Next week sees the upgrading of cannabis to its new position as a “special” class B drug which, unlike all other class B drugs, will see “offences” of possession dealt with by a warning or – if they can get the measure approved in time – by a spot fine of £80. UKCIA isn’t alone in thinking this looks more like a a downgrading than an upgrading, but a “strengthening” of the law is what it’s being dressed up as, so we’ll go with that for now. Next week will probably be a better time to comment on the move to “Bspecial” so we’ll leave it as something to look forward to.

But for now, it’s opportune to take a look at some of the evidence the government has about the cannabis trade in the UK, which will throw some light on just how little they really know about the euphemistically termed “controlled” drugs scene.

A week or so back the ACMD study into cannabis potency was published- known as the “HOME OFFICE CANNABIS POTENCY STUDY 2008” (PDF here).

Now, we would naturally expect that any properly controlled drug would be subjected to rigorous standards of manufacture, so that the authorities would know exactly was was being sold. But of course, illegal drugs aren’t like that, not being controlled in any meaningful sense of the word.

But it would be reasonable at least to expect that any product on sale in anything like as widespread a way as cannabis is would at least be properly monitored so that we know what is on sale, how strong it is, what form it takes and so on. Fact is though, with illegal drugs including something as common as cannabis, we don’t know any of these things.

So this Home Office study set out to fill in some of the blanks in our knowledge of just what isn’t really being controlled.  Now the first problem they had is how to collect samples. For any study that wants to be taken seriously, the sampling method used is all important. You have to be sure of taking a representative sample or your results will be meaningless. The ACMD had no choice but to ignore this important prerequisite and to go with the best they could get, which was police seizures from people who had been given a warning for possession.

On the subject of data collection the report has this to say:

The numbers of samples from each force should not be seen as a measure of the ‘cannabis problem’ in their area.

Translation: The number of samples from an area can’t be considered representative of cannabis trade in that area. Now if this sampling method had any validity it would perfectly match the “cannabis problem” in those areas, if it doesn’t, the data is rubbish. The reason given for this was:

For operational reasons some forces chose to send in material from only one Borough Command Unit or from one of several forces collection points. Some forces experienced internal logistics problems; others were very enthusiastic and sent in everything received during the trial period.

So the data collection is at best not very good, which would make any kind of real statistical analysis very suspect indeed on the basis of the mantra of  “garbage in, garbage out”.

Simple yet very important criticism of the data collection would include:

* Depending on on the police catching someone, hence the type of people caught and the environment they are caught in is likely to be not representative of cannabis users in that area.

* Several samples could have come from the same area at about the same time, therefore coming from the same batch and so on, whereas other areas or user profiles went undetected.

People more qualified than I could no doubt make many more a detailed criticism of the data collection method .  It is normal for any scientific study to address the weaknesses of the study, to identify them and make some estimate as to how they could have skewed the results. It is very noticeable that this Home Office study makes no attempt to do so, this is a serious failing.

Having collected the “data”, the sample were then analyised.

Initial laboratory examination showed that 80.8% were herbal cannabis and 15.3% were cannabis resin.
The remaining 3.9% were either indeterminate or not cannabis

So nearly 4% of people were given a warning for possession of something that wasn’t cannabis? That’s pretty bad really.

Before any kind of analysis can be done, an important term has to be defined:

The potency of cannabis is defined as the concentration (%) of Δ9-tetrahydrocannabinol (THC)

That is a partial and wholly unacceptable definition of potency. What it specifically does not mean is the “concentration” of THC in the whole sample as seized. The reason for this is that herbal cannabis – and to a lesser extent hashish – contains a lot of water and plant material. As the report goes on to explain somewhat badly, the samples were subjected to an extraction process. This was done to remove the oils from the sample – the oils of course contain the THC. They have therefore made no measurement of how much oil there was in the sample, so a sample containing mostly vegetable matter which yields a tiny amount of oils could well produce a very high potency result, yet actually be quite weak.

The relationship of “potency” to “strength” – if there indeed is one – isn’t explained.

What potency does mean though is the amount of THC  in comparison with everything else in the oils of the plant. That could be a useful thing to know if we know what the other constituents are, but otherwise it isn’t really a lot of use. One other ueful figure to know of course would be the “potency” of CBD and, for the first time, that was also measured, but there are no historical recordings of CBD content to show any changes.

So what of the results?

ACMD Potency results

The mean THC concentration (potency) of the sinsemilla samples was 16.2% (range = 4.1 to 46%). The median potency was 15.0%

The Sensi samples are the blue bars. A range of 4.1 to 46 percent is almost unbelievable but if true, shows the shockingly unpredictable nature of the cannabis trade, that is a huge range.  Actually, the results are questionable because a potency of 46% is very high, probably about as high as nature can produce (according to the ACMD presentation given back in March last year). The distribution curve also shows an unexplained “tip up” at the high end which is not a feature of a normal distribution. Something is wrong there and bad data is the most likely explanation.

The results for the “traditional imported” cannabis (green bars) can probably be ignored, based as they are on what the study accepts is a very small sample size. The hashish results (yellow bars) are worth a comment though

The mean potency of cannabis resin was 5.9% (range = 1.3 to 27.8%). The median = 5.0% was typical of values reported by others over many years.

A quick look at the results for resin in the graph above shows a normal distribution curve, which is to be expected, but with a couple of data points outside the normal curve. A possible explanation for this is that these are samples of “bubble hash” – hash made from “home grown” plants. It should be remembered that all forms of cannabis can be used to make hash, it’s just that traditionally the plants used to do so would have produced a product too weak to be used as the herbal variety. This could be one explanation as to why “old skool” hash – good old “rocky”, “Leb” and so on, seems to have had a lower potency and probably contained higher levels of CBD. Hence:

Cannabis resin had a mean CBD content of 3.5% (range = 0.1 to 7.3%), but the CBD content of herbal cannabis was less than 0.1% in nearly all cases.

They also report that

There was a weak, but statistically-significant, correlation (r = 0.48; N = 112; P < 0.001) between the THC and the CBD content of resin.

So the CBD concentration mirrored THC concentration to an extent. This could possibly imply the samples all came from the same same producer region (probably Morocco) and hence had originally the same profile, but have since been diluted. These were maybe samples of so-called “soap bar” hash.

The report notes that

It will be seen from Figure 8 that three samples of herbal cannabis had anomalously high CBD values. The reason for this is not understood.

It is, of course, entirely possible for a plant to produce high values of THC and/or CBD, or more likely there’s something wrong with the data.

That is just about the extent of our knowledge of the UK cannabis industry and the product it supplies. In all honestly, it’s not very good but it’s about as good as we can expect under prohibition.Of course, if cannabis were legal and properly regulated we would know exactly what was on sale, where it was on sale from and everything else. Controlled drugs? Don’t make me laugh.

Footnote and one last gripe. One of the terms long used in the cannabis culture is “Sensi” – meaning cannabis grown without seeds. This is a long established word and is well understood across generations of cannabis users. However, the ACMD and the Home Office, having no connection with the “street” culture they are investigating use they use the term “sinsemilla”. “Sinsemella” is derived from the Spanish meaning “without seeds”, so while it is perhaps strictly speaking correct, it isn’t the term used. This is an annoying detail perhaps, but they should be aware that the type of cannabis most commonly available is  known as “sensi” and that “Sensi” has always attracted a premium and is certainly not a new development.