Gastrointestinal-specific anxiety and smoking abstinence expectancies among persons with irritable bowel syndrome.

Participants were recruited from across the United States (US) using Qualtrics Panels, an online survey management system. The name of the study (i.e., Gastrointestinal Disorders and Smoking Investigation) and inclusion/exclusion criteria were listed in the Qualtrics Panels Management system and adults were able to self-select into the study if they believed they may be eligible. Participants who self-selected into the study first completed a brief eligibility screening form. If deemed eligible, participants were asked to provide informed consent and were directed to the study survey. Participants could opt to receive the equivalent of $10.75 in compensation via cash-based incentives (i.e., gift cards), reward miles, or reward points. To ensure quality responses, a speeding check (i.e., one-half the median survey completion time) was included and IP addresses were recorded to prevent multiple attempts to complete the survey by the same respondent. The study was approved by the Institutional Review Board of the university where the study took place. A total of 2,858 individuals responded to the study advertisement, of which 2,141 were excluded because they did not meet self-reported ROME-IV criteria for IBS ( Palsson et al., 2016 ), 181 were excluded because they reported smoking less than 5 cigarettes per day for the past year, 84 were excluded because they were under 18 years of age, and 189 were excluded because they did not pass data quality checks, resulting in a final sample of 263 individuals for the current analyses. Participants were 263 adults who met criteria for IBS and endorsed smoking 5 or more cigarettes per day (52.1% female; M age = 44.1 years, SD = 12.71). Inclusion criteria was as follows: (1) reported being 18–75 years of age, (2) endorsing current daily smoking (≥ 5 cigarettes per day), (3) GI symptoms consistent with Rome IV criteria for IBS including recurrent abdominal pain occurring at least weekly and two or more of the following that occur at least 30% of the time: pain related to defecation, pain associated with a change in frequency of stool, pain associated with a change in the appearance of stool; and symptom onset at least 6 months ago ( Drossman & Tack, 2022 ; Lacy et al., 2016 ), and (4) having internet access (to ensure ability to complete the study survey). Exclusionary criteria included: (1) lack of English proficiency (to ensure comprehension of survey items), and (2) an inability to provide voluntary informed consent. Demographic information collected included sex, age, race/ethnicity, annual income, and highest level of education. Age, sex, and race/ethnicity were included as covariates. The Chronic Overlapping Pain Conditions Screener (COPCS; Schrepf et al., 2024 ) is designed to identify the presence of nociplastic pain conditions that frequently co-occur in an individual. Respondents are presented with a body map and asked to indicate areas in which they experience chronic pain. Based on the indicated area(s), they respond to a series of diagnostic questions corresponding to a relevant pain condition (i.e. if an individual marks jaw pain on the body map, they will be presented with questions corresponding to the criteria for temporomandibular disorder [TMD]). As the COPCS provides a comprehensive assessment of 10 different chronic pain conditions, it was used in this study to describe the sample in terms of reporting prevalence of specific pain conditions, including but not limited to TMD, IBS, and fibromyalgia (FM). The Overall Depression Symptom and Impairment Scale (ODSIS; Bentley et al., 2014 ) is a 5-item self-report measure assessing symptoms and functional impairment associated with depressive symptoms ( Bentley et al., 2014 ). Items of the ODSIS are summed to create total score ranging from 0 to 25; higher scores indicate more severe symptoms and impairment. The ODSIS has shown strong convergent and discriminant validity across both clinical and non-clinical samples and demonstrates measurement invariance across diverse samples ( Ito et al., 2015 ). Additionally, a clinical cutoff for ODSIS has been established, such that a score of 8 or above indicates likely depression. The ODSIS total score was utilized in descriptive analyses and as a covariate in all models in the current study. ODSIS evidenced excellent internal reliability in the current sample (α = 0.95). The Smoking History Questionnaire (SHQ; Brown et al., 2002 ) assesses history of smoking, including age of onset, products used, and information about past quit attempts. The SHQ also measures the severity of problems when attempting to quit smoking. The scale is comprised of 17-items rated on a 5-point scale (1 = not at all , 5 = extremely ) asking participants how serious each item was in their attempts to quit smoking. Items are averaged for a total severity score, where higher scores indicate more problems experienced when attempting to quit. The SHQ was used to describe smoking history and for the number of cigarettes per day as a covariate. The Fagerström Test for Cigarette Dependence-Revised (FTCD-R; Korte et al., 2013 ) is a 6-item self-report measure of cigarette dependence. Items on the FTCD-R vary in scale and grading (e.g., “What cigarette would you most hate to give up? 1 = first in the morning, 0 = all others” and “How soon after you wake up do you have your first cigarette? 3 = within 5 minutes, 2 = 6–30 minutes, 1 = 21–30 minutes, 0 = after 60 minutes”). All items are summed for a total score and higher scores indicate a greater degree of cigarette dependence. In the current study, the FTCD-R was utilized in descriptive analyses (α = 0.51). The Visceral Sensitivity Index (VSI; Labus et al., 2004 ) is a 15-item measure of GI-specific anxiety (e.g., “I often worry about problems in my belly”). Items are rated on a 6-point Likert scale (1 = ‘ strongly agree’ to 6 = ‘strongly disagree’ ). In the current analyses, the VSI were reverse coded, then summed to create a total score such that higher scores indicate more severe GI-specific anxiety. The VSI total score was used in the primary analyses as a predictor variable (α = 0.95). The Smoking Abstinence Expectancies Questionnaire (SAEQ; Abrams et al., 2011 ) was utilized to assess the short-term experiences resulting from smoking abstinence amongst individuals who engage in daily smoking. This questionnaire consists of 28-questions on a 7-point Likert scale from 0 ( very unlikely ) to 6 ( very likely ). These questions, divided into 4 subscales, encompass the many possible psychological and physiological consequences of smoking abstinence. The SAEQ yields four factors, with higher scores representing stronger beliefs about abstinence expectancies. Three subscales represent negative abstinence expectancies, including negative mood (e.g., “I would feel tense.”), somatic symptoms (e.g., “My chest would feel tight.”), and harmful consequences (e.g., “I would feel like I’m losing control.”). A fourth subscale reflects positive expectancies (positive consequences; e.g., “I would feel happy”). For each of the subscales, greater scores indicate stronger smoking abstinence expectancies (α range = 0.91–0.95). In this current study, the SAEQ’s four subscales were each used as criterion variables. Using SPSS Version 29, first, sample descriptive statistics and zero-order correlations among study variables were examined. Second, four identical 2-step hierarchical regression analyses were conducted with the abstinence expectancies: (1) negative consequences, (2) somatic symptoms, (3) harmful consequences, and (4) positive consequences. The first step included the following covariates: age ( Stake-Nilsson et al., 2013 ), sex (0 = male, 1 = female; Smith et al., 2016 ), race/ethnicity (0 = non-Hispanic White, 1 = all other races/ethnicities; Ballou & Keefer, 2013 ), cigarettes per day ( Farzaneh et al., 2013 ), and depressive symptoms ( Mendelsohn, 2012 ). The VSI was then entered in the second step. Model fit for each of the steps was evaluated with the F statistic and increase in variance accounted for (change in R 2 ). Squared semi-partial correlations ( sr 2 ) were used as measures of effect size for each of the individual predictors and are interpreted as small (0.01), medium (0.09), and large (0.25) ( Ferguson, 2016 ).

Detailed descriptive information for the sample is presented in Table 1 . Participants reported smoking an average of 18.49 ( SD = 16.42) cigarettes per day. The average age of smoking onset in the sample was 17.8 years ( SD = 6.93). The average level of cigarette dependence in the sample, as measured by the FTCD-R ( Korte et al., 2013 ), was in the moderate range ( M = 5.54; SD = 2.01). In addition to smoking combustible cigarettes, approximately one quarter (24.3%) of the sample reported daily use of electronic cigarettes. Additionally, from a lifetime perspective, 51.0% reported cigar use, 38.4% reported smokeless tobacco use, and 30.0% reported pipe tobacco use; the frequency of use for these products was not assessed. Regarding depression symptoms, 55.1% of the sample ( N = 145) met criteria for probable clinical depression based on the ODSIS ( Fig. 1 ). Bivariate correlations are presented in Table 2 . GI-specific anxiety was statistically significantly and positively correlated with cigarettes per day, depressive symptoms, and all four smoking abstinence expectancies (negative consequences, somatic symptoms, harmful consequences, and positive consequences). Regression results for all models are presented in Table 3 . For the smoking abstinence negative consequences expectancies, covariates entered in the first step accounted for approximately 13% of the variance ( R 2 = 0.13, F [5, 257] = 7.64, p <.001), and sex and depressive symptoms emerged as statistically significant predictors. At step two, the addition of the GI-specific anxiety (Δ R 2 = 0.07, Δ F [1, 256] = 21.85, p <.001) accounted for approximately 7% of additional variance. For smoking abstinence somatic symptom expectancies, covariates entered in the first step accounted for 19% of the variance ( R 2 = 0.19, F [5, 257] = 12.16, p <.001), and age and depressive symptoms emerged as statistically significant predictors (see Table 3 ). At step two, the addition of GI-specific anxiety (Δ R 2 = 0.12, Δ F [1, 256] = 45.95, p <.001) accounted for an additional 12% of variance. For smoking abstinence harmful consequence expectancies, covariates entered in the first step accounted for 17% of the variance ( R 2 = 0.17, F [5, 257] = 10.25, p <.001), and cigarettes per day and depressive symptoms emerged as statistically significant predictors (see Table 3 ). At step two, the addition of GI-specific anxiety (Δ R 2 = 0.15, Δ F [1, 256] = 54.78, p <.001) accounted for an additional 15% of variance. For smoking abstinence positive consequence expectancies, covariates entered in the first step accounted for 20% of the variance ( R 2 = 0.20, F [5, 257] = 13.19, p <.001), and age, sex, race/ethnicity, cigarettes per day, and depressive symptoms emerged as statistically significant predictors (see Table 3 ). At step two, the addition of GI-specific anxiety (Δ R 2 = 0.04, ΔF[1, 256] = 13.92, p <.001) accounted for an additional 4% of variance.

IBS is a prevalent chronic illness that has profound negative consequences across life domains ( Oka et al., 2020 ). Although health behaviors (e.g., eating, physical activity) have been routinely studied and targeted in the context of the clinical management of IBS, less focus has been oriented on substance use behavior. This study aims to expand on previous research by evaluating the association of GI-specific anxiety, an established vulnerability factor for severe IBS ( Farzaei et al., 2016 ; Sibelli et al., 2016 ), with smoking abstinence expectancies in adults with IBS who smoke. This approach is potentially fruitful given psychological-based interoceptive concerns are linked to IBS ( Deiteren et al., 2016 ) and smoking ( Zvolensky et al., 2017 ) and may reflect a broadband construct relevant to understanding this comorbid population ( Nord & Garfinkel, 2022 ). As hypothesized, GI-specific anxiety was significantly related to greater abstinence expectancies for negative mood, somatic symptoms, and harmful consequences. The sizes of the incremental effects for GI-specific anxiety ranged from small to medium (7% for negative mood to 15% of unique variance for somatic symptoms). Further, the observed effects were evident after accounting for a wide range of covariates (e.g., depressive symptoms, cigarettes per day; 13 − 20% of variance across models) that exerted clinically important effects on the outcomes ( Abelson, 1985 ). Of note, the current study sample was comprised of majority of individuals who met criteria for clinical depression (i.e., 55.1%). The prevalence of depression in the current study was higher than prevalence estimates in the general population with IBS. For example, meta-analytic findings estimate that between 20 and 40% of individuals with IBS have depressive symptoms ( Hu et al., 2021 ; Nikolova et al., 2022 ; Tarar et al., 2023 ; Zhang et al., 2018 ). Importantly, a wealth of literature also documents high rates of depression among smokers ( Han et al., 2022 ; Wu et al., 2023 ), and depression places individuals who smoke at risk for poorer smoking outcomes ( Mathew et al., 2017 ). Therefore, the combined impact of IBS and clinical depression may contribute to worse outcomes related to smoking behavior among this subgroup of the population. These are the first data to suggest individual differences in GI-specific anxiety among adults with IBS who smoke are important to understanding abstinence-based cognition. These findings align with the limited existing research on GI-specific anxiety in the general population ( Zvolensky et al., 2017 ) and notable given that smoking abstinence expectancies are linked to the maintenance and relapse of smoking in previous work ( Hendricks et al., 2011 ). For example, persons with IBS who smoke with greater GI-specific anxiety who expect more negative consequences during abstinence may be less likely to engage in a quit attempt or have less success in maintaining abstinence. In contrast to prediction, however, GI-specific anxiety was positively associated at the bivariate level of positive abstinence expectancies and showcased an incremental effect with this smoking expectancy construct. Notably, while statistically significant, of the four criterion variables modeled, the addition of GI-specific anxiety in the model examining positive expectancies showed the weakest change variance. Previous research has tended to show minimal or negative relations between negative affect constructs and positive abstinence expectancies ( Zvolensky et al., 2023a ). Because of the limited work on IBS and smoking, caution should be exercised in overinterpreting this finding at this stage of research development. Nonetheless, one working perspective on this result may be that GI-specific anxiety may prompt ‘health-oriented attention’ to smoking, and thereby elicit beliefs that abstinence can have positive consequences ( Horenstein & Heimberg, 2020 ). The question arises how GI-specific anxiety may be related to abstinence expectancies. GI-specific anxiety, as a threat-based response to interoceptive sensations, may drive escape and avoidance tendencies, including smoking as a means to manage discomfort ( Deutsch & Strack, 2006 ; Kakoschke et al., 2017 ; Rooke et al., 2008 ; Wiers et al., 2007 ). Among people who smoke, such motivation may elicit the propensity to smoke to regulate aversive psychosomatic symptoms ( Fillo et al., 2016 ). Thus, from a negative reinforcement perspective ( Wise & Koob, 2014 ), periods of abstinence may induce greater worry and concerns (e.g., negative mood, negative consequences, and somatic sensations). In this way, GI-specific anxiety could be related to negative beliefs about abstinence, as was evident in the current work. Alternatively, dual process models suggest anxiety can trigger avoidance and approach behaviors, leading to individuals to recognize the negative and positive effects of abstinence ( Wiers & Stacy, 2006 ). Further, GI-specific anxiety may create ambivalence wherein individuals acknowledge harms and benefits of smoking (Leventhal, 2011), theoretically leading to negative and positive abstinence expectancies. However, future work is needed to further validate this theory and determine the explanatory impact of smoking abstinence expectancies (both negative and positive) in the context of GI-specific anxiety and smoking cessation. Clinically, the present findings highlight the potential importance of screening for elevated GI-specific anxiety in adults with moderate to severe IBS who smoke, as it may represent a critical factor in smoking cessation efforts. Indeed, heightened levels of GI-specific anxiety could identify a specific vulnerability that may fuel dysfunctional abstinence-based cognitions linked to smoking maintenance and relapse among the IBS population. At present, there are no targeted or specialized smoking cessation programs for individuals with IBS. As such, this chronic illness population is apt to experience continued challenges in quitting and adverse health consequences related to smoking ( United States Department of Health and Human Services, 2014 ). The present work suggests that one route to change may be to develop targeted intervention protocols that seek to engage GI-specific anxiety among the IBS population to improve abstinence-based cognitions. Future research is needed to explore if reducing GI-specific anxiety improves abstinence expectancies, and by extension, clinical outcomes for smoking cessation among the IBS population. There are study limitations that warrant comment. First, the study design was cross-sectional. Therefore, causal and directional inferences cannot be made. Future research could build from this work and test the validity of the present model using longitudinal or experimental methodologies. Second, the nationally recruited sample may not be representative of the larger population of individuals in the United States with IBS. It should be noted that participants selected into the study, and there was a high prevalence of depression and other chronic pain conditions in the sample. IBS commonly co-occurs with other chronic pain conditions ( Riedl et al., 2008 ), although epidemiological studies reporting prevalence rates are limited. Rates of chronic low back pain in the current sample (38.8%) is similar to what has been reported in other studies (38%; Jones et al., 2001 ), but prevalence of chronic fatigue syndrome in the current sample (39.2%) is somewhat higher than previous estimates (14%; Jones et al., 2001 ). Importantly, rates of cigarette use were not documented in these studies and it is possible that the clinical picture of the present sample differs from previously studied sample, given that these are individuals who smoke and have IBS (e.g., individuals who smoke may be more likely to experience fatigue; Kahraman et al., 2021 ). Additionally, the sample was comprised of majority non-Hispanic White To test the generalizability of the model to other racial/ethnic groups, research could sample a more racially/ethnically diverse group. This effort is important given inequities in smoking ( Nguyen-Grozavu et al., 2020 ) and IBS ( Sasegbon & Vasant, 2021 ). Third, the assessment approach that was employed involved validated self-report instruments. Moreover, the smoking processes included were cognitive processes known to be associate with smoking behavior, albeit, not objective outcomes of smoking behavior. To cross index volitional reporting, multimethod assessment protocols utilizing objective smoking outcomes (e.g., CO analysis; Mattes et al., 2014 ) could be used in future research. Fourth, the current study controlled for depressive symptoms. However, we recognize that a number of potential confounding variables may be relevant. As such, future work should consider controlling for other relevant psychological variables (e.g., Posttraumatic stress symptoms) to isolate the unique impact of GI-Specific anxiety. Fifth, participants were included in the current study if they experienced GI symptoms. Specifically, individuals had to experience symptoms consistent with Rome IV criteria for IBS, including recurrent abdominal pain occurring at least weekly and two or more of the following that occur at least 30% of the time: pain related to defecation, pain associated with a change in frequency of stool, pain associated with a change in the appearance of stool; and symptom onset at least 6 months ago ( Drossman & Tack, 2022 ; Lacy et al., 2016 ). Future research would benefit from confirming eligibility criteria via medical record review to ensure IBS diagnosis as well as control for other medical conditions. Sixth, the current study examined one theoretically relevant construct in smoking behavior (i.e., GI-specific anxiety). However, it is important to note that GI-specific anxiety is just one of several affective vulnerability factors (e.g., neuroticism, general anxiety) that may be clinically relevant to examine in the context of smoking processes among individuals with IBS. Future work is needed to evaluate the concurrent role of these various constructs to determine the strongest indicators of smoking abstinence expectancies and to further improve theoretical knowledge and understanding. Finally, we did not sample persons engaged in a quit attempt or who were seeking treatment for smoking cessation. Future research may benefit from testing the current model among people with IBS who smoke involved in a quit attempt. Overall, the current investigation found that GI-specific anxiety was associated with negative and positive abstinence expectancies among adults with IBS who smoke. This study is the first to suggest that individual differences in GI-specific anxiety can impact abstinence expectancies in a population with IBS. Future research is now needed to explore the potential of reducing GI-specific anxiety to improve abstinence expectancies among this chronic health population in efforts to improve cessation success. Specifically, it would be fruitful to further expand the current model to include smoking-related outcomes (e.g., number of cigarettes smoked per day, attempts to quit smoking) to determine if smoking abstinence expectancies serves as a mechanism in the relationship between GI-Specific anxiety and smoking cessation.