This article was downloaded by: [McMaster University] On: 21 March 2013, At: 03:29 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK Cognitive Behaviour Therapy Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/sbeh20 The Role of Anger in Generalized Anxiety Disorder Sonya S. Deschênes a , Michel J. Dugas a b , Katie Fracalanza c & Naomi Koerner c a Department of Psychology, Concordia University, 7141 Sherbrooke Street West, Montréal, Québec, Canada, H4B 1R6 b Hôpital du Sacré-Coeur de Montréal, Montréal, Québec, Canada c Ryerson University, Toronto, Ontario, Canada Version of record first published: 19 Mar 2012. To cite this article: Sonya S. Deschênes , Michel J. Dugas , Katie Fracalanza & Naomi Koerner (2012): The Role of Anger in Generalized Anxiety Disorder, Cognitive Behaviour Therapy, 41:3, 261-271 To link to this article: http://dx.doi.org/10.1080/16506073.2012.666564 PLEASE SCROLL DOWN FOR ARTICLE Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. The publisher does not give any warranty express or implied or make any representation that the contents will be complete or accurate or up to date. The accuracy of any instructions, formulae, and drug doses should be independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims, proceedings, demand, or costs or damages whatsoever or howsoever caused arising directly or indirectly in connection with or arising out of the use of this material.
Transcript
This article was downloaded by: [McMaster University]On: 21 March 2013, At: 03:29Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954 Registeredoffice: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Cognitive Behaviour TherapyPublication details, including instructions for authors andsubscription information:http://www.tandfonline.com/loi/sbeh20
The Role of Anger in GeneralizedAnxiety DisorderSonya S. Deschênes a , Michel J. Dugas a b , Katie Fracalanza c &Naomi Koerner ca Department of Psychology, Concordia University, 7141 SherbrookeStreet West, Montréal, Québec, Canada, H4B 1R6b Hôpital du Sacré-Coeur de Montréal, Montréal, Québec, Canadac Ryerson University, Toronto, Ontario, CanadaVersion of record first published: 19 Mar 2012.
To cite this article: Sonya S. Deschênes , Michel J. Dugas , Katie Fracalanza & Naomi Koerner(2012): The Role of Anger in Generalized Anxiety Disorder, Cognitive Behaviour Therapy, 41:3,261-271
To link to this article: http://dx.doi.org/10.1080/16506073.2012.666564
PLEASE SCROLL DOWN FOR ARTICLE
Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions
This article may be used for research, teaching, and private study purposes. Anysubstantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,systematic supply, or distribution in any form to anyone is expressly forbidden.
The publisher does not give any warranty express or implied or make any representationthat the contents will be complete or accurate or up to date. The accuracy of anyinstructions, formulae, and drug doses should be independently verified with primarysources. The publisher shall not be liable for any loss, actions, claims, proceedings,demand, or costs or damages whatsoever or howsoever caused arising directly orindirectly in connection with or arising out of the use of this material.
1Department of Psychology, Concordia University, 7141 Sherbrooke Street West, Montreal,Quebec, Canada H4B 1R6; 2Hopital du Sacre-Coeur de Montreal, Montreal, Quebec,
Abstract. Little is known about the role of anger in the context of anxiety disorders, particularly withgeneralized anxiety disorder (GAD). The aim of study was to examine the relationship betweenspecific dimensions of anger and GAD. Participants (N ¼ 381) completed a series of questionnaires,including the Generalized Anxiety Disorder Questionnaire (GAD-Q-IV; Newman et al., 2002,Behavior Therapy, 33, 215–233), the State-Trait Anger Expression Inventory (STAXI-2; Spielberger1999, State-Trait Anger Expression Inventory-2: STAXI-2 professional manual, Odessa, FL:Psychological Assessment Resources) and the Aggression Questionnaire (AQ; Buss & Perry 1992,Journal of Personality and Social Psychology, 63, 452–459). The GAD-Q-IV identifies individualswho meet diagnostic criteria for GAD (i.e. GAD analogues) and those who do not (non-GAD). TheSTAXI-2 includes subscales for trait anger, externalized anger expression, internalized angerexpression, externalized anger control and internalized anger control. The AQ includes subscales forphysical aggression, verbal aggression, anger and hostility. The GAD-Q-IV significantly correlatedwith all STAXI-2 and AQ subscales (r’s ranging from .10 to .46). Multivariate analyses of variancerevealed that GAD analogues significantly differed from non-GAD participants on the combinedSTAXI-2 subscales (h 2 ¼ .098); high levels of trait anger and internalized anger expressioncontributed the most to GAD group membership. GAD analogue participants also significantlydiffered from non-GAD participants on the combined AQ subscales (h 2 ¼ .156); high levels of anger(affective component of aggression) and hostility contributed the most to GADgroup membership. Within the GAD analogue group, the STAXI-2 and AQ subscales significantlypredicted GAD symptom severity (R 2 ¼ .124 and .198, respectively). Elevated levels of multipledimensions of anger characterize individuals who meet diagnostic criteria for GAD. Key words:anxiety; generalized anxiety disorder; anger; hostility; aggression
Received 22 July, 2011; Accepted 7 February, 2012
Correspondence address: Michel J. Dugas, PhD, Department of Psychology, Concordia University,7141 Sherbrooke Street West, Montreal, Quebec, Canada, H4B 1R6. Tel: þ 1 514 848 2424. Ext. 2215,Fax: þ 1 514 848 4537. E-mail: [email protected]
Anger and anxiety have historically beenlinked through shared physiological reactionsto stress (e.g. Cannon, 1929). In particular,anger and anxiety may be related throughunderlying biological vulnerabilities, such thatwhen confronted with threat, individuals reacteither with anger or anxiety, i.e. “fight” or“flight” (Barlow 2002). According to thistheory, the individual’s sense of mastery overthe threatening situation predicts the type ofreaction with high perceived mastery predict-ing anger and low perceived mastery predict-
ing anxiety. This suggests that appraisals arean important feature of both emotions.Despite this, anger, defined as an emotionelicited by perceptions of threat caused by themisdeeds of others (DiGiuseppe & Tafrate2007), has received little empirical attention inthe context of anxiety disorders.
However, there is some evidence to suggestthat elevated levels of anger are present inindividuals with anxiety disorders. In additionto trait and state anger, some dimensions ofanger that have been examined include
q 2012 Swedish Association for Behaviour Therapy ISSN 1650-6073http://dx.doi.org/10.1080/16506073.2012.666564
Cognitive Behaviour Therapy Vol 41, No 3, pp. 261–271, 2012
hostility (the cognitive component of anger),aggression (the behavioural component ofanger), internalized anger expression (thetendency to suppress angry feelings), externa-lized anger expression (the tendency to out-wardly express angry feelings) and angercontrol (the ability to regulate anger). Specifi-cally, Moscovitch, McCabe, Antony, Rocca,and Swinson (2008) found elevated levels ofhostility in individuals with social anxietydisorder, obsessive-compulsive disorder andpanic disorder, relative to non-anxious con-trols. They also found that individuals withpanic disorder reported higher levels ofaggressive anger and that individuals withsocial phobia reported lower verbal aggressioncompared to non-anxious controls. Erwin,Heimberg, Schneier, and Liebowitz (2003)also found elevated trait anger and interna-lized anger expression in individuals withsocial anxiety disorder, relative to non-anxious individuals. Of the anxiety disorders,anger has mostly been examined in relation topost-traumatic stress disorder (PTSD; Novaco2010). For example, Meffert and colleagues(2008) found that greater levels of angermediated the relationship between traumaexposure and PTSD symptoms among policeofficers. In addition, meta-analytical findingssuggest that PTSD symptoms are related tovarious dimensions of anger, particularlyinternalized anger expression, with largeeffects (Olatunji, Ciesielski, & Tolin, 2010;Orth & Wieland 2006). However, a recentstudy showed that, after controlling fordemographic variables, PTSD did not signifi-cantly predict anger expression, but didsignificantly predict anger experience over a30-day period (Hawkins & Cougle 2011).Only a few studies have examined the
relationship between anger and generalizedanxiety disorder (GAD). This is surprisingbecause irritability, which is characterized by alowered threshold for anger (DiGiuseppe &Tafrate 2007), is a symptom of GAD (APA,2000). Erdem, Celik, Yetkin, and Ozgen(2008) found that individuals with GAD hadgreater levels of trait anger, externalized angerexpression, internalized anger expression, aswell as lower anger control (i.e. lower self-regulation of anger), than did non-anxiousindividuals. More recently, Hawkins andCougle (2011) found that greater angerexpression, as well as anger experience over a
30-day period, was associated with GADindependently of shared associations withother psychiatric conditions. Overall, thesefindings suggest that anger may be animportant emotion associated with GAD.Although the above-mentioned studies exam-ined the relationship between specific angerdimensions (e.g. trait anger, anger expression)and GAD, the relative contributions of eachanger dimension to GAD is largely unknown.The goal of this study was to examine the
relationship between specific dimensions ofanger, and the presence and severity of GADby (1) comparing individuals who meetdiagnostic criteria for GAD to individualswho do not meet diagnostic criteria for GADon a combination of anger dimensions; (2)exploring the relative contribution of eachanger dimension to GAD status and (3)examining the extent to which anger dimen-sions predict GAD symptom severity inindividuals who meet diagnostic criteria forGAD. The anger dimensions examined werebased on the subscales of the questionnairesused in this study (see below).It was hypothesized that individuals who
meet diagnostic criteria for GAD would differfrom individuals who do not meet diagnosticcriteria for GAD on a combination of angerdimensions, and that lower scores on externa-lized and internalized anger control and higherscores on all other anger dimensions wouldpredict greater GAD symptom severity.Although the examination of the relativecontribution of anger dimensions to GADstatus was largely exploratory, we expectedthat trait anger and internalized anger (i.e.inwards anger expression, hostility) wouldcontribute more to GAD than would externa-lized anger (i.e. outwards anger expression,physical aggression and verbal aggression).This hypothesis was derived from the evidencesuggesting that internalized anger (e.g. Ola-tunji et al., 2010) and hostility (e.g. Moscov-itch et al., 2008) are strong predictors of otheranxiety disorders.
Method
ParticipantsThree hundred and eighty-one (N ¼ 381)undergraduate students, between the ages of18 and 57 (M ¼ 23.49, SD ¼ 6.27), partici-pated in the study in exchange for course
262 Deschenes, Dugas, Fracalanza and Koerner COGNITIVE BEHAVIOUR THERAPY
Dow
nloa
ded
by [
McM
aste
r U
nive
rsity
] at
03:
29 2
1 M
arch
201
3
credit. The majority of the sample was female(85.79%) and studying in the field ofpsychology (71.39%). Most participants(38.10%) were in their first year of study,25.93% were in their second year, 20.63%were in their third year and 15.34% were intheir fourth year, with 87.73% of the samplestudying full-time. The majority of the sample(63.47%) reported English as their firstlanguage, 14.67% reported French and21.87% reported “other” as their firstlanguage. The majority of participantsreported their ethnicity as Caucasian(65.00%), 7.10% as Asian, 6.84% as Multi-Racial, 5.53% as Black, 5.00% as Latino and5.00% as Middle Eastern, whereas 5.53%reported “other” as their ethnicity.
MeasuresGeneralized Anxiety Disorder Questionnaire.The Generalized Anxiety Disorder Question-naire (GAD-Q-IV; Newman et al., 2002) wasdeveloped as a screening tool for the diagnosisof GAD. It is composed of 14 self-reporteditems that assess the symptoms of GAD basedon the DSM-IV; APA, 2000 diagnosticcriteria. Eleven of the items are rateddichotomously (i.e. the presence or theabsence of symptoms), one item requiresparticipants to list worry topics, and twoitems assess the degree of interference and thedegree of distress resulting from worrying on aLikert scale ranging from 0 (None) to 8 (Verysevere). The GAD-Q-IV demonstrates con-vergent and discriminant validity, a kappaagreement of .67 with a structured diagnosticinterview of GAD and test–retest reliability of92% over 2 weeks. The recommended cut-offscore for individuals meeting diagnosticcriteria is 5.7 out of a total of 13 (Newmanet al., 2002), with scores of 5.7 and aboveindicating the presence of GAD (i.e. GADanalogues) and a score below 5.7 indicatingthe absence of GAD (i.e., non-GAD).Penn State Worry Questionnaire. The PennState Worry Questionnaire (PSWQ; Meyer,Miller, Metzger, & Borkovec, 1990) is a 16-item self-report measure that assesses thetendency to worry. Participants rate the extentto which items are typical of themselves on aLikert scale ranging from 1 (Not at all typical)to 5 (Very typical), with items such as “I amalways worrying about something.” ThePSWQ has demonstrated convergent and
divergent validity, excellent internal consist-ency (a ¼ .93) and test–retest reliability over8–10 weeks (r ¼ .92).State-Trait Anger Expression Inventory-2. TheState-Trait Anger Expression Inventory-2(STAXI-2; Spielberger 1999) is a 57-itemself-report measure with scales developed toassess anger as a dispositional characteristic(trait anger scale), situational anger (stateanger scale) and the expression of anger (angerexpression scale). Given the goals of thisstudy, only the trait anger and angerexpression scales were included. The 10-itemTrait Anger scale (T-ANG) assesses thefrequency and intensity of anger experiences,with items such as “I get angry when I’mslowed down by other’s mistakes” rated on aLikert scale ranging from 1 (Almost never) to 4(Almost always). The 32-item AngerExpression scale is composed of four subscalesthat assess how people react when they areangry. For each eight-item subscale, the extentto which participants generally react whenangry is rated on a Likert scale ranging from 1(Almost never) to 4 (Almost always). TheAnger Expression-Out (AX-O) subscalemeasures the expression of anger towardsobjects or other individuals with the use ofphysically or verbally aggressive behavioursand includes items such as “I strike out atwhatever infuriates me.” The AngerExpression-In (AX-I) subscale measures theextent to which angry feelings are experiencedyet suppressed (i.e. lack of expression) andincludes items such as “I boil inside, but I don’tshow it.” The Anger Control-Out (AC-O)subscale assesses the extent to which a personcontrols his or her anger by preventing theexternalized expression of anger, and includesitems such as “I keep my cool.” Finally, theAnger Control-In (AC-I) subscale assesses theextent to which a person controls angryfeelings by attempts to calm down and cooloff, and includes items such as “I dosomething relaxing to calm down.” TheSTAXI-2 subscales have demonstrated con-struct validity and adequate internal consist-ency with a’s ranging from .70 to .85.Aggression Questionnaire. The AggressionQuestionnaire (AQ; Buss & Perry 1992) is a29-item self-report measure that assesses thedisposition of aggression, and is composed offour subscales. For each subscale, the extent towhich each statement is characteristic or
VOL 41, NO 3, 2012 Anger in Generalized Anxiety Disorder 263
Dow
nloa
ded
by [
McM
aste
r U
nive
rsity
] at
03:
29 2
1 M
arch
201
3
uncharacteristic of participants is rated on aLikert scale ranging from 1 (Extremelyuncharacteristic of me) to 5 (Extremelycharacteristic of me). The Physical Aggression(AQ-PA) subscale is composed of eight itemsand assesses the motor component of aggres-sive behaviour, which involves the desire toharm others, with items such as “Once in awhile, I can’t control the urge to strike anotherperson.” The Verbal Aggression (AQ-VA)subscale is composed of five items and assessesinstrumental aggression with items such as“When people annoy me, I may tell them whatI think of them.” The Anger (AQ-ANG)subscale is composed of seven items andassesses the affective component of aggres-sion, including physiological arousal andpreparation for aggression, with items suchas “I have trouble controlling my temper.”Finally, the Hostility (AQ-HOST) subscale iscomposed of eight items and measures thecognitive component of aggression, includingfeelings of injustice, with items such as “Whenpeople are especially nice, I wonder what theywant.” The AQ has good internal consistency(a ¼ .89) and test–retest reliability over 9weeks (r ¼ .80).
ProcedureParticipants were recruited in psychologyundergraduate classes or through the Depart-ment of Psychology’s Participant Pool atConcordia University. They were invited tocomplete a series of questionnaires on angerand anxiety, administered in a quasi-counter-balanced order, either individually or ingroups of up to 10 participants. The exper-imenter (the first author) obtained informedconsent and debriefed all participants.
Results
Preliminary analysesData were normally distributed (all skewnessvalues ,3.0 and kurtosis values ,10.0; Kline2009), and therefore no outliers were removed.The correlations between the GAD-Q-IV andall STAXI-2 and AQ subscales were statisti-cally significant (r’s ranging from .10 to .46,p’s , .05). Male sex was significantly corre-lated with greater externalized anger controland greater physical aggression, whereasfemale sex was significantly correlated withhigher scores on the GAD-Q-IV. Age was
negatively related to hostility. Because thestrengths of the correlations were weak(r’s , .17), we did not statistically control forage and sex in subsequent analyses. See Table 1for a correlation matrix.Next, we used the recommended cut-off
score (5.7; Newman et al., 2002) on the GAD-Q-IV to create the GAD analogue (n ¼ 131)and non-GAD (n ¼ 250) groups. Given thatworry is the primary feature of GAD, weexamined the validity of the GAD-Q-IV in oursample by conducting an independent samplet-test between the groups on PSWQ scores.We found that, as expected, the GADanalogue group had significantly higher scores(M ¼ 63.00, SD ¼ 10.64) than did the non-GAD group (M ¼ 46.00, SD ¼ 12.44),t(378) ¼ 213.26, p , .001. These means andstandard deviations are comparable to thoseof clinical samples of individuals with GAD(e.g. M ¼ 65.27, SD ¼ 8.50; Ladouceur et al.,2000) and samples of non-anxious individuals(e.g. M ¼ 47.08, SD ¼ 13.24; Behar, Alcaine,Zuellig, & Borkovec, 2003).
Anger and GAD group membershipTo examine the difference between the GADanalogue group and the non-GAD group onthe STAXI-2 subscales, a multivariate analysisof variance (MANOVA) was conducted.GAD group membership served as theindependent variable, and the STAXI-2 sub-scales served as the dependent variables. Asexpected, there was a statistically significantdifference between the GAD analoguegroup and the non-GAD group on thecombined STAXI-2 anger subscales, L ¼ .90,F(5, 374) ¼ 8.09, p , .001, h 2 ¼ .098. SeeTable 2 for means and standard deviations ofthe STAXI-2 subscales by GAD group -membership.A discriminant function analysis was con-
ducted to examine the relative contribution ofeach STAXI-2 subscale to GADgroup membership. The correlations betweenthe predictors and the discriminant function(i.e. the structure matrix) suggest that elevatedT-ANG and AX-I (Anger Expression-In)accounted for the most variance in GADgroup membership (loadings less than .50 arenot interpreted; Tabachnick & Fidell 2007).See Table 3 for canonical coefficients and thestructure matrix. Using Jackknife classifi-cation, a method used to classify each case
264 Deschenes, Dugas, Fracalanza and Koerner COGNITIVE BEHAVIOUR THERAPY
VOL 41, NO 3, 2012 Anger in Generalized Anxiety Disorder 265
Dow
nloa
ded
by [
McM
aste
r U
nive
rsity
] at
03:
29 2
1 M
arch
201
3
by the functions derived from all other cases,the discriminant function could be used tocorrectly classify 63.95% (n ¼ 243) of individ-uals into their respective groups, with 57.25%(n ¼ 75) correctly classified as GAD analogueand 67.47% (n ¼ 168) correctly classified as
non-GAD. The difference in the proportionsof correct classification was significant,x 2 ¼ 3.89, p ¼ .049, suggesting that theSTAXI-2 subscales can better identify indi-viduals who do not meet diagnostic criteria forGAD than those who do.To assess the effect of anger on GAD
symptom severity, a multiple regressionanalysis predicting GAD-Q-IV continuousscores was conducted within the GADanalogue group (n ¼ 131), with the STAXI-2subscales entered as predictors. As expected,the regression model was statistically signifi-cant, F(5, 125) ¼ 3.54, R 2 ¼ .124, p ¼ .005.Of the predictor variables, only AX-I signifi-cantly predicted GAD symptom severity(b ¼ .22, p ¼ .017). See Table 4 for detailedresults of the multiple regression.
Aggression andGADgroupmembershipTo examine the difference between the GADanalogue group and the non-GAD group onthe AQ subscales, a MANOVA was con-ducted. GAD group membership served as theindependent variable, whereas AQ subscalesserved as the dependent variables. Asexpected, there was a statistically significantdifference between the GAD analoguegroup and the non-GAD group on thecombined AQ subscales, L ¼ .84, F(4,376) ¼ 17.34, p , .001, h 2 ¼ .156. See Table5 for means and standard deviations of AQsubscales by GAD group membership.A discriminant function analysis was con-
ducted to examine the relative contribution ofeach AQ subscale to GAD group membership.The correlations between the predictors andthe discriminant function suggest that elevatedAQ-HOST and AQ-ANG accounted for themost variance in GAD group membership.See Table 6 for canonical coefficients and thestructure matrix. Using Jackknife classifi-cation, the discriminant function could beused to correctly classify 66.93% (n ¼ 255) ofindividuals into their respective groups, with64.89% (n ¼ 85) correctly classified as GADanalogue and 68.00% (n ¼ 170) correctlyclassified as non-GAD. The difference in theproportions of correct classification was notstatistically significant, x 2 ¼ .38, p ¼ .54.To assess the association of aggression to
GAD symptom severity, a multiple regressionanalysis with AQ subscales predicting GAD-Q-IV continuous scores was conducted within
Table 2. Means and standard deviations for theSTAXI-2 by GAD group membership
GAD
analogue
(n ¼ 131)
Non-GAD
(n ¼ 250)
STAXI-2 subscales Mean SD Mean SD
T-ANGa 21.50b 6.19 18.52c 4.73
AX–O 16.35b 4.91 14.74c 3.59
AX–I 19.42b 4.62 16.86c 4.83
AC–O 22.18b 5.22 23.39c 4.51
AC–I 20.86b 5.06 22.38c 4.71
Notes. Means with different (b,c) superscripts aresignificantly different (p , .05). STAXI-2, State-Trait Anger Expression Inventory, second edition;GAD, generalized anxiety disorder; T-ANG, State-Trait Anger Expression Inventory II–Trait Scale;AX–O, State-Trait Anger Expression Inventory II–Anger Expression-Out subscale; AX-I, State-TraitAnger Expression Inventory II–Anger Expression-Insubscale; AC-O, State-Trait Anger ExpressionInventory II–Anger Control-Out subscale; AC-I,State-Trait Anger Expression Inventory II–AngerControl-In subscale.a Data missing for one participant (non-GAD:n ¼ 249).
Table 3. Standardized canonical coefficients andstructure matrix for the STAXI-2 predicting GADgroup status (N ¼ 380a)
266 Deschenes, Dugas, Fracalanza and Koerner COGNITIVE BEHAVIOUR THERAPY
Dow
nloa
ded
by [
McM
aste
r U
nive
rsity
] at
03:
29 2
1 M
arch
201
3
the GAD analogue group. As expected, theregression model was statistically significant,F(4, 126) ¼ 7.80, R 2 ¼ .198, p , .001. Of thepredictor variables, only AQ-HOST signifi-cantly predicted GAD symptom severity(b ¼ .39, p , .001). See Table 7 for detailedresults of the multiple regression.
Discussion
The aim of this study was to examine therelationship between specific dimensions ofanger, and the presence and severity of GAD.Overall, our results suggest that heightenedlevels of anger, in particular trait anger,internalized anger expression, anger as the
affective component of aggression and hosti-lity are uniquely related to GAD status. Ourresults also suggest that, when controlling forshared variance between the subscales, onlyinternalized anger expression from the STAXIand hostility from the AQ uniquely contributeto the severity of GAD symptoms withinindividuals who meet diagnostic criteria.These findings are broadly consistent withour hypotheses.
The current findings are also in keeping withprevious research on anger and anxietydisorders demonstrating that elevated angerlevels, particularly internalized angerexpression (e.g. Bridewell & Chang 1997;Orth & Wieland 2006) and hostility (e.g.Moscovitch et al., 2008), are present inanxious individuals. Also in accordance with
Table 4.Multiple regression for the STAXI-2 predicting GAD symptom severity in GAD analogues (n ¼ 131)
[Lower, upper]STAXI-2 subscales R 2 B SE b 95% confidence interval for B
VOL 41, NO 3, 2012 Anger in Generalized Anxiety Disorder 267
Dow
nloa
ded
by [
McM
aste
r U
nive
rsity
] at
03:
29 2
1 M
arch
201
3
our findings, Erdem and colleagues (2008)found that individuals with GAD haveelevated levels of trait anger and angerexpression. Although Hawkins and Cougle(2011) showed that a diagnosis of GAD wasrelated to elevated anger experience and agreater tendency to express anger externally,they did not assess the tendency to expressanger internally. Our results suggest that whenthe shared variance between internal andexternal anger expression is controlled, inter-nalized anger expression is a stronger pre-dictor of GAD.Although our results do not address the
question of why anger and GAD tend to co-occur, one possibility is that they arefunctionally related due to shared informationprocessing biases. For example, Barrazoneand Davey (2009) found that both angry andanxious mood inductions led to increasedthreat interpretations of ambiguous homo-phones (e.g. slay/sleigh). Relatedly, Owen(2011) concluded based on a review of thepublished literature that high trait anger ischaracterized by similar trans-diagnostic cog-nitive processes (e.g. selective attention) asother emotional disorders such as depressionand anxiety disorders. In addition, anger andGAD may share underlying cognitive vulner-abilities such as intolerance of uncertainty.Intolerance of uncertainty arises from a set ofnegative beliefs, including the belief thatuncertainty is unfair (Sexton & Dugas 2009).Similarly, anger has been associated withperceived unfairness (e.g. Barclay et al., 2005).One possibility is that perceiving a state ofuncertainty as unfair can lead to anger,anxiety, or both, in individuals who do not
cope well with uncertainty. Future studiesshould aim to examine the role of intoleranceof uncertainty in anger. Overall, it seemspossible that similar cognitive processescontribute to both anger and anxiety. Anotherpossibility relates to a model of GAD thatposits that the heightened intensity of manyemotions contribute to GAD (Mennin, Heim-berg, Turk, & Fresco, 2005). Thus, individualswith GAD may find anger and other emotionsoverwhelming, and these individuals maytherefore worry about the consequences oflosing control over their anger.Anger may be particularly important to
examine in the context of anxiety disorders asit can interfere with cognitive-behaviouraltreatment (CBT). For instance, one studyfound that pre-treatment anger predictedpoorer response to CBT for individuals withsocial anxiety disorder (Erwin et al., 2003).Although the mechanisms by which angerleads to poor CBT responses are unknown,one possibility is that anger interferes withcommon therapy factors in the treatment ofanxiety disorders. For example, anger mayinterfere with the development of a strongtherapeutic alliance, as suggested by DiGiu-seppe, Tafrate, and Eckhardt (1994). Inaddition, anger may lead to lower motivationin treatment or resistance to change, or a lesscollaborative approach to goal setting, all ofwhich are known to affect treatment response(Hubble, Duncan, & Miller, 2004).It is currently unknown whether anger leads
to poor responses in the CBT of GAD.We canpostulate, however, that anger may interferewith some components of empirically sup-ported CBT protocols for GAD. For example,
Table 7. Multiple regression for the AQ predicting GAD symptom severity in GAD analogues (n ¼ 131)
[Lower, Upper]AQ subscales R 2 B SE b 95% confidence interval for B
268 Deschenes, Dugas, Fracalanza and Koerner COGNITIVE BEHAVIOUR THERAPY
Dow
nloa
ded
by [
McM
aste
r U
nive
rsity
] at
03:
29 2
1 M
arch
201
3
Roemer and Orsillo (2007) developed atreatment protocol that targets experientialavoidance, which is characterized by attemptsto reduce the intensity and frequency ofnegative internal experiences. Anger mayinterfere with clients’ ability to focus aware-ness on the present moment and acceptinternal experiences. Another empiricallysupported CBT protocol for GAD includesproblem-solving training as a component oftreatment (Dugas & Robichaud 2007). Giventhat high levels of anger and hostility havebeen found to predict poor social problem-solving skills (D’Zurilla, Chang, & Sanna,2003), individuals with GAD who haveelevated anger may be faced with particularchallenges when attempting to solve their day-to-day problems. The effect of anger onspecific components of treatment, however,requires further exploration.
The finding that scores on measures ofanger and aggression correctly classifiedindividuals meeting diagnostic criteria forGAD at a greater than chance level (57.3%and 64.9%, respectively) is noteworthy. Thesefindings suggest that it may be valuable forclinicians to enquire about anger difficulties inclients with GAD to obtain a more completeunderstanding of potential emotional pro-blems, particularly given that difficulties withanger management are not screened for incommon diagnostic assessments, with theexception of borderline personality disorder(e.g. The Structured Clinical Interview forDSM-IV Axis-II Disorders; First, Spitzer,Gibbon, & Williams, 1996).
LimitationsA possible caveat to empirically investigatinganger is the lack of a consistent definition ofanger and its related constructs (Eckhardt,Norlander, & Deffenbacher, 2004). There iscurrently little agreement on definitions for thedimensions of anger, and this likely affects thedevelopment of self-report anger assessments.Thus, the reliance on such self-reportmeasures in this study is a limitation. Futurestudies could improve on this by using multi-method assessments of anger. In addition, ourstudy is limited by the use of an analoguesample of GAD composed of universitystudents enrolled in at least one psychologycourse. Although analogue samples have beenshown to be similar to clinical samples of
individuals with GAD on measures of worryand anxiety (Roemer, Borkovec, Posa, &Borkovec, 1995), we cannot be certain that theanger levels reported by our GAD analoguegroup would be comparable to individualswith GAD who were recruited from a clinicalsetting.
Arguably, another limitation is that ourstatistical analyses did not control fordepression. Elevated anger levels have beenfound in individuals with major depression(e.g. Riley, Treiber, & Woods, 1989), andGAD and major depression are highlycomorbid (e.g. Brown, Campbell, Lehman,Grisham, & Mancill, 2001). It is thereforepossible that our results were in part due toshared variance between anger anddepression. However, the decision to excludedepression as a covariate was made to increasethe ecological validity of our results. Specifi-cally, there are a number of symptoms ofGAD and depression that overlap, such asdifficulty concentrating, fatigue, andsleep disturbance (APA, 2000), and thesecriteria were included in our measure of GAD.In addition, negative affect is common to bothanxiety and depression, as suggested by thetripartite model of depression and anxiety(Clark & Watson 1991). Relatedly, depressivesymptoms are important features of theclinical presentation of GAD, and controllingfor these would “exclude” a number ofsymptoms that make up the diagnostic criteriafor GAD, thereby limiting the generalizabilityof our results. Furthermore, Miller andChapman (2001) suggested that statistically“removing” shared variance between twoconceptually similar constructs (e.g. anxietyand depression) leads to poor constructvalidity of the target construct. In summary,we chose not to control for depression, giventhe overlapping nature of GAD anddepression.
Conclusions
The potential link between anger and GAD incognitive-behavioural contexts has not beengiven much attention. This is reflected in thescarce literature on anger and GAD, and thelack of recommendations for addressinganger-related symptoms in evidence-basedtreatments for GAD. The current findingshighlight the importance of examining the
VOL 41, NO 3, 2012 Anger in Generalized Anxiety Disorder 269
Dow
nloa
ded
by [
McM
aste
r U
nive
rsity
] at
03:
29 2
1 M
arch
201
3
co-occurrence of anger andGAD. Overall, ourresults suggest that multiple facets of anger arerelated to GAD symptoms; although furtherresearch is needed to identify the mechanismsby which high trait anger, internalized angerexpression and hostility are related to GAD.
Acknowledgements
This study was supported by the Fonds de laRecherche en Sante du Quebec (FRSQ). Wethank Stephanie Correia, Stella Dentakos andAndrea Tkalec for their assistance with testingand data entry.
ReferencesAmerican Psychiatric Association (APA) (2000).
Diagnostic and statistical manual of mentaldisorders: Text Revision (4 ed. APA) (DSM-IV-TR). Washington, DC: Author.
Barazzone, N., & Davey, G. L. (2009). Angerpotentiates the reporting of threateninginterpretations: An experimental study. Journalof Anxiety Disorders, 23, 489–495,doi:10.1016/j.janxdis.2008.10.007.
Barclay, L. J., Skarlicki, D. P., & Pugh, S. D.(2005). Exploring the role of emotions ininjustice perceptions and retaliation. Journal ofApplied Psychology, 90, 629–643,doi:10.1037/0021-9010.90.4.629.
Barlow, D. H. (2002). Anxiety and its disorders: Thenature and treatment of anxiety and panic (2nded.). New York, NY: Guilford Press.
Behar, E., Alcaine, O., Zuellig, A. R., & Borkovec,T. D. (2003). Screening for generalized anxietydisorder using the Penn State Worry Ques-tionnaire: A receiver operating characteristicanalysis. Journal of Behavior Therapy andExperimental Psychiatry, 34, 25–43,doi:10.1016/S0005-7916(03)00004-1.
Bridewell, W. B., & Chang, E. C. (1997).Distinguishing between anxiety, depression,and hostility: Relations to anger-in, anger-out,and anger control. Personality and IndividualDifferences, 22(4), 587–590.
Brown, T. A., Campbell, L. A., Lehman, C. L.,Grisham, J. R., & Mancill, R. B. (2001).Current and lifetime comorbidity of the DSM-IV anxiety and mood disorders in a large clinicalsample. Journal of Abnormal Psychology, 110,585–599, doi:10.1037//0021-843X.110.4.585.
Buss, A., & Perry, M. (1992). The AggressionQuestionnaire. Journal of Personality and SocialPsychology, 63, 452–459, doi:10.1037/0022-3514.63.3.452.
Cannon, W. B. (1929). Bodily changes in pain,hunger, fear and rage (2nd ed.). New York:Appleton-Century-Crofts.
Clark, L. A., &Watson, D. (1991). Tripartite modelof anxiety and depression: Psychometric evi-dence and taxonomic implications. Journal of
D’Zurilla, T., Chang, E. C., & Sanna, L. J. (2003).Self-esteem and social problem solving aspredictors of aggression in college students.doi:10.1521/jscp.22.4.424.22897 Journal ofSocial and Clinical Psychology, 22, 424–440.
DiGiuseppe, R., & Tafrate, R. (2007). Under-standing anger disorders. New York, NY:Oxford University Press.
DiGiuseppe, R., Tafrate, R., & Eckhardt, C. (1994).Critical issues in the treatment of anger.Cognitive and Behavioral Practice, 1, 111–132,doi:10.1016/S1077-7229(05)80089-6.
Dugas, M. J., & Robichaud, M. (2007). Cognitive-behavioral treatment for generalized anxietydisorder: From science to practice. New York:Routledge/Taylor & Francis Group.
Eckhardt, C., Norlander, B., & Deffenbacher, J.(2004). The assessment of anger and hostility: Acritical review. Aggression and Violent Behavior,9, 17–43, doi:10.1016/S1359-1789(02)00116-7.
Erwin, B. A., Heimberg, R. G., Schneier, F. R., &Liebowitz, M. R. (2003). Anger experience andexpression in social anxiety disorder: Pretreat-ment profile and predictors of attrition andresponse to cognitive-behavioral treatment.Behavior Therapy, 34(3), 331–350.
First, M. B., Spitzer, R. L., Gibbon, M., &Williams, J. B. W. (1996). Structured ClinicalInterview for DSM-IV Axis I Disorders –Patient edition. SCID-I/P, Version 2.0.New York: Psychiatric Institute, BiometricsResearch Department.
Hawkins, K. A., & Cougle, J. R. (2011). Angerproblems across the anxiety disorders: Findingsfrom a population-based study. Depression andAnxiety, 28, 145–152, doi: 10.1002/da.20764.
Hubble, M. A., Duncan, B. L., & Miller, S. D.(2004). The heart and soul of change: Whatworks in therapy. Washington, DC: AmericanPsychological Association.
Kline, R. B. (2009). Becoming a behavioural scienceresearcher: A guide to producing research thatmatters. New York: The Guilford Press.
Ladouceur, R., Dugas, M. J., Freeston, M. H.,Leger, E., Gagnon, F., & Thibodeau, N. (2000).Efficacy of a cognitive-behavioral treatment forgeneralized anxiety disorder: Evaluation in acontrolled clinical trial. Journal of Consultingand Clinical Psychology, 68(6), 957–964.
Meffert, S., Metzler, T., Henn-Haase, C., McCa-slin, S., Inslicht, S., Chemtob, C., & Marmar,C. R. (2008). A prospective study of trait angerand PTSD symptoms in police. Journal ofTraumatic Stress, 21, 410–416, doi:10.1002/jts.2035.
Mennin, D. S., Heimberg, R. G., Turk, C. L., &Fresco, D. M. (2005). Preliminary evidence foran emotion regulation deficit model of general-ized anxiety disorder. Behaviour Research and
270 Deschenes, Dugas, Fracalanza and Koerner COGNITIVE BEHAVIOUR THERAPY
Meyer, T. J., Miller, M. L., Metzger, R. L., &Borkovec, T. D. (1990). Development andvalidation of the Penn State Worry Question-naire. Behaviour Research and Therapy, 28,487–495, doi:10.1016/0005-7967(90)90135-6.
Miller, G. A., & Chapman, J. P. (2001). Misunder-standing analysis of covariance. Journal ofAbnormal Psychology, 110, 40–48,doi:10.1037/0021-843X.110.1.40.
Moscovitch, D., McCabe, R., Antony, M., Rocca,L., & Swinson, R. (2008). Anger experience andexpression across the anxiety disorders.Depression and Anxiety, 25, 107–113,doi:10.1002/da.20280.
Newman,M., Zuellig, A., Kachin, K., Constantino,M., Przeworski, A., Erickson, T., & Cashman-McGrath, L. (2002). Preliminary reliability andvalidity of the Generalized Anxiety DisorderQuestionnaire-IV: A revised self-report diag-nostic measure of generalized anxiety disorder.Behavior Therapy, 33, 215–233, doi:10.1016/S0005-7894(02)80026-0.
Novaco, R. W. (2010). Anger and psychopathol-ogy. In M. Potegal, G. Stemmler, & C.Speilberger (Eds.), International handbook ofanger: Constituent and concomitant biological,psychological, and social processes(pp. 467–497). New York, NY: SpringerScience þ Business Media.
Olatunji, B. O., Ciesielski, B. G., & Tolin, D. F.(2010). Fear and loathing: A meta-analyticreview of the specificity of anger in PTSD.Behavior Therapy, 41, 93–105, doi:10.1016/j.-beth.2009.01.004.
Orth, U., & Wieland, E. (2006). Anger, hostility,and posttraumatic stress disorder in trauma-exposed adults: A meta-analysis. Journal ofConsulting and Clinical Psychology, 74,698–706, doi: 10.1037/0022-006X.74.4.698.
Owen, J. M. (2011). Transdiagnostic cognitiveprocesses in high trait anger. Clinical Psychol-ogy Review, 31, 193–202, doi:10.1016/j.cpr.2010.10.003.
Riley, W. T., Treiber, F. A., & Woods, M. (1989).Anger and hostility in depression. Journal ofNervous and Mental Disease, 177, 668–674,doi:10.1097/00005053-198911000-00002.
Roemer, L., Borkovec, M., Posa, S., & Borkovec,T. (1995). A self-report diagnostic measure ofgeneralized anxiety disorder. Journal of Beha-vior Therapy and Experimental Psychiatry, 26,345–350, doi:10.1016/0005-7916(95)00040-2.
Roemer, L., & Orsillo, S. M. (2007). An open trialof an acceptance-based behavior therapy forgeneralized anxiety disorder. Behavior Therapy,38, 72–85, doi:10.1016/j.beth.2006.04.004.
Sexton, K., & Dugas, M. (2009). Defining distinctnegative beliefs about uncertainty: Validatingthe factor structure of the Intolerance ofUncertainty Scale. Psychological Assessment,21, 176–186, doi:10.1037/a0015827.
Spielberger, C. D. (1999). State-Trait AngerExpression Inventory-2: STAXI-2 professionalmanual. Odessa, FL: Psychological AssessmentResources.
Tabachnick, B. G., & Fidell, L. S. (2007). Usingmultivariate statistics (5th ed.). Boston, MA:Allyn & Bacon/Pearson Education.
VOL 41, NO 3, 2012 Anger in Generalized Anxiety Disorder 271
