Psychosomatic Medicine Tips for Better Browsing
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Labus, J. S.
Right arrow Articles by Naliboff, B. D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Labus, J. S.
Right arrow Articles by Naliboff, B. D.
Psychosomatic Medicine 69:89-98 (2007)
© 2007 American Psychosomatic Society

ORIGINAL ARTICLES

The Central Role of Gastrointestinal-Specific Anxiety in Irritable Bowel Syndrome: Further Validation of the Visceral Sensitivity Index

Jennifer S. Labus, PhD, Emeran A. Mayer, MD, Lin Chang, MD, Roger Bolus, PhD and Bruce D. Naliboff, PhD

From the Departments of Medicine (L.C., E.A.M.), Physiology (E.A.M.), Psychiatry and Biobehavioral Sciences (J.S.L., E.A.M.), and Brain Research Institute (E.A.M., B.D.N.), Center for Neurovisceral Sciences and Women’s Health, David Geffen School of Medicine at UCLA; VA Greater Los Angeles Healthcare System (L.C., B.D.N.), Los Angeles, California.

Address correspondence and reprint requests to Jennifer Labus, Center for Neurovisceral Sciences and Women’s Health, Peter V. ueberroth Bldg., Rm. 2338c2, 10946 Le Conte Avenue, Los Angeles, CA 90095. E-mail: jlabus{at}ucla.edu


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 NOTES
 Appendix
 REFERENCES
 
Objectives: The Visceral Sensitivity Index (VSI) was developed as the first instrument to assess gastrointestinal-specific anxiety, the cognitive, affective, and behavioral response to fear of gastrointestinal sensations, symptoms, and the context in which these visceral sensations and symptoms occur. The purpose of the current study was to a) replicate the previously reported psychometric properties of the VSI, b) assess the known-groups and concurrent validity of the instrument, and c) test conceptual hypotheses regarding gastrointestinal-specific anxiety in comparison to other general measures of psychological distress as a crucial mechanism (mediator/moderator) underlying irritable bowel syndrome diagnosis and its symptoms.

Methods: Two undergraduate student samples (n > 500) were administered the VSI along with measures assessing presence of lower gastrointestinal symptoms, nongastrointestinal pain, health-service utilization, anxiety, depression, vitality, neuroticism, and anxiety sensitivity. Path analyses tested the hypothesis that gastrointestinal-specific anxiety mediates the relationship between affective variables and irritable bowel syndrome diagnosis and symptoms. A ‘known-groups’ validity approach elucidated the relevance of gastrointestinal-specific anxiety across a spectrum of irritable bowel syndrome patients.

Results: Good concurrent, divergent and discriminant validity was demonstrated. Logistic regression revealed that gastrointestinal-specific anxiety is the key explanatory variable of irritable bowel syndrome diagnostic status. Path analysis demonstrated that gastrointestinal-specific anxiety mediates the relationship between general psychological distress measures and gastrointestinal symptom severity. The VSI was related to gastrointestinal, but not nongastrointestinal, symptom severity.

Conclusions: Overall, the VSI demonstrated excellent psychometric properties providing further support for its use in mechanistic studies of the role of anxiety in irritable bowel syndrome presentation.

Key Words: gastrointestinal • anxiety • assessment • irritable bowel syndrome

Abbreviations: ANX = anxiety; ASI = Anxiety Sensitivity Index; BSQ-SF = Bowel Symptom Questionnaire-Short Form; DEP = depression; EPQN = Eysenck Personality Questionnaire-Short Form; GI = gastrointestinal; HAD = Hospital Anxiety and Depression Scale; IBS = irritable bowel syndrome; VIT = Medical Outcomes Study-SF-36 energy/fatigue subscale; VSI = Visceral Sensitivity Index.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
METHODS
 RESULTS
 DISCUSSION
 NOTES
 Appendix
 REFERENCES
 
Anxiety about specific gastrointestinal (GI) sensations is thought to play a key role in the maintenance and perhaps development of irritable bowel syndrome (IBS) symptoms (1). GI-specific anxiety can be defined as the cognitive, affective, and behavioral response stemming from fear of GI sensations, symptoms, and the context in which these visceral sensations and symptoms occur. Examples of GI symptom-related contexts include situations involving food and eating, like restaurants and parties or locations in which bathroom facilities are unknown or difficult to reach. GI-specific anxiety includes hypervigilance to, and fear, worry, and avoidance of, GI sensations and contexts. As a marker of over responsiveness to GI sensations, GI-specific anxiety is hypothesized to perpetuate IBS symptoms through alterations in autonomic and pain facilitation, as well as cognitive mechanisms (1). GI-specific anxiety may be associated with beliefs of poor symptom control and high illness impact and may, therefore, be a critical element in the severe decrements in quality of life found in IBS patients (2).

Analogous models of stress sensitivity have been proposed for certain anxiety disorders (3,4) and more recently for chronic somatic pain (5,6). Anxiety sensitivity has been postulated as a constitutional predisposition characterized by hypersensitivity to anxiety-related somatic sensations (autonomic arousal) based on beliefs that these sensations have harmful somatic, psychological, or social consequences (7,8). Anxiety sensitivity has been conceptualized as a stable personality trait and empirically distinguished from trait anxiety (9). That is, anxiety sensitivity refers to the specific tendency to respond fearfully to the interoceptive sensations of anxiety rather than the predisposition to respond anxiously to a wide range of stressors that characterizes trait anxiety. Anxiety sensitivity has been found to be a risk factor for anxiety pathology and is commonly measured by the Anxiety Sensitivity Index (ASI) (10), a multi-dimensional instrument assessing panic/anxiety-specific somatic sensations (i.e., heart palpitations), cognitive dyscontrol (i.e., inability to concentrate), and fear of publicly observable dysfunction (e.g., sweating) (8). In addition, high scores on the ASI have been found to be predictive of negative experience of somatic pain as well as presence of a functional GI disorder (i.e., globus and functional dyspepsia) (5,11–15).

Neuroticism or trait anxiety has also been identified as a predisposing condition for the development of IBS symptoms (and GI symptom severity) in some individuals (16). Neuroticism is defined by the tendency to experience a broad spectrum of negative emotions, has been associated with exaggerated response to bodily sensations, and is a predictor of anxiety sensitivity. Anxiety sensitivity is thought to arise when the general tendency to experience negative emotions (neuroticism) interacts with specific learning experiences (i.e., anxiety is harmful) (17,18).

A recent factor analysis of a more comprehensive measure of anxiety sensitivity developed as an alternative to the ASI, the Anxiety Sensitivity Profile, revealed four specific somatic mechanisms or domains of anxiety sensitivity in addition to a single higher order factor (general anxiety sensitivity trait). These corresponded to a) fear of respiratory symptoms, b) fear of cardiovascular symptoms, c) fear of cognitive dyscontrol, and d) fear of GI symptoms (7). These findings support hypotheses of specific manifestations of anxiety sensitivity related to clusters of symptoms. Given its focus on panic related content, the ASI may index a panic-specific mechanism of anxiety sensitivity, whereas the Pain Anxiety Symptom Scale (19,20), which has been developed and validated for assessment of anxiety related to somatic pain, may assess somatic pain-specific anxiety sensitivity (21,22).

Recently, the Visceral Sensitivity Index (VSI) was developed as the first instrument to assess GI-specific anxiety. Initial validation efforts indicate that the VSI is an efficient, reliable, and valid measure of GI-specific anxiety in an IBS patient population (23). GI-specific anxiety, as measured by the VSI, may be the GI pain-specific expression of anxiety sensitivity in IBS. As such, the VSI may prove useful for clinical assessment, mechanistic studies of the role of anxiety in IBS presentation, and as a measure of treatment outcome. Although promising, the VSI requires further cross-validation information, especially in broader, nonpatient and nonIBS populations.

The aim of the current study is to further demonstrate the psychometric properties of the VSI in nonpatient IBS and community samples using a known-groups validity approach (24). Several specific questions are addressed in this cross-validation study. a) Is the internal reliability and factor structure of the instrument preserved across IBS and nonIBS populations? b) Is GI-specific anxiety, as assessed by the VSI, a distinct construct from state anxiety, anxiety sensitivity, and neuroticism? The initial validation data indicated moderately high correlations between general anxiety symptoms and the VSI, but it is hypothesized that this finding is population specific. In IBS patients, GI-specific anxiety may generalize to global anxiety symptoms, but in a broader population, fear of GI sensations should be more distinct from other fears and worries. c) Does the VSI differentiate subjects with and without IBS symptoms and levels of IBS severity? In particular, we anticipate that for individuals meeting Rome II diagnostic criteria for IBS (25), GI-specific anxiety levels will be highest for individuals endorsing IBS-related health-care seeking. Given its hypothesized role as a GI-specific manifestation of anxiety sensitivity, we also predict that the VSI will not relate to nonGI pain severity indices. In sum, the purpose of the current study was to a) replicate the previously reported psychometric properties of the VSI, b) assess the known-groups and concurrent validity of the instrument, and c) test conceptual hypotheses regarding GI-specific anxiety in comparison to other general measures of psychological distress as a crucial mechanism (mediator/moderator) underlying IBS diagnosis and its symptoms.


   METHODS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
RESULTS
 DISCUSSION
 NOTES
 Appendix
 REFERENCES
 
Participants
Three samples were utilized. Samples 1 and 2 were obtained in 2004 from university undergraduate students enrolled in introductory psychology courses at the University of California, Los Angeles (UCLA). The study was approved by the UCLA Institutional Review Board. Undergraduate students were asked to voluntarily complete questions during their regularly scheduled class time and were under no obligation to participate. Regardless of major study area, psychology classes are among the most widely attended of all undergraduate classes. Our questionnaire was part of a larger package of questionnaires pertaining to other psychological research areas. For empirical comparison, data from individuals meeting IBS Rome diagnostic criteria from the initial validation sample of IBS patients (primarily recruited through community advertisement) was also extracted (Sample 3) (23).

Measures

UCLA Bowel Symptom Questionnaire-Short Form (BSQ-SF)
The BSQ-SF included the 7 questions comprising the Rome II survey criteria (25,26), as well as an assessment of presence of nonGI pain such as headache, back or neck pain, and GI-related health-service utilization. These questions are a survey format designed to assess the primary diagnostic symptom criteria recommended for determining a diagnosis of IBS by an evidence and expert-based consensus (26).

Symptom Severity Indices
A 21-point numerical rating scale was used to obtain a general index of overall GI symptom severity. Individuals endorsing IBS symptoms were asked to ‘rate the overall severity of [their] gastrointestinal symptoms during the past week.’ The scale was anchored on the left with ‘no symptoms’ and on the right with ‘the most intense symptoms imaginable.’ A second 21-point numerical rating scale was also utilized to assess severity of nonGI pain during the past week for individuals endorsing experiencing symptoms of neck/shoulder pain, headache, muscle/joint pain, lower back/hip pain, and nonspecific aching on a regular basis.

Medical Outcomes Study SF-36-Energy/Fatigue Subscale (VIT)(27,28)
The 4-item energy/fatigue subscale assesses self-perceived levels of energy (2) and includes the degree of feeling ‘full of life,’ feeling ‘worn out,’ and feeling ‘tired.’ Decrements in energy/fatigue levels are considered a critical component of health status in patients with IBS (2,29). High scores on this measure indicate high energy and low fatigue.

Visceral Sensitivity Index (VSI)
The 15-item self-report questionnaire is designed to measure those unique aspects of fear, anxiety, and hypervigilance that can accompany misappraisals of visceral sensations and discomfort. This scale has demonstrated reliability and validity in a sample of IBS patients (23). Items on the VSI are reverse scored (i.e., 1–6 becomes 5–0) and totaled to yield a range of possible scores from 0 (no GI-specific anxiety) to 75 (severe GI-specific anxiety).

Hospital Anxiety and Depression Scale (HAD)
The HAD is a self-assessment mood scale specifically designed for use in nonpsychiatric settings (30). It has been extensively validated and is one of the most widely used brief inventories for assessment of symptoms of anxiety and depression. The HAD provides two 7-item subscales: anxiety (ANX) and depression (DEP) with scores ranging from 0 to 21. To provide a context for the scores for both subscales, scores from 0 to 7 represent ‘noncases,’ 8 to 10 ‘doubtful cases,’ and 11 to 21 ‘cases.’

Anxiety Sensitivity Index (ASI) (31)
The ASI is the most widely used self-report measure of anxiety sensitivity (32) and assesses panic-specific somatic sensations (i.e., heart palpitations), cognitive dyscontrol (i.e., inability to concentrate), and fear of publicly observable dysfunction (e.g., sweating). Conceptualized as a stable personality trait, anxiety sensitivity as measured by the ASI has been conceptually and empirically distinguished from trait anxiety (32).

Neuroticism Scale of the Eysenck Personality Questionnaire-Short Form (EPQN) (33)
The EPQN assesses the general trait of stress responsiveness. This is a widely used instrument in psychological and functional medical disorders, and is a strong predictor of functional GI disorders in the general population and development of IBS following acute infectious diarrhea (34). It has adequate internal consistency, good test–retest reliability, and the factor structure for the whole instrument is cross-culturally consistent (35).

Procedures
Sample 1 was administered a questionnaire packet that included the VSI, BSQ-SF, HAD-ANX, HAD-DEP, VIT, and EPQN. To provide data for confirmatory analysis, Sample 2 was also administered the VSI, BSQ-SF, HAD-ANX, and EPQN. Additionally, Sample 2 completed the ASI and the GI and pain symptom severity scales. From the database containing Sample 3, we extracted data from the VSI, BSQ-SF, HAD-ANX and ASI, and GI symptom severity from the subsample of individuals meeting Rome II diagnostic criteria.

Statistical Analysis
Factor analysis was used to assess internal consistency, reliability, and construct validity of the VSI. Divergent and convergent validity of the VSI were examined using bivariate correlations. Although there are potential pitfalls of stepwise regression (36), this method was considered the most appropriate technique for testing the specific hypothesis that the VSI would be the single best predictor among the other general measures of psychological distress. Specifically, stepwise logistic regression was applied regressing IBS diagnostic status onto the VSI and the general measures of psychological distress. To rule out the possibility that the results might have been due to selection of predictors for entry into the statistical model, we used stepwise regression in conjunction with an all possible subsets approach known as a dominance analysis (37). A dominance analysis is an empirical solution to determining the relative importance of predictors in multiple regression analysis. This statistical procedure is based on an examination of the R2 values for all possible subset models. A dominance analysis identifies the importance of predictors in a general sense as well as the predictor that adds the most to the prediction after controlling for any specific subset of the remaining variables.

The ability of the VSI to discriminate between known groups was examined by comparing subject groups differing in known health status. Known-groups validity compares mean scale scores across subject groups known to differ in the underlying health construct being investigated. The known groups validity approach is founded on the principle that certain specified groups of patients might be anticipated to score differently from others (24). To test the known-groups validity of the VSI, we examined mean scores for groups that are known to differ clinically: a) IBS positive (+) patients, individuals meeting Rome II diagnostic criteria who reported having seen either a GI specialist at any time or nonspecialist within the past year for their IBS symptoms; b) IBS+ nonpatients defined as Rome II positive individuals who did not endorse IBS health-care utilization; and c) nonIBS individuals who did not meet full Rome II diagnostic criteria for IBS. We examined the differences between these IBS groups because it was expected that these groups would have differing levels of GI symptom anxiety. Before analyses, information regarding Rome II diagnostic criteria was extracted from the BSQ-SF data for Samples 1 and 2 and was used to classify participants into two groups: a) IBS+ individuals meeting Rome II diagnostic criteria, and b) nonIBS individuals who did not endorse abdominal pain or discomfort during at least one day in at least 3 separate weeks in the last 3 months; a necessary, but not sufficient, condition for a Rome II diagnosis of IBS. Nineteen participants from Sample 1 and five participants from Sample 2 endorsed abdominal pain, but did not meet full diagnostic criteria for IBS. Because these participants could not reliably be classified into a "known" group, IBS+ or nonIBS, their data were not considered in this analysis.

Mediation analyses were conducted to determine the potential mediating effect of GI-specific anxiety on the relationships between affective variables and IBS symptoms. Specifically, path analyses via a structural equation modeling framework and bootstrap estimation of standard errors were performed in Amos version 4.0 (SPSS Inc., Chicago, IL). The significance of mediation was evaluated using the empirical sampling distribution of the product of the coefficients (the estimate of the intervening variable or the indirect effect, {alpha}ß) divided by the standard error of the indirect effect, {sigma}{alpha}ß (38). Similar to significance testing using the standard normal distribution, z' = {alpha}ß/{sigma}{alpha}ß is calculated and tested against the critical value from the empirical sampling distribution of the product of coefficients (H0: {alpha}ß/{sigma}{alpha}ß = 0). The empirical critical value at {alpha} = 0.05 is 0.97. This test is considered the most powerful method for testing intervening effects (38). In lieu of p values, 95% bias corrected confidence intervals are presented.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
DISCUSSION
 NOTES
 Appendix
 REFERENCES
 
Demographics
Characteristics of the final samples can be seen in Table 1. Consistent with prevalence estimates in college student samples (39), and comparable to general population estimates (40), 11% of the individuals met criteria for IBS in both samples. The ratio of females to males was approximately 2:1. Sample 3 reported, on average, an initial diagnosis of IBS at age 37 and a mean duration of about 12 years. A large proportion of participants (a little less than 50%) endorsed an Asian ethnicity. To examine the potential threat to external validity posed by combining data from individuals with distinct and strongly represented ethnicities, independent t-tests were used to test the homogeneity of the group on study variables. Specifically, potential differences between individuals endorsing Asian ethnicity and the next largest ethnic group, Caucasians, were examined. On average, independent t-tests detected no differences between Caucasian and Asian ethnic groups on the VSI, HAD-ANX, HAD-DEP, neuroticism, or percentage of individuals diagnosed with IBS. Within the subgroup of patients diagnosed with IBS, there were no differences in male to female ratio, characterization of IBS symptoms, psychological measures, or endorsement of nonGI pain between Asian and Caucasian ethnicity.


View this table:
[in this window]
[in a new window]

  		TABLE 1. Characteristics of the Samples

 

Reliability of the VSI
To determine the internal consistency of the VSI, Chronbach’s {alpha} was computed for each sample. The VSI demonstrated excellent internal consistency (Sample 1 Chronbach’s {alpha} = 0.92, Sample 2 Chronbach’s {alpha} = 0.90). Factor analysis supported the previously demonstrated uni-dimensional factor structure for the VSI in both samples (n = 505). That is, one main factor (eigenvalue = 7.2) accounted for 48% of the total variance in VSI scores. The range of factor loadings across all 15 items was greater than 0.60 for all items except item 2, which had a factor loading of 0.43. A second nonunique factor (eigenvalue = 1.1) accounted for 8% of the variance (23).

Divergent Validity
To assess divergent validity for the VSI, a correlation matrix was constructed for each sample. As can be seen in Table 2, the VSI demonstrated only moderate positive correlations with measures of state anxiety (HAD-ANX), fear of anxiety symptoms (ASI), neuroticism (EPQN), and depression (HAD-DEP), as well as a negative correlation with vitality (VIT). Although the VSI strongly demonstrates some relationship to relevant measures of psychological distress, the magnitude of these relationships suggests the VSI is measuring something distinctly different from these measures. Indeed, the correlation coefficient for the dichotomous variable, presence of a Rome II diagnosis for IBS, was strongest for the VSI in both samples, with higher scores associated with meeting diagnostic criteria for IBS. Notably, sex was not a significant correlate of VSI scores. To examine differences between IBS and nonIBS groups in the relationship between GI-specific anxiety (VSI) and state anxiety (HAD-ANX), fear of anxiety symptoms (ASI), depression (HAD-DEP), and neuroticism (EPQN), bivariate correlations were computed for each sample by IBS diagnosis (see Table 3). As can be seen, the largest correlations were demonstrated in the IBS recruited sample. To further elucidate the potential differences in the magnitude of the correlations by IBS group, Samples 1 and 2 were combined and bivariate correlations between the VSI and the psychosocial measures were computed. Next, to allow direct comparisons of the IBS groups, bivariate correlations were transformed into effect sizes using Fisher’s Zr-transform and 95% normal confidence intervals were computed (see Table 4). Because small sample sizes decrease the precision of effect size estimates, we provide the estimates from the IBS+ groups from Samples 1 and 2 (n = 54), but place a caution on interpretation of any differences. As can be seen in Table 4, the magnitude of the relationship between the VSI and the psychosocial distress measures was significantly greater in the IBS+ group compared with the nonIBS group. Of particular interest, the correlation between the VSI and state anxiety were significantly different between groups (IBS+ Sample 3 > IBS+ Sample 2 > nonIBS Sample 2). These correlations demonstrate both divergence between the VSI and the other psychosocial measures and the stronger overlap of GI-specific anxiety and general anxiety in IBS subjects.


View this table:
[in this window]
[in a new window]

  		TABLE 2. Correlation Matrices for Construct Validation

 

View this table:
[in this window]
[in a new window]

  		TABLE 3. Bivariate Correlations Between the VSI and Psychological Distress Measures

 

View this table:
[in this window]
[in a new window]

  		TABLE 4. Effect Sizes Characterizing the Relationship Between the VSI and the Psychological Distress Measures by IBS Group

 

Concurrent Validity

IBS Diagnosis
To further elucidate the importance of the VSI as an explanatory variable of IBS diagnosis in each sample, the dichotomous indicator variable, presence of a Rome II IBS diagnosis (0 = no, 1 = yes), was regressed onto the VSI and the other available measures in Sample 1 and 2 using stepwise logistic regression. For Sample 1, the presence of IBS was regressed onto the VSI, HAD-ANX, HAD-DEP, EPQN, and VIT. The VSI was the sole independent predictor of presence of IBS explaining 20% (Nagelkerke’s R2) of the variance in diagnostic status (ßVSI = 0.07 [0.01], p < .001). Similarly, for Sample 2, regressing the presence of IBS onto the VSI, HAD-ANX, EPQN, and ASI revealed that the VSI was the sole predictor of diagnostic status accounting for 31% of the variance in diagnostic status (ßVSI = 0.11 [0.02], p < .001). To further demonstrate the relative importance of the predictors and to rule out the possibility of alternative results due to selection predictors for entry into the statistical model, a dominance analysis was performed in both samples. Results from this analysis support the VSI as the dominant predictor of IBS diagnostic status in both samples (see Appendices 1 and 2).

Symptom Severity
To determine the relationship between VSI scores and symptom severity, we examined the relationship between GI symptom severity and VSI scores for individuals meeting Rome II diagnostic criteria for IBS (n = 23) in Sample 2.1 Consistent with the results from the initial validation study, the bivariate correlation between symptom severity and VSI score was strong, r = 0.43, p < .05. Of interest, the mean GI symptom severity rating was moderate in this IBS+ group (M = 10.6, SE = 0.98) and comparable to the symptom severity levels reported by the IBS+ group from Sample 3 (M = 11.3, SE = 0.47).

In Sample 2, 135 individuals endorsed nonGI pain and reported a mean symptom severity of 8.3 (SE = 0.55). As predicted, the VSI was less related to nonGI symptom severity for subjects endorsing nonGI pain, r = 0.17, p = .051.

Does GI-Specific Anxiety Mediate the Relationship Between State Anxiety and GI Symptom Severity?
In the initial validation sample, GI-specific anxiety was found to mediate the relationship between state anxiety and GI symptom severity and we attempted to replicate this analysis in the community IBS+ group from Sample 2. As can be seen in Figure 1, the impact of state anxiety, as measured by the HAD-ANX, on symptom severity through the mediator, GI-specific anxiety, was significant, {alpha}ß = 0.43 (95% bias-corrected confidence interval, 0.10–1.30). The test for intervening effects indicated that the mediating effect for GI-specific anxiety was statistically significant, z' = 1.64 and accounted for 98% of the effect of state anxiety on symptom severity. The direct effect of state anxiety was not statistically significant, c' = 0.01 (–1.24–0.54). The model accounted for 22% of the variance in GI symptom severity. In this small sample of 23 IBS+ patients, path analyses replicated the results of the initial validation study and provide further support for the notion that the effect of state anxiety on GI symptom severity is mediated by GI-specific anxiety.


Figure 115
View larger version (11K):
[in this window]
[in a new window]

  		Figure 1. Mediation models for the intervening variable models. Unstandardized path coefficients are denoted in parentheses and mediated effects in brackets.

 

Does GI-Specific Anxiety Mediate the Relationship Between Anxiety Sensitivity and GI Symptom Severity?
In the initial validation sample, GI-specific anxiety was also found to mediate the relationships between the ASI and GI symptom severity, and this analysis was replicated using the IBS+ group from Sample 2. As can be seen in Figure 1B, the direct effect of ASI on symptom severity controlling for GI-specific anxiety was not significant, c' = –0.09 (–0.29–0.17). However, the impact of the ASI on symptom severity through the mediator, GI-specific anxiety, was significant, {alpha}ß = 0.11 (0.02–0.27). The test for intervening effects indicated significant mediating effects for GI-specific anxiety, z' = 2.02 (95% critical value = 0.97). Although the coefficient for the direct effect estimate was negative, neither examination of the confidence intervals nor significance testing indicated evidence of significant suppression effects.

Does GI-Specific Anxiety Mediate the Relationship Between Neuroticism and GI Symptom Severity?
We examined GI-specific anxiety as a potential mediator of the relationship between neuroticism and GI symptom severity in the IBS+ group from Sample 2 (n = 23). As can be seen in Figure 1C, visceral anxiety completely mediated the relationship between neuroticism and symptom severity, {alpha}ß = 0.45 (0.12–0.94). Again, the test for intervening effects indicated that the mediating effect for GI-specific anxiety was statistically significant, z' = 2.58. The direct effect of state anxiety was not statistically significant, c' = –0.32 (–0.99–0.30). This model explained 27% of the variance in symptom severity. In this small sample, path analyses tentatively suggested that the relationship between neuroticism and GI symptom severity is completely mediated by GI-specific anxiety.

Does GI-Specific Anxiety Mediate the Relationship Between Neuroticism and IBS Diagnosis?
We combined Samples 1 and 2 (n = 493) to test GI-specific anxiety as a potential mediator between neuroticism and IBS diagnosis. Indeed, as seen in Figure 1D, visceral anxiety mediated the relationship between neuroticism and IBS diagnosis, {alpha}ß = 0.01 (0.01–0.02). Again, the test for intervening effects indicated that the mediating effect for GI-specific anxiety was statistically significant, z' = 5.58. The direct effect of neuroticism was not statistically significant, c' = 0.00 (–0.01–0.01). The model accounted for only 16% of the variance in IBS diagnosis. Arguably, the mediation was quite small, however, the analysis supported the notion that any relationship between neuroticism and IBS diagnosis is completely mediated by GI-specific anxiety.

Discriminant Validity
A total of 136 individuals met Rome II diagnostic criteria for IBS. In the group of IBS+ patients and nonpatients (n = 136), all individuals reported some type of information regarding health-care utilization. Approximately 38% (33/86) reported visiting a physician for GI problems in the past year and about 44% (47/105) reported having had consulted a gastroenterologist at some time.

To examine the discriminative nature of the VSI, we examined the mean scores in three "known" groups drawn across the 3 samples: nonIBS (n = 451), IBS+ patients (n = 54), and IBS+ nonpatients (n = 82). IBS+ patients were individuals meeting Rome II diagnostic criteria who reported having seen either a GI specialist at any time or nonspecialist within the past year for their IBS symptoms. IBS+ nonpatients were defined as Rome II positive individuals who did not endorse IBS health-care utilization. Table 5 presents the means and 95% confidence intervals of the VSI and the other psychological distress measures in the three "known" groups. Scores on the VSI discriminated the three groups (IBS+ patients > IBS+ nonpatients > nonIBS). Not surprisingly, scores on the VSI were slightly higher for the IBS+ patient as compared with the IBS+ nonpatient group. Scores on the EPQN were greater in the IBS+ groups versus nonIBS group. Scores on the HAD-ANX, EPQN, HAD-DEP, and ASI did not differentiate the IBS+ groups. The VIT measure did not discriminate between the nonIBS and IBS+ groups. Of particular note, the analysis of the EPQN and VIT was limited due to the small sample sizes in the IBS+ groups and should be interpreted with caution.


View this table:
[in this window]
[in a new window]

  		TABLE 5. Means and 95% Confidence Intervals for the VSI and Psychological Distress Measures in VSI Validation Samples

 


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
NOTES
 Appendix
 REFERENCES
 
The current study provided further evidence for the sound psychometric properties of the VSI, a newly developed measure for the assessment of GI-specific anxiety. VSI scores were examined in several "known" groups (IBS+ and nonIBS groups, as well as patient and nonpatient IBS+ groups) from data collected from two surveys of approximately 500 undergraduate students and the initial VSI validation sample. Our main findings were a) the VSI was the key predictor of presence of IBS accounting for 20% to 31% of the variance in IBS diagnostic status; b) GI-specific anxiety, as assessed by the VSI, was a significant mediator of the relationship between both neuroticism and anxiety symptoms and IBS symptom severity; c) the VSI was less related to nonGI symptoms, depression symptoms, and vitality; and d) the VSI discriminated IBS patients from IBS nonpatients and nonIBS controls. Overall, the VSI demonstrated excellent psychometric properties when administered to nonIBS, nonpatient IBS, and patient IBS samples, providing further support for its use as a survey instrument and an outcome measure in pharmacological and psychological treatment trials. In addition, the VSI may be a valuable cognitive marker for evaluation of pathophysiologic mechanisms in IBS patients.

The results demonstrate that the construct of GI-specific anxiety was adequately assessed in two predominantly nonIBS samples. Specifically, the VSI retained its original uni-dimensional factor structure and demonstrated excellent internal consistency. The VSI successfully discriminated between nonIBS, nonpatient IBS+, and patient IBS+ groups. As expected, VSI scores were highest for individuals meeting Rome II diagnostic criteria for IBS who also endorsed IBS health-care seeking.

Good divergent validity was also demonstrated. As hypothesized, in a small group of relatively young IBS+ patients, the VSI was related to GI, but not nonGI, pain severity. The findings also indicated that the VSI measures a unique construct, distinct from measures of state anxiety, fear of panic/anxiety symptoms, neuroticism, depression, and vitality. As anticipated, GI-specific anxiety generalized to global anxiety in IBS+ groups, however, in nonIBS groups GI-specific anxiety was distinct from other fears or worries. That is, an effect size comparison demonstrated that the magnitude of the relationship between the VSI and the psychological distress measures was significantly stronger for the IBS+ as compared with the nonIBS groups. Indeed, logistic regression indicated that in the presence of the other psychological distress measures, the VSI was the key predictor of IBS status accounting for 20% to 31% of the variance in diagnostic status.

These findings are consistent with the hypothesis that GI-specific anxiety is the mechanism through which other measures of psychological distress such as neuroticism, anxiety sensitivity, and state anxiety relate to IBS presentation. Consistent with previous findings (16), neuroticism demonstrated a small but significant relationship with IBS diagnostic status (ß = 0.14 [0.10–0.18], n = 453). As predicted, GI-specific anxiety completely mediated this relationship replicating the results from the initial VSI validation study (23). GI-specific anxiety was also a significant mediator of the relationship between neuroticism and IBS symptom severity. Overall, these findings support the notion that neuroticism is a vulnerability factor for IBS symptoms (16), but that GI-specific anxiety mediates this relationship and serves as an IBS-specific expression of neuroticism. That is, GI-specific anxiety may arise when neuroticism interacts with specific learning experiences (i.e., GI symptoms are harmful). Our findings are in line with the findings of Goubert and colleagues, who found that the effects of neuroticism on pain vigilance (and pain severity) in chronic lower back pain patients was largely mediated by pain catastrophizing and pain-related fear (6).

GI-specific anxiety also mediated the relationship between the anxiety measures (state anxiety and anxiety sensitivity) and GI symptom severity. Thus, our findings support hypotheses of disease-specific expressions of anxiety sensitivity and suggest that generalized measures of anxiety symptoms are important in disorders like IBS only when accompanied by assessment of GI-specific anxiety (7). In sum, GI-specific anxiety and not general anxiety or neuroticism is the best available marker of the characteristic cognitive and affective processes in IBS.

Because patients who engage in health-care seeking behavior tend to report high psychological distress (41,42), the general survey known-group approach was a particular strength of the study. However, this approach also necessitated that IBS diagnosis be made without confirmatory data on the lack of organic findings which may have led to inclusion of a small number of cases that would not fit the clinical Rome II criteria (43). Additionally, the scope of inferences from this study is limited to a young adult university student population. Although none of the students knew the hypotheses of the study or the content of the survey when agreeing to participate, they did voluntarily fill out the survey, and therefore self-selection may also have influenced the resulting sample characteristics. The sample, while seemingly representative of the UCLA undergraduate population did have a very high percentage of young adults endorsing Asian ethnicity. However, analyses did not indicate that combining data across Asian and Caucasian ethnicity threatened the external validity or generalization of the results. Although differences in the epidemiology of IBS have failed to be supported with emerging Asian studies (44), evidence does suggest that the symptom presentation of IBS for Asian versus Western civilizations may be different and characterized by a greater frequency of upper abdominal pain and defecatory symptoms perceived as being less bothersome (45). Incidence of depression and anxiety (46–48) and the impact of health-related quality of life (2) does not appear to differ between Asian and Caucasian ethnicity. This study found no differences on the major supporting symptom characteristics, male to female ratio, or health-care seeking between young Asian and Caucasian adults attending a university in the United States and meeting the diagnostic criteria for IBS. However, this study was not designed to assess these differences and provides only tentative evidence regarding the similarities of IBS presentation across young Asian and Caucasian adults attending a university in the United States. Arguably, generalization of the results for less represented minorities such as African American and Latino is questionable.

The sample utilized for the mediation analyses of GI symptom severity was small and replication in a larger sample is necessary. Additionally, evidence for the predictive and mediator role of GI-specific anxiety in determining IBS diagnosis relied on correlation and covariance techniques which cannot establish causal links definitively. Clearly, a prospective study is required to assess the predictive capabilities of the VSI. Based on studies of early life trauma, life stress, and co-morbidity of mood disorders in IBS the present study examined a theoretical model of trait and symptom-specific anxiety as mechanisms that may maintain or exacerbate IBS symptoms. It is also possible and not mutually exclusive with the current findings that symptom severity may to some extent affect GI-specific anxiety and general anxiety (but probably not trait anxiety), although this nonrecursive model was not specifically tested.

Overall, the results provided evidence that the concept of GI-specific anxiety is relevant to a relatively young population and can be reliably measured using the VSI. In addition, our results strengthened the evidence that GI-specific anxiety may promote, exacerbate, and maintain the symptoms of IBS, consistent with current hypotheses regarding the role of central hyperexcitability in these disorders (29). Future studies are needed to assess the role of GI-specific anxiety in other functional GI disorders and in organic GI conditions such as inflammatory bowel disease. In addition, further study of the VSI as an outcome measure in pharmacological and psychological treatment trials is needed to verify its sensitivity to change.

We would like to thank Teresa Olivas for her editorial assistance in the preparation of this manuscript.


   Appendix
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 NOTES
 Appendix
REFERENCES
 


View this table:
[in this window]
[in a new window]

  		APPENDIX 1. Dominance Analysis for Predictors of IBS Diagnostic Status From Sample 1.

 


View this table:
[in this window]
[in a new window]

  		APPENDIX 2. Dominance Analysis for Predictors of IBS Diagnostic Status From Sample 2

 


   NOTES
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 NOTES
Appendix
 REFERENCES
 
1Unfortunately, a measure of GI symptom severity was not obtained in Sample 1. Back

Received for publication July 15, 2005; revision received August 31, 2006.

Supported in part by grants P50 DK64539, R24 AT002681, NR 04881, and VA Medical Research.

DOI:10.1097/PSY.0b013e31802e2f24


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 NOTES
 Appendix
 REFERENCES
 

  1. Mayer EA, Craske MG, Naliboff BD. Depression, anxiety and the gastrointestinal system. J Clin Psychiatry 2001;62:28–36.
  2. Gralnek IM, Hays RD, Kilbourne A, Naliboff B, Mayer EA. The impact of irritable bowel syndrome on health-related quality of life. Gastroenterology 2000;119:654–60.[CrossRef][Medline]
  3. Reiss S, McNally RJ. The expectancy model of fear. In: Reiss S, Bootzon RR, eds. Theoretical Issues in Behavior Therapy. New York: Academic Press; 1985;107–21.
  4. Reiss S, Peterson RA, Gursky DM, McNally RJ. Anxiety sensitivity, anxiety frequency and the predictions of fearfulness. Behav Res Ther 1986;24:1–8.[CrossRef][Medline]
  5. Asmundson GJ, Norton PJ, Norton GR. Beyond pain: the role of fear and avoidance in chronicity. Clin Psychol Rev 1999;19:97–119.[CrossRef][Medline]
  6. Goubert L, Crombez G, Van Damme S. The role of neuroticism, pain catastrophizing and pain-related fear in vigilance to pain: a structural equations approach. Pain 2004;107:234–41.[CrossRef][Medline]
  7. Taylor S, Cox BJ. Anxiety sensitivity: multiple dimensions and hierarchic structure. Behav Res Ther 1998;36:37–51.[CrossRef][Medline]
  8. Reiss S. Expectancy model of fear, anxiety, and panic. Clin Psychol Rev 1991;11:141–53.
  9. Reiss S. Trait anxiety: it’s not what you think it is. J Anxiety Disord 1997;11:201–14.[CrossRef][Medline]
  10. Taylor S. Nature and measurement of anxiety sensitivity: reply to Lilienfeld, Turner, and Jacob (1996). J Anxiety Disord 1996;10:425–51.[CrossRef]
  11. Asmundson GJ. Commentary: anxiety, sensitivity and the pain experience. Eur J Pain 2001;5:23–5.[CrossRef][Medline]
  12. Keogh E, Cochrane M. Anxiety sensitivity, cognitive biases, and the experience of pain. J Pain 2002;3:320–9.[CrossRef][Medline]
  13. Keogh E, Mansoor L. Investigating the effects of anxiety sensitivity and coping on the perception of cold pressor pain in healthy women. Eur J Pain 2001;5:11–22.[CrossRef][Medline]
  14. Keogh E, Birkby J. The effect of anxiety sensitivity and gender on the experience of pain. Cognition and Emotion 1999;13:813–29.[CrossRef]
  15. Norton GR, Norton PJ, Asmundson GJ, Thompson LA, Larsen DK. Neurotic butterflies in my stomach: the role of anxiety, anxiety sensitivity and depression in functional gastrointestinal disorders. J Psychosom Res 1999;47:233–40.[CrossRef][Medline]
  16. Talley NJ, Boyce PM, Jones M. Is the association between irritable bowel syndrome and abuse explained by neuroticism? A population based study. Gut 1998;42:47–53.[Abstract/Free Full Text]
  17. Cox BJ, Parker JD, Swinson RP. Anxiety sensitivity: confirmatory evidence for a multidimensional construct. Behav Res Ther 1996;34:591–8.[CrossRef][Medline]
  18. Cox BJ. The nature and assessment of catastrophic thoughts in panic disorder. Behav Res Ther 1996;34:363–74.[CrossRef][Medline]
  19. McCracken LM, Zayfert C, Gross RT. The Pain Anxiety Symptoms Scale: development and validation of a scale to measure fear of pain. Pain 1992;50:67–73.[CrossRef][Medline]
  20. Kori SH, Miller RP, Todd DD. Kinesiophobia: a new view of chronic pain behavior. Pain Manage 1990;1:35–43.
  21. Asmundson GJ, Norton GR. Anxiety sensitivity in patients with physically unexplained chronic back pain: a preliminary report. Behav Res Ther 1995;33:771–7.[CrossRef][Medline]
  22. Greenberg J, Burns JW. Pain anxiety among chronic pain patients: specific phobia or manifestation of anxiety sensitivity? Behav Res Ther 2003;41:223–40.[CrossRef][Medline]
  23. Labus J, Bolus R, Chang L, Wiklund I, Naesdal J, Mayer E, Naliboff B. The visceral sensitivity index: development and validation of a gastrointestinal symptom-specific anxiety scale. Aliment Pharmacol Ther 2004;20:89–97.[Medline]
  24. Kerlinger FN, Lee HB. Foundations of Behavioral Research. 4th ed. Fort Worth, TX: Harcourt College Publishers; 2000.
  25. Drossman DA, Corazziari E, Talley NJ, Thompson WG, Whitehead WE. ROME II. The Functional Gastrointestinal Disorders. Diagnosis, Pathophysiology and Treatment: A Multinational Consensus. 2nd ed. McLean, VA: Degnon Associates; 2000.
  26. Fass R, Longstreth GF, Pimentel M, Fullerton S, Russak SM, Chiou CF, Reyes E, Crane P, Eisen G, McCarberg B, Ofman J. Evidence- and consensus-based practice guidelines for the diagnosis of irritable bowel syndrome. Arch Intern Med 2001;161:2081–8.[Abstract/Free Full Text]
  27. Ware JE, Sherbourne C. The MOS 36-item short-form health survey (SF-36) I: Conceptual framework and item selection. Med Care 1992;30:473–83.[Medline]
  28. Ware JE, Snow KK, Kosinski M, Gandek B. SF-36 Health Survey. Manual and Interpretation Guide. Boston: The Health Institute, New England Medical Center; 1993.
  29. Mayer EA. The neurobiology of stress and gastrointestinal disease. Gut 2000;47:861–9.[Free Full Text]
  30. Zigmond AS, Snaith RP. The hospital anxiety and depression scale. Acta Psychiatr Scand 1983;67:361–70.[Medline]
  31. Peterson RA, Reiss S. Anxiety Sensitivity Index manual. Palo Heights: International Diagnostic Systems, 1987.
  32. Reiss S. Trait anxiety: it’s not what you think it is. J Anxiety Disorder 1997;11:201–214.[CrossRef][Medline]
  33. Eysenck SB, Eysenck HJ, Barrett P. A revised version of the Psychoticism Scale. Pers Individ Dif 1985;6:21–9.
  34. Gwee KA, Graham JC, McKendrick MW, Collins SM, Marshall JS, Walters SJ, Read NW. Psychometric scores and persistence of irritable bowel after infectious diarrhoea. Lancet 1996;347:150–3.[CrossRef][Medline]
  35. Eysenck HJ, Eysenck MW. Personality and Individual Differences: A Natural Science Approach. New York: Plenum Press; 1985.
  36. Thompson B. Stepwise regression and stepwise discriminant analysis need not apply here: a guidelines editorial. Educ Psychol Meas 1995;55:525–34.[Abstract]
  37. Azen R, Budescu DV. The dominance analysis approach for comparing predictors in multiple regression. Psychol Methods 2003;8:129–48.[CrossRef][Medline]
  38. MacKinnon DP, Lockwood CM, Hoffman JM, West SG, Sheets V. A comparison of methods to test mediation and other intervening variable effects. Psychol Methods 2002;7:83–104.[CrossRef][Medline]
  39. Hazlett-Stevens H, Craske MG, Mayer EA, Chang L, Naliboff BD. Prevalence of irritable bowel syndrome among university students: the roles of worry, neuroticism, anxiety sensitivity and visceral anxiety. J Psychosom Res 2003;55:501–5.[CrossRef][Medline]
  40. Thompson WG, Longstreth GF, Drossman DA, Heaton KW, Irvine EJ, Müller-Lissner SA. C. Functional bowel disorders and D. functional abdominal pain. In: Drossman DA, Corazziari E, Talley NJ, Thompson WG, Whitehead WE, eds. Rome II. The Functional Gastrointestinal Disorders. Diagnosis, Pathophysiology and Treatment: A Multinational Consensus. McLean: Degnon Associates; 2000;351–432.
  41. Drossman DA, McKee DC, Sandler RS, Mitchell CM, Cramer EM, Lowman BC, Burger AL. Psychosocial factors in the irritable bowel syndrome. A multivariate study of patients and nonpatients with irritable bowel syndrome. Gastroenterology 1988;95:701–8.[Medline]
  42. Whitehead WE, Bosmajian L, Zonderman AB, Costa PT Jr. Schuster MM. Symptoms of psychologic distress associated with irritable bowel syndrome. Comparison of community and medical clinic samples. Gastroenterology 1988;95:709–14.[Medline]
  43. Cash BD, Chey WD. Irritable bowel syndrome-an evidence-based approach to diagnosis. Aliment Pharmacol Ther 2004;19:1235–45.[CrossRef][Medline]
  44. Cremonini F, Talley NJ. Irritable bowel syndrome: epidemiology, natural history, health care seeking and emerging risk factors. Gastroenterol Clin North Am 2005;34:189–204.[CrossRef][Medline]
  45. Gwee KA. Irritable bowel syndrome in developing countries - a disorder of civilization or colonization? Neurogastroenterol Motil 2005;17:317–24.[CrossRef][Medline]
  46. Zhang AY, Snowden LR. Ethnic characteristics of mental disorders in five U.S. communities. Cultur Divers Ethnic Minor Psychol 1999;5:134–46.[Medline]
  47. Lam CY, Pepper CM, Ryabchenko KA. Case identification of mood disorders in Asian American and Caucasian American college students. Psychiatr Q 2004;75:361–73.[CrossRef][Medline]
  48. Gee CB. Assessment of anxiety and depression in Asian American youth. J Clin Child Adolesc Psychol 2004;33:269–71.[CrossRef][Medline]



This article has been cited by other articles:


Home page
 Psychosom. Med.Home page
B. D. Naliboff, A. M. Waters, J. S. Labus, L. Kilpatrick, M. G. Craske, L. Chang, H. Negoro, H. Ibrahimovic, E. A. Mayer, and E. Ornitz
Increased Acoustic Startle Responses in IBS Patients During Abdominal and Nonabdominal Threat
Psychosom Med, October 1, 2008; 70(8): 920 - 927.
[Abstract] [Full Text] [PDF]

Home page
 GutHome page
E A Mayer, S Bradesi, L Chang, B M R Spiegel, J A Bueller, and B D Naliboff
Functional GI disorders: from animal models to drug development
Gut, March 1, 2008; 57(3): 384 - 404.
[Abstract] [Full Text] [PDF]

Home page
 J. Neurosci.Home page
S. M. Berman, B. D. Naliboff, B. Suyenobu, J. S. Labus, J. Stains, G. Ohning, L. Kilpatrick, J. A. Bueller, K. Ruby, J. Jarcho, et al.
Reduced Brainstem Inhibition during Anticipated Pelvic Visceral Pain Correlates with Enhanced Brain Response to the Visceral Stimulus in Women with Irritable Bowel Syndrome
J. Neurosci., January 9, 2008; 28(2): 349 - 359.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Labus, J. S.
Right arrow Articles by Naliboff, B. D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Labus, J. S.
Right arrow Articles by Naliboff, B. D.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS