Abstract

Tobacco use in mental health in general and bipolar disorder in particular remains
disproportionally common, despite declining smoking rates in the community. Furthermore,
interactions between tobacco use and mental health have been shown, indicating the outcomes
for those with mental health disorders are impacted by tobacco use. Factors need to be explored
and addressed to improve outcomes for those with these disorders and target specific
interventions for people with psychiatric illness to cease tobacco smoking. In the context of
bipolar disorder, this review explores; the effects of tobacco smoking on symptoms, quality of
life, suicidal behaviour, the biological interactions between tobacco use and bipolar disorder, the
interactions between tobacco smoking and psychiatric medications, rates and factors surrounding
tobacco smoking cessation in bipolar disorder and suggests potential directions for research and
clinical translation. The importance of this review is to bring together the current understanding
of tobacco use in bipolar disorder to highlight the need for specific intervention.
Keywords: Bipolar disorder, Smoking, Smoking cessation, Quality of life, Psychotropic drugs
INTRODUCTION

Tobacco smoking is a leading preventable cause of poor health and premature mortality.1) While
tobacco smoking rates in the general population are declining, they remain raised in cohorts with
psychiatric problems.2) Nicotine, the primary active compound found in tobacco, is highly
addictive.3) High rates of tobacco smoking have been described in bipolar disorder and are
associated with poorer outcomes.4) Tobacco smoking is approximately 2–3 times more prevalent
in bipolar disorder than in the general population, with estimates ranging from 60–70% in
bipolar patients compared to 25–30% in the general population.2,4,5) This elevated smoking rate
may not only impact on the course of the disorder but may also contribute to higher levels of
tobacco-related morbidity and mortality.6)

Despite high rates of comorbidity and related morbidity, there is a only modest field of research
focusing on smoking among individuals with bipolar disorder and the relationship with important
clinical and public health outcomes including: the effects on mood symptoms, quality of life,
suicidal behaviour, implications for pharmacological treatment, barriers to cessation, and the
biological pathways modulating the expression of pathophysiology. Moreover, results from
existing studies spanning these topics appear conflicting, suggesting uncertainly exists in the
literature.

Some authors have argued that tobacco smoking results in poorer treatment outcomes in bipolar
disorder,7–12) whereas other researchers have not found such a relationship.13,14) Variance
between studies may be due to design factors including the sample size, the sample selection
inconsistent measurements of smoking behaviour, and the methodological design of the studies.
Given this, the following review aims to explore the issues specific to smoking cessation in
bipolar disorder and provide the most clarity regarding treatment options and outcomes.
A comprehensive search using PubMed and Google Scholar was conducted using the terms
“smoking and mania”, “smoking and bipolar disorder” and “nicotine dependence and bipolar
disorder”. The purpose of the current review was to explore the effects of tobacco smoking in
bipolar disorder, examining the relationship between smoking and mood symptoms (e.g.,
depression and mania), quality of life, suicidal behaviour, biological interactions between the
pathophysiology of the disorder and the addiction to tobacco, as well as the effects of tobacco
smoke on medication and of smoking cessation. Specifically, this paper discusses pertinent
studies spanning clinical and public health topics, addressing current gaps in the field and
extending previous reviews conducted, with an aim of prompting further research in the nascent
field. Such an evidence base may also provide a context for guiding policy makers and informing
clinical practice.
MAIN SUBJECTS
Tobacco Smoking Cessation and Effects on Mood
Symptoms

At a population level, mood symptoms such as mania or depression appear to be associated with
tobacco use. The presence of mania and hypomania is associated with 3.9 and 3.5 times greater
likelihood of tobacco dependence in the psychiatric population, respectively.15) Additionally,
during acute tobacco smoking cessation, those with bipolar disorder may have higher risks of
experiencing mania,16,17) and transient depressive recurrence.18) Furthermore, Adan et al.19)
have reported low mood scores in highly dependent smokers, intermediate mood scores in low
dependent smokers and high mood scores in non-smokers, suggesting the role nicotine
dependence plays in mood regulation. Another issue regarding bipolar disorder and tobacco
smoking cessation is the potential effect of cessation on mood symptoms. Although there have
been case reports of mania,16,17) and depression,18) following abrupt nicotine withdrawal, there
are few double blind placebo controlled studies that have looked at the effect of tobacco
cessation on mood. A key recent meta-analysis has shown that on aggregate, mental health
symptoms, particularly depression, improve in those who quit smoking.20)

Those with bipolar disorder might thus be proposed to have an increased susceptibility to
tobacco smoking despite the observation that many patients start to smoke before the onset of
their illness.21–23) Research shows that most smokers take up the habit as teenagers,24) with
recent statistics in Australia revealing that in 2010, the age of initiation was 16.25) Regarding the
age of onset in bipolar disorder, for bipolar type I the average age of onset is 23 years for males
and 26 years for females, and for bipolar type II, the age of onset is 29.7 years for males and 30.1
years for females.26) A line of research has postulated that enduring alterations in
neurophysiology as a result of tobacco abuse may unmask an underlying vulnerability to
affective disorders, including bipolar disorder.19,27,28) However, the mechanisms underlying
the relationship between tobacco smoking and bipolar disorder are clearly bidirectional, complex
and multifactorial.13,29) Therefore, while it is probably correct that in most instances tobacco
smoking behaviour pre-dates and is a risk factor for bipolar disorder, caution is recommended
when interpreting these findings and implying causality.
Specific associations between tobacco smoking status and symptom characteristics of the
disorder have been described.7,8,11,12,30,31) In an early study looking at the role of tobacco
smoking and psychotic symptom status in bipolar disorder, Corvin et al.12) examined the effect
of smoking severity on a sample of 92 patients with bipolar disorder. Tobacco smoking was less
prevalent in less symptomatic patients whereas pervasiveness of smoking was correlated with
severity of symptoms (particularly psychotic symptoms). This was a cross-sectional study with a
small sample size using self-report measures which should be noted when interpreting the
results, as no objective measure such as blood test were used to detect nicotine use. A recent
prospective and naturalistic study over 24 months examined the effect of tobacco smoking on
mental health outcomes in 239 patients with schizo-affective and bipolar disorder,11) and found
that in both schizoaffective and bipolar disorder, daily smokers had poorer prognosis in terms of
depression and overall bipolar symptomatology (assessed using the Clinical Global Impressions
— Depression and Clinical Global Impressions — Overall Bipolar scales) and longer stays in
hospital compared with non-smokers. An important limitation, similar to Corvin and
colleagues,12) is that the study did not objectively measured nicotine dependence and used self-
report measures. This is again an important issue as varying levels of nicotine dependence may
affect overall scores and mood symptomatology more than simply assessing daily smoking rates
reported by the patients. Daily smoking rates do not necessarily address smokers’ levels of
nicotine addiction amongst patients with bipolar disorder.

Other studies have reported a link between tobacco smoking and worse outcomes in both mania
and depression in bipolar disorder.7,8,10,30,31) In particular, a large study of 399 patients with
bipolar disorder reported that a lifetime history of tobacco smoking was significantly related to
earlier onset of the first depressive or manic episode, greater symptom severity, poorer
functioning, a lifetime history of suicide attempt comorbid anxiety and substance abuse.10) In
addition, results from the Systematic Treatment Enhancement Program for Bipolar Disorder
(STEP-BD) have linked tobacco smoking to clinical variables including more lifetime manic and
depressive symptoms, rapid cycling illness, comorbid mental disorders, substance abuse and
active illness.8) However cross-sectional studies have showed inconsistent findings.13,14) In the
most recent study, illness severity, number of episodes, history of psychotic symptoms, history
of suicide attempts, age at onset of the illness, abuse of psychoactive drugs, current clinical status
and other demographic variables and their relationship with tobacco smoking were examined in a
sample of 102 participants with bipolar disorder. There was a higher tobacco smoking prevalence
in this population and a higher presence of alcohol and other drugs use and earlier age of illness
onset in those patients with bipolar disorder who smoked more. No other association with regard
to illness severity was found in this study. However, the authors caution against generalisation
due to the small sample size and the cross sectional study design. Furthermore, nicotine levels
were not measured in this study. Similarly, Cassidy et al.13) reported no association between
smoking and psychosis in patients with bipolar disorder, but no measures of mood were included
in the analysis.

Inconsistent findings on the impact of tobacco smoking on prognosis in bipolar disorder have
also been recently reported in a comprehensive review on the effects of smoking on the health
outcomes in bipolar disorder.32) In the reviewed studies, tobacco smoking was correlated with
rapid cycling,4,8,33) more affective episodes,32,33) elevated risk of psychotic episodes,12) and
more time spent in hospital,20,30) in some studies but not in others. Individual differences in
nicotine dependence may account for the variability of the findings between studies.
Methodological complexities including limited number of participants, bipolar patients who are
nicotine dependent versus smokers who are non-nicotine dependent, admixtures of prospective
and cross-sectional studies, diverse measures and lack of objective biological markers might
account for the difference observed.
Quality of Life and Tobacco Smoking in Bipolar Disorder

Tobacco use enhances mortality and decreases quality of life through a number of smoking-
related health diseases such as cardiovascular, cancer or respiratory conditions.34) Tobacco
smoking has also been related to more severe course, relapse and higher number of
hospitalizations,20) and is also associated with an elevated risk of concomitant substance use
disorders.24) Few studies have examined the quality of life in smokers versus non-smoker with
bipolar disorder. Consistent findings across studies have shown that quality of life is higher in
non-smokers and occasional smokers compared with nicotine dependent bipolar
patients.11,21,22) Smoking has also been associated with low socioeconomic status in patients
with bipolar disorder,8) and with the risk of anxiety disorders in the general population,23)
which may be moderating or mediating pathways.

In a study examining the quality of life in bipolar disorder in comparison with that of the general
population, 48 euthymic and 60 non-euthymic bipolar patients were compared with 1210
controls using the physical and mental component scores of the Medical Outcomes Survey 36-
item Short Form Health-Survey (SF-36).21) Clinical and course of illness variables and their
association with low physical and mental quality of life were explored including manic and
depressive symptoms and consumption of other addictive substances. Lower mental quality of
life was associated with longer duration of illness and early onset in bipolar patients. A higher
presence of depressive symptoms, nicotine dependence and lack of social support was also
reported in the bipolar group. Further studies have replicated this association. Quality of life in
terms of physical health, psychological wellbeing and environmental and social relationships was
lower in bipolar patients with nicotine dependence than in those without nicotine dependence.22)
Worse health related quality of life scores were also found for daily smokers.11)

It is plausible that tobacco smoking in bipolar disorder might be indirectly influencing this
construct through modifiable lifestyle factors that have also shown to be related to bipolar
disorder. Poorer diet and lower levels of physical activity have been associated with smokers, in
addition to these factors already being present in bipolar disorder.24–27) Furthermore, tobacco
smoking may play a detrimental role in sleep quality for those with bipolar disorder. Sleep
disturbance has comprehensively and independently been associated with poor health-related
quality of life.35) In a study including 252 participants with mood disorder, the severity of the
disorder was positively associated with the mean number of cigarettes smoked per day and with
greater levels of fatigue. Additionally, mild to severe symptoms of insomnia were related to the
level of nicotine dependence.28) Differences between smokers and non-smokers in circadian
rhythmicity and mood variation have been reported in the general population.19) Altered sleep
patterns play a detrimental role in bipolar disorder,29) as disrupted night time sleep and day time
mood have been proposed to be mutually reinforcing.36) This bidirectional model, known as the
sleep-mood cycle, is of particular importance to bipolar disorder and insomnia.36) Disruptions in
circadian rhythms have been identified as a major trigger for the onset and relapse of bipolar
disorder.37–40) These modifiable lifestyle factors not only play a role in quality of life for
bipolar disorder, but these relationships imply that they could play a critical part in the
management of the illness.
The Relationship between Tobacco Smoking and Suicidal
Behaviour in Bipolar Disorder

To date, many studies have identified associations between psychiatric disorders, comorbidities
and suicide attempts. This has been demonstrated across several studies utilizing adequate study
designs with large sample sizes.41) In particular, tobacco smoking has also been independently
linked to increased risk of suicidality, an association that was sustained after adjustment for
psychiatric disorders.42)

Individuals with bipolar disorder, have higher rates of suicide than the general population.43)
Moreover, the contribution of disability and mortality related to physical causes of death, suicide
and injuries incurred via self-harm in this group are significant.10) Due to the deleterious
outcomes for individuals and significant burden that both tobacco smoking and suicidal
behaviour incurs, it is necessary to understand the role of this potentially modifiable lifestyle
factor and its link to suicidal behaviour in this group to reduce mortality and disability.

Several studies have demonstrated associations between tobacco smoking in bipolar disorder and
suicidal behaviour.8,10,33,44,45) In one study, predictors of suicidal acts were investigated in a
2-year follow-up study of patients (n=308) presenting with major depressive disorder or bipolar
disorder at baseline.44) Cox proportional hazards regression analyses revealed that tobacco
smoking was among the strongest predictors for future suicidal acts, alongside history of suicidal
acts and self-reported severity of depression. Furthermore, among other significant clinical
predictors for suicidal acts were presence of a comorbid Cluster B personality disorder, and
aggressive and impulsive personality traits. These personality features and traits have been
shown to be risk factors for self-mutilation, help-seeking suicide threats and/or attempts and
completed suicide,46–48) and frequently co-occur with bipolar disorder.49) It is possible that
comorbidity among impulsive/aggressive features and smoking increases the risk for suicidal
behaviours in bipolar disorder.

A retrospective clinical study (1999–2004) also demonstrated significant associations between

lifetime tobacco smoking and suicide attempts in patients with bipolar disorder, after adjusting
for other psychiatric comorbidities and measures of disability and functional impairment.10)
Finally, subsequent prospective analyses have also reported that tobacco smoking was associated
with increased risk of suicidality in individuals with bipolar disorder (n=106), throughout the
course of a nine-month follow-up.35) These baseline associations were not sustained after
adjustment for impulsivity in the models. Moreover, small sample sizes precluded further
analyses examining associations between impulsivity, smoking and suicide attempts.
The increased risk for suicide attempts among those with bipolar disorder, and who smoke,
appears to true even for youths. The Course and Outcome of Bipolar Youth Study examined 212
youths (7–17 years) with bipolar disorder and correlates. Differences between three groups of
tobacco smoking status (daily, never smoked, ever smoked) were observed, with daily smokers
having a higher lifetime prevalence of suicide attempts. Logistic regression found those with a
history of tobacco smoking had approximately a three-fold increased risk of lifetime suicide
attempt compared to those without. However, it was beyond the scope of the study to examine
the prevalence and correlates of nicotine dependence among the youths with bipolar disorder.
Further, longitudinal analyses of subsequent follow-up visits may elucidate temporal
relationships smoking and suicidality in the younger with bipolar disorder.

Contrary these findings, others have reported no significant associations for tobacco smoking
and suicidality in bipolar disorder in longitudinal analyses.50–52) In one study, 40.5% of those
who had attempted and/or completed suicide over the observational period (2 years) were
smokers.48) Whilst there was an initial trend for associations between baseline tobacco smoking
status and increased risk for suicide attempts, the association was not sustained in logistic
regression models,52) as was the case in another study with a similar follow-up period.51) An
additional prospective study also reported that tobacco smoking did not predict suicide attempts
over the observational period (18 months).50) A review examining the clinical and health
outcomes of a number of psychiatric disorders, also suggested that associations between tobacco
smoking and suicidal behaviour in patients with bipolar disorder was inconclusive.32)

These studies highlight that tobacco smoking may be independently associated with suicide
attempts in individuals with bipolar disorder. However, there remains areas of uncertainty about
the relationship between tobacco smoking and bipolar disorder in regards to suicidal behaviour.
It may be that personality variables (linked to impulsivity) may be important effect modifiers in
the relationship between tobacco smoking and suicidality, particularly during acute phases of the
disorder. Examining the potential role of personality as a mediating and/or moderating variables
in future studies examining tobacco smoking and comorbidities is warranted.
Biological Interactions between the Pathophysiology of
Bipolar Disorder and Tobacco Addiction

To date, operative mechanisms linking tobacco smoking and bipolar disorder remain unclear. In
unipolar depression, the role of neurotransmitters has been hypothesized. Associations may be
related to lower levels of brain serotonergic function in smokers.53) Lower levels of serotonin in
both humans and primates have been linked with higher levels of impulsive, novelty seeking
behaviour.54) However, given discrepancies between the underlying biology in unipolar and
bipolar depression (see next section), caution needs to be taken when considering these links.
Additionally, dysregulation of the dopamine and glutamatergic systems and their inputs to
different regions of the brain (particularly prefrontal cortex and basal ganglia) have been linked
to the maintenance and relapse of addictive behaviour.55–57) Nicotine is a nicotinic cholinergic
receptor agonist that causes the release of both dopamine and norepinephrine. These two
neurotransmitters are implicated in the pathophysiology of bipolar disorder, and are also thought
to play a key role in addiction. This may influence susceptibility of people with bipolar disorder
to the rewarding effects of nicotine. Furthermore, smokers have reduced monoamine oxidase
which is a potential pharmacological effect of tobacco smoke as it prolongs the synaptic presence
of neurotransmitters involved in mood regulation (i.e., serotonin and dopamine).58) Cholinergic-
monoamine pathway interactions may link nicotine dependence and bipolar disorder given the
putative role of dopamine, serotonin nor-epinephrine and glutamine,4,59) in the pathogenesis of
both bipolar disorder and nicotine addiction and the role of monoamine neurotransmitters in
mood disorders.

In this regard, cyclical dopamine dysregulation has been proposed as a central mechanism
underlying bipolarity.60) The activation of the mesolimbic dopaminergic pathways through the
diffuse cholinergic innervations in the brain is thought to play a salient role in the generation of
reward and reinforcement of tobacco smoking.61) Further to this, dynamic desensitization of
nicotinic cholinergic, dopaminergic and other downstream receptors may account not only for
nicotine tolerance and withdrawal symptoms but also for smoking conditioning through the
rewarding effects of nicotine in the brain. An increase in dopaminergic transmission would
predominate during the manic phase there is whereas a secondary down regulation of receptors
would eventually dampen the cycle learning to a phase of decreased dopaminergic transmission,
characteristic of the depressive phase.60) This theory of the role of the dopaminergic system
provides support for the relationship between tobacco smoking and mood disorders. There might
be a number of other underlying neurobiological mechanisms to account for this relationship
given the vast array of other neuro-chemicals triggered by nicotine. Enduring changes in
excitatory glutamatergic synapses on the nucleus accumbens are induced by nicotine use,55,62)
and have been shown to be involved in drug relapse.63) Recently, disturbances in glutamatergic
function have also been implicated in the pathophysiology of mood disorders (major depressive
disorder and bipolar disorder).46–49) The role of inflammation and oxidative and nitrosative
stress as a plausible mediator for the development of depressive and bipolar disorders and its
relationship to tobacco use have also been examined in recent years. The literature also provides
some support for a potential interaction between tobacco smoking and the neuroprogressive
course of the disorders as these pathways are affected by exposure to tobacco smoke components
such as nicotine and free radicals.64–66)
Effects of Tobacco Smoking on Pharmacological
Treatment in Bipolar Disorder

The effect of tobacco smoking on the impact of the metabolism of many psychotropic
medications used in the treatment of bipolar disorder is well documented.4,11,67) Polycyclic
aromatic hydrocarbons in tobacco smoke, increase the metabolism of some psychotropic
medications, including olanzapine, clozapine, haloperidol and fluvoxamine.4) This is largely
mediated via the induction of the hepatic cytochrome P450 enzyme 1A2 by the polycyclic
aromatic hydrocarbons found in tobacco smoke, leading to lower serum levels of drugs including
olanzapine as well as other antipsychotics, antidepressants and benzodiazepines. This may be a
pathway whereby tobacco smoking reduces the therapeutic benefits of taking these medications.
This results in smokers with bipolar requiring an increased dose, approximately 50% higher, to
achieve an adequate therapeutic response.67)
Additionally, some case reports suggests that a substantial increase in tobacco smoking can lead
to an increase in psychiatric symptoms among individuals with bipolar disorder who are taking
olanzapine, at least in part due to the reduction in medication levels explored the dose dependent
effect of tobacco smoking on serum concentrations of clozapine and olanzapine.67) The mean
serum concentration to dose ratio was twice as high in non-smokers compared to smokers for
both olanzapine and clozapine, leading to authors to consider differential starting doses in
smokers compared to non-smokers.

It has been further reported that nicotine’s activation of the sympathetic nervous system may
reduce the sedating effects of benzodiazepines and the capacity of beta blockers to lower heart
rate and blood pressure.4) The evidence that tobacco smoking not only interferes with a range of
different medications, points to the widespread effect that nicotine and tobacco smoke plays on
multiple neurotransmitter and brain systems.11,67–69) Dani and De Biasi68) suggest fast
nicotinic transmission is used throughout the peripheral nervous system and at the neuromuscular
junctions, leading to body-wide impacts. This view is supported by Laviolette and van der
Kooy,69) who state that nicotine produces diverse neurophysiological, motivational and
behavioural effects throughout several brain regions via multiple neurochemical pathways.

The far-reaching ability of nicotine to produce changes in multiple neurotransmitter systems is

possibly through the widespread cholinergic innervations that nicotine plays on the brain. As
nicotine is a nicotinic cholinergic receptor agonist it enhances the release of a cascade of
neurotransmitters including dopamine, serotonin, norepinephrine and glutamine. Agents that
drive dopamine such as cocaine and amphetamine are among the best models of mania, while
dopamine D2 antagonists are robustly anti-manic and include several antipsychotic agents.61)
Nicotine may thus indirectly drive dopamine itself as well as counteract the effects of
pharmacological interventions, firstly via its psychostimulant properties, and secondly via
pharmacokinetic interactions as highlighted above.

Hutchison et al.70) looked at the atypical antipsychotic, olanzapine, in attenuating cue elicited
craving for tobacco. The hypothesis was that if mesolimbic dopaminergic activity is associated
with cue-elicited cravings for tobacco, a dopaminergic antagonist should attenuate cue-elicited
cravings for tobacco. Olanzapine attenuated cue elicited craving for tobacco, but did not
moderate the subjective effects of tobacco smoking in a randomly assigned five day placebo
controlled trial. However, dopamine antagonists decrease the pleasure, reinforcement and reward
of tobacco smoking, which might lead to increase smoking behaviour.71)
Tobacco Smoking Cessation and Bipolar Disorder

The rate of successful cessation attempts for people with mental illness is low. Data from the
United States 2002 National Survey on Drug Use and Health demonstrated that the quit rate was
49% for former smokers with no serious psychological distress, determined using the K6 scale,
whereas the quit rate was 29% for former smokers with serious psychological distress.72) The
reason for these reduced quit rates has been examined by tobacco control researchers, with one
view describing the concept of ‘hardening’ and ‘hard-core smokers’, where reduced quit rates are
attributed to greater difficulty in quitting or less willingness to do so.73) In the general
population, evidence of hardening remains elusive as tobacco smoking rates continue to decline;
however, amongst people with bipolar and other major psychiatric disorders smoking rates have
not declined at the same rate and hardening may be a factor. However, it is very unlikely to fully
explain the disparity in quit rates between the mentally ill and general populations. Smokers with
mental illness may be as motivated to quit smoking tobacco as smokers without a current mental
illness. This has been demonstrated in studies of an inpatient population58) in Australia. There is
some data indicating less readiness to quit amongst bipolar patients in India74); however, this
may be attributed to a lesser awareness of the health risks of tobacco smoking in that specific
population. This suggests that willingness to quit may not be a factor in the discrepancies
between quit rates between mentally ill and not mentally ill smokers. Greater difficulty in
quitting smoking may still be a ‘hardening’ factor in this population as most readily available
tobacco smoking cessation strategies have been developed for the general population and may
not be as effective for mentally ill smokers.

Other non-hardening factors may also be applicable to this population, including less awareness
of the health risks associated with tobacco smoking, a factor likely to vary considerably between
and even within different countries. There are reports that some people with mental illness may
feel alienated from approaching mainstream tobacco smoking cessation services.8) However, the
reasons for the disparity in smoking cessation rates between mentally ill and not mentally ill
people have not been sufficiently researched.

In a review of tobacco smoking cessation in bipolar disorder, George et al.33) found just 11
clinical studies on bipolar disorder and tobacco addiction showing that people with bipolar
disorder have much lower quit rates than smokers without a comorbid condition. Several case
reports as reviewed by Dome et al.32) suggested that varenicline treatment, for tobacco
cessation, may provoke mania in patients with or without a history mood disorders. However this
has not been borne out by large scale well designed randomized trials.60)

There have only been four published controlled smoking cessation treatment studies in bipolar
smokers. The first two are relatively small studies, with two recent more definitive studies with
varenicline. The initial small pilot study (n=5) but showed buproprion was well-tolerated and
potentially useful during 10 weeks of treatment.75) Following on from this, Wu et al.76)
conducted another tiny (n=5) study of varenicline for tobacco cessation in those with bipolar
disorder. Over the 10-week treatment phase, reduction in tobacco smoking in the varenicline
group was found and it was well tolerated. In the first large randomized controlled trial of
varenicline for tobacco cessation maintenance in individuals diagnosed with bipolar disorder and
schizophrenia, 87 participants were included in the maintenance phase, of which only 13 were
diagnosed with bipolar disorder (the remaining were diagnosed with schizophrenia). The study
included an open-label phase for 12 weeks following which participants who were abstinent for
at least 14 days were then randomized to receive either continued varenicline treatment, or
placebo up to 52 weeks of treatment.60) Varenicline was efficacious as a maintenance treatment
in the study, and this effect remained when those with bipolar disorder were analysed separately.
The largest and most rigorous study was that by Evins et al.,60) who enrolled 247 smokers with
schizophrenia or bipolar disorder from 10 community mental-health centres. Of these, 203
received 12-week treatment with both open-label varenicline and cognitive behavioral therapy,
after which 87 met abstinence criteria and entered the relapse prevention intervention. After a
year, the point-prevalence abstinence rates were 60% with varenicline treatment and 19% with
placebo, giving an odds ratio of 6.2. Importantly, there were no impacts on psychiatric
symptoms.

This paper has highlighted that tobacco smoking is deleterious to mental health, but a critical
question is whether mental health benefits from tobacco smoking cessation. A critical review and
meta-analysis has recently been published, with striking findings. Firstly, quitting has been
linked to reduced anxiety and stress, depression, and improved positive mood as well as quality
of life compared with ongoing tobacco smoking. Secondly, effect sizes are equivalent for people
with psychiatric disorders compared to those without. Lastly, effect sizes appear similar or
greater than those of antidepressant treatment for the abovementioned disorders.77) This finding
highlights the necessity to work towards cessation.

It is not clear how best to assist smokers with mental illness to quit tobacco smoking. There are
additional factors that may further contribute to the success or failure of smoking cessation in
those particularly, diagnosed with bipolar disorder. Given the high rates of comorbid anxiety
disorders and substance abuse the complexities of treatment for smoking cessation are increased.
George et al.33) have suggested that the alternating mood fluctuations (ranging from irritable
and elevated mood to severe depression) may contribute to differences in smoking cessation
rates found in bipolar disorder. Similarly, Ostacher et al.10) reported a relationship between
comorbid anxiety disorders and substance abuse in lifetime smokers with bipolar disorder,
suggesting further difficulty in smoking cessation. Current smoking cessation interventions for
individuals with bipolar disorder include several first line pharmacotherapies that include
nicotine replacement therapy, bupropion, sustained release and varenicline.3) It has also been
reported that olanzapine can assist in cue prompted cravings for tobacco.70) The problem with
most of these therapies is that it is unclear how safe and efficient these aids are for people with
bipolar disorder.3) Unlike pharmacotherapy, behavioural interventions have been tested but
unfortunately, current evidence is insufficient to recommend these approaches as alternatives to
standard counselling.3) However, behavioural interventions improve the chances of quitting,
especially when used in conjunction with pharmacotherapy.78) Examples of behavioural
interventions include cognitive behaviour therapy, mood management interventions and
motivational interviewing.

Most large studies use mixed populations that are generally heavily weighted towards high
prevalence disorders. El-Guebaly et al.79) reviewed 24 studies of tobacco smoking cessation in
people with mental disorder, where the majority of interventions combined pharmacotherapies
and psychoeducation, and found that quit rates for people with mental illnesses were similar to
those expected for the general population. On the basis of these findings, Australian tobacco
smoking cessation programs give recommendations for smoking cessation strategies for people
with mental illnesses that are largely identical to recommendations given or people without
mental illnesses; apart from warning about altered pharmacokinetics of some psychotropic
medications, the possibility of withdrawal symptoms being mistaken for symptoms of mental
illness and a risk of exacerbation of depressive symptoms. A government funded tobacco
smoking cessation program in Brazil identifies people with mental illnesses and substance abuse
and tailored their smoking cessation intervention to their needs.80) Kerr and colleagues81)
describe barriers to mainstream tobacco smoking cessation programs for people with mental
illnesses that include; perceived benefits of smoking to their symptoms of mental illness, fear of
illness worsening, social circles where smokers predominate, low self-confidence, health
professionals who are not encouraging their cessation efforts and are not trained in tobacco
smoking cessation and reluctance to access mainstream smoking cessation services. Furthermore,
health professionals are not always aware of the importance of addressing smoking cessation in
this vulnerable group.82) While there is insufficient evidence to recommend any specific tobacco
smoking cessation technique for people with bipolar disorder, there is sufficient evidence to
conclude that tobacco smoking cessation interventions should be tailored to the individual needs
and preferences of the person being treated. For some individuals with bipolar disorder
mainstream tobacco smoking cessation techniques may be adequate. For other individuals it may
be more effective to provide tobacco smoking cessation support delivered through their usual
mental health care provider and providing support techniques additional to those provided by
mainstream services. Health professionals should encourage smokers with mental illness to cease
tobacco smoking.
FUTURE DIRECTIONS FOR RESEARCH: TOBACCO
SMOKING IN BIPOLAR DISORDER

In this review, an attempt has been made to understand the relationship between smoking and
bipolar disorder. As Pasco et al.83) note, it is becoming increasingly clear that tobacco smoking
is not only harmful to our physical health, but also for our mental health, which may in fact
contribute to its onset or aggravate current symptoms.

It is possible that certain individuals are more vulnerable to nicotine’s psychostimulant properties
than others and researchers have implied that there is the possibility that separate causal
mechanisms may operate in a bi-directional manner.84) Many of the studies evaluated did not
assess for nicotine dependence, which may play a bigger role on nicotine’s effect on mood
symptomatology that rates of daily smoking. It is also clear that more studies need to be
conducted to accurately assess the consequences that dependent tobacco smoking places on
bipolar disorder. It is suggested that consistent measures of mood symptoms as they apply to
bipolar disorder are assessed, as well as consistent measures to assess nicotine dependence
amongst individuals with bipolar disorder.

In conclusion, people with bipolar disorder have much higher tobacco smoking rates compared
to the general population. Furthermore, although the relationship between tobacco smoking and
bipolar disorder may be bidirectional and that cause and effect may be difficult to determine, the
majority of studies reviewed suggest that tobacco smoking appears to play a mediating role in
the severity of bipolar disorder. Clear gaps exist in the literature, and there is a need for
researchers to look at formal smoking cessation programs and the relationship between tobacco
smoking, insomnia, lifestyle factors and quality of life in bipolar disorder. Caggiula et al.85)
suggest, the abuse liability associated with tobacco smoking equals or exceeds that of other
addictive drugs. It is also suggested that tobacco smoking is screened by mental health
professionals and targeted during routine treatment, with the implementation of more formal
cessation programs.86)