The following is an edited transcript of an interview with Dr Robin Murray, Head of Psychiatry at the Institute of Psychiatry, in which he discusses the effect cannabis can have on our brain and its relationship with our genetic make-up.
Cannabis and the Brain
There is a huge amount for us to understand about cannabis and the brain.
Essentially almost nothing was known until the 1990s and really the bulk of the research has only happened in the last five years.
The cannabanoid receptor system has been discovered and people are beginning to understand how it works, it's important in the brain as a whole. But if you compare it with what we know about amphetamines or LSD or alcohol, we know hardly anything about the way it is metabolised in the brain.
That's extraordinary because this is a drug that's been used for 1000s of years.
Sure, but this is not a drug that anybody paid any attention to.
This is a drug which everybody thought was safe and indeed most people don't even think it's a drug.
For Rastafarians this is a sacred herb, for teenagers in south London if you ask them "Do you take any drugs?", they say "No no, I don't take drugs",
"Do you take cannabis?",
"Oh of course I take cannabis, but that's not a drug, that's a natural substance."
Cannabis consumption in Europe has only increased rapidly since the 1960s, so in some societies cannabis has been used a lot, in other societies it hasn't been used.
But of course one of the differences is that not only has cannabis consumption become much commoner in Europe and Westernised societies, consumption has been starting at an earlier age and of course then more and more stronger varieties.
So it's increasing consumption, starting early and the availability of much stronger variants.
This is one of the things that puzzled a lot of liberal parents that took cannabis themselves in the 1960s and 1970s.
It never did them any harm.
Of course, they just tended to take a joint when they went to a festival or at weekends and so on and of course this was something that contained 2% - 4% of THC (tetrahydrocannabinol) which is the active ingredient.
Whereas, when their kids started taking cannabis, they though "that's not so bad, at least they are not going out and getting drunk at night", but they didn't know they were smoking every night, several joints and perhaps were then going onto skunk varieties which may contain 10%, 12%, 15% or more of THC.
So the actual cannabis has changed significantly over the last 40 years?
Well you can still get the old fashioned varieties and in a sense they are still the most popular, but people who take a lot of cannabis tend to migrate towards the more concentrated forms, just like the person who just takes a little alcohol tends to take a lager or shandy or glass of wine, whereas the dependent person tends to take vodka or whisky.
It's the same with cannabis. As you become more dependent you need to take more cannabis, then you shift more towards the skunk variety.
Background to the research
I see people with schizophrenia and other psychosis, so people who behave in a bizarre fashion, who think that other people are out to get them, who see or hear things.
So I suppose I saw people like that, but traditionally people with schizophrenia, many of them have had problems right throughout their life.
They've maybe either had some developmental problems, or as children they've been to see child psychiatrists, they didn't get on with the other boys and girls at their school and then they gradually slipped into developing schizophrenia.
What then happened was that I was seeing patients with psychosis who had been very normal as children.
Children who had been great at school, who'd been academically successful, who'd been very sociable, who'd been very sporty, really kids that every parent would be proud of and wouldn't think there were any problems and then they went psychotic. Now that is much more unusual.
In the late 1980s and the early 1990s we started seeing lots more patients who were psychotic who had delusions and hallucinations who were also taking a lot of cannabis.
And at first we thought, well maybe they are taking this because they have so many problems, it's going to help them, they're trying to calm themselves down, maybe it helps their hallucinations.
But then we realised that the people who were taking the cannabis actually did worse. The psychotics who were taking cannabis did worse than the people who were not taking cannabis.
We followed up about 100 people for four years and we found that the people who kept taking cannabis were three times more likely still to be psychotic, still to be hearing voices, still to be being deluded after four years.
So we got to thinking, maybe if once you're psychotic it makes you worse, could it actually have contributed to developing psychosis in the first place?
Cannabis and psychosis
Psychosis is due to an excess of dopamine in the middle of the brain and we know that if you have an excess of dopamine for any reason this can make people psychotic.
Similarly all the drugs that we use to treat psychosis, they decrease dopamine in the mid-sections of the brain.
What does the dopamine system do?
Well dopamine is the chemical which grabs your attention, so if suddenly something exciting happens, then this is mediated by dopamine, you pay attention to something when dopamine is released.
And in psychosis, what people find is that everything seems to catch their attention.
A little noise or the way that somebody grimaces or somebody new comes in the door, or they hear something on the television, all of these things seem to be tremendously important and it's because the dopamine makes them important.
They think, "Gosh, these are really important to me" and begin to connect them.
They think one is connected to the other, they think that television is talking to them, they think the neighbours are conspiring against them.
Stick them all together and they develop a sort of delusional explanation that all of these things are happening to me because I am tremendously important, or because the CIA are ganging up against me.
And it's the dopamine release that causes people to develop these illusions.
And what role does cannabis play in the dopamine system?
Well the cannabis just revs up the dopamine.
We know that from animal studies, there have been fewer human studies, but there are now increasing studies examining that question, exactly what happens when you give cannabis and put somebody in a scanner.
Of course this is one of the reasons why cannabis is useful for people with multiple sclerosis or people with chronic pain.
That is to say one of the beneficial effects is the effect of the dopamine and this can help relax some of the muscles of people with neurological disease.
So this is one of the reasons why we like taking cannabis, we feel it's therapeutically beneficial.
The other function of dopamine of course is that it's a reward chemical, it's a pleasure chemical.
All the drugs that people take, everything from alcohol to cannabis to amphetamines, they all rev up the dopamine system and in most people that just gives pleasure, but some people appear to be vulnerable, that revving up the dopamine system can induce a psychosis.
So the fact that cannabis directly affects the dopamine system suggests to you that it's plausible that it plays a role in psychosis?
Cannabis and genetic susceptibility to psychosis
The only way you can actually find out whether cannabis actually causes psychosis is to start with people who are healthy, just like the only way you can find out whether smoking tobacco causes lung cancer is to start with teenagers who are smoking and follow them through their life.
So really that's what we and other people have done.
It dawned on us as it dawns on everybody, everybody knows people who smoke cannabis every day and there doesn't really have any bad effects on them, so they just get the beneficial effects.
Whereas, a minority of people go psychotic, so there has to be some difference in genetic susceptibility. So the obvious thing to do was to look at the genes which are involved dopamine.
We think that cannabis has it's effect on the dopamine system, therefore we wondered whether people who differed in the genes which metabolise dopamine might contribute to the susceptibility.
There's a gene called COMT and COMT metabolises dopamine, breaks down dopamine in the frontal lobes of the brain.
So what we did was to look at the distribution of the different types of COMT. There's a MET type and there's a VAL type.
People who had a mixture of a MET and a VAL, their risk went up two-fold if they took cannabis.
People who had just a VAL/VAL type of the gene, their risk went up 10 times if they took cannabis in their adolescence
So it made it more clear why some people can just smoke every day for years and they don't go psychotic, but other people are much more vulnerable and they can go psychotic after a much shorter period.
How many are in that very vulnerable group, the VAL/VAL group?
A quarter of the population are VAL/VAL, a quarter of the population are MET/MET and the rest, half the population are a mixture.
How much does it increase your risk if you are VAL/VAL?
Well if you take cannabis, then you would be 10 times more likely to go psychotic if you are a VAL/VAL than if you didn't take cannabis.
Normally being a VAL/VAL has no consequences whatsoever, it's just that it makes people more vulnerable to the effects of cannabis.
So we can say that this is people who took a lot of cannabis, but we also know from the work of one of our colleagues in Holland, Jim van Os, he did an experimental study and he gave cannabis, a standardised cannabis cigarette to volunteers.
Those with the VAL/VAL unit type, they had a much bigger effect on their memory and on their perceptions immediately following taking the cannabis, than the MET/METs.
So is that the end of the story? We know now that some people have a genetic predisposition to mental illness if they smoke cannabis, is that where the research stops?
We know that people vary in their susceptibility to mental illness in general and then cannabis is just one other factor.
I mean I am not, wouldn't want to suggest that cannabis is the major cause of schizophrenia.
Schizophrenia is a bit like heart disease, there are a number of factors that predispose, you get some genetic predisposition, children who have had a difficult birth, who have been hypoxic, short of oxygen at birth have an increased risk and then there are a range of social situations that can increase their risk of psychosis.
So cannabis is just one other factor on top of this that can push people into psychosis.
What we really need to know now, and what we and others are doing is to study the exact metabolism and to know exactly in humans, as opposed to rats, what happens when you do take cannabis on a scanner.
So there are a number of studies now going on looking at exactly what is happening in the brain following taking cannabis.
Cannabis and adolescents
Does it make a difference when you smoke your cannabis on the likelihood of developing problems with mental health?
It seems to, and we don't have definitive knowledge about this, but the way that we have been able to show that cannabis can contribute to the onset of psychosis is by doing big, what we call cohort studies.
By taking a whole series of people in childhood or adolescence and then following them up and comparing the ones who go onto to take cannabis to the ones who don't, because there's always been this argument, maybe the people who go psychotic were always peculiar.
And what we've been able to show is that in a series of about six different studies that adolescents who take a lot of cannabis are more likely to go on and develop psychosis.
And even if you take any of the adolescents who seem to be a bit odd, or seem to have the possibility they might go psychotic and you put them aside, even if you follow up the healthy 15 year olds, in our study a healthy 15-year-old, normal, successful 15-year- old, those who took cannabis on a regular basis, they were three times more likely to develop psychosis by the time they were 26.
Now if they started by 18 they were only about one-and-a-half times more likely to go on to develop psychosis, so it does seem that the earlier you start, the bigger the risk.
Another possibility is if you start early, in our studies, you are more likely to become dependent.
So you start early and you stay with it and it may be that you become more dependent.
So there are these two possibilities.
The other and perhaps more worrying possibility is that if you start while your brain is still developing, we know that in adolescence there are a lot of changes in the dopamine receptors, that these are being sculpted into the final sort of pattern, adult pattern and maybe if you throw cannabis at your brain cells while they are still changing and maturing, maybe this can alter them permanently.
So that is the worrying possibility. We don't know this for sure, but it's clearly a possibility that has to be investigated.
How plausible is that?
We know that from studies of rats that if you give a range of drugs from cocaine, amphetamines or cannabis to adolescent rats then you can alter their neuronal system permanently, yes.
So children, young adolescents who smoke cannabis may be permanently altered in their brain, the way they think?
It's a possibility and again I would say, this is not a couple of puffs, this is taking cannabis every day for a prolonged period.
Psychosis, schizophrenia and cannabis
Schizophrenia is a severe type of psychosis, so by and large many people with psychosis can recover, most people with schizophrenia have it for life, they may have a chronic illness or they may have repeated relapses, so schizophrenia really is a severe type of psychosis.
If there is this change in cannabis use, young people are smoking more cannabis and smoking it earlier and if that cannabis use is associated with an increase risk of mental illness, wouldn't we expect to see a huge explosion in the number of people exhibiting the symptoms of psychosis becoming schizophrenic?
Well, at first I should re-iterate that in our study in New Zealand we found that cannabis accounted for 8% of schizophrenia, so if you'd abolished all cannabis consumption in New Zealand, you'd have got rid of only 8% of schizophrenia.
In Holland, Van Os found it was about 13%.
However these studies were some time ago and the proportion may be increasing.
In fact, in south London where I work, the frequency of schizophrenia has doubled since the 1960s.
Now there may be several reasons for that, but I think drug consumption is one factor.
But it could just as well be that you are better at diagnosing schizophrenia, that people are more alert to the symptoms of schizophrenia?
No, it couldn't.
Schizophrenia is not like the psychiatric conditions which are difficult to diagnose.
People who go psychotic, their friends, their relatives, they notice that they go psychotic, they get taken along to see a psychiatrist and the criteria are fairly clear.
It's not that difficult to diagnose psychosis, so it's not a change in the criteria.
We've actually looked at that question, that it is genuinely is that psychosis in south London is twice as common as it used to be, which is one of the reasons why the mental health services in our big cities are in such a mess.
It's sad to say this, but the treatment that we can offer to people coming into many psychiatric units in our big cities is rather poor and one of the reasons for that is that the service is overwhelmed by the number of people with psychosis and I am sure that drug consumption in general and cannabis consumption contributes to this, not only in increasing the number of people coming, but recycling people through the unit.
Many people who develop psychosis would normally recover, but the people who develop psychosis, recover and then go down the road and meet their dealer and start smoking again.
They come back into the unit, we treat them, they get better, they go out, they come back in.
So our wards are now full of people who are re-cycling through them because they are dependent on cannabis and they don't give up.
So this has meant that the services are under a much greater strain than they used to be.
Does the relationship between cannabis and psychosis become more obvious once somebody's had a psychotic episode?
Yes, once you have been psychotic then your dopamine receptors are sensitised and a number of factors can trigger off the psychosis again and it's quite common that people who have been psychotic, who are partially recovered, for example in our unit, where somebody gets better we say well, why don't you go out for the evening, meet your friends, they come back and the next morning they are psychotic again and if we say, well did you have a joint and they say yes, they are really surprised it could be just one joint.
They say, well it used to be that I smoked every day and I didn't go psychotic, once you have de-sensitised your dopamine system it can be sort of triggered off relatively easily.
Some people have compared the influence of cannabis on mental health and the influence of tobacco on lung cancer, do you think that is a fair comparison?
I think it's perhaps more like alcohol actually than smoking, because cigarette smoking account for nearly all of lung cancer, whereas cannabis accounts only for a proportion of schizophrenia.
Most people can drink alcohol and they don't get liver failure, they don't become an alcoholic, the more they drink, the more likely they are to develop problems and similarly with cannabis, that most people will get nothing but benefits from smoking cannabis, but a proportion of people will develop problems particularly if they smoke very heavily.
You talk as if it's certain that the causal direction... the fact that cannabis may have a causal role in some mental illness is a matter of fact, yet the studies are actually very small populations of people who actually developed psychosis. Were they statistically significant, how robust is this research?
Well the studies are not all small.
The Swedish army study was a study of 50,000 men inducted into the Swedish army and then followed up for 25 years and they found that several hundred people went psychotic. And the people who had smoked cannabis when they were 18 heavily, they were six times more likely to go psychotic.
Some of the other studies have been smaller and more intensive. But it's relatively unusually in epidemiology for all the studies to point in the same direction and since 2002 there have been six studies and they've al pointed in the same direction.
Now I would accept that epidemiological studies cannot absolutely prove something.
In order to prove that, if we go back to the link between cigarette smoking and lung cancer, the epidemiological studies could show a link, but it wasn't until animal studies were done, where rats were made to smoke cigarettes and they were shown to develop lung cancer.
We really have to do some equivalent studies where we actually understand the mechanism in the brain in much greater detail than we understand it at present before it can be 100% proven.
How confident are you that cannabis plays a role in the development of some mental illness, some psychosis?
Oh I think I am much more confident.
I would say I am convinced. Why am I convinced? Partly because of the evidence, but partly because of the patients I see and also the fact that when people who are psychotic, when they stop taking cannabis you can have such a dramatic improvement, so it seems fairly logical, if you stop doing something and you improve, maybe it contributed to it in the first place.
You say that the epidemiological studies don't give us proof as yet, so are you convinced that cannabis plays a causal role in the development of some mental illness?
I think that as a clinician one pays a lot of attention to the epidemiology, but one also pays attention to the patients one sees and what they and their relatives tell one.
And in particular I've been impressed by some people with very severe psychosis, who I really was rather pessimistic about their future, but then they stopped taking cannabis and they did so much better.
So if stopping banging your head against the wall relives the pain, well maybe banging your head against the wall actually contributed to the pain in your head.