Cogn Ther Res (2009) 33:323–333 DOI 10.1007/s10608-007-9181-7 ORIGINAL ARTICLE Distinctiveness of Perceived Health in Panic Disorder and Relation to Panic Disorder Severity Vladan Starcevic Æ David Berle Æ Pauline Fenech Æ Denise Milicevic Æ Claire Lamplugh Æ Anthony Hannan Published online: 13 January 2008 Ó Springer Science+Business Media, LLC 2008 Abstract The aims of this study were to investigate to what extent perceived health in panic disorder (PD) is distinct from related constructs and to examine the relationship between perceived health and various domains of PD severity. Seventy-five PD outpatients were administered instruments measuring perceived health, anxiety sensitivity, hypochondriacal fears and beliefs, catastrophic appraisals of the consequences of physical symptoms of anxiety and panic, various aspects of psychopathology, and PD severity. Results of the correlation, partial correlation, and multiple regression analyses suggest the following: (1) among PD patients there is much overlap between negatively perceived health, catastrophic appraisals of the consequences of physical symptoms of anxiety and panic, and hypochondriacal beliefs; (2) negatively perceived health is not significantly related to PD severity nor to any aspect of PD severity, whereas anxiety sensitivity and hypochondriacal beliefs have a significant relationship with the severity of anticipatory anxiety, and in that regard may be more clinically useful than perceived health before the beginning of treatment. V. Starcevic Discipline of Psychological Medicine, Nepean Hospital, University of Sydney, Sydney/Penrith, NSW, Australia D. Berle
D. Milicevic
C. Lamplugh
A. Hannan Nepean Anxiety Disorders Clinic, Sydney West Area Health Service, Penrith, NSW, Australia V. Starcevic (&)
P. Fenech Department of Psychological Medicine, Nepean Hospital, P.O. Box 63, Penrith, NSW 2751, Australia e-mail: starcev@wahs.nsw.gov.au Keywords Panic disorder
Panic disorder severity
Perceived health
Anxiety sensitivity
Hypochondriacal beliefs
Catastrophic appraisals Introduction Health perceptions are defined as appraisals, evaluations or interpretations of one’s health (Schmidt and Telch 1997). Perceived health can be broadly positive or negative; regardless of how it correlates with the actual health, it may be important to understand its role in certain types of psychopathology. For example, negatively perceived health has been proposed to represent a cognitive risk factor for panic disorder (PD), separate from heightened anxiety sensitivity (Yartz et al. 2005). Consequently, PD may be more likely to occur on a background of negative perceptions of one’s health. A negatively perceived health may also have prognostic implications for PD patients, considering that negatively perceived health has been found to be a significant predictor of mortality in general (Idler and Angel 1990; Kaplan and Camacho 1993; Mossey and Shapiro 1982) and that people with panic-like anxiety symptoms, panic attacks, and PD have higher mortality rates, mainly due to cardiovascular and cerebrovascular diseases (Allgulander and Lavori 1991; Coryell et al. 1982, 1986; Kawachi et al. 1994; Smoller et al. 2007; Weissman et al. 1990). Given a possible role of negatively perceived health in the pathogenesis and prognosis of PD, it is surprising that it has not received as much attention as some other putative pathogenetic and prognostic factors. This may to some extent be due to the conceptual overlap and lack of clarity about the relationship between perceived health and similar 123 324 constructs and psychopathological states, especially anxiety sensitivity (defined as the fear of anxiety-related sensations based on beliefs that these sensations have harmful consequences; Reiss and McNally 1985) and hypochondriasis (defined as a maladaptive preoccupation with health, which manifests itself through worries and/or beliefs about having a serious disease; Starcevic 2001). Negatively perceived health may be similar to anxiety sensitivity and hypochondriacal phenomena, in that persons who feel as though their health status is vulnerable, may also be fearful of any somatic symptoms which signal a threat to their health. Moreover, there has been debate as to whether anxiety sensitivity and hypochondriacal phenomena represent distinct fears or different aspects of a more general fear of somatic sensations (Otto and Pollack 1994; Otto et al. 1992; Taylor 1994, 1995). Several studies have examined the relationship between perceived health in PD and anxiety sensitivity. Significant correlations between the two (ranging from -0.28 to -0.51; Gregor et al. 2005; McLeish et al. 2006; Schmidt and Telch 1997; Yartz et al. 2005) have been reported. However, the interpretations of these correlations varied from one study to another, contributing to some ambiguity. For instance, a correlation of -0.30 was interpreted as ‘‘moderate,’’ but it was suggested that perceived health and anxiety sensitivity are distinct concepts (Yartz et al. 2005). A similar degree of correlation (-0.31) was found in another study (Schmidt et al. 2003), but there was no association between perceived impact of physical health and anxiety sensitivity. This finding was interpreted to suggest that a negative perception of the impact of physical health—but not necessarily negatively perceived health— was distinguishable from anxiety sensitivity. It is questionable, though, to what extent a negative perception of the impact of physical health can be separated and differentiated from negatively perceived health, especially outside research settings. There have been no studies of the relationship between perceived health in PD and hypochondriacal tendencies or hypochondriasis. This is surprising in view of the fact that PD patients often have prominent hypochondriacal features (Otto et al. 1992; Starcevic et al. 1992) and that not infrequently, PD co-occurs with full-blown hypochondriasis (Bach et al. 1996; Barsky et al. 1994; Benedetti et al. 1997; Furer et al. 1997). As already suggested, individuals with PD who also have negative perceptions of their health may worry more about their health and be more prone to develop hypochondriacal fears and beliefs. Likewise, people with PD and prominent hypochondriacal concerns may be more likely to perceive their health negatively. Not only is it unclear whether negatively perceived health may lead to hypochondriasis or vice versa, but also how it is related and whether the two are 123 Cogn Ther Res (2009) 33:323–333 distinct from each other. Furthermore, anxiety sensitivity may play a certain role in the relationship between negative perceptions of one’s health and hypochondriasis in PD, considering that anxiety sensitivity was found to be the strongest predictor of hypochondriacal concerns in PD patients (Otto et al. 1992). Yet another aspect of PD that has not been directly studied in relation to perceived health is a tendency to make catastrophic appraisals of the consequences of physical symptoms of anxiety and panic. This tendency may be heightened in PD individuals with negatively perceived health because they may harbor a strong sense of physical vulnerability (Schmidt and Telch 1997; Schmidt et al. 2003). Likewise, catastrophic interpretations of anxiety-related physical sensations and anticipation of somatic catastrophes may lead to negative perceptions of one’s health. There may also be a link between negatively perceived health in PD and more prominent depression, anxiety, and somatization (defined as the tendency to experience and communicate somatic distress and symptoms unaccounted for by pathological findings, to attribute them to physical illness, and to seek medical help for them; Lipowski 1988). While PD individuals who are also more anxious and more depressed and who somatize more may perceive their health more negatively, negatively perceived health may increase the levels of anxiety and depression. The somatizing tendencies in PD are often related to bodily preoccupation and hypochondriacal fears and beliefs, which may lead to negatively perceived health. Negatively perceived health in PD was predicted by higher levels of depression (Schmidt et al. 1996) and made the experience of negative affect, including that of depression and anxiety, more likely (Gregor et al. 2005). In a study in which somatization was measured via panic-related physical concerns, somatization did not prove to be an independent predictor of negatively perceived health (Schmidt et al. 1996). Alongside the distinctiveness of the concept, the clinical utility of perceived health needs to be established. For example, if a construct is related to the severity of a disorder, or is useful in predicting treatment response or prognosis, then it is worthy of consideration by clinicians. Gregor et al. (2005) reported that perceived health was associated (r = -0.39) with the total score on the Panic Disorder Severity Scale (PDSS; Shear et al. 1997) and that it significantly predicted disability in family and home responsibilities (adjusted R2 = 0.19) and social and interpersonal disability (adjusted R2 = 0.14), but not occupational or school impairment. Gregor et al., however, did not report correlations for various domains of PD severity, such as the frequency and intensity of panic attacks, the extent of anticipatory anxiety, or agoraphobic avoidance, Cogn Ther Res (2009) 33:323–333 although these domains constitute the ‘‘items’’ or subscales of the PDSS. The relative utility of perceived health in predicting the severity of PD, when compared with anxiety sensitivity, hypochondriacal beliefs, and similar variables, also remains unclear. A central question is whether perceived health is a more useful and precise indicator of PD severity than are these other variables. In view of the above, the present study, conducted in a sample of PD patients, had three aims. First, we sought to clarify the relationships between perceived health on one hand and related constructs and psychopathological states (anxiety sensitivity, catastrophic appraisals of the consequences of physical symptoms of anxiety and panic, henceforth referred to as catastrophizing of physical symptoms, hypochondriacal fears and beliefs, anxiety, depression, and somatization) on the other. Second, we wanted to determine whether any significant relationships between perceived health and variables such as anxiety and depression could be better accounted for by associations between anxiety sensitivity, hypochondriacal beliefs, or catastrophizing of physical symptoms, and these other variables. This controlling for the effects of anxiety sensitivity, hypochondriacal beliefs, and catastrophizing of physical symptoms was considered necessary because of the apparently greatest conceptual overlap between perceived health and these constructs, as discussed above. Third, we wanted to further investigate the relationship between perceived health and various domains of PD severity, especially in relation to the aforementioned similar constructs. On the basis of previous research (Gregor et al. 2005; McLeish et al. 2006; Schmidt and Telch 1997; Schmidt et al. 1996, 2003; Yartz et al. 2005) and stated clinical observations, we hypothesized that a measure of perceived health would correlate significantly with measures of anxiety sensitivity, catastrophizing of physical symptoms, hypochondriacal fears and beliefs, anxiety, depression, and somatization. However, when controlling for anxiety sensitivity, hypochondriacal beliefs, and catastrophizing of physical symptoms, we expected perceived health to no longer have significant relationships with other variables. We expected that, consistent with previous research (Gregor et al. 2005), negatively perceived health would be associated with increased overall PD severity and greater PD-related disability. We also considered it likely that negatively perceived health would be associated with specific domains of PD severity, such as degree of anticipatory anxiety. However, we were unable to hypothesize about the relative utility of perceived health in predicting PD severity when compared with the seemingly similar constructs of anxiety sensitivity, hypochondriacal beliefs, and the catastrophizing of physical symptoms. 325 Method Participants Seventy-five participants with a diagnosis of PD with or without agoraphobia were recruited from patients attending the Nepean Anxiety Disorders Clinic (Penrith, NSW, Australia), a specialist outpatient anxiety disorders service. They were usually referred to the clinic by their general practitioners. Some patients were referred by mental health professionals and other health care workers, whereas others were self-referred. Participants took part in this study voluntarily after the procedures had been fully explained to them and after they had signed consent forms. Clinic attendees were considered eligible to participate if the condition for which they sought help was PD with or without agoraphobia, or if this was causing them the most distress or functional impairment. Patients with a history of psychosis or bipolar disorder, as well as patients with current psychosis, bipolar disorder, substance abuse or dependence, severe depression, severe personality disorder, self-harming behavior, and suicidality, are not treated in the clinic, and therefore did not participate in the study. Those with a current depressive disorder (major depressive disorder or dysthymic disorder) were considered eligible, so long as PD with or without agoraphobia was considered their principal diagnosis. Measures Clinician-administered Measures Besides a routine clinical interview, two clinician-rated instruments were administered: the Mini International Neuropsychiatric Interview (MINI; Sheehan et al. 1999) and the Panic and Agoraphobia Scale (PAS; Bandelow 1995). The MINI is a semi-structured diagnostic interview, which establishes both the principal and co-occurring DSM-IV diagnoses. In addition to the diagnoses ‘‘covered’’ by the MINI, we have added assessment of specific phobia and hypochondriasis according to the DSM-IV criteria. The administration of the MINI is simple and usually takes between 30 and 60 min. The MINI has been validated against other widely used structured diagnostic interviews, and it was found to have good psychometric properties (Lecrubier et al. 1997; Sheehan et al. 1997, 1998). For example, the concordance (kappa value) for ‘‘current’’ PD between the MINI and the Structured Clinical Interview for DSM-III-R, Patient Version, was 0.76 (Sheehan et al. 1997). A test–retest reliability figure reported for the MINIbased diagnoses of ‘‘panic disorder and/or agoraphobia’’ was 0.76 (Lecrubier et al. 1997). There is evidence to 123 326 suggest that the MINI is well accepted by patients (Pinninti et al. 2003). The PAS is a measure of PD severity based on clinician assessment, which incorporates five subscales: Panic Attacks (consisting of panic attack frequency, severity, and duration items); Agoraphobic Avoidance; Anticipatory Anxiety; Disability and Functional Impairment; and Worries About Health. A total score is calculated by summing the scores on all of the items, across all five subscales. Bandelow (1999) has reported the sound internal (Cronbach a = 0.85), interrater (j = 0.78), and test–retest reliability (r = 0.73) of the scale. The PAS also correlates well with similar measures of panic symptomatology (Bandelow 1999). For all analyses in this study, our PAS ‘‘total’’ score did not include scores on any Worries About Health items, because their content is similar to that of our measure of perceived health; the inclusion of these scores might have artificially increased the chances of finding a significant relationship between negatively perceived health and the overall PD severity. Self-report Instruments Several self-report instruments were used in the study: the Medical Outcomes Study 36-item Short-Form Health Survey, Symptom Checklist 90-Revised (SCL-90R), Agoraphobic Cognitions Questionnaire (ACQ), Anxiety Sensitivity Index (ASI), and Illness Attitude Scales (IAS). The Medical Outcomes Study 36-item Short-Form Health Survey (SF-36; Ware and Sherbourne 1992; Ware et al. 1993) assesses general physical and mental health on eight subscales: Physical Functioning, Social Functioning, Bodily Pain, Role Impairment due to Physical Factors, Vitality, Mental Health, General Health, and Role Impairment due to Emotional Factors. For this study, we focused on the 5-item General Health subscale that we felt best corresponds with the perceived health status. Participants were required to respond to item 1 (‘‘In general, would you say your health is:’’) using a 5-point Likert scale (1 = excellent, 2 = very good, 3 = good, 4 = fair, 5 = poor). Items 2–5 of the General Health subscale are listed as follows, ‘‘I seem to get sick a little easier than other people’’; ‘‘I am as healthy as anybody I know’’; ‘‘I expect my health to get worse,’’ and ‘‘My health is excellent.’’ Participants responded to items 2–5 using a 5point Likert scale (1 = definitely true, 2 = mostly true, 3 = don’t know, 4 = mostly false, 5 = definitely false). Items 1, 3, and 5 are reverse scored. The total scores, representing the sum of scores on all five items, range between 5 and 25, with lower scores denoting more negatively perceived health (Ware et al. 1993). The subscale scores can be transformed for comparison to norms for the US population, but this is an optional step (Ware et al. 123 Cogn Ther Res (2009) 33:323–333 2000), and was deemed unnecessary for the purposes of the study. The General Health subscale appears to have satisfactory internal consistency, ranging from 0.78 to 0.90, as well as evidence of validity (Ware et al. 1993). In our sample, the internal consistency of the General Health subscale was 0.81. The SCL-90R (Derogatis 1994) is a 90-item instrument that assesses overall distress and psychopathology during the preceding 7 days. Each item is rated on a 5-point Likert scale (0–4) yielding scores on nine subscales (Somatization, Obsessive-Compulsive, Interpersonal Sensitivity, Depression, Anxiety, Hostility, Phobic Anxiety, Paranoid Ideation, and Psychoticism). Scores on each subscale are calculated by obtaining the mean of the corresponding items. All subscales appear to have good convergent and discriminant validity (Peveler and Fairburn 1990) and satisfactory internal consistency (ranging from 0.77 for Psychoticism to 0.90 for Depression in a study by Derogatis et al. 1976; and from 0.79 for Paranoid Ideation to 0.90 for Depression in a study by Horowitz et al. 1988). Test–retest reliability over 10 weeks ranged from 0.68 for Somatization to 0.83 for Paranoid Ideation (Horowitz et al. 1988). In the analyses, we included only the Somatization, Depression, and Anxiety subscales, because they were most pertinent to PD and perceived health. The ACQ (Chambless et al. 1984) is a 14-item instrument that assesses current frequency of catastrophic appraisals of anxiety and panic. Each item is rated on a 5-point scale ranging from 1 (‘‘thought never occurs’’) to 5 (‘‘thought always occurs when I am nervous’’). Scores on the subscales for Loss of Control (the mean of seven items) and Physical Concerns (the mean of seven items) can be calculated. The ACQ appears to have good internal consistency (Cronbach a = 0.8; Chambless et al. 1984) and adequate test–retest reliability over a 3-month period (r’s = 0.79, 0.70, and 0.85 for ACQ Total, Loss of Control, and Physical Concerns scores, respectively; Arrindell 1993a). The bi-dimensional factor structure of Loss of Control and Physical Concerns has been replicated in numerous studies (e.g., Arrindell 1993b; Stephenson et al. 1999). In this study, we only investigated the Physical Concerns subscale, as we felt that this was the most relevant to perceived health. The ASI (Reiss et al. 1986) is a 16-item scale assessing the fear of anxiety-related symptoms. Items are rated on a 5-point scale, from 0 (very little) to 4 (very much); a total score (ranging from 0 to 64) is calculated by summing the item scores. The ASI has good to excellent internal consistency (ranging from 0.82 in a study by Telch et al. 1989 to 0.88 in a study by Peterson and Heibronner 1987), and adequate test–retest reliability over a 2-week period (r = 0.75; Reiss et al. 1986). The scale also has a satisfactory degree of criterion validity and construct validity (Peterson and Heibronner 1987). Cogn Ther Res (2009) 33:323–333 The IAS (Kellner 1986) measures psychopathology associated with hypochondriasis and abnormal illness behavior. The IAS consists of nine subscales: Worry about Illness, Concerns about Pain, Health Habits, Hypochondriacal Beliefs, Thanatophobia, Disease Phobia, Treatment Experience, Effects of Symptoms, and Bodily Preoccupations. Items are rated on a 5-point scale, from 0 (no) to 4 (most of the time). Bouman and Visser (1998) have reported a test–retest reliability of 0.96 across a 4-week period and internal consistency ranging from 0.87 to 0.90. Disease Phobia and Hypochondriacal Beliefs subscales were considered most relevant for the assessment of hypochondriacal fears and beliefs (Kellner et al. 1987), and this is the reason for our focus on these two subscales. Study Procedures Upon contacting the clinic, patients were screened by phone to determine their suitability for the clinic and to establish a provisional diagnosis. Every case was then discussed at an intake meeting, and patients were subsequently assigned for full assessment to one of the clinicians, often not the same one who did the initial telephone screen. At the assessment appointment, study procedures were explained and patients agreeing to participate signed consent forms. All participants were then administered the MINI and the PAS by one of the clinical psychologists experienced in the area of anxiety and related disorders and trained in the use of diagnostic interviews. Following the administration of the MINI and the PAS, participants completed the SF-36, SCL-90R, ACQ, ASI, and IAS. Clinical file notes for each participant were reviewed for reports of the presence or absence of current chronic and/or significant medical conditions, and this information was recorded in the database. Data Analyses All analyses were performed using the Statistical Package for the Social Sciences (SPSS 13.0). Pearson’s r correlations were calculated to evaluate the relationships between participants’ perceived health on one hand and anxiety sensitivity, hypochondriacal fears and beliefs, catastrophizing of physical symptoms, and levels of psychopathology, on the other. Three partial correlation analyses were then conducted to measure these same associations while adjusting for the effects of anxiety sensitivity, hypochondriacal beliefs, and catastrophizing of physical symptoms, respectively. The rationale for conducting the latter analyses was explained in Introduction; in addition, anxiety sensitivity, hypochondriacal beliefs, and catastrophizing of physical symptoms correlated strongly with perceived health in our initial correlation analyses. Given 327 that multiple correlation coefficients were considered, Bonferroni corrections were used for each set of analyses. Pearson’s correlations were calculated to determine the relationships between the cognitive variables (perceived health, hypochondriacal beliefs, anxiety sensitivity, and catastrophizing of physical symptoms), overall severity of PD, and each of the PAS subscales, except for Worries About Health. Multiple regression analyses were then conducted to determine whether perceived health predicted PD severity independently of anxiety sensitivity, hypochondriacal beliefs, and catastrophizing of physical symptoms. The scores on the SF-36 General Health subscale, ASI, IAS Hypochondriacal Beliefs subscale, and the ACQ Physical Concerns subscale were entered as independent variables in each analysis and used to predict scores on the four PAS subscales and PAS total score (which served as dependent variables). For each regression, we also report partial correlations. These represent the correlations between each independent and dependent variable while holding all other independent variables constant. Finally, to establish whether the findings were influenced by the presence of current chronic and/or significant medical conditions, independent samples t-tests were performed to compare participants with and without such medical conditions on perceived health, on each of the other cognitive variables, and on each of the PD severity domains. The correlation and regression analyses were then repeated for those without current medical conditions. Results Participant Characteristics Fifty-five (73.3%) participants were women, 49 (65.3%) were married or living in a de facto relationship, 12 (16.0%) had a post-secondary school education, and 40 (53.3%) were engaged in paid employment. The mean age of the sample was 36.52 years (SD = 11.07), and their median age was 36 years. Twenty (26.7%) patients were identified as having at least one current chronic and/or significant medical condition. Examples of such conditions include hypertension, asthma, and arthritis. Fifty-eight (77.3%) participants had agoraphobia. With regards to current co-occurring diagnoses, there were 21 (28.0%) patients with generalized anxiety disorder, 18 (24.0%) with major depressive disorder, 15 (20.0%) with specific phobia, 6 (8.0%) with social anxiety disorder, 5 (6.7%) with dysthymic disorder, 5 (6.7%) with obsessive-compulsive disorder, 1 (1.3%) with posttraumatic stress disorder, and 1 (1.3%) with hypochondriasis. Two (2.7%) patients had past alcohol abuse, 2 (2.7%) were diagnosed with past alcohol dependence, and 4 (5.0%) had past substance abuse. 123 328 Cogn Ther Res (2009) 33:323–333 Table 1 Descriptive data and bivariate correlations for self-report measures Variables Mean (SD) 1. GHS SF-36 16.00 (4.30) 1 2 3 4 5 6 2. IAS (HB) 2.91 (3.11) -0.53** 3. IAS (DP) 3.92 (3.43) -0.50** 0.74** 4. SCL-90R DEP 1.65 (1.00) -0.49** 0.46** 0.40** 5. SCL-90R ANX 1.89 (0.99) -0.45** 0.42** 0.45** 0.81** 6. SCL-90R SOM 1.55 (0.95) -0.42** 0.48** 0.50** 0.74** 0.79** 37.08 (12.82) -0.47** 0.45** 0.59** 0.57** 0.66** 0.64** 2.20 (0.70) -0.56** 0.59** 0.63** 0.43** 0.49** 0.58** 7. ASI 8. ACQ-Physical 7 0.60** n = 75 ** P \ 0.01 after a Bonferroni correction for 28 comparisons GHS SF-36 = General Health subscale of the Short Form Health Survey IAS (HB) = Hypochondriacal Beliefs subscale of the Illness Attitude Scales IAS (DP) = Disease Phobia subscale of the Illness Attitude Scales SCL-90R DEP = Depression subscale of the Symptom Checklist 90-Revised SCL-90R ANX = Anxiety subscale of the Symptom Checklist 90-Revised SCL-90R SOM = Somatization subscale of the Symptom Checklist 90-Revised ASI = Anxiety Sensitivity Index total score ACQ-Physical = Fear of Physical Consequences subscale of the Agoraphobic Cognitions Questionnaire Distinctiveness of Perceived Health The means and standard deviations of the main variables, and the correlations between each variable, are reported in Table 1. Participants’ perceived health was found to be significantly negatively correlated with catastrophizing of physical symptoms (r = -0.56, P \ 0.01), hypochondriacal beliefs (r = -0.53, P \ 0.01), disease phobia (r = -0.50, P \ 0.01), depression (r = -0.49, P \ 0.01), anxiety sensitivity (r = -0.47, P \ 0.01), anxiety (r = -0.45, P \ 0.01), and somatization (r = -0.42, P \ 0.01). The results of the partial correlation analyses controlling for anxiety sensitivity, hypochondriacal beliefs, and catastrophizing of physical symptoms are presented in Table 2. When anxiety sensitivity was partialled out, only hypochondriacal beliefs (r = -0.41, P \ 0.01) and catastrophizing of physical symptoms (r = -0.39, P \ 0.05) remained significantly associated with perceived health. When we controlled for hypochondriacal beliefs, only catastrophizing of physical symptoms (r = -0.35, P \ 0.05) remained significantly associated with perceived health. When catastrophizing of physical symptoms was controlled for, there were no significant associations between perceived health and any other variable. Prediction of PD Severity Correlations between the various cognitive variables (perceived health, anxiety sensitivity, hypochondriacal beliefs, 123 and the catastrophizing of physical symptoms) and each component of the PD severity, as well as overall PD severity, are shown in Table 3. Anxiety sensitivity was significantly correlated with frequency, severity, and duration of panic attacks, as measured by PAS Panic Attacks (r = 0.35, P \ 0.05), and with severity of anticipatory anxiety, as measured by PAS Anticipatory Anxiety (r = 0.51, P \ 0.01). The only other significant correlation was between hypochondriacal beliefs and severity of anticipatory anxiety (r = 0.39, P \ 0.05). Table 4 shows the results of the multiple regression analyses. All four cognitive variables, in combination, predicted frequency, severity, and duration of panic attacks (18% of variance accounted for), severity of anticipatory anxiety (30% of variance accounted for), and overall severity of PD (15% of variance accounted for). However, the only independent predictors of any specific domain of PD severity were anxiety sensitivity and hypochondriacal beliefs, which each predicted severity of anticipatory anxiety (partial r’s of 0.40 and 0.24, respectively). Presence of Current Chronic and/or Significant Medical Conditions There were no significant differences between participants with a current medical condition (n = 20; mean = 16.44; SD = 4.16) and those without (n = 55; mean = 14.75; SD = 4.55) on perceived health, on any other cognitive variable (hypochondriacal beliefs, anxiety sensitivity, and Cogn Ther Res (2009) 33:323–333 329 Table 2 Partial correlations for self-report measures controlling for anxiety sensitivity, hypochondriacal beliefs, and catastrophic appraisals of the consequences of physical symptoms of anxiety and panic GHS SF-36 controlling for ACQ-Physical GHS SF-36 controlling for IAS (HB) GHS SF-36 controlling for ASI 1. ASI -0.31 -0.20 2. IAS (HB) -0.41** 3. IAS (DP) 4. SCL-90R DEP -0.31 -0.31 -0.19 -0.32 5. SCL-90R ANX -0.21 -0.30 -0.25 6. SCL-90R SOM -0.18 -0.22 -0.14 7. ACQ-Physical -0.39* -0.35* -0.30 -0.23 -0.33 n = 75 * P \ 0.05 after a Bonferroni correction for 21 comparisons ** P \ 0.01 after a Bonferroni correction for 21 comparisons GHS SF-36 = General Health subscale of the Short Form Health Survey IAS (HB) = Hypochondriacal Beliefs subscale of the Illness Attitude Scales IAS (DP) = Disease Phobia subscale of the Illness Attitude Scales SCL-90R DEP = Depression subscale of the Symptom Checklist 90-Revised SCL-90R ANX = Anxiety subscale of the Symptom Checklist 90-Revised SCL-90R SOM = Somatization subscale of the Symptom Checklist 90-Revised ASI = Anxiety Sensitivity Index total score ACQ-Physical = Fear of Physical Consequences subscale of the Agoraphobic Cognitions Questionnaire Table 3 Correlations between cognitive variables, panic disorder severity subscale scores, and total panic disorder severity score GHS SF-36 PAS Panic Attacks PAS Agoraphobic Avoidance PAS Anticipatory Anxiety PAS Disability and Functional Impairment PAS Totala -0.31 -0.10 -0.20 -0.26 -0.26 0.32 ASI 0.35* 0.05 0.51** 0.26 IAS (HB) 0.34 0.13 0.39* 0.31 0.34 ACQ-Physical 0.24 0.08 0.34 0.25 0.26 n = 75 * P \ 0.05 after a Bonferroni correction for 20 comparisons ** P \ 0.01 after a Bonferroni correction for 20 comparisons a PAS total score was calculated with Worries About Health items excluded GHS SF-36 = General Health subscale of the Short Form Health Survey ASI = Anxiety Sensitivity Index total score IAS (HB) = Hypochondriacal Beliefs subscale of the Illness Attitude Scales ACQ-Physical = Fear of Physical Consequences subscale of the Agoraphobic Cognitions Questionnaire PAS = Panic and Agoraphobia Scale catastrophizing of physical symptoms), nor on any of the PD severity domains. When the correlation and regression analyses were repeated only for participants without a current medical condition, the results were almost the same: all correlations were of a similar magnitude to those of the whole sample (although not all remained significant); for the regression analyses (where cognitive variables were used to predict PD severity), anxiety sensitivity remained a significant predictor of the severity of anticipatory anxiety (partial r = 0.41), while hypochondriacal beliefs were no longer a significant predictor of the severity of anticipatory anxiety, despite being correlated to a similar extent as in the whole sample (partial r = 0.26). 123 330 Cogn Ther Res (2009) 33:323–333 Table 4 Summary of multiple regression analyses predicting panic disorder severity Dependent variables PAS Panic Attacks Variables in the equation B GHS SF-36 SE B b Partial r -0.10 0.10 -0.14 -0.12 IAS (HB) 0.23 0.14 0.23 0.19 ASI 0.06 0.03 0.26 0.22 -0.54 0.70 -0.12 -0.91 -0.48 0.14 -0.51 -0.04 0.15 0.20 0.12 0.09 -0.008 0.05 -0.03 -0.02 -0.03 0.99 -0.005 -0.003 0.06 0.09 0.15 0.27* ACQ-Physical Overall R2 = 0.18, F = 3.84, P = 0.007 PAS Agoraphobic Avoidance GHS SF-36 IAS (HB) ASI ACQ-Physical 2 Overall R = 0.02, F = 0.33, P = 0.86 PAS Anticipatory Anxiety GHS SF-36 IAS (HB) 0.07 0.18 ASI ACQ-Physical 0.47** 0.14 0.24 0.07 0.02 -0.04 0.42 -0.01 -0.01 0.40 Overall R2 = 0.30, F = 7.58, P \ 0.0001 PAS Disability and Functional Impairment GHS SF-36 -0.07 0.11 -0.09 -0.07 IAS (HB) 0.22 0.16 0.20 0.16 ASI 0.03 0.04 0.14 0.11 -0.02 0.76 -0.004 -0.003 ACQ-Physical Overall R2 = 0.12, F = 2.37, P = 0.06 PAS Total a -0.14 0.33 -0.06 -0.05 IAS (HB) GHS SF-36 0.77 0.46 0.24 0.20 ASI 0.16 0.11 0.21 0.18 -0.62 2.26 -0.04 -0.03 ACQ-Physical Overall R2 = 0.15, F = 3.20, P = 0.02 n = 75 * P \ 0.05; ** P \ 0.01 a PAS total score was calculated with Worries About Health items excluded GHS SF-36 = General Health subscale of the Short Form Health Survey IAS (HB) = Hypochondriacal Beliefs subscale of the Illness Attitude Scales ASI = Anxiety Sensitivity Index total score ACQ-Physical = Fear of Physical Consequences subscale of the Agoraphobic Cognitions Questionnaire Discussion As hypothesized and reported by others (Gregor et al. 2005; McLeish et al. 2006; Schmidt and Telch 1997; Schmidt et al. 1996, 2003; Yartz et al. 2005), perceived health of patients with PD was significantly and moderately negatively correlated with a range of variables: anxiety sensitivity, catastrophizing of physical symptoms, hypochondriacal fears and beliefs, anxiety, depression, and somatization. However, the results of partial correlation analyses when we controlled for anxiety sensitivity, hypochondriacal beliefs, and the catastrophizing of physical symptoms, respectively (Table 2), suggest that relationships between perceived health and anxiety, depression, somatization, and hypochondriacal fears may be better accounted for by the catastrophizing of physical symptoms and hypochondriacal beliefs. 123 These results also indicate that among people with PD, there is a significant overlap between negatively perceived health, catastrophizing of physical symptoms, and hypochondriacal beliefs. In other words, the more negatively people with PD perceive their health, the more likely they are to appraise physical consequences of their symptoms catastrophically and to harbor beliefs about having a serious disease. This may also be the case the other way around, in all possible combinations of these variables. Since the correlations do not imply causation, we cannot speculate about the chronological order in the relationships between these variables. The results of partial correlation analyses when hypochondriacal beliefs and the catastrophizing of physical symptoms were controlled for did not confirm a significant relationship between perceived health and anxiety Cogn Ther Res (2009) 33:323–333 sensitivity in PD. This finding is somewhat different from previous research results (Gregor et al. 2005; McLeish et al. 2006; Schmidt and Telch 1997; Yartz et al. 2005), although the latter relied only on standard correlation analyses. This may suggest some conceptual difference and less overlap between negatively perceived health and anxiety sensitivity; alternatively, there may be a significant relationship and greater overlap between negatively perceived health and one component of anxiety sensitivity— fear of anxiety-related sensations, as measured by the Physical Concerns subscale of the ASI. Unlike Gregor et al. (2005) who reported that perceived health had a medium strength of association with PD severity (r = -0.39), we did not find a significant correlation (r = -0.26) between perceived health and overall PD severity. This can partially be attributed to the different instruments used in our study (the General Health subscale of the SF-36 and the PAS) compared with the study by Gregor et al. [Perceived Physical Health subscale of the General Health Survey (Stewart et al 1988) and the PDSS]. Another difference pertains to the way the participants were recruited: our sample consisted of people who sought treatment for PD, whereas Gregor et al. recruited local community members with PD through advertisements. It is possible that because of this recruitment difference, our sample was characterized both by a more severe PD and by more negatively perceived health than the sample in the Gregor et al. study, yet we did not find a significant relationship between negatively perceived health and PD severity. Whether our failure to find this relationship in a clinical sample reflects a true lack of such a relationship is a possibility that requires further study. The fact that we did not find significant associations between negatively perceived health and any of the domains of PD severity (frequency, severity and duration of panic attacks, severity of anticipatory anxiety, extent of agoraphobic avoidance, and disability levels) may also support the idea that no significant relationship exists between negatively perceived health and the overall severity of PD. Besides perceived health, anxiety sensitivity, hypochondriacal beliefs, and catastrophizing of physical symptoms also failed to be significantly associated with overall PD severity. However, anxiety sensitivity and hypochondriacal beliefs were significantly related to one aspect of PD severity, that is, the severity of anticipatory anxiety. In the full sample, they were both significantly correlated with the severity of anticipatory anxiety and were the only independent predictors of it; in a subsample of patients without any current medical conditions, anxiety sensitivity remained the only significant predictor of the severity of anticipatory anxiety, while hypochondriacal beliefs were still similarly correlated with the severity of anticipatory anxiety. This is not surprising, because the 331 essence of anticipatory anxiety is the fear of panic attacks and their consequences, which largely overlaps with anxiety sensitivity and to some degree, also with hypochondriacal beliefs. Our finding that negatively perceived health was not significantly associated with anticipatory anxiety casts some doubt on the utility of perceived health compared with anxiety sensitivity and hypochondriacal beliefs at the time when patients present to the clinic and before they begin treatment. Stated otherwise, anxiety sensitivity and hypochondriacal beliefs initially seemed to provide meaningful information on one important aspect of PD, whereas negatively perceived health did not appear to be useful in this regard. Although the present findings do not suggest that perceived health is a particularly distinct or clinically useful construct, a number of questions remain unanswered. One is to what extent perceived health predicts the treatment response and longer-term prognosis for PD. The 6-month treatment follow-up data of Schmidt and Telch (1997) provide a preliminary suggestion that more positively perceived health may be a predictor of successful treatment and of the maintenance of treatment gains. However, studies with longer-term follow-up are needed. Further research might also investigate whether there is any link between negatively perceived health and subsequent mortality in PD patients and if so, whether this link is independent from the effects of other cognitive variables. Our study has several limitations. The first pertains to the instrument that we used for the assessment of perceived health. To the best of our knowledge, the General Health subscale of the SF-36 has not been used in studies of perceived health in PD. Therefore, it is difficult to make direct comparisons with results of other studies, which were based on a different, although related instrument, the General Health Survey. However, the General Health Survey was not used in a consistent manner across the studies, and the more fundamental problem appears to be a lack of consensus on the conceptualization of perceived health and on how best to measure it. One way of improving assessment of perceived health would be to obtain additional information by means of a structured interview, instead of relying solely on self-report instruments. Second, we do not know whether disorders cooccurring with PD, although not as prevalent as in some other clinical samples of people with PD, might have affected perceived health and other variables. Considering that there is a relationship between depression and anxiety sensitivity (e.g., Taylor et al. 1996) and depression and hypochondriacal phenomena (e.g., Demopulos et al. 1996), future studies may benefit from investigating perceived health and other related variables in samples of the relatively ‘‘pure’’ PD individuals, who do not have a concurrent depressive disorder. At least as far as medical 123 332 conditions are concerned, our findings suggest that their presence is not necessarily associated with the more negatively perceived overall health in people with PD, nor does it appear to influence significantly the relationships among the aforementioned variables. Third, ours was a cross-sectional study, unable to investigate how perceived health and other cognitive variables may change in PD individuals over time and affect one another, with and without treatment. Further research in this area should use a longitudinal design. Finally, it is uncertain to what extent our findings, derived from a clinical study of PD patients, can be generalized to non-clinical populations; therefore, the results of this study should best be interpreted within a clinical context. In summary, we found a significant overlap between negatively perceived health, the catastrophizing of physical symptoms, and hypochondriacal beliefs among PD patients. Negatively perceived health was not significantly associated with the overall severity of PD and was not significantly related to any domain of PD severity. In contrast, anxiety sensitivity and hypochondriacal beliefs were significantly related to the severity of anticipatory anxiety before the beginning of treatment, and in this domain they seem to exhibit greater clinical utility than perceived health. Future research should investigate whether perceived health might be clinically useful in other domains of PD, and in a way that is independent of other related variables. References Allgulander, C., & Lavori, P. W. (1991). Excess mortality among 3302 patients with ‘‘pure’’ anxiety neurosis. Archives of General Psychiatry, 48, 599–602. Arrindell, W. A. (1993a). The fear of fear concept: Stability, retest artefact and predictive power. Behaviour Research and Therapy, 31, 139–148. Arrindell, W. A. (1993b). The fear of fear concept: Evidence in favour of multidimensionality. Behaviour Research and Therapy, 31, 507–518. Bach, M., Nutzinger, D. O., & Harti, L. (1996). Comorbidity of anxiety disorders and hypochondriasis considering different diagnostic systems. Comprehensive Psychiatry, 37, 62–67. Bandelow, B. (1995). The assessment of efficacy of treatments for panic disorder and agoraphobia. II. The Panic and Agoraphobia Scale. International Clinical Psychopharmacology, 10, 73–82. Bandelow, B. (1999). Panic and agoraphobia scale (PAS). Seattle, WA: Hogrefe & Huber Publishers. Barsky, A. J., Barnett, M. C., & Cleary, P. D. (1994). Hypochondriasis and panic disorder: Boundary and overlap. Archives of General Psychiatry, 51, 918–925. Benedetti, A., Perugi, G., Toni, C., Simonetti, B., Mata, B., & Cassano, G. B. (1997). Hypochondriasis and illness phobia in panic-agoraphobic patients. Comprehensive Psychiatry, 38, 124–131. 123 Cogn Ther Res (2009) 33:323–333 Bouman, T. K., & Visser, S. (1998). Cognitive and behavioral treatment of hypochondriasis. Psychotherapy and Psychosomatics, 67, 214–221. Chambless, D. L., Caputo, G. C., Bright, P., & Gallagher, R. (1984). Assessment of ‘‘fear of fear’’ in agoraphobics: The Body Sensations Questionnaire and the Agoraphobic Cognitions Questionnaire. Journal of Consulting and Clinical Psychology, 52, 1090–1097. Coryell, W., Noyes, R., & Clancy, J. (1982). Excess mortality in panic disorder: A comparison with primary unipolar depression. Archives of General Psychiatry, 39, 701–703. Coryell, W., Noyes, R., & House, J. D. (1986). Mortality among outpatients with anxiety disorders. American Journal of Psychiatry, 143, 508–510. Demopulos, C., Fava, M., McLean, N. E., Alpert, J. E., Nierenberg, A. A., & Rosenbaum, J. F. (1996). Hypochondriacal concerns in depressed outpatients. Psychosomatic Medicine, 58, 314–320. Derogatis, L. R. (1994). Symptom checklist-90-R: Administration, scoring and procedures manual. Minneapolis, MN: NCS Pearson. Derogatis, L. R., Rickels, K., & Rock, A. (1976). The SCL-90R and the MMPI: A step in the validation of a new self-report scale. British Journal of Psychiatry, 128, 280–289. Furer, P., Walker, J. R., Chartier, M. J., & Stein, M. B. (1997). Hypochondriacal concerns and somatization in panic disorder. Depression and Anxiety, 6, 78–85. Gregor, K. L., Zvolensky, M. J., & Yartz, A. R. (2005). Perceived health among individuals with panic disorder: Associations with affective vulnerability and psychiatric disability. Journal of Nervous and Mental Disease, 193, 697–699. Horowitz, L. M., Rosenberg, S. E., Baer, B. A., Ureno, G., & Villasenor, V. S. (1988). Inventory of interpersonal problems: Psychometric properties and clinical applications. Journal of Consulting and Clinical Psychology, 56, 885–892. Idler, E. L., & Angel, R. J. (1990). Self-rated health and mortality in the NHANES-I epidemiologic follow-up study. American Journal of Public Health, 80, 446–452. Kaplan, G. A., & Camacho, T. (1993). Perceived health and mortality: Nine-year follow-up of the Human Population Laboratory cohort. American Journal of Epidemiology, 117, 292–304. Kawachi, I., Colditz, G. A., Ascherio, A., Rimm, E. B., Giovannucci, E., Stampfer, M. J., et al. (1994). Prosepective study of phobic anxiety and risk of coronary heart disease in men. Ciculation, 89, 1992–1997. Kellner, R. (1986) Somatization and hypochondriasis. New York: Praeger. Kellner, R., Abbott, P., Winslow, W. W., & Pathak, D. (1987). Fears, beliefs, and attitudes in DSM-III hypochondriasis. Journal of Nervous and Mental Disease, 179, 647–655. Lecrubier, Y., Sheehan, D. V., Weiller, E., Amorim, P., Bonora, I., Sheehan, K. H., et al. (1997). The Mini International Neuropsychiatric Interview (MINI). A short diagnostic structured interview: Reliability and validity according to the CIDI. European Psychiatry, 12, 224–231. Lipowski, Z. J. (1988). Somatization: The concept and its application. American Journal of Psychiatry, 145, 1358–1368. McLeish, A. C., Zvolensky, M. J., Bonn-Miller, M. O., & Barnstein, A. (2006). Perceived health moderates the association between smoking rate and panic vulnerability variables among daily smokers. Depression and Anxiety, 23, 257–265. Mossey, J. M., & Shapiro, E. (1982). Self-rated health: A predictor of mortality among the elderly. American Journal of Public Health, 71, 800–808. Otto, M. W., & Pollack, M. H. (1994). Panic disorder and hypochondriacal concerns: A reply to Taylor. Journal of Anxiety Disorders, 8, 101–103. Cogn Ther Res (2009) 33:323–333 Otto, M. W., Pollack, M. H., Sachs, G. S., & Rosenbaum, J. F. (1992). Hypochondriacal concerns, anxiety sensitivity, and panic disorder. Journal of Anxiety Disorders, 6, 93–104. Peterson, R. A., & Heibronner, R. (1987). The anxiety sensitivity index: Construct validity and factor analytic structure. Journal of Anxiety Disorders, 1, 117–121. Peveler, R. C., & Fairburn, C. G. (1990). Measurement of neurotic symptoms by self-report questionnaire: Validity of the SCL-90R. Psychological Medicine, 20, 873–879. Pinninti, N. R., Madison, H., Musser, E., & Rissmiller, D. (2003). MINI international neuropsychiatric schedule: Clinical utility and patient acceptance. European Psychiatry, 18, 361–364. Reiss, S., & McNally, R. J. (1985). Expectancy model of fear. In: S. Reiss, & R. R. Bootzin (Eds.), Theoretical issues in behavior therapy (pp. 107–121). New York: Academic Press. Reiss, S., Peterson, R. A., Gursky, D. M., & McNally, R. J. (1986). Anxiety sensitivity, anxiety frequency, and the prediction of fearfulness. Behaviour Research and Therapy, 24, 1–8. Schmidt, N. B., Joiner, T. E., Staab, J. P., & Williams, F. M. (2003). Health perceptions and anxiety sensitivity in patients with panic disorder. Journal of Psychopathology and Behavioral Assessment, 25, 139–145. Schmidt, N. B., & Telch, M. J. (1997). Nonpsychiatric medical comorbidity, health perceptions, and treatment outcome in patients with panic disorder. Health Psychology, 16, 114–122. Schmidt, N. B., Telch, M. J., & Joiner, T. E. (1996). Factors influencing health perceptions in patients with panic disorder. Comprehensive Psychiatry, 37, 253–260. Shear, M. K., Brown, T. A., Barlow, D. H., Money, R., Sholomskas, D. E., Woods, S., et al. (1997). Multicenter collaborative Panic Disorder Severity Scale. American Journal of Psychiatry, 154, 1571–1575. Sheehan, D., Janavs, J., Baker, R., Harnett-Sheehan, K., Knapp, E., Sheehan, M., et al. (1999). The Mini-International Neuropsychiatric Interview (M.I.N.I). English Version 5.0.0. DSM-IV. Journal of Clinical Psychiatry, 60(Suppl. 18), 39–62. Sheehan, D. V., Lecrubier, Y., Sheehan, K. H., Amorim, P., Janavs, J., Weiller, E., et al. (1998). The Mini-International Neuropsychiatric Interview (M.I.N.I): The development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. Journal of Clinical Psychiatry, 59(Suppl. 20), 22–33. Sheehan, D. V., Lecrubier, Y., Sheehan, K. H., Janavs, J., Weiller, E., Bonara, L. I., et al. (1997). Reliability and validity of the MINI International Neuropsychiatric Interview (MINI) according to the SCID-P. European Psychiatry, 12, 232–241. 333 Smoller, J. W., Pollack, M. H., Wassertheil-Smoller, S., Jackson, R. D., Oberman, A., Wong, N. D., et al. (2007). Panic attacks and risk of incident cardiovascular events among postmenopausal women in the Women’s Health Initiative Observational Study. Archives of General Psychiatry, 64, 1153–1160. Starcevic, V. (2001). Clinical features and diagnosis of hypochondriasis. In: V. Starcevic, & D. R. Lipsitt (Eds.), Hypochondriasis: Modern perspectives on an ancient malady (pp. 21–60). New York: Oxford University Press. Starcevic, V., Kellner, R., Uhlenhuth, E. H., & Pathak, D. (1992). Panic disorder and hypochondriacal fears and beliefs. Journal of Affective Disorders, 24, 73–85. Stephenson, R., Marchand, A., & Lavallée, M.-C. (1999). A Canadian French adaptation of the Agoraphobic Cognitions Questionnaire: Cross-cultural validation and gender differences. Scandinavian Journal of Behavior Therapy, 28, 58–69. Stewart, A. L., Hays, R. D., & Ware, J. E. (1988). The MOS shortform general health survey. Medical Care, 26, 724–735. Taylor, S. (1994). Comment on Otto et al. (1992): Hypochondriacal concerns, anxiety sensitivity, and panic disorder. Journal of Anxiety Disorders, 8, 97–99. Taylor, S. (1995). Panic disorder and hypochondriacal concerns: Reply to Otto and Pollack (1994). Journal of Anxiety Disorders, 9, 87–88. Taylor, S., Koch, W. J., Woody, S., & McLean, P. (1996). Anxiety sensitivity and depression: How are they related? Journal of Abnormal Psychology, 105, 474–479. Telch, M. J., Shermis, M. D., & Lucas, J. A. (1989). Anxiety sensitivity: Unitary personality trait or domain specific appraisals? Journal of Anxiety Disorders, 3, 25–32. Ware, J. E., & Sherbourne, C. D. (1992). The MOS 36-item shortform health survey (SF-36). Medical Care, 30, 473–483. Ware, J. E., Snow, K. K., Kosinki, M., & Gandek, B. (1993). SF-36 Health survey: Manual and interpretation guide. Boston, MA: The Health Institute. Ware, J., Kosinki, M., & Gandek, B. (2000). SF-36 Health survey: Manual and interpretation guide. Lincoln: Quality Metric Incorporated. Weissman, M. M., Markowitz, J. S., Ouellette, R., Greenwald, S., & Kahn, J. P. (1990). Panic disorder and cardiovascular/cerbrovascular problems: Results from a community survey. American Journal of Psychiatry, 147, 1504–1508. Yartz, A. R., Zvolensky, M. J., Gregor, K., Feldner, M. T., & LeenFeldner, E. W. (2005). Health perception is a unique predictor of anxiety symptoms in non-clinical participants. Cognitive Behaviour Therapy, 34, 65–74. 123