Smoking and mental health - a
review
of the literature
Dr
Ann McNeill
Independent
Consultant
&
Honorary
Senior Lecturer in Public Health
St
George’s Hospital Medical School
London
Click
here for a
printable pdf version
Table of Contents
1 Association of mental health problems and
smoking
1.1
Comparison with the general population. 5
1.2
Association of smoking with people living in institutions
1.3
Proportion of all cigarettes smoked by people with mental health problems
1.4
The role of the tobacco industry
1.5
Nicotine dependence is a psychiatric disorder
1.6
The relationship between smoking and different mental health disorders
1.6.2
Bipolar affective disorder
2
The impact of smoking on people with mental health problems
2.1
People with mental health problems can ill afford to smoke
2.2
High mortality and morbidity associated with smoking
3 Why do people with mental health problems
smoke more?
3.1
Association of cigarette smoking and mental health problems with deprivation
3.2
The relationship between smoking and the environment
3.3
Does smoking causes mental illness?
3.4
Does the illness cause smoking? Nicotine use as self-medication
3.4.1
Relationship between nicotine and neurotransmitters
3.4.3
Nicotine and schizophrenia
3.4.4
Nicotine and adult attention deficit hyperactivity disorder
3.4.6
Interactions between nicotine and medications for schizophrenia
3.5
Difficulties with cessation and withdrawal effects
4
Reducing smoking in people with mental health problems
4.1
Do people with mental health problems want to quit smoking?
4.2 Treatment for smokers with mental health
problems
4.2.2
Brief opportunistic advice to quit for smokers with mental health problems
4.2.3
Cognitive and behavioural therapy for smokers with mental health problems
4.2.4
Group therapy and nicotine replacement therapy (NRT)
4.2.6
Offer smokers with mental health problems the best
4.2.7
Smoking reduction and nicotine replacement therapy
4.3.1
Smoking policies in psychiatric institutions in the UK
4.4
Smoking among mental health professionals
I am extremely grateful for the diligent research
assistance of Karen Richardson from ASH.
I would also like to acknowledge various reviews from
the US and France that provided many of the references cited in this review.
Two by Tony George and Jennifer Vessicchio are available free on-line[1],[2]. Reviews by Gregory Dalack[3] and Jacques le Houezec[4] were also very useful. I am also very grateful to Professors John
Hughes, Michael Farrell, Martin Jarvis and Debra Malpass for providing
literature and contacts with international experts on these issues. The tables are adapted from material
prepared by Professor Martin Jarvis. Thanks also go to Ben Ayliffe and Naj
Dehlavi from ASH for technical support.
The report was commissioned and managed by Judith Watt, SmokeFree London
Programme.
December 2001
SmokeFree London Programme
London Region NHS
40 Eastbourne Terrace
London W2 3QR
Tel: 020 7725 5436
Fax: 020 7725 5393
sflondon@doh.gsi.gov.uk
This report aims to:
o
discuss the relationship
between smoking and clinically diagnosed mental health problems
o
explain why this is an
important yet poorly addressed public health issue
o
suggest approaches
to reduce the problem.
Key issues identified in this report:
o
Nicotine dependence is
the most prevalent, deadly and yet most treatable of all psychiatric disorders
but is often overlooked by the psychiatric professions
o
Smoking prevalence is
significantly higher among people with mental health problems than among the
general population
o
Smoking prevalence is
highest among those with a diagnosis of a psychotic disorder
o
Studies have shown
smoking rates to be as high as 80% among schizophrenics although increased
smoking might be associated with psychosis or severity of psychosis rather than
with a specific psychotic illness
o
People with psychotic
disorders who live in institutions are particularly vulnerable: over 70% of
this group smoke including 52% who are heavy smokers
o
Daily cigarette
consumption is considerably higher among smokers with mental health problems
who may also inhale more deeply from their cigarettes
o
In the US it has been
estimated that just under half of all the cigarettes smoked are smoked by
people who have had a psychiatric or substance abuse disorder in the past
month. This may be an overestimate but illustrates that a significant
proportion of cigarettes smoked are likely to be done so by people with mental
health problems.
o
Smoking related fatal
diseases have been shown to be commoner among schizophrenics than among the
general population
o
Some of the excess
mortality of people with mental health problems is potentially preventable if
they are given support to stop smoking
o
A significant
proportion of people with schizophrenia recognises that smoking is a problem,
want to quit and will attend smoking cessation therapy
o
More than half (52%) of
schizophrenic smokers living in institutions wanted to give up smoking
o
Effective treatments
exist to help people stop smoking and are not yet being routinely offered to people
with mental health problems
o
All health
professionals working with smokers with mental health problems should encourage
smokers to quit and refer those needing further support to specialist smoking
cessation services
o
There is some evidence
from other countries to suggest that smokers with mental health problems feel
excluded from mainstream stop smoking programmes. Smoking cessation services
need to be offered guidance on how to be accessible to and how best to support
smokers with mental health problems.
o
Nicotine may help
alleviate some of the positive and negative symptoms associated with
psychiatric illnesses and may also help to alleviate the side effects
associated with their medications
o
Attempts to stop
smoking do not appear to exacerbate psychotic symptoms
o
Many mental health
institutions at best condone and at worst encourage smoking as cigarettes are
sometimes used to reward or punish patients.
o
Smoke-free policies
encourage smokers to quit, make non-smoking the norm and reduce the harmfulness
of environmental tobacco smoke.
o
People with mental
health problems may be more vulnerable to the misleading messages about tobacco
promoted by the tobacco industry
o
In the UK, people with
schizophrenia who smoke contribute an estimated Ł139m each year to the
Treasury.
1 Association of mental health problems and smoking
1.1 Comparison with the general population
Smoking prevalence is significantly higher among people with mental
health problems than among the general population. The most recent data for
England illustrating this are from the Psychiatric Morbidity Survey which
involved over 10,000 adults and was carried out by the then Office of
Population Censuses and Surveys (OPCS) in 1993[5]. Sixteen per cent of this general population sample had
a current psychiatric disorder. The vast majority of these were neurotic
disorders with nearly half being accounted for by mixed anxiety and depressive
disorders. Less than 0.5% of the sample had a psychotic disorder. A similar survey has been carried out by the
Office of National Statistics (ONS) in 2000, but data on smoking prevalence are
not yet available.
Table 1 shows a breakdown of smoking prevalence among six categories of
neurotic disorder and one category of any psychotic disorder (due to the
prevalence of psychotic disorders in the general population being very low).
The overall smoking prevalence for the sample was 32%, which compared well with
a national smoking prevalence of 30% from the 1993 Health Survey5. (National smoking prevalence in 1998 was 26% in
women and 28% in men[6]).
All categories of mental health problems had higher levels of smoking than the
general population. Forty four per cent of those diagnosed as having any
neurotic disorder were smokers. The highest smoking rates were among those with
a diagnosis of psychosis.
Table 1
For the neurotic disorders, a clear relationship was observed between
smoking and the number of neurotic symptoms as measured by the revised version
of the Clinical Interview Schedule (CIS-R)[a].
This score ranges from 0 to 57 and the overall threshold score for significant
psychiatric morbidity is 12. Current smoking increased in a stepwise fashion
from 28% for a CIS-R score of 0-5, to 34% for a score of 6-11, 42% for a score
of 12-17 and 48% for a score of 18 plus5. Higher scores were also associated with heavier
smoking[7].
Smoking prevalence also varied with the number of neurotic disorders. For those
with one neurotic disorder 43% were current smokers, for those with two or more
disorders, 54% were current smokers5.
A dose response relationship has also been observed between the number
of psychiatric diagnoses and smoking in a recent US study[8].
1.2 Association of smoking with people living in institutions
People with severe mental health disorders will be under-represented in
national household surveys as many of them will be living in institutions or be
homeless. Hence in 1996, in addition to the national household survey, the OPCS
also published a national survey of around 1,200 residents in psychiatric
institutions (including recognised lodgings, group homes, hostels, residential
care homes and hospitals)[9]
and a national survey of over 1,000 homeless people[10].
Among those living in institutions, three categories of disorder were
examined:
o
schizophrenia,
delusional or schizoaffective disorders (around 70% of the sample);
o
affective psychoses,
i.e. mania and bipolar affective disorder (around 8%);
o
and neurotic disorders
i.e. generalised anxiety disorder, depressive episode, mixed anxiety and
depressive disorder, phobia, obsessive-compulsive disorder and panic (around
8%).
Five per cent had other mental health disorders and for the remaining 9%
insufficient information was obtained to enable classification. Table 2 gives
the smoking prevalence for these different disorders.
Table 2
As can be seen in the graph, people with mental health disorders who
live in institutions have particularly high rates of smoking. Among homeless
people smoking prevalence was 70%, the highest rates being observed among
residents of nightshelters (84%) and rough sleepers (91%).
1.3 Proportion of all cigarettes smoked by people with mental health problems
In the US it was recently estimated that nearly 45% of all the
cigarettes smoked were smoked by individuals with a psychiatric (ranging from
psychotic illnesses to anxiety disorders and phobias) or substance abuse
disorder in the past month8. These figures may be an overestimate because
cigarette consumption figures were based on peak consumption, ie how many
cigarettes per day people smoked when they were smoking most (daily cigarette
consumption was not assessed). In addition, the definition of psychiatric
illness was quite broad such that 28% of the US population was estimated to
have had a mental illness in the last month. Nevertheless, the proportion of
cigarettes smoked by people with mental health problems is startling. However,
over one third of patients with any lifetime history of mental illness, and 30%
of those having a mental illness in the last month, reported to have quit
smoking. A similar analysis has not yet been carried out in England but it
seems likely that it would yield results of a similar magnitude.
A further report in the US has queried whether smoking in pregnancy
should be considered an indicator for screening for the additional risk to
infant health from a psychiatric disorder[11].
Preliminary data from this study indicated an association between smoking in
pregnancy and a current psychiatric illness.
1.4 The role of the tobacco industry
Lasser and
colleagues8 commented that the tobacco industry has ‘identified
psychologically vulnerable persons as a part of their tobacco market’. Building
on this, Boyd and Lasser[12] suggested that those with mental
health problems might find it more difficult to recognise ‘the specious claims
and implications of tobacco advertising’. They argued that we should not assume
that the tobacco industry does not target those with mental illnesses who are
now a large part of their market.
1.5 Nicotine dependence is a psychiatric disorder
Nicotine dependence is a recognised psychiatric illness and has been
referred to by John Hughes as ‘the most prevalent, most deadly, most costly,
yet most treatable of all psychiatric disorders’[13]. However, it is often overlooked by
the mental health professions[14].
Hughes[15] has argued that it would be
preferable to distinguish between smoking and nicotine dependence, as not all
smokers are nicotine dependent. He has remarked[16]
that a recent US study showed that, despite decreases in smoking prevalence
among young adults up until 1992, the prevalence of nicotine dependence
symptoms increased over time such that the incidence of acquiring nicotine
dependence symptoms by age 24 was in fact increasing. The importance of this
illustrates the utility of the psychiatric diagnosis of nicotine dependence.
Some mental health problems such as depression have been shown to be more
associated with nicotine dependence than smoking[17].
1.6 The relationship between smoking and different mental health disorders
A review of the relationship between smoking and some
mental health disorders follows.
The OPCS survey of people living in institutions found that 74% of people
with schizophrenic disorders were current smokers.
Strength of nicotine dependence is strongly related to cigarette
consumption, perceived difficulty in stopping and time to first cigarette[18].
On all of these indicators, smokers with schizophrenic illnesses show signs of
high dependence. Just over half (51%) of the sample were heavy smokers (defined
as smoking more than 20 cigarettes a day); 55% of men, and 39% of women. This
compares with 8% in the general population[19].
Eighty-two per cent of the smokers felt that it was difficult not to smoke for
a whole day, compared to 57% of smokers in the general population. Seventy-two
per cent smoked their first cigarette within 30 minutes, compared with 41% in
the general population[20].
Despite this high dependence however, there is still a reasonable degree
of motivation to quit. More than half (52%) said they would like to give up
smoking, compared with 69% in the general population.
Smoking prevalence in people with schizophrenia living in institutions
may inflate smoking rates, as people with schizophrenia living in general
households are omitted. One study[21]
which did examine all people with schizophrenia (n=169) in a discrete
geographical area (Nithsdale in SW Scotland) found that 58% were smokers. This
figure although lower than that given above is still considerably higher than
the smoking rate in the general population. Furthermore, over two-thirds of the
smokers in the Nithsdale sample were smoking 25 or more cigarettes a day
(compared with 51% smoking more than 20 cigarettes a day in the OPCS
institutional survey).
Elevated smoking rates among people with schizophrenia have also been
observed in other countries1,[22].
In addition to higher cigarette consumption, Olincy and colleagues[23]
found significantly higher urinary cotinine levels among schizophrenic smokers
than nonschizophrenic smokers with similar smoking histories suggesting that
those with schizophrenia consumed higher doses of nicotine from their
cigarettes.
Smoking rates may vary depending on the type of schizophrenia the person
is suffering from. A Greek study found that smoking rates differed between
different subtypes of schizophrenia[24].
There may also be gender differences in the smoking
patterns of schizophrenics. In the OPCS institutional sample, 62% of women with
a schizophrenic disorder were current smokers, compared to 78% of men. One US
hospital study found that male schizophrenic patients had the highest frequency
of smoking, followed by male nonschizophrenic patients, female schizophrenics
and female nonschizophrenic smokers respectively[25].
Given these data it is not surprising that smoking related fatal
diseases are commoner among schizophrenics than among the general population[26]. This study followed a cohort of 370
schizophrenics living outside hospitals, from 1981for 12 years. During this
time 79 died, 73% from natural causes (which includes smoking induced diseases)
and 24% from unnatural causes such as suicides or accidents. Standardised
mortality rates (SMRs) were significantly raised for smokers (but not for
non-smokers), and for smoking-related diseases. The SMR for lung cancer was
twice the expected value.
1.6.2 Bipolar affective disorder
The OPCS survey of people living in institutions found that 70% of
people with affective psychosis (including mania and bipolar disorder) were
smokers. Again, within this population of smokers, there appears to be high
levels of dependence. Nearly half of the population (49%) were heavy smokers,
the vast majority of all smokers (82%) felt that it would be difficult not to
smoke for a whole day and 61% had their first cigarette less than 30 minutes
after waking. However, there was still interest in quitting with again more
than half (58%) stating that they would like to give up smoking.
Elevated rates of smoking among those with bipolar disorder have also
been observed in other countries22 [27]
[28]. One study however found little
increase in smoking rates among those with bipolar disorder although it had a
very small sample size[29].
A preliminary study of 92 patients with a diagnosis of bipolar affective
disorder in Ireland may shed light on these potentially contradictory findings[30].
A relationship was found between smoking and heavy smoking and a history of
psychotic symptoms. The group with no history of psychotic symptoms had a
smoking prevalence similar to the national smoking prevalence figure. The
authors hypothesised that increased smoking is associated with psychosis or
severity of psychosis rather than with a specific psychotic illness.
One study has identified an increased risk for psychiatric complications
after smoking cessation among smokers with histories of major depression, in
particular bipolar disease[31].
The OPCS survey of people living in institutions categorised people with
various depressive and anxiety disorders together as neurotic disorders so
prevalence data for the various categories are not available separately.
Smoking prevalence was high (74%) for people with these neurotic disorders:
with women and men having the same smoking prevalence rates. Once again,
indicators of dependence were very high: 77% of these found it difficult not to
smoke for a whole day and 85% smoked their first cigarette within 30 minutes of
waking. Yet nearly half (41%) said that they would like to give up smoking.
Studies examining only people with major depression from other countries
have also shown that they are more likely to smoke2 [32].
In Great Britain, smoking among people with depressive episodes living at home5
was 56% and was higher among women (53%) than among men (46%). There is also
some evidence from an Australian study of an association between regular
smoking in teenage girls and high levels of depression and anxiety[33].
Studies have also shown that people with major depression have
difficulty when they try to stop[34]. As reported by George and Vessichio2, one trial examining the efficacy of clonidine as an
aid to smoking cessation in 71 heavy smokers who had failed in previous quit
attempts found unexpectedly that 61% had a history of major depression[35].
History of depression also had a significant negative effect on success at
quitting. Another study has found that symptoms of depression assessed at baseline
predicted time to first cigarette smoked after attempts at quitting,
illustrating that symptoms of depression predict failure to quit[36].
More severe withdrawal symptoms have been observed when people with histories
of major depression or indeed any anxiety disorder attempted unsuccessfully to
stop smoking or to cut down, than among those without such a history[37].
However, severity of withdrawal did not account for the relationship
between major depression and continued smoking. Smokers with a history of
depression have also been shown to be more likely to experience more persistent
withdrawal discomfort over a longer period of time[38].
In addition, smokers with a history of depression who stop appear to be
at a significantly increased risk of developing a new episode of major
depression which remains high for at least six months[39].
Covey and colleagues34 described the relationship between smoking and
depression as: ‘complex, pernicious, and potentially life long’.
Other studies however have shown no effect of depression on rate of
quitting. For example, Breslau and colleagues (1998) in a prospective study of
over 1000 adults over a period of years found a history of major depression did
not affect smokers rate of quitting[40].
Finally, one study whilst finding that the risk of experiencing major
depressive episodes following quitting was twice as high among smokers with a
history of depression, also found that developing depressive episodes was
similar whether patients succeed in quitting or fail[41].
This suggests that it might be the attempt at quitting or fear of failure that
might precipitate the depressive episodes.
The OPCS survey of people living in general households5 found
a smoking prevalence of 55% among people with panic disorder (52% among women,
59% among men). Higher rates of smoking have also been found in people
suffering from panic disorder in other countries[42]
although this finding has not always been consistent in men[43].
Smoking rates of 53%-60%, and higher cigarette consumption, have been
found among people with post-traumatic stress disorder (PTSD)[44],[45].
As mentioned above, one study has found withdrawal symptoms more severe in
people with anxiety disorders who had unsuccessfully tried to reduce their
smoking, than among those without such disorders37.
The OPCS survey of people living in general households5 found
a smoking prevalence of 47% among those with a generalised anxiety disorder,
42% among those with mixed anxiety and depressive disorders, 48% among those
with a phobia, 40% among those with obsessive-compulsive disorder. Among those
with any neurotic disorder 44% were current smokers, 43% of women, 47% of men.
These figures compare with 29% smokers among those with no neurotic disorder
(28% of women, 30% of men).
As described earlier smoking (both current and heavy) was related to a
higher numbers of neurotic symptoms on the CIS-R scale, and the number of
neurotic disorders in the OPCS Psychiatric Morbidity Survey5.
Smoking has also been associated with adult attention deficit disorder[46],
eating disorders[47]
and substance abuse disorders2.
An inverse relationship has been observed between cigarette smoking and
some diseases eg Alzheimer’s disease4, Parkinson’s disease[48]
and Tourette’s syndrome4. George and Vessichio2 report that nicotine may have a neuroprotective
effect, as it appears to alleviate the neurological impairment involved with
these diseases.
Le Houezec4 reported that a preliminary study has shown cognitive
functioning of people with dementia improved after short term (up to four weeks[49])
nicotine administration and that an improvement has also been demonstrated in
symptoms of Tourette’s syndrome following nicotine administration. Similarly,
one initial case study has illustrated that long-term nicotine administration
in the form of nicotine patch improved Parkinson’s Disease[50].
In summary, higher
smoking prevalence and levels of nicotine dependence are associated with
various mental health illnesses. In general, a greater severity of the illness
is associated with higher rates of smoking. There are gender differences in
smoking rates for some of the illnesses.
Psychiatric complications when stopping have also been reported with
depressive and anxiety disorders.
2 The impact of smoking on people with mental health problems
2.1 People with mental health problems can ill afford to smoke
People with mental health problems spend a significant proportion of
their income on smoking. In one case study in the US a schizophrenic smoker was
estimated to spend just over one third of weekly income on cigarettes12.
McCreadie and Kelly[51]
estimated that there were at least 200,000 people with schizophrenia in the UK
and based on estimates of 60% smoking an average of 26 cigarettes per day,
estimated that they contributed Ł139m each year to the Treasury. They commented
that the costs of schizophrenia have been estimated in the UK to be between
Ł397m and Ł714m each year indicating that people with schizophrenia were,
through their smoking, contributing substantially to the cost of their care. They
concluded that,
‘This double blow of smoking and
unemployment hits the patients very hard, financially. The state loses less;
what it gives to this vulnerable group in our community with one hand, it takes
back with the other’.
McDonald’s paper[52]
substantiates these figures and comments that money spent on cigarettes ‘is not
being spent on clothing, leisure pursuits and personal possessions, which could
help to increase the quality of life of these patients’.
2.2 High mortality and morbidity associated with smoking
Patients with mental health problems who smoke will get the same smoking
related diseases as those who do not have such problems.
Many people with mental health problems die from smoking related
diseases such as cardiovascular and respiratory diseases, the rates of which
can be twice as high among schizophrenics as in age-matched control populations26.
There may also be interactions between the mental health illness and
smoking. For example, Jung & Irwin found that depressed smokers had lower
natural killer cell activity than control smokers and depressed and control
non-smokers[53]. Dysfunctional natural killer cell activity
may contribute to primary tumour development and metastatic cancer risk.
Dalack and colleagues comment that recent data indicate that smoking is
a risk factor for dyskinesia (defined as unusual body movements such as tremor)
independent of medication exposure3. Tardive dyskinesia (tardive
means delayed as the movement may become apparent only after long term
treatment) has been linked to increased morbidity and mortality in those with
chronic schizophrenia3.
The burden of smoking related illnesses on top of mental health problems
may be huge. Boyd and Lasser have commented ‘those with mental illness are
often the least capable of coping with the devastating medical illnesses caused
by smoking12.
Finally, smoking is also a cause of many fires. A small-scale study by
Strathclyde Fire Brigade found that 12% of fires involving care in the
community patients over a two year period were due to careless dispersal of
smoke materials[54].
In summary, the effects of smoking
among people with mental health problems will be of a similar or possibly
greater magnitude to those without such problems. The cost of smoking, both
financially and to physical health, is very high.
3 Why do people with mental health problems smoke more?
Several hypotheses have
been proposed to account for higher smoking among those with mental health
problems. Research is ongoing in this area and there may well be more than one
explanation or there may be different explanations for the different types of mental
health problem. Some of these hypotheses are discussed in more detail below.
3.1 Association of cigarette smoking and mental health problems with deprivation
Cigarette smoking has become increasingly concentrated in the most deprived
groups. A clear inverse relationship exists between smoking prevalence and
social class6. Table 3 shows smoking prevalence by a deprivation
score (taking into account occupation, educational level, housing tenure, car
ownership, unemployment, and living in crowded accommodation) developed by
Jarvis[55].
Table 3
Smokers in more highly deprived groups have also been shown to have
higher nicotine dependence levels55 as shown in Table 4.
Table 4
Similarly a relationship has been demonstrated between having a
psychiatric disorder and deprivation. One study has demonstrated that both
social class and area level deprivation had independent associations with
mental health problems[56]
and the authors suggested that both personal disadvantage and the deprivation
of the surrounding area were having an effect.
So could the association between smoking and mental health problems be
explained by deprivation factors? Perhaps people suffering from mental health
illnesses smoke in a similar way to other deprived groups as a coping mechanism
to deal with the stresses of their everyday lives.
One study among a cohort of 16 year old New Zealanders[57]
examined depression and nicotine dependence and prospectively measured risk
factors including sociodemographics, family history of criminality, life events
etc. Much of the co-morbidity between depression and nicotine dependence could
be explained by common or correlated risk factors associated with either
disease. Contrary to an early study of twins that suggested a strong genetic
association between smoking and major depression[58],
the New Zealand study suggested a substantial input of social and childhood
factors which were antecedent to either nicotine dependence or depression.
However, even after adjustment for these other risk factors there was a
significant association between depressive disorders and nicotine dependence.
This latter finding has been demonstrated elsewhere: cigarette smoking
is associated with mental health problems even after controlling for
sociodemographic factors[59],[60]
suggesting that factors other than deprivation are involved.
3.2 The relationship between smoking and the environment
The data shown in Tables 1 and 2 demonstrate that the rates of smoking
observed among people living in institutions were considerably higher than for
those with similar illnesses living at home. The highest rates were observed
for those homeless who were sleeping rough. These data suggest that the
environment plays a role in smoking prevalence although it is possible that
those with more severe forms of mental health diseases are those that are
institutionalised or are homeless.
One study carried out in a US hospital25 suggested that it was unlikely that boredom or
institutionalisation contributed to smoking. In this hospital, schizophrenic
inpatients had higher smoking rates than nonschizophrenic inpatients and in
addition, a long duration of hospitalisation was associated with being a
nonsmoker among nonschizophrenic patients.
High rates of smoking in psychiatric institutions and hospitals may also
reflect lax smoking policies. Anecdotal evidence suggests that cigarettes have
been perceived as providing an area for negotiation between staff and patients
and cigarette privileges can be offered as rewards for good behaviour12.
An Australian qualitative study of 24 patients living
in the community with various psychotic disorders supported this. The smokers
in this study perceived their smoking as a core need[61]. One consistent theme that emerged
in the research was that smoking was a means of control in an otherwise
uncontrollable environment, although some patients reported that this could be
used against them as cigarettes were sometimes used to reward or punish
patients. Another theme was that smoking was reinforced by people and places in
their environment:
‘The impression gained
from all participants was that, if people went into hospital as non-smokers, in
all probability they would leave as smokers, literally because of peer pressure
to smoke, the lack of other activities to occupy them while there, and reinforcement
by the institution.’
Some mental health professionals will also be smokers themselves and may
indeed smoke with their patients[62].
Mental health professionals rarely discuss smoking with their patients12 14 26. This may be because they lack the skills and
knowledge, or because they do not think that their patients can quit, or
because they believe that smoking is one of the few pleasures that people with
severe mental health problems can have[63].
3.3 Does smoking causes mental illness?
Prenatal exposure to nicotine or cigarette smoke may cause mental health
problems later in life. There is evidence that prenatal nicotine exposure
disrupts neuronal development hence making such a theory biologically plausible[64]
[65]. Some studies have found an association
between smoking during pregnancy and increased risk of adult deficit
hyperactivity disorder[66].
As smoking precedes many mental health illnesses, smoking by the
individual might contribute to or cause the illness. However, this would not
preclude the possibilities that there might be common aetiologies (such as
social, familial and individual risk factors) causing both smoking and the
illness, or that smoking is used to ameliorate symptoms of the illness before
it becomes overt or diagnosed.
Smoking has been found to predate
illness in many different diseases, for example, bipolar illness28 and schizophrenia15. Studies examining whether smoking precedes
depressive and anxiety illnesses have thus far been contradictory.
Prospective data suggest that the association between nicotine
dependence and major depression probably reflects common factors that
predispose to both disorders[67],40 For example, Breslau and colleagues’
longitudinal five-year study40 reported that a history of major depression at
baseline increased significantly the risk for progression to daily smoking
(although it did not significantly decrease the rate of quitting). History of
daily smoking at baseline significantly increased the risk for major
depression. The authors suggested that shared aetiologies might be important.
However, Wu and colleagues[68]
in a US study of nearly 2,000 youths assessed from 1989 to 1994, found that
tobacco smoking predicted a slight increase in the risk of a subsequent onset
of depressed mood, but depressed mood was not associated with a later risk of
initiating cigarette smoking. This supported a possible causal link from
tobacco smoking to later depressed mood in childhood and early adolescence but
not vice versa.
Similarly, Johnson and colleagues in a study of a sample of nearly 700
youths from New York interviewed at ages 16 and 22 found that heavy cigarette
smoking during adolescence was associated with an increase in various anxiety
disorders at age 22 after controlling for various confounding factors[69].
However, anxiety disorders during adolescence were not significantly associated
with chronic cigarette smoking during early adulthood. In this study, an
association was found for agoraphobia, generalised anxiety disorder and panic
disorder but not for obsessive-compulsive or social anxiety disorders. The
authors suggested that cigarette smoking might increase risk of certain anxiety
disorders during late adolescence and early adulthood. They hypothesised that
the links with some anxiety disorders but not others may be linked to impaired
respiration and the potential anxiogenic (i.e. increases anxiety) effects of
sustained nicotine intake.
Breslau and colleagues also found that daily smoking was associated with
an increased risk for the first occurrence of a panic attack[70].
These preliminary data also suggested that the link between smoking and lung
disease could account for the relationship between smoking and panic attacks.
There was also some evidence of a reduced risk of panic disorder after smoking
cessation.
In common with these findings, West and colleagues report that anxiety
decreased following the first week of smoking abstinence[71].
The authors suggested that their findings supported the view that smoking is
chronically anxiogenic rather than being anxiolytic or reducing anxiety and
that is consistent with the finding that anxiety is increased among smokers in
general population studies.
In summary there is some evidence that
smoking may affect the body to increase vulnerability to some mental health
disorders. There is some indication from these studies that smoking may cause
an increase in anxiety.
3.4 Does the illness cause smoking? Nicotine use as self-medication
Even if smoking did not cause a mental illness, smoking might be
maintained if it had a positive effect on symptoms of the illness. A self-medication view is consistent with
several hypotheses about the effects of nicotine: that it helps to alleviate
some of the positive and negative symptoms of mental health problems; that it
improves cognition; that it may also help to alleviate the side effects
associated with anti-psychotic medications. In the first hypothesis where
nicotine alleviates some of the symptoms, then abstinence from smoking and
nicotine would result in increased symptoms on quitting[72].
This is also addressed in later sections.
These hypotheses are explored briefly below following a description of
the biological processes related to nicotine intake.
3.4.1 Relationship between nicotine and neurotransmitters
Nicotine interacts with nicotinic
receptors on nerves throughout the body and brain. In the brain, nicotine acts
on nicotinic acetylcholine receptors (nAchR) causing transmitter release and
metabolism. Chronic nicotine use causes inactivation of the receptors which
might cause a subsequent increase in their number22. The brains of smokers have an increased number of
high affinity nicotinic receptors[73].
People with schizophrenia however
have a lower number of nicotinic receptors. Leonard hypothesised that this
might be due to an abnormality of the genes relating to neuronal nicotinic
receptors in schizophrenia[74].
Nicotine use affects the release of
different neurotransmitters, including acetylcholine, dopamine, norepinephrine,
serotonin (5-HT), glutamate, and
aminobutyric acid (GABA)2. These neurotransmitters have various effects and may
play a role in some mental health disorders.
Reward pathways in the brain use
dopamine (and possibly also norepinephrine and 5-HT) as the neurotransmitter
and it has therefore been hypothesised that systems involving these
neurotransmitters may be involved in depression in humans. Indeed,
antidepressant drugs commonly increase dopamine, norepinephrine or 5-HT.
Similarly, nicotine stimulates the release of these neurotransmitters, thereby
counteracting depression4. Indeed, some small-scale pilot studies have shown
that nicotine improves depression in never-smokers who are depressed[75]. These findings
could explain why smoking cessation is associated with major depression in
depressed smokers2.
Klimek
and colleagues[76]
found that long-term smokers had neurochemical abnormalities in a part of their
brains, the locus coerulus, similar to the brains of animals treated with
antidepressant drugs and opposite to those observed in people with major
depression. The direction of causality here remains unclear however. Smoking
could cause the neurochemical abnormalities or people with these abnormalities
may be more susceptible to becoming smokers.
Chronic smoking also appears to
inhibit (non-nicotine) monoamine oxidase B (MAO-B) activity in the brain[77]. MAO-B is
involved in the breakdown of dopamine and therefore if MAO-B is inhibited,
dopamine levels would increase which would be rewarding. MAO inhibitors are
effective antidepressants and together these findings also suggest that smoking
(but not nicotine in this case) has some antidepressive effects4,13.
Finally, a study by Epping-Jordan and colleagues[78]
found that when nicotine was administered to rats who had been chronically exposed
to nicotine, the rats sensitivity to rewarding electrical impulses remained
stable. During withdrawal however, the intensity of the electrical impulse had
to be increased by 40% to maintain the rewarding response. This decreased
sensitivity may correspond to the depression/anhedonia experienced by smokers
during withdrawal.
3.4.3 Nicotine and schizophrenia
Dysfunction of dopamine and other neurotransmitters is associated with
schizophrenia3. Positive symptoms (such as voices, delusion,
confusion etc) of schizophrenia are thought due to excess dopamine in
substantia nigra, whereas negative symptoms (such as withdrawal, inertia, lack
of motivation) are due to concurring deficits of dopamine in the cortex3. These aspects are referred to further below.
People with schizophrenia who smoke present more positive psychiatric
symptoms of schizophrenia than
nonsmokers[79],[80]. A recent study involved 101
patients with schizophrenia in which antipsychotic medication was discontinued.
At baseline smokers had more positive symptoms and were apparently more
functionally impaired than nonsmokers. This difference was however no longer
evident after a 30-day medication discontinuation period[81].
This suggests an interaction between the medication, smoking and positive
symptoms of schizophrenia. Interactions with medication are discussed further
below.
Findings for negative symptoms have been contradictory with increased79, similar80 or decreased24 negative symptoms being observed in smokers compared
with nonsmokers.
One study found a relationship between heavy smoking and positive and
negative symptoms. Heavy smokers (here defined as >25 cigarettes per day)
had the most positive symptoms and a significantly lower number of negative
symptoms80.
Smoking in schizophrenia could therefore
be related to ameliorating positive symptoms. In the small-scale qualitative
study by Lawn and colleagues, schizophrenic smokers (more so than other
diagnostic groups) reported that their smoking was self-medicating and they
reported alleviating positive symptoms61. Smoking could also be serving to reduce the number
of negative symptoms, although it may simply be associated with greater
severity of the disease. Although the evidence and the direction of causality
is unclear, Dalack and colleagues3 have concluded that smoking in schizophrenia may
represent attempts to self-medicate symptoms of the illness, in particular
negative symptoms. Negative symptoms associated with schizophrenia may be
related to a lowered activity of the systems involving dopamine. Smoking may
reverse this effect by stimulating the release of dopamine4. An example of this is given below.
3.4.3.1 Links with negative symptoms – P50 – schizophrenia and other mental health problems
Patients with schizophrenia are unable to ignore distracting stimuli,
for example, abnormal auditory filtering is found in patients with
schizophrenia. The abnormal filtering is often referred to as an auditory
sensory gating deficit. To measure this abnormal filtering, paired auditory
stimuli are delivered half a second apart. Subjects with no schizophrenia
inhibit the response to the second stimulus. Schizophrenics show similar
responses to both stimuli. This is referred to as diminished suppression of
auditory-evoked P50 response.
This deficit may be related to nicotine receptors in the brain. Adler
and colleagues (as reported by Dalack3) have shown that nicotine and cigarettes can
transiently reverse these deficits. George and Vessachio1 point out that atypical medications may also normalise
P50 deficits in schizophrenia.
Genetic evidence links nicotinic function with the diminished P50
response as P50 is related to alpha nicotinic receptors in the brains of
families affected with schizophrenia. Smoking also improved smooth pursuit eye
movement abnormalities in schizophrenic patients and it has been suggested that
dysfunctional nicotine receptors may be involved[82].
3.4.4 Nicotine and adult attention deficit hyperactivity disorder
Levin and colleagues in small-scale studies with adults with attention
deficit hyperactivity disorder have found that acute administration of nicotine
improved the symptoms of this disorder[83],[84].
The authors postulate that this may be due to nicotine receptors and
neurotransmitters. This finding may be largely due to nicotine’s affect on
improving attention discussed in the next section below.
In summary, there is some evidence that
nicotine impacts on the symptoms of various mental illnesses. The mechanisms
involved appear to involve neurotransmitters and nicotine receptors, but the
exact nature of these relationships is still being explored.
If nicotine improved general cognitive functioning, this could explain
why many people with mental health problems smoke. A recent review[85]
of the cognitive effects of nicotine however, concluded that nicotine does
improve aspects of cognitive functioning in smokers forced to abstain. Benefits
of nicotine for nonsmokers or smokers who were not abstaining from cigarettes
were inconclusive. This suggests that
many benefits of smoking may be simply relief of incipient adverse withdrawal
effects that the smokers have wrongly labelled. Nevertheless, there may be
positive benefits, for example, there is some evidence of improved focusing of
attention in smokers over non-smokers[86].
3.4.6 Interactions between nicotine and medications for schizophrenia
3.4.6.1 Antipsychotic medications for schizophrenia
People with schizophrenia and some other mental health disorders are
often treated with antipsychotic (i.e. reduces psychotic symptoms) also known
as neuroleptic (i.e. acts on the nervous system) medication. Generally,
antipsychotic drugs block dopamine receptors in the brain and hence block the passage
of nerve signals by dopamine. In so doing they reduce the symptoms of
schizophrenia[87]. As the serotonin (5-HT) system
interacts with the dopaminergic system, 5-HT may also be involved.
Antipsychotic medications are divided into typical and atypical types.
Typical medications such as chlorpromazine were introduced into the market from
the 1950s onwards. Atypical medications were developed during the 1990s and
work on a wider range of symptoms and tend to be associated with fewer side
effects.
Antipsychotic medication more effectively controls positive symptoms of
schizophrenia than negative symptoms. Newer drugs appear also to more
effectively act on negative symptoms. The medication also causes side effects.
These include unusual body movements, such as dystonia (prolonged muscle
spasms), akathisia (restlessness and agitation), and Parkinsonian movement
disorders (including stiffness, shakiness, tardive dyskinesia, feeling drowsy).
3.4.6.2 Smoking and neuroleptics
Cigarette smokers are generally prescribed higher neuroleptic doses,
which might be a due to smoking increasing the metabolism of the neuroleptic
medication80 although not all studies are consistent with this4. If quicker
metabolism of antipsychotic drugs is a factor, when people stop smoking they
may experience an increase in drug levels in their blood and this may require a
reduction in the amount of antipscyhotic drug they should be prescribed87.
The choice of neuroleptic medication influences smoking behaviour. For
example, the atypical treatment clozapine is associated with a decrease in smoking
and haloperidol is associated with an increase in smoking and nicotine blood
levels[88].
McEvoy and colleagues studied 70 patients with schizophrenia who were
receiving conventional antipsychotics and then switched to clozapine[89].
Smokers showed significantly greater therapeutic response to clozapine than
nonsmokers and smoked less when treated with clozapine than with conventional
drugs. The authors concluded that certain patients with schizophrenia responded
favourably to either nicotine or clozapine.
A more recent study on schizophrenic patients in British Columbia found
that patients treated with clozapine had significantly lower expired air carbon
monoxide values (an indicator of smoke inhalation) than patients treated with
depot (injected) neuroleptics and patients receiving clozapine reported smoking
less than patients treated with depot neuroleptics[90].
Decreased smoking rates do not appear to be associated with other atypical
antipsychotic medications[91].
3.4.6.3 Smoking may reduce the side effects of medication
As mentioned above antipsychotic drugs block dopamine receptors. In
doing this, they will reduce pleasure. Smoking may help to counteract this
blockade of dopamine receptors.
Regarding the typical antipsychotic medications, Levin and colleagues,
for example, have found that nicotine reversed some of the adverse side effects
of haloperidol treatment and improved cognitive performance in schizophrenia[92].
One small single-blind study with 16 patients has also found that nicotine
patches significantly reduced akathisia associated with antipsychotic
medication[93].
McEvoy and colleagues hypothesised that if patients
with schizophrenia smoked primarily to reverse the effects of their
antipsychotic medications, then chronic schizophrenics should smoke at
substantially higher prevalence rates than first-episode patients[94].
Their study however found no difference in smoking rates between chronic and
first-episode schizophrenics, suggesting that it is the schizophrenia, not its
treatment with antipsychotic drugs that was determining smoking prevalence.
In summary, there is a relationship
between smoking and neuroleptic doses and the type of neuroleptic medication
being administered. Smoking may reduce the side effects of some forms of
medication. Other factors also appear to play a role.
3.5 Difficulties with cessation and withdrawal effects
It has also been hypothesised that higher smoking rates in psychotic
patients may be due to their experiencing greater difficulties with cessation
and withdrawal effects when they try to stop. As stated above, these
difficulties may vary depending on the mental illness and have been reported
for depressive and anxiety symptoms37. Exacerbation of symptoms has not yet been detected
with schizophrenics on abstinence[95].
However, although the data are limited, studies have suggested that cessation
rates for schizophrenics are considerably lower than among smokers with other
psychiatric diagnoses[96].
In summary, Section 3 has explored
possible reasons for the links between smoking and mental health disorders. The
evidence would support a multiplicity of factors, varying by the type of mental
illness and probably the severity of the illness.
4 Reducing smoking in people with mental health problems
4.1 Do people with mental health problems want to quit smoking?
Around a half of people with mental health problems in the British
surveys mentioned in section 1, expressed a desire to quit. Canadian and US
studies have shown that people with schizophrenia recognise that smoking is a
problem, are interested in attending smoking cessation groups and appear to be
appropriately motivated[97],[98].
There are little data on effectiveness of different treatments for
smokers with mental health problems from the UK. Most of the work in this field
originates from the US and has been carried out with people with schizophrenia.
This work is discussed below following a brief outline of what is known about
effective smoking cessation treatments for all smokers.
4.2 Treatment for smokers with mental health problems
4.2.1 Effective smoking
cessation treatments
Evidence based and professionally endorsed national smoking cessation
guidelines in England[99]
emphasised an integrated smoking cessation strategy involving brief
opportunistic advice to quit from health professionals, with a prescription for
effective pharmacological treatments, backed up by intensive specialist
cessation support for those smokers who need it. Brief opportunistic advice has
a low efficacy but because of the huge number of people health professionals
see in the course of any one year, it can have a very significant public health
impact. Brief advice mainly triggers attempts to quit (and may do so in 40% of
smokers given such advice) but many smokers will need further support.
Intensive support has a much higher efficacy but will reach fewer smokers.
Two pharmacological treatments have proven efficacy for smoking
cessation. Over one hundred trials involving nicotine replacement therapies
(NRT) have been carried out and these have demonstrated that NRT is effective
in smoking cessation, approximately doubling the success rate of any treatment.
NRT has an excellent safety record and given that smokers are already inhaling
nicotine there are no significant new risks involved when smokers use NRT.
There are currently six products available: gum, patch, nasal spray, inhaler,
sublingual tablet, and lozenges. There is little evidence of any difference in
these products, and little scientific support for matching treatments to
smokers except the 4mg gum is better than the 2mg gum for heavy smokers and the
standard strength patch is more effective than the low dose patch for moderate
to heavy smokers. There is also some evidence that combinations of different
types of NRT may be better than one alone[100].
Several trials have also demonstrated the effectiveness of bupropion (or
amfebutamone, trade name Zyban). All of these trials involved combining
treatment with bupropion with behavioural support. Again, bupropion
approximately doubles the success rate. Bupropion has a more complex
side-effect profile, with more drug interactions and contraindications, than
NRT, so care is needed with prescribing.
All of these smoking cessation treatments have been demonstrated to be
extremely cost-effective costing less than Ł1000 per life year saved[101],
considerably below the informal NICE threshold for cost effective NHS
treatments[102].
Following publication of the Government’s White Paper, Smoking Kills[103],
a network of NHS smoking cessation services have been set up across England.
These services offer specialist intensive support for smokers as well as
encouraging, training and supporting primary care health professionals to give
brief opportunistic advice on smoking.
It is important for these services to be accessible to smokers with
mental health problems and anecdotally there are reports of this happening in
the UK. However, there is no guidance for those running the services as to how
best to support smokers with mental health problems. It is important to be able
to respond appropriately to requests for support and also to be proactive in
reaching out to smokers with mental health problems in the community. There is
some evidence from the qualitative Australian study that smokers with mental
health problems feel excluded from mainstream smoking cessation programmes61.
4.2.2 Brief opportunistic advice to quit for smokers with mental health problems
The US guidelines published by the American
Psychiatric Association[104]
recommended the routine treatment of smoking for patients with psychiatric
diagnoses. Data in the US indicate that physicians were identifying smoking status
of people with such diagnoses but were not then providing counselling or
pharmacotherapies to encourage and support smokers in stopping[105]. One US commentator has suggested that
psychiatric nurses are uniquely placed to intervene with smokers and should
also initiate and support wider tobacco control policies[106].
A recent British editorial reported that although the
national service framework for mental health states that people with a severe
mental illness should have physical assessments, there was little evidence that
this was happening[107].
They went on to comment that this is in spite of frequent contact with primary
care services. We highlighted earlier reasons why mental health professionals
do not intervene with smokers. However, it is clear from the evidence above
that smokers with mental health problems are motivated to quit.
There is some suggestion that mental health problems
may undermine attempts at quitting rather than ability to stop[108].
It is therefore critical that health professionals who come into contact with
smokers with mental health problems routinely ask about smoking and advise
their patients to stop. Many smokers will however need further support and this
is outlined below.
Finally, it is important not to offer treatment when
the mental health illness is florid or
very active[109].
A note can be made in the patient’s notes however to intervene when the
patient’s condition stabilises. Patients should probably be followed up more
closely, in order to monitor antipsychotic medication and as a precaution in
case symptoms of the illness become exacerabated .
4.2.3 Cognitive and behavioural therapy for smokers with mental health problems
One study[110]
with cigarette smokers who had experienced major depressive disorder in the
past examined the effects of a standard cognitive-behavioural smoking cessation
treatment programme and this standard programme plus cognitive-behavioural
treatment for depression. No pharmacological treatments were involved.
Abstinence rates at one year were high (25 - 33%) and there was no difference
between the treatments. However, secondary analyses revealed that smokers with
recurrent major depression and heavy smokers were significantly more likely to
be abstinent if they received the two treatment programmes rather than the
standard programme only. The authors suggested that the additional programme
for depression might provide benefits for some smokers with major depression.
4.2.4 Group therapy and nicotine replacement therapy (NRT)
Hall and colleagues[111]
reported a preliminary study comparing a standard treatment programme using
groups support with a depression prevention programme. Both programmes involved
the use of nicotine gum. Abstinence rates were higher among smokers with a
history of major depression in the depression prevention programme. A
subsequent study however failed to replicate this finding with the two
interventions producing similar abstinence rates[112].
An uncontrolled trial to measure the efficacy of a smoking cessation
group programme involving group therapy and nicotine patches, which had been
modified for individuals with schizophrenia, found that a significant
proportion stopped smoking[113].
Fifty subjects completed a programme of seven weekly sessions (out of 65 who
attended initial assessments) and 21 (42%) had quit smoking at the end. All but
one of the subjects who had quit used the nicotine patch. There was no change
in the positive or negative symptoms of schizophrenia.
A randomised controlled trial of 45 smokers diagnosed with schizophrenia
or schizoaffective disorder assigned to a standard group therapy programme or a
specialised group therapy programme for smokers with schizophrenia was carried
out in the US[114].
All subjects participated in 10 weekly group therapy sessions and treatment with
the nicotine patch and continued to receive their pre-study atypical or typical
antipsychotic medications. Smoking abstinence rates did not differ between the
two types of group therapy programmes although continuous smoking abstinence
rates in the last four weeks of the trial were higher with the specialised
programme. However, atypical antipsychotic medications in combination with the
nicotine patch significantly enhanced the rate of smoking cessation (56% vs.
22% at the end of the programme) compared with those receiving typical
antipsychotic medications. In the atypical group, carbon monoxide levels
decreased significantly. Risperidone and olanzapine were associated with the
highest quit rates although sample sizes here are very small. Again, there were
no changes in psychiatric symptoms during the trial.
This study built on the previous studies described earlier showing that
the atypical medication clozapine may reduce cigarette consumption in
schizophrenics and that the typical medication haloperidol could increase
cigarette smoking. The authors suggested that medications targeting specific
clinical symptoms and neurochemical aspects of schizophrenic illness could
improve drug dependence.
A recent study examining retrospective utilisation of nicotine
replacement therapies in a psychiatric unit in the US found that more patients
preferred to use the nicotine inhaler over the nicotine transdermal patch[115].
The most relevant contraindications for bupropion in this context
include: in patients with a history of
bipolar disease, in patients with a current seizure disroder or any history of
seizures, in patients with a current or previous diagnosis of bulimia or
anorexia nervosa. There are also several potential drug interactions between
bupropion and anti-depressants and anti-psychotics which also need to be taken
into account before prescribing bupropion.
Bupropion will therefore be unsuitable for many people taking
concomitant medication. It is worth noting however, that bupropion is licensed
as an antidepressant in the US (but not in the UK) and hence it has been used
extensively with patients with depression in the US. The efficacy of bupropion
however appears to be independent of a past history of depression and is not
due to a reduction in depression following cessation[116].
Interestingly, SSRIs have also been shown to be ineffective in helping smokers
stop[117].
Studies in the US have also been carried out with bupropion and smokers
with other mental health problems. A case report and preliminary
placebo-controlled trials have supported the efficacy of bupropion enhancing
smoking cessation rates in schizophrenic patients[118],[119].
A further study included eight patients with schizophrenia who
participated in a 14-week open-label trial of bupropion and supportive group
therapy[120].
The goal was stopping smoking but patients were encouraged to continue to
participate even if they were not successful in complete cessation. None of the
patients stopped smoking during the study (one subsequently stopped smoking
completely) but the treatment package helped patients decrease their cigarette
consumption as measured by a decrease in expired air CO levels. No worsening of
positive symptoms or cognition or anxiety was observed. There was no change in
suppression of the P50 event-related potential, which might have been expected
if nicotine intake decreased. The authors commented that this latter finding
was hard to interpret although it is not known how much nicotine is needed to
enact the relevant change. There was some evidence of an improvement in
negative symptoms during the treatment period.
Preliminary work in 15 veteran outpatients with chronic post-traumatic
stress disorder who wanted to stop smoking has shown that bupropion is
generally well tolerated in combination with a stable psychotropic medication
regimen, and may be effective for smoking cessation in those with this disorder[121].
4.2.6 Offer smokers with mental health problems the best
Professor
John Hughes, one of the leading experts in this field, advises that if health
professionals manage to encourage their smoking patients to make an attempt to
quit, they should be offered the best treatment, as it may be some years before
they try to quit again. Given that these patients tend to be heavier smokers,
for example, stronger doses of NRT may therefore be preferable and should be
combined with specialist cessation support.
In
addition, there is increasing evidence of the effectiveness of using more than
one NRT for treatment100. One study also suggested that combining NRT and
bupropion might enhance success[122].
Given the greater nicotine dependence in this population, combination therapy
may increase the chances of successful stopping.
4.2.7 Smoking reduction and nicotine replacement therapy
Given the high rates of smoking and the low rates of stopping in this
population a harm reduction approach might be appropriate in parallel with
encouraging cessation. There is good evidence that smoking-related morbidity
and mortality are related to the dose or amount of smoking, so that if some
cigarettes could be replaced with less harmful forms of nicotine delivery,
there might be an overall benefit to the smoker’s health100. The
potential downside to this approach is that it might discourage quit attempts.
However, there is some evidence to suggest that by being able to control their
smoking, using a less harmful form of nicotine delivery might actually encourage
the smoker to quit100. It can be argued that smokers who are unable or
unwilling to quit should at least be given the choice of which form of nicotine
delivery to use.
Studies have therefore examined the impact of NRT on ad libitum smoking in psychiatric
patients and preliminary evidence suggests that this may be a useful approach.
One small-scale study of 13 psychiatric patients found they smoked
significantly fewer cigarettes whilst receiving the nicotine patch than when
they had the placebo patch[123].
A further exploratory study found that wearing a nicotine patch was safe
and well tolerated over a 32 hour period among a group of ten heavy smokers
with schizophrenia who were not actively trying to cut down or quit and who
therefore smoked ad libitum during
the study[124].
Nicotine levels increased during active patch treatment without evidence of
nicotine toxicity. Psychiatric symptoms, carbon monoxide and cigarettes per day
did not change, although eight subjects had a decrease in expired air CO on the
active patch. The heaviest smokers (n=5) had a statistically significant
decrease in expired air CO of at least 20% suggesting a reduction in smoke
inhalation. Dyskinesias showed a small, but significant, increase during
smoking when wearing the active patch.
Smokefree policies reduce the harmful effects of secondhand tobacco
smoke, encourage smokers to quit and help to make non-smoking the norm. Many psychiatric
patients who are ex-smokers may relapse when hospitalised and vulnerable to
constant smoking stimuli61.
Research from Canada and the United States shows that a smokefree policy
can be implemented in psychiatric institutions with careful planning and
consistency by all staff. One study in the US concluded that a smokefree policy
produced significantly fewer adverse effects than the staff anticipated[125].
Staff attitudes also changed to favour a smokefree environment.
Vancouver General Hospital in Canada[126]
implemented a complete indoor smokefree policy in its psychiatric assessment
and inpatient psychiatry units. It is
reported that workplace conditions notably improved and some long-standing beliefs
about psychiatric patients were disproved.
The major concern was whether psychiatric patients could be prevented
from smoking without major behavioural consequences: ‘There seemed to be a long
standing belief in the hospital community that psychiatric patients could not
tolerate a non-smoking policy: many staff anticipated a resultant increase in
violence and elopement and widespread surreptitious smoking.’
Nurses formed a committee and facilitated the implementation of the ban.
Materials on smoking cessation were gathered and doctors were introduced to
prescribing nicotine gum and clonidine hydrochloride to reduce withdrawal
symptoms. The impending change was advertised to patients, other departments
and other hospitals. Numerous signs were put up – initially designed and
produced by the patients themselves. The hospital found that open discussion
was an effective way of addressing the ethical differences.
‘The approach of policy implementation seemed to produced more anxiety
in the staff than in the patients: policy implementation itself was
considerably less dramatic. Arguments over cigarettes continue to occur, but
staff almost universally agree that problems overall have been fewer than
before the policy’.
A psychiatric hospital in Tuebingen, Germany is currently introducing a
complete smokefree policy in a staged manner[127].
Researchers carried out a survey initially among employees and found that a
smoke-free environment was thought to be a reasonable and achievable goal in
these settings.
4.3.1 Smoking policies in psychiatric institutions in the UK
Following a recent assessment of tobacco control policies in the NHS[128],
psychiatric units and long-stay units were identified as posing particular
challenges to successful policies limiting smoking. Subsequently, the Health
Development Agency published the results of a consultation process tackling
tobacco control within these units[129].
Forty NHS Trust tobacco control policies and statements were analysed.
The main emphasis of the final report is on staff at
all levels in the organisation. The
reason given for this emphasis is that staff are responsible for implementation
of the policies: ‘Policies will flourish or flounder over time depending on how
competently they are managed’.
Although encouraging smokers to stop was not the main emphasis in the
tobacco control policies of the 40 trusts analysed, it was often recognised
that a period of hospitalisation is an opportunity for health promotion on
smoking.
One general medical trust’s policy indicated the need for staff to
ascertain patients’ smoking status: ‘All in-patients and out-patients should be
asked whether they are smokers. Where
appropriate, advice on giving up should be provided. A supply of relevant literature should be maintained in each
ward, clinic and health centre’.
With few exceptions the tobacco control policies allowed for designated
smoking areas for users, either on the ward or nearby within the building.
None of the trusts explicitly offered NRT to patients within the 40
policy documents examined.
4.4 Smoking among mental health professionals
Smoking policies may also encourage psychiatric staff to stop smoking.
There is a dearth of recent data on smoking among nurses[130].
Studies in the 1980s found conflicting evidence as to whether smoking is higher
among psychiatric nursing than among other psychiatric professions[131]
[132].
However, one small-scale study in the UK, which examined employment influences
on women’s smoking, found that psychiatric nurses had a smoking prevalence rate
twice that found among other groups of nurses[133].
In summary, smokers with mental health
problems are motivated to quit and can be successfully helped to stop smoking.
Effective treatments include group therapy, NRT and bupropion. There is some evidence that adapting the
treatment programme for smokers with mental health problems may enhance
efficacy although the data are limited. Harm reduction approaches are also
worth considering for smokers with mental health problems who are unable or
unwilling to quit. Smokefree policies can be successfully introduced in
psychiatric institutions and will
encourage smokers to quit. Mental health professionals have an important
role to play in encouraging and supporting smokers’ attempts to stop.