Heart-Focused Anxiety in Patients With Noncardiac Chest Pain

Abstract

Heart-focused anxiety (HFA) is a fear of cardiac sensations driven by worries of physical health catastrophe. HFA is impairing and distressing and has been shown to disproportionately affect individuals with noncardiac chest pain (NCCP), chest pain that persists in the absence of an identifiable source. The Cardiac Anxiety Questionnaire (CAQ) is a measure designed to assess HFA. The aim of this study was to evaluate the psychometric properties and factor structure of the CAQ in a sample of 229 adults diagnosed with NCCP. Results demonstrated that the CAQ is a useful measure of HFA in patients with NCCP and that a four-factor model including fear of cardiac sensations, avoidance of activities that elicit cardiac sensations, heart-focused attention, and reassurance seeking was the best fit for the data. Additionally, associations between CAQ subscales and two measures of health-related behaviors—pain-related interference and health care utilization—provided evidence of concurrent validity. Treatment implications are also discussed.

Keywords

Cardiac Anxiety Questionnaire heart-focused anxiety noncardiac chest pain pain-related impairment health care utilization

Heart-focused anxiety (HFA) is a fear of cardiac sensations driven by worries of physical health catastrophe (e.g., heart attack; Eifert, 1992). These worries are often chronic (Tew et al., 1995) and are related to increased pain frequency and intensity (Fleet & Beitman, 1998; Zvolensky, Eifert, Feldner, & Leen-Feldner, 2003). Individuals with HFA commonly avoid activities associated with cardiac arousal (e.g., exercise, sex; Eifert, 1992), which works to maintain anxiety and may constitute a substantial health risk in its own right (Eifert, Zvolensky, & Lejuez, 2000). Furthermore, cardiac-related fears are associated with elevated levels of health care utilization, regardless of an identifiable cardiac disease (Aikens, Michael, Levin, & Lowry, 1999; Realini & Katerndahl, 1993; Strik, Denollet, Lousberg, & Honig, 2003) and may contribute to excessive medical costs.

HFA overlaps with a range of mental and physical health conditions (Zvolensky, Feldner, Eifert, Vujanovic, & Solomon, 2008). However, this distinct form of anxiety is characterized by worry specific to cardiac functioning as opposed to the broader health concerns typical of hypochondriasis (Barsky et al., 2001) or the generalized misinterpretation of bodily arousal central to panic disorder (Barlow, 1988). To this point, empirical findings have demonstrated that individuals with HFA are uniquely vigilant to and exceptionally fearful of cardiac sensations (Eifert, Hodson, Tracey, Seville, & Gunawardane, 1996; White, Craft, & Gervino, 2010).

Fear of cardiac symptoms was initially identified as a component of anxiety sensitivity (AS; Taylor & Cox, 1998)—the global fear of physical anxiety symptoms based on beliefs that they negatively affect social, physical, and psychological functioning (Reiss & McNally, 1985). However, HFA may be more accurately viewed as the theoretical evolution of that AS lower order factor as it more fully accounts for the behavioral, cognitive, and affective aspects of cardiac-centric worry (Eifert, Zvolensky, et al., 2000).

Research on HFA has in large part been spurred by the development of the Cardiac Anxiety Questionnaire (CAQ; Eifert, Thompson, et al., 2000). In addition to being utilized by studies focusing on HFA (e.g., Fischer et al., 2012; Hoyer et al., 2008; Van Beek, Mingles, et al., 2012), the CAQ has served as a measure of treatment outcome in trials of medical (e.g., Aicher, Holz, Feldner, Kollner, & Schafers, 2011) and psychological interventions (e.g., Pelland et al., 2011; Spinhoven, Van der Does, Van Dijk, & Van Rood, 2010).

Initial psychometric evaluation of the CAQ involving 188 cardiology patients revealed three underlying factors: (a) fear about heart functioning, (b) avoidance of activities believed to elicit cardiac symptoms, (c) and heart-focused attention (Eifert, Thompson, et al., 2000). These lower order factors have allowed for a more detailed understanding of HFA. For example, there is evidence that heart-focused attention and fear (but not avoidance) predict chest pain severity in patients with coronary artery disease (Zvolensky, Eifert, et al., 2003). Other findings have illustrated that in individuals self-referred for coronary calcium screening, those without signs of coronary artery calcium endorsed higher levels of heart-focused attention than those identified as having coronary artery calcium (Carmin, Wiegartz, Hoff, & Kondos, 2003).

Subsequent validity studies of the CAQ have indicated that HFA is more precisely described by a four-factor solution. This updated model includes the three originally proposed factors as well as a fourth “reassurance-seeking” factor. In a study of 658 individuals undergoing cardiac evaluation, the four-factor model proved to be a better fit in patients with and without significant coronary atherosclerosis (Marker, Carmin, & Ownby, 2008). Investigation of a Dutch version of the CAQ involving 237 cardiac patients also found support for a reassurance-seeking factor (Van Beek, Voshaar, et al., 2012). These findings fit with current theory, which holds that reassurance seeking plays a key role in the maintenance of HFA. The transient reduction of anxiety that results from reassurance seeking may actually perpetuate cardiac fears by strengthening maladaptive illness beliefs and reinforcing the value of ineffective safety behaviors (Eifert, Zvolensky, et al., 2000).

HFA has been proposed as a key factor in the development and persistence of noncardiac chest pain (NCCP) or chest pain without a detectible cardiac origin (Eifert, 1992). Elevated HFA in patients with NCCP is associated with higher reported levels of pain, functional impairment (Eifert et al., 1996), and health care seeking (Eslick, Jones, & Talley, 2003). Moreover, HFA has proven amenable to treatment. In a trial of cognitive behavioral therapy for NCCP, treatment participants evidenced significant reductions in HFA, which in turn mediated decreases in chest pain (Spinhoven et al., 2010). Given the importance of HFA to NCCP theory and treatment, a more in-depth understanding of the function of lower order HFA factors is needed. In addition, as NCCP patients are disproportionately affected by HFA, it is crucial that HFA assessment instruments be validated in this patient group.

The aim of this study was to evaluate the construct validity of the CAQ. First, we examined the factor structure, internal consistency, test–retest reliability, and convergent and discriminant validity of the CAQ in a sample of adult patients diagnosed with NCCP. With regard to the factor structure, we investigated whether the addition of a reassurance-seeking factor would result in a stronger model than the one originally proposed by Eifert, Thompson, et al. (2000). Additionally, we appraised concurrent validity by examining the relation of CAQ subscales to two health-related behavioral indices—pain-related interference and health care utilization.

Method

Participants

The sample consisted of 229 patients undergoing cardiac evaluation at a university-affiliated medical center as part of a larger study on the clinical course of NCCP. Eligible participants met the following inclusion criteria: (a) chief complaint of chest pain or discomfort; (b) complete cardiac evaluation, including a general physical exam and exercise stress test that revealed no abnormalities; (c) at least 18 years of age; and (d) English fluency. Average age of participants was 50 years (SD = 10.3). Approximately half of the sample was female (56%), married (56%), and working full time (58%). The group was predominantly Caucasian (83%) and well educated (62% with a bachelor’s degree; 98% with a high school diploma). Participants endorsed a history of various medical conditions, including asthma (15%), diabetes (9%), cancer (7%), and hypertension (33%). In addition, 11% of the sample reported smoking cigarettes, and 55% reported regularly exercising.

Measures

Demographics and Medical History

Demographic and medical history information, including details regarding chest pain, was collected by self-report questionnaire. Of the chest pain characteristics assessed here, frequency of chest pain, duration of chest pain condition, and number of chest pain–specific doctor visits in the past year were included as study variables. Specifically, participants answered, “How often do you experience the chest pain?” by selecting from a 6-point scale (1 = several times per day; 6 = never or rarely). Participants responded to the question, “How long have you had the pain?” on a 5-point scale (1 = seven days or less; 5 = more than 1 year). Last, participants responded to the question, “In the past year, how many doctor visits did you have because of chest pain?” on a 5-point scale (1 = never; 5 = several times a month).

Cardiac Anxiety Questionnaire

The CAQ (Eifert, Thompson, et al., 2000) is an 18-item self-report questionnaire that assesses HFA. Participants are asked to indicate how often they currently experience certain heart-related concerns on a 5-point Likert-type scale (0 = never; 4 = always). Higher scores indicate greater HFA. This measure has demonstrated good psychometric properties in a number of cardiac patient samples (Eifert, Thompson, et al., 2000; Van Beek, Voshaar, et al., 2012).

West Haven Multidimensional Pain Inventory

The West Haven Multidimensional Pain Inventory (MPI; Kerns, Turk, & Rudy, 1985) is a self-report questionnaire that consists of 61 items assessing social, psychological, and behavioral consequences of pain. Items are divided into three domains including, (a) perceptions of pain and impact of pain on daily life, (b) pain-related social experience, and (c) influence of pain on daily activities. The MPI has been used extensively with a variety of medical populations (Bernstein, Jaremko, Hinkley, 1995; Piotrowski, 1998) and has demonstrated good internal consistency, test–retest reliability, and convergent validity (Kerns et al., 1985; Thompson, 1990). The MPI has also been noted to have strong face validity and patient acceptability (Bernstein et al., 1995). This study utilized the pain-related interference scale from the first domain.

Body Vigilance Scale

The Body Vigilance Scale (BVS; Schmidt, Lerew, & Trakowski, 1997) is a four-item self-report scale that measures attentional focus to internal bodily sensations. The first three items measure degree of attentional focus, perceived sensitivity to changes in bodily sensations, and average time spent attending to body sensations on an 11-point Likert-type scale (0 = not at all like me; 10 = extremely like me). Item 4 asks participants to rate the extent to which they attend to 15 anxiety-related bodily sensations on an 11-point Likert-type scale (0 = none; 10 = extreme). Responses to Item 4 are averaged to yield a single score, and BVS total score is calculated by summing all four items. The BVS has been shown to demonstrate good internal consistency and convergent and discriminant validity. The single-factor structure of the BVS has been confirmed, and BVS scores have been found to predict rates of health care use. Furthermore, these findings have been replicated in clinical and nonclinical samples (Olatunji, Deacon, Abramowitz, & Valentiner, 2007; Schmidt et al., 1997).

Anxiety Sensitivity Index

The Anxiety Sensitivity Index (ASI; Reiss, Peterson, Gursky, & McNally, 1986) is a 16-item self-report scale measuring fear of potential negative implications of anxiety-related bodily sensations. Respondents indicate the degree of their fear on a 5-point Likert-type scale (0 = very little; 4 = very much). The scale consists of three subscales (i.e., physical, cognitive, and social concerns) each representing a lower order factor of AS (Zinbarg, Brown, Barlow, & Rapee, 2001). The ASI has been widely used and has demonstrated good temporal stability (Rodriguez, Bruce, Pagano, Spencer, & Keller, 2004), internal consistency, incremental validity, and convergent and discriminant validity (Vujanovic, Arrindell, Bernstein, Norton, & Zvolensky, 2007). In addition, ASI scores have been shown to predict intensity of physical sensation in response to hyperventilation (Donnell & McNally, 1989).

Albany Panic and Phobia Questionnaire–Revised

The Albany Panic and Phobia Questionnaire–Revised (APPQ-R; Brown, White, & Barlow, 2005) is a 24-item self-report questionnaire that was modified from the original scale (Rapee, Craske, & Barlow, 1994) and measures situational, interoceptive, and social fear and avoidance. Respondents rate the degree of fear they would expect to experience while engaging in certain activities on a 9-point Likert-type scale (0 = no fear; 8 = extreme fear). This study utilized the interoceptive fear subscale, which consists of five items that assess fear of activities that induce somatic sensations. Research with clinical samples supports the factor structure, reliability, and convergent and discriminant validity of the APPQ-R (Brown et al., 2005).

Depression Anxiety Stress Scale

The Depression Anxiety Stress Scale (DASS; Lovibond & Lovibond, 1995) is a 42-item self-report questionnaire that measures the severity of symptoms on three 14-item subscales: depression, anxiety, and stress. Participants rate the degree to which items applied to them over the past week on a 4-point Likert-type scale (0 = did not apply to me; 3 = applied to me very much or most of the time). The DASS has demonstrated good reliability and validity in both clinical and nonclinical samples (Antony, Bieling, Cox, Enns, & Swinson, 1998; Lovibond & Lovibond, 1995). Specifically, results have shown that the DASS has good factor structure stability across a variety of samples, high internal consistency, and strong concurrent validity. This study utilized the depression subscale.

Procedure

Cardiac clinic technicians invited patients to participate in a longitudinal assessment study after a cardiac exam revealed no abnormalities. Interested patients completed a phone screen with research staff to ensure eligibility. Trained clinicians (i.e., clinical psychology doctoral students or a licensed psychologist) obtained informed consent from participants. As part of a larger study, participants then completed the self-report questionnaires described above at four time points and returned the questionnaire to the research office in a prepaid envelope. Apart from test–retest reliability, analyses were conducted using data obtained at baseline. Following completion of the study, participants were paid $25 for their time and effort.

An expert panel was tasked with identifying which of the 18 CAQ items should be included in a reassurance-seeking subscale. The panel was provided with the 18 CAQ items and asked which fit best into a reassurance-seeking factor. We planned to include items on the reassurance-seeking subscale that received at least three out of five votes. Experts included two master’s degree–level clinical psychology doctoral students and three doctoral-level clinical psychologists, all of who have extensive research experience in NCCP and have published on the subject. This process allowed us to develop a more theoretically informed model as opposed to the data-driven model that results from exploratory factor analysis (Pedhazur & Schmelkin, 1991).

Results

Analysis of Expert Panel Responses

Of the 18 CAQ items, three were unanimously selected by all five panel members. One item was selected by two of the panel members, and four items were selected by only one of the panel members. Those three items with a panel selection rate of 100% were used to form the reassurance-seeking subscale. The three items selected were (a) “When I have chest discomfort, or when my heart is beating fast, I like to be checked out by a doctor”; (b) “When I have chest discomfort, or when my heart is beating fast, I tell my family or friends”; and (c) “When I have chest discomfort, or when my heart is beating fast, I feel safe around a hospital, physician, or other medical facility.”

Factor Structure

Maximum likelihood factor analysis in AMOS 18.0 was used to conduct confirmatory factor analysis using the three-factor model originally proposed by Eifert, Thompson, et al. (2000) and the four-factor model that includes a reassurance-seeking factor. Confirmatory factor analysis allowed us to examine a priori hypotheses regarding those two differing models. Models were evaluated using the overall chi-square, the Tucker–Lewis index (TLI), the comparative fit index (CFI), and the root mean square error of approximation (RMSEA). Four sets of error terms were allowed to correlate: Items 3 and 4, Items 2 and 7, Items 14, 15, and 16, and Items 17 and 18.

Neither the three-factor model, χ²(126) = 258.11, p < .001, nor the four-factor model, χ²(123) = 243.68, p < .001, were a strong fit for the data. Although rejecting the null hypothesis of this chi-square test indicates poor model fit, it should be noted that the chi-square test is overly sensitive to sample size, such that large samples can lead to false rejection of the null hypothesis (Bentler & Bonett, 1980). Fit indices for both the three-factor (TLI = .89, CFI = .91, RMSEA = .07) and four-factor (TLI = .90, CFI = .92, RMSEA = .07) models were in the acceptable range. A chi-square comparison test indicated that the four-factor model was a significant improvement over the three-factor model, change in χ²(3) = 14.43, p < .01. Furthermore, the three items included in the reassurance-seeking subscale exhibited stronger loadings on that new factor than on the heart-related fear factor of the original model. Table 1 contains the standardized regression coefficients for the items of both models.

Table 1.

Confirmatory Factor Analyses.

	Factor loadings
	Eifert, Thompson, et al.’s (2000) three-factor model			Expert panel’s four-factor model
	Fear	Avoidance	Attention	Fear	Avoidance	Attention	Reassurance-seeking
10. If tests come out normal, I still worry about my heart.	.78			.79
13. I worry that doctors do not believe my chest pain/discomfort is real.	.53			.53
When I have chest discomfort, or when my heart is beating fast:
14. . . . I worry that I may have a heart attack.	.64			.63
15. . . . I have difficulty concentrating on anything else.	.59			.58
16. . . . I get frightened.	.62			.62
17. . . . I like to be checked out by a doctor.	.43						.50
18. . . . I tell my family or friends.	.31						.66
11. I feel safe around a hospital, physician, or other medical facility.	.31						.48
2. I avoid physical exertion.		.75			.75
5. I take it easy as much as possible.		.58			.58
7. I avoid exercise or other physical work.		.77			.77
9. I avoid activities that make my heart beat faster.		.83			.83
12. I avoid activities that make me sweat.		.79			.79
1. I pay attention to my heartbeat.			.83			.84
3. My racing heart wakes me up at night.			.55			.55
4. Chest pain/discomfort wakes me up at night.			.36			.35
6. I check my pulse.			.40			.40
8. I can feel my heart in my chest.			.65			.64

Convergent and Discriminant Validity, Internal Consistency, and Test–Retest Reliability

Correlations between the CAQ and BVS, ASI, APPQ-behavioral avoidance subscale, and DASS-depression subscale were used to determine convergent and discriminant validity (see Table 2). CAQ total and subscales scores demonstrated significant correlations with anxiety-related constructs as measured by the BVS, ASI, and APPQ-behavioral avoidance subscale. All were in the moderate range except for the correlation between CAQ-avoidance and BVS and the correlation between CAQ-reassurance seeking and AAPQ-behavioral avoidance. Correlations between CAQ total and subscales and DASS-Depression were also significant but fell in the small-sized range. Cronbach’s alpha was used as an index of internal consistency. Cronbach’s alphas for the CAQ total and subscale scores indicated adequate to good internal consistency (alpha’s ranged from .66-.86; see Table 3). In addition, correlations between baseline and 6-month follow-up scores indicated low to adequate test–retest reliability (rs ranged from .52 to .79; see Table 3).

Table 2.

Convergent and Discriminant Validity Analyses and CAQ Subscale Intercorrelations.

		CAQ
Measure	M (SD)	Total	Fear	Avoidance	Attention	Reassurance-seeking
CAQ
Total	1.18 (0.58)	—	.84**	.68**	.73**	.57**
Fear	1.36 (0.89)	—	—	.41**	.53**	.41**
Avoidance	1.03 (0.82)	—	—	—	.34**	.13
Attention	0.91 (0.67)	—	—	—	—	.27**
Reassurance seeking	1.65 (0.96)	—	—	—	—	—
Validity measures
Depression	5.62 (7.47)	.26**	.32**	.23**	.21*	−.01
Body vigilance	42.41 (37.57)	.49**	.44**	.25*	.42**	.35**
Anxiety sensitivity	17.13 (11.84)	.56**	.62**	.30**	.36**	.29**
Behavioral avoidance	8.96 (9.75)	.54**	.48**	.54**	.32**	.16**

Note. CAQ = Cardiac Anxiety Questionnaire; Body vigilance = Body Vigilance Scale; Anxiety Sensitivity = Anxiety Sensitivity Index; Behavioral Avoidance = Albany Panic and Phobia Questionnaire interoceptive subscale; Depression = Depression Anxiety Stress Scales depression subscale.

p < .05. **p < .01.

Table 3.

Reliability Analyses.

Subscale	Cronbach’s alpha	Test–retest correlation
Fear	.82	.73**
Avoidance	.86	.79**
Attention	.72	.73**
Reassurance seeking	.66	.52**
Total	.86	.79**

p < .01.

Concurrent Validity

We conducted two hierarchical linear regression analyses to examine the unique contributions of CAQ subscales to the prediction of pain-related interference and number of chest pain–specific doctor visits in the past year. Both hierarchical models were structured such that demographic variables (i.e., age and gender) were entered at Step 1, pain variables (i.e., chest pain frequency and total duration of chest pain condition) at Step 2, and CAQ subscales at Step 3. Dummy codes were used for gender (male = 1, female = 0). Squared semipartial correlations are provided as a measure of effect size for predictors that were significant at Step 3, as this metric represents the percentage of total variance in the dependent variable uniquely accounted for by an independent variable. Tables 4 and 5 present the results from each of these hierarchical linear regression analyses.

Table 4.

Hierarchical Regression Analysis for Pain Interference.

	Β	SE Β	β	T	R ²	ΔR²	F change
Step 1: Demographics					<.01	<.01	0.13
Age	.03	.05	.04	0.64
Gender	.85	1.05	.05	0.81
Step 2: Clinical variables					.06	.06	5.65**
Pain frequency	−1.06	.32	−.22	−3.33**
Pain total duration	−.18	.38	−.03	−0.474
Step 3: CAQ subscales					.25	.19	11.31**
Fear	.37	.15	.21	2.39*
Avoidance	.35	.14	.18	2.49*
Attention	.22	.19	.10	1.18
Reassurance seeking	.35	.20	.13	1.73

p < .05. **p < .01.

Table 5.

Hierarchical Regression Analysis for Number of Chest Pain–Specific Doctor Visits in Past Year.

	Β	SE Β	β	T	R ²	ΔR²	F change
Step 1: Demographics					.01	.01	1.30
Age	<.01	<.01	−.08	−1.07
Gender	−.02	.09	−.02	−0.20
Step 2: Clinical variables					.05	.04	3.36*
Pain frequency	−.07	.03	−.19	−2.63**
Pain total duration	−.01	.03	−.02	−0.27
Step 3: CAQ subscales					.11	.06	2.88*
Fear	.01	.01	.10	1.10
Avoidance	<.01	.01	.01	0.11
Attention	<.01	.02	−.01	−0.16
Reassurance seeking	.04	.02	.19	2.29*

p < .05. **p < .01.

CAQ subscales made significant contributions to the prediction of both criterion variables. With regard to pain-related interference, demographic and clinical variables accounted for a cumulative 6% of variance, ΔR² = .06; F change(2, 179) = 5.65; p change < .01. After controlling for the effects of these variables, CAQ subscales accounted for an additional 19% of the variance, ΔR²; F change(4, 175) = 11.31; p change < .01. With all variables entered, only CAQ-Fear and CAQ-Avoidance were significant predictors, with squared semipartial correlations of .02 and .03, respectively. The final regression equation accounted for 25% of the variance (R²) in pain-related interference, F(8, 175) = 7.44; p < .01.

For number of chest pain–specific doctor visits in the past year, demographic and clinical variables accounted for a cumulative 5% of variance, ΔR² = .05; F change(2, 179) = 3.36; p change < .05. After controlling for the effects of these variables, CAQ subscales accounted for an additional 6% of the variance, ΔR²; F change(4, 175) = 2.88; p change < .05. With all variables entered, only CAQ-Reassurance Seeking was a significant predictor, with a squared semipartial correlation of .03. The final regression equation accounted for 33% of the variance (R²) in number of chest pain–specific doctor visits, F(8, 175) = 2.66; p < .01.

Discussion

Our objective was to evaluate the factor structure and psychometric properties of the CAQ in a sample of patients with NCCP. Results indicated that the CAQ has good construct validity in this particular patient group, and a four-factor model was determined to be the best fit for the data. In addition, lower order factors of HFA demonstrated significant relationships with pain-related impairment and health care utilization, indicating good concurrent validity.

Three-factor and four-factor models of the CAQ were evaluated using confirmatory factor analysis. The three-factor model was derived from an exploratory factor analysis by Eifert, Thompson, et al. (2000) using responses from a sample of 188 patients, of which the majority had tested positive for coronary artery disease. The four-factor version was based on that originally proposed three-factor model and included an additional reassurance-seeking factor composed of items identified by an expert panel of NCCP researchers. Notably, our expert panel identified the same three items as belonging to the reassurance-seeking subscale as those identified in previous studies by Marker et al. (2008) and Van Beek, Voshaar, et al. (2012). Three items from the original heart-related fear subscale were taken to form the reassurance-seeking subscale. Although neither model was an ideal fit for the data, results demonstrated that the four-factor model was an improvement over the three-factor model.

Unimpressive fit indices indicate the need for additional examination of the CAQ factor structure in NCCP patient samples. These results may be due to methodological issues related to the inherent differences between exploratory and confirmatory factor analyses. Many of the parameters that were unconstrained in the initial exploratory analysis conducted by Eifert, Thompson, et al. (2000) were constricted in our follow-up confirmatory analysis, meaning that our analyses were simply more conservative. Sample size may have also influenced these findings. While there is support that our sample is adequately sized for these analyses (Shah & Goldstein, 2006), others have suggested that given the number of parameters specified in the models, a larger sample would have been more appropriate (Jackson, 2003). Nevertheless, what is of primary importance is that the inclusion of a reassurance-seeking factor resulted in a model that is not only more psychometrically sound but also better aligned with current theory, which holds that reassurance seeking is a key behavioral component and maintenance factor of HFA (Eifert, Zvolensky, et al., 2000).

Reliability analyses provided some additional support of the psychometric basis of the CAQ. Test–retest correlations indicated that the CAQ total score and all but one subscale have good temporal stability. The test–retest correlation for the reassurance-seeking subscale is only moderately sized, indicating insufficient reliability. Estimates of internal consistency, based on Cronbach’s alphas, indicated that the CAQ total score and all but one subscale have acceptable internal consistency. Again, it was the reassurance-seeking subscale that demonstrated low reliability. However, because Cronbach’s alpha is a function of the number of items contained in a scale (Cortina, 1993), this finding may in part be influenced by the fact that there are only three items on the reassurance-seeking subscale.

The CAQ demonstrated good convergent and discriminant validity. CAQ scores correlated more highly with measures of anxiety-related constructs than with a measure of depression. Several correlations deserve specific mention. Heart-related fear correlated most highly with AS or a fear of anxiety-related sensations. Avoidance of behaviors thought to elicit cardiac arousal demonstrated the strongest correlation with behavioral avoidance of anxiety-provoking circumstances. Last, heart-focused attention was most strongly associated with body vigilance. These three correlations were moderately sized.

Hierarchical regression analyses were conducted to investigate the contribution of lower order CAQ factors to two health-related markers of behavior. Results revealed that the CAQ has good concurrent validity. Cardiac-related fear and cardiac-related avoidance demonstrated significant relationships with pain-related life interference, the degree to which pain impairs daily functioning. In addition, higher scores on the reassurance-seeking subscale were associated with greater number of chest pain–specific doctor visits in the past year. These relationships persisted after controlling for relevant demographic (i.e., age and gender) and chest pain variables (i.e., chest pain frequency and total duration of chest pain condition). Our findings speak to the utility of the CAQ by demonstrating that CAQ subscales are related to experiential components of NCCP. Furthermore, these findings may have implications for the development of future NCCP treatments.

Psychological interventions for NCCP have led to reductions in chest pain severity and frequency as well as chest pain–related functional impairment (Jonsbu, Dammen, Morken, Moum, & Martinsen, 2011; Kisely, Campbell, Skerritt, & Yelland, 2010). However, significant room for improvement remains. Recent evidence has shown that the effectiveness of intervention approaches based in the attribution model suffer from low acceptability. That is, although dysfunctional misattributions of chest pain etiology may stand at the center of NCCP, attempts to modify patients’ beliefs regarding the source and cause of chest pain are generally not well received (Esler & Bock, 2004). Our findings illuminate targets for intervention that fall outside of the attribution model and are consistent with cognitive behavioral therapy recommendations for treatment of medically unexplained symptoms (Kent & McMillan, 2009). Cognitive interventions designed to reduce heart-related fears and exposure-based interventions aimed at decreasing heart-related behavioral avoidance could be avenues to alleviate functional impairment. Additionally, interventions to modify reassurance-seeking behaviors could decrease rates of excessive health care use, a well-documented problem in the NCCP patient population (Tew et al., 1995).

Several study limitations exist. First, due to a fairly homogenous sample, the generalizability of these results may be restricted. Second, as the regression analyses were part of a cross-sectional design, we are unable to determine causality. Third, data were collected entirely through self-report, and method variance may have influenced our results. Fourth, although the reassurance-seeking subscale demonstrated a significant relationship to doctor visits, it should be noted that other factors known to influence health care utilization, such as insurance status, education, and ethnicity were not accounted for. Fifth, the 6-month interim period between CAQ administrations may have been too lengthy to accurately determine test–retest reliability. This time interval may have had a particularly problematic impact on the reassurance subscale’s test–retest correlation, as reassurance may fluctuate substantially over time.

Additional research is required to further evaluate the psychometric properties of the CAQ. In particular, future studies should examine factor structure invariance over time as well as across gender. Theory-driven tests of HFA lower order factors are another critical next step in this area.

Our findings illustrate that the CAQ is a useful measure of HFA in patients with NCCP, and further, that CAQ lower order factors are related to health-related outcomes in this population. Strengthening HFA assessment instruments is a critical step toward expanding our understanding of NCCP as well as improving diagnostic and intervention methods for this patient group.

Footnotes

Authors’ Note

The contents herein are solely the responsibility of the authors and do not necessarily represent the views of the funding sources.

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: Preparation of this article was supported in part by grants from the National Institute of Mental Health (MH63185) and the University of Missouri–St. Louis (University Research Award) awarded to Kamila S. White.

References

Aicher

Holz

Feldner

Kollner

Schafers

H. J.

(2011). Quality of life after aortic valve surgery: Replacement versus reconstruction. Journal of Thoracic and Cardiovascular Surgery, 142, e19-e24.

Aikens

J. E.

Michael

Levin

Lowry

(1999). The role of cardioprotective avoidance beliefs in noncardiac chest pain and associated emergency department utilization. Journal of Clinical Psychology in Medical Settings, 6, 317-332.

Antony

M. M.

Bieling

P. J.

Cox

B. J.

Enns

M. W.

Swinson

R. P.

(1998). Psychometric properties of the 42-item and 21-item versions of the Depression Anxiety Stress Scales in clinical groups and a community sample. Psychological Assessment, 10, 176-181.

Barlow

D. H.

(1988). Anxiety and its disorders. New York, NY: Guildford Press.

Barsky

A. J.

Ahern

D. K.

Bailey

E. D.

Saintfort

Liu

E. B.

Peekna

H. M.

(2001). Hypochondriacal patients’ appraisal of health and physical risks. American Journal of Psychiatry, 158, 783-787.

Bentler

P. M.

Bonett

D. G.

(1980). Significance tests and goodness of fit in the analysis of covariance structures. Psychological Bulletin, 88, 588-606.

Bernstein

I. H.

Jaremko

M. E.

Hinkley

B. S.

(1995). On the utility of the West Haven-Yale Multidimensional Pain Inventory. Spine, 20, 956-963.

Brown

T. A.

White

K. S.

Barlow

D. H.

(2005). A psychometric reanalysis of the Albany Panic and Phobia Questionnaire. Behaviour Research and Therapy, 43, 337-355.

Carmin

C. N.

Wiegartz

P. S.

Hoff

J. A.

Kondos

G. T.

(2003). Cardiac anxiety in patients self-referred for electron beam tomography. Journal of Behavioral Medicine, 26, 67-80.

10.

Cortina

J. M.

(1993). What is coefficient alpha? An examination of theory and applications. Journal of Applied Psychology, 78, 98-104.

11.

Donnell

C. D.

McNally

R. J.

(1989). Anxiety sensitivity and history of panic as predictors of response to hyperventilation. Behaviour Research and Therapy, 27, 325-332.

12.

Eifert

G. H.

(1992). Cardiophobia: A paradigmatic behavioral model of heart-focused anxiety and non-anginal chest pain. Behaviour Research and Therapy, 30, 329-345.

13.

Eifert

G. H.

Hodson

S. E.

Tracey

D. R.

Seville

J. L.

Gunawardane

(1996). Heart-focused anxiety, illness beliefs, and behavioral impairment: Comparing healthy heart-anxious patients with cardiac and surgical inpatients. Journal of Behavioral Medicine, 19, 385-399.

14.

Eifert

G. H.

Thompson

R. N.

Zvolesnsky

M. J.

Edwards

Frazer

N. L.

Haddad

J. W.

Davig

(2000). The Cardiac Anxiety Questionnaire: Development and preliminary validity. Behaviour Research and Therapy, 38, 1039-1053.

15.

Eifert

G. H.

Zvolensky

M. J.

Lejuez

C. W.

(2000). Heart-focused anxiety and chest pain: A conceptual and clinical review. Clinical Psychology: Science and Practice, 7, 403-417.

16.

Esler

J. L.

Bock

B. C.

(2004). Psychological treatments for noncardiac chest pain: Recommendations for a new approach. Journal of Psychosomatic Research, 56, 263-269.

17.

Eslick

G. D.

Jones

M. P.

Talley

N. J.

(2003). Non-cardiac chest pain: Prevalence, risk factors, impact and consulting—A population-based study. Alimentary Pharmacology & Therapeutics, 17, 1115-1124.

18.

Fischer

Kindermann

Karbach

Herzberg

P. Y.

Ukena

Barth

. . . Kollner

(2012). Heart-focused anxiety in the general population. Clinical Research in Cardiology, 101, 109-116.

19.

Fleet

R. P.

Beitman

B. D.

(1998). Cardiovasucular death from panic disorder and panic-like anxiety: A critical review of the literature. Journal of Psychosomatic Research, 44, 71-80.

20.

Hoyer

Eifert

G. H.

Einsle

Zimmermann

Krauss

Knaut

. . . Kollner

(2008). Heart-focused anxiety before and after cardiac surgery. Journal of Psychosomatic Research, 64, 291-297.

21.

Jackson

D. L.

(2003). Revisiting sample size and number of parameter estimates: Some support for the N:q hypothesis. Structural Equation Modeling, 10, 128-141. doi:10.1207/S15328007SEM1001_6

22.

Jonsbu

Dammen

Morken

Moum

Martinsen

E. W.

(2011). Short-term cognitive behavioral therapy for non-cardiac chest pain and benign palpitations: A randomized controlled trial. Journal of Psychosomatic Research, 70, 117-123.

23.

Kent

McMillan

(2009). A CBT-based approach to medically unexplained symptoms. Advances in Psychiatric Treatment, 15, 148-151.

24.

Kerns

R. D.

Turk

D. C.

Rudy

T. E.

(1985). The West Haven-Yale Multidimensional Pain Inventory (WHYMPI). Pain, 23, 345-356.

25.

Kisely

S. R.

Campbell

L. A.

Skerritt

Yelland

M. J.

(2010). Psychological interventions for symptomatic management of non-specific chest pain in patients with normal coronary anatomy. Cochrane Database of Systematic Reviews, 20, 1-47.

26.

Lovibond

P. F.

Lovibond

S. H.

(1995). The structure of negative emotional states: Comparison of the Depression Anxiety Stress Scales (DASS) with the Beck Depression and Anxiety Inventories. Behaviour Research and Therapy, 33, 335-343.

27.

Marker

C. D.

Carmin

C. N.

Ownby

R. L.

(2008). Cardiac anxiety in people with and without coronary atherosclerosis. Depression and Anxiety, 25, 824-831.

28.

Olatunji

B. O.

Deacon

B. J.

Abramowitz

J. S.

Valentiner

D. P.

(2007). Body vigilance in non-clinical and anxiety disorder samples: Structure, correlates, and prediction of health concerns. Behavior Therapy, 38, 392-401. doi:10.1016/j.beth.2006.09.002

29.

Pedhazur

Schmelkin

L. P.

(1991). Measurement, design, and analysis: An integrated approach. Hillsdale, NJ: Lawrence Erlbaum.

30.

Pelland

Marchand

Lessard

Belleville

Chauny

Vadeboncoeur

. . . Lavoie

K. L.

(2011). Efficacy of 2 interventions for panic disorder in patients presenting to the ED with chest pain. American Journal of Emergency Medicine, 29, 1051-1061.

31.

Rapee

R. M.

Craske

M. G.

Barlow

D. H.

(1994). Assessment instrument for panic disorder that includes fear of sensation-producing activities: The Albany Panic and Phobia Questionnaire. Anxiety, 1, 114-122.

32.

Realini

J. P.

Katerndahl

D. A.

(1993). Factors affecting the threshold for seeking care: The Panic Attack Care-Seeking Threshold (PACT) study. Journal of the American Board of Family Practice, 6, 215-223.

33.

Reiss

McNally

R. J.

(1985). The expectancy model of fear. In Reiss

Bootzin

(Eds.), Theoretical issues in behavior therapy (pp. 107-121). New York, NY: Academic Press.

34.

Reiss

Peterson

R. A

Gursky

D. M.

McNally

R. J.

(1986). Anxiety sensitivity, anxiety frequency, and the predction of fearfulness. Behaviour Research and Therapy, 24, 1-8.

35.

Rodriguez

B. F.

Bruce

S. E.

Pagano

M. E.

Spencer

M. A.

Keller

M. B.

(2004). Factor structure and stability of the Anxiety Sensitivity Index in a longitudinal study of anxiety disorder patients. Behaviour Research and Therapy, 42, 79-91.

36.

Schmidt

N. B.

Lerew

D. R.

Trakowski

J. H.

(1997). Body vigilance in panic disorder: Evaluating attention to bodily perturbations. Journal of Consulting and Clinical Psychology, 65, 214-220. doi:10.1037/0022-006X.65.2.214

37.

Shah

Goldstein

S. M.

(2006). Use of structural equation modeling in operations management research: Looking back and forward. Journal of Operations Management, 24, 148-169. doi:10.1016/j.jom.2005.05.001

38.

Spinhoven

Van der Does

A. J. W.

Van Dijk

Van Rood

Y. R.

Heart-focused anxiety as a mediating variable in the treatment of noncardiac chest pain by cognitive-behavioral therapy and paroxetine. Journal of Psychosomatic Research, 69, 227-235.

39.

Strik

J. J. M. H.

Denollet

Lousberg

Honig

(2003). Comparing symptoms of depression and anxiety as predictors of cardiac events and increased health care consumption after myocardial infarction. Journal of the American College of Cardiology, 42, 1801-1807.

40.

Taylor

Cox

B. J.

(1998). Anxiety sensitivity: multiple dimensions and hierarchic structure. Behaviour Research and Therapy, 36, 37-51.

41.

Tew

Gutherie

E. A.

Creed

F. H.

Cotter

Kisely

Tomenson

(1995). A long-term follow-up study of patients with ischemic heart disease versus patients with nonspecific chest pain. Journal of Psychosomatic Research, 39, 977-985.

42.

Thompson

(1990). Investigation into the reliability and validity of a Multidimensional Pain Inventory in an oncology population. Oncology Nursing Forum, 17, 160.

43.

Van Beek

M. H. C. T.

Mingels

Voshaar

R. C. O.

van Balkom

A. J. L. M.

Lappenschaar

Pop

Speckens

A. E. M.

(2012). One-year follow up of cardiac anxiety after a myocardial infarction: A latent class analysis. Journal of Psychosomatic Research, 73, 362-368.

44.

Van Beek

M. H. C. T.

Voshaar

R. C. O.

Van Deelen

F. M.

van Balkom

A. J. L. M.

Pop

Spekens

A. E. M.

(2012). The Cardiac Anxiety Questionnaire: Cross-validation among cardiac patients. International Journal of Psychiatry in Medicine, 43, 349-364.

45.

Vujanovic

A. A.

Arrindell

W. A.

Bernstein

Norton

P. J.

Zvolensky

M. J.

(2007). Confirmatory factor analytic evidence, internal consistency, and construct validity in a young adult sample from the Netherlands. Assessment, 14, 129-143.

46.

White

K. S.

Craft

J. M.

Gervino

E. V.

(2010). Anxiety and hypervigilance to cardiopulmonary sensations in pain patients with and without psychiatric disorders. Behaviour Research and Therapy, 48, 394-401.

47.

Zinbarg

R. E.

Brown

T. A.

Barlow

D. H.

Rapee

R. M.

(2001). Anxiety sensitivity, panic, and depressed mood: A reanalysis teasing apart the contributions of the two levels in the hierarchical structure of the anxiety sensitivity index. Journal of Abnormal Psychology, 110, 372-377. doi:10.1037/0021-843X.110.3.372

48.

Zvolensky

M. J.

Eifert

G. H.

Feldner

M. T.

Leen-Feldner

(2003). Heart-focused anxiety and chest pain in postangiography medical patients. Journal of Behavioral Medicine, 26, 197-209.

49.

Zvolensky

M. J.

Feldner

M. T.

Eifert

G. H.

Vujanovic

A. A.

Solomon

S. E.

(2008). Cardiophobia: A critical analysis. Transcultural Psychiatry, 45, 230-252.