Apples and oranges, beer and ecstasy
Most people commenting on the sacking of the government’s drug czar, Professor Nutt, have taken the view that one has to choose between science and prejudice.
However, I think he can be criticised also on the basis of bad science.
I admit I haven’t read any of his reports, so my comments are based on quotes in articles, and they might of course have been misleading.
It seems to be well-established that he thinks horse-riding is more dangerous than ecstasy:
Nutt wrote in the Journal of Psychopharmacology that 10 people in Britain died a year from horse riding — or “equasy” as he called it — and that it was associated with more than 100 road traffic accidents annually.
“Based on these harms, it seems likely that the ACMD would recommend control (for equasy) under the Misuse of Drugs Act perhaps as a Class A drug given it appears more harmful than ecstasy,” he said.
Between 35 and 50 people die each year in England and Wales from ecstasy, currently ranked as one of the most dangerous Class A drugs, along with heroin and cocaine, according to government figures.
But Nutt said ecstasy was proportionately less dangerous, causing acute harm only in one in every 10,000 cases, compared to one in 350 cases for horse riding.
Unless I’m misunderstanding him, he seems simply to count the number of deaths and accidents without comparing them to the number of people who take part in these activities
Although I haven’t found a good quotation, it seems to me he’s doing the same for alcohol, i.e., he says it’s really dangerous because it kills so many alcoholics and causes so many accidents.
However, alcohol is used by the vast majority of the population, and on a very regular basis by many of them.
So to compare alcohol with cannabis, or horse-riding with ecstasy, you have to find two comparable groups, e.g., 100 teenagers drinking alcohol but not using cannabis at all and another 100 teenagers using cannabis but never touching alcohol.
Otherwise you’re comparing apples and oranges, which is useless.
If Professor Nutt had it his way, alcohol would be completely banned, and the entire population would move to the new legal drugs such as ecstasy. After a few years, there wouldn’t be any deaths caused by alcohol any more, but hundreds of people would be crashing their car after taking ecstasy, and his classification would have to be changed.
Firstly, I know nothing about this guy, but I’d trust a person going by the name Professot Nutt any day 🙂
Secondly, concerning whether he compared the actual numbers, I don’t read the quite like that; I read that horses cause 10 deaths and Ecstasy 35-50 (though in different regions) in absolute numbers. On the other hand (last paragraph of quote), horses cause deaths once out of 350 (maybe only one in 3850 assuming the 350 number is from both deaths and traffic accidents) whereas Ecstasy only causes death once in every 10000.
In this case he is sort of correct, even though the results may be polluted by the intersection of horse riders and Ecstasy users. Most likely that intersection is negligible, though.
Also, I believe that alcohol actually does cause more deaths proportionally (didn’t check it).
I know it’s possible that Nutt is right and I am wrong — I should look up his actual publications, I guess.
However, I still think it would be useful to know how many riders and ecstasy users there are before starting to compare mortality rates.
Also, I’m sure alcohol does cause many deaths, but given that it’s the only ubiquitous legal drug, it’s no wonder it’s causing many problems. How does our society compare with a country that has a different legal drug (e.g., khat)?
According to Wikipedia (http://en.wikipedia.org/wiki/List_of_preventable_causes_of_death) around 85000 dies pr year because of alcohol, yielding around a 1 in 3500 chance of dying of alcohol (assuming around 300 mio people use alcohol in the US (the entire population is around 307 mio according to Wikipedia), so this is underestimating the risk. This, of course, is a completely different country, so reasonable comparison is probably not justified.