Dysfunctional Beliefs In Group And Individual Cognitive Behavioral Therapy For Obsessive Compulsive Disorder

Transcript

  See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/49753452 Dysfunctional beliefs in group and individualcognitive behavioral therapy for obsessivecompulsive disorder  Article   in  Journal of anxiety disorders · December 2010 DOI: 10.1016/j.janxdis.2010.12.001 · Source: PubMed CITATIONS 11 READS 88 3 authors , including:Hjalti JónssonAarhus University Hospital 6   PUBLICATIONS   94   CITATIONS   SEE PROFILE Birgit Egedal BennedsenAarhus University 16   PUBLICATIONS   622   CITATIONS   SEE PROFILE All content following this page was uploaded by Hjalti Jónsson on 21 August 2015. The user has requested enhancement of the downloaded file. All in-text references underlined in blue are added to the srcinal documentand are linked to publications on ResearchGate, letting you access and read them immediately.  This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institutionand sharing with colleagues.Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third partywebsites are prohibited.In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further informationregarding Elsevier’s archiving and manuscript policies areencouraged to visit:http://www.elsevier.com/copyright  Author's personal copy  Journal of Anxiety Disorders 25 (2011) 483–489 Contents lists available at ScienceDirect  JournalofAnxietyDisorders Dysfunctional beliefs in group and individual cognitive behavioral therapy forobsessive compulsive disorder Hjalti Jónsson a , b , ∗ , Esben Hougaard a , Birgit E. Bennedsen b a Department of Psychology, Aarhus University, Jens Chr, Skous Vej 4, 8000 Aarhus, Denmark b Clinic for Obsessive Compulsive Disorder, Aarhus University Hospital, Risskov, Skovagervej 2, 8240 Risskov, Denmark a r t i c l e i n f o  Article history: Received 9 June 2010Received in revised form 8 December 2010Accepted 11 December 2010 Keywords: Obsessive-compulsive disorderCognitive therapyCognitive theoryResponsibilityThought action fusionGroup therapy a b s t r a c t Theprimaryaimofthestudywastoinvestigatedysfunctionalbeliefsintheformofinflatedresponsibility(IR)andthoughtactionfusion(TAF)aspredictiveandmediatingvariablesinindividual( n =33)andgroup( n =37)cognitivebehavioraltherapy(CBT)forobsessivecompulsivedisorder(OCD).IRandTAFdeclinedsignificantlyduringCBT,andthedeclinewaspositivelyassociatedwithchangeinOCDsymptoms.How-ever, when controlling for change in depressive symptoms, only change in IR remained significantlyassociated with OCD symptom change. The moral subtype of TAF predicted poorer treatment outcome,butonlyingroupCBT.Bothtreatmentsproducedasimilaramountofchangeinthedysfunctionalbeliefs.The results provide some, preliminary evidence that IR, but not TAF, may be specifically involved in thechange mechanisms of both individual and group CBT for OCD, although the design of the study withpre- and post-therapy measurements only does not allow for a causal mediator analysis. © 2010 Elsevier Ltd. All rights reserved. 1. Introduction Cognitive behavioral therapy (CBT) including exposure andresponseprevention(ERP)forobsessivecompulsivedisorder(OCD)has been found to be effective in a number of randomized con-trolled clinical studies (Eddy, Dutra, Bradley, & Westen, 2004;Rosa-Alcázar, Sánchez-Meca, Gómez-Conesa, & Marín-Martínez,2008; van Blakom, van Oppen, Vermeulen, & van Dyck, 1994). CBTis generally recommended as the first choice among psychologicaltreatments for OCD, although there is insufficient evidence of theimportance of the specific cognitive methods in CBT (Abramowitz,2006; National Institute for Health and Clinical Excellence, 2006).Contemporary cognitive models of OCD imply that dysfunctionalbeliefs underlying negative appraisals of intrusive thoughts play apivotal role in the maintenance of OCD in that they lead to emo-tionaldiscomfortandcounterproductivecontrolstrategies,suchascompulsivebehaviors(e.g.Rachman,1997,1998;Salkovskis,1985,1999).An international working group of experts in the field, the ObsessiveCompulsiveCognitionsWorkingGroup[OCCWG](1997),srcinally identified six rationally derived belief constructs sup-posed to constitute a core cognitive profile for OCD: inflated  Abbreviations:  IR, inflated responsibility; I-CBT, individual cognitive behavioraltherapy; G-CBT, group cognitive behavioral therapy. ∗ Corresponding author. Tel.: +45 8942 4996; fax: +45 8942 4901. E-mail address:  [email protected] (H. Jónsson). responsibility; over importance of thoughts; need to controlthoughts; intolerance of uncertainty; overestimation of threat;and perfectionism. However, these theoretically derived belief domains appear to overlap (OCCWG, 2005). Based on studieswith the obsessive beliefs questionnaire (OBQ; OCCWG, 2003;OCCWG, 2005), srcinally constructed to measure the six cog-nitive domains, three broader belief domains were empiricallyderived:responsibility/threatestimation;perfectionism/certainty;andimportance/controlofthoughts(OCCWG,2005).StudiesbytheOCCWG (2005) seemed to imply that especially responsibility andimportanceofthoughtswerecharacteristicforOCDpatients.Thesetwo belief domains correspond to central belief constructs in thetwomajorcognitiveappraisalmodelsofOCD;inflatedresponsibil-ity(IR)inthemodelof Salkovskis(1985,1999),andthoughtactionfusion (TAF) in Rachman’s model (1997, 1998).IR, defined as “the belief  that one has power which is pivotal to bring about, or prevent, subjectively crucial negatively outcomes”(Salkovskis, 1999, p. 32), plays a central role in Salkovskis’ modelof OCD. According to the model, it is the specific interpretation of responsibility for harm to oneself or others that is hypothesizedto link intrusive cognitions with both the experience of discom-fort and the overt or covert neutralizing (compulsive) behaviorsin OCD. Studies with both clinical and non-clinical participantshavefoundsignificantcorrelationsbetweenIRandOCDsymptoms(e.g. Freeston, Ladouceur, Thibodeau, & Gagnon, 1992; O’Leary,Rucklidge,&Blampied,2009;Rhéaume,Freeston,Dugas,&Letarte,1995; Salkovskis et al., 2000) and that experimental manipulationofIRhasleadtoincreasedurgestorectifythesituation,andtomore 0887-6185/$ – see front matter © 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.janxdis.2010.12.001  Author's personal copy 484  H. Jónsson et al. / Journal of Anxiety Disorders  25 (2011) 483–489 distress (Arntz, Voncken, & Goosen, 2007; Bouchard, Rhéaume, &Ladouceur, 1999; Ladouceur, Rhéaume, Freeston, & Aublet, 1995;Lopatka & Rachman, 1995).Rachman’s model (1997, 1998) of OCD stresses the importance of distorted beliefs in the form of TAF, defined as the experienceof thoughts and actions as equivalent. TAF appears in two forms:the likelihood type of TAF – the belief that having an unaccept-ablethoughtmayactuallyenhancetheprobabilitythattheaversiveeventwilloccur;andthemoraltypeofTAF–thebeliefthathavingan unacceptable thought is morally equivalent to carrying out theaction concerned. Consistent with this theoretical account, stud-ies have repeatedly found a relationship between TAF and OCDsymptoms, with stronger correlations for likelihood TAF than formoralTAF(Amir,Freshman,Ramsey,Neary,&Brigidi,2001;Rassin,Diepstraten, Merckelbach, & Muris, 2001; Shafran, Thordarson,& Rachman, 1996). Furthermore, experimental studies with non- clinical participants indicate that TAF promotes intrusive thinking,and leads to more discomfort, resistance and attempts to neutral-ize the content of thoughts (Rassin, Merckelbach, Muris, & Spaan,1999).However,TAFhasalsobeenimplicatedinotherpsychologi-caldisorderssuchasdepression(Abramowitz,Whiteside,Lynam,& Kalsy, 2003) and other anxiety disorders (Hazlett-Stevens, Zucker,& Craske, 2002), suggesting that TAF may be a more general fea- ture common to other disorders characterized by negative af fect, ormayreflectamoregeneralpropensitytomagicalthinkingorlackof insight (Einstein & Menzies, 2004a,b).As suggested by the OCCWG (2005), IR and TAF are related the-oretical constructs in that both may be considered beliefs aboutthe importance of controlling intrusive thoughts. According toRachman(1997)IRcouldbebothacontributingfactortotheoccur-rence of TAF, and a consequence of this or other cognitive biases.A recent study ( n =63) including both OCD patients and patientswithotheranxietydisorders,aswellasnon-clinicalcontrols,founda large correlation of 0.60 between IR and TAF after controlling forseverity of depression (O’Leary et al., 2009), and a Tûrkish studyof 51 OCD patients found a correlation of 0.68 (Yorulmaz, Karanci,Bastug,Kisa,&Goka,2008).Thesestudiesusedthesamemeasuresof IR (responsibility attitude scale [RAS]; Salkovskis et al., 2000) and TAF (thought action fusion scale [TAFS]; Shafran et al., 1996) as the present one. In the study by O’Leary et al. (2009) IR and TAF were both significantly correlated with OCD symptoms, but onlythe correlation affecting IR remained significant after controllingfor depression.Since dysfunctional beliefs like IR and TAF are considered coreconstructsincognitivetheoriesofmaintenancefactorsinOCD,suchdysfunctionalbeliefsshouldbeexpectedtomediatechangesinOCDsymptomsinCBT;andprobablyalsotopredictpooreroutcome(cf.Keeley, Storch, Merlo, & Geffken, 2008), as strength of beliefs maybe assumed to relate to persistence of maintenance mechanisms.Treatment studies have demonstrated significant reductions inIR and TAF (as well as other OCD-related beliefs) following CBT,and that such changes are associated with improvement in OCDsymptoms(Bouvard,2002;Emmelkamp,vanOppen,&vanBalkom,2002; Rassin et al., 2001; Whittal, Woody, McLean, Rachman, &R obichaud,2010).However,noneofthesestudiescontrolledforthe effects of changes in depression, necessary for conclusion regard-ing the specific effects of dysfunctional belief for OCD, and notmore generally for general distress or negative affect. A recentstudy by Manos et al. (2010) examined whether changes in obses-sive beliefs (as measures by the OBQ) predicted changes in OCDsymptoms from pre- to post treatment among a severe clinicalsample ( n =108). When controlling for symptoms of depressionand anxiety, changes in the perfectionism/certainty subscale of OBQpredictedchangesinOCDseverity,whilechangesintheotherOBQ subscales (i.e., importance/control of thoughts or responsibil-ity/threat estimation) did not.Veryfewstudieshaveinvestigatedspecificdysfunctionalbeliefsas pre-treatment predictors of outcome in CBT for OCD. A recentreview by Keeley et al. (2008) only mentions one study on theOCCWG core cognitive constructs as predictors of response to CBTfor OCD; the study by McLean, Whittal, Thordarson et al. (2001)mentions no relationship between pre-treatment TAF or IR andoutcome.Eventhoughspecificbeliefdomainshavenotbeenmuchstudied for their unique association with CBT treatment failure inOCD,ithasbeensuggestedthatthestrengthorfixityofbeliefs,oftenreferred to as lack of insight or presence of overvalued ideation,may be related to poorer outcome (Foa, Abramowitz, Franklin, &Kozak, 1999; Kozak & Foa, 1994; Neziroglu, Pinto, Yaryura-Tobias,& McKay, 2004; Wiegartz, Carmin, & Pollard, 2002).Group CBT (G-CBT) has been found to achieve outcome compa- rable to individual CBT (I-CBT) (Anderson & Rees, 2007; Jaurrieta et al., 2008; Jónsson, Hougaard, & Bennedsen, 2010), but only onestudy known to the authors has investigated the role of dysfunc- tional beliefs in group therapy. In a randomized controlled study of group ERP versus G-CBT, McLean et al. (2001) included threebelief measures (i.e., an earlier version of RAS, TAFS, and inven-tory of beliefs related to obsessions: Freeston, Ladouceur, Gagnon,&Thibodeau,1993)findingthatonlyRASwassignificantlyreducedfollowing both group treatments. In a later study of I-CBT versusindividual ERP at the same centre, a significant decline on all sub-scales of the OBQ was reported, with no difference between ERPand CBT (Whittal, Thordarson, & McLean, 2005). Based on resultsfromthesetwostudies,itwassuggestedbytheauthorsthatG-CBTfor OCD may lead to less change in dysfunctional beliefs due to itsinsufficient time to identify and challenge individual participants’idiosyncratic beliefs.In summary, there is some evidence that changes in dysfunc-tional beliefs are related to symptomatic change in I-CBT for OCD,butalmostnoevidenceonsuchbeliefsaspre-treatmentpredictorsof outcome of CBT. The found relationship between dysfunctionalbeliefs and symptomatic change may indicate that such beliefsfunction as a mediator of change in CBT for OCD as implied bycognitive theory. However, only one of the prior studies took intoaccountchangeindepressivesymptomsduringtreatmentasapos-sibly contaminating factor, and none used methodologically strictdesignstoinvestigatethecausalrelationshipbetweenvariables(cf.Baron & Kenny, 1986; Kraemer, Wilson, Fairburn, & Agras, 2002).There are almost no studies of dysfunctional beliefs in G-CBT forOCD.The primary aim of the study was to investigate how IR andTAF were related to treatment outcome of CBT, both as a possiblemediator of change in that change in the variables is associ-ated with change in OCD symptoms (also after controlling forchangeindepressivesymptoms),andasapre-treatmentpredictivepatient variable. Secondary aims were to explore the relation-ship between IR and TAF in an OCD sample, and to comparethe role of these beliefs in G-CBT and I-CBT. It was hypothe-sized that IR and TAF would significantly decline following CBTtreatment with the decline being positively associated with OCDsymptom reduction; and that stronger beliefs in IR and TAF at pre-treatment would predict a less favorable outcome of treatment.Due to the paucity of prior studies, no pre-study hypotheses as tothe relative role of dysfunctional beliefs in G-CBT and I-CBT wasprovided. 2. Method  2.1. Study design This study was a part of a randomized comparative study of the effectiveness of group versus individual CBT for OCD. The  Author's personal copy H. Jónsson et al. / Journal of Anxiety Disorders  25 (2011) 483–489 485 overall design of the study and its main outcome results arereported elsewhere ( Jónsson et al., 2010).  2.2. Participants ParticipantswereconsecutivelyrecruitedfrompatientsreferredtoTheClinicforOCDatAarhusUniversityHospital,Denmark,from January2005untilJuly2007.TheywerediagnosedbytheuseoftheanxietydisordersinterviewscheduleforDSM-IV(ADIS-IV)(Brown,DiNardo, & Barlow, 1994) and the structured clinical interview forDSM-IVAxisIIPersonalityDisorders(SCID-II)(First&Gibbon,2004)by clinical psychologists supervised by the senior psychiatrist aspart of the ordinary routines at the clinic.A sub-sample of 86 participants with available data on the OCDbeliefmeasuresatpre-treatmentwasdrawnfromatotalsampleof 93participantsassessedatpre-treatment.Pre-andpost-treatmentdata on the belief measures were available for 70 participants; 33in I-CBT, and 37 in G-CBT.Themaincriteriaforinclusioninthestudywereaprimarydiag-nosisofOCD;age20–70;aYale-Brownobsessivecompulsivescale(Y-BOCS)scoreof  ≥ 16;sufficientproficiencyinDanishlanguageto join a group; and acceptance of being randomly assigned to treat-ments.PatientswhodeclinedrandomizationwereofferedordinaryI-CBTattheclinic.Patientswereexcludedfromthestudyiftheyhadany of the following diagnoses: organic brain disease; current psy-chotic episode, bipolar affective disorder, severe major depressiveepisode, severe substance use disorder; and cluster A personalitydisorder. Patients were also excluded if their medication dosagehadbeenunstableinathree-monthperiodpriortotreatmentstart.  2.3. Measures The following measures were included in the study.  2.3.1. The Yale Brown obsessive compulsive scale (Y-BOCS) The Yale Brown obsessive compulsive scale (Y-BOCS)(Goodman, Price, Rasmussen, & Mazure, 1989a) measures theseverity of OCD symptoms with respect to time spent, interfer-ence, distress, resistance and control. Obsessions and compulsionsare rated separately, each on 5 items ranging from 0–4, giving asub-total for severity of obsessions and compulsions as well as atotalscoreforall10items(thetotalscorewasusedintheanalyses).The Y-BOCS has become the gold standard for assessing outcomein OCD research, and has shown adequate inter-rater agreement,internal consistency and validity (Goodman, Price, Rasmussen,& Mazure, 1989b). A clinical psychologist (HJ) not involved inthe participants’ treatment performed the Y-BOCS ratings on allparticipants, except for seven participants rated by the seniorpsychiatrist at the clinic (BB). The raters were not blinded as to theparticipants’ treatment condition. An inter-rater reliability checkwas done on a sub sample of 13 patients interviewed jointly bythe two raters, finding an intra-class correlation of 0.72 (two-waysmixed for individual raters, absolute agreement).  2.3.2. Responsibility attitude scale (RAS) Responsibility attitude scale (RAS) (Salkovskis et al., 2000a) is a26-item self-report scale designed to assess general assumptions,attitudesandbeliefsheldaboutresponsibility.Itemsarescoredon7pointscales.AtotalRASscoreisobtainedbysummingalltheitemsand dividing by 26, resulting in a final score between 1 and 7 withhigher values representing higher levels of perceived responsibil-ity. The RAS has been reported to have good reliability (Cronbach’s ˛ =0.92),aswellasdiscriminativevalidity(Salkovskisetal.,2000).Cronbach’s ˛ in this study was 0.95.  2.3.3. Thought action fusion scale (TAFS) Thought action fusion scale (TAFS) (Shafran et al., 1996) is a19-item self-report measure that assesses the tendency to fusethoughts and actions. Each item is rated on a 5-point scale rang-ing from 0 (disagree strongly) to 4 (agree strongly), with a totalscore range of 0–76. Twelve items assess moral type of TAF (i.e.,TAFS-Moral); and seven items assess likelihood type of TAF (i.e.,TAFS-Likelihood). TAFS has been found to have good reliability(Cronbach’s  ˛ =0.85–0.96), and a principal components analysissupported the two-factor solution into TAFS-Moral and TAFS-Likelihood in an OCD sample (Shafran et al., 1996). Cronbach’s  ˛ in this study was 0.93 for the total scale; 0.93 for TAFS-Moral; 0.95for TAFS-Likelihood.  2.3.4. Beck depression inventory-second version (BDI-II) Beck depression inventory-second version (BDI-II) (Beck, Steer,& Brown, 1996) is a self-report measure with 21 items each scoredfrom 0–3. The BDI-II is the latest version of BDI, the most widelyused self-report scale for depressive symptoms, with revisionaimedatmakingthescalemoreconsistentwithDSM-IVcriteriaformajor depressive disorders. The BDI has been widely researched,and it is known to have good reliability and validity (Beck et al.,1996; Beck & Steer, 1984). Cronbach’s ˛ in this study was 0.89.  2.4. Treatment  All participants received one initial individual session aimedat deciding treatment objectives, and, thereafter, 15 weekly ses-sions, and three extra booster sessions at 1, 3, and 6 monthspost-treatment (follow-up data are not included in this study).All group sessions (including the booster sessions) were 2h inlengthwithtwotherapistsandsixparticipants.AllI-CBT(includingthe booster sessions) were of 1h length.The treatment program was primarily inspired by Salkovskis’model for OCD (1985, 1999) and it consisted of psycho-education,cognitive restructuring, and ERP exercises/behavior experiments(introduced as both an opportunity to test beliefs and to habituateto anxiety). A more detailed description of the treatment protocolis provided elsewhere ( Jónsson et al., 2010).  2.5. Statistical analysis At first, bivariate correlations were computed among baselinescores on the TAFS, RAS and Y-BOCS to investigate the degree of association between these variables, followed by an analysis of partialcorrelationscontrollingfortheeffectsofdepression(BDI-II).Within-group changes on outcome measures and OCD beliefswere analyzed by paired  t  -tests, and magnitude of change wasestimated according to Cohen’s formula:  d =( M  before − M  after )/SD (within)pooled  (Cohen, 1987). To analyze the association betweenchange in OCD beliefs and symptomatic change, at first, bivariatecorrelations between variables were calculated followed by hier-archical, stepwise regression analyses with change scores on theY-BOCS as a criterion variable entering change scores on the BDI-IIin step one, and change scores on RAS and TAFS in step two.To analyze OCD beliefs as predictors of treatment outcome,hierarchical, stepwise regression analyses were conducted withend-state and change scores on the Y-BOCS as criterion variables,entering pre-treatment scores on the Y-BOCS and on the BDI-II instep one, and pre-treatment scores on the RAS and TAFS in steptwo.Between-group differences in the magnitude of change onsymptomorbeliefmeasureswereanalyzedbymeansofmultivari-ate analysis of variance (MANOVA).All analyses were performed on the sample of 70 participantswith both pre- and post-treatment data on the OCD belief mea-