A PSYCHOMETRIC AND CLINCIAL INVESTIGATION OF ANXIETY SENSITIVITY IN ANXIETY DISORDERS Kerry Ann Armstrong, B.SocSc (Psych) (Hons) Dissertation Submitted to the Queensland University of Technology for the Degree of Doctor of Philosophy School of Psychology and Counselling Faculty of Health 2004
327
Embed
A PSYCHOMETRIC AND CLINCIAL INVESTIGATION OF ANXIETY …eprints.qut.edu.au/15931/1/Kerry_Armstrong_Thesis.pdf · Revised in Australian Clinical and Normative Populations. European
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
A PSYCHOMETRIC AND CLINCIAL INVESTIGATION OF ANXIETY SENSITIVITY IN ANXIETY DISORDERS
Kerry Ann Armstrong, B.SocSc (Psych) (Hons)
Dissertation Submitted to the Queensland University of Technology for the Degree of Doctor of Philosophy School of Psychology and Counselling
Faculty of Health
2004
ii
iii
ACKNOWLEDGEMENTS
There have been many people who have provided their support, assistance
and understanding whilst I completed this degree. I would like to take this
opportunity to say thank you to everyone who helped me along the way.
First and foremost I would like to say thank you to my principal supervisor
Dr Nigar Khawaja, whom without her supervision, support and guidance this
dissertation would not be possible. Nigar is perhaps the most generous and
professional person that I have had the pleasure of working with and I have learned a
great deal from her, as both a student and from her supervising me for registration as
a Psychologist.
I would also like to thank Professor Tian Oei from the University of
Queensland, not only for being an associate supervisor but also for allowing me to
collect data from his CBT clinic. Tian was always available for a supervision
meeting when I needed one and was helpful and kind in his encouragement in all
aspects of the dissertation. Also, a thank you goes to Ms Sue Fell from the Toowong
CBT clinic for all the support and encouragement she offered to me during the
course of the PhD, as well as all of her assistance in the collection of data.
A big thank you to Professor Mary Sheehan, for not only being an associate
supervisor, but for all the assistance and support she showed me when unforseen
circumstances arose.
I would like to thank the late Professor Larry Evans for his assistance in
collecting data from his clinic. Larry would always take the time to ask me how I
iv
was coming along in the dissertation and was very generous in helping me in my
data collection efforts. I would also like to thank Dr Stephen Cox for all his
assistance in his role as Statistics Advisor in the Confirmatory Factor Analysis
section of this PhD.
Further, I don’t think an acknowledgements page would be complete without
a note made regarding the ‘behind the scenes’ team that offer their encouragement
and support on a daily basis.
First, I would like to thank my fiancé Dan for always taking an interest in
what I do, for listening to me when I talk, and for understanding that a dissertation is
not a Monday to Friday, 9 to 5 job.
Thank you also to Kim Johnston who, although she has her own PhD to
complete, has always taken the time to talk to me about my dissertation, help me sort
out my thinking and even enter data. Kim was always available, at any time of day
or night, to help in anyway she could.
I would like to thank all the staff from the Centre for Accident Research and
Road Safety – Queensland (CARRS-Q) for their assistance, encouragement, and
understanding over the years. In particular I would like to thank Cynthia Schonfeld,
who has supported me in every way since day one. I would also like to thank
Dianne Jensen for being ‘Aunty Di’ and for showing me how to format documents
properly.
Finally, I would like to thank the Post Grad Community, in particular Eve
Dwyer and Trish Obst for reading the final draft; as well as Jenny Summerville,
v
Collette Roos, Jane Shakespeare-Finch, Katina Damoulius, and Leith Harding for
being the supportive people that they are and always being available for anything.
vi
vii
CERTIFICATION OF ORIGINAL WORK
I, Kerry Ann Armstrong, certify that to the best of my knowledge this dissertation,
which is submitted in partial fulfilment of the requirements of Doctor of Philosophy
degree undertaken at the Queensland University of Technology, is my own work,
except as acknowledged in the text. The material contained within has not been
submitted, in whole or in part, for a degree at this or any other university.
Publications arising from this dissertation are specified.
Signed……………………………………………………………………………….. Kerry Ann Armstrong School of Psychology and Counselling Faculty of Health Queensland University of Technology February, 2004
viii
ix
PARTS OF THE THESIS SUBMITTED FOR PUBLICATION OR PRESENTED AT CONFERENCES
Papers Submitted for Publication Armstrong, K. A., Khawaja, N. G., & Oei, T. P. S. (Submitted). Confirmatory Factor Analysis and Psychometric Properties of the Anxiety Sensitivity Index – Revised in Australian Clinical and Normative Populations. European Journal of Psychological Assessment. Papers Presented at Conferences International Armstrong, K. A., & Khawaja, N. G. (2003). An investigation of Anxiety Sensitivity: Confirmatory factor analysis and psychometric properties of the 21- item Anxiety Sensitivity Index. Poster Presented at the 37th Annual Convention for the Association for Advancement of Behavior Therapy, Boston, MA. National Armstrong, K. A., & Khawaja, N. G. (2003). Anxiety Sensitivity and Differences in Diagnostic Groups. Paper presented as part of a symposium entitled ‘Questionnaires: Are they of any use?’ at the Australian Association for Cognitive Behavioural Therapy, Adelaide, 42, South Australia. Armstrong, K. A., & Khawaja. N. G. (2002). An investigation of the Anxiety Sensitivity Index – Revised (ASI-R) in a clinical population. Australian Journal of Psychology, 54 (Supplement), (Combined Abstracts for the Australian Psychology Conference), Gold Coast, 10, Queensland. Armstrong, K. A., & Khawaja. N. G. (2002). An investigation of Anxiety Sensitivity: Confirmatory factor analysis and psychometric properties of the Anxiety Index – Revised (ASI-R). Australian Association for Cognitive and Behavioural Therapy, 25th National Conference, Brisbane, 41, Queensland.
x
xi
ABSTRACT Anxiety sensitivity is a cognitive, individual difference variable that is differentiated
by an individual’s fear of anxiety sensations and centred on the belief that such
sensations result in harmful consequences. In order to test anxiety sensitivity, Reiss,
Peterson, Gursky, and McNally (1986) developed the Anxiety Sensitivity Index
(ASI). However, one contentious issue in the area concerns the factor analytic
structure of anxiety sensitivity and this has important consequences for the construct.
Numerous investigations have been conducted using the ASI, and the results have
varied appreciably with some researchers arguing for a unidimensional construct.
However the general consensus now is that anxiety sensitivity is multidimensional.
It has been argued that the repeated attempts to clarify the dimensionality of anxiety
sensitivity, using the 16- item ASI, is problematic because the scale was never
designed to measure a multidimensional construct in the first instance. Thus, the
objective of the dissertation was to critically examine the anxiety sensitivity
construct by using an expanded, multidimensional measure of anxiety sensitivity
referred to as the Anxiety Sensitivity Index – Revised ([ASI-R] Taylor & Cox, 1998)
and establish the psychometric properties of the measure by conducting a series of
empirical investigations to assess the clinical utility of the measure.
A series of three empirical investigations are presented in the current
dissertation. The first investigation aimed to critically examine the factor structure
and psychometric properties of the ASI-R. Confirmatory factor analysis using a
clinical sample of adults revealed that the ASI-R could be improved substantially
through the removal of 15 problematic items in order to account for the most robust
xii
dimensions of anxiety sensitivity. The modified measure was re-named the 21- item
Anxiety Sensitivity Index (21- item ASI) and re-analysed with a large sample of
nonclinical adults, revealing configural and metric invariance across groups.
Further, comparisons with other alternative models that also include comparisons
with previous published ASI models indicated the 21- item ASI to be the best fitting
model for both groups. There was also evidence of internal consistency, test-retest
reliability, and construct va lidity for both samples. The aim of the second
investigation was to critically examine differences between and within various
anxiety classifications, a mood disorder classification, and a nonclinical control
sample, with respect to both general and specific dimensions of anxiety sensitivity as
identified by the 21- item ASI. In most instances, the results revealed that the
differences between and within the diagnostic groups were consistent with
theoretical expectations. Finally, the third investigation aimed to examine
differences within each diagnostic category before and after cognitive behavioural
therapy in order to provide a further test of validity for the revised 21- item ASI. The
results revealed significant differences within all but one diagnostic group on the pre
and post-treatment scores, using the global and specific dimensions of the 21-item
ASI.
The strengths, theoretical contribution, limitations, and directions for future
research are discussed. It is concluded that the overall findings relating to the series
of empirical investigations presented in the current dissertation make a significant
and valid theoretical contribution to the field of anxiety sensitivity in particular, and
anxiety research in general, by enhancing our understand ing of anxiety sensitivity
xiii
and how the 21- item ASI can be used to improve therapeutic interventions in clinical
practice.
xiv
xv
TABLE OF CONTENTS
Acknowledgements iii Certification of Original Work vii Parts of Dissertation Submitted for Publication and
Presented at Conferences ix Abstract xi Table of Contents xv List of Tables xxi List of Figures xxiii Introduction xxv Chapter 1: Theoretical approaches to the fear of anxiety: Introduction to
Chapter 2: Review of the literature on anxiety sensitivity 2.1 Introduction 15
2.2 Introduction to the Anxiety Sensitivity Index 15 2.3 Anxiety Sensitivity and Trait Anxiety 18 2.4 Factor Analytic Investigations of the ASI 20 2.5 The Anxiety Sensitivity Index – Revised 40 2.6 Predictive Validity of Anxiety Sensitivity 48 2.7 Anxiety Sensitivity and Nonclinical Populations 51
2.8 Anxiety Sensitivity and Clinical Populations 54 2.8.1 Panic Disorde r with or without Agoraphobia 54 2.8.2 Posttraumatic Stress Disorder 58 2.8.3 Social Phobia 60 2.8.4 Obsessive-Compulsive Disorder and Generalised Anxiety Disorder 65 2.8.5 Specific Phobia 66 2.8.6 Major Depressive Disorder 68
2.9 Anxiety Sensitivity and Cognitive Behavioural Therapy 73 2.9.1 Anxiety Sensitivity and Combined Cognitive Behavioural and Pharmacological Therapy 77
2.10 Rationale for Study One 80 2.11 Rationale for Study Two 83 2.12 Rationale for Study Three 84 2.13 Chapter Summary 85
xvi
Chapter 3: Method
3.1 Participants 89 3.1.2 Participants from the clinical setting 89 3.1.3 Participants from the nonclinical setting 90
4.3 Results 111 4.3.1 Clinical and nonclinical groups CFA 111 4.3.2 Model modifications of Taylor and Cox’s (1998) Model 114 4.3.3 Test of Configural Invariance of Modified Model 116 4.3.4 Test of Metric of Modified Model 116 4.3.5 Comparisons of Alternative Models 121 4.3.6 Comparisons of the 21- item ASI and competing ASI models 123 4.3.7 Internal consistency and test-retest reliability of the 21-item ASI 127 4.3.8 Concurrent Validity of the 21- item ASI 128 4.3.9 Discriminant Validity of the 21-item ASI 130
4.4 Discussion 133 4.5 Chapter Summary 137
xvii
Chapter 5: Differences in anxiety sensitivity between and within diagnostic and nonclinical groups using the 21-item Anxiety Sensitivity Index
5.1 Introduction 141 5.2 Method 144
5.2.1 Participants 144 5.2.2 Design 147
5.2.3 Measures 148 5.2.4 Procedure 148
5.3 Results 148 5.3.1 Data Screening and Cleaning 148 5.3.2 Preliminary Analyses 151 5.3.3 Tests of Hypotheses 153
5.3.3.1 Analysis of differences between individuals with a primary diagnosis of panic disorder, GAD, PTSD, depression, and nonclinical controls on the total score of the 21- item ASI 153
5.3.3.2 Analysis of differences between individuals with a primary diagnosis of panic disorder, GAD, PTSD, depression, and nonclinical controls on the specific dimensions of the 21-item ASI 155
5.3.3.3 Analysis of differences for the panic disorder diagnostic group on the first- order dimensions of anxiety sensitivity, as assessed by the 21- item ASI 164
5.3.3.4 Analysis of differences for the GAD diagnostic group on the first-order dimensions of anxiety sensitivity, as assessed by the 21- item ASI 166
5.3.3.5 Analysis of differences for the PTSD diagnostic group on the first-order dimensions of anxiety sensitivity, as assessed by the 21- item ASI 167
5.3.3.6 Analysis of differences for the depression diagnostic group on the first-order dimensions of anxiety sensitivity, as assessed by the 21- item ASI 169
5.3.3.7 Analysis of differences for the nonclinical control group on the first-order dimensions of anxiety sensitivity, as assessed by the 21-item ASI 171
5.4 Discussion 173 5.5 Chapter Summary 183
xviii
Chapter 6: Examination of differences within diagnostic categories after cognitive behavioural therapy using the 21-item Anxiety Sensitivity Index
6.1 Introduction 189 6.2 Method 191
6.2.1 Participants 191 6.2.2 Measures 194
6.2.3 Design 194 6.2.4 Treatment 194
6.3 Results 195 6.3.1 Data Screening and Cleaning 195 6.3.2 Reliability Analyses of Scale 197 6.3.3 Preliminary Analyses 197 6.3.4 Tests of Hypotheses 200
6.3.4.1 Analysis of differences for each diagnostic group before and after CBT for overall level of anxiety sensitivity 200
6.3.4.2 Analysis of differences for the panic disorder group before and after CBT for the first-order dimensions of anxiety sensitivity, as assessed by the 21-item ASI 201
6.3.4.3 Analysis of differences for the GAD group before and after CBT for the first-order dimensions of anxiety sensitivity, as assessed by the 21- item ASI 203
6.3.4.4 Analysis of differences for the PTSD group before and after CBT for the first-order dimensions of anxiety sensitivity, as assessed by the 21- item ASI 205
6.3.4.5 Analysis of differences for the depression group before and after CBT for the first- order dimensions of anxiety sensitivity, as assessed by the 21- item ASI 206
6.3.4.6 Analysis of differences on the BAI, CCQ-M, FQ and Zung – SDS after CBT 208
Appendix D: Questionnaire Booklet for clinical and nonclinical Participants’ 261
Appendix E: List of Contents of Cognitive Behavioural Therapy Manuals for Anxiety Disorders at the Toowong Private Hospital 277
Appendix F: List of Contents of Cognitive Behavioural Therapy Manuals for Depression at the Toowong Private
Hospital 283 Appendix G: Telephone Screening Sheet for Clinical Participants
Recruited from Advertisements in Local Media 291
Appendix H: Non-published handbook for Information about Anxiety Disorders prepared by Dr Nigar Khawaja and Kerry Ann Armstrong 297
Appendix I: Instructions for participants from First Year Subject Pool 327
xx
xxi
LIST OF TABLES
Table 2.1 Factor analytic studies of the Anxiety Sensitivity Index 21 Table 2.2 Factor analytic studies of the Anxiety Sensitivity Index-
Revised 41
Table 4.1 Clinical group CFA of three hypothesised ASI-R models 113
Table 4.2 Nonclinical group CFA of three hypothesised ASI-R models 114
Table 4.3 Nonclinical group CFA of the 21-item ASI Hierarchical model 116
Table 4.4 Metric Invariance CFA statistics between the clinical and nonclinical group datasets using the 21- item ASI 117
Table 4.5 Clinical groups CFA of the unifactorial, orthogonal 4- factor model and 3-factor hierarchical models of the 21- item ASI 121
Table 4.6 Nonclinical groups CFA of the unifactorial, orthogonal 4-factor model and 3-factor hierarchical models of the 21- item ASI 123
Table 4.7 Clinical groups CFA of alternative hypothesised ASI models 125
Table 4.8 Nonclinical groups CFA of alternative hypothesised ASI models 126
Table 4.9 Internal Consistency and Test-Retest Reliability of the 21-item ASI for the Clinical and Nonclinical Groups 127
Table 4.10 Pearson Correlations among the 21- item ASI, 16- item ASI, BAI, CCQ-M, FQ, Zung-SDS, and DASS-Stress Scale for the clinical group 129
Table 4.11 Pearson Correlations among the 21- item ASI, 16- item ASI, BAI, CCQ-M, FQ, Zung-SDS, and DASS-Stress Scale for the nonclinical group 130
Table 4.12 Pearson Correlations among the 21- item ASI, Self-Efficacy, and COPE questionnaires for the clinical group 131
Table 4.13 Pearson Correlations among the 21- item ASI, Self-Efficacy, and COPE questionnaires for the non clinical group 132
xxii
Table 4.14 Between Groups Differences of the 21- item ASI and first-order dimensions: Means, Standard Deviations and Univariate Results for the Clinical and Nonclinical groups 133
Table 5.1 Means, Standard Deviations and Age Range for diagnostic groups 146
Table 5.2 Sample size of Demographic Variables for the diagnostic groups 147
Table 5.3 Means and Standard Deviations for the Diagnostic and Nonclinical Control Groups on the 21-item ASI dimensions 152
Table 6.1 Means, Standard Deviations and Age Range for d iagnostic groups 192
Table 6.2 Sample size of Demographic Variables for the diagnostic groups 193
Table 6.3 Internal Consistency of the 21- item ASI for the Clinical Group Pre and Post-Cognitive-Behavioural Therapy 197
Table 6.4 Means and Standard Deviations for the Diagnostic Groups before and after Cognitive-Behavioural Therapy 199
xxiii
LIST OF FIGURES
Figure 4.1 Clinical group factor loadings for the items of the 21- item ASI with a hierarchical model of four first-order dimensions and a general higher-order dimension 119 Figure 4.2 Nonclinical group factor loadings for the items of the 21-item ASI with a hierarchical model of four first-order dimensions and a general higher-order dimension 120 Figure 5.1 Mean differences and standard error between panic disorder,
GAD, PTSD, depression, and nonclinical control groups and the 21-item ASI total score 155
Figure 5.2 Mean differences and standard error between panic disorder, GAD, PTSD, depression, and nonclinical control groups and the 21-item ‘Fear of Respiratory Symptoms’ dimension 158
Figure 5.3 Mean differences and standard error between panic disorder, GAD, PTSD, depression, and nonclinical control groups and the 21-item ‘Fear of Publicly Observable Symptoms’ dimension 160
Figure 5.4 Mean differences and standard error between panic disorder, GAD, PTSD, depression, and nonclinical control groups and the 21-item ‘Fear of Cardiovascular/Stroke Symptoms’ dimension 162
Figure 5.5 Mean differences and standard error between panic disorder, GAD, PTSD, depression, and nonclinical control groups and the 21-item ‘Fear of Cognitive Dyscontrol’ dimension 163
Figure 5.6 Mean differences and standard error for the panic disorder group and the four first-order dimensions of the 21-item ASI 165
Figure 5.7 Mean differences and standard error for the GAD group and the four first-order dimensions of the 21- item ASI 167
Figure 5.8 Mean differences and standard error for the PTSD group and the four first-order dimensions of the 21- item ASI 169
Figure 5.9 Mean differences and standard error for the depression group and the four first-order dimensions of the 21- item ASI 171
xxiv
Figure 5.10 Mean differences and standard error for the nonclinical control group and the four first-order dimensions of the 21-item ASI 173
Figure 6.1 Mean differences and standard error on the 21-item ASI total score for the panic disorder, GAD, PTSD and depression diagnostic groups before and after Cognitive- Behavioural Therapy 201
Figure 6.2 Mean differences and standard error for the panic disorder group on the four first order dimensions of the 21- item ASI before and after Cognitive-Behavioural Therapy 203
Figure 6.3 Mean differences and standard error for the GAD group on the four first order dimensions of the 21-item ASI before and after Cognitive-Behavioural Therapy 205
Figure 6.4 Mean differences and standard error for the PTSD group on the four first order dimensions of the 21-item ASI before and after Cognitive-Behavioural Therapy 206
Figure 6.5 Mean differences and standard error for the Depression group on the four first order dimensions of the 21-item ASI before and after Cognitive-Behavioural Therapy 208
xxv
INTRODUCTION
Since the concept of anxiety sensitivity was initially presented as part of
Reiss and McNally’s (1985) interactional model of expectancy theory, it has
received considerable attention in the extant literature. According to Reiss’s
expectancy theory (Reiss & McNally, 1985), anxiety sensitivity is a cognitive,
individual difference variable that is characterised by the individual’s fear of anxiety
related sensations and based on the belief that such sensations result in harmful
consequences.
In order to test anxiety sensitivity, Reiss, Peterson, Gursky, and McNally
(1986) developed the Anxiety Sensitivity Index (ASI). Several investigations, using
the ASI, have revealed that anxiety sensitivity is capable of not only distinguishing
panic disorder from other anxiety disorders (see Apfledorf, Shear, Leon, & Portera,
1994); but also of predicting who will respond fearfully in panic provocation
procedures (see Rapee, Brown, Antony, & Barlow, 1992) as well as who will
develop panic attacks as part of prospective research (see Schmidt, Lerew, &
Jackson, 1997). Thus the construct validity of anxiety sensitivity, using the ASI, has
been well documented. The construct of anxiety sensitivity has also been linked to
other clinical problems such as major depression (Otto, Pollack, Fava, Uccello &
As a result of the numerous investigations into the classification of anxiety
sensitivity, a general consensus now exists within the available literature that
suggests anxiety sensitivity represents a genuine individual difference variable,
which is not only conceptually but also empirically distinct from trait anxiety.
While the above research appears to provide support for the usefulness of
the anxiety sensitivity construct as measured by the ASI, it is important to note
that one of the most contentious issues in the area of this construct concerns the
factor analytic structure of anxiety sensitivity. According to Reiss and
McNally’s (1985) Expectancy Theory, anxiety sensitivity is conceptualised as
unidimensional. Accordingly, Taylor, Koch, McNally, and Crockett (1992) have
argued that the ASI is comprised of a unidimensional factor structure. While
some research has been conducted that supports this view of anxiety sensitivity,
others have found support for a multidimensional structure, with all
investigations using varying sample sizes, factor analytic techniques as well as
varying rationales as to how factors are to be extracted. The following sections
presents the empirical research to date regarding the internal structure, items and
psychometric properties of the original ASI as well as the relatively new Anxiety
Sensitivity Index – Revised (ASI-R) measure developed by Taylor and Cox
(1998).
2.4 Factor Analytic Investigations of the ASI
The majority of factor analytic investigations of the ASI have employed
the method of exploratory factor analysis, with a limited number using
confirmatory factor analysis. This section will go through each investigation of
anxiety sensitivity in a chronological order. A table detailing each study in brief
is provided (see Table 2.1).
Anxiety Sensitivity 21
Table 2.1. Factor analytic studies of the Anxiety Sensitivity Index Author (s)
Sample Method of Factor Extraction
Method of Factor Rotation
Number of Factor (s)
Reiss et al. (1986) Two independent samples of university students (N = 49 and N = 98)
PCA None 1
Peterson and Heilbronner (1987)
122 self described ‘tense, anxious, or nervous’ university students
PCA Oblique and orthogonal 1
Telch, Shermis, and Lucas (1989)
840 university students PAF and PCA Varimax 4
Wardle, Ahmad, and Hayward (1990)
160 individuals with a diagnosis of agoraphobia and 120 nonclinical controls
PCA Varimax 4
Anxiety Sensitivity 22
Taylor, Koch, and Crockett (1991)
142 spider phobic university students and 93 individuals with a diagnosis of anxiety, depression or other disorder
PCA Oblique 1
Taylor et al. (1992)
142 spider phobic university students and 327 individuals with a diagnosis of an anxiety or stress-related disorder
CFA Oblique and orthogonal 4 (as previously identified
by Telch et al., 1989)
Cox, Parker, and Swinson (1996)
365 university students and 216 individuals with a diagnosis of panic disorder
CFA Orthogonal and oblique 4 (as previously identified by Wardle et al., 1990)
Taylor, Koch, Wood, and McLean (1996)
52 individuals with a diagnosis of panic disorder, 46 individuals with a diagnosis of major depression, and 37 individuals with a dual diagnosis of both
PCA Oblique 3
Zinbarg, Barlow, and Brown (1997)
432 individuals with a diagnosis of an anxiety or depressive disorder
PFA and CFA Oblique 3
Anxiety Sensitivity 23
Blais et al. (2001) 340 individuals with panic disorder, social phobia, or major depression and 50 university students
PCA Orthogonal and Oblique 2 (using 11 of the original
16 items of the ASI)
Schmidt and Joiner (2002)
233 individuals from the general community and 809 university students
PCA and PAF and CFA
Orthogonal and Oblique 2 (using 10 of the original
16 items of the ASI)
Note. PCA = Principal components analysis; PAF = Principal axis factor analysis; CFA = Confirmatory factor analysis.
Anxiety Sensitivity 24
In the original study of the ASI, Reiss et al. (1986) employed principal
components analysis (PCA) and reported a unidimensional solution, with 13 of
the 16-items loading .4 or more on the first factor and a high degree of inter-
relation between scale items. They concluded this study by arguing that the
results provided evidence that the ASI measures a coherent factor. However, this
investigation is flawed because they offered no rationale as to how many factors
were to be extracted as well as the very small sample size employed.
Peterson and Heilbronner (1987) investigated the factor analytic structure
of the ASI using a sample of 122 self-described ‘tense, anxious, or nervous’
university students who volunteered to take part in a relaxation study. Using
PCA, four factors were extracted using Kaiser’s criterion of eigenvalues ≥ 1 rule.
These four factors were subjected to an oblique simple structure rotation, in
which two factors associated with an a priori ‘fear of consequences’ and two
factors with an a priori ‘fear of physical sensations’ emerged. Using a specified
two-factor structure to further evaluate the a priori factors, an orthogonal rotation
was conducted which resulted in 11 of the 16-items loading on the first factor
and five items loading onto the second factor. Three items cross-loaded
substantially onto the first factor. Factor loadings for each analysis were not
provided and a rational was not offered for employing both rotational analyses.
As such, it is unclear what Peterson and Heilbronner (1987) were trying to show
as the implications for employing both rotational analyses are contradictory
because the oblique rotational analysis specifies that the dimensions of the ASI
are related whereas the orthogonal rotational analysis specifies that the
dimensions of the ASI are unrelated. Peterson and Heilbronner (1987) argued
that their results provided support for a strong single factor ASI and suggested
Anxiety Sensitivity 25
that only the total scale score should be used in future clinical and research work
due to the oblique rotation resulting in four factors with only a few items in each
factor and the high cross-loadings from the second factor of the two-factor
orthogonal rotation. The results of Peterson and Heilbronner (1987) study are
not definitive as they are vague as to how many factors should be retained as a
result of their conflicting rotational analyses and they concluded that the ASI was
best viewed as unidimensional when the unidimensional model was not tested.
In order to test the hypothesis that the ASI measured a unitary personality
variable, Telch, Shermis, and Lucas (1989) administered the ASI to 840
university students. In the first of two separate factor analyses of the ASI, Telch
et al. (1989) employed an interated principal axis factor (PAF) analysis with a
varimax rotation and Kaiser’s criterion of eigenvalues ≥ 1 as the factor extraction
rule. The procedure resulted in one-factor solution with all but three items
loading onto the principal factor. However, as the proportion of variance
explained by the unidimensional structure was relatively low (25%), PCA
followed by varimax rotation and Kaiser’s criterion of eigenvalues ≥ 1 was
performed in order to maximise the variance explained in the data. This second
procedure resulted in four factors capable of explaining 53.5% of the total
variance. The four factors obtained were (a) concern about physical sensations,
(b) concern about mental/cognitive incapacitation, (c) concern about loss of
control, and (d) concern about heart/lung failure. Factor three and four only
consisted of two and three items respectively and two items from factor two and
two items from factor four had moderate to high loadings on the first factor.
Telch et al. (1989) concluded by challenging the view in which ASI is regarded
as unidimensional and instead argued that the ASI measures several loosely
Anxiety Sensitivity 26
associated cognitive appraisal domains involved with the anticipated negative
consequences of anxiety. However, it should be noted that this interpretation
may be limited by the fact that Telch et al. (1989) imposed an orthogonal
rotation, in which dimensions are forced to be unrelated to one another. As such,
it would have been interesting, based on their conclusion, to investigate the factor
structure after an oblique rotation was applied and compare it to the orthogonal
solution.
Wardle, Ahmad, and Hayward (1990) assessed the factor structure of the
ASI in 160 individuals with agoraphobia and 120 nonclinical controls. Using
PCA with varimax rotation and Kaiser’s criterion of eigenvalues ≥ 1 rule for each
sample separately resulted in a four-factor solution that was capable of
explaining more than 60% of the total variance. For the agoraphobic sample, the
pattern of factor loadings was consistent with factors reflecting (a) fear of
physical sensations related to cardiovascular health, (b) fear of loss of mental
control, (c) fear of gastrointestinal upset, and (d) concern of publicly observable
anxiety symptoms. The item loadings for the nonclinical sample were regarded
as meaningless and uninterpretable. Examination of a two and three-factor
solution did not improve the meaningfulness of the internal structure of the ASI
and the four-factor solution was retained as it was regarded as a “very clear
factor structure” (Wardle et al., 1990, p. 330), despite the presence of six cross-
loading items in the agoraphobic sample and 11 cross-loading items in the
nonclinical sample. Wardle et al. (1990) concluded their investigation by
arguing that the findings supported a multidimensional view of anxiety
sensitivity without discussing the implications of the cross-loaded items.
Anxiety Sensitivity 27
Taylor, Koch, and Crockett (1991) examined the factor structure of the
ASI using 142 university students who indicated that they were spider phobic
according to the Fear Survey Schedule III (FSS-III) and 93 individuals from a
clinical setting who were diagnosed according to the DSM-III-R (APA, 1987)
criteria for anxiety, depression or other disorder. As Taylor et al. (1991) were
aware of the risk of over-extraction of factors using Kaiser’s criterion, they also
employed Cattell’s scree test as well as Thurstone’s criteria for simple structure
for determining the number of factors to retain in each sample. Using these three
rules of factor extraction in addition to PCA with oblique rotation, Taylor et al.
(1991) concluded that the ASI represented a unidimensional solution that
measures a fear of bodily sensations.
In another examination, Taylor et al. (1992) used confirmatory factor
analysis (CFA) to evaluate competing models previously published in the
available literature. The four models identified for inclusion in their study
included Reiss and McNally’s (1995) unifactorial solution, as well as Peterson
and Heilbronner (1987), Telch et al. (1989) and Wardle et al. (1990) four-factor
solutions. Taylor et al. (1992) examined responses from two samples including
142 spider-phobic university students and 327 individuals from a clinical setting
who met the DSM-III-R (APA, 1987) criteria for an anxiety or stress-related
disorder. For each four-factor model, Taylor et al. (1992) examined both the
oblique and orthogonal forms of each solution using a cut-off of at least .30 to
define a salient factor loading. Taylor et al. (1992) choose to retain the
orthogonal solutions because the oblique solutions for both the clinical and
student samples resulted in highly correlated factors (clinical sample range rs =
.34 to .92; student sample range rs = .03 to .97). They argued that the presence
Anxiety Sensitivity 28
of highly correlated factors indicated that the oblique factors tended to collapse
on one another. However it is not clear why the researchers rejected this solution
in favour of the orthogonal models as it is inconsistent with theoretical view of
anxiety sensitivity. Using only the orthogonal solutions and unifactorial model,
Taylor et al. (1992) identified the Telch et al. (1989) solution as the best-fitting
model for the clinical sample. It is important to note that examination of the chi-
square statistic as well as the Tucker-Lewis and Mulaik Normed Fit Index for all
models revealed that each model was well below the criteria of .8 to .9 for
retaining a specified model. In concluding their investigation, Taylor et al.
(1992) argued that the oblique solution could be seen as part of a hierarchical
factor structure in which all the oblique factors load onto a higher-order factor.
However, because the factors were highly correlated, Taylor et al. (1992) argued
they could best be regarded as measuring a single facet of anxiety sensitivity
because the four-factor model appeared to be the result of constraining the
factors to be orthogonal. Thus, these investigators proposed that the ASI
measures a single facet of anxiety sensitivity and is best regarded as unifactorial.
As such, it is unclear what Taylor et al’s. (1992) investigation revealed as their
findings and conclusions are confusing and arguably, not consistent with the
theory in which anxiety sensitivity is based.
In an investigation similar to Taylor et al. (1992), Cox, Parker, and
Swinson (1996) investigated the ASI by employing CFA to the unifactorial
model of Reiss and McNally (1985) and four-factor models of Peterson and
Heilbronner (1987), Telch et al. (1989) and Wardle et al. (1990). Each model
was tested using orthogonal and oblique solutions and using responses from two
samples consisting of 365 university students and 216 individuals who met the
Anxiety Sensitivity 29
DSM-IV (APA, 1994) criteria for panic disorder. Using multiple goodness-of-fit
indicators, Cox et al. (1996) reported the unidimensional model did not meet the
criteria standards for adequacy of fit in either the clinical or student sample.
Conversely, for both the clinical and student samples, the Peterson and
Heilbronner (1987) four-factor oblique model met all criteria for the goodness-
of-fit indices; however the chi-square statistic was significant. The orthogonal
model was not found to be an adequate fit of the data in either sample. The
Telch et al. (1989) four-factor oblique model was found to be a good fit of the
data in the student sample and an adequate fit of the clinical sample. Again, the
orthogonal model was not found to be a good fit of the data in either the clinical
or student sample. The four-factor oblique model based on Wardle et al.’s
(1990) agoraphobic sample met the criteria for the goodness-of-fit indices in the
student sample; however the chi-square statistic was significant. Neither the
orthogonal or oblique models were a good fit of the data for the clinical sample.
Finally, for both the clinical and student samples, the four-factor oblique and
orthogonal models based on Wardle et al’s. (1990) analysis with nonclinical
adults did not meet the criteria for retaining a specified model. Cox et al. (1996)
compared the goodness-of-fit indices for all three models that indicated
acceptable fit indices and found that the four-factor model of Peterson and
Heilbronner (1987) provided the best fit of the data in both the clinical and
student samples.
It is important to note that cut-off parameters for the fit indices Cox et al.
(1996) used were not as stringent as those used today. As such, not one of the
four models tested would have been retained as satisfying today’s goodness-of-fit
criteria. Cox et al’s. (1996) investigation found empirical support for a
Anxiety Sensitivity 30
multidimensional model of the ASI and they argued that there may be some
important domains within the anxiety sensitivity construct. However, it was
noted that one limitation to using the ASI in a multidimensional manner is that
there are too few items on the scale to produce reliable anxiety sensitivity
subscales. As such, Cox et al. (1996) argued that future work in the area of
anxiety sensitivity would be enhanced if the “ASI item pool was expanded to
allow for reliable subscales to be developed” (p. 596).
Taylor, Koch, Woody, and McLean (1996) examined the relationship
between anxiety sensitivity and depression in 52 individuals who met the DSM-
III-R (APA, 1987) criteria for panic disorder, 46 who met the criteria for major
depression and 37 individuals who met the criteria for both. Using the rule of
parallel analysis in order to determine the number of factors to retain, Taylor et
al. (1996) used PCA with oblique rotation and reported a three-factor solution
that accounted for 59% of the total variance. The three factors retained were (a)
fear of publicly observable arousal-related reactions, (b) phrenophobia or fear of
cognitive dyscontrol, and (c) fear of somatic sensations. Although all items had
salient loading of ≥ .40, it is important to note that factor two had only three
salient loadings and two items from factor three significantly cross-loaded onto
factor one. Taylor et al. (1996) concluded that the ASI as well as the construct of
anxiety sensitivity is multidimensional and posited that Lilienfeld et al’s. (1993)
argument, in which anxiety sensitivity is viewed as hierarchically arranged, may
be correct.
In order to test the hypothesis that the ASI is hierarchically arranged,
Zinbarg, Barlow, and Brown (1997) conducted both an exploratory and
confirmatory factor analytic investigation. Examining responses of 432
Anxiety Sensitivity 31
individuals from a clinical setting who met the DSM-III-R (APA, 1987) criteria
for an anxiety or depressive disorder, Zinbarg et al.’s (1997) first study employed
a first-order principal factor analysis and oblimin rotation. Three factors that
accounted for 44% of the total variance were extracted. Rather than employing
the Kaiser’s criterion to determine the number of factors to extract, Zinbarg et al.
(1997) decided to extract the maximum number of factors that would allow
replicability to other multidimensional solutions of the ASI already published in
the available literature. Using coefficients of convergence, Zinbarg et al. (1997)
reported that the three-factor solution converged well with the results of other
investigations, but became unstable when a fourth factor was extracted. The
three factors retained were labelled (a) AS-Physical Concerns, (b) AS-Mental
Incapacitation Concerns, and (c) AS-Social Concerns. Factor one was correlated
r = .44 with factor two and r = .36 with factor three, while the second factor was
correlated r = .33 with the third factor. In order to test the presence of a higher-
order general factor, Zinbarg et al. (1997) factor analysed the first-order
correlation matrix using principal factor extraction combined with Kaiser’s
criterion and scree test and reported a single general order factor. The loadings
of the three first-order factors on the high-order factor were .69, .64, and .52
respectively. Zinbarg et al. (1997) argued that the higher-order factor was indeed
a general factor evidenced by 15 of the 16 ASI items having a loading of .30 or
greater on it.
For their second investigation, Zinbarg et al. (1997) employed CFA in
order to test the hierarchical model of anxiety sensitivity identified in their
exploratory work as well as a unifactorial model and an alternative model with
three factors constrained to be orthogonal. Examining responses from the same
Anxiety Sensitivity 32
432 individuals used in their previous exploratory investigation, Zinbarg et al.
(1997) found that the three-factor hierarchical model provided a better fit of the
data than either the model with three factors constrained to be orthogonal or the
unifactorial model. It is important to note however that examination of the
goodness-of fit parameters for retaining a specified model were not as stringent
as they are today and it is entirely possible that all three models would have been
rejected using current standards. Further, Zinbarg et al. (1997) did not use a
separate sample on which to confirm their exploratory work. Thus, it is unclear
whether the hierarchical model would be invariant in another clinical sample if
the factor structure had not already been explored and fitted to the data.
Zinbarg et al. (1997) concluded their investigation by arguing that their
model is dissimilar to the description offered by Taylor et al. (1992) because the
latter did not perform the appropriate statistical tests of comparing the fit
between oblique multidimensional, unifactorial and orthogonal multidimensional
models. Similarly, they argued that as the estimation of the general factor
saturation exceeded .50, the total score of the ASI provides an interpretable
measure of anxiety sensitivity because it is consistent with the hypothesis of a
single construct that accounts for the majority of the variance in ASI total scores.
Therefore, Zinbarg et al. (1997) interpreted their results as countering the
hypothesis that the ASI contains multiple orthogonal factors and instead is best
viewed as hierarchical containing three first-order factors that load onto a single
second-order factor. The researchers suggested that future investigations should
focus on revising the ASI with the aim of producing higher general factor
saturation and increasing the reliability of the AS-Social Concerns subscale
through the inclusion of additional items.
Anxiety Sensitivity 33
Recently two other exploratory investigations have emerged that have
utilised different methodologies and rationales than the approach taken by
Zinbarg et al. (1997). The first of these was conducted by Blais et al. (2001).
In order to evaluate the ASI through the process of item analysis, Blais et
al. (2001) examined responses from 340 outpatients who met the DSM-III-R
(APA, 1987) criteria for panic disorder, social phobia, or major depression as
well as 50 university students. Initial item-to-scale correlations revealed that
items 1, 5, and 7 had adjustable correlations below the acceptable value of .30.
Follow-up between groups’ tests revealed that items 1, 5, 7, and 8 consistently
failed to discriminate between the groups in a theoretically expected manner,
while item 13 mainly reflected social phobia. Use of PCA with Kaiser’s
criterion, scree test, and root curve criterion for all of the items of the ASI
revealed either a three or four-factor solution. Although the two-factor structures
were similar, the four-factor solution was retained as it produced more primary
loadings and also had fewer hyperplaning items than did the three-factor
solution. Factor one contained 7 items and accounted for 37% of the variance.
Items on this factor were consistent with the assessment of fear of arousing
sensations. Factor two contained only 2 items and accounted for 9% of the
variance. Items on this factor appeared to measure fear related to loss of control.
Both items on factor two were shown to have adjusted item-to-scale correlations
below the cut-off level of .30. Factor three contained 3 items and accounted for
8% of the variance. This factor was considered stable as all three items loaded
exclusively on it with a loading of .70 or greater and was consistent with the
assessment of fear of loss of mental control. Finally, factor four contained only 2
items and accounted for 7% of the variance. Items on this factor appeared to
Anxiety Sensitivity 34
measure concern of gastrointestinal symptoms. Item 13 failed to load
significantly on any of the four factors and item 11 was significantly cross-
loaded on both factors four and one. Orthogonal or oblique rotations contained
essentially identical factor loadings.
Blais et al. (2001) has argued that as factors two and four only contained
two items each, it was possible that these factors would be unstable and would
have a reduced probability of replication in future investigations. As such, a
follow-up study using data from three previously published investigations (Ball,
al., 1997) was performed using the original 16-item ASI and the modified ASI
with five items eliminated (1, 5, 7, 8, and 13). Blais et al. (2001) referred to this
scale as the 11-item ASI. Results from all three investigations revealed that the
11-item ASI performed as well as the original 16-item ASI. Thus, Blais et al.
(2001) results revealed that through the removal of five problematic items (31%
of the scale); improvements could be made to the original ASI without loss of
construct validity.
Whilst the results from Blais et al. (2001) investigation showed some
promise regarding the 11-item ASI, it is important to note that it is not without
limitations. Firstly, in their examination of the ASI, Blais et al. (2001) retained
11 of the 16 original items that corresponded to one dimension relating to a fear
of somatic sensations of anxiety and another representing fear of loss of mental
control. Whilst these two dimensions are similar to those observed in other
investigations of the original ASI it is important to note that Blais et al’s. (2001)
investigation fails to provide clearer understanding of anxiety sensitivity than has
previously been reported in the literature. The two dimensions retained by Blais
Anxiety Sensitivity 35
et al. (2001) do not add significantly to the existing literature as they are too
general to highlight the specific anxiety sensitivity fears. Similarly, as the
dimension ‘fear of loss of mental control’ contains only three items, this
dimension could benefit from further item expansion, whereas the absence of a
dimension capable of accounting for fears related to publicly observable anxiety
symptoms excludes potentially important information from investigation using
the current scale.
Another limitation in their investigation concerns the proposition that the
methodology was to some extent flawed. In their analyses, Blais et al. (2001)
applied factor analysis to a participant pool comprised of clinical and nonclinical
participants. Whilst they assumed that it was logical to combine these two
groups into one heterogeneous sample to improve the generalisability of results,
the investigation could have been improved by retaining the clinical and
nonclinical participants as two separate testing groups. As their investigation
involved modifications to the original ASI item pool, a more rigorous test would
have been to make item modifications based on one a priori group of participants
and repeat the analysis in the second group of participants in order to test the
configural and metric invariance of their modified ASI model. As studies of the
original ASI suggest that the factor structure is convergent across clinical and
nonclinical populations (see Zinbarg, Mohlman, & Hong, 1999), this would have
been a more stringent test of the 11-item ASI model.
In a similar investigation to the above, Schmidt and Joiner (2002)
evaluated the ASI in a community sample of 233 individuals with no history of
psychiatric illness or spontaneous panic. Initial examination of the corrected
item-total correlations revealed that items 1, 5, and 7 were found to have the
Anxiety Sensitivity 36
fewest correlations greater than .20. Results of PCA and PAF using both
orthogonal and oblique rotations suggested retaining two factors after it was
determined that a four-factor solution would not be stable. As such, Schmidt and
Joiner (2002) argued that items 1, 5, and 7 be removed from the measure.
Additional analyses revealed that through the removal of these three problematic
items, the internal consistency of the scale improved slightly. PCA and PAF of
the revised 13-item ASI revealed that, in general, analyses using orthogonal
rotations were consistent with a three-factor solution consisting of dimensions
relating to ‘fears of mental catastrophe’, ‘cardiopulmonary fears’, and ‘vasovagal
fears’. Conversely, analyses using oblique rotations were consistent with a two-
factor solution as the items from the ‘vasovagal fears’ factor loaded onto the
‘fears of mental catastrophe’ and ‘cardiopulmonary fears’ factors.
Schmidt and Joiner (2002) retained both models and examined them
again through the process of confirmatory factor analysis. For comparison
purposes, the orthogonal and oblique multifactorial models of Peterson and
Heilbronner (1987), Telch et al. (1989), Blais et al. (2001) and four separate
unifactorial solutions were also examined. Using the same population on which
the exploratory analyses were conducted, Schmidt and Joiner (2002) found that a
two-factor oblique model produced the best overall fit of all models tested, by
meeting the various goodness-of-fit cut-off criteria using only 10 items from the
ASI. The first factor, labelled ‘fear of mental catastrophe’ contained 6 items (2,
11, 12, 13, 15, and 16); while the second factor, labelled ‘fear of
cardiopulmonary symptoms’ contained 4 items (6, 8, 9, and 10). While the fit
indices for this model were marginal at best, Schmidt and Joiner (2002) retested
Anxiety Sensitivity 37
this 10-item, two-factor model on another population of 809 undergraduate
university students and again found a marginal fit of the student data.
This investigation is an improvement on the Blais et al. (2001) study as
the methodology was enhanced by employing the hypothesis testing approach of
confirmatory factor analysis and retesting using a separate sample of participants.
The investigation served to highlight the inadequacy of the previously identified
ASI models in the available literature and revealed that the shortfalls of the ASI
could be improved by expanding the item pool in order to construct more reliable
subscales.
In summary, it is important to note that Reiss et al. (1986) originally
developed the ASI to measure the single anxiety sensitivity component of their
interactional model of Expectancy Theory. Thus, the ASI was originally
developed to measure a unidimensional construct. According to Reiss et al.
(1986), anxiety sensitivity was defined as a cognitive, individual difference
variable delineated by an individual’s fear of anxiety related sensations and based
on the belief that such sensations result in harmful consequences. Thus the ASI
was developed in line with this conceptual definition. However, results of
exploratory and confirmatory factor analytic investigations of the ASI reviewed
at this juncture have revealed that most investigators have found that anxiety
sensitivity is in fact comprised of multiple dimensions. There has been some
argument over the exact number and nature of dimensions with some
investigators finding support for as few as two dimensions (Blais et al., 2001;
Schmidt & Joiner, 2002), and others finding support for three (Stewart et al.,
1997; Taylor, 1996; Zinbarg et al., 1997), or four dimensions (Peterson &
Heilbronner, 1987; Telch et al., 1989). Further, while different methods of
Anxiety Sensitivity 38
analysis have been employed, some inappropriately and some without a sound
theoretical rationale Zinbarg et al. (1997) contended that anxiety sensitivity, as
measured by the 16-item ASI, is hierarchically structured, consisting of multiple
first-order dimensions that relate to a fear of social, somatic, or psychological
consequences of anxiety that load onto a single second-order dimension.
However, while some level of agreement is emerging regarding this
interpretation, it is important to note that a consensus has not yet been reached on
this key issue.
While the ASI is important in furthering our understanding of anxiety
sensitivity, the repeated attempts to clarify the dimensionality of anxiety
sensitivity using the 16-item ASI is problematic, as it was never designed to
measure a multidimensional construct. As such, the original ASI contains a
number of flaws that require attention. As an unresolved question concerns the
number and nature of the first-order anxiety sensitivity dimensions, it is
important to note that several investigations have repeatedly revealed that the
ASI contains too few items to reliably identify the major anxiety sensitivity
facets. For example results from several investigations (see Deacon &
Valentiner, 2001; McWilliams & Cox, 2001; Stewart et al., 1997) have revealed
that the ASI contains only two items capable of measuring the domain of ‘fear of
publicly observable anxiety and anxiety-related symptoms’. A further limitation
with the ASI is that it also contains too few items to determine whether the ‘fear
of physical sensations’ dimension consists of a number of dimensions such as
fear of respiratory symptoms, fear of cardiovascular symptoms and fear of
gastrointestinal symptoms. Thus, precisely which physical sensations are feared
is unclear. Examination of the scale reveals that the ASI only contains two items
Anxiety Sensitivity 39
capable of assessing cardiovascular concerns, two items capable of assessing
gastrointestinal concerns and one item capable of assessing respiratory concerns.
Tabachnick and Fidell (2001) have argued that interpretation of factors defined
by only one or two variables is problematic for psychological test measurement
research as it results in poorly defined constructs. As such, it is clear that the
ASI contains too few items for these specific dimensions to be assessed
adequately as the emergence of a factor with only one or two items is
conceptually meaningless. Further, caution should be used in the interpretation
of scores from the various dimensions that have been derived from use with the
original ASI, because of the instability that may result from dimensions that do
not contain an adequate number of items.
A further problem with the ASI is that it contains a number of ambiguous
statements (e.g., ‘unusual body sensations scare me’; ‘it is important for me to
stay in control of my emotions’; ‘it embarrasses me when my stomach growls’;
or ‘it scares me when I am nervous’). Items such as these are problematic
because they do not act as reliable markers for the specific anxiety sensitivity
dimensions as individuals could obtain high scores for a number of different
reasons.
Given that investigations of the first-order dimensions of the ASI have
validated the importance of the multidimensional perspective of anxiety
sensitivity (see Zinbarg et al., 1999 for review), Taylor and Cox (1998)
developed the 36-item Anxiety Sensitivity Index – Revised (ASI-R), in order to
provide a more comprehensive measure of the first-order anxiety sensitivity
dimensions. While the 36-item ASI-R retains the same instructions and response
format as the 16-item ASI, it contains a broader selection of items, and hence
Anxiety Sensitivity 40
dimensions, for assessing the explicit domains of anxiety sensitivity. In addition,
the ASI-R contains 10 items which have been adopted from the original ASI.
These items were included because they were considered reliable markers for
specific anxiety sensitivity dimensions (The ASI-R is presented in Appendix B).
A table detailing the three published investigations of the ASI-R is provided (see
Table 2.2).
2.5 The Anxiety Sensitivity Index – Revised
In their first study designed to examine the internal structure of the ASI-
R, Taylor and Cox (1998) examined responses from 155 individuals attending
outpatient clinics who also met the DSM-IV (APA, 1994) criteria for an anxiety
disorder, trichotillomania, major depression, Tourette’s disorder, kleptomania,
somatoform pain disorder, adjustment disorder, and psychological factors
affecting a medical condition. The ASI-R is a self-report measure containing 36
items, 10 of which are from the 16-item ASI, that relate to a fear of anxiety
symptoms (e.g., ‘smothering sensations scare me’; it is important for me not to
appear nervous in public’; ‘when I feel a pain in my chest, I worry that I’m going
to have a heart attack’; ‘When my thoughts seem to speed up, I worry that I
might be going crazy’). It contains the same instructions and format of the ASI,
with participants rating each item on a five-point scale that ranges from (0) ‘very
little’ to (4) ‘very much’.
Drawing on the domains derived from previous factor analytic
investigations of the ASI, Taylor and Cox (1998) constructed the ASI-R to
measure six a priori dimensions of anxiety sensitivity that related to a fear of
cardiovascular, respiratory, gastrointestinal, publicly observable, dissociative and
neurological anxiety symptoms. Using the rule of parallel analysis in order to
Anxiety Sensitivity 41
Table 2.2. Factor analytic studies of the Anxiety Sensitivity Index-Revised Author (s)
Sample Method of Factor Extraction
Method of Factor Rotation
Number of Factor (s)
Taylor and Cox (1998)
155 outpatients PCA and PAF Oblique 4
Zvolensky et al. (2003) Nonclinical participants from six countries, including Canada (N = 478), France (N = 701), Mexico (N = 418), Netherlands (N = 536), Spain (N = 480), and the United States (N = 173)
PFA Oblique 2
Deacon, Abramowitz, Woods, and Tolin (2003)
Two independent samples of university students ( N = 558 and N = 444)
PCA and PAF Oblique 4
Note. PCA = Principal components analysis; PAF = Principal axis factor analysis.
Anxiety Sensitivity 42
determine the number of factors to retain at both the mean and 95th percentile
eigenvalue, Taylor and Cox (1998) used both PCA and PAF with oblique
rotation and reported a four-factor solution that was just below the cut-off values
of the mean and 95th percentiles. The four-factor solution accounted for 60% of
the total variance for PCA and 55.1% of variance for PAF. The four factors
retained were (a) fear of respiratory symptoms, (b) fear of publicly observable
anxiety reactions, (c) fear of cardiovascular symptoms, and (d) fear of cognitive
dyscontrol. The pattern of salient loadings was very similar across both PCA and
PAF with only a small number of hyperplane items.
The matrix of correlations for the four first-order factors computed from
the PCA solution was analysed again via PCA in order to identify the presence of
second-order factors. Similarly, the matrix of correlations for the four first-order
factors computed from the PAF solution was also analysed again via PAF in
order to identify any second-order factors. For each analysis, only one
eigenvalue was greater than 1, which led to a single second-order factor being
extracted. This second-order or general factor accounted for 52% of the variance
in the PCA solution and 44.9% of the variance in the PAF solution. As such, the
results indicated the presence of a hierarchical solution in which four first-order
dimensions loaded on a single second-order or general anxiety sensitivity
dimension. In order to account for the individual variance of the first and
second-order factors as well as to compute the loadings of each item on the
second-order factor and residualised loadings of each item on the first-order
factors, a Schmidt and Leiman transformation and application of Jensen and
Weng’s formulae were conducted. This process revealed that the pattern of
loadings of the first-order factors were similar to those obtained by PCA and
Anxiety Sensitivity 43
PAF with oblique rotation and that the second-order factor accounted for
substantially more variance (40.8% for PCA) than the four first-order factors
(6.7%, 5.7%, 6.6%, and 4.6% for PCA respectively). As a result, Taylor and
Cox (1998) argued that these findings suggest “anxiety sensitivity arises largely
from a general anxiety sensitivity factor, with more modest contributions from
the four specific factors” (p. 475).
In their second study, correlations between the ASI-R first and second-
order factors and the original ASI, Beck Anxiety Inventory (BAI), Beck
Depression Inventory (BDI) and medication status were investigated (Taylor &
Cox, 1998). There was evidence of large correlations (rs ≥ .60) with the ASI,
which was predictable given that the ASI-R contains 10 of the 16-items from the
ASI. The scale’s first and second-order dimensions were also moderate to highly
correlated with the BAI and moderately correlated with the BDI. The ASI-R
factors were uncorrelated with medication status. Thus, there was sufficient
support for the concurrent validity of the ASI-R.
Taylor and Cox’s (1998) third investigation compared the factor scores of
the ASI-R across four groups of participants that included individuals with a
diagnosis of panic disorder and assessed before treatment; individuals with a
diagnosis of panic disorder and assessed after treatment; a group of individuals
classified as ‘other anxiety disorders’ and assessed before treatment; and the
remaining individuals who were classified as ‘non anxiety disorders’ and were
assessed before treatment. Using one way analysis of variance for each factor
score with Newman-Keuls post hoc comparisons for significant differences
between groups, Taylor and Cox (1998) reported that the panic disorder group
who were assessed at pre-treatment scored significantly higher on all of the first
Anxiety Sensitivity 44
and second-order dimensions when compared to the other diagnostic groups.
Conversely, a significant difference was not observed between the ‘other anxiety
disorders’ group assessed at pre-treatment when compared to the ‘nonanxiety
disorder’ group who were also assessed at pre-treatment for any of the first or
second-order ASI-R dimensions. Conversely, it is interesting to note that both
the ‘other anxiety disorders’ and ‘nonanxiety disorder’ groups assessed at pre-
treatment scored significantly higher on the ‘fear of respiratory symptoms’
dimension as well as on the second-order or total score dimension when
compared to the panic disorder group assessed at post-treatment.
Since Taylor and Cox (1998), only two investigations have appeared in
the extant literature that attempt to determine the factor structure and
psychometric properties of the expanded ASI-R. The first investigation,
conducted by Zvolensky et al. (2003), sought to determine the factor structure
and internal consistency of the ASI-R using a large, diverse sample of
nonclinical participants from six countries, including Canada, France, Mexico,
the Netherlands, Spain and the United States. PFA with oblique rotation was
applied to each of the six samples as well as a within-group correlation matrix.
Parallel analysis found that a two-factor solution was the most replicable across
all seven datasets. Zvolensky et al. (2003) reported that the two dimensions
retained related to (1) fear of somatic sensations and (2) fear of social-cognitive
concerns. Subsequent investigations revealed that the two dimensions contained
an acceptable level of internal consistency; the two dimensions were moderate to
strongly correlated in each dataset, denoting the presence of a hierarchical
structure; and factor comparability coefficients indicated that the different
versions of both dimensions were, in most cases, consistent with one another
Anxiety Sensitivity 45
across datasets. Zvolensky et al. (2003) concluded their investigation by
proposing that although their investigation offered important insight into the
nature of some aspects of anxiety sensitivity, it was an early first attempt to
examine the construct using the ASI-R in a large, diverse sample and as such,
should be regarded in this context.
Deacon, Abramowitz, Woods, and Tolin (2003) set out to replicate
Taylor and Cox’s (1998) results and provide an adequate evaluation of the ASI-
R’s psychometric properties. Using both PCA and PAF with oblique rotation on
a large sample of undergraduate university students, these researchers reported
that parallel analysis using the mean and 95th percentile eigenvalues revealed
four dimensions relating to (1) beliefs about the harmful consequences of
somatic sensations, (2) fear of publicly observable anxiety reactions, (3) fear of
cognitive dyscontrol, and (4) fear of somatic sensations without explicit
consequences. The presence of a hierarchical solution was examined using the
same methodology as Taylor and Cox (1998) and again, a single second-order
dimension was extracted indicating that the ASI-R is hierarchically arranged.
Coefficients of congruence indicated that while the dimensions relating to fear of
publicly observable anxiety reactions and cognitive dyscontrol were highly
replicable to the same dimensions obtained by Taylor and Cox (1998); the two
somatic dimensions showed less convergence. In addition, correlations between
the ASI-R dimensions and other related measures were theoretically consistent
with the anxiety sensitivity construct. However, due to differences between the
somatic factors obtained in their study when compared to the results obtained by
Taylor and Cox (1998); Deacon et al. (2003) conducted another investigation
Anxiety Sensitivity 46
using a second, independent sample of undergraduate university students in order
to increase confidence in the reliability of their factor analytic findings.
Employing the same methodology to identify the number of factors to
retain, Deacon et al. (2003) again found that the ASI-R contained a hierarchical
solution of four first-order dimensions that loaded onto a single, second-order
dimension. Essentially, this solution was identical to the solution identified in
their previous investigation with the dimensions assessing cognitive dyscontrol
and publicly observable anxiety symptoms closely replicating those obtained by
Taylor and Cox (1998). Conversely, the two dimensions relating to somatic
sensations consisted of a blend of the two somatic dimensions reported by Taylor
and Cox (1998) and therefore did not replicate their findings. Deacon et al.
(2003) concluded their investigation by arguing that future studies of the ASI-R
using diverse populations were warranted.
Results of the above investigations of the ASI-R reveal that the revised
measure has the potential to open up new and important avenues for further
investigation. Although a great deal of research has been conducted into the
anxiety sensitivity construct, investigators have repeatedly noted the inherent
problems of measuring the construct when exclusively using the 16-item ASI.
Thus, in an effort to present a more comprehensive measure of the anxiety
sensitivity construct, Taylor and Cox (1998) developed the 36-item ASI-R, with
initial research showing promise.
It is important to reiterate that most of the 16-item ASI models evaluated
in the literature have revealed poor fit of the data and, at best, the best-fitting
models have provided only a marginal fit of the data. The 16-item ASI was
constructed to measure what was originally conceptualised as a unitary construct
Anxiety Sensitivity 47
(Reiss et al., 1986) and because the 16-item ASI contains a relatively small
number of items, most of which measure fears of somatic sensations (e.g.,
Stewart et al., 1997), the scale is too abbreviated to adequately measure the major
anxiety sensitivity facets. It has been ascertained that the 16-item ASI contains
too few items to determine whether the ‘fear of somatic sensations’ dimension,
for example, may actually consist of several factors, such as fears of cardiac
symptoms and fears of gastrointestinal symptoms. Further, past research has
repeatedly revealed that the 16-item ASI dimension related to ‘Social Concerns’
tends to be plagued by low levels of internal consistency when compared to the
other dimensions and usually explains the least amount of variance in the overall
solution (Peterson & Plehn, 1999; Stewart et al., 1997, Taylor et al., 1996).
Finally, evidence of a number of ambiguous statements contained in the 16-item
ASI is problematic because such items do not act as reliable markers for the
specific anxiety sensitivity dimensions. Overall, the numerous limitations of the
16-item ASI are of concern as they reduce an investigator’s confidence when
making any theoretical claims about the anxiety sensitivity construct.
In light of the limitations of the 16-item ASI, the 36-item ASI-R
developed by Taylor and Cox (1998) is a promising instrument for measuring
anxiety sensitivity. The research presented at this point, using the ASI-R, has
revealed that the lower-order dimensions demonstrate theoretically consistent
relationships with criterion variables and that the second-order dimension
correlate very well with the 16-item ASI, indicating that both instruments
measure the same construct. Further, given that the content validity of the ASI-R
is an improvement on the 16-item ASI, it is contended that it is a better measure
for use in studies investigating anxiety sensitivity dimensions.
Anxiety Sensitivity 48
Given the important role that anxiety sensitivity plays in anxiety
pathology, the following sections aim to discuss the research and findings related
to how the construct is manifested between and within various anxiety and mood
pathologies as well as normative groups. As a psychological construct, anxiety
sensitivity has become associated with understanding panic pathology in
particular and understanding negative emotional functioning found in other
anxiety and mood psychopathology in general. Further, as the degree to which
treatment, and in particular cognitive-behavioural therapy (CBT), successfully
impacts upon anxiety sensitivity will also be examined and discussed. It is
important to note that in all instances the research presented from this point
forward has been conducted using the 16-item ASI only.
2.6 Predictive Validity of Anxiety Sensitivity
From the time when Reiss and McNally (1985) first proposed the
expectancy model of fear, extensive research has culminated to demonstrate a
relationship between anxiety sensitivity and various anxiety-related phenomena.
Some of the most persuasive evidence of the predictive validity of anxiety
sensitivity and the ASI has come from prospective research. For example, in a
prospective study designed to predict the frequency and intensity of panic attacks
after a 3-year period, Maller and Reiss (1992) administered the ASI to 151
university students in 1984 and retested these same students in 1987 using the
form the first dimension ‘fear of somatic sensations’ and 17 items (items 29, 13,
9, 17, 14, 16, 19, 18, 15, 11, 12, 31, 34, 35, 33, 36, 32) to form the second
dimension ‘fear of social-cognitive concerns’. For each of the hypothesised
models under investigation the variance of the first item for each of the first-
order dimensions was set to 1 and the remaining items were allowed to vary
freely in order to set the scale. The variance from the second-order dimension to
the first-order dimensions was also set to 1.
Overall, confirmatory factor analysis of the three hypothesised
hierarchical models identified by Deacon et al. (2003); Taylor and Cox (1998);
and Zvolensky et al. (2003) failed to provide an adequate fit of the data in either
the clinical group (see Table 4.1) or nonclinical group (see Table 4.2). It can be
Anxiety Sensitivity 113
seen that all of the goodness-of-fit indices are well out of range of the
recommended cutoff criteria for retaining a hypothesised model.
Table 4.1
Clinical group CFA of three hypothesised ASI-R models
Note. *p < .001
Hypothesised Model
χ² SBχ² df Model AIC
NNFI CFI RMSEA
Deacon et al. (2003)four-factor hierarchical model
3898.50*
1511.95*
590
2718.50
.546
.575
.152
Taylor and Cox (1998) four-factor hierarchical model
2170.15*
1722.68*
590
990.15
.781
.795
.105
Zvolensky et al. (2003) two-factor hierarchical model
3226.83*
2562.26*
592
2042.83
.639
.661
.136
Anxiety Sensitivity 114
Table 4.2
Nonclinical group CFA of three hypothesised ASI-R models
Note. *p < .001
Note. *p < .001
Note. *p < .001
4.3.2 Model modifications of Taylor and Cox’s (1998) Model
As the four-factor hierarchical model identified by Taylor and Cox (1998)
resulted in the best available goodness-of-fit indices for all three hypothesised
models under investigation, further analyses were performed using this model as
a basis on which to determine structural modifications. Additionally it was
determined that the clinical group data was the most representative group on
which to base model modifications. Therefore, further analyses of the clinical
group data, using EQS, were performed in an attempt to develop a more robust
and possibly more parsimonious model.
Initially, a unidimensional model was identified in order to test the fit of
the clinical dataset. However, this model failed to converge. As such,
Hypothesised Model
χ² SBχ² df Model AIC
NNFI CFI RMSEA
Deacon et al. (2003) four- factor hierarchical model
4604.23*
1033.73*
590
3427.23
.532
.562
.125
Taylor and Cox (1998) four-factor hierarchical model
2420.99*
1773.57*
590
1240.99
.801
.788
.085
Zvolensky et al. (2003) two-factor hierarchical model
4110.22*
2897.90*
592
2926.22
.592
.616
.117
Anxiety Sensitivity 115
modifications were made on the basis of the four-factor model identified by
Taylor and Cox (1998). Firstly, the covariance between items within each of the
factors was examined by inspecting items with large standardised residuals.
Bentler (1995) has argued “large values of standardised residuals point to the
variables that are not being well explained by the model” (p. 91). For example, it
was found that item 15 ‘I think it would be horrible for me to faint in public’
exhibited a large unmodelled covariance with item 14 ‘I believe it would be
horrible to vomit in public’ and item 5 ‘It scares me when I feel faint’. However,
there was not a large unmodelled covariance between item 14 and 5. This
revealed that item 15 was indicating a possible relationship between itself and
these two items when such a relationship was not hypothesised to exist. As such
the decision was made to remove item 15 from the model. Other items that were
removed from the model because they were resulting in large unmodelled
covariances were items 11, 12, 28 and 29.
The remaining 31 items were further evaluated on the basis of the
Lagrange Multiplier (LM) test. Parameters that were identified as improving
model fit at the multivariate level were examined to determine whether they were
significantly cross-loading onto more than one or more factors. Although some
cross-loading may be important for measuring overall anxiety sensitivity,
significant cross-loadings were removed in order to improve model fit and
develop a more parsimonious measure with a clean factor structure (see
Anderson & Gerbing, 1998). Further, unless a theory suggests items may be
cross-loaded, the presence of cross-loading items could be attributed to a
statistical artifact (Anderson & Gerbing, 1998). Therefore, in order to generate
factors that measured distinct aspects of anxiety sensitivity, any item that resulted
Anxiety Sensitivity 116
in a z score of >1.96 on more than one factor was removed from the solution.
This process resulted in items 5, 6, 7, 8, 9, 10, 20, 25, 26, and 30 being removed,
even though they also loaded onto the original factors. As such a more
parsimonious, 21-item four-factor hierarchical model was identified as the best
possible fit of the clinical data.
4.3.3 Test of Configural Invariance of Modified Model
In order to validate the solution obtained from the clinical dataset and
determine whether the same items from the clinical sample loaded onto the same
dimensions as the nonclinical sample, a separate CFA was performed using the
nonclinical sample data in order to test the configural invariance of the model.
Results revealed that the 21-item four-factor hierarchical model was an
acceptable fit of the nonclinical population data (see Table 4.3).
Table 4.3
Nonclinical group CFA of the 21-item ASI Hierarchical model
χ² SBχ² df Model
AIC
NNFI CFI RMSEA
21-item ASI
Hierarchical
Model
537.32*
403.04*
185
167.32
.922
.932
.066
Note. *p < .001
4.3.4 Test of Metric Invariance of Modified Model
Similarly, a multiple group CFA was performed in order to test the metric
invariance of the model between both the clinical and nonclinical group datasets.
Anxiety Sensitivity 117
The results revealed that the 21-item four-factor hierarchical model was again an
acceptable fit of both the clinical and nonclinical group datasets (see Table 4.4).
Table 4.4
Metric Invariance CFA statistics between the clinical and nonclinical group
datasets using the 21-item ASI
χ² df Model
AIC
NNFI CFI RMSEA
21-item ASI
Hierarchical
Model
1163.25*
391
381.25
.913
.919
.05
Note. *p < .001
Examination of each constraint indicated significantly different factor
loadings existed between the clinical and nonclinical groups on items 2, 16, 23,
24, 32, 33, 34, and 36 as well between the first-order dimensions 1, 3, and 4 and
the second-order or general anxiety sensitivity factor. Specifically items 2, 16,
24, 33, 34, and 36 were stronger indicators for the clinical group when compared
to the nonclinical groups’ factor loadings. Conversely, items 23 and 32 were
stronger indicators for the nonclinical groups’ when compared to the clinical
group. In addition, the loadings on dimensions 1, 3, and 4 were also stronger
indicators of the general anxiety sensitivity dimension for the nonclinical group
when compared to the clinical groups’ loadings. The loadings for the clinical
group are reported in Figure 4.1 and the loadings for the nonclinical group are
reported in Figure 4.2. It can be seen that the 21 items from the ASI-R resulted
in four first-order dimensions comprising of (1) Fear of respiratory symptoms
Anxiety Sensitivity 118
(four items); (2) Fear of publicly observable symptoms (six items); (3) Fear of
cardiovascular/stroke symptoms (five items); and (4) Fear of cognitive
dyscontrol (six items) that loaded onto a single, second-order dimension for both
the clinical and nonclinical samples.
Anxiety Sensitivity 119
Figure 4.1. Clinical group factor loadings for the items of the 21-item ASI with a hierarchical model of four first-order dimensions and a general higher-order dimension.
Publicly Observable
V 4
V 2
V 3 Respiratory
V 1
V 13
V 14
V 16
V 17
V 18
V 19
.60
.59
Cardiovascular/Stroke
V 27
V 24
V 23
V 22
V 21
Cognitive Dyscontrol
V 36
V 35
V 34
V 33
V 32
V 31
.55
.52
.82
.89
.68
.90
.71
.56
.60
.74
.87
.85
.77
.76
.91
.90
.75
.74
.82
.88
.91
.84
.83
Anxiety Sensitivity
Anxiety Sensitivity 120
Figure 4.2. Nonclinical group factor loadings for the items of the 21-item ASI with a hierarchical model of four first-order dimensions and a general higher-order dimension.
Publicly Observable
V 4
V 2
V 3 Respiratory
V 1
V 13
V 14
V 16
V 17
V 18
V 19
.62
.62
Cardiovascular/Stroke
V 27
V 24
V 23
V 22
V 21
Cognitive Dyscontrol
V 36
V 35
V 34
V 33
V 32
V 31
.79
.79
.74
.88
.67
.89
.67
.50
.62
.76
.83
.85
.83
.74
.83
.87
.79
.80
.87
.84
.86
.73
.81
Anxiety Sensitivity
Anxiety Sensitivity 121
For the clinical group, all items loaded very well (86% were above 0.70)
on their respective factors. The four first-order dimensions also loaded well (>
0.50) on the second-order anxiety sensitivity dimension. Similarly, for the
nonclinical group all items loaded very well (81% were above 0.70) on their
respective dimensions and the four first-order dimensions also loaded well (>
0.60) on the second-order anxiety sensitivity dimension. However, it is
important to note that while the number of first-order anxiety sensitivity
dimensions is identical to those identified by Taylor and Cox (1998) the number
of items unique to first-order dimension has changed because redundant or
problematic items were removed from the scale. There was also a change of one
of the dimension names from ‘Fear of Cardiac Sensations’ to ‘Fear of
Cardiovascular/Stroke Sensations’ to be more representative of the items
underlying that dimension. As a result of these changes the decision was made
to call this modified scale the 21-item ASI.
4.3.5 Comparisons of alternative models
In order to determine whether any alternative models could provide a
better fit of the data, the obtained 21-item four-factor hierarchical model was
compared with additional models. Firstly, a unifactorial model that contained all
of the items from the 21-item ASI was selected with the variance of the single
factor set to 1 in order to allow the individual items to vary freely. The second
alternative model tested was one that constrained the four first-order dimensions
of the 21-item ASI to be orthogonal. In order to set the scale, the path from the
first item for each of the first-order dimensions was set to 1 and the remaining
items were allowed to vary freely. Both the unifactorial and orthogonal models
failed to provide an adequate fit of the data in either the clinical or nonclinical
Anxiety Sensitivity 122
samples. In addition, a three-factor hierarchical model was also tested whereby
the ‘Fear of Respiratory Symptoms’ and ‘Fear of Cardiovascular/Stroke
Symptoms’ dimensions were combined to make one ‘Fear of Physiological
Symptoms’ dimension containing nine items (items 1, 2, 3, 4, 21, 22, 23, 24, and
27). The goodness-of fit indices for all three models for the clinical group are
reported in Table 4.5 and Table 4.6 for the nonclinical group. It can be seen that
while the orthogonal model is an improvement upon the unifactorial model, it too
does not meet the stringent criteria for adequate model fit as neither model was
capable of accounting for the increments in variance that is explained by the
hierarchical model. Likewise, the three-factor hierarchical model was not an
adequate fit of either the clinical or nonclinical datasets.
Table 4.5
Clinical groups CFA of the unifactorial, orthogonal 4-factor model and 3-factor
hierarchical models of the 21-item ASI.
Note. *p < .001
χ² SBχ² df Model AIC
NNFI CFI RMSEA
Unifactorial
21-item ASI 2266.93* 1926.17* 189 1888.98 .470 .523 .211
Multifactorial
21-item ASI Orthogonal 4-factor model
719.78*
624.71*
189
341.77
.865
.878
.107
Anxiety Sensitivity 123
Table 4.6
Nonclinical groups CFA of the unifactorial, orthogonal 4-factor model and 3-
factor hierarchical models of the 21-item ASI.
Note. *p < .001
4.3.6 Comparisons of the 21-item ASI and competing ASI models
In order to compare the 21-item ASI to previously validated ASI models
in the available literature, CFA was conducted on the unifactorial 16-item (Reiss
et al., 1986), 14-item (Taylor et al., 1992), 11-item (Blais et al., 2001), and 10-
item (Schmidt & Joiner, 2002) ASI models. Similarly, previous multifactorial
models of the ASI consisting of four-factors (Peterson & Heilbronner, 1987;
Telch et al., 1989), three-factors (Zinbarg et al., 1997) and two-factors (Blais et
al., 2001; Schmidt & Joiner, 2002) were also evaluated as orthogonal, oblique
and hierarchical structures. The goodness-of-fit indices for the clinical group are
reported in Table 4.7 and the goodness-of-fit indices for the nonclinical group are
reported in Table 4.8. It can be seen that for all models, the goodness-of-fit
indices were well below the cut-off criteria recommended for retaining a
χ² SBχ² df Model AIC
NNFI CFI RMSEA
Unifactorial
21-item ASI 2689.97* 1855.95* 189 2311.87 .462 .515 .175
Multifactorial
21-item ASI Orthogonal 4-factor model
826.31*
616.78*
189
448.31
.863
.876
.088
Anxiety Sensitivity 124
hypothesised model for both populations (CFI range .586 to .833; RMSEA range
.95 to .21).
Anxiety Sensitivity 125
Table 4.7
Clinical groups CFA of alternative hypothesised ASI models
Penava, S. J., Otto, M. W., Maki, K. M., & Pollack, M. H. (1998). Rate of
improvement during cognitive-behavioral group treatment for panic
disorder. Behavior Research and Therapy, 36, 665-673.
Peterson, R. A., & Heilbronner, R. L. (1987). The Anxiety Sensitivity Index:
Construct validity and factor analytic structure. Journal of Anxiety
Disorders, 1(2), 117-121.
Peterson, R. A., & Plehn, K. (1999). Measuring Anxiety Sensitivity. In S.
Taylor (Ed.), Anxiety Sensitivity: Theory, Research and Treatment of the
Fear of Anxiety (pp. 61-82). New Jersey: Lawrence Erlbaum
Associates.
Plehn, K., & Peterson, R. A. (2002). Anxiety sensitivity as a predictor of the
development of panic symptoms, panic attacks, and panic disorder: A
prospective study. Journal of Anxiety Disorders, 16, 455-474.
Rapee, R. M., Ancis, J. R., & Barlow, D. H. (1988). Emotional reactions to
physiological sensations: panic disorder patients and non-clinical Ss.
Behavior Research and Therapy, 26 (3), 265-269.
Rapee, R. M., Brown, T. A., Antony, M. M., & Barlow, D. H. (1992). Response
to hyperventilation and inhalation of 5.5% carbon dioxide-enriched air across the DSM-III-R anxiety disorders. Journal of Abnormal Psychology, 101, 538–552.
Anxiety Sensitivity 243
Rapee, R. M., & Medoro, L. (1994). Fear of psychical sensations and trait
anxiety as mediators of the response to hyperventilation in nonclincial
subjects. Journal of Abnormal Psychology, 103 (4), 693-699.
Razan, G. (1961). The observable unconscious and the inferable conscious in
current Soviet psychophysiology: Interoceptive conditioning, semantic
conditioning, and the orienting reflex. Psychological Review, 68, 81-147.
Reiss, S. (1991). Expectancy theory of fear, anxiety, and panic. Clinical
Psychology Review, 11, 141-153.
Reiss, S., & Havercamp, S. (1996). The sensitivity theory of motivation:
Implications for psychopathology. Behavior Research and Therapy, 34,
621-632.
Reiss, S., & McNally, R. J. (1985). The expectancy model of fear. In S.
Reiss., & R. R. Bootzin (Eds.), Theoretical issues in Behaviour Therapy
(pp. 107-121). NY: Academic Press.
Reiss, S., Peterson, R. A., Gursky, D. M., & McNally, R. J. (1986). Anxiety
sensitivity, anxiety frequency and prediction of fearfulness. Behavior
Research and Therapy, 24, 1-8.
Sandin, B., Chorot, P., & McNally, R. J. (1996). Anxiety Sensitivity Index:
Normative data and its differentiation from trait anxiety. Behavior
Research and Therapy, 39, 213-219.
Schmidt, N. B., & Joiner, T. E. (2002). Structure of the Anxiety Sensitivity
Index psychometrics and factor structure in a community sample.
Journal of Anxiety Disorders, 16(1), 33-49.
Anxiety Sensitivity 244
Schmidt, N. B., Lerew, D. R., & Jackson, R. J. (1997). The role of anxiety
sensitivity in the pathogenesis of panic: Prospective evaluation of
spontaneous panic attacks during acute stress. Journal of Abnormal
Psychology, 106(3), 355-364.
Schmidt, N. B., Lerew, D. R., & Joiner, T. E. (1998). Anxiety sensitivity and
the pathogenesis of anxiety and depression: Evidence for symptom
specificity. Behavior Research and Therapy, 36, 165-177.
Schmidt, N. B., & Telch, M. J. (1994). Role of fear of fear and safety
information in moderating the effects of voluntary hyperventilation.
Behavior Therapy, 25, 197-208.
Scott, E. L., Heimberg, R. G., & Jack, M. S. (2000). Anxiety sensitivity in
social phobia: Comparison between social phobics with and without
panic attacks. Depression and Anxiety, 12, 189-192.
Sherer, M., Maddux, J. E., Mercadante, B., Prentice-Dunn, S., Jacobs, B., &
Rogers, R. W. (1982). The self-efficacy scale: Construction and
Validation. Psychological Reports, 51, 663-671.
Shear, M. K., Pilkonis, P. A., Cloitre, M., & Leon, A. (1994). Cognitive
Behavioral Treatment Compared With Nonprescriptive Treatment of
Panic Disorder. Archives of General Psychiatry, 51, 395-401.
Shostak, B. B., & Peterson, R. A. (1990). Effects of anxiety sensitivity on
emotional response to a stress task. Behavior Research and Therapy, 28,
513-521.
Anxiety Sensitivity 245
Spiegel, D. A., Bruce, T. J., Gregg, S. F., & Nuzzarello, A. (1994). Does
Discontinuation in Panic Disorder? American Journal of Psychiatry, 151,
876-881.
Stein, D. J., & Bower, C. (1997). A neuro-evolutionary approach to the anxiety
disorders. Journal of Anxiety Disorders, 11, 4, 409-429.
Stewart, S. H., Samoluk, S. B., & MacDonald, A. B. (1999). Anxiety
sensitivity and substance use and abuse. In S. Taylor (Ed.), Anxiety
Sensitivity: Theory, Research, and Treatment of the Fear of Anxiety (pp.
287-319). New Jersey: Lawrence Erlbaum Associates.
Stewart, S. H., Taylor, S., & Baker, J. M. (1997). Gender differences in
dimensions of anxiety sensitivity. Journal of Anxiety Disorders, 11(2),
179-200.
Tabachnick, B. G., & Fidell, L. S. (2001). Using Multivariate Statistics (4th
ed.). NY: Harper Collins Publishers, Inc.
Taylor, S. (2003). Anxiety Sensitivity and its implications for understanding
and treating PTSD. Journal of Cognitive Psychotherapy, 17 (2), 179-186.
Taylor, S. (1996). Nature and measurement of anxiety sensitivity: Reply to
Lilienfeld, Turner, and Jacob (1996). Journal of Anxiety Disorders, 5,
425-451.
Taylor, S. (1995). Anxiety sensitivity: Theoretical perspectives and recent
findings. Behavior Research and Therapy, 33(3), 243-258.
Taylor, S., & Cox, B. J. (1998). An expanded anxiety sensitivity index:
Evidence for a hierarchic structure in a clinical sample. Journal of
Anxiety Disorders, 12(5), 463-483.
Anxiety Sensitivity 246
Taylor, S., Koch, W. J., & Crockett, D. J. (1991). Anxiety sensitivity, trait
anxiety, and the anxiety disorders. Journal of Anxiety Disorders, 5, 293-
311.
Taylor, S., Koch, W. J., & McNally, R. J. (1992). How does anxiety
sensitivity vary across the anxiety disorders? Journal of Anxiety
Disorders, 6, 249-259.
Taylor, S., Koch, W. J., McNally, R. J., & Crockett, D. J. (1992).
Conceptualizations of anxiety sensitivity. Psychological Assessment, 4,
245-250.
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.
Taylor, S., Rabian, B., & Fedoroff, I. C. (1999). Anxiety Sensitivity: Progress,
Prospects, and Challenges. In S. Taylor (Ed.), Anxiety Sensitivity:
Theory, Research and Treatment of the Fear of Anxiety (pp. 339-353).
New Jersey: Lawrence Erlbaum Associates.
Telch, M. J., Lucas, J. A., Schmidt, N. B., Hanna, H. H., Jaimez, T., & Lucas,
R. A. (1993). Group cognitive-behavioral treatment of panic disorder.
Behavior Research and Therapy, 31, 279-87.
Telch, M. J., Shermis, M. D., & Lucas, J. A. (1989). Anxiety Sensitivity:
Unitary personality trait or domain-specific appraisal? Journal of Anxiety
Disorders, 3, 25-32.
Wardle, J., Ahmad, T., & Hayward, P. (1990). Anxiety sensitivity in
agoraphobia. Journal of Anxiety Disorders, 4, 325-333.
Anxiety Sensitivity 247
Wells, A. (1997). Cognitive therapy of anxiety disorders: A practice manual
andconceptual guide. NY: John Wiley & Sons.
Westra, H. A., Stewart, S. H., & Conrad, B. E. (2002). Naturalistic manner of
benzodiazepine use and cognitive behavioral therapy outcome in panic
disorder with agoraphobia. Anxiety Disorders, 16, 233-246.
Wolman, B. B., & Stricker, G. (Eds.). (1994). Anxiety and related disorders: A
handbook. New York: John Wiley & Sons, Inc.
Zeitlin, S. B., & McNally, R. J. (1993). Alexithymia and anxiety sensitivity in
panic disorder and obsessive- compulsive disorder. American Journal of
Psychiatry, 150, 658-660.
Zinbarg, R. E., Barlow, D. H., & Brown, T. A. (1997). Hierarchical structure
and general factor saturation of the Anxiety Sensitivity Index: Evidence
and implications. Psychological Assessment, 9(3), 277-284.
Zinbarg, R. E., Mohlman, J., & Hong, N. N. (1999). Dimensions of anxiety
sensitivity. In S. Taylor (Ed.), Anxiety Sensitivity: Theory,
Research and Treatment of the Fear of Anxiety (pp. 83-114). New
Jersey: Lawrence Erlbaum Associates.
Zung, W. W. K. (1965). A self-rating depression scale. Archives of General
Psychiatry, 12, 63-70.
Zvolensky, M. J., Arrindell, W. A., Taylor, S., Bouvard, M., Cox, B. J.,
Stewart, S. H., Sandin, B., Cardenas, S. J., & Eifert, G. H. (2003).
Anxiety sensitivity in six countries. Behavior Research and Therapy,
41(7), 841-859.
Anxiety Sensitivity 248
Anxiety Sensitivity 249
Appendix A
Anxiety Sensitivity Index
Anxiety Sensitivity 250
Anxiety Sensitivity 251
Anxiety Sensitivity Index The items listed below reveal fears and worries, which we sometimes experience. Please read each item carefully and using the scale given below, rate it by circling the appropriate answer.
Very A little Some Much Very Little Much
1. It is important for me not to appear nervous 0 1 2 3 4 2. When I cannot keep my mind on a task, I worry that I might be going crazy 0 1 2 3 4 3. It scares me when I feel “shaky” (trembling) 0 1 2 3 4 4. It scares me when I feel faint 0 1 2 3 4 5. It is important to me to stay in control of my emotions 0 1 2 3 4 6. It scares me when my heart beats rapidly 0 1 2 3 4 7. It embarrasses me when my stomach growls 0 1 2 3 4 8. It scares me when I am nauseous 0 1 2 3 4 9. When my heart is beating rapidly, I worry that I might be having a heart attack 0 1 2 3 4 10. It scares me when I become short of breath 0 1 2 3 4 11. When my stomach is upset, I worry that I might be seriously ill 0 1 2 3 4 12. It scares me when I am unable to keep my mind on a task 0 1 2 3 4 13. Other people usually notice when I feel shaky 0 1 2 3 4 14. Unusual body sensations scare me 0 1 2 3 4 15. When I am nervous, I worry that I might be mentally ill 0 1 2 3 4 16. It scares me when I am nervous 0 1 2 3 4
Anxiety Sensitivity 252
Anxiety Sensitivity 253
Appendix B
Anxiety Sensitivity Index – Revised
Anxiety Sensitivity 254
Anxiety Sensitivity 255
Anxiety Sensitivity Index - Revised The items listed below reveal fears and worries, which we sometimes experience. Please read each item carefully and using the scale given below, rate it by circling the appropriate answer.
Very A little Some Much Very Little Much
1. When I feel like I’m not getting enough air, I get scared that I might suffocate 0 1 2 3 4 2. Smothering sensations scare me 0 1 2 3 4 3. It scares me when I become short of breath 0 1 2 3 4 4. When my chest feels tight, I get scared that I won’t be able to breath properly 0 1 2 3 4 5. It scares me when I feel faint 0 1 2 3 4 6. When my throat feels tight, I worry that I could choke to death 0 1 2 3 4 7. It scares me when my heart beats rapidly 0 1 2 3 4 8. When my breathing become irregular, I fear that something bad will happen 0 1 2 3 4 9. It scares me when I feel “shaky” (trembling) 0 1 2 3 4 10. When I have trouble swallowing, I worry that I could choke 0 1 2 3 4 11. It frightens me when my surroundings seem strange or unreal 0 1 2 3 4 12. It scares me when my body feels strange or different in some way 0 1 2 3 4 13. It is important for me not to appear nervous 0 1 2 3 4 14. I believe it would be awful to vomit in public 0 1 2 3 4 15. I think that it would be horrible for me to faint in public 0 1 2 3 4 16. I worry that other people will notice my anxiety 0 1 2 3 4 17. When I tremble in the presence of others I fear what people might think of me 0 1 2 3 4 18. When I begin to sweat in a social situation, I fear that people will think negatively of me 0 1 2 3 4 19. It scares me when I blush in front of people 0 1 2 3 4 20. When I feel a strong pain in my stomach, I worry it could be cancer 0 1 2 3 4 21. When my head is pounding, I worry I could have a stroke 0 1 2 3 4 22. When my heart is beating rapidly, I worry that I might be having a heart attack 0 1 2 3 4 23. When my face feels numb, I worry that I might be having a stroke 0 1 2 3 4 24. When I feel a pain in my chest, I worry that I’m going to have a heart attack 0 1 2 3 4 25. When I feel dizzy, I worry that there is something wrong with my brain 0 1 2 3 4 26. When my stomach is upset, I worry that I might be seriously ill 0 1 2 3 4
Anxiety Sensitivity 256
Very A little Some Much Very Little Much
27. When I notice my heart skipping a beat, I worry that something is seriously wrong with me 0 1 2 3 4 28. When I get diarrhoea, I worry that I might have something wrong with me 0 1 2 3 4 29. It scares me when I am nauseous 0 1 2 3 4 30. It scares me when I feel tingling or prickling sensation in my hands 0 1 2 3 4 31. When I feel “spacey” or spaced out I worry that I may be mentally ill 0 1 2 3 4 32. When my thoughts seem to speed up, I worry that I might be going crazy 0 1 2 3 4 33. When I have trouble thinking clearly, I worry that there is something wrong with me 0 1 2 3 4 34. When I cannot keep my mind on a task, I worry that I might be going crazy 0 1 2 3 4 35. It scares me when I am unable to keep my mind on a task 0 1 2 3 4 36. When my mind goes blank I worry that there is something terribly wrong with me 0 1 2 3 4
Anxiety Sensitivity 257
Appendix C
Items that form the 21-item Anxiety Sensitivity Index
Anxiety Sensitivity 258
Anxiety Sensitivity 259
Items that form the 21-item Anxiety Sensitivity Index The items listed below reveal fears and worries, which we sometimes experience. Please read each item carefully and using the scale given below, rate it by circling the appropriate answer.
Very A little Some Much Very Little Much
1. When I feel like I’m not getting enough air, I get scared that I might suffocate 0 1 2 3 4 2. Smothering sensations scare me 0 1 2 3 4 3. It scares me when I become short of breath 0 1 2 3 4 4. When my chest feels tight, I get scared that I won’t be able to breath properly 0 1 2 3 4 5. It is important for me not to appear nervous 0 1 2 3 4 6. I believe it would be awful to vomit in public 0 1 2 3 4 7. I worry that other people will notice my anxiety 0 1 2 3 4 8. When I tremble in the presence of others I fear what people might think of me 0 1 2 3 4 9. When I begin to sweat in a social situation, I fear that people will think negatively of me 0 1 2 3 4 10. It scares me when I blush in front of people 0 1 2 3 4 11. When my head is pounding, I worry I could have a stroke 0 1 2 3 4 12. When my heart is beating rapidly, I worry that I might be having a heart attack 0 1 2 3 4 13. When my face feels numb, I worry that I might be having a stroke 0 1 2 3 4 14. When I feel a pain in my chest, I worry that I’m going to have a heart attack 0 1 2 3 4 15. When I notice my heart skipping a beat, I worry that something is seriously wrong with me 0 1 2 3 4 16. When I feel “spacey” or spaced out I worry that I may be mentally ill 0 1 2 3 4 17. When my thoughts seem to speed up, I worry that I might be going crazy 0 1 2 3 4 18. When I have trouble thinking clearly, I worry that there is something wrong with me 0 1 2 3 4 19. When I cannot keep my mind on a task, I worry that I might be going crazy 0 1 2 3 4 20. It scares me when I am unable to keep my mind on a task 0 1 2 3 4 21. When my mind goes blank I worry that there is something terribly wrong with me 0 1 2 3 4
Anxiety Sensitivity 260
Anxiety Sensitivity 261
Appendix D
Questionnaire Battery
Anxiety Sensitivity 262
Anxiety Sensitivity 263
CONSENT FORM
Chief Investigator Kerry Ann Armstrong School of Psychology and Counselling (Ph. 3864-4686) Associate Investigator Dr Nigar Khawaja School of Psychology and Counselling (Ph. 3864-4757) An Investigation of Anxiety Sensitivity and the role it plays in the Development of Anxiety and Anxiety Disorders You are invited to participate in a research project which aims to advance our understanding Anxiety Sensitivity and the role it plays in the development and maintenance of anxiety and anxiety disorders. If you agree to participate you will be asked to complete a series of questionnaires. These questionnaires will consist of questions regarding how you felt and what you thought at a time when you possibly experienced an episode of anxiety. Although your participation may have no direct benefit to you, it is hoped that the findings will be useful in advancing our understanding Anxiety Sensitivity and the factors that contribute to the development and maintenance of this construct as well as add to the current research on this topic. We anticipate that the questionnaire package take approximately 30 minutes to complete. There are no right or wrong answers when completing the questionnaire material. Additionally there is no time limit for responses; therefore you can work at a pace that is comfortable to yourself. However, do not spend too much time on any one question and please ensure that all questions are answered before returning your responses. Your participation in this project is entirely voluntary and you are able to discontinue your involvement in this study at anytime without explanation or penalty. Your confidentiality will be preserved and no identifying information will be made public, as only your student number will be recorded on the questionnaire material. Copies of the research report will be made available for interested persons upon request. This study is being conducted as part of a PhD project by Kerry Ann Armstrong, and is supervised by Dr Nigar Khawaja. You may contact Kerry Armstrong or Dr Khawaja during the study if any matter of concern arises on the phone number given above. You may also contact the secretary of the QUT Research Ethics Committee on 3864-2902 if you have any further concerns about the ethical conduct of this research.
Anxiety Sensitivity 264
I, _________________________________________________ (name) consent to participate in the research described above. I have read the information provided above and I have had the opportunity to ask questions. I also understand that I am able to withdraw from this study at any time without explanation, and any information I provide is treated as confidential. I am aged 18 years or over at the time of this study. _____________________________________________________ Signature of participant Date Kerry Ann Armstrong Researcher’s Name Signature of Researcher Date
5. In the home do you regularly speak a language other than English? (Circle one number only) No ....................................................................................... 0
Yes (please specify which language) .................................
6. What is Your Highest Level of Education Attained? (Circle one number only)
High School.......................................................................... 1 Tafe Certificate..................................................................... 2 Tafe Diploma / Associate Diploma....................................... 3 University Undergraduate.................................................... 4 University Postgraduate....................................................... 5 Other (please specify).......................................................... 6
Anxiety Sensitivity 267
Anxiety Sensitivity Index and Anxiety Sensitivity Index - Revised The items listed below reveal fears and worries, which we sometimes experience. Please read each item carefully and using the scale given below, rate it by circling the appropriate answer.
Very A little Some Much Very Little Much
1. When I feel like I’m not getting enough air, I get scared that I might suffocate 0 1 2 3 4 2. Smothering sensations scare me 0 1 2 3 4 3. It scares me when I become short of breath 0 1 2 3 4 4. When my chest feels tight, I get scared that I won’t be able to breath properly 0 1 2 3 4 5. It scares me when I feel faint 0 1 2 3 4 6. When my throat feels tight, I worry that I could choke to death 0 1 2 3 4 7. It scares me when my heart beats rapidly 0 1 2 3 4 8. When my breathing become irregular, I fear that something bad will happen 0 1 2 3 4 9. It scares me when I feel “shaky” (trembling) 0 1 2 3 4 10. When I have trouble swallowing, I worry that I could choke 0 1 2 3 4 11. It frightens me when my surroundings seem strange or unreal 0 1 2 3 4 12. It scares me when my body feels strange or different in some way 0 1 2 3 4 13. It is important for me not to appear nervous 0 1 2 3 4 14. I believe it would be awful to vomit in public 0 1 2 3 4 15. I think that it would be horrible for me to faint in public 0 1 2 3 4 16. I worry that other people will notice my anxiety 0 1 2 3 4 17. When I tremble in the presence of others I fear what people might think of me 0 1 2 3 4 18. When I begin to sweat in a social situation, I fear that people will think negatively of me 0 1 2 3 4 19. It scares me when I blush in front of people 0 1 2 3 4 20. When I feel a strong pain in my stomach, I worry it could be cancer 0 1 2 3 4 21. When my head is pounding, I worry I could have a stroke 0 1 2 3 4 22. When my heart is beating rapidly, I worry that I might be having a heart attack 0 1 2 3 4 23. When my face feels numb, I worry that I might be having a stroke 0 1 2 3 4 24. When I feel a pain in my chest, I worry that I’m going to have a heart attack 0 1 2 3 4 25. When I feel dizzy, I worry that there is something wrong with my brain 0 1 2 3 4 26. When my stomach is upset, I worry that I might be seriously ill 0 1 2 3 4
Anxiety Sensitivity 268
Very A little Some Much Very Little Much
27. When I notice my heart skipping a beat, I worry that something is seriously wrong with me 0 1 2 3 4 28. When I get diarrhea, I worry that I might have something wrong with me 0 1 2 3 4 29. It scares me when I am nauseous 0 1 2 3 4 30. It scares me when I feel tingling or prickling sensation in my hands 0 1 2 3 4 31. When I feel “spacey” or spaced out I worry that I may be mentally ill 0 1 2 3 4 32. When my thoughts seem to speed up, I worry that I might be going crazy 0 1 2 3 4 33. When I have trouble thinking clearly, I worry that there is something wrong with me 0 1 2 3 4 34. When I cannot keep my mind on a task, I worry that I might be going crazy 0 1 2 3 4 35. It scares me when I am unable to keep my mind on a task 0 1 2 3 4 36. When my mind goes blank I worry that there is something terribly wrong with me 0 1 2 3 4 37. It is important to me to stay in control of my emotions 0 1 2 3 4 38. It embarrasses me when my stomach growls 0 1 2 3 4 39. Other people usually notice when I feel shaky 0 1 2 3 4 40. Unusual body sensations scare me 0 1 2 3 4 41. When I am nervous, I worry that I might be mentally ill 0 1 2 3 4 42. It scares me when I am nervous 0 1 2 3 4
Anxiety Sensitivity 269
Beck Anxiety Inventory
Listed below are a variety of symptoms. Please read each symptom and by using the given scale rate how much you have been bothered by it over the past week. Read each item carefully and encircle the appropriate answer.
Not at all A Little Quite Severely, I Could Barely Stand It
1. Numbness or tingling 0 1 2 3
2. Feeling hot 0 1 2 3
3. Wobbliness in legs 0 1 2 3
4. Unable to relax 0 1 2 3
5. Fear of worst happening 0 1 2 3
6. Dizzy or lightheaded 0 1 2 3
7. Heart pounding or racing 0 1 2 3
8. Unsteady 0 1 2 3
9. Terrified 0 1 2 3
10. Nervous 0 1 2 3
11. Feeling of choking 0 1 2 3
12. Hands trembling 0 1 2 3
13. Shaky 0 1 2 3
14. Fear of losing control 0 1 2 3
15. Difficulty in breathing 0 1 2 3
16. Fear if dying 0 1 2 3
17. Scared 0 1 2 3
18. Indigestion or discomfort 0 1 2 3
19. Faint 0 1 2 3
20. Face flushed 0 1 2 3
21. Sweating not due to heat 0 1 2 3
Anxiety Sensitivity 270
COPE Questionnaire
We are interested in how people respond when they confront difficult or stressful events in their lives. There are a lot of ways to try and deal with stress. This questionnaire asks you to indicate whet you generally do and feel, when you experience stressful events. Obviously, different events bring out somewhat different responses, but think about what you usually do when you are under a lot of stress. Please respond to each of the following items by using the given scale. Choose your answers thoughtfully, and make your answers as true FOR YOU as you can. Please answer every item. There are no “right” or “wrong” answers, so choose the most accurate answer for YOU – not what you think “most people” would say or do. Indicate what YOU usually do when YOU experience a stressful event.
I usually don’t I usually do I usually do I usually don’t do this do this a this a medium do this a at all little bit amount lot
1. I try to grow as a person as a result of the experience 1 2 3 4 2. I turn to work or other substitute activities to take my mind off things 1 2 3 4 3. I get upset and let my emotions out 1 2 3 4 4. I try to get advice from someone about what to do 1 2 3 4 5. I concentrate my efforts on doing something about it 1 2 3 4 6. I say to myself “this isn’t real” 1 2 3 4 7. I admit to myself that I can’t deal with it, and quit trying 1 2 3 4 8. I restrain myself from doing anything too quickly 1 2 3 4 9. I discuss my feelings with someone 1 2 3 4 10. I get used to the idea that it happened 1 2 3 4 11. I talk to someone to find out more about the situation 1 2 3 4 12. I keep myself from getting distracted by other thoughts or activities 1 2 3 4 13. I daydream about things other than this 1 2 3 4 14. I get upset, and I am really aware of it 1 2 3 4 15. I seek God’s help 1 2 3 4 16. I make a plan of action 1 2 3 4 17. I accept that this has happened and that it can’t be changed 1 2 3 4 18. I put off doing anything until the situation permits 1 2 3 4 19. I try to get emotional support from friends and relatives 1 2 3 4 20. I just give up trying to reach my goal 1 2 3 4 21. I take additional action to try and get rid of the problem 1 2 3 4 22. I refuse to believe that it has happened 1 2 3 4
Anxiety Sensitivity 271
I usually don’t I usually do I usually do I usually don’t do this do this a this a medium do this a at all little bit amount lot
23. I let my feelings out 1 2 3 4 24. I try to see it in a different light, to make it seem more positive 1 2 3 4 25. I talk to someone who could do something concrete about the problem 1 2 3 4 26. I sleep more than usual 1 2 3 4 27. I try to come up with a strategy about what to do 1 2 3 4 28. I focus on dealing with this problem, and if necessary let other things slide a little 1 2 3 4 29. I get sympathy and understanding from someone 1 2 3 4 30. I give up the attempt to get what I want 1 2 3 4 31. I look for something good in what is happening 1 2 3 4 32. I think about how I might best handle the problem 1 2 3 4 33. I make sure not to make matters worse by acting too soon 1 2 3 4 34. I try hard to prevent other things from interfering 1 2 3 4 35. I go to movies or watch TV, to think about it less 1 2 3 4 36. I accept the reality of the fact that it happened 1 2 3 4 37. I ask people who have had similar experiences 1 2 3 4 38. I feel emotional distress and express those feelings 1 2 3 4 39. I force myself to wait for the right time to do something 1 2 3 4 40. I reduce the amount of effort I’m putting into solving the problem 1 2 3 4 41. I learn to live with it 1 2 3 4 42. I put aside other activities in order to concentrate on this 1 2 3 4 43. I think hard about what steps to take 1 2 3 4 44. I act as though it hasn’t ever happened 1 2 3 4 45. I do what has to be done, one step at a time 1 2 3 4
46. I learn something from the experience 1 2 3 4 47. I take direct action to get around the problem 1 2 3 4
Anxiety Sensitivity 272
DAS S – Stress Sub Scale
Please read each statement and circle a number 0, 1, 2 or 3 which indicates how much the statement applied to you over theThere are no right or wrong answers. Do not spend too much time on any statement.
The rating scale is as follows:
0 Did not apply to me at all; 1 Applied to me to some degree, or some of the time; 2 Applied to me to a considerable degree, or aof time; 3 Applied to me very much, or most of the time
I found it hard to wind down 0 1 2 3 I tended to over-react to situations 0 1 2 3 I felt that I was using a lot of nervous energy 0 1 2 3
I found myself getting agitated 0 1 2 3
I found it difficult to relax 0 1 2 3
I was intolerant of anything that kept me from getting on with what I was doing 0 1 2 3
I felt that I was rather touchy 0 1 2 3
Anxiety Sensitivity 273
Zung - SDS
Please indicate for each of these questions which answer best describes how you have been feeling during the past 4 weeks. Circle the most appropriate response. Circle only one response in each line. Please answer all the questions.
Rarely Some of A good part All or most the time of the time of the time
1. I feel downhearted, blue and sad 1 2 3 4
2. Morning is when I feel best 1 2 3 4
3. I have crying spells or feel like it 1 2 3 4
4. I have trouble sleeping through the night 1 2 3 4
5. I eat as much as I used to 1 2 3 4
6. I enjoy looking at, talking to and being with attractive men/women 1 2 3 4
7. I notice that I am losing weight 1 2 3 4
8. I have trouble with constipation 1 2 3 4
9. My heart beats faster than usual 1 2 3 4
10. I get tired for no reason 1 2 3 4
11. My mind is as clear as it used to be 1 2 3 4
12. I find it easy to do the things I used to 1 2 3 4
13. I am restless and can’t keep still 1 2 3 4
14. I feel hopeful about the future 1 2 3 4
15. I am more irritable than usual 1 2 3 4
16. I find it easy to make a decision 1 2 3 4
17. I feel that I am useful and needed 1 2 3 4
18. My life is pretty full 1 2 3 4
19. I feel that others would be better off if I were dead 1 2 3 4
20. I still enjoy the things I used to 1 2 3 4
Anxiety Sensitivity 274
The Self-Efficacy Scale To what extent do each of the following statements apply to you. Please circle the number that indicates your level of agreement.
1. When I make plans I am certain I can make them work 1 2 3 4 5 2. I feel insecure about my ability to do things 1 2 3 4 5 3. When I set important goals for myself, I rarely achieve them 1 2 3 4 5 4. I give up on things before completing them 1 2 3 4 5 5. If something looks too complicated, I will not even bother to try it 1 2 3 4 5 6. One of my problems is that I cannot get down to work when I should 1 2 3 4 5 7. I avoid facing difficulties 1 2 3 4 5 8. When I have something unpleasant to do, I stick to it until I finish it 1 2 3 4 5 9. I give up easily 1 2 3 4 5 10. When I decide to do something, I go right to work on it 1 2 3 4 5 11. When trying to learn something new, I soon give up if I am not initially successful 1 2 3 4 5 12. When unexpected problems occur, I do not handle them very well 1 2 3 4 5 13. I avoid trying to learn new things when they look too difficult for me 1 2 3 4 5 14. Failure just makes me try harder 1 2 3 4 5 15. If I can’t do a job the first time, I keep trying until I can 1 2 3 4 5 16. I am a self-reliant person 1 2 3 4 5 17. I do not seem capable of dealing with most problems in life 1 2 3 4 5
Anxiety Sensitivity 275
Fear Questionnaire (How You Feel in Certain Situations)
Choose a number from the scale below to show how much you would avoid each of the situations listed because of fear or other unpleasant feelings. Please circle the number opposite each situation.
0 1 2 3 4 5 6 7 8 Would not Slightly Definitely Markedly Always avoid it avoid it avoid it avoid it avoid it
1. Main phobia (fear) you want treated (describe in your own words and rate) 0 1 2 3 4 5 6 7 8 2. Injections or minor surgery 0 1 2 3 4 5 6 7 8 3. Eating or drinking with other people 0 1 2 3 4 5 6 7 8 4. Hospitals 0 1 2 3 4 5 6 7 8 5. Traveling alone by bus or coach 0 1 2 3 4 5 6 7 8 6. Walking alone in busy streets 0 1 2 3 4 5 6 7 8 7. Being watched or stared at 0 1 2 3 4 5 6 7 8 8. Going into crowded shops 0 1 2 3 4 5 6 7 8 9. Talking to people in authority 0 1 2 3 4 5 6 7 8 10. Sight of blood 0 1 2 3 4 5 6 7 8 11. Being criticised 0 1 2 3 4 5 6 7 8 12. Going alone far from home 0 1 2 3 4 5 6 7 8 13. Thought of injury or illness 0 1 2 3 4 5 6 7 8 14. Speaking or acting to an audience 0 1 2 3 4 5 6 7 8 15. Large open spaces 0 1 2 3 4 5 6 7 8 16. Going to the dentist 0 1 2 3 4 5 6 7 8 17. Other situations (describe and rate) 0 1 2 3 4 5 6 7 8
Anxiety Sensitivity 276
Anxiety Sensitivity 277
Appendix E
Outline of the manual based cognitive behavioural therapy
for anxiety employed in the current dissertation
Anxiety Sensitivity 278
Anxiety Sensitivity 279
OUTLINE OF THE MANUAL BASED COGNITIVE BEHAVIOURAL THERAPY FOR ANXIETY EMPLOYED IN THE CURRENT
DISSERTATION Day 1 Schedule
• Philosophy of the Program
• What is Anxiety?
• Main Types of Anxiety Disorders Panic Disorder (with or without Agoraphobia)
• Other Types of Anxiety Disorders Generalised Anxiety Disorder Simple Phobia Social Phobia Obsessive Compulsive Disorder (OCD) Post Traumatic Stress Disorder (PTSD) Differential Diagnosis
• Organic Disorders and Conditions Associated with Panic Diagnostic Criteria for Social Phobia Diagnostic Criteria for Generalised Anxiety Disorder Comorbidity Who Suffers from Panic Disorder? How can Panic Disorder be treated? Well-Established Treatments for Anxiety Disorders Symptom Control
• Understanding Anxiety I: Cognitions
• Understanding Anxiety II: Symptoms
• Understanding Panic Attacks
• Whys and the Why
• Understanding Fear: The Fire Analogy
• The Fire Analogy: Explanation
• Day 1 Homework
Anxiety Sensitivity 280
Day Two Schedule
• The ABC model of emotion • Symptoms Vs Fear Vs Cognition
• Three Phases of a Panic Attack
Phase One Phase Two Phase Three
• Different Techniques for each phase
Techniques for Phase One Techniques for Phase Two Techniques for Phase Three
• Body, Behaviour, Cognition
• Learning to Control Physical Sensations Slow Breathing exercise Relaxation Techniques – SECTION 1
Relaxation Techniques 2 – Rag Doll Technique Relaxation Techniques 3 – Relaxation by using Imagery Relaxation Techniques 4 – Autogenic Training When and Where Should Relaxation Strategies be Used?
• Day Two Homework Day Three Schedule
• The Flip-Flop Model
• Dangerous Mind Games that We Play: A Lose-Lose game
• A New Mind Game that I Need to Play: A Win-Win Game
• Day Three Homework Day Four Schedule
• Hyperventilation and Breathing Retraining • Hyperventilation
• Breathing Retraining
• Cognitive Distortions
Anxiety Sensitivity 281
• Questions to Help You Challenge Negative Thinking
• Positive Self-Statements
1. Preparation for the Event or Situation 2. Confronting the Event or Situation 3. Coping with Fear 4. Evaluation of Coping and Rewards for Successful Coping
• Cognitive Rehearsal • Problem Solving
• Day Four Homework
Day Five Schedule
• In Vivo Situational Exposure
• Interoceptive Exposure
• More Advanced Relaxation Techniques
• Day Five Homework
Day Six Schedule
• Changing the Way You Think
• Five Falsity Tests for Beliefs
• The Vertical Arrow Technique
• Distraction Techniques
• Thought Stopping
• Past, Present and Future
• Balance Sheet
• More Advanced Relaxation Techniques Instructions for Quick Isometric Relaxation
• Relaxation by Recall
• Day Six Homework
Anxiety Sensitivity 282
Day Seven Schedule
• Building a Support System
• Sources of Support
• Day Seven Homework
Day Eight Schedule
• Maintenance
• Relapse Prevention
• Day 8 Homework
All Information Contained in the Full Manual (including this Appendix) is
Protected by Copywrite to the Cognitive Behavioural Therapy (CBT) Day
Treatment Program at the Toowong Private Hospital 2000.
Anxiety Sensitivity 283
Appendix F
Outline of the manual based cognitive behavioural therapy for depression employed in the current dissertation
Anxiety Sensitivity 284
Anxiety Sensitivity 285
OUTLINE OF THE MANUAL BASED COGNITIVE BEHAVIOURAL THERAPY FOR DEPRESSION EMPLOYED IN THE CURRENT
DISSERTATION Day 1 Schedule
• Introduction and Ground Rules • Philosophy of the Program
• Cognitive Behavioural Therapy (CBT)
• Spring Cleaning
• Four Levels of Change
• Depression: Types and variations
• Depression is treatable
• Evidence based treatment for depression
• Criteria for Major Depressive Episode (from DSM-IV)
• Diagnostic criteria for Dysthymic Disorder (from DSM-IV)
• The 3 Ingredients of Emotion
• Common Errors regarding Emotion
• “You” and “Events / Others”
• Past events cannot be changed, but our memories of events can be
changed
• Pharmacotherapy
• Medication: Regular Review
• Medication
• Information about Medication
• Some Adverse Effects of Pharmacological Agents
Anxiety Sensitivity 286
• Medication commonly used in the Treatment of Depression and their
side-effects
• Pharmacologic Properties of Antidepressants
• Preparations and Dosage of Antidepressants
• Benzodiazepines used in the Treatment of Depression and Anxiety
Disorders
• Imagery Exercise – Day One
• Homework: Do Not A Do, A Dodo
Day 2 Schedule
• The Behavioural Component: Behavioural Do
• Some Activities are Useful to Combat Depression
• Reasons for Planning an Doing Activities that Combat Depression
• “Cookie Jar” Activities
• Suggestions for Pleasant Events
• Daily Schedule of Activities
• Exercise: Do It
• Daily Exercise Record
• Decision Making about Problems
• Homework Activity
Day 3 Schedule
• Cognitions
• Problems of Social Comparisons
Anxiety Sensitivity 287
• Present Characteristics Vs Future Goals
• Overwhelmed by Responsibility
• Automatic Thoughts (AT)
• The Link between Schemas and Automatic Thoughts
• Cognitive Triad and Schema
• Information from your Environment
• Automatic Thoughts Link Together or Talk to Each Other
• Fighting Automatic Thoughts
• Cognitive Distortions
• Thought that Cause People Problems
• Dangerous Mind Games that We Play: A lose-lose game
• Whys and the Why
• New Mind Games That I Need to Play: The win-win Game
• Thought Processes Vs Thought Content
• Daily Beliefs Monitoring Schedule Daily Record of Dysfunctional Thoughts 1 Example: Daily Record of Dysfunctional Thoughts 1
• Day 3 Homework
Day 4 Schedule
• Cognitive Restructuring Questions
• Counters
• Types of Counters
• Practice Counters
• Thoughts that Cause People Problems
• Mantras
Anxiety Sensitivity 288
Day 5 Schedule
• Five Falsity Tests for Beliefs
• Balance Sheet
• Changing our Memory of Unpleasant Events
• A Process of Change: Yes I Can
Day 6 Schedule
• The Vertical Arrow Technique
• The Automatic Thoughts – Iceberg Analogy
• The Characteristics of the Vertical Arrow
• Counters to Combat Automatic Thoughts
• Process of Combating Automatic Thoughts: Planting a New Tree
Analogy
• Reasons for Turning Counters to Mantras
• Logical Analysis Logical Analysis Worksheet Logical Analysis Worksheet Example 1 Logical Analysis Worksheet Example 2 Day 7 Schedule
• Problem Solving
• Past, Present, and Future
• Balance Sheet
• Building a Support System
• Sources of Support
• The Importance of Understanding Cost
Anxiety Sensitivity 289
Day 8 Schedule
• Relapse Prevention
• A Smorgasbord of Strategies to Life Depression
• Reading List
• Self Help books about Depression and Bipolar Disorder
• Autobiographical Accounts of Depression or Bipolar Disorder
• For your Supporters
• Self help books about Stress and anxiety
• Author Publications
All Information Contained in the Full Manual (including this Appendix) is
Protected by Copywrite to the Cognitive Behavioural Therapy (CBT) Day
Treatment Program at the Toowong Private Hospital 2000.
Anxiety Sensitivity 290
Anxiety Sensitivity 291
Appendix G
Preliminary screening sheet
Anxiety Sensitivity 292
Anxiety Sensitivity 293
PRELIMINARY SCREENING SHEET Date: …………………………………………………………………. Name: …………………………………………………………………. Sex: M F Age: …………… Contact Phone No: (W)…………………………….. (H)…………………………. Address:…………………………………………………………………… …………………………………………………………………….. Q1: Do you have a major physical problem? Y N If yes, what is the problem……………………………………………….. Are you receiving any treatment for it …..……………………..………... Q2: Do you have any other problem (emotional, neurological)? Y N If yes, what is the problem……………………………………………….. Are you receiving any treatment for it .…………………………………... Q3: Have you been hospitalised recently? Y N If yes, what was the problem…………………………………………….. Are you receiving any treatment for it..…………………………………... Q4: Do you drink alcohol? Y N If yes, has alcohol affected your health? Y N your family? Y N work? Y N If yes, are you receiving treatment for it?………………………………..
CONTINUE OVER PAGE
Anxiety Sensitivity 294
Q5: Do you take any social or recreational drugs? Y N
If Yes, which one/s……………………………………………………….
Has taking this drug made you confused? Y N See things? Y N Hear Voices? Y N Have strange ideas? Y N
Any other?………………………………………………………………… Did you have to go to hospital because of taking this drug? Y N
Q6: Are you depressed? Y N If yes, how is your daily life affected? …………………………………… (a) Do you have a loss of appetite Y N (b) Do you have trouble sleeping Y N (c) Do you wake up early in the morning Y N (d) Do you feel worthless Y N (e) Do you feel like life is not worth living Y N
Q7: Do you feel elated (hyper)? Y N If yes, have you been… (a) Over talkative Y N (b) Shouting at people Y N (c) Arguing with people Y N (d) Experiencing thoughts racing through your mind Y N (e) Involved in reckless activity Y N Q8: Do you feel very high and very low at times? Y N Q9: Now I am going to ask you about some unusual experiences which people
sometimes have. (a) Do you hear voices/sounds that others can’t hear? Y N (b) Do you see things that other’s can’t see? Y N (c) Do you smell things that other’s can’t smell? Y N (d) Do you think that you are especially important
in some way and have powers to do things that others can’t do? Y N
Anxiety Sensitivity 295
(e) Do you think that people are taking special notice of you, or talking about you? Y N
(f) Is anybody going out of the way to give you a hard time, or to hurt you Y N
(g) Do you receive special messages from other people, newspapers, radio or T.V? Y N
Notes on Yes responses …………………………………………………………………………………… …………………………………………………………………………………… …………………………………………………………………………………… …………………………………………………………………………………… Q 10: Do senseless thoughts keep on recurring Y N Do you keep on repeating acts such as: (a) washing self/object (s) Y N (b) checking objects Y N (c) counting objects Y N (d) any other? ………………………………………… Y N Q11: Do you ever have a sudden rush of anxiety? Y N If yes, do you at that time feel (a) shaky / trembling Y N (b) sweating Y N (c) out of breath Y N (d) choking Y N (e) detached from surroundings Y N (f) fear of dying Y N (g) fear of going crazy Y N (h) fear of going out of control Y N Q12: Are you anxious when (a) being alone Y N (b) going to crowded shopping centres Y N (c) using public transport Y N (d) using lifts / elevators Y N (e) crossing a busy street Y N (f) any other?...................................................................Y N
Anxiety Sensitivity 296
Anxiety Sensitivity 297
Appendix H
Handbook on the effects and treatment
of anxiety and anxiety disorders
halla
This appendix is not available online. Please consult the hardcopy thesis available from the QUT Library
Anxiety Sensitivity 327
Appendix I
Instruction Sheet for Participants
Anxiety Sensitivity 328
Anxiety Sensitivity 329
Instructions for Participants
Thank you for appearing for this study. My name is Kerry Ann Armstrong
and the current project is being conducted for the purposes of my PhD
research, which is supervised by Dr Nigar Khawaja. The study is
concerned with the measurement of anxiety sensitivity and its relationship
to other measures.
As part of this study your task is to complete the questionnaires in the
order they appear in the booklet. As there is no set time limit, you may
work at a speed which you feel comfortable. There are no right or wrong
answers to items in the questionnaires. Your responses will be
completely confidential. When you have completed the questionnaires, I
will initial your ‘Applied Research Experience’ sticker and you are free to
leave. There will be a 20 minute debriefing session after all
questionnaires have been returned for those of you who are interested.
Does anyone have any questions before we start?
If you have any questions about the task or materials as you are
completing the questionnaire items, please raise your hand and I will
attempt to answer your question.
Again, thank you for participating in this study.
Queensland University of Technology School of Psychology and Counselling