742 P. R o b s o n
is withdrawn, difficulty in keeping consumption under control, and a preoccupa-
tion with the drug that interferes with the normal activities of living.
Tolerance to the subjective effects of marijuana has been reported (Georgotas
and Zeidenberg 1979), and a minority (16%) of regular smokers experienced at
least one of the following symptoms following abrupt withdrawal of cannabis:
irritability, insomnia, tremor, sweating, gastro-intestinal disturbance or appetite
change (Wisbeck et al. 1996). These effects peak between 2 and 6 days after abrupt
withdrawal (Budney et al. 2003). It has been reported that a third of regular
users experienced some difficulty in controlling their use of the drug (Thomas
1996). All research in this area is dogged by serious methodological problems,
including highly selected samples, non-validated measures, poor response rates in
community surveys, and the existence of many confounding variables. However,
it seems reasonable to accept that psychological dependence will occur in a small
minority of cannabis smokers. The existence of a clear-cut physical dependence
syndrome is much less convincing on the basis of the published literature. If it
exists at all, it is probably mild and transient, and is likely to consist of a few days
of sleep disturbance and somatic symptoms of anxiety in heavy daily users who
abstain abruptly.
In an interview study (Robson and Bruce 1997), the dependence potential of
various street drugs was assessed in 201 problem and 380 “social” users of heroin,
cocaine or amphetamine using the well-validated Severity of Dependence Scale
(SDS). Scores (maximum = 15) in the problem group were 12.9 for heroin, 9.6 for
other opioids, 6.1 for amphetamine and 5.5 for crack cocaine. All of these scores
were consistent with findings in other studies. Cannabis SDS score was 2.6 and
comparable with those of LSD (3.1) and ecstasy (1.3), two drugs that are generally
not associated with physical or psychological dependence. In the parallel sample
of social users, the cannabis SDS was similar at 3.4.
Attempting to define and investigate cannabis dependence in patients is still
more challenging, especially if the individual is experiencing a beneficial thera-
peutic effect. Developing an emotional attachment or preoccupation with a drug
that has helped with previously intractable, life-impairing symptoms is a very dif-
ferent matter from becoming over-preoccupied with a recreational drug. It would
hardly be surprising for a patient abruptly denied such a medicine to yearn for
it and become preoccupied with re-establishing a supply. Is the diabetic addicted
to insulin? Experience in the therapeutic setting with much more powerfully ad-
dictive drugs than cannabis is encouraging. For example, the abuse of opiates is
extremely unusual among patients treated appropriately for pain and other symp-
toms (Porter and Jick 1980; Portenoy 1990), and this is very likely to be the case
with cannabis-based medicines. Support for this is provided by the intoxication
data from a recent study (Wade et al. 2004) comparing a THC-containing cannabis
extract (Sativex) with placebo for a 6-week treatment period in patients with MS.
At the end of the trial, all patients re-titrated on to the active medicine for a fur-
ther 4 weeks. Intoxication scores were recorded in a daily diary on a 100 mm VAS
scale shown in Fig. 1. Average peak scores reached only around 20/100, and levels
appeared to diminish over time. There was no evidence that Sativex was abused by
any of these patients.