eg a national decrease in worrying]. However, the recent reduction in Dutch cannabis use might have more to do with the police decision to close half the cannabis cafés than a decrease in worrying. If this can be shown to be the case, and a reduction of schizophrenia occurs in ten years time, direction would have been proved. One could then talk about the reduction in illness as having been caused by the closure of the cafés.
It is indeed difficult to prove that the use of cannabis causes long-term schizophrenia ten or so years later. We saw that the difficult part is to prove direction. This is not to say that it is difficult to design the required experiments. All we need is to assemble a group of volunteers 15 years old, blindly select half who ingest a placebo, leaving the rest to ingest the active components of cannabis [already identified as one of the causes of short term symptoms of mental illness]. The dose should be the equivalent of one spliff per week during the test period, say the teenage years. Which child gets the palcebo, and which gets the drug, is to be chosen at random, and kept secret from the participants (but is recorded by the researcher). I have thus described the standard structure of a double-blind experiment. We then measure whether the difference in the number of cases of schizophrenia, suicide, religious mania etc ten years later is significantly different between the two groups. If it is, and the placebo-takers have less illness, then direction is proved, within the level of significance chosen.
The difficult part is to get the proposal past the Ethics Committee of King's College London. In fact, much more recently (2011), a group working at King's College and the Maudsley have indeed shown that smoking cannabis does cause mental illness.
In the study in New Zealand, analysed in [2], 3 out of 29 patients who were heavy users of cannabis suffered long-term psychotic damage. This is more than one in ten. In [3], other studies are included in an analysis, and adjustments made, and the authors conclude that the association overall for heavy users is 3%, compared with 1% for light users and non-users. Thus 97% of heavy users might expect not to suffer ill effects ten years later. This shows that, if heavy use is a cause, it is not effective on its own. One could form the hypothesis that there is a predisposition [present in 3% of the population] for cannabis to cause schizophrenia, and for those without the predisposition, smoking pot is not going to cause long-term schizophrenia. Since we do not know how to identify this condition, each of us has a risk of being among "the chosen few". If cannabis use is a cause, then by using cannabis (fifty times a year for several teenage years) we increase our risk by a factor of about 3.
But if cannabis were not a cause of long term illness, but is merely associated with it, then if a person is among the "chosen few", and he is persuaded not to use cannabis, this decision would not do him any good; he would get ill anyway. This is the implication of the failure of direction.
In August 2001, Prof. Mathew [sic] Martin-Iverson, of the University of Western Australia, asked for volunteers to help decide this question. If he got any, we must wait 15 years for the results.
[1] Andréasson, S et al., LANCET, 1987, (ii), 1483-5.
[2] Arsenault, L. et al., BMJ, Vol 325, 2002, 1212-1213.
[3] Arsenault, L., et al The British Journ. of Psychiatry, Vol 184, (2004), 110-117.
[4] D'Souza, C. et al, Marijuana and Madness, ed. D. J. Castle and R. Murray, Cambridge University Press, 2004. Degenhardt, L. and Hall, W., Australian and New Zealand Journal of Psychiatry, Vol 34, 26-34, 2004.
Return to EPR, Cot Deaths, and the Dangers of Cannabis
Return to Good Cures
Go to my HOME PAGE for links to my co-authors, friends, and others.
© by Ray Streater, 29 Dec 2003.