744 P. R o b s o n
presence or direction of causality. Five prospective studies have been subjected to
critical review (Arsenault et al. 2004). The authors’ conclusion was that cannabis
smoking by young adolescents confers an overall twofold increase in the risk of
developing schizophrenia. However, they state that “cannabis use appears to be nei-
ther a sufficient nor a necessary cause for psychosis. It is a component cause, part
of a complex constellation of factors leading to psychosis”. They further conclude:
Although themajority ofyoungpeopleareabletousecannabisinadolescence
without harm, a vulnerable minority experiences harmful outcomes. The
epidemiological evidence suggests that cannabis use among psychologically
vulnerable adolescents should be strongly discouraged by parents, teachers
and health practitioners alike.However, the five studies reviewed in this paper have been criticised elsewhere for
methodological shortcomings including: presence of clinical or sub-clinical psy-
chiatricillnesspriortocannabisconsumption;lackofacleartemporallinkbetween
cannabis use and subsequent psychiatric illness; poor reliability of the diagnosis of
schizophrenia; confusion between acute toxic states and functional mental illness;
confusion of association with causation; confounding effects of other recreational
drugs and environmental risk factors for mental illness; unreliability of self-report
of an illegal activity; and a lack of a correlation in epidemiological studies between
prevalence of cannabis consumption and schizophrenia. The UK Advisory Council
on the Misuse of Drugs (ACMD) reviewed the evidence in depth and concluded
(2002, p. 8) “... no clear causal link has been demonstrated.” Degenhardt and Hall
reached a similar conclusion (2002): “Time trends in schizophrenia and cannabis
use are not consistent with the hypothesis that cannabis use causes schizophrenia
de novo.”
In conclusion, the link between functional mental illness and recreational
cannabis use in previously healthy subjects with no psychiatric history remains
controversial, and a causative link has not yet been established. However, it would
seem advisable for individuals with existing psychiatric illness or a strong family
history to avoid THC-containing medicines.
4
Future Directions
Notwithstanding all the hard work summarised above, the scientific evaluation
of medicinal cannabis in humans is in its infancy. The role of cannabis-based
medicines in all the clinical indications so far discussed requires clarification
throughfurtherwell-controlled,adequatelypoweredrandomisedtrials.Therapidly
expandingknowledgeofthestructureandfunctionoftheendocannabinoidsystem
raises the hope of exciting new pharmacological entities. To give a few examples: It
may be possible to enhance the activity of endocannabinoids by inhibiting degra-
dation mechanisms such as fatty acid amide hydrolase, and since there appears
to be local up-regulation of endocannabinoids in certain pathological conditions,
this gives the added possibility of site selectivity (Baker et al. 2001); the discovery