Abstract
Introduction:
Mindful attention to breath (MAB) is a common component of mindfulness-based interventions for smoking. Other types of breathing strategies including taking deep breaths, diaphragmatic breathing, and yogic breathing have also been advised for smoking cessation. Surprisingly, there has been little research directly investigating the efficacy of breathing techniques on smoking outcomes. The aim of this laboratory study was to test the effects of MAB with and without a controlled yogic breathing component on smoking behavior.
Methods:
Twelve-hour abstinent daily smokers (N = 33; 39% female; M = 9.61 cigarettes per day; 67% African American) were randomly assigned to a 20-min intervention consisting of (1) MAB, (2) mindful Ujjayi yogic breathing (MYB), or (3) no treatment (NT).
Results:
Compared with NT, MAB (Hazard ratio [HR] = 0.187, p = 0.023) and MYB (HR = 0.266, p = 0.048) both reduced the risk of smoking during a 50-min smoking versus money choice task. Furthermore, the MAB group reported significantly greater postmanipulation serenity (i.e., relaxed, calm, and at ease) than MYB and NT. No differences among the conditions were observed for ratings of smoking craving, affect, or withdrawal symptoms. Qualitative data indicated high acceptability of the MAB exercise with participants reporting uniformly positive descriptors of MAB in terms of satisfaction (e.g., “relaxing”), utility (e.g., “very helpful”), and ease of use (“not at all difficult”).
Discussion:
Findings suggest MAB is an easily disseminable and cost-effective strategy that may promote smoking abstinence and should be further explored.
Introduction
Smoking causes high rates of premature death and disability in the United States and worldwide. 1 Although most smokers want to quit and many try, even the best available treatments result in high relapse rates. 2 Improved treatments for quitting smoking are needed, including strategies that focus on increasing the awareness and acceptance of the aversive effects of abstinence and are feasible to use in everyday life. 3 Evidence is growing for the efficacy and effectiveness of mindful-based strategies for smoking cessation.4–7
Mindfulness practices aim to develop a nonjudgmental attitude and nonreactive response to thoughts, feelings, and sensations as they arise.8–10 Training in mindfulness is believed to facilitate smoking cessation by teaching individuals to be in the moment and aware and accepting of the discomfort associated with nicotine abstinence (e.g., craving, negative affect). These practices increase one's tolerance for negative internal states and break the cycle of reacting to such states by smoking.11–14
Various mindfulness-based treatments for smoking cessation have been developed in recent years (e.g., Mindfulness Based Addiction Treatment), 14 or integrated into other treatment strategies (e.g., Mindfulness-Based Relapse Prevention 15 or Acceptance and Commitment Therapy). 16 Reviews and meta-analyses have generally concluded that such multicomponent mindfulness treatments are as effective or more effective than other active (e.g., Freedom from Smoking) 17 or sham treatments.18,19
However, there has been limited research examining the efficacy and utility of individual treatment components and treatment mechanisms associated with good outcomes. 4 Although different mindfulness-based treatments have varying treatment components, mindful meditation appears to be a core component. 20 Mindful meditation has been described as “the awareness that arises from paying attention, on purpose, in the present moment and nonjudgmentally.” 21 Mindful attention to breath (MAB) is a strategy believed to help anchor one's attention to the present moment experience by observing and accepting the experience of breathing, specifically attending to the natural inhalations and exhalations of the breath. In controlled studies, MAB improves various aspects of mood22,23 and cognitive functioning. 24
MAB may be a particularly effective and efficient strategy to increase one's ability to resist smoking.13,25 MAB has been hypothesized to increase self-regulation through various mechanisms including greater body awareness and attentional control.26,27 A recent randomized clinical trial by Spears et al., that tested a text messaging program as an adjunct to Mindfulness Based Addiction Treatment found that among five mindfulness practices, participants reported using MAB most often during daily life. 13
Moreover, MAB was uniquely associated with better smoking outcomes at the end of treatment and at the 1-month follow-up. To the authors' knowledge, there are no prior controlled studies examining the isolated effects of MAB on smoking or other addiction-related outcomes. Prior laboratory-based studies have demonstrated that listening to a 10-min body scan audio, including attention to breath and other parts of the body, resulted in greater decreases in desire to smoke, irritability, tension, and restlessness, as well as greater decreases in smoking craving and withdrawal compared with a control audio.28,29
Strategies involving an active change in breathing style have also been commonly suggested as methods to cope with smoking abstinence. For example, taking deep breaths is a widely recommended coping strategy for smoking.30,31 Moreover, deep breathing has been reported as one of the most useful coping strategies among individuals attempting smoking abstinence. 32 Various types of yoga practices (e.g., pranayama) also include a controlled breathing component and prior research has demonstrated beneficial effects of yoga for smoking cessation, although with varying rigor across studies.33–35
Ujjayi pranayama breathing is associated with increased calmness, both physically and mentally.36,37 There is evidence for the therapeutic effects of yogic breathing on other health outcomes including but not limited to stress, anxiety, depression, and cardiovascular disease.36,38–40
Only a few prior studies have directly examined controlled breathing techniques on smoking-related outcomes. McClernon et al. found that controlled deep breathing in the laboratory over a period of 4 h reduced withdrawal symptoms, craving, and negative affect compared with a control condition. 41 Shahab et al. demonstrated that 10 min of yogic breathing reduced craving, but not other withdrawal symptoms compared with a 10-min video control. 42 Finally, Lotfalian et al. compared a 20-min exercise involving mindful Ujjayi yogic breathing (MYB) with an active cognitive strategy (i.e., thinking about the negative effects of smoking) and no treatment. 25
MYB and the cognitive strategy both decreased smoking craving and withdrawal relative to no treatment. However, only MYB decreased negative affect and reduced the risk of smoking during a smoking choice procedure (SCP) relative to the other conditions.
Given that components of MAB and mindful yogic breathing have been shown to decrease smoking craving, withdrawal, and reduce the risk of smoking, this study examines the effects of MAB with and without a deep yogic breathing component on subjective states and smoking behavior. It was hypothesized that both MYB and MAB would have beneficial effects on mood, craving, and withdrawal relative to no treatment (NT) among abstaining smokers. It was also hypothesized that MYB and MAB would lower the risk of smoking during a 50-min SCP versus money choice procedure.
Methods
Participants
Adult regular cigarette smokers were recruited through flyers in the Washington D.C. metro area and web-based advertisements. Participants were screened by phone (N = 111) and eligibility was determined as follows: (1) 18 years of age or older, (2) able to read and write in English, (3) smoke at least five cigarettes per day for at least 1 year, (4) not currently seeking smoking cessation treatment or using smoking pharmacotherapy, (5) not pregnant or planning to become pregnant, (6) absence of smoking-related health problems (e.g., asthma), and (7) no prior participation in related studies in this laboratory.
Seventy-eight individuals were ineligible because of insufficient smoking rate or duration (n = 49), health problems (n = 13), use of smoking pharmacotherapy (n = 3), or participation in another related study in this laboratory (n = 9). Due to COVID-19, 43 data collection ceased in March 2020. There were 33 study completers (13 women and 20 men) with a mean age of 43.12 years (standard deviation [SD] = 12.69).
The sample included individuals who identified as Black/African American (67%), White (24%), Asian (3%), and American Indian or Alaska Native (6%) with 6% identifying as Hispanic or Latino/a/x. Participants smoked a mean of 9.61 (SD = 4.51) cigarettes per day for 17.94 years (SD = 11.21) with 64% preferring menthol cigarettes. This study was approved by the institutional review board at American University.
Procedure
After phone screening, eligible participants were required to abstain from smoking for 12 h before the scheduled session and to bring their preferred brand of cigarettes. Upon arrival, participants were taken to an experimental room and informed consent was obtained. Researchers collected a breath sample for carbon monoxide (CO) testing (cutoff <10 parts per million [ppm]) (M = 3.97 ppm, SD = 2.10) and recorded the reported time of last cigarette (M = 16.76 h, SD = 5.09). Participants completed baseline measures before random assignment (based on a random number generator) to one of three experimental manipulations: (1) MYB (n = 12), (2) MAB (n = 11), or (3) NT (n = 10).
Participants were run by four research assistants who consisted of a graduate student, two senior undergraduate students, and a postbaccalaureate student. All were supervised by a doctoral-level clinical psychologist. The MYB condition was taken directly from Lotfalian et al. 25 The researcher instructed the participant to sit in a relaxed position with eyes closed. Participants were told to concentrate on their natural inhalations and exhalations of breath without changing anything.
Then they were guided through an Ujjayi pranayama breathing exercise, a slow diaphragmatic breath that involves gentle contraction of the throat muscles to create resistance. Participants were also given instructions to observe, be present, and accept the experience without judgment, and, if their mind wandered, to gently guide their attention back to breathing.
In the MAB condition, the researcher instructed the participant to sit in a relaxed position with eyes closed. The breathing script was modified from both Wherever You Go There You Are: Mindfulness Meditation in Everyday Life 9 and Lotfalian et al. 25 Participants were instructed to concentrate on their natural inhalations and exhalations of breath without changing anything. They were told to bring their full attention to breath and the experience of breathing (e.g., “notice each in breath as it enters your nostrils”). They were told to breathe as they normally would and not strive to change anything about their breathing. They were also given instructions to observe, be present, and accept the experience without judgment, and, if their mind wandered, to gently guide their attention back to breathing.
In both the MYB and MAB conditions, the researcher spent the first 5 min instructing the participant and then 5 min practicing the exercises with the participants. Then participants performed the breathing exercises for 10 min alone. A soft tone sounded every 30 sec as a gentle reminder for participants to focus their attention on the breathing exercises. 25 In the NT control condition, participants sat quietly for 20 min. After the experimental manipulation, participants completed measures of smoking craving, affect, and state mindfulness. Subsequently participants completed a smoking choice laboratory task to assess their willingness to abstain from smoking. 44
This validated laboratory procedure is intended to model real-world smoking behavior. 44 The cigarette pack, a lighter, and an ashtray were placed next to the participant. Participants were told that they could choose to smoke or earn $0.50 for each 5-min period they chose to delay smoking up to 50 min. 44 A different tone sounded every 5 min to signal participants to make a choice to smoke a cigarette or earn $0.50. Latency to smoke was recorded. Participants completed measures immediately after they smoked and were then fully debriefed and discharged. See Supplementary Data for the intervention protocol.
Measures
Smoking abstinence
Smoking abstinence was verified through self-report and biochemically using a Bedfont Micro III Smokerlyzer 45 to capture expired CO levels in ppm.
Demographics and smoking history questionnaire
This measure assessed demographic variables and smoking history along with the contemplation ladder, 46 a single item used to identify intention and assess readiness to quit smoking and the Fagerstro¨m Test for Nicotine Dependence (FTND), a commonly used 6-item measure of nicotine dependence. 47
Five-factor mindfulness questionnaire
The five-factor mindfulness questionnaire (FFMQ) is a widely used measure of trait mindfulness. 48 Participants rated 39 statements from 1 (never or rarely true) to 5 (very often or always true). The FFMQ yields five subscales: (1) Observing (Cronbach's α = 0.746), (2) Describing (Cronbach's α = 0.865), (3) Acting with Awareness (Cronbach's α = 0.915), (4) Nonjudging (Cronbach's α = 0.857), and (5) Nonreactivity (Cronbach's α = 0.778), and a total score (Cronbach's α = 0.898).
State mindfulness scale
Participants rated 21 items assessing state mindfulness on a scale from 1 (not at all) to 5 (very well). 49 A total score was computed (Cronbach's α = 0.927).
Positive and negative affect schedule
The positive and negative affect schedule (PANAS) consists of 10 positive affect and 10 negative affect items that are rated on a scale from 1 (very slightly or not at all) to 5 (extremely). 50 Scores were calculated for positive affect (Cronbach's α = 0.897) and negative affect (Cronbach's α = 0.724). Three additional items (relaxed, calm, at ease) that comprise the serenity factor of the PANAS-X (Cronbach's α = 0.918) were also administered based on prior findings. 51
Urge rating scale
Participants rated their cravings, wants, and desires to smoke from 0 (none at all) to 10 (greatest ever experienced). 52 A mean score was calculated (mean Cronbach's α = 0.981).
Minnesota Tobacco Withdrawal Scale-Modified
A modified version of the Minnesota Tobacco Withdrawal Scale was used to assess withdrawal symptoms. 53 The following five items were administered: irritability/frustration/anger, anxiety, difficulty concentrating, restlessness, and depressed/sad mood. Items were rated on a 5-point scale ranging from 0 (none) to 4 (severe) and asked participants to report on the present moment. A mean score was calculated (Cronbach's α = 0.910).
Treatment acceptability/feasibility
This measure was developed by this laboratory to assess participants' reactions to the experimental conditions. Participants were asked to provide open-ended responses to the following questions: (1) how useful did you find the technique you learned today, (2) how satisfied are you with the technique, and (3) how difficult did you find the technique?
Analytic plan
All analyses were conducted using SPSS version 27. There were no baseline differences among the groups (see Table 1) on any demographic variables or smoking history variables, or trait mindfulness. One-way analyses of covariance were conducted for smoking urge, affect factors, and withdrawal with the postmanipulation score entered as the dependent variable and the premanipulation score on that variable and baseline nicotine dependence (FTND score) entered as covariates. Significant omnibus tests were followed up with simple comparisons. A Cox proportional hazards model was generated to examine the effects of the treatment condition on smoking choice during the 50-min SCP controlling for baseline nicotine dependence.
Baseline Demographic Characteristics and Smoking Variables
Note: There were no baseline differences among the three groups, all p's < 0.05.
FFMQ, five-factor mindfulness questionnaire; FTND, Fagerstro¨m Test for Nicotine Dependence; MAB, mindful attention to breath; MYB, MAB combined with controlled yogic breathing; NT, no treatment; SD, standard deviation.
Results
Self-reported mood, craving, and withdrawal measures
As given in Table 2, there was a significant difference among the conditions on the serenity factor, F(2, 28) = 4.93, p = 0.015, η 2 = 0.260. Follow-up comparisons revealed that those in the MAB condition reported greater postmanipulation serenity than those in the MYB condition, F(1, 19) = 8.39, p = 0.009, η 2 = 0.306, and those in the NT condition, F(1, 17) = 5.10, p = 0.037, η 2 = 0.231. There was no difference in serenity between the MYB condition and NT condition. There were no significant differences for affect, withdrawal, or craving.
Dependent Variables Pre- and Postintervention
MTWS-M, Minnesota Tobacco Withdrawal Scale-Modified; PANAS, positive and negative affect schedule; URS, urge rating scale.
Smoking behavior
Fourteen of 33 participants (42.4%) chose to smoke during the SCP. More than twice as many participants in the MAB (72.7%) and MYB (66.7%) groups abstained for the full 50-min money versus SCP than in the NT condition (30%). Among those who smoked, the average delay was 7.58 min, (SD = 10.57), smoking duration was 6.31 min (SD = 3.02), and participants took a mean of 14.29 puffs (SD = 6.04).
As shown in Figure 1, the Cox proportional hazards model revealed an estimated risk of smoking for those in the MAB condition significantly less than the NT condition (Wald = 5.19; Hazard's ratio [HR] = 0.187; p = 0.023; confidence interval [CI] = 0.044–7.92). The MYB group also had a significant reduction of risk of smoking relative to the NT group (Wald = 3.92; HR = 0.266; p = 0.048; CI = 0.072–0.987). The MAB and MYB conditions did not differ.

Estimated survival function for smoking latency during the choice procedure.
Treatment acceptability/feasibility
Qualitative data indicated high acceptability of the MAB exercise with participants reporting uniformly positive descriptors of MAB in terms of satisfaction (e.g., “relaxing”), utility (e.g., “very helpful”), and ease of use (“not at all difficult”). The full set of responses are given in the Supplementary Data.
Discussion
This controlled preliminary laboratory study examined the effects of MAB with and without a controlled yogic breathing component on smoking behavior and subjective effects. In this small sample, both MAB and MYB reduced smoking risk compared with NT. The findings partially replicate Lotfalian et al. who tested the same MYB intervention compared with a cognitive intervention and NT control. 25 They found that the risk of smoking in the NT group was 2.89 times that of the MYB condition. In this study, the risk of smoking in the NT group was 3.75 times greater than the MYB intervention.
However, these findings should be interpreted with caution due to the small sample size and shorter postintervention observation period. Furthermore, this study extended Lotfalian et al. by testing the effects of MAB without a change in breath. Risk of smoking in the NT group was more than five times that of the MAB group. 25 The findings, although preliminary, suggest that MAB with or without a controlled breathing component facilitates the ability to refrain from smoking during a subsequent short observation period.
As far as subjective effects, the only difference observed among the conditions was in self-reported ratings of serenity. MAB produced significantly greater feelings in serenity than NT, but MYB did not in this sample. An evaluation of pre- to postmeans reveals that serenity increased in the MAB group, slightly decreased in the MYB group, and did not change in the NT group. It is possible that MYB intervention was not as easily learned and executed among participants resulting in decreased serenity. Future research should replicate these findings and more closely evaluate the acceptability, ease of learning, and subjective effects of different breathing techniques.
There were no effects of the breathing manipulations on the positive or negative affect factors of the PANAS, tobacco withdrawal symptoms (e.g., irritability), or craving. This is somewhat inconsistent with Lotfalian et al. who found that MYB produced greater negative affect and craving relief than NT. 25 Another controlled study found that yogic breathing produced greater craving relief than a control condition but found no differences in affect. 42 McClernon et al. reported that controlled deep breathing over a period of 4 h resulted in significant reduction in negative affect and craving compared with a control condition. 41
Large differences in methodologies and measures used across the studies make it difficult to reconcile these inconsistencies. Furthermore, the small sample in this study limited power to detect smaller effects. It is important to note that mindfulness approaches in general encourage an awareness and acceptance of a current mood state rather than any direct attempts to change it. Mindfulness approaches may prevent unhealthy reactions to mood states even in the absence of a change in mood state. Nevertheless, a reduction in stress reactivity has been observed in some prior research examining mindfulness approaches for smoking. 54
This study has various limitations. Due to COVID-19, data collection had to be stopped prematurely. This resulted in a smaller sample size than initially planned. Although moderate effects on the SCP and replicated findings were observed from a similar prior study testing MYB, the results need to be interpreted with caution and further replicated. 25 Furthermore, the MAB and MYB breathing interventions were a total of 20 min and only immediate outcomes were assessed. Future research should examine the dose and timing of such interventions in a clinical setting that will result in the highest rates of compliance, patient acceptability, and maximal long-term benefits.
The 12-h abstinence period is a limitation to the study. Further studies should examine abstinence periods longer than 12 h. The postintervention observation period should be extended in future studies. In addition, future research could also examine whether breathing techniques readily transfer to a virtual environment such as texting, web-based information, and telehealth. 13 Future research could also examine physiological reactions and other mechanisms underlying the beneficial effects of MAB. The control condition involved sitting in a room quietly for 20 min.
Scores on the subjective measures do not suggest that this experience was aversive. Nevertheless, future studies could consider the inclusion of an active control condition. This sample consisted mostly of African American smokers who generally have lower odds of cessation success and represent a group in need of effective treatment strategies to reduce smoking-related health disparities.
In conclusion, this controlled laboratory investigation found that among this small diverse sample, those who were taught to engage in MAB reported high treatment acceptability and demonstrated lower risk of smoking during the observation period than the control group. These findings should be further explored as breathing strategies are feasible to perform at a moments' notice, brief, easily disseminable, cost-effective, and may be particularly useful for underserved and low-income populations.
Footnotes
Acknowledgments
These data formed the basis of the first author's master's thesis leading to the degree of master of arts in psychology at American University in Washington D.C. The authors wish to thank Rebecca Goyette, Melissa Baney, and Olivia Richardson for their assistance with data collection and Dr. Tommy Gunawan for his assistance with study design and statistical analyses.
Authors' Contributions
J.C.H. and L.M.J. designed the study and wrote the protocol. J.C.H. collected the data. J.C.H. and L.M.J. conducted the statistical analysis and wrote the article. J.C.H. and L.M.J. have read and approved the final article.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon request.
Author Disclosure Statement
The authors have no real or potential conflicts of interest to disclose.
Funding Information
Funding for this study was provided by a Graduate Student Research Award provided by the College of Arts and Sciences at American University. This funding source had no other role other than financial support for participant payments.
Abbreviations Used
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
