I don’t pay a lot of attention to statistics about psychoactive substances. They are often based on incomplete data or skewed research conducted to ‘prove’ a prejudice concerning prohibition. And they can’t alter my central belief that substances that have the potential to do harm should not be left in the hands of criminals rather than being regulated by governments. Only when governments accept responsibility for minimising the harms caused is the true extent of those harms worth more scrupulous study.
But there are numbers that bounce around and I, sometimes, wonder how reliable they are. One such is the number of cannabis users who will become dependent (often described as ‘addicted’). The number favoured by drug aunt website ‘Smart Approaches to Marijuana’ is 1 in 10 but this is a rounding up from the 9% (or 1 in 11) number usually given.
My temptation is always to consider the rest. If only 9% become dependent, then 91% are not suffering serious harm. (If we could say that 91% of people are not obese, we wouldn’t be seeing any fuss made about the problems of being overweight.) But, a piece by Sunil Kumar Aggarwal for Huffington Post looking at the origin of the 9% claim, interested me.
When reading Aggarwal’s piece it is important to remember that, in the USA, healthcare is a business rather than being about caring for people’s health. Like any business, it strives to create demand for its products and, rather like the makers of Listerine who created a condition called halitosis in order to sell a product to combat it, healthcare providers in the USA have sought to create conditions in order to sell the treatment to go with them.
The American Psychiatric Association’s Diagnostic and Statistical Manual (DSM) defines the symptoms to look for when diagnosing a huge range of mental health issues. The most recent revision to this document came in for a lot of criticism over its attempt to create all manner of addictions (though not the addiction to creating new profit opportunities) but it was the 1987 revised third edition that led to the 9% statistic.
Aggarwal demonstrates how a survey, conducted between 1990 and 1992 and based on diagnostic criteria from DSM3, used biased questions to measure dependence and biased attitudes to reduce the role of alcohol in respondents’ problems.
I’m not a statistician. I did get a good grade in ‘A’ level pure maths in 1968 but Mr Phasey, whose catchphrase ‘By the well-known formula’ would produce blank stares and blushes more often than not, would be appalled at how little of it I’ve retained. But I do like to play around with numbers and see if my residual knowledge and common sense can shed light on the reliability of numerical claims.
I decided to see what it would mean if the 9% claim was true for the UK. It is important to stress that what follows has numerous broad assumptions and that, at times, figures for different years have been mixed because I couldn’t find all the data I wanted for one single year. But given that year on year variations are not dramatic, I think it is possible to reach broad conclusions based on broad assumptions.
It seems to be reasonable to say that you’d need to be using cannabis frequently to even be considered as a candidate for the term ‘dependent’. The European Monitoring Centre for Drugs and Drug Addiction (EMCDDA), in its 2012 report ‘Prevalence of daily cannabis use in the European Union and Norway’, says, in Table 2 on page 16, that, for all adults in the UK, 3.8% report last month prevalence and 15% of those monthly users report use at more than 20 days a month. 0.6% of all adults report daily use.
The CIA World Factbook estimates the UK population as 41 million 15-64 year olds.
3.8% of 41 million is 1.5 million and 15% of that is 225,000 people using cannabis more than 20 times per month.
0.6% of 41 million is 246,000 daily users.
Clearly, there are approximations at work because the number of daily users has to be lower that the number of people using more than 20 times a month but less than daily. Rather than try and resolve the discrepancy I’m going to proceed on the basis that there are between 200,000 and 250,000 very frequent cannabis users in the UK.
According to the most recent Crime Survey for England and Wales (CSEW), 30% of adults have used cannabis at least once in their life. The CSEW deals only with England & Wales but if we apply that 30% to the whole of the UK we get 12 million adults in the UK who have used cannabis at least once in their lives. If 9% of users become dependent, as claimed, then there should be 1,000,000 dependent users in the UK. Instead there are 200-250,000 very frequent users. The next stage is to see whether even that number of very frequent users can be accurately described as dependent.
The National Treatment Agency for Substance Misuse (NTA) said in its 2012 annual report that, during 2011/2, 15,750 people received treatment for cannabis dependence. 10,544 of those were new entrants. These figures are for England only. Scaling that number to take account of Wales, Scotland & Northern Ireland suggests that around 20,000 people in the UK are in treatment for cannabis dependence.
The NTA makes a broad brush assumption that half of all heroin users are in treatment so, applying that estimate to all substances, shouldn’t there be 100-125,000 cannabis users in treatment? The UK now has one of the most efficient treatment systems in Europe so it won’t do to suggest that there is a huge waiting list of dependent cannabis users waiting to enter.
As far as the UK is concerned, then, the 9% claim falls on two counts; the number of very frequent users is in the region of 1-2% of adults who have ever used cannabis and, based on numbers in treatment, dependent users account for only 0.2% of ever users.
Going back to looking at the majority percentage, it would seem that 99.8% of cannabis users do not become dependent. That number makes current policy look exceedingly stupid.
You can send comments via the contact page but please be sure to say what blog entry you are commenting on.