APPENDIX 4
Memorandum from Rethink
EXECUTIVE SUMMARY
Cannabis is the most widely used illegal drug
in the UK. There is a now a substantial body of scientific research
indicates a positive relationship between cannabis use in adolescence
and the later onset of psychosis; other research establishes a
positive relationship between cannabis use and relapse by people
with psychotic illnesses; a small amount of research has now been
undertaken on the relationship between cannabis use and wider
mental health problems.
The UK Government has not contributed to the
expansion of this evidence base, either through commissioning
evidence or through funding applications submitted to it. It took
a significant decision to reclassify cannabis to a class C drug
in 2004, but did not use the subsequent information campaign to
communicate the evidence. A project to provide information materials
to people with mental illness was vastly under-funded and flawed.
School drugs education has not covered this evidence either. The
body charged with reviewing evidence by the Government, the Advisory
Council on the Misuse of Drugs, was only asked to review this
evidence following media pressure.
This poor record has been followed in January
2006 by a decision to keep cannabis at class C, invest further
in health education and review evidence on the recommendation
of ACMD and the majority of mental health and drug charities.
Rethink believes that the failures of Government
policy to reflect the evidence base has been due to a number of
factors, including the politicisation of the issue, the cross-cutting
nature of the issue, a reliance on single experts in departments
and professionals rather than people with mental illness and their
carers to direct policy.
Hence, we make the following recommendations.
Recommendation 1:
Guidance to civil servants and Ministers stress
the importance of considering and commissioning evidence on all
aspects of cross-cutting issues.
Recommendation 2:
Guidance to civil servants should stress the
importance of consulting a variety of resident experts and practicing
professionals.
Recommendation 3:
A mechanism be created through which service
users, carers and organisations representing these groups can
inform Departmental decisions on research funding, commissioning
and determining future priorities.
Recommendation 4:
Users of drug and mental health services, their
carers and organisations representing these groups to be included
in the make-up of committees such as ACMD.
Recommendation 5:
Guidance to civil servants to stress the need
to consult service users and carers as well as professionals,
at all stages of the policy-making process.
Recommendation 6:
A mechanism be created through which service
users, carers and organisations representing these groups can
inform Departmental decisions on research funding, commissioning
and determining future priorities.
Recommendation 7:
The advice given by Government-appointed bodies
such as ACMD and Government policy to be regularly evaluated by
external organisations.
INTRODUCTION
Rethink is a national charity, with over 8,000
members comprising both people with severe mental illness and
their carers. Rethink operates almost 400 services for people
with severe mental illness across England and Northern Ireland.
Rethink has campaigned on the issue of cannabis
for many years because our members have consistently identified
it as a major issue and because of emerging scientific evidence
linking cannabis and mental illness.
Cannabis is the most widely used illegal drug
in the UK—rates of use among young people are said to be
falling a small amount (Health and Social Care Information Centre,
2005), but are still the highest in Europe (European Monitoring
Centre for Drugs and Drug Addiction, 2002).
A. WHAT IS
THE EVIDENCE
ON CANNABIS?
(1) Cannabis use and the onset of psychosis
(a) epidemiology
(i) association
A clear association between use of cannabis
and psychosis has been established by several longitudinal population
cohort studies. The US National Epidemiological Catchment Area
Study examined 20,000 community and institutional residents in
the early 1980s. Using this sample and matching cases and controls
for social and demographic characteristics, Tien and Anthony (1990)
reported that people who used cannabis on a daily basis were 2.4
times more likely to report psychotic experiences than non-daily
cannabis users (after adjusting for alcohol use and psychiatric
diagnoses). Similarly, the Australian National Survey of Mental
Health and Well-being 1997 (Hall et al, 1998) found that
those who met the International Classification of Disease
(ICD)-10 criteria for cannabis dependence were nearly three times
more likely to report that they had been diagnosed with schizophrenia
than those without cannabis dependence disorder.
These studies do not, however indicate the direction
of the association between cannabis and psychosis, suggesting
two possible hypotheses:
1. "Temporal priority hypothesis"—Cannabis
use precedes development of psychosis.
2. "Self-medication hypothesis"—Cannabis
use is a consequence of psychosis with people using it to self-medicate
their symptoms.
(ii) temporal priority
Other population studies provide evidence for
cannabis use preceding the development of psychotic symptoms.
The Swedish Conscript Cohort (Andréasson et al,
1987) followed up 50,087 Swedish conscripts and found evidence
for a "dose-dependent" relationship between cannabis
use at 18 years and diagnosis of schizophrenia 15 years later.
Heavy users of cannabis, with no psychiatric diagnosis at conscription,
were 2.3 times more likely to be diagnosed with schizophrenia
later in life (after adjustment for confounding variables). The
authors noted, however, that only 3% of heavy cannabis users went
on to develop schizophrenia suggesting that it may only affect
those who have some other pre-existing vulnerability to psychosis.
One of the limitations of this study is the
large temporal gap between cannabis use at 18 years and onset
of schizophrenia 15 years later, with no assessment of cannabis
use or other drug use in the intervening period.
The sample was also followed up by Zammit et
al (2002) across the period 1970 to 1996. They found that
the risk of developing schizophrenia was increased (odds ratio
= 1.9) in those who had ever reported cannabis use at baseline.
A dose-dependent effect was again found, with those who had used
cannabis more than 50 times prior to assessment having a further
increased risk of developing schizophrenia (odds ratio = 6.7).
This study used a more complete psychiatric
register and controlled better for confounding variables such
as other drug use, known risk factors for schizophrenia, IQ and
social integration, but still found a relationship between cannabis
use and schizophrenia. The authors estimated that 13% of schizophrenia
could be averted if all cannabis use were prevented.
The Netherlands Mental Health and Incidence
Study (Van Os et al, 2002) examined the relationship between
cannabis use and psychosis amongst the general population (n=4,045)
and subjects with self-reported symptoms of psychosis (n=59).
They found that users of cannabis at baseline were nearly three
times more likely to show psychotic symptoms at follow-up three
years later. This risk remained significant even after a variety
of confounding factors were controlled for. They also found evidence
for a "dose-dependent" relationship with the heaviest
users showing the highest risk. The authors estimated the attributable
risk of cannabis to psychosis to be 13%, similar to Zammit's earlier
finding. The relationship between cannabis use and psychotic symptoms
was found to be even stronger for people with more severe psychotic
symptoms who required care. The attributable risk of cannabis
to severe psychotic symptoms was estimated at 50%. This study
is limited, however, by the short follow up period.
The Dunedin Multidisciplinary Health and Development
Study examined a general population birth-cohort of 1,037 subjects
born in Dunedin in 1972-73 with follow up at age 26. The key advantage
of this study is that the authors collected data on self-reported
psychotic symptoms at age of 11, before the onset of cannabis
use. They found that individuals reporting cannabis use at ages
15 and 18 had higher rates of psychotic symptoms at age 26 when
compared to non-users. This association remained significant after
controlling for psychotic symptoms before the onset of cannabis
use (Arseneault et al, 2002). A significant effect of age
was also found, with cannabis use at 15 resulting in an increased
likelihood of meeting diagnostic criteria for schizophreniform
disorder at 26. Further, 10.3% of age 15 cannabis users were diagnosed
with schizophreniform disorder at age 26 compared to 3% of controls.
This suggests a strong developmental effect of early cannabis
use.
In addition to establishing temporal priority,
the Dunedin Study also found evidence for specificity of outcome,
as cannabis use at age 15 did not predict depressive symptoms
at age 26, and specificity of exposure, as the use of other illicit
drugs did not predict schizophrenia outcomes over and above cannabis
use. The authors concluded that "using cannabis in adolescence
increases the likelihood of experiencing symptoms of schizophrenia
in adulthood".
A significant effect of age was replicated in
a recent study by Stefanis et al (2004), which examined
3,500 subjects who formed part of the Greek Birth Cohort Study.
Participants were administered a postal questionnaire which examined
drug use and psychotic symptoms at age 19. Cannabis life-time
frequency use was associated positively with positive psychotic
symptoms. This effect size was much larger for those who had started
cannabis use earlier in adolescence (pre-15 years). This evidence
is limited as it is cross-sectional only, although the significant
effect of age suggests that cannabis use preceded the development
of psychotic symptoms.
A second general population birth-cohort study,
the Christchurch Health and Development Study, was conducted in
New Zealand, which followed up 1,265 children at ages 18 and 21.
As part of the study, data was collected on cannabis use and psychotic
symptoms. They found that young people meeting DSM-IV criteria
for cannabis dependence had elevated rates of psychotic symptoms
at both age 18 (rate ratio = 3.7) and age 21 (rate ratio = 2.3)
after adjusting for many variables, including self-reported psychotic
symptoms, other drug use and other psychiatric disorders. The
authors concluded that this showed that the development of cannabis
dependence is associated with increased rates of psychotic symptoms.
More recently, Ferdinand et al (2005)
conducted a longitudinal population based study with 2,076 young
children and adolescents recruited in 1983 from the province of
Zuid-Holland. Subjects were followed up in 1997, when they were
between the ages of 18 and 30. They found that cannabis use was
a risk factor for psychotic symptoms in initially psychosis-free
individuals and that this risk was increased almost three-fold
when compared to non-users. They also found some support for the
self-medication hypothesis, with psychotic symptoms predicting
future cannabis use. The hazard ratio for cannabis use preceding
psychotic symptoms was higher than that for psychotic symptoms
preceding cannabis use (2.81 versus 1.70).
Thus several studies have suggested clear temporal
priority for cannabis use. We do recognise some problems with
these studies, including heterogeneity of outcome across studies,
the use of self-report measures and limited statistical power.
However, conclusively demonstrating the causal role of cannabis
in the development of psychosis is necessarily difficult, given
the practical difficulties of using animal models and ethical
impossibility of human controlled trials. Adjusting epidemiological
data for confounding risk factors for psychosis also presents
enormous statistical difficulties. Given these constraints, we
find the evidence for a relationship between early cannabis use
and later psychotic symptoms compelling.
We also believe that the level of evidence required
should be set against the level of risk identified by these studies.
The development of a psychotic disorder is a serious and significant
experience in an individual's life. The studies presented above
indicate that cannabis use significantly increases the risk of
this outcome. It is in this context that we make our recommendations
for policy. However, we would support further epidemiological
research to confirm the results of these studies.
(iii) other findings
A variety of other epidemiological research
weakens the self-medication hypothesis.
A follow-up study in 1989 of the Swedish Conscript
Cohort (Andreasson et al, 1989) found that cannabis users
who developed schizophrenia had better premorbid personalities,
a more abrupt onset of the condition and more positive symptoms
than non-cannabis users who had schizophrenia. Earlier research
also suggested that cannabis users who develop schizophrenia have
better premorbid adjustment as well as having fewer negative symptoms
and better treatment outcomes (Allebeck, 1991). More recently,
a study over five years of Issac (1995) found that among inpatients
in South London, with the exception of patients with diabetes,
cannabis users tended to have more severe psychotic symptoms on
admission.
An innovative study (Verdoux, 2002) used self-reports
of drug use and psychotic symptoms from 79 college students, taken
at random times over seven days. A positive association was found
between cannabis use and unusual perceptions and a negative association
between cannabis use and hostility. There was no temporal relationship
between reporting unusual experiences and cannabis use, as the
self-medication hypothesis would predict.
A number of studies have found that people with
schizophrenia give similar reasons to other substance users for
their use of cannabis and other drugs, eg to relax or socialise,
to feel good, relieve boredom or provide stimulation. (Dixon et
al, 1990; Bergman et al 1985; Noordsky et al,
1991; Test et al, 1989).
Two reviews of the evidence concluded respectively
that: "on the basis of the best evidence currently available,
that cannabis use is likely to play a causal role with regard
to schizophrenia" (Arsenault et al, 2004) and "cannabis
is an independent risk factor both for psychosis and development
of psychotic symptoms" (Semple et al, 2005).
(iv) outstanding issues
If cannabis were a risk factor for schizophrenia,
one would expect that rates of schizophrenia would increase as
cannabis use increases. In Britain, cannabis use amongst young
people appears to have increased substantially over the past 30
years, from around 10% reporting lifetime use in 1969-70 to 50%
reporting lifetime use in 2001.
Initial data on the incidence of schizophrenia
suggests that it has not increased, but instead stabilised or
slightly decreased over the relevant time period. However, there
are a number of factors which may account for this data, in particular
changes in service design and a narrowing of the diagnostic criteria
for schizophrenia. Hence Kendell (1993) concluded that despite
reports of a falling incidence for schizophrenia in the UK, it
would be rash to conclude that rates of schizophrenia were falling
(Kendell, 1993). In some specific geographical areas, it seems
that the incidence of schizophrenia has increased significantly.
Boydell (2003) concluded that the incidence of schizophrenia had
doubled in thirty years in Camberwell, South East London (Boydell,
2003). This study included all psychiatric contracts, rather than
just admissions, and thus minimised the effects of changes in
service provision. It also identified all possible cases of psychosis
in the first instance, to minimise the effect of diagnostic delay
or administrative inaccuracy.
Given the difficulties in establishing changes
in the incidence of schizophrenia, we do not believe that the
current evidence on incidence refutes the significant amount of
epidemiological evidence pointing to a relationship between both
adolescent cannabis use and heavy cannabis use and later psychotic
symptoms.
(b) Neuroscience
Neuroscientific research gives evidence of mechanism
by which cannabis use may give rise to psychotic symptoms.
Two cannabinoid receptors have been identified:
CB1 and CB2 (Institute of Medicine, 1999; Pertwee, 2002), though
others may exist (Wiley and Martin, 2002). The CB1 receptor is
responsible for the psychological effects of THC (Heustis et
al, 2001), whereas the role of CB2 is less clear.
The CB1 receptor is most heavily concentrated
in the mesolimbic and mesocortical pathways, both believed to
be important for the development of schizophrenia (Ameri, 1999).
Interaction between CB1 and dopamine D2 receptors has been documented
in rats and monkeys (Meschler et al, 2001). Cannabis increases
dopaminergic activity in the mesolimbic system (Ameri, 1999).
This research gives some biological plausibility
to the temporal priority hypothesis discussed above.
(c) Conclusion
Epidemiological evidence, underpinned by neuroscientific
research, suggests that there is a relationship between both adolescent
cannabis use and heavy cannabis use and the onset of psychosis.
However, many questions remain and require further study. We do
not believe that the current evidence on incidence convincingly
refutes the temporal priority hypothesis.
(2) Other mental health problems
Some epidemiological studies have also established
an association between cannabis use and poor mental health more
generally.
A cross-sectional study has found an association
between cannabis use and low life-satisfaction, contact with mental
health services and hospitalisation (Kandel, 1984). Fergusson,
Horwood and Swain-Campbell (2002) found relationship between cannabis
use and suicidal behaviour after adjusting for confounding variables,
which was both dose-responsive and stronger the earlier the onset
of cannabis use. Rey et al (2002) found that in a nationally
representative sample of adolescent Australians, cannabis users
were three times more likely than non-cannabis users to experience
depression. Fergusson et al (1997) found evidence for a
relationship between cannabis use and major depression among the
Christchurch birth cohort, with heavy users (defined as having
used 10+ times) twice as likely as non-heavy users (having used
one to nine times) and three times more likely than non-users
to meet criteria for mood disorders. The Zurich cohort study found
that those meeting criteria for depression by age 30 were 2.3
times more likely than the general population to use cannabis
regularly (Angst, 1996). Another study found that 68% of female
cannabis users were depressed (Patton et al, 2002).
However, other studies have not found a relationship
between adolescent and depression or found that it is insignificant
after adjusting for confounding variables (Fergusson and Horwood,
1997; Brook, Cohen and Brook, 1998; McGee et al, 2000).
Some studies have also found evidence for a
link between cannabis use and suicide among adolescents, which
remains after adjusting for confounding variables (Borges et
al, 2000; Beautrais et al, 1999, Andreasson and Allebeck,
1990). Other studies have found an association but not a relationship
which remains after adjustment (Fergusson and Horwood, 1997; Patton
et al, 1997).
In both these areas, there is a need for more
research and better designed studies to clarify the relationship
between cannabis use in adolescence and suicide and depression/affective
disorders.
(3) Relapse
The negative effects of cannabis use on people
with psychotic illness have been well-established, initially through
case studies. In a retrospective study of people with schizophrenia,
Negrete et al (1986) found higher rates of continuous hallucinations
and delusions, and more hospitalisations amongst active users.
Jablensky et al (1992) replicated these findings in a two-year
follow up study of 1,202 patients with first-episode schizophrenia
enrolled in 10 countries as part of a World Health Organisation
(WHO) Collaborative Study. Linszen et al (1994) conducted
the first large prospective cohort study, comparing 24 users with
69 non-users over a year with assessments of mental state on a
monthly basis. Cannabis users experienced significantly more,
and earlier, psychotic relapses or exacerbation of symptoms over
the 12 month period and the effect was dose-responsive. Martinez-Arevalo
et al (1994) followed up 62 young adults with schizophrenia
over a one year period and found that cannabis use was the best
predictor of relapse and hospitalisation during this time.
In a longer term prospective study, Caspari
(1999) followed up 39 patients with schizophrenia over 68 months
and found a significantly higher rate of rehospitalisation. Cannabis
users also tended to have poorer psychosocial functioning than
non-users and higher scores on the "thought disturbance"
and "hostility" items of the Brief Psychiatric Rating
Scale (BPRS), though the strength of these findings is weakened
by the fact that only one assessment of mental state was made
once at the end of the 68 months.
More recently, Issac et al (2005) studied
115 patients admitted to a psychiatric intensive care unit in
South London, assessing mental state using the BPRS every two
weeks during their admission period. People with a history of
cannabis abuse were found to be younger on first admission and
had more previous hospital admissions. Urinanalysis indicated
that 25% of the sample used cannabis during admission, and those
that did use during admission tended to spend longer in hospital.
There is clear evidence to support the hypothesis
that the use of cannabis by patients with a diagnosis of schizophrenia
does result in an exaceberation of psychotic symptoms. This mitigates
against the self-medication hypothesis with patients using cannabis
to alleviate their symptoms.
B. USE OF
THIS EVIDENCE
BY GOVERNMENT
(a) Government policy on cannabis
In October 2001, the Home Secretary asked the
Advisory Council no the Misuse of Drugs to review the classification
of cannabis. In March 2002, the Advisory Council reported and
concluded that:
"no clear causal link has been
demonstrated. The onset of schizophrenia often occurs in the late
teens, when cannabis use is most common, so that an association
is inevitable."
The report goes on to recommend that cannabis
be reclassified from class B to C, on the basis that the harm
associated with it was less than other class B drugs. It was subsequently
reclassified in January 2004.
At the time this report was written, only the
Andreasson (87) study had been made public, so there was not a
large evidence base from which to make this judgement. However,
by the time that cannabis was in actually reclassified to class
C, in January 2003, a number of other studies had been published,
including Zammit (02), Van Os (02), Arsenault (02). In the light
of this, we find it surprising that a further review of the evidence
was not ordered before the reclassification decision was implemented.
In January 2005, following an extensive media
campaign by Rethink on the anniversary of reclassification, the
Department of Health announced a review of epidemiological evidence
on cannabis and the aetiology of mental illness. In March 2005,
the Home Secretary asked ACMD to look again at evidence on cannabis
and reconsider its classification.
In January 2006, ACMD's report was released,
which recommended that cannabis remain a class C drug, that a
sustained public education and information strategy about the
hazards of cannabis be created, that services for individuals
with cannabis problems be reviewed, measures to protect people
with schizophrenia on in-patient wards be strengthened and a research
programme on cannabis and mental health be instituted.
(b) Government's role in increasing
the evidence base
The Government has singularly failed to commission
looking at the impact of cannabis on mental health. No major study
so far on this issue has hence originated from the UK. This seems
a significant failure on the part of the Government, given that
hints of an important impact on mental health date back to 1987,
as noted above. ACMD in 2001 too failed to recommend more research
on the issue, even though it noted that the debate on it was long-running.
Applications to the Department of Health for
funding for studies on this question have also been consistently
rejected, even though some were strong, in Rethink's view.
Government has also failed to commission studies
looking at the impact of legal penalties or classification on
the use of cannabis. There is very little knowledge globally on
how the relative effectiveness of legal status, drugs education
and information campaigns on reducing usage levels.
(c) Communication of evidence
(i) To the general public
To our knowledge, the British Government has
never attempted to communicate the mental health risks of cannabis
use to the wider public. Indeed, opportunities to do so have been
missed.
For example, when cannabis was reclassified
to Class C in the UK January 2004, the public health campaign
that accompanied reclassification did not mention the possible
mental health effects of cannabis, but instead concentrated solely
on the physical health effects of use and its continued illegality.
This contrasts with the action of the French
Government, which in 2005 invested €3.8 million in the communications
side of its cannabis campaign. This is particularly noteworthy,
given that the latest evidence suggests that France has a lower
level of cannabis use among young people than the UK. In France,
35.7% of young adults report lifetime use (compared with 40.4%
in the UK) and 4.9% of young adults report use in the last year
(compared with 16.6% in the UK) (European Monitoring Centre for
Drugs and Drug Addiction, 2002).
(ii) To school-age children
Opportunities within school drugs education
have also been missed. Current DfES drugs guidance stresses the
physical health effects of cannabis, the possibility of dependence
and especially its illegality, but dismisses the evidence on mental
health effects:
". . . there has been a lot of
debate about whether the use of cannabis can lead to mental illness,
especially schizophrenia. However, no clear causal link has been
proven for the latter, although cannabis can worsen existing schizophrenia
and other mental illnesses and lead to relapse in some people.
It is important for schools to reinforce
to pupils the message that cannabis is harmful to health and is
still an illegal drug, and that possession remains a criminal
offence leading to a possible criminal conviction" (DfES,
2004;25)(emphasis DfES)
Cannabis education in schools has also been
conducted too late. Research suggests that one factor determining
the success of drugs education is ensuring that it is delivered
at a relevant time in young people's development (McBride, 2005)
and the current statistics on cannabis use among young people
in the Britain. The latest statistics reveal that 1% of 11 year
olds, 2% of 12 year olds, 7% of 13 year olds, 17% of 14 year olds
and 26% of 15 year olds had used cannabis in the last year (Health
and Social Care Information Centre, 2005). This would suggest
that any intervention in British schools should take place before
the age of 14, possibly before the age of 13, with booster sessions
following this, in order to maximize effectiveness.
Studies have also shown that school drugs education
can delay the age of first use, can reduce the number of young
people who go on to frequent or high use and reduce drugs-related
harms (Maggs and Schulenberg, 1998; Dijkstra et al, 1999;
McBride et al, 2004; DfES, 2004). Given that the major
risks involved with cannabis are dependent on the age and quantity
of use, drugs education in this area seems a particularly appropriate
intervention.
Furthermore, there is evidence of an inverse
relationship between recall of drugs cannabis education and cannabis
use. Among Year 11 pupils, those who did not remember having lessons
about drugs in the last year were more likely than those who did
to have used cannabis in the last month (21% compared with 16%).
This contrasts with the evidence on recall of tobacco and alcohol-related
lessons, where recall was found to have no impact on rates of
either tobacco or alcohol use (Health and Social Care Information
Centre, 2005), suggesting that education in relation to cannabis
may be even more effective in deterring use among young people
than tobacco or alcohol education.
(iii) To people with mental illness
The 2001 ACMD report does note the potential
risk of cannabis use to people with existing mental health problems.
Its view on this was very clear:
"Cannabis can unquestionably worsen
schizophrenia (and other mental illnesses) and lead to relapse
in some patients. Its use should therefore be particularly discouraged
in all people with mental health problems."
Despite this, there was no attempt until 2004
to create information materials for people with mental illness
about the risks of cannabis. Even this attempt was flawed, as
only £230,000 was allocated to the project. With approximately
1% of the population currently experiencing psychotic symptoms,
this equates to a spend of less than one pence per head of the
population experiencing psychosis and in touch with mental health
services.
As part of this project, research was commissioned
from Cragg, Ross and Dawson (unpublished), to look at people's
information needs and to make recommendations as to how the evidence
on cannabis might be communicated. This researched highlighted
an explanation used successfully by many psychiatrists in explaining
the mental health risks of cannabis to their patients, especially
those who were embedded in cannabis culture and hence could not
believe that it was causing them harm when others seemed to be
able to tolerate it well. The explanation used was that some people
had a "cannabis allergy"—this was said to work
very well as awareness of food allergies and intolerances was
growing among the public. Despite this evidence from professionals
of its usefulness, it was rejected by officials working on the
materials after advice from the Department of Health. The grounds
for rejecting it were that whilst people who had an allergy to
a food experienced an adverse reaction to it very quickly, sometimes
even instantaneously, those who experienced adverse mental health
effects from cannabis often would not feel them until years later.
Whilst there is some truth in this argument, there was no further
work done to try and present this analogy in a way which would
avoid this problem.
Once draft materials had been produced, further
research was done with people with mental illness, their carers
and professionals to test out the materials. The feedback gained
from most groups was negative and hence it was decided to rework
the materials. In this case, the decision was well grounded in
evidence.
The project has now been put on hold, pending
the advice of ACMD, despite the fact that the evidence on cannabis
use by people with mental health problems was never in question.
As well as specific communications campaigns,
there is an opportunity for Government to use existing health
awareness programmes to communicate messages on cannabis. However,
so far, such programmes (eg the Expert Patient Programme) do not,
in our view, cover the issue of cannabis sufficiently, if at all.
C. WHY HAS
EVIDENCE NOT
BEEN REFLECTED
IN GOVERNMENT
ACTION?
(a) Politicised debate
Cannabis has been a politicised issue since
the beginning of the twentieth century and "reefer madness".
In this contested arena, it has been difficult for Departmental
Advisors and experts of all kinds to look objectively at evidence.
At a conference in September 2005 on cannabis, Griffith Edwards,
the founder of the National Addiction Centre, pinpointed the two
possible errors made in the cannabis policy arena: the positive
error where too much credence is given to findings and the negative
error, where findings are dismissed too easily. He concluded that
20 years ago, the positive error had been rife; now, it is the
negative error that is rife. Hence the evidence has often not
been looked at objectively. The ACMD report of 2005 is a notable
exception to this trend.
(b) Role of the media
Because of the politicisation of this issue,
there has been a high level of media interest. This has meant
that the Government has sometimes been under pressure to make
rapid decisions to respond to criticism in the media. Hence, the
announcement of the Department of Health's review of evidence
in January 2005. However, in the case of the decision to refer
the issue to ACMD, this has led to well evidenced policy-making.
The media has, however, played an extremely
important role in communicating evidence on cannabis and psychosis
to the general public. Coverage of the issue has been significant
and has made a valuable contribution to educating the public about
this issue and in promoting discussion of it.
(c) Cross-over between departments
The issue of cannabis and mental illness does
not fit easily into Governmental or Departmental structures, lying
between the Home Office and the Department of Health and between
public health, mental health, and substance misuse. Hence, monitoring
research on this issue does not seem to have been part of the
core function of any one team—officials seem to have "dipped
into" the issue at certain points, because they were asked
for advice, but not followed the succession of findings on the
matter closely. This "dipping in and out" has allowed
people to look at individual pieces of evidence within the context
of their pre-conceived ideas on the issue, rather than questioning
their view of cannabis.
Cannabis is part of both the law enforcement
and health agendas. There has been no attempt to look at cannabis
policy "in the round", to consider the interaction of
public health education initiatives, information provision in
mental health services, drug service provision and law enforcement.
Officials and Ministers need to be encouraged to look at such
cross-cutting issues in a more coherent and comprehensive manner.
Recommendation 1: Guidance to civil servants and
Ministers stress the importance of considering and commissioning
evidence on all aspects of cross-cutting issues.
(d) Reliance on a single experts in
Government
Despite this, we have experience of officials
relying on only one expert, often an internal expert, to provide
advice on cannabis. Often, officials do not then challenge this
advice—it is regarded as an "expert view" and
is seen as absolute, though in fact it may be partial.
This seems to be particularly problematic if
the "expert" in the Government department has a professional
background linked to the issue—in this case, as a psychiatrist,
mental health nurse or researcher. Often, people with a professional
background are employed by Government departments and are seen
as resident experts in that field by career civil servants, because
of their experience "in the field". Whilst their advice
can be valuable, it is too often seen by officials as absolute.
Some of these experts are consulted on too wide a range of issues
than they can reasonably be expected to have mastered, a far wider
range than any official would be. The view of any one professional
would be considered a useful, but partial view, if they were responding
to a Government consultation—it does not seem right that
a single official's view is prioritised purely because of their
prior professional experience.
Furthermore, it is questionable how far the
views of these experts reflect current professional practice—the
longer they work in Government, the more removed are their experiences
from current practice and experience. Given that cannabis use
has increased significantly in the past two decades, professionals'
experience in mental health facilities has changed also—experts
who are out of touch with current professional practice are likely
to be out of touch with these experiences.
Experts are also far more likely than practicing
professionals to influence decisions on research funding and commissioning.
Given that practicing professionals have more relevant experience,
they should have a means in which to influence such decisions
and recommend future priorities.
Recommendation 2: Guidance to civil servants should
stress the importance of consulting a variety of resident experts
and practicing professionals.
Recommendation 3: A mechanism be created through
which service users, carers and organisations representing these
groups can inform Departmental decisions on research funding,
commissioning and determining future priorities
(e) Reliance on professionals, rather
than service users
To our knowledge, there is no-one with personal
experience of using drug or mental health services involved in
making cannabis policy. This seems a significant omission especially
in the make-up of ACMD. Including people with mental illness and/or
substance use problems on such bodies could help ensure that they
are more in touch with current issues for people and that views
are grounded in experience, rather than preconceived ideas. Organisations
which represent service users could also play an important role.
A similar case could be made for carers playing a role on such
bodies.
Service users also seem to be the last port
of call for officials making cannabis policy—this was certainly
our experience in the COI project to create information materials
on cannabis. Given that these materials were destined for people
with mental illness, it seemed foolish not to consult people with
mental illness at an earlier stage about what kind of information
they needed, as well as design and other issues. In the research
on information needs, mentioned above, service users were not
as well-represented as professionals.
There is currently no mechanism for service
users, carers and organisations representing them to make suggestions
and recommendations for future research funding.
Recommendation 4: Users of drug and mental health
services, their carers and organisations representing these groups
to be included in the make-up of committees such as ACMD.
Recommendation 5: Guidance to civil servants to
stress the need to consult service users and carers as well as
professionals, at all stages of the policy-making process.
Recommendation 6: A mechanism be created through
which service users, carers and organisations representing these
groups can inform Departmental decisions on research funding,
commissioning and determining future priorities.
(f) Lack of evaluation
The policy making process on cannabis does not
seem to be evaluated in a systematic or formal way. ACMD's advice,
for example, has never been evaluated by an external body. We
believe that these processes, like other Government processes,
deserve to be reviewed by an external body.
Recommendation 7: The advice given by Government-appointed
bodies such as ACMD and Government policy to be regularly evaluated
by external organisations.
(g) Time lag
As noted above, there was a significant time
lag between the review of ACMD in 2001 and the implementation
of reclassification in 2003. In this period, a significant amount
of new evidence emerged about cannabis and mental illness, but
the cannabis decision was not revisited in the light of this.
This did not happen despite the efforts of Rethink to bring the
new evidence to the attention of Government. There needs to be
a mechanism for reviewing evidence and updating recommendations
between the point where policy recommendations are made and they
are implemented. Once a major policy decision (such as that to
re-classify) has been taken, further research relating to that
decision should be systematically monitored and reported to Ministers
responsible.
Recommendation 8: Major policy decisions to be
accompanied by a commitment to monitor research developments until
and following implementation.
January 2006
| 
