Pontifications on Poison
Being some ramblings on events associated with poisonous plants.
Friday 29th July 2011
Yesterday, I did some blatant cherry-picking from the British Crime Survey’s ‘Drug Misuse Disclosed’ report and, today, I’ll do the same for the NHS report entitled ‘Smoking, drinking and drug use among young people in England in 2010’.
I mentioned that it is hard to do a direct read across from one report to the other so it is not easy to see how attitudes formed by the 11-15-year old group develop or change for the 16-24-year olds. One of the constraints is that the BCS covers England and Wales whereas, as its title states, the NHS report is only about England.
As an aside, given that health is dealt with by the devolved administrations, it might be time to stop talking about a National Health Service and instead refer to the EHS, WHS, SHS and NIHS to reflect that health policy may be different in England, Wales, Scotland and Northern Ireland.
The first thing I noticed in the summary of the report is a tendency to want to produce a good read rather than presenting realistic comparisons. Thus, the summary states ‘It is estimated that in 2010 around 150,000 young people in this age group were regular smokers, around 400,000 had drunk alcohol in the last week, around 200,000 had taken drugs (including glue, gas and other volatile substances) in the last month and around 380,000 had taken drugs in the last year’*.
*Copyright © 2011, Health and Social Care Information Centre, Lifestyles Statistics.
You have to look elsewhere to establish that ‘regular smokers’ are those who smoke at least once a week. This may lead you to conclude that the 150,000 regular smokers is a direct comparison with the 400,000 drinkers but that is not the case. There is a difference between drinking in the last week and drinking at least once a week. In fact the survey finds that 13% of respondents had drunk in the last week but only 8% said they normally drank at least once a week. That suggests that the direct comparison is between 150,000 ‘regular’ i.e. weekly smokers and 250,000 ‘regular’ drinkers.
Then, when it comes to drugs, the switch is from weekly use to monthly prevalence. I’m afraid I suspect that the weekly prevalence doesn’t give a dramatic enough number for the report’s purposes.
As far as accuracy is concerned, I do have concerns about whether this survey achieves a high enough standard. Unlike the BCS survey, which involves participants entering responses directly into a computer, the NHS survey is conducted on paper and I can’t help thinking that some pupils would be concerned about their responses being identifiable from their writing. (Though most responses are of the ‘tick box’ variety, there are some occasions where a number of word has to be entered.) Like the BCS, the survey includes Semeron, a fictional substance, and the conclusion that responses are honest because only 13 of the 7,296 respondents said they had used it could be false. It may simply be that the respondents know enough about drugs to recognise that Semeron is a phony.
All these concerns make me think that, perhaps even more so than with the BCS survey, only trends may be useful rather than absolute numbers on an annual basis.
And because some data is presented by age, it is possible to examine whether the reported trends ‘feel’ right. Since last year’s 14-year olds, for example, become this year’s 15-year olds, it may be possible to see if there is an inconsistency in the develop of both usage of substances and attitudes towards them. Take the number of regular smokers at age 14 versus 15. You would expect the number of smokers to increase with age, and it does, but the increase varies from 15% to 67% year on year but with no trend.
The last week prevalence of drinking alcohol seems to display an odd result. In every year, previously, the figure for 15 year-olds was higher than 14-year olds by at least 9 percentage points and up to 15 points. But the 2010 15-year olds have only 5 percentage points higher prevalence than 2009’s 14-year olds. The lower prevalence figures that seem to apply throughout for 2010 may, therefore, be an anomaly and, to be fair to the authors, they do seem to acknowledge this by talking about the need to see if next year’s results continue the trends.
This is a 233 page report against ‘Drug Misuse Disclosed’s 47 pages so there is a great deal of detail but I can’t help wondering if the data has been over-analysed. I may return to specific parts of it in coming blogs but, for now, I’ll just return to that misleading paragraph I quoted in full, above, and try and find proper comparisons between smoking, alcohol and drugs.
Looking just at last week prevalence, 7% of respondents reported smoking, 13% reported drinking alcohol and X% reported using drugs. I’ve put ‘X%’ because the survey does not collect weekly prevalence figures for drug use. The monthly prevalence for drug use, at 7%, indicates that both smoking and drinking alcohol are bigger issues for the education system than drugs. Though there is a clear trend toward reduced smoking and drinking, there is no room for complacency.