Formulating policy in relation to drugs is obviously quite a difficult thing to do. I comment on it, as I always have, from the perspective of a psychiatrist who is interested in drugs and drugs and the brain.
We have a range of expertise on the Advisory Council on the Misuse of Drugs (ACMD); we're very strong in terms of chemistry and pharmacology, and psychology; and we have a definite knowledge, interest and responsibility to look at social harms as well. We provide one arm of the policy-formulating perspective. In addition, there are a number of other agencies, organisations and individuals who contribute to policy formation.
There are also international partners – we have signed up to international treaties – which determine that, in essence, the UK follows United Nations policy on drugs. This can be quite a tough constraining influence on how countries regulate drugs (although some countries, such as the Netherlands, have managed to be more flexible, even though they still sign up to the international conventions).
Then, of course, there are other factors feeding into political decisions about drugs: what the general public thinks (or is thought to think); and then there's the media. In recent years, the whole process of determining drug classification has become quite complex and highly politicised.
Cannabis – a potent problem
I am going to focus on cannabis because it is the only drug that has been downgraded in the history of the 1971 Misuse of Drugs Act, an interesting point in itself. The issues relating to cannabis pose a challenge to whether the act is working as it was originally intended.
The ACMD was requested by the home secretary in 2007 to review the status of cannabis because: "Though statistics show that cannabis use has fallen significantly, there is real public concern about the potential mental health effects of cannabis use, in particular the use of stronger forms of the drug, commonly known as skunk."
So there was a skunk scare. Cannabis had gone from class B to C, but, supposedly, skunk use had been increasing and it was getting stronger, so we were asked to review whether the decision to go from B to C was still appropriate. In our report we came to several conclusions:
• Cannabis is a harmful drug and there are concerns about the widespread use of cannabis amongst young people.
• A concerted public health response is required to drastically reduce its use.
• Current evidence suggests a probable, but weak, causal link between psychotic illness and cannabis use.
• The harms caused by cannabis are not considered to be as serious as drugs in class B and therefore it should remain a class C drug.
There has been a lot of commentary and some research as to whether cannabis is associated with schizophrenia, and the results are really quite difficult to interpret. What we can say is that cannabis use is associated with an increased experience of psychotic disorders. That is quite a complicated thing to disentangle because, of course, the reason people take cannabis is that it produces a change in their mental state. These changes are akin to being psychotic – they include distortions of perception, especially in visual and auditory perception, as well as in the way one thinks. So it can be quite hard to know whether, when you analyse the incidence of psychotic disorders with cannabis, you are simply looking at the acute effects of cannabis, as opposed to some consequence of cannabis use.
If we look on the generous side there is a likelihood that taking cannabis, particularly if you use a lot of it, will make you more prone to having psychotic experiences. That includes schizophrenia, but schizophrenia is a relatively rare condition so it's very hard to be sure about its causation. The analysis we came up with was that smokers of cannabis are about 2.6 times more likely to have a psychotic-like experience than non-smokers. To put that figure in proportion, you are 20 times more likely to get lung cancer if you smoke tobacco than if you don't.
There is a relatively small risk for smoking cannabis and psychotic illness compared with quite a substantial risk for smoking tobacco and lung cancer.
The other paradox is that schizophrenia seems to be disappearing (from the general population) even though cannabis use has increased markedly in the last 30 years. When we were reviewing the general practice research database in the UK from the University of Keele, research consistently and clearly showed that psychosis and schizophrenia are still on the decline. So, even though skunk has been around now for 10 years, there has been no upswing in schizophrenia. In fact, where people have looked, they haven't found any evidence linking cannabis use in a population and schizophrenia.
I want to move on now to look at how people gather information about drugs and the challenges of communicating the best evidence relating to drug harms to the public. This is difficult in the face of what you might call a peculiar media imbalance in relation to drugs. The following data illustrates a remarkable finding. It derives from the PhD of a Scottish graduate, Alasdair JM Forsyth, who looked at every single newspaper report of drug deaths in Scotland from 1990 to 1999 and compared them with the coroners' data.
Over the decade, there were 2,255 drug deaths, of which the Scottish newspapers reported 546. For aspirin, only one in every 265 deaths were reported. For morphine, one in 72 deaths were reported, indicating that editors were not interested in this opiate. They were more interested in heroin, where one in five deaths were reported, and methadone, where one in 16 deaths were reported.
They were also more interested in stimulants. With amphetamines, deaths are relatively rare at 36, but one in three were reported; for cocaine it was one in eight. Amazingly, almost every single ecstasy death – that is, 26 out of 28 of those where ecstasy was named as a possible contributory factor – was reported. So there's a peculiar imbalance in terms of reporting that is clearly inappropriate in relation to the relative harms of ecstasy compared with other drugs. The reporting gives the impression that ecstasy is a much more dangerous drug than it is. This is one of the reasons I wrote the article about horse riding that caused such extreme media reactions earlier this year. The other thing you'll notice is that there is a drug missing, and that's cannabis. Also missing is alcohol, which will have killed a similar number – 2,000-3,000 people – in Scotland over that time, maybe more. Of course, cannabis wouldn't have killed anyone because it doesn't kill. And that's one of the reasons why we thought cannabis should be class C, because you cannot die of cannabis overdose.
We've tried very hard for at least the last 10 years to put together a structure for assessing drug harms. This eventually became a research paper, Development of a Rational Scale to Assess the Harms of Drugs of Potential Misuse, published in the Lancet in 2007. Despite – or perhaps because of – its novelty and remit, it was very hard to get a paper published that challenged some of the current (mis)perceptions about drugs.
In principle, we broke down drug harms into the following parameters: physical harm (acute, chronic and intravenous), dependency (intensity of pleasure, psychological dependence, physical dependence), and social harms (intoxication, other social harms and health-care costs).
We looked at all the drugs in the Misuse of Drugs Act and added some others that weren't already covered by it. For example, we included ketamine, which wasn't covered by the act at the time, solvents, and tobacco and alcohol, because we thought it was very important that harms of illicit drugs were assessed against the harms of drugs that people know and use. This analysis eventually established a ranking order presented opposite.
A number of important points emerged. The ranking suggested that there are clearly some very harmful drugs (you might say these would be class A drugs) and there are some drugs that aren't very harmful, such as khat or alkyl nitrites, which aren't controlled by the act at all.
Interestingly, some class A drugs scored much lower than other class A drugs, suggesting that there is some anomaly in terms of that part of the current statutory classification system.
The ranking also suggests that a tripartite classification system might make sense, with drugs ranking as more harmful than alcohol being class A and those ranking lower than tobacco as class C. The exercise also highlighted how dangerous alcohol is. I believe that dealing with the harms of alcohol is probably the biggest challenge that we have in relation to drug harms today.
One problem is that sometimes you get into what I think of as an illegality–logic loop. This is an example of a conversation I've had many times with many people, some of them politicians:
MP "You can't compare harms from a legal activity with an illegal one."
Professor Nutt "Why not?"
MP "Because one's illegal."
Professor Nutt "Why is it illegal?"
MP "Because it's harmful."
Professor Nutt "Don't we need to compare harms to determine if it should be illegal?"
MP "You can't compare harms from a legal activity with an illegal one."
I have been surprised how difficult this concept is to get across to some people, whether they are politicians, fellow scientists or members of the general public.
This is an edited extract from a July 2009 lecture by Professor David Nutt, a transcript of which was published last week by the Centre for Crime and Justice Studies at Kings College London. crimeandjustice.org.uk/estimatingdrugharms.html
November 3, 2009