Abstract
Mind–body interventions can improve vulnerabilities that underlie smoking behavior. The characteristics of smokers who use mind–body medicine have not been explored, preventing the development of targeted interventions. Patients (N = 593) presenting to a mind–body medicine clinic completed self-report measures. Patients were 67 percent never smokers, 27 percent former smokers, and 6 percent current smokers. Current smokers were younger; more likely to be single, unemployed, or on disability; and report greater depression symptoms, greater pain, and lower social support (ps < .05).Current smokers who use mind–body medicine have unique psychosocial needs that should be targeted in mind–body smoking cessation interventions.
Introduction
Cigarette smoking remains the leading cause of preventable death and disease in the United States, accounting for more than 480,000 deaths each year and US$300 billion a year in smoking-related healthcare costs (Centers for Disease Control and Prevention, 2015a; US Department of Health and Human Services, 2014; Xu et al., 2015). Given the negative health effects, many smokers are motivated to quit smoking. Out of the 40 million current adult smokers in the United States, nearly 70 percent report a desire to quit in any given year, and of those, approximately 40 percent make a quit attempt each year (Centers for Disease Control and Prevention, 2013, 2015a, 2015b; US Department of Health and Human Services, 2014). Unfortunately, despite multiple evidence-based treatment options, smoking cessation rates remain low and many smokers do not make use of treatment services (US Department of Health and Human Services, 2014). If the current standards of treatment are maintained, the projected prevalence of smoking among adults could be as high as 15%in 2050 (US Department of Health and Human Services, 2014).
There are known demographic trends in smoking behavior. Smoking rates are higher among males compared to females (18.8% vs 14.8%, respectively), and among individuals with a disability or limitation (21.9%), who are uninsured (27.9%), have less than high school education (22.9%), and live below the poverty line (26.3%) (Centers for Disease Control and Prevention, 2015a). Smokers, particularly those from lower socioeconomic status (SES) backgrounds, are also more likely to live in smoking households and to have stressful living and work conditions (Hiscock et al., 2012; Twyman et al., 2014). These patterns influence smoking cessation outcomes.
Despite having higher prevalence rates, smokers from lower SES backgrounds exhibit lower quit rates, have less access to quit resources, and are less likely to participate and adhere to smoking cessation interventions compared to smokers from higher SES backgrounds (Bock et al., 2014; Hiscock et al., 2012, 2015). Furthermore, they are less likely to receive provider referral or support for tobacco treatment (Twyman et al., 2014). Understanding demographic patterns of smokers is therefore important for developing treatment outreach efforts that are targeted toward relevant groups.
Psychological factors play a key role in smoking behaviors. Smokers report greater levels of stress and emotional problems than non-smokers (US Department of Health and Human Services, 2014). Nicotine dependence is associated with multiple psychological problems including anxiety disorders, major depression, stress, and pain, and reducing negative emotions is a primary reason why individuals smoke (Goodwin et al., 2008; Kassel et al., 2003; Piper et al., 2004; Zvolensky et al., 2009, 2010). The tendency to smoke to reduce negative emotions becomes particularly problematic during a quit attempt, given that increased negative affect is a common symptom of nicotine withdrawal (Hughes et al., 1990). Thus, many smokers are inherently vulnerable to smoking relapse because, during a quit attempt, they are confronted with increases in the exact symptoms they typically manage by smoking.
Social support also impacts smoking outcomes. Social support refers to an individual’s ability to seek and receive help from others across multiple domains (e.g. instrumental support with everyday tasks, emotional support during periods of distress) and is an important factor in quitting smoking (Chandola et al., 2004; Fiore et al., 2008; Lancaster and Stead, 2005; Sherbourne and Stewart, 1991). Smokers who perceive greater social support are more likely to initiate a quit attempt and maintain abstinence for at least 3 months, while those who perceive low social support are unlikely to successfully quit (Mermelstein et al., 1986) and maintain being quit (Park et al., 2009).
Mind–body medicine offers several techniques that can reduce smoking-related psychosocial problems. Mind–body medicine, which includes meditation practices, yoga, and tai chi, focuses on the interactions between emotional, psychological, spiritual, behavioral, and social factors to promote health and well-being (National Center for Complementary and Integrative Health, 2011; Park et al., 2013). These interventions counter the negative physiological effects of stress by eliciting the relaxation response, which is the physiological state responsible for rest and restoration (i.e. parasympathetic dominance), opposite to the “fight-or-flight” stress response (i.e. sympathetic dominance). In addition, mind–body interventions reduce psychological stress, anxiety, depression, and pain (Astin, 2004; Hofmann et al., 2010; Jacobs, 2001; Luberto et al., 2013; Miller et al., 2015; Park et al., 2013; Sylvia et al., 2015), and many mind–body interventions are practiced in a group format, which can improve social support (Barrera et al., 2002; Cattan et al., 2005; Samuelson et al., 2010; Vranceanu et al., 2014). Recent systematic reviews also concluded that yoga and meditation-based interventions, and mindfulness meditation interventions specifically (Oikonomou et al., 2017), improve smoking cessation rates (Carim-Todd et al., 2013). Indeed, mindfulness training is associated with significantly higher long-term abstinence rates (25%) as compared to standard care tobacco treatments (14%; Oikonomou et al., 2017). Improvements in emotional functioning are thought to be a key mechanism of these effects (Abrantes et al., 2008; Adams et al., 2014; Schuman-Olivier et al., 2014; Vidrine et al., 2009).
Taken together, demographic and psychosocial factors play a key role in smoking behavior, and mind–body interventions can improve many of these core vulnerabilities. However, the characteristics of smokers who seek mind–body services remain unclear, which limits the ability to develop attractive outreach efforts and prevents the development of mind–body interventions targeted specifically to their psychosocial needs. Indeed, studies of mind–body interventions rarely report on the prevalence of smokers seeking mind–body medicine services, or report on smokers’ characteristics and whether or not they differ from non-smokers. Smokers who seek mind–body services cannot be assumed to be similar to the general population of smokers because individuals who use mind–body interventions differ from those who do not (Bishop et al., 2007; Clarke et al., 2015; Honda and Jacobson, 2005). Moreover, individuals who utilize mind–body medicine tend to be of higher income and education levels (Berz et al., 2015; Bishop and Lewith, 2010), suggesting that smokers might be less likely to access mind–body medicine services. We recently developed the Stress Management and Resiliency Training–Relaxation Response Resiliency Program (SMART-3RP) (Park et al., 2013), a multi-modal mind–body intervention that significantly reduces physical and mental health symptoms (Kuo et al., 2015; Miller et al., 2015; Sylvia et al., 2015) and is easily modifiable for different clinical groups; however, it has not yet been adapted for smoking cessation due to the lack of available information regarding the characteristic needs of smokers who might utilize this treatment.
The purpose of this study is therefore to explore the prevalence of current smokers who present to a mind–body medicine clinic, as well as describe their demographic and psychosocial characteristics and how these characteristics differ as compared to never and former smokers. Specifically, we explore differences among smoking status groups (i.e. current, former, and never smokers) in terms of demographic variables (i.e. age, gender, race, marital status, employment status, and education level), psychological variables (i.e. depression, anxiety, pain, and stress), and social support–related variables (i.e. emotional support, instrumental support, and social isolation) in order to gain a better understanding of smokers’ needs and thus inform future intervention development efforts. The hypothesis is that the prevalence of current smokers will be lower than the prevalence of never and former smokers, and that current smokers will report greater levels of psychosocial problems.
Method
Participants
Patients were 593 adults who presented to a mind–body medicine clinic at a large academic medical center in the northeast for the treatment of stress-related problems.
Measures
Demographics
Patients completed a demographic questionnaire. They were asked to report age, gender, race, marital status, employment status, and education level.
Tobacco use
Three questions from the Behavioral Risk Factor Surveillance System (BRFSS) were used to assess current and past smoking status and patterns. The BRFSS is a widely used questionnaire developed by the Centers for Disease Control and Prevention (CDC) to assess health risk behaviors across multiple domains. It shows strong reliability and validity in general, and for the tobacco use questions in particular (Pierannunzi et al., 2013). Participants were asked to indicate whether they never smoked, smoked in the past, currently smoke some days, or currently smoke everyday. Those who indicated that they smoked in the past were asked when they quit, and those who indicated any current smoking were asked to indicate their time of smoking onset and current smoking rate (less than half a pack/day, 1 pack/day, 1.5 pack/day, 2 packs/day, or more than 3 packs/day). Some-day and everyday current smokers were collapsed for the analyses.
Psychosocial outcomes
Pain and stress were each assessed using a single-item 10-point analog scale, with higher scores indicating greater levels of pain and stress.
Four-item short-forms from the Patient-Reported Outcome Measurement Information System (PROMIS) were used to assess anxiety, depression, social isolation, instrumental support, and emotional support. PROMIS is a National Institutes of Health (NIH)-funded initiative that uses modern measurement theory and mixed-methods research to provide valid, reliable, and efficient item banks of patient-reported health outcomes (Cella et al., 2007, 2010; Kroenke et al., 2014). Items are assessed on a 5-point scale where higher scores indicate greater levels of the construct being measured over the past week. T-scores are reported for PROMIS measures, which standardize scores around the general population mean of 50 and standard deviation (SD) of 10 (Liu et al., 2010; Schalet et al., 2014).
Procedure
Patients presented to a mind–body medicine clinic at a large academic medical center in the Northeast between March 2015 and October 2015 for the treatment of stress-related symptoms. All participants completed the self-reported measures on an iPad prior to their initial intake session as part of standard clinical practice. The data were stored in a hospital-wide secure database within the electronic medical record system. Only study staff and medical providers within the hospital system had access to the data. The data were extracted, de-identified, and cleaned for the current analyses. The hospital institutional review board (IRB) approved all study procedures.
Data analysis
Frequencies and descriptive statistics were examined for all continuous variables. All variables were normally distributed, with skewness and kurtosis values below 1. One-way analysis of variance (ANOVA) was used to compare never, former, and current smokers on continuous variables. Post hoc pairwise analyses adjusted for multiple comparisons were used to examine significant overall group differences (p < .05 with two-tailed hypothesis testing). Confidence intervals are presented as an estimate of effect size. Chi-square (χ2) was used to compare smoking status groups on categorical variables. For categorical variables that differ by smoking status, percentages of current smokers within key categories are reported.
Results
The mean age of participants was 50.15 years (SD = 16.15) (see Table 1 for demographic details). In total, 70% of participants were female, 85% were White, 58% were married or living with a partner, 56% were currently employed, and 78% had completed college or higher degrees. In terms of smoking status, 66% were never smokers, 27% were former smokers, and 6% were current smokers. Former smokers had been quit for an average of 22.31 (SD = 14.73) years. Among the current smokers, 58% were daily smokers, and 64% smoked less than half a pack per day. Current smokers had been smoking for an average of 21.78 (SD = 13.43) years.
Demographic characteristics of patients by smoking status.
The test statistic for age is the F value from a one-way ANOVA; Fisher’s exact test is reported for education; all other test statistics are χ2. Percentages are valid percentages reflecting missing data and do not always sum exactly to 100 percent.
In examining demographic characteristics by smoking status, there were significant group differences in terms of age, marital status, employment, and education level (all ps < .05). Current smokers and never smokers were younger than former smokers (ps < .001). Current smokers were more likely to be single (45%), unemployed (18%) or on disability (36%), and relatively less likely to be college graduates (61%) as compared to the other groups. There were no significant group differences in terms of gender (p = .96) or race (p = .92).
Table 2 displays psychosocial characteristics by smoking status. Overall, for the full sample, participants reported elevations in anxiety symptoms (M = 59.72, SD = 8.85) and some minor elevations in depression symptoms (M = 55.94, SD = 9.30). Stress levels were also slightly above the mid-point of the 11-point scale (M = 6.41, SD = 2.30). Scores for pain, social isolation, emotional support, and instrumental support were similar to the general population means.
Psychosocial characteristics of patients by smoking status.
PROMIS: Patient-Reported Outcome Measurement Information System.
Mean values (standard deviations) are presented, with higher scores indicating greater levels of each outcome. Depression, anxiety, social isolation, emotional support, and instrumental support were measured using PROMIS subscales based on T-scores (population mean = 50, standard deviation = 10). Pain and stress were each measured using a single-item 11-point scale (0 = none to 10 = extremely). Post hoc pairwise comparisons were examined with Tukey corrections for multiple comparisons. There were no significant differences between never and former smokers.
When comparing these outcomes by smoking status groups, there were significant differences in terms of depression symptoms, F(2) = 3.37, p = .03, with current smokers reporting greater levels of depression symptoms than never smokers (95% confidence interval (CI) = .07–8.07). For pain, there was a significant difference between smoking status groups, F(2) = 11.21, p < .001, with current smokers reporting greater pain as compared to never smokers (95% CI = 1.23–3.68) and former smokers (95% CI = .84–3.42). For social isolation, there was a significant effect of smoking status, F(2) = 4.94, p = .01, with current smokers reporting greater social isolation compared to never smokers (95% CI = 1.14–9.20). Instrumental support also differed by smoking status, F(2) = 5.62, p < .001, with current smokers reporting less instrumental support than never smokers (95% CI = −9.49 to −1.28).
There were no significant differences between never and former smokers in any outcome (all ps > .05). There were no significant overall group differences in terms of anxiety, F(2) = 1.06, p = .35; stress, F(2) = .91, p = .40; or emotional support, F(2) = 1.33, p = .26.
Discussion
The purpose of this study was to explore the prevalence of smokers presenting to a mind–body medicine clinic, as well as explore demographic and psychosocial characteristics of current smokers as compared to never and former smokers, in order to inform future intervention development efforts. Consistent with prediction, results suggest that there was a low prevalence (6%) of current smokers, and current smokers were more likely to be unemployed or on disability, and had elevated psychosocial problems compared to never or former smokers.
Overall, the patients presenting to this mind–body clinic were primarily White, female, and highly educated. These sample characteristics are consistent with the characteristics of individuals who tend to use mind–body interventions (Barnes et al., 2004, 2008; Clarke et al., 2015), and they likely explain the observed smoking prevalence rate of 6%. Indeed, while this smoking rate is lower than that of the general population (i.e. 18%), it is the same as the national smoking rate among highly educated groups (Centers for Disease Control and Prevention, 2013). Thus, smokers presented to our mind–body clinic at the rate that would be expected for individuals who use mind–body services. The observed prevalence rate and sample characteristics might therefore accurately reflect the population of smokers who seek mind–body services. The current findings also suggest that smokers who present to mind–body medicine clinics have been smokers for many years but are light smokers, which might reflect efforts to minimize or compensate for the negative health effects of smoking given that mind–body medicine users are generally health conscious (Bishop and Lewith, 2010; Bishop et al., 2007).
In terms of psychosocial variables, consistent with the literature, current smokers reported greater depression symptoms, pain, and social isolation, and less instrumental support compared to never or former smokers. They also reported higher rates of disability, higher rates of unemployment, and less education, though in general current smokers were still highly educated. Thus, the current findings indicate that there is a significant minority of highly educated smokers who seek mind–body interventions and experience psychosocial challenges that are likely to complicate smoking cessation efforts. These findings provide important information for developing targeted mind–body smoking cessation interventions: they suggest that these interventions focus on psychoeducation and skills training regarding the application of mind–body techniques for depression, pain, and social support problems. For example, interventions such as the SMART-3RP, which teach deep breathing and yoga as general coping strategies, might enhance their self-efficacy for smoking cessation by focusing on how these strategies can be used for pain and mood management in the context of smoking cessation (Pilkington et al., 2005; Wren et al., 2011). These findings also support a group format for mind–body smoking cessation interventions, as the group setting would be useful for promoting social support and reducing feelings of social isolation (Barrera et al., 2002); the SMART-3RP, accordingly, is delivered in a group format and focuses on enhancing and utilizing one’s social support.
Interestingly, inconsistent with the literature, current smokers did not differ from former or never smokers in terms of anxiety or stress. This lack of significant findings is surprising given the large literature documenting associations between smoking and anxiety (Breslau, 1995; Zvolensky et al., 2005). One explanation for these findings may be that smokers were light smokers, as higher smoking rates have been associated with greater anxiety-related problems (McLeish et al., 2007). Another possible explanation is that all groups had elevated anxiety symptoms, thereby obscuring any between-group differences. Indeed, anxiety, regardless of smoking status, was the only outcome for which all groups scored approximately 1 SD above the normative mean. Given these overall elevations, which are not uncommon for patients seeking mind–body treatment (Bishop and Lewith, 2010), it is likely important for mind–body smoking cessation interventions to target anxiety or stress symptoms. There were also no differences between smoking status groups in terms of emotional support, suggesting that while current smokers may feel isolated and perceive limited support in managing tangible tasks, they do believe there are others they can turn to during times of distress. Given that former smokers had been non-smokers, on average, over two decades, it is understandable that there were no significant differences in any outcome among former smokers and never smokers.
The fact that smokers, particularly those from disadvantaged groups, are less likely to use mind–body medicine services warrants consideration. Recent research suggests that low-SES individuals may be less likely to utilize both mind–body interventions and traditional tobacco treatments due to differences in treatment awareness and access, rather than differences in income level (Bishop and Lewith, 2010). Mind–body interventions, especially those delivered in group settings, have the potential to benefit low-SES smokers by teaching coping skills for managing psychosocial stressors and withdrawal symptoms, reducing nicotine withdrawal symptoms, increasing quitting self-efficacy, and promoting emotional and smoking-related social support, all of which have the potential to promote smoking cessation. Accordingly, it is important to develop outreach and recruitment efforts that enhance awareness and overcome barriers to mind–body treatment access for less educated, low-income, or ethnic minority smokers. Including images that represent ethnic minorities and language that is clear to lower reading levels, and integrating services into local clinics, may also increase program relevance, heighten program awareness and accessibility, and ultimately improve treatment uptake. Highlighting other contextual factors of mind–body smoking cessation interventions, particularly the social interactions and supportive nature of the program, might also appeal to low-SES smokers who have less social support.
To our knowledge, there is little research examining the characteristics of patients who present to mind–body medicine clinics, particularly in the United States. Two of these previous studies were from our research group and examined the same mind–body clinic examined in this study. Results similarly indicated that most patients were females, married, and highly educated, and they also demonstrated that 41 percent of participants reported high somatic symptoms (Nakao et al., 2001c), and diagnoses of irritable bowel syndrome (7%), anxiety (6%), and insomnia (5%; Nakao et al., 2001a). Other studies conducted within mind–body clinics were conducted in Japan. These studies also found a high prevalence of somatic symptoms (e.g. 86% fatigue, 79% insomnia) and mood disorder symptoms, including 31 percent of patients reporting suicidal ideation and 4 percent meeting diagnostic criteria for depression (Nakao et al., 2001d; Nakao et al., 2002). Women (Nakao et al., 2001b) and patients with depression (Nakao et al., 2001d) were shown to report greater somatic symptoms than men and non-depressed patients. No previous studies, however, examined the prevalence or characteristics of patients presenting to these clinics in terms of their smoking history.
Limitations to this study are worth noting. First, we were unable to measure smoking-specific factors that may be relevant for further targeting mind–body interventions (e.g. smoking-specific social support, quitting self-efficacy). Second, we were also unable to explore the use of electronic cigarettes or other tobacco products, which are potentially important targets for tobacco interventions. Third, although representative of the individuals who tend to use mind–body interventions, the current results cannot generalize to racial minority smokers or those with less education. Research should focus on outreach efforts for low-SES groups and work to identify their specific tobacco treatment needs. The findings are also from a single clinic site in a single North-eastern state, potentially limiting generalizability to clinics in other areas. Future research should examine a wider range of smoking-related outcomes among smokers who seek mind–body interventions across diverse clinics, as well as examine how these factors influence smoking cessation outcomes.
Overall, the current results suggest that smokers could benefit from mind–body smoking cessation interventions that target depression symptoms, pain, and social support in particular. Work is clearly needed, however, to engage smokers in mind–body interventions. These findings support the rationale for developing mind–body smoking cessation interventions and provide important information for targeting these interventions to the specific psychosocial problems that underlie smoking behavior.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by funds from the National Center for Complementary and Integrative Health (NCCIH 2T32AT000051-6; Dr. Luberto) and National Cancer Institute (1K24CA197382; Dr. Park).