Sleep as a Target for Optimized Response to Smoking Cessation Treatment

Abstract

Declining national rates of current tobacco use to an all-time low of 15.1% represents a public health victory. Undermining this progress, however, are smoking rates of up to 50% among high-risk, low-income populations. Current FDA-approved treatments for nicotine dependence are ineffective with between 70–95% of treatment-seekers relapsing within the first year of attempted abstinence. Thus, identification of novel intervention targets to optimize response to currently available treatments for nicotine dependence is a critical next step. One such target may be sleep insomnia. Insomnia is a clinically verified nicotine withdrawal symptom but, to date, addressing insomnia or other sleep disturbance symptoms as an adjunctive smoking cessation therapy has yet to be fully considered. To this end, this manuscript presents a narrative review of: (1) sleep continuity and architecture in smokers versus nonsmokers; (2) effects of nicotine abstinence on sleep; (3) possible mechanisms linking sleep with smoking cessation outcomes; (4) plausible adjunctive sleep therapies to promote smoking cessation; (5) possible treatments for unhealthy sleep in smokers; and (6) directions for future research. Taken together, this will provide conceptual support for sleep therapy as an adjunctive treatment for smoking cessation.

Implications

This narrative literature review presents a comprehensive discussion of the relationship between habitual sleep and cigarette smoking. The extent to which unhealthy sleep in smokers may be a viable intervention target for promoting response to smoking cessation treatment is considered. Ultimately, this review provides conceptual support for sleep therapy as an adjunctive treatment for smoking cessation.

Introduction

Despite declines in adult cigarette smoking prevalence during the past 50 years, cigarette smoking remains the leading cause of preventable death and disability in the United States. Data show that cigarette smoking and secondhand smoke exposure are accountable for at least 443000 premature deaths and up to $289 billion in direct health care expenditures and productivity losses each year.¹ Mortality associated with continued tobacco use is well-documented: 33% of cardiovascular and metabolic diseases, 32% of all cancers (including 87% of lung cancer), and 62% of pulmonary and respiratory diseases are attributable to cigarette smoking.² In spite of these adverse health effects, 15.1% of adults in the United States (~36.5 million people) are current smokers, with rates of 33–48% reported among demographic subgroups including those who are uninsured, low-income, and low-education.³ Data also show that adults with a mental health disorder (eg, anxiety disorder) are twice as likely to smoke than those in the general population.⁴ Thus, smoking cessation remains a public health priority.

Current FDA-approved treatments for nicotine dependence, including nicotine replacement therapies (eg, nicotine patch, spray, gum, lozenge) and nonnicotinic treatments (eg, bupropion, varenicline) are suboptimally effective. Although these treatments do double the odds of 6-month abstinence compared to placebo, less than one quarter can expect to remain abstinent.⁵Healthy People 2020 has set the national goal of a 12% smoking prevalence rate for all demographic groups; achieving this goal will require the development of more effective treatments for smoking cessation, as well as strategies to optimize current treatments.⁶ As a common biologic function that plays a central role in metabolic regulation, emotion regulation, performance, memory consolidation, brain recuperation processes, and learning, sleep may be such an intervention target that may optimize nicotine dependence treatment response. For example, insomnia (difficulty falling and/or staying asleep) is a clinically recognized nicotine withdrawal symptom⁷ that is not addressed in the clinical guidelines for nicotine dependence treatment.⁵ On this basis, it is the purpose of the current manuscript to consider the extent to which addressing sleep behaviors in treatment-seeking smokers could be an effective adjunctive treatment approach for nicotine dependence. To achieve this, we will first consider differences in sleep quality metrics in smokers versus nonsmokers. Next, a review of the effects of smoking abstinence on sleep quality and a brief overview of the possible mechanisms that may link sleep with smoking cessation outcomes will be provided. Following this, a review of evidence-based treatments for sleep disturbances will be considered with the goal of identifying sleep therapies that could be used in the context of smoking cessation. Last, future research directions needed to validate the extent to which poor sleep quality may be a viable target with which to optimize response to standard nicotine dependence treatment, will be considered.

Sleep Continuity and Architecture in Smokers Versus Nonsmokers

Overview of Sleep Continuity and Architecture

Sleep is quantified by metrics of sleep continuity and sleep architecture. Sleep continuity refers to the timeline of when an individual is asleep, compared to the time when they are intending to sleep. For example, key metrics within sleep continuity include the timing of sleep, the total amount of time spent in bed (time in bed, or TIB), sleep latency (time to fall asleep, or SL), number of awakenings, total time awake after sleep onset (also referred to as “wake after sleep onset” or WASO), time of final awakening, total sleep time (computed as TST = TIB − SL − WASO), and sleep efficiency (the proportion of time spent in bed actually asleep, computed as [TST/TIB]*100).⁸

Sleep architecture represents the cyclical pattern of sleep as it shifts among the various sleep stages, including non-rapid eye movement (NREM) and rapid eye movement (REM) sleep. Polysomnography (PSG) provides objective assessment of the different sleep stages that are detailed in Table 1; the temporal and percentage of time in each of these stages are key markers of individual sleep quality. Briefly, the three NREM stages (N1, N2, and N3) roughly parallel a depth-of-sleep continuum, with arousal thresholds generally lowest in N1 and highest in N3 sleep. N1 and N2 sleep stages are associated with minimal or fragmentary neuronal activity. REM sleep is characterized by heart rate, breathing rate and brain wave activity that is similar to waking levels, compared to other stages of sleep.⁹ REM sleep (as with N3) is important for cognitive tasks such as memory consolidation and information processing; dreaming predominantly occurs during REM sleep.¹⁰ Throughout the sleep period, adults will cycle between stages of NREM and REM, spending 75–80% of sleep time in NREM and the remainder in REM sleep.⁹

Table 1.

Stages of Sleep

Sleep Stage	Characteristics	Neuronal Activity	Muscular/Breathing/Heart Rate Activity	Brain Waves
NREM Stage 1 (N1)	• Drowsy Sleep • Lasts less than 10 min; ~5% of total sleep time	• Transition between waking and sleep	• Muscles active, eyes rolling slowly • Breathing becomes more regular; heart rate slows	• Transition from unsynchronized beta and gamma to slower alpha waves
NREM Stage 2 (N2)	• Light Sleep; first stage of sleep • 45–50% of adult sleep time in this stage	• Information processing and memory consolidation occurs	• Muscle activity decreases • Breathing becomes more regular; heart rate slows	• Theta wave range • Stage characterized by short bursts of brain activity (sleep spindles) and K complexes • K complexes are short negative high-voltage peaks followed by slower positive complex and then a final negative peak • Sleep spindles and K complexes protect sleep state and suppress outside stimuli
NREM Stage 3 (N3)	• Deep Sleep • 15–20% of adult sleep time	• Sleeper unaware of any sounds or stimuli • Information processing and memory consolidation occurs	• Brain temperature, breathing and heart rate, and blood pressure are at their lowest	• Slow-wave delta
REM	• Occurs in cycles of 90–120 min throughout the sleep period • Dominates latter part of sleep period; accounts for 20–25% of adult sleep time	• Majority of dreams occur in this stage • Lack of REM sleep shown to impair performance in complex tasks • Procedural and spatial memory consolidation	• Rapid and possibly random side-to- side, intermittent eye movements • Breathing becomes more rapid and irregular • Heart rate and blood pressure increase to near waking levels • Musculature unresponsive in this sleep stage	• Brain activity mimics waking state: theta, alpha and high-frequency beta waves present

Self-reported perceptions regarding sleep are also valuable metrics. Sleep disruptive events (i.e., sleep walking, night terrors) and daytime sleepiness or dysfunction (i.e., sleepiness, lack of energy, drowsiness that may prevent the completion of daytime tasks) are commonly measured characteristics of sleep.¹¹ A growing body of literature has compared these and other sleep variables in smokers and nonsmokers; a review of this work is provided below.

Sleep Architecture in Smokers versus Nonsmokers

Five studies were found that used polysomnography to examine sleep architecture in smokers and nonsmokers.^12–16 Three of the five studies found that, compared to nonsmokers, smokers had a significantly higher percentage of time in N1. For example, Zang and colleagues found that among 779 smokers and 2916 never smokers, current smokers accrued 24% more N1 sleep;¹² this would indicate shallower, more disturbed sleep. In another study of women (N = 63 smokers and N = 323 nonsmokers) the mean time in minutes in N1 was 31 for smokers and 21 for nonsmokers.¹⁵ Similarly, smokers in these studies were reported to have a significantly higher percentage of N2 sleep, but significantly lower percentage of N3 sleep.^12,16 Jaehne and colleagues reported that in a laboratory-conducted PSG assessment of 44 smokers and 44 matched nonsmokers, smokers reported a higher REM density than their counterparts.¹³ Collectively, this small body of work suggests that smokers may spend less time in deeper, more restful sleep-states than nonsmokers.

Sleep Continuity in Smokers versus Nonsmokers

In terms of sleep onset latency, there is consensus across PSG verified studies that smokers (vs. nonsmokers) have a longer sleep onset latency,^12–16 shorter sleep duration,^12,13 and later sleep timing.¹⁷ PSG verified sleep onset latency has been reported to be 5.4,¹² 8.0,¹³ 10.0,¹⁵ and 24.9¹⁴ min longer in current versus nonsmokers. Mean total sleep time/duration has also been found to differ between smokers and nonsmokers, with smokers having shorter sleep. In one study, smokers reported 13.3 fewer minutes of total sleep time¹³ and 14.0 min in another study.¹² Overall, smokers recorded significantly more time awake after sleep onset.¹⁴

Findings from these PSG studies showing longer sleep onset latency and shorter duration in smokers versus nonsmokers are consistent with the self-report literature. Using data from the National Health and Nutrition Examination study, Branstetter and Colleagues found that current smokers took almost 25.9 (SD = 21.3) min to fall asleep compared to 21.5 (SD = 19.5) min in former smokers, and 22.1 (SD = 19.3) min in never smokers.¹⁸ Other studies have found self-reported sleep latency to be significantly longer in smokers than nonsmokers.^19–21

In terms of differences in sleep duration and sleep timing, smokers report shorter sleep duration, and later sleep timing than nonsmokers. For example, population level data from the National Health and Nutrition Examination Survey showed that mean sleep duration in smokers is 6.6 h versus 6.9 h in non/never smokers.¹⁸ Although data from the United Kingdom Biobank showed that in a sample of 34401 smokers, 30.8% reported short sleep (≤6 h), and 9.3% reported long sleep (≥9 h) duration.¹⁷ Several other studies found self-reported sleep duration to be significantly shorter in adult smokers than nonsmokers,²¹ with one study showing significance for light smokers (<15 cigarettes per day) versus nonsmokers, only.²² Using data from N = 323047 adult respondents of the 2009 Behavioral Risk Factor Surveillance System, Grandner and colleagues found that self-reported insufficient sleep was highest among daily current smokers and lowest among those who never smoked.²³ In terms of sleep timing, data from a national sample of adults showed that current smokers had a more than two-fold greater odds of having an evening versus intermediate timing preference,¹⁷ as well as a 40% greater odds of waking up too early.²¹

When the relationship between smoking status and sleep efficiency is considered, two of the five studies that used PSG assessment reported differences. Jaehne and Colleagues found that smokers had poorer sleep efficiency that was not significantly different from nonsmoker levels (87.08% vs. 89.84%, respectively),¹³ whereas Redline and Colleagues report that sleep efficiency was significantly lower in smokers than nonsmokers.¹⁶

Together, these objective assessments of sleep continuity markers indicate that smokers have poorer sleep continuity than nonsmokers as suggested by longer sleep latency and shorter sleep duration. Lower sleep efficiency was indicated by some, but not all studies reviewed.

Sleep Fragmentation in Smokers versus Nonsmokers

One PSG study of smokers and nonsmokers observed that smokers had significantly more disruptive events such as general leg movements and a higher leg movement index as compared to nonsmokers.¹³ In one of the more comprehensive studies from the self-report literature examining the relationship between sleep and smoking status, a global disturbed sleep quality index was found to be significantly more prevalent in smokers versus nonsmokers (28.1% vs. 19.1%).¹⁹ Other data show only male smokers to have significantly greater prevalence of nightmares and disturbing dreams as compared to nonsmokers.²⁴

Among smokers, nocturnal awakenings to smoke are common, reported in 19–51% of smokers.^25–27 One study showed that among night smokers, night smoking occurred on one-in-four nights (26%) and averaged two episodes per night.²⁶ Night-time smokers are more nicotine dependent^25,26 and, following a cessation attempt, are more likely to relapse.²⁶ These studies indicate that smokers may be vulnerable to sleep fragmentation and disruptive events.

Daytime Sleepiness in Smokers versus Nonsmokers

Across several longitudinal and cross-sectional studies, smokers are more likely to report daytime sleepiness than nonsmokers. In one longitudinal, observational study of 3516 adults, excessive daytime sleepiness was related to current smoking in females and not males.²⁴ In a study that used self-report NHANES data to examine sleep characteristics of current (N = 2015), former (N = 2741), and never smokers (N = 5752), results showed that current smokers reported significantly more occurrences of feeling unrested and overly sleepy during the day as compared to the comparison groups.¹⁸ Cross-sectional data from the Behavioral Risk Factor Surveillance System also showed that smokers reported significantly more daytime sleepiness.²⁰

Summary

Together, these data suggest that smokers are vulnerable to deficits in sleep continuity and architecture. From a sleep continuity perspective, smokers are more vulnerable to longer sleep latency, more awakenings, poorer sleep quality, and shorter sleep time. From a sleep architecture perspective, shorter percentage of time in slow wave sleep is more common in smokers than nonsmokers while subjective reports indicate that smokers have more restless sleep and greater daytime drowsiness and sleepiness.

Smoking Abstinence and Sleep

As a clinically verified symptom of nicotine withdrawal, insomnia is reported by up to 42% of abstinent smokers,^28–30 while up to 80% of smokers habitually experience sleep disturbances,³¹ that then become exacerbated following cessation.³² Nicotine withdrawal is a robust predictor of relapse to former smoking practices³³ and as such withdrawal symptoms are primary intervention targets. Elucidating the extent to which insomnia and other sleep deficits change following abstinence, and relate to smoking status and cessation outcome, is critical to quantifying the extent to which sleep may be a valid intervention target to promote cessation.

Changes in Sleep Following Abstinence

Three studies have objectively assessed sleep patterns (using polysomnography [PSG]) following cessation in treatment-seeking smokers. In the larger of the two studies, 33 smokers completed a PSG assessment at baseline, 24–36 h, and three-months following cessation.³² Results showed a significantly increased percentage of wake time after sleep onset and night-time arousal in the first 24–36 h of quitting; no significant differences were seen at the 3-month follow-up.³² In another study that included an analytic sample of seven treatment-seeking smokers, data showed that sleep duration and efficiency declined significantly in the first month of abstinence, however, by 1 year after cessation, sleep metrics had improved with reductions in latency to REM sleep and stage 1 (light) sleep and increases in REM (deep) sleep.³⁴ Wetter and Colleagues (1995) reported on a double-blind randomized trial that compared sleep architecture in 34 treatment-seeking smokers who received either active or placebo nicotine patches.³⁵ Sleep was PSG monitored for two nights before smoking cessation and three nights afterwards. The results showed that while sleep fragmentation significantly increased among placebo patch users, the active patch users did not demonstrate significant increases in sleep fragmentation following cessation.³⁵ Converging with these data from treatment-seeking smokers are data from a within-subject laboratory study that objectively compared the effects of smoking abstinence versus smoking-as-usual on sleep quality, daytime sleepiness and mood in a sample of 18 non–treatment-seeking smokers. Results showed that as compared to smoking-as-usual, nicotine abstinence significantly increased relative arousals, sleep stage changes, and awakenings in the first week of abstinence.³⁶ Collectively, these objective assessments of sleep metrics across the quitting period suggest that sleep deficits in the form of longer sleep latency, decreased sleep duration and efficiency are likely in the first weeks of quitting, but that these deficits are ameliorated 3–12 months after quitting.

Several studies have also examined self-reports of the natural history of withdrawal in abstinent smokers. Cummings and colleagues reported on a sample of 33 smokers who completed withdrawal diaries daily for a 21-day period following cessation.²⁸ Difficulty sleeping and daytime sleepiness in this sample did not show significant declines across the 21-day observation period as compared to the other withdrawal symptoms measured (ie, craving, irritability). Meanwhile, heavier smokers reported significantly higher mean scores of difficulty sleeping and daytime sleepiness than light smokers.²⁸ By contrast, electronic diary assessment of nicotine withdrawal duration and symptom severity showed that in 214 treatment-seeking smokers, sleep disturbances did dissipate in a 21-day monitoring period after abstinence.³⁷ Data from these self-report studies converge with findings from studies using objective measures of sleep by showing that following nicotine abstinence, smokers experience an exacerbation of insomnia-type symptoms (ie, longer sleep onset latency, more frequent awakenings) and shorter sleep duration and that cross time, these symptoms may dissipate.

Relationship Between Sleep and Cessation Outcome

Ten studies that explicitly examined one or more sleep metrics in relation to smoking cessation outcomes were reviewed.^{27,30,38–45} While the range of sleep metrics measured, the use of different tools to measure the same sleep metrics, the variability in smoking cessation treatments used, and time-period of assessment pre- and post-cessation across these studies makes direct comparison challenging, some points of commentary can be raised (see Supplementary Table 1).

First, eight studies showed that sleep metrics measured immediately before cessation and/or during cessation predicted relapse. For example, Peltier and colleagues reported that in a sample of 139 treatment-seeking smokers, increased sleep latency, reduced subjective sleep quality and increased daytime dysfunction in the first week of quitting were predictive of relapse 4-weeks after treatment while increased sleep disturbances were predictive of relapse 12-weeks after treatment.³⁸ Sleep disturbances alone did not predict relapse in a different sample of 385 treatment-seeking smokers. Instead, pre-cessation sleep disturbances interacted with waking at night to smoke (pre-cessation) to predict relapse 6, 24, and 48 weeks post-quitting.³⁹

Second, pre-treatment sleep habits are relevant to smoking outcomes. Four of the studies found that pre-cessation (versus abstinence-induced) sleep deficits were predictive of relapse.^39,41,42,45 In a sample of 579 smokers who received a 12-week anxiety-related smoking cessation program versus a control condition, the results showed that smokers who self-reported pre-cessation insomnia symptoms had a 11% greater odds (aOR = 1.11; 95% CI = 1.01–1.22) of relapsing 3-months following cessation than those who did not have pre-cessation insomnia symptoms. Post-quit insomnia was not related to cessation outcome.⁴¹ Likewise, in another study of 1136 smokers who received pharmacotherapy and counseling, data showed that smokers reporting more sleep disturbance pre-treatment were less likely to be quit at the end of treatment (OR = 0.79; 95% CI = 0.67–0.93).⁴² Dorner and colleagues reported that greater nocturnal awakenings at baseline was an independent predictor of relapse 5-weeks following cessation in a sample of 2471 treatment-seeking smokers.⁴⁵ The remaining studies reviewed either did not have an assessment of sleep in the first week(s) of cessation,⁴⁰ found that sleep patterns both before and after cessation predicted cessation outcome,³⁸ did not report results in sufficient detail to ascertain whether sleep quality before or after cessation was related most to cessation,⁴³ or did not find that sleep related to cessation outcome.³⁰

Third, only one of the studies reviewed was designed specifically to test the efficacy of a behavioral sleep intervention on cessation outcome in a small sample of 19 smokers with a clinical diagnosis of insomnia.⁴⁴ Fucito and Colleagues compared quit rates in 9 participants who received a cognitive-behavioral treatment for insomnia + smoking cessation counseling + transdermal nicotine versus smoking cessation counseling + transdermal nicotine. The results of this small study showed that participants receiving the experimental insomnia treatment reported better sleep quality and efficiency; they also had more days to relapse.⁴⁴

Some of the take-home points from this literature are that sleep deficits (ie, insomnia-type symptoms of longer sleep latency, night-time awakenings, difficulty staying asleep) both before and after a quit attempt may predict relapse in treatment-seeking smokers. Importantly, not all studies found these associations, suggesting that there may be subgroups of smokers (ie, those with higher levels of pre-treatment insomnia symptoms) who may be more vulnerable to the exacerbated sleep deficits following cessation. Cognitive-behavioral treatment for insomnia as an adjunctive treatment for smoking cessation may be a plausible approach to delaying relapse. The characterization or phenotype of treatment-seeking smokers most vulnerable to relapse because of sleep deficits, and the extent to which cognitive-behavioral therapy for insomnia increased days of abstinence in this population warrants consideration.

Effects of Pharmacotherapy on Sleep

Sleep disturbances are a recognized side effect of the FDA-approved treatments for nicotine dependence including nicotine replacement therapies (patch, spray, gum, lozenge), bupropion and varenicline. One placebo-controlled trial that utilized bupropion and varenicline treatment arms showed that these active treatments did not ameliorate withdrawal-related sleep disturbance, thus strategies to address sleep disturbances induced by smoking cessation pharmacologic treatments are needed to promote cessation.⁴² Characterizing the sleep disturbances presented by each of the pharmacologic treatments is therefore necessary to informing the design of adjunctive nicotine dependence treatments.

Nicotine Replacement Therapy

Nicotine replacement therapies (NRTs; transdermal patch, gum, spray, lozenge) provide partial nicotine replacement upon cessation of smoking and in doing so, ameliorate nicotine craving and pharmacologic withdrawal symptoms.⁴⁶ Up to 50% of treatment-seeking smokers using nicotine replacement therapies report sleep disturbances that start on the day of use.⁴⁷ Disturbed sleep, vivid dreams and daytime drowsiness are some of the more commonly reported side effects from using nicotine replacement therapies. In one study, 6.4% of participants reported disturbed sleep, 4.4% reported vivid dreams, and 1.5% reported daytime drowsiness while using NRT.⁴⁸ Meta-analytic data of 120 studies involving 177390 individuals, showed that the prevalence of insomnia among individuals using nicotine replacement therapy for smoking cessation was 11.4%.⁴⁹ High levels of pre-treatment nicotine dependence, continued cessation, and female gender were found to significantly predict sleep disturbances 4-weeks after quitting in a sample of 1392 treatment-seeking smokers.⁴⁷

Studies examining the trajectory of NRT sleep related side effects suggest that sleep disturbances among NRT users may take some time to subside. In one cohort study, instances of sleep disturbance (vivid dreams, other sleep disturbances) were still being reported by up to 50% of abstinent smokers after 12-weeks of treatment.⁴⁷ This is consistent with another study that showed no change in reports of sleep disturbance in the 21-days following cessation,³⁷ but inconsistent with data showing that use of transdermal nicotine actually ameliorates sleep disturbances following cessation compared to placebo.³⁵ Collectively, these studies reporting on NRT use and sleep in smokers suggest that up to one-in-ten treatment-seeking smokers can experience NRT-induced sleep disturbance following cessation that may last well into the quitting period (ie, up to 12 weeks).

Bupropion

Sustained release bupropion (bupropion SR) is an aminoketone anti-depressant that is hypothesized to promote smoking cessation and delay relapse⁵⁰ to smoking by inhibiting dopamine reuptake in the reward center of the brain. Compared to placebo, bupropion increases the relative risk of cessation by 1.62 (95% CI = 1.49–1.76).⁵¹

Between 4–21% of treatment-seeking smokers using bupropion SR report disturbed sleep including insomnia, abnormal dreams and daytime fatigue.⁵² Some studies show that sleep disturbances associated with bupropion are significantly higher than those found in placebo, and varenicline.⁵³ Conversely, other studies show no significant increases in sleep disturbances associated with bupropion treatment.⁵⁴ Although this evidence reporting on the increases of sleep disturbances following cessation using bupropion is mixed, that up to one-in-five bupropion users report an increase in sleep disturbances is clinically meaningful.

Varenicline

Varenicline is an a4B2 partial agonist medication indicated for the treatment of nicotine dependence. As a a4B2 partial agonist, varenicline stimulates sufficient dopamine to reduce craving while simultaneously acting as a partial antagonist by blocking reinforcement from smoked nicotine.⁵⁵ Double-blind, randomized trials show varenicline to outperform bupropion and placebo in producing higher quit rates. For example, Gonzales and Colleagues report that following a 12-week treatment period, varenicline quit rates were 50.3% as compared to 33.5% in the bupropion arm and 14.5% in the placebo arm.⁵⁶ Compared to placebo, meta-analytic data show bupropion to increase the odds of cessation by 1.84 and varenicline by 2.88,⁵⁷ thus, varenicline is considered the most effective FDA-approved treatment for nicotine dependence.

Listed side effects of Varenicline include insomnia, vivid or lucid dreams and other sleep disturbances such as difficulty staying asleep. McLure and Colleagues reported that 39–46% of treatment-seeking smokers using varenicline reported difficulty sleeping, while 56–68% reported a change in dreaming, and that these sleep disturbances were retained 21-days after cessation.⁵⁸ Meta-analysis of clinical trials that compared the efficacy of varenicline to placebo, show that disturbed sleep, specifically insomnia symptoms of difficulty falling and staying asleep, as well as the incidence of abnormal dreams were between 50% and 70% higher in varenicline recipients.^59,60 One study that prospectively evaluated changes in sleep insomnia and dreams among treatment-seeking smokers using varenicline (N = 38), showed that, based on daily sleep diaries over a 7-day period, participants retained excellent sleep efficiency (>90%) and that while overall sleep measures did not change significantly, an increased number of awakenings and reports of dreams was observed.⁶¹ Prospective studies suggest that insomnia-related symptoms peak in the first week of quitting and then progressively decline until pre-treatment levels are achieved at 2–12 weeks.⁶² Together, these studies reporting on the relationship between varenicline use and sleep disturbances show that while as many as seven-in-ten treatment-seeking smokers using varenicline report sleep symptoms, that the symptoms to dissipate across time.

Take-Home Points: Relationship between Sleep and Cessation Outcome

Poor sleep health as characterized by shorter sleep duration, difficulty falling asleep, difficulty staying asleep, early awakenings and night-time awakenings are more common in smokers than nonsmokers. Of particular relevance to smoking cessation efforts, sleep health deteriorates following cessation in many smokers, and this in turn is implicated in relapse. Importantly, FDA-approved treatments for nicotine dependence may also impede healthy sleep. Varenicline, the most effective smoking cessation treatment, in particular has insomnia symptoms and abnormal dreams as a notable side effect. These different lines of evidence converge to underscore sleep as an intervention target for treatment-seeking smokers, particularly for those using pharmacotherapy. Another question raised by this body of work is whether there are sub-groups of smokers (ie, those with higher nicotine dependence, those with poorer pre-cessation sleep health) who are particularly vulnerable to sleep deficits and poorer sleep health following cessation, and therefore might be a higher-priority for a sleep health intervention.

Possible Mechanisms Linking Poor Sleep to Smoking Cessation Outcomes

To further understand the possible relationship between sleep and smoking cessation, it is important to consider the different mechanisms through which sleep may impact smoking behavior and vice versa. Plausible mechanisms through which tobacco use and sleep interact include cognition, affective (ie, mood, depressive symptoms) and emotional (ie, emotional dysregulation) states.

Unhealthy sleep has been associated with cognitive deficits,⁶³ whereas cognitive impairment following cessation predicts relapse to former smoking habits.⁶⁴ For example, short (≤6 h) and long (≥9 h) sleep has predicted poorer cognitive function.⁶³ Even an extra 6 h of wakefulness can produce deficits in alertness and working memory.⁶⁵ Adverse changes in sleep (either substantial increases or decreases in sleep duration) have been associated with compromised cognitive function.⁶⁶

Corroborating these data are a large literature on neurocognitive effects of experimentally-induced short sleep. Overall, sleep loss leads to impairments in vigilance and sustained attention,⁶⁷ as well as executive function and decision making,⁶⁸ which could plausibly lead to unhealthy decision making. For example, Greer and colleagues⁶⁹ showed that sleep loss led to worse food-related decision making. However, studies specifically linking sleep loss due to smoking and decision making around smoking have not yet been conducted.

Disruption in cognitive processing is a common nicotine abstinence symptom in clinical studies,²⁹ with up to one-half of abstinent smokers reporting difficulty concentrating.⁷⁰ Compared to a smoking state, abstinent smokers experience specific deficits in sustained attention⁷¹ working memory^72,73 and executive function.⁶⁴ In turn, nicotine use can ameliorate these deficits.⁷⁴ Importantly, attention and concentration deficits following a quit attempt increase risk of smoking relapse in clinical studies.^75–77 Thus, cognitive deficits and disturbed sleep are both abstinence symptoms in habitual smokers that may interact to increase the likelihood of relapse.

Similar to cognition, there are data to suggest that depressive symptoms and emotional dysregulation are associated with continued smoking^78,79 and habitually poor sleep.^80–82 As reviewed thus-far, unhealthy sleep, including insomnia symptoms (difficulty getting to and staying asleep) is prevalent in smokers,^12–16 and exacerbated sleep deficits following cessation are common.³⁸ Unhealthy sleep is highly prevalent among adults with depressive disorders. Even among non-depressed adults, poor sleep quality precedes a subsequent increase in depressive symptoms and negative mood.⁸⁰ As a group, smokers are more likely to suffer from depressive and mood disorders than nonsmokers. This complex interplay between sleep, smoking, and depressive symptomology is likely agitated upon smoking cessation when abstinence from nicotine leads to increases in negative mood and insomnia symptoms,⁸⁰ both of which have been shown to relate to relapse among treatment-seeking smokers.^38,83 The temporal sequence of changes in depressive symptoms and sleep habits following cessation has yet to be fully understood, but such information would inform upstream intervention targets for smoking behavior.

Likewise, emotional dysregulation, or the ability to regulate emotions and control behavioral responses, has been implicated as a mechanism for how sleep may relate to smoking cognitions and quitting outcomes. From the outset, poor sleep quality has been highly correlated with emotion dysregulation in smokers.⁷⁹ Recent data has extended this work to show that emotional dysregulation mediates the relationship between insomnia symptoms and several smoking cognition variables including, negative reinforcement smoking outcome expectancies, negative reinforcement smoking motives and negative reinforcement expectancies from smoking abstinence. Importantly these associations were adjusted for other demographic and smoking behavior variables.⁸¹ Converging with this study are data from Filio and colleagues, showing that in a sample of 128 treatment-seeking smokers, higher levels of emotion dysregulation was associated with lower levels of self-efficacy for remaining abstinent, more quit-related problems, and a lower likelihood of having had a quit attempt of 24 h or greater.⁷⁹ This small body of work converges to suggest that emotional regulation may be an important mechanism linking sleep with cigarette smoking behaviors and quitting.

Cognitive, affective, and emotional states present plausible pathways through which sleep and tobacco use may interact. This area of work is severely under-developed, and longitudinal studies are needed to quantify the association, and the temporal relationships, between these variables across time. Testing the extent to which improving sleep ameliorates deficits in cognitive, affective, and emotional states in smokers across the smoking cessation process will help determine if sleep improvement is a viable adjunctive therapy for smoking cessation.

Plausible Adjunctive Sleep Therapies to Promote Smoking Cessation

Overview

Sleep patterns characteristic of smokers congeal around insomnia-type symptoms including difficulty falling asleep (long sleep latency) and difficulty staying asleep (short sleep duration, frequent awakenings and arousal during the night), that are amplified following cessation.³² Some studies suggest that increases in disturbed sleep following cessation is attributed to the use of pharmacotherapy, whereas others suggest that disturbed sleep following cessation is attributable to nicotine-withdrawal.⁴² In both scenarios, disturbed sleep before,^39,41,42 and after cessation determines relapse, and as such, warrants treatment as part of the cessation process. There are a range of behavioral and pharmacological treatments for insomnia-type symptoms that may be suitable for use in conjunction with standard nicotine dependence treatment (counseling + pharmacotherapy); an overview is provided here.

Behavioral Treatments

Cognitive-Behavioral therapy for insomnia (CBT-I) is a first-line treatment for chronic insomnia⁸⁴ that improves sleep outcomes for up to two years after treatment⁸⁵ and is preferred by patients with a clinical diagnosis of insomnia to drug therapy.⁸⁶ CBT-I is comprised of two core components (stimulus control and sleep restriction therapy), as well as several optional components including cognitive therapy, sleep hygiene, and relaxation.⁸⁷ Stimulus control techniques work to strengthen the association between the bed and bedroom with sleep, and to establish a consistent sleep schedule. Sleep restriction therapy is a specific approach that addresses the mismatch between sleep ability and sleep opportunity by reducing sleep opportunity to match ability and then slowly upwardly titrating sleep opportunity as long as the individual is able to maintain high sleep efficiency. Cognitive therapy seeks to identify and replace dysfunctional beliefs and attitudes about sleep and insomnia. Sleep hygiene works to address environmental factors, physiologic factors, and behavioral components (ie, regular sleep scheduling, limiting alcohol intake). Relaxation training seeks to address the high levels of physiologic, cognitive, and/or emotional arousal, both at night and during the daytime, which is exhibited by individuals who have difficulty falling and/or staying asleep.^84,88 Deep breathing, progressive relaxation, and meditation are relaxation techniques that haven been shown to lower pre-sleep arousal (eg, racing thoughts) and improve sleep metrics.⁸⁹ In a recent meta-analysis of 20 studies that examined the efficacy of CBT-I among patients with chronic insomnia, Sleep onset latency, wake after sleep onset, total sleep time, and sleep efficiency, were all significantly improved by multi-modal CBT-I.⁸⁹ This is in the context of several other meta-analyses and systematic reviews of CBT-I showing that not only is it superior to placebo⁸⁸ and equivalent or superior to pharmacotherapy for insomnia,⁹⁰ but it is effective even in the presence of comorbid conditions such as depression and chronic pain.⁹¹

To date, only one study has examined the effects of a CBT-I intervention on smoking cessation outcomes.⁹² Nineteen treatment seeing smokers were randomized to receive eight sessions of CBT-I, transdermal nicotine patch and smoking cessation counseling (N = 9) versus transdermal nicotine patch and smoking cessation counseling (N = 10) alone. While the results showed no difference in smoking cessation rates between the groups, participants receiving the CBT-I had a longer time to relapse.⁹² A fully-powered examination of the effects of CBT-I on smoking cessation outcomes is warranted.

Pharmacological Treatments

Benzodiazepines are a pharmacologic first-line treatment for insomnia. Currently there are five FDA-approved benzodiazepines for this indication: estazolam, flurazepam, quazepam, temazepam, and triazolam.⁹³ These medications act by increasing the activity of the inhibitory neurotransmitter GABA to inspire drowsiness or sedation. Consistent with this mechanism, sleep latency (time to sleep) and wake after sleep onset, are both significantly reduced while sleep duration and sleep quality are significantly increased using these therapeutics in the short term. However with increased tolerance of these pharmaceutics, sleep improvements may be curtailed.⁹⁴ Of particular relevance to smokers, cigarette smoke contains beta carbolines that block the actions of benzodiazepines at the GABA-A receptors,⁹⁵ thus higher doses may be needed in smokers versus nonsmokers to observe comparable effects. Benzodiazepines are not contraindicated with any of the FDA-approved treatments for nicotine dependence and their role in promoting smoking cessation through improved sleep has yet to be evaluated.

Melatonin is a hormone normally secreted from the pineal gland at night that serves as the signal of darkness in the organism and as such plays a pivotal role in the physiological regulation of circadian rhythms, including sleep. Several melatonin receptor agonists have recently become available for treatment of sleep disorders: ramelteon for the treatment of insomnia characterized by difficulty with sleep onset, prolonged-release melatonin for treatment of primary insomnia characterized by poor quality of sleep in patients who are aged 55 or over, agomelatine for the treatment of depression and associated sleep disorder, and tasimelteon for the treatment of non-24 h sleep-wake disorder in the blind.⁹⁶ Given that longer sleep latency (difficulty falling asleep) is a characteristic of smokers (versus nonsmokers) that is exacerbated following cessation, the reported reductions in sleep latency in Ramelteon users⁹⁷ may be particularly beneficial to curbing sleep deficits following smoking cessation. These melatonin receptor agonists are not contraindicated with the FDA smoking cessation medications and their efficacy as adjunctive smoking cessation treatments warrants investigation.

Directions for Future Research

On the basis of this narrative review, it could be argued that sleep is an understudied and underutilized intervention target for promoting smoking cessation and preventing relapse in treatment-seeking smokers. As demonstrated, sleep deficits in terms of shorter sleep duration and insomnia symptoms (difficulty getting to sleep and staying asleep) are a sleep phenotype of smokers that may become exacerbated following cessation, both as an abstinence symptom, and, as a side effect of quit-smoking medications. There are several directions for future work that are needed to quantify the prospective relationship between sleep and tobacco use and advance our understanding for how improved sleep may promote smoking abstinence.

First, the temporal relationship between smoking and sleep needs further consideration. As discussed in this review, sleep may be disrupted because of the physiological effects of nicotine and nicotine withdrawal upon abstinence. Conversely, smokers may use their smoking habit to counter the effects of daytime sleepiness because of poor sleep. Prospective, observational studies examining the temporal relationship underpinning this complex interplay between sleep and smoking are needed. Meta-analytic studies to quantify the relationship between sleep deficits with smoking behaviors and cessation outcomes would also be valuable.

Second, laboratory and clinical studies to examine the effects of sleep pharmacologics on tobacco consumption in a natural setting and as an adjunctive treatment for smoking cessation are needed. Such studies could also be used to inform the mediating role of cognitive, affective, and nicotine dependence factors on the relationship between tobacco use and sleep.

Third, behavioral therapies for insomnia targeting treatment-seeking smokers need to be developed and evaluated in the context of a smoking cessation interventions. For example, physical activity and mindfulness based approaches have independently been shown to reduce insomnia symptoms⁹⁸ and promote smoking abstinence,^99,100 thus, these approaches may have promise in treating smokers with higher levels of sleep deficits.

It is to be hoped that the culmination of evidence to elucidate the relationship between sleep, tobacco use, and cessation outcomes will ultimately inform a sleep phenotype of risk for continued smoking. This knowledge base will in turn inform targeted intervention approaches to promote cessation outcomes in smokers most vulnerable to sleep deficits.

Funding

Research reported in this publication was supported by an Institutional Development Award (IDeA) Center of Biomedical Research Excellence from the National Institute of General Medical Sciences of the National Institutes of Health under grant number P20GM113125 and by the University of Delaware Research Foundation grant number 16A01366.

Declaration of Interests

Dr. Patterson receives medication free of charge from Pfizer. Dr. Rizzo has been compensated as a consultant and/or a promotional speaker for Astra-Zeneca, Novartis, and Genentech. Dr. Malone, Dr. Grandner, and Dr. Edwards have no conflicts of interest.