Intolerance of uncertainty and social anxiety

Paul A. Boelen
a,
*, Albert Reijntjes
b
a
Department of Clinical and Health Psychology, Utrecht University, P.O. Box 80140, 3508 TC Utrecht, The Netherlands
b
Department of Psychosocial Development in Context, Utrecht University, The Netherlands
1. Introduction
Anxiety disorders are among the most prevalent of all
psychiatric disorders (Kessler, Berglund, Demler, Jin, & Walters,
2005). The recent decades have seen advancement of our
understanding of the etiology of anxiety disorders. An important
premise in theories about anxiety disorders is that they contain
both common and unique components (Brown & Barlow, 2002;
Mineka, Watson, & Clark, 1998). Specically, theorists have
proposed and research has conrmed that all anxiety disorders
(as well as mood disorders) involve elevated levels of negative
affectivity or neuroticism (Mineka et al., 1998), whereas specic
cognitive factors are responsible for the development of specic
anxiety disorders. Examples of such specic factors include anxiety
sensitivity in panic disorder, fear of negative evaluation in social
phobia, excessive responsibility beliefs in obsessive compulsive
disorder (OCD), and intolerance of uncertainty in generalized
anxiety disorder (GAD) (cf. Starcevic & Berle, 2006).
Identifying factors that are unique to specic anxiety disorders
is important for understanding and treatment of these disorders. In
recent years, there is growing interest in one of such specic
factors, namely intolerance of uncertainty (IU). IUhas been dened
as a cognitive bias that affects how a person perceives, interprets,
and responds to uncertain situations on a cognitive, emotional, and
behavioral level (Dugas, Schwarzt, &Francis, 2004, p. 835). People
high in IU experience the possible occurrence of future negative
events as stressful, believe that uncertainty is negative, reects
badly on a person, and should be avoided, and have difculties
functioning well in uncertain situations (Buhr & Dugas, 2002).
In studies with non-clinical samples (e.g., Holaway, Heimberg,
& Coles, 2006) and clinical samples (e.g., Dugas & Ladouceur, 2000)
IU has consistently been found to be correlated with GAD.
Nevertheless, studies examining specicity of IU to GAD have
yielded mixed results (Starcevic & Berle, 2006). On the one hand,
there is evidence that IU is a relatively unique component of GAD.
For instance, studies have shown that high levels of IU distinguish
patients with GAD from patients with panic disorder (Dugas,
Marchand, & Ladouceur, 2005) and from a mixed group of patients
with other anxiety disorders (Ladouceur et al., 1999). Moreover, IU
has been found to be more strongly related with pathological
worry which is a key symptom of GAD than with depression
(Dugas et al., 2004). On the other hand, there are studies suggesting
that IU is critical to both GAD and OCD. For instance, in a study
among undergraduate students, Holaway et al. (2006) found that
people with clinical signicant levels of GAD and OCD reported
higher levels of IU than controls, but did not differ signicantly
from each other in terms of IU levels (also see Steketee, Frost, &
Cohen, 1998; Tolin, Abramowitz, Brigidi, & Foa, 2003).
To our knowledge, no studies have yet examined the relation-
ship of IU with social anxiety. Yet, it is possible that IU contributes
Journal of Anxiety Disorders 23 (2009) 130135
A R T I C L E I N F O
Article history:
Received 6 March 2008
Received in revised form 29 April 2008
Accepted 29 April 2008
Keywords:
Social-anxiety
Intolerance-of-uncertainty
A B S T R A C T
Research has shown that intolerance of uncertainty (IU) the tendency to react negatively to situations
that are uncertain is involved in generalized anxiety disorder (GAD). There is uncertainty about the
specicity of IU. Some studies have shown that IU is specic for GAD. Other studies have shown that IU is
also involved in obsessive compulsive disorder (OCD). No studies have yet examined IU in social anxiety,
although it is possible that IU plays a role in anxiety responses that can be experienced in social-
evaluative situations. This study examined the relationship between IU and social anxiety among 126
adults. Findings revealed that IU explained a signicant amount of variance in social anxiety severity
when controlling for established cognitive correlates of social anxiety (e.g., fear of negative evaluation)
and for neuroticism. Furthermore, it was found that IUwas related with symptomlevels of GAD, OCD, and
social anxiety, but not depression, when controlling the shared variance among these symptoms.
2008 Elsevier Ltd. All rights reserved.
* Corresponding author. Tel.: +31 30 2533021; fax: +31 30 2534718.
E-mail address: P.A.Boelen@uu.nl (P.A. Boelen).
Contents lists available at ScienceDirect
Journal of Anxiety Disorders
0887-6185/$ see front matter 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.janxdis.2008.04.007
to the severity of social anxiety symptoms. People with elevated
social anxiety experience a marked and persistent fear in social
situations in which they are exposed to possible scrutiny by others
(Hofmann & Barlow, 2002). Clearly, uncertainty, ambiguity, and
unpredictable change are inherent to such social-evaluative
situations. Thus, it seems plausible that intolerance of such
uncertainty, ambiguity, and unpredictable change in the form
of nding that uncertainty is stressful, reects badly on a person,
and blocks constructive action (i.e., IU) is associated with the fear
(e.g., of being criticized), avoidance (e.g., of talking to strangers),
and physical discomfort (e.g., blushing, trembling) that some people
experience in such situations.
The present study, conducted in The Netherlands, addressed
several interrelated issues, the overarching aim of which was to
improve our understanding of the role of IU in social anxiety.
Specically, this study had two goals.
The rst goal was to examine the degree to whichIUwas related
to social anxiety severity, when taking into account a number of
cognitive variables that have been found to be associated with
social anxiety. Cognitive behavioral models of social anxiety have
proposed (Clark & Wells, 1995; Rapee & Heimberg, 1997) and
research has conrmed (Weeks et al., 2005) that fear of negative
evaluation (FNE), dened as fear of being judged disparagingly,
critically, or hostilely by others, is strongly linked with social
anxiety. Furthermore, social anxiety has been found to be
associated with anxiety sensitivity (Orsillo, Lilienfeld, & Heimberg,
1994), low self-esteem (Kocovski & Endler, 2000), various dimen-
sions of perfectionism (Juster et al., 1996), and with pathological
worry (Starcevic et al., 2007). Our rst goal was to examine if IU
explained variance in social anxiety severity, above and beyond the
variance explained by these established cognitive correlates of
social anxiety. Because we wished to examine the contribution of
IU and the other cognitive correlates to the explained variance in
social anxiety independent of neuroticism, this variable was
controlled for as well.
The second goal of this study was to further our understanding
of the generality vs. specicity of IUby examining the specicity of
the relationship between IU and GAD, with regards to social
anxiety. For comparative reasons, we also included measures of
OCDand depression symptoms, expecting that IUwould be related
with symptom levels of GAD, social anxiety, and OCD, but not
depression (Dugas et al., 2004), when controlling for the shared
variance between symptoms.
2. Method
2.1. Participants and procedure
Data were available from 126 adults. Participants were
originally recruited for a longitudinal study on coping with loss,
primarily designed to examine the predictive effects of various
types of coping behaviors on the development of complicated grief
(Boelen and Van den Hout, 2008). All participants were recruited
through an advertisement on an Internet-site that briey
explained aims of the study on grief and invited people to
participate by lling in questionnaires. In total, 568 questionnaires
were sent to people who expressed their interest in participation
and 404 (71%) questionnaires were returned. Participants who had
suffered a loss less than 5 months prior to completion of the initial
questionnaires (N = 121) were invited to complete additional
questionnaires for the study on grief. All remaining participants
(N = 283) were invited to participate in the present study on IU.
This invitation and the questionnaires for the current study were
sent to them together with a brief written report on the outcomes
of the grief study for which they were originally recruited. Of these
283 participants, 126 (44.5%) completed the battery of ques-
tionnaires for the current study. Participants did not receive any
nancial compensation in return for their participation.
On average, participants were 47.7 (S.D. = 11.7) years of age.
Most (91.3%) were woman. With respect to educational level, one
participant had nished primary school, 47.2% had nished high
school, and 52.0% had been to college or university. Losses that
participants had suffered all had occurred at least 18 months prior
to participation in the present study.
2.2. Measures
The questionnaire package included a consent form, an
information letter, and the following measures. Questionnaires
were administered in the same order across participants.
2.2.1. Social Phobia Inventory (SPIN)
The SPINis a 17-itemquestionnaire constructed by Connor et al.
(2000). It was designed as a concise measure for the assessment of
the fear, avoidance, and physiological symptoms that characterize
social phobia. Respondents rate the presence of symptoms in the
preceding week, on 4-point scales ranging from 0 (not at all) to 4
(extremely). Items are summed to form an overall social anxiety
severity score. The measure has been found to have good
psychometric properties (Connor et al., 2000). To construct the
Dutch version, the rst and second author translated the English
version of the SPIN into Dutch. Then, an independent researcher
who was uent in Dutch and English and familiar with the concept
of social anxiety checked if each question was properly translated.
Discrepancies between both Dutch versions were then discussed
and cleared to make sure that the Dutch translation resembled the
original formulation as closely as possible. In the present sample,
Cronbachs a of the SPIN was 0.93.
2.2.2. Intolerance of Uncertainty Scale (IUS)
The IUS is a 27-item measure of IU developed by Freeston,
Rheaume, Letarte, Dugas, and Ladouceur (1994). It taps different
aspects of IU such as the idea that uncertainty is unacceptable,
reects badly on a person, and interferes with active coping.
Respondents rate the degree to which each of 27 items apply to
them on 5-point scales ranging from 1 (not at all characteristic of
me) to 5 (entirely characteristic of me). The original French version
(Freeston et al., 1994) as well as the English version (Buhr & Dugas,
2002), and the Dutch version used in this study (De Bruin, Rassin,
van der Heiden, & Muris, 2006) have demonstrated adequate
psychometric properties. In the present sample, Cronbachs a was
0.95.
2.2.3. Brief Fear of Negative Evaluation Scale (BFNE)
The BFNE (Weeks et al., 2005) is a 12-itemmeasure tapping the
fear to be judged negatively by others. Respondents rate the extent
to which items tapping this fear are characteristic of them on 5-
point scales ranging from 0 (not at all characteristic of me) to 4
(extremely characteristic of me). The scale has been found to have
adequate psychometric properties (Weeks et al., 2005). The Dutch
version was obtained from Bo gels (2004) and, in the present
sample, had an a of 0.97.
2.2.4. Anxiety Sensitivity Index (ASI)
The ASI is a 16-item questionnaire developed by Peterson and
Reiss (1992) designed to tap the tendency to fear anxiety-related
bodily sensations. Respondents rate the extent to which they
experienced fears represented in the items in the preceding week
on 5-point scales ranging from0 (very little) to 4 (very much). Both
the English (Peterson & Reiss, 1992) and Dutch version (Vujanovic,
P.A. Boelen, A. Reijntjes / Journal of Anxiety Disorders 23 (2009) 130135 131
Arrindell, Bernstein, Norton, & Zvolensky, 2007) of the measure
have good psychometric properties. In this sample, the total score
was used as a global index of anxiety sensitivity and Cronbachs a
was 0.90.
2.2.5. Rosenberg Self-Esteem Scale (SES)
The SES is a 10-item measure of self-esteem developed by
Rosenberg (1989). Respondents rate their agreement with each of
10 positively and negatively worded self-statements on 4-point
scales ranging from 0 (strongly agree) to 3 (strongly disagree). The
total score is calculated such that higher scores reect lower self-
esteem. Both the English (Rosenberg, 1989) and Dutch version
(Franck, De Raedt, Barbez, & Rosseel, in press) have good
psychometric properties. Cronbachs a in the present sample
was 0.88.
2.2.6. Multidimensional Perfectionism Scale (MPS)
The MPS is a 35-item questionnaire of perfectionismdeveloped
by Frost, Marten, Lahart, and Rosenblate (1990). It can be used to
obtain an overall index of perfectionism, as well as separate scores
for six dimensions of perfectionism: Concern over Mistakes (CM),
Doubts about Actions (DA), Personal Standards (PS), Parental
Expectations (PE), Parental Criticism (PC) and Organization (OR).
Respondents rate their level of agreement with each item on 5-
point scales ranging from 0 (strongly disagree) to 4 (strongly
agree). Psychometric properties of the measure are adequate (e.g.,
Frost et al., 1990) with the 6-factor structure being conrmed
(Juster et al., 1996). We only used three subscales that have been
found to be correlated with social anxiety: CM, DA, and PC (cf.
Juster et al., 1996; Rosser, Issakidis, & Peters, 2003). These
subscales were specically translated for the current study, using
the procedure described with the SPIN. Cronbachs as of the
subscales in this sample were 0.92 for CM, 0.87 for DA, and 0.71
for PC.
2.2.7. Penn State Worry Questionnaire (PSWQ)
The PSWQ is a 16-item measure of pathological worry
developed by Meyer, Miller, Metzger, and Borkovec (1990).
Respondents rate degree to which items reecting worry are
typical for them on 5-point scales ranging from 1 (not at all typical
of me) to 5 (very typical of me). Psychometric properties of the
original English version (Meyer et al., 1990) and Dutch version
(Rijsoort, van Emmelkamp, & Vervaeke, 1999) are adequate.
Cronbachs a in the present sample was 0.94.
2.2.8. Generalized Anxiety Disorder Questionnaire for
DSM-IV (GAD-Q-IV)
The GAD-Q-IV is a 9-item measure constructed by Newman
et al. (2002) to assess symptoms of GADas dened in DSM-IV(APA,
1994). It contains ve dichotomous (yes/no) items tapping the
occurrence of worry, one itemassessing worry topics, a checklist of
six dichotomously scored somatic symptoms related to GAD, and
two questions assessing the degree of distress and disability linked
with the worry, scored on 9-point scales ranging from0 (none) to 8
(very severe). The GAD-Q-IVcan be used to either classify cases and
non-cases of GAD or to obtain a continuous score of GAD severity.
In the current study, this latter scoring format was used. The
measure was specically translated for the current study using the
procedure that was described with the SPIN. Cronbachs a in the
present sample was 0.89.
2.2.9. Depression Subscale of the Hospital Anxiety and Depression
Scale (HADS-D)
The 7-itemdepression subscale of the HADS (Zigmond &Snaith,
1983) was used to assess depressive symptom severity. Respon-
dents rate the presence of seven depressive symptoms on 4-point
scales, which are summed to yield a depression severity score. The
English (Zigmond & Snaith, 1983) and Dutch versions (Spinhoven
et al., 1997) have adequate psychometric properties. In the current
sample, Cronbachs a was 0.90.
2.2.10. Obsessive Compulsive Inventory revised version (OCI-R)
The OCI-R is an 18-item measure constructed by Foa et al.
(2002) to assess OCD symptomatology. Respondents rate the
presence of OCD symptoms on 5-point scales ranging from not at
all to extremely. The OCI-R consists of six subscales (i.e.,
washing, checking, ordering, obsessing, hoarding, and neutraliz-
ing) and its total score provides a general index of OCD severity.
This total score was used in this study. Psychometric properties of
the English (Foa et al., 2002; Hajcak, Huppert, Simons, & Foa, 2004)
and Dutch version (Cordova-Middelbrink et al., 2007) of the OCI-R
have been found to be good. In this sample, Cronbachs a was 0.85.
2.2.11. Shortened Eysenck Personality Questionnaire (EPQ)
neuroticism subscale
The neuroticism subscale from the shortened EPQ (Eysenck,
Eysenck, & Barrett, 1985), the Dutch version of which was
developed and validated by Sanderman et al. (1995), was used
to asses the level of neuroticism. Respondents indicate their
agreement with items, using a forced-choice response format (yes
vs. no). In this sample, Cronbachs a was 0.87.
3. Results
3.1. Preliminary analyses
Table 1 shows means and S.D.s for all measures. The mean score
on the SPIN (tapping social anxiety) in the current sample was
M = 11.2 (S.D. = 11.2). This score was similar to the mean score of
M = 12.1 (S.D. = 9.3) of 68 healthy volunteers in the study of Connor
et al. (2000); (t(125) = 0.87, p = 0.39). IUS, PSWQ, and GAD-Q-IV
scores resembled those of student samples (cf. Dugas et al., 2004;
Holaway et al., 2006). The same applied to scores on the BFNE (cf.
Weeks et al., 2005), OCI-R (Hajcak et al., 2004) and MPS subscales
(Juster et al., 1996). Mean HADS-D scores in this sample (M = 4.1,
S.D. = 4.3) did not differ signicantly from the mean score of
M = 3.4 (S.D. = 3.3) of a non-clinical sample of 199 adults drawn
from the general Dutch population (t(125) = 1.80, p = 0.08)
(Spinhoven et al., 1997). Findings suggest that, although all
participants had experienced the death of a close relative, their
psychological distress was at the level of the normal population.
None of the scores on the study measures differed as a function
of gender or age of the participants. Normality data (e.g.,
skewnesses) indicated that all measures, except the GAD-Q-IV,
SES, and EPQ neuroticism subscale were positively skewed.
Therefore, these variables were log-transformed in all regression
analyses presented below.
To be able to explore associations between the study measures,
nonparametric correlations (Spearman r) were calculated. These
are shown in Table 1. To examine the statistical signicance of the
78 correlations that were calculated, a Bonferroni correction was
applied with the alpha level being set at 0.0006 (i.e., 0.05/78). All
correlations were signicant, except the correlations between the
MPS subscale Parental Criticism and most other measures.
3.2. Regression with social anxiety severity as dependent variable
Next, we examined the extent to which IUexplained variance in
social anxiety, when controlling for neuroticism and for the
following seven cognitive correlates of social anxiety: FNE, anxiety
P.A. Boelen, A. Reijntjes / Journal of Anxiety Disorders 23 (2009) 130135 132
sensitivity, low self-esteem, the three perfectionism subscales
fromthe MPS (i.e., CM, DA, and PC), and pathological worry. To this
end, we conducted hierarchical regression analysis predicting SPIN
scores, in which neuroticism was entered in block 1, the seven
cognitive variables were included in a stepwise analysis in block 2,
and IU was added to the equation in block 3. We used the stepwise
procedure in block 2 because we had a relatively large number of
correlated independent variables and a relatively small sample.
Findings are summarized in Table 2.
Outcomes showed that neuroticism (entered in block 1)
explained 38.1% of the variance in social anxiety. FNE and anxiety
sensitivity were consecutively selected into the equation in block 2
explaining an additional 14.7% and 2.3% of the variance,
respectively. IU, entered in block 3, accounted for an additional
4% of the variance in social anxiety severity. In the nal model,
neuroticism, FNE, and IU emerged as unique correlates of social
anxiety. Anxiety sensitivity no longer had a unique association
with social anxiety.
3.3. Relative importance of FNE and IU
Given that FNE and IU were the only two cognitive variables
that explained unique variance in social anxiety severity over and
above neuroticism, it was considered relevant to examine the
relative contribution of both variables to the explained variance in
SPIN scores when controlling for neuroticism. To this end, two
hierarchical regression analyses predicting SPIN scores were
carried out, one in which neuroticism and IU were entered in
the rst block followed by FNE in block 2, and one in which
neuroticismand FNE were entered in the rst block, followed by IU
in block 2.
The rst of these regressions showed that FNE explained 6.6% of
the variance in social anxiety over and above the 51.6% explained
by neuroticism and IU (F change (1, 121) = 18.96, p < 0.001). The
second regression showed that IUexplained 5.4% of the variance in
SPINscores over and above the 52.8%explained by neuroticismand
FNE (F change (1, 121) = 15.52, p < 0.001). Together, neuroticism,
FNE, and IU explained 58.2% of the variance in social anxiety
severity (F(3, 124) = 56.14, p < 0.001).
3.4. Specicity of IU to social anxiety, GAD, OCD, and depression
In our nal round of analyses, we examined the degree to which
IU was specically related to symptom levels of GAD, social
anxiety, OCD, and depression when controlling for the shared
variance among these symptoms. To this end, IUS scores were
regressed on GAD-Q-IV, SPIN, OCI-R, and HADS-D scores entered
simultaneously. Outcomes are summarized in Table 3. Together,
these four symptomclusters explained 66.2% of the variance in IUS
scores (F(4, 113) = 53.48, p < 0.001). Levels of social anxiety, GAD,
and OCD symptoms, but not the level of depression, were uniquely
related with scores on the IUS.
Hierarchical regression showed that the contribution to the
explained variance in IU, over and above the variance explained by
the other three symptomlevels, of social anxiety was 4.9% (F change
(1, 109) = 15.75, p < 0.001), of GADwas2.6%(Fchange(1, 109) = 8.47,
p < 0.01), of OCDwas 6.3%(F change (1, 109) = 20.38, p < 0.001), and
of depression was 1.2% (F change (1, 109) = 3.98, p = 0.05)
Table 2
Summary of hierarchical regression analysis predicting Social Phobia Inventory
scores (log)
Variables entered
(in order)
DR
2
DF B S.E. B b
Block 1 (enter) EPQ-N 0.38 75.67 0.02 0.01 0.17
*
Block 2 (stepwise) BFNE (log) 0.15 38.12 0.63 0.18 0.28
**
ASI (log) 0.02 6.19 0.50 0.30 0.15
Block 3 (enter) IUS (log) 0.04 11.71 0.92 0.27 0.31
**
Note: ASI = Anxiety Sensitivity Index. BFNE = Brief Fear of Negative Evaluation Scale.
EPQ-N = Shortened Eysenck Personality Questionnaire, neuroticism subscale.
IUS = Intolerance of Uncertainty Scale. Bs, S.E.s, and bs are from the nal step,
whereas DR
2
and DF values are from each step.
*
p < 0.05.
**
p < 0.01.
Table 3
Summary of hierarchical regression analysis predicting Intolerance of Uncertainty
Scale scores (log)
Variables added B S.E. B b
SPIN 0.10 0.03 0.30
**
GAD-Q-IV 0.01 0.004 0.26
*
OCI-R 0.42 0.09 0.29
**
HADS-D 0.07 0.04 0.15
Note: GAD-Q-IV = Generalized Anxiety Disorder Questionnaire for DSM-IV. HADS-
D = Depression Subscale of Hospital Anxiety and Depression Scale. OCI-R = Revised
Obsessive Compulsive Inventory. SPIN = Social Phobia Inventory.
*
p < 0.01.
**
p < 0.001.
Table 1
Mean scores of variables and spearman correlations among variables
M (S.D.) 1 2 3 4 5 6 7 8 9 10 11 12
1 SFI 11.2 (11.2)
2 BFNE 26.2 (12.7) 0.63
3 ASI 25.4 (9.2) 0.64 0.63
4 SES 19.8 (5.1) 0.46 0.47 0.42
5 MPS CM 18.4 (7.3) 0.57 0.68 0.54 0.43
6 MPS D 8.4 (3.1) 0.47 0.57 0.48 0.54 0.70
7 MPS PC 9.4 (4.3) 0.30
NS
0.33 0.28
NS
0.19
NS
0.55 0.49
8 PSWQ 43.9 (14.2) 0.55 0.58 0.60 0.56 0.45 0.47 0.25
NS
9 GAD-Q-IV 4.9 (3.2) 0.68 0.57 0.68 0.54 0.40 0.39 0.22
NS
0.74
10 OCI-R 9.4 (7.7) 0.46 0.44 0.43 0.40 0.40 0.45 0.17
NS
0.49 0.46
11 HADS-D 4.1 (4.3) 0.48 0.35 0.48 0.59 0.33 0.36 0.15
NS
0.54 0.64 0.38
12 EPQ-N 5.4 (3.7) 0.64 0.48 0.64 0.58 0.42 0.40 0.22
NS
0.75 0.77 0.48 0.64
13 IUS 54.4 (19.9) 0.70 0.59 0.64 0.53 0.53 0.55 0.22
NS
0.75 0.68 0.67 0.58 0.67
Note: All correlations are signicant at p < 0.0006 unless otherwise noted. ASI = Anxiety Sensitivity Index. BFNE = Brief Fear of Negative Evaluation scale. EPQ-N = Shortened
Eysenck Personality Questionnaire, neuroticism subscale. GAD-Q-IV = Generalized Anxiety Disorder Questionnaire for DSM-IV. HADS-D = Depression scale of Hospital
Anxiety and Depression Scale. IUS = Intolerance of Uncertainty Scale. MPS CM = Multidimensional Perfectionism Scale Concern over Mistakes. MPS DA = Multidimensional
Perfectionism Scale Doubts about Actions. MPS PC = Multidimensional Perfectionism Scale Parental Criticism. NS = Not signicant. OCI-R = Revised Obsessive Compulsive
Inventory. PSWQ = Penn State Worry Questionnaire. SES = Rosenberg Self-Esteem Scale. SPIN = Social Phobia Inventory.
P.A. Boelen, A. Reijntjes / Journal of Anxiety Disorders 23 (2009) 130135 133
4. Discussion
This study had two goals. The rst goal was to examine the
association of IU with social anxiety severity, when controlling for
neuroticism and several established correlates of social anxiety
(FNE, anxiety sensitivity, low self-esteem, three aspects of
perfectionism, and pathological worry). The second goal was to
enhance knowledge on the generality vs. specicity of IU by
examining the specicity of the relationship of IU with symptom
levels of GAD, social anxiety, OCD, and depression.
As a rst step in our analyses, zero-order correlations of social
anxiety severity with the other indices examined in this study
were calculated (among other correlations). All these correlations,
except the correlation of social anxiety with Parental Criticism (a
component of perfectionism), were signicant.
1
These ndings are
in agreement with earlier studies showing that social anxiety is
correlated with FNE (Weeks et al., 2005), anxiety sensitivity
(Orsillo et al., 1994), low self-esteem (Kocovski & Endler, 2000),
some components of perfectionism (Juster et al., 1996), patholo-
gical worry (Starcevic et al., 2007), and neuroticism (Brown &
Barlow, 2002).
Next, hierarchical regression predicting social anxiety severity
was carried out to examine how much of the variance in social
anxiety severity was explained by IU, when controlling for
neuroticism (entered in block 1 of the regression equation) and
for the aforementioned seven correlates of social anxiety (sub-
jected to a stepwise procedure in block 2). Outcomes showed that
FNE and anxiety sensitivity were the only two cognitive correlates
that were selected into the equation in block 2, explaining 17% of
the variance in social anxiety over and above the 38.1% explained
by neuroticism. Low self-esteem, different aspects of perfection-
ism, and pathological worry were not selected into the equation.
These ndings link up with earlier studies that have cast doubts on
the importance of perfectionism in social anxiety (Rosser et al.,
2003). In addition, they show that, although people high in social
anxiety also display elevated levels of pathological worry (cf.
Starcevic et al., 2007) and low self-esteem (Kocovski & Endler,
2000), worry and low self-esteem are not specic correlates of
social anxiety, when taking into account other correlates. Most
pertinent to the rst goal of this study, IU, added to the equation in
block 3, explained a signicant additional 4% of the variance in
social anxiety severity, above and beyond neuroticism, FNE, and
anxiety sensitivity. In the nal model, neuroticism, FNE, and IU, but
not anxiety sensitivity explained unique variance in social anxiety.
When comparing the relative contribution of FNE and IU to the
explained variance in social anxiety, independent of neuroticism, it
was found that FNE explained 6.6% of the variance in social anxiety
over and above neuroticism and IU. IU explained 5.4% of the
variance in social anxiety over and above neuroticism and FNE.
Altogether, these ndings replicate earlier studies which have
shown that FNE is critical to social anxiety (Weeks et al., 2005) and
complement earlier ndings in showing that IU possibly plays an
important role in social anxiety as well.
To achieve our second goal, levels of IU were regressed on
symptom levels of social anxiety, GAD, OCD, and depression.
Outcomes showed that, after controlling for the shared variance
among the symptom clusters, GAD, social anxiety, and OCD, but
not depression were uniquely related with IU. When comparing
the relative contribution of symptom clusters to the explained
variance in IU, OCD severity was found to have the strongest link
with IU, followed by social anxiety, GAD, and depression.
Altogether, these ndings link up with earlier ndings that IU is
specic for GAD and OCD (Holaway et al., 2006) but not for
depression (Dugas et al., 2004). Importantly, the present ndings
add to earlier studies in showing that IU is also specically related
to social anxiety.
Several limitations should be kept in mind when interpreting
outcomes of this study. A rst limitation is that the cross-sectional
design precludes any conclusions about causality. Thus, the degree
to which IU is causally related to the development and
maintenance of social anxiety awaits further examination. A
second limitation concerns the composition of the sample. The
sample included self-selected, predominantly relatively highly
educated people with internet access, most of whom were female.
This restricts the generalizability of the results and comparability
with other samples. Moreover, all participants had suffered the
death of a relative. Although one could argue that specic aspects
of the participants mourning may have affected the current
results, this seems unlikely given that losses occurred at least 18
months prior to this study and, more importantly, depression
severity was at a non-clinical level. Nevertheless, conclusions of
this study must remain tentative pending replication with other
samples. A third limitation is that we administered a relatively
large number of questionnaires. This may have lowered the
response rate and may have been tiring to the people who
participated. A fourth limitation is that we only examined the
association of IU with four clinical phenomena (i.e., symptoms of
GAD, social anxiety, OCD, depression). It would be relevant for
future studies to assess a wider range of clinical symptoms in the
context of a single study to further examine the specicity of IU to
different psychological disorders.
Notwithstanding these limitations, the current study adds to
our knowledge of factors contributing to social anxiety severity
and the specicity vs. generality of IU. At the very least, ndings
indicate that a more comprehensive examination of the role of IU
in social anxiety is warranted. Cognitive behavioral models of
social anxiety suggest that safety behaviors and self-focused
attention play a key role in social anxiety (Clark & Wells, 1995;
Rapee & Heimberg, 1997). Among other things, it would be
interesting for future studies to explore how IU works together
with these maladaptive behaviors in maintaining social anxiety.
One might speculate that, in some way, these behaviors serve to
decrease the presumably unbearable feelings of uncertainty in
social-evaluative situations, in the same way that worrying serves
to reduce the uncertainty that GAD patients experience when a
potential future catastrophe comes to mind. With respect to the
generality vs. specicity of IU, several challenging research
questions remain as well. For instance, if future studies conrm
that IUis related to different psychological disorders and not just to
GAD, it becomes important to identify factors that contribute to the
differential development of GAD, OCD, or social anxiety among
those with elevated levels of IU.
If future studies show that IU is causally related to social
anxiety, this would have clinical implications. For instance, it
would suggest that augmenting peoples tolerance for the
uncertainty that is inherent to many social-evaluative situations
is a potentially fruitful intervention in the treatment of social
anxiety. Currently, effective treatments of social anxiety mostly
target self-focused attention, avoidance and safety behaviors, FNE,
and distorted self-imagery, using interventions such as cognitive
1
Although one could argue that it is very logical that FNE and social anxiety
severity are correlated because there is conceptual overlap between both concepts,
it is important to note that several authors have claimed that the concepts are
related, but distinct. For instance, Weeks et al. (2005, p. 179) state that fear of
negative evaluation pertains to the sense of dread associated with being evaluated
unfavorably while anticipating or participating in a social situation, whereas social
anxiety pertains to affective reactions to these situations. Indeed, examination of the
items of the BFNE and SPIN used in this study revealed little item content overlap,
suggesting that the strong correlation between the measures is not just the result of
confounds in content.
P.A. Boelen, A. Reijntjes / Journal of Anxiety Disorders 23 (2009) 130135 134
restructuring, task concentration training, and exposure (Mulkens,
Bo gels, de Jong, & Louwers, 2001), as well as experiential exercises
and video feedback training (Clark et al., 2006). It would be
interesting for future studies to examine the effect of directly
targeting IU on the severity of social anxiety symptoms. The work
of Dugas and coworkers indicate that IUcan be successfully curbed
using specic cognitive behavioral interventions such as problem
orientation training and exposure to uncertainty (e.g., Dugas &
Ladouceur, 2000). Among other things, one might speculate that
repeated imaginal exposure to fear-evoking social scenarios that
include different elements of uncertainty may help to increase
tolerance for the uncertainty that is inherent to many social-
evaluative situations and, along that pathway, reduce social
anxiety. More research on the role of IU in social anxiety is likely
to augment current approaches to the assessment and treatment of
this prevalent, debilitating clinical phenomenon.
References
American Psychiatric Association. (1994). Diagnostic and statistical manual of mental
disorders (4th ed.). Washington, DC: American Psychiatric Association.
Boelen, P. A., & van den Hout, M. A. (2008). The role of threatening misinterpretations
and avoidance in emotional problems after loss. Behavioural and Cognitive Psy-
chotherapy, 36, 7188.
Bo gels, S. M. (2004). De Nederlandstalige Fear of Negative Evaluation Scale [Dutch version
of the Fear of Negative Evaluation Scale]. Maastricht: Vakgroep Differentie le en
Experimentele Psychologie.
Brown, T. A., & Barlow, D. H. (2002). Classication of anxiety and mood disorders. In: D.
H. Barlow (Ed.), Anxiety and its disorders: the nature and treatment of anxiety and
panic (2nd ed., pp. 292327). New York, NY: Guilford Press.
Buhr, K., & Dugas, M. J. (2002). The Intolerance of Uncertainty Scale: psychometric
properties of the English version. Behaviour Research and Therapy, 40, 931945.
Clark, D. M., Ehlers, A., McManus, F., Fennell, M., Grey, N., Waddington, L., et al. (2006).
Cognitive therapy versus exposure and applied relaxation in social phobia: a rando-
mised controlled trial. Journal of Consulting and Clinical Psychology, 74, 568578.
Clark, D. M., & Wells, A. (1995). A cognitive model of social phobia. In: R. G. Heimberg,
M. R. Liebowitz, D. A. Hope, & F. R. Schneier (Eds.), Social phobia: diagnosis,
assessment, and treatment (pp. 6993). New York: Guilford Press.
Connor, K. M., Davidson, J. R. T., Churchill, L. E., Sherwood, A., Foa, E., & Weisler, R. H.
(2000). Psychometric properties of the Social Phobia Inventory (SPIN). British
Journal of Psychiatry, 176, 379386.
Cordova-Middelbrink, J. A., Dek, E. C. P., &Engelbarts, M. M. B. (2007). Automatisering bij
subklinische obsessief-compulsieve stoornis [automatic processes in subclinical obses-
sive compulsive disorder]. Unpublished master thesis. Utrecht: Utrecht University.
De Bruin, G. O., Rassin, E., van der Heiden, C., & Muris, P. (2006). Psychometric
properties of a Dutch version of the Intolerance of Uncertainty Scale. Netherlands
Journal of Psychology, 62, 9197.
Dugas, M. J., & Ladouceur, R. (2000). Treatment of GAD: targeting intolerance of
uncertainty in two types of worry. Behavior Modication, 24, 635657.
Dugas, M. J., Marchand, A., & Ladouceur, R. (2005). Further validation of a cognitive-
behavioral model of generalized anxiety disorder: diagnostic and symptom spe-
cicity. Journal of Anxiety Disorders, 19, 329343.
Dugas, M. J., Schwarzt, A., & Francis, K. (2004). Intolerance of uncertainty, worry, and
depression. Cognitive Therapy and Research, 28, 835842.
Eysenck, S. B. G., Eysenck, H. J., & Barrett, P. (1985). A revised version of the psychoti-
cism scale. Personality and Individual Differences, 6, 2129.
Foa, E. B., Huppert, J. D., Leiberg, S., Langner, R., Kichic, R., Hajcak, G., et al. (2002). The
Obsessive-Compulsive Inventory: development and validation of a short version.
Psychological Assessment, 14, 485496.
Franck, E., De Raedt, R., Barbez, C., & Rosseel, Y. (in press). Psychometric properties of
the Dutch Rosenberg self-esteem scale. Psychologica Belgica.
Freeston, M. H., Rheaume, J., Letarte, H., Dugas, M. J., & Ladouceur, R. (1994). Why do
people worry? Personality and Individual Differences, 17, 791802.
Frost, R. O., Marten, P., Lahart, C., & Rosenblate, R. (1990). The dimensions of perfec-
tionism. Cognitive Therapy and Research, 14, 449468.
Hajcak, G., Huppert, J. D., Simons, R. F., & Foa, E. B. (2004). Psychometric properties of
the OCI-R in a college sample. Behaviour Research and Therapy, 42, 115123.
Hofmann, S. G., & Barlow, D. H. (2002). Social phobia (social anxiety disorder). In: D. H.
Barlow (Ed.), Anxiety and its disorders: the nature and treatment of anxiety and panic
(2nd ed., pp. 454476). New York, NY: Guilford Press.
Holaway, R. M., Heimberg, R. G., & Coles, M. E. (2006). A comparison of intolerance of
uncertainty in analogue obsessive-compulsive disorder and generalized anxiety
disorder. Journal of Anxiety Disorders, 20, 157174.
Juster, H., Heimberg, R., Frost, R., Holt, C., Mattia, J., &Faccenda, K. (1996). Perfectionism
and social phobia. Personality and Individual Differences, 21, 403410.
Kessler, R. C., Berglund, P., Demler, O., Jin, R., &Walters, E. E. (2005). Lifetime prevalence
and age-of-onset distributions of DSM-IV disorders in the National Comorbidity
Survey Replication. Archives of General Psychiatry, 62, 593602.
Kocovski, N. L., & Endler, N. S. (2000). Social anxiety, self-regulation, and fear of
negative evaluation. European Journal of Personality, 14, 347358.
Ladouceur, R., Dugas, M. J., Freeston, M. H., Rheaume, J., Blais, F., Boisvert, J., et al.
(1999). Specicity of generalized anxiety disorder symptoms and processes.
Behaviour Research and Therapy, 30, 191207.
Meyer, T. J., Miller, M. L., Metzger, R. L., & Borkovec, T. D. (1990). Development and
validation of the Penn State Worry Questionnaire. Behaviour Research and Therapy,
28, 487495.
Mineka, S., Watson, D., & Clark, L. A. (1998). Comorbidity of anxiety and unipolar mood
disorders. Annual Review of Psychology, 49, 377412.
Mulkens, S., Bo gels, S. M., de Jong, P. J., & Louwers, J. (2001). Fear of blushing: effects of
task concentration training versus exposure in vivo on fear and physiology. Journal
of Anxiety Disorders, 15, 413432.
Newman, M. G., Zuelling, A. R., Kachin, K. E., Constantino, M. J., Przeworski, A., Erickson,
T., et al. (2002). Preliminary reliability and validity of the GAD-Q-IV: a revised self-
report diagnostic measure of generalized anxiety disorder. Behavior Therapy, 33,
215233.
Orsillo, S. M., Lilienfeld, S. O., & Heimberg, R. G. (1994). Social phobia and response to
challenge procedures: examining the interaction between anxiety sensitivity and
trait anxiety. Journal of Anxiety Disorders, 8, 247258.
Peterson, R. A., & Reiss, S. (1992). Anxiety sensitivity index manual (2nd ed.). Worthing-
ton, OH: International Diagnostic Systems.
Rapee, R. M., &Heimberg, R. G. (1997). Acognitive-behavioral model of anxiety in social
phobia. Behaviour Research and Therapy, 35, 741756.
Rijsoort, S. N., van Emmelkamp, P. M. G., & Vervaeke, G. A. C. (1999). The Penn State
Worry Questionnaire and the Worry Domains Questionnaire: structure, reliability,
and validity. Clinical Psychology and Psychotherapy, 6, 297307.
Rosenberg, M. (1989). Society and the adolescent self-image (Rev. ed.). Middletown, CT:
Wesleyan University Press.
Rosser, S., Issakidis, C., & Peters, L. (2003). Perfectionism in social phobia: relationship
between the constructs and impact on cognitive behaviour therapy. Cognitive
Research and Therapy, 27, 143151.
Sanderman, R., Arrindell, W. A., Ranchor, A. V., Eysenck, H. J., & Eysenck, S. B. G. (1995).
Het meten van persoonlijkheidskenmerken met de Eysenck Personality Questionnaire
(EPQ), een handleiding [measuring personality characteristics using the Eysenck
Personality Questionnaire (EPQ), a manual]. Groningen: Noordelijk Centrum voor
Gezondheidsvraagstukken.
Spinhoven, P., Ormel, J., Sloekers, P. P. A., Kempen, G. I. J. M. , Speckens, A. E. M., &
Hemert van, A. M. (1997). A validation of the Hospital Anxiety and Depression
Scale (HADS) in different groups of Dutch subjects. Psychological Medicine, 27,
363370.
Starcevic, V., & Berle, D. (2006). Cognitive specicity of anxiety disorders: a review of
selected key constructs. Depression and Anxiety, 23, 5161.
Starcevic, V., Berle, D., Milicevic, D., Hannan, A., Lamplugh, C., & Eslick, G. (2007).
Pathological worry, anxiety disorders and the impact of co-occurrence with
depressive and other anxiety disorders. Journal of Anxiety Disorders, 21, 1016
10127.
Steketee, G., Frost, R. O., & Cohen, I. (1998). Beliefs in obsessive compulsive disorder.
Journal of Anxiety Disorders, 12, 525537.
Tolin, D. F., Abramowitz, J. S., Brigidi, B., &Foa, E. B. (2003). Intolerance of uncertainty in
obsessive-compulsive disorder. Journal of Anxiety Disorders, 17, 233242.
Vujanovic, A. A., Arrindell, W. A., Bernstein, A., Norton, P. J., & Zvolensky, M. J. (2007).
Sixteen-item Anxiety Sensitivity Index. Conrmatory factor analytic evidence,
internal consistency, and construct validity in a young adult sample from The
Netherlands. Assessment, 14, 129143.
Weeks, J. W., Heimberg, R. G., Fresco, D. M., Hart, T. A., Turk, C. L., & Schneier, et al.
(2005). Empirical validation and psychometric evaluation of the brief fear of
negative evaluation scale in patients with social anxiety disorders. Psychological
Assessment, 17, 179190.
Zigmond, A. S., & Snaith, R. P. (1983). The hospital anxiety and depression scale. Acta
Psychiatrica Scandinavica, 67, 361370.
P.A. Boelen, A. Reijntjes / Journal of Anxiety Disorders 23 (2009) 130135 135