A feasibility study of Acceptance and Commitment Therapy for emotional dysfunction following psychosis

The experience of psychosis can lead to depression, anxiety and fear. Acceptance and Commitment Therapy (ACT) facilitates individuals to accept difficult mental experiences and behave in ways that are consistent with personally held values. This
of 7
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  Shorter communication A feasibility study of Acceptance and Commitment Therapy for emotionaldysfunction following psychosis Ross White a , * , Andrew Gumley a , Jackie McTaggart a , Lucy Rattrie a , Deirdre McConville a ,Seonaid Cleare a , Gordon Mitchell b a  Academic Unit of Mental Health and Well-being, The University of Glasgow, Glasgow G12 0XH, United Kingdom b Department of Clinical Psychology, Stratheden Hospital, Cupar, Fife KY15 5RR, United Kingdom a r t i c l e i n f o  Article history: Received 16 February 2011Received in revised form2 August 2011Accepted 9 September 2011 Keywords: PsychosisAcceptanceMindfulnessDepressionTherapy a b s t r a c t The experience of psychosis can lead to depression, anxiety and fear. Acceptance and CommitmentTherapy (ACT) facilitates individuals to accept dif  󿬁 cult mental experiences and behave in ways that areconsistent with personally held values. This study was a single (rater) blind pilot randomised controlledtrial of ACT for emotional dysfunction following psychosis. Twenty-seven participants with psychosiswere randomised to either: ten sessions of ACT plus treatment as usual (TAU) or TAU alone. The HospitalAnxiety and Depression Scale, Positive and Negative Syndrome Scale, Acceptance and Action Ques-tionnaire, Kentucky Inventory of Mindfulness Skills and Working Alliance Inventory were used. Indi-viduals were assessed at baseline and 3 months post-baseline. The individuals randomised to receive ACTfound the intervention acceptable. A signi 󿬁 cantly greater proportion of the ACT group changed frombeing depressed at time of entry into the study to not being depressed at follow-up. The ACT groupshowed a signi 󿬁 cantlygreater increase in mindfulness skills and reduction in negative symptoms. Resultsindicated that individuals randomised to ACT had signi 󿬁 cantly fewer crisis contacts over the study.Changes in mindfulness skills correlated positively with changes in depression. ACT appears to offerpromise in reducing negative symptoms, depression and crisis contacts in psychosis.   2011 Elsevier Ltd. All rights reserved. Introduction The experience of psychosis has been shown to be associatedwith increased levels of depression (Birchwood et al., 2000; Rooke& Birchwood, 1998), hopelessness (White, McCreery, Gumley, &Mulholland, 2007), social anxiety (Cosoff & Hafner, 1998; Gumley, O ’ Grady, Power, & Schwannauer, 2004) and traumatic reactionssuf  󿬁 cient to qualify for PTSD (Shaw, McFarlane, Bookless, & Air,2002; White & Gumley, 2009). Randomised clinical trials havefound that Cognitive Behaviour Therapy for psychosis (CBTp) isef  󿬁 cacious for treating residual distressing positive and negativesymptoms (Wykes, Steel, Everitt, & Tarrier, 2008). However, theevidence for treating emotional dysfunction (such as anxiety,depression and hopelessness) is less clear (Birchwood, 2003). Although,Wykeset al.(2008) foundamoderatelystrongeffectsizeof CBTp on mood, when studies with  ‘ poor ’  methodological qualitywere controlled for, the weighted effect size on mood in theadequate quality studies was not signi 󿬁 cant.In recent years there has been a move towards incorporatingacceptance-based approaches into cognitive-behavioural frame-works to help alleviate distress associated with psychologicaldisorders (e.g. Rapgay & Bystrisky, 2009; Segal, Williams, &Teasdale, 2002). Acceptance and Commitment Therapy (ACT;Hayes, Strosahl, & Wilson,1999) is one such psychological therapy.ACT is derived from Relational Frame Theory (RFT; Hayes, Barnes-Holmes, & Roche, 2001); a behavioural theory concerned with thenature oflanguage andcognition.ACTconceptualises psychologicalsuffering as being largely caused by cognitive entanglement,experientialavoidance,andtheresultingpsychologicalrigiditythatimpedes people ’ s ability to take behavioural steps that are consis-tentwiththeindividual ’ scorevalues(Hayes&Smith,2005).Ratherthan altering the content or frequency of cognitions, ACT seeks toalter the individual ’ s  psychological relationship  with thoughts,feelingsandsensationstopromote greaterpsychological 󿬂 exibility.Pull (2008), summarising the  󿬁 ndings from previous reviews,concludedthatACThasevidencedeffectivenessfortreatingarangeof psychological disorders, but there is a need for more wellcontrolled studies to be conducted. *  Corresponding author. Academic Unit of Mental Health and Well-being, TheUniversity of Glasgow,1st Floor, Administration Building, Gartnavel Royal Hospital,1055 Great Western Road, Glasgow G12 0XH, United Kingdom. Tel.:  þ 44 01412113918; fax:  þ 44 0141 211 0356. E-mail address: (R. White). Contents lists available at SciVerse ScienceDirect Behaviour Research and Therapy journal homepage: 0005-7967/$  e  see front matter    2011 Elsevier Ltd. All rights reserved.doi:10.1016/j.brat.2011.09.003 Behaviour Research and Therapy xxx (2011) 1 e 7 Please cite this article in press as: White, R., et al., A feasibility study of Acceptance and Commitment Therapy for emotional dysfunctionfollowing psychosis,  Behaviour Research and Therapy  (2011), doi:10.1016/j.brat.2011.09.003  Two previous randomised controlled trials have investigatedACT for psychosis (Bach & Hayes, 2002; Gaudiano & Herbert, 2006). Both were non-blind trials that focused on inpatients withpsychotic disorders. Bach and Hayes (2002) found that individualsreceiving ACT demonstrated signi 󿬁 cantly lower belief in positivesymptoms at follow-up compared to the treatment as usual onlygroup (TAU). The re-hospitalisation rate in the ACT group was half that of the TAU onlygroup. Similarly, Gaudiano and Herbert (2006)found that signi 󿬁 cant decreases in beliefs about hallucinationsduring treatment were only observed in the ACT condition. Thesechanges in positive symptom conviction were strongly associatedwith changes in levels of distress.As yet, no blind randomised controlled trials of ACT forpsychosis havebeen conducted.Furthermore, no researchhas beenconducted to determine whether ACT is effective for addressingemotionaldysfunction(e.g.depressionandanxiety)thatcanfollowan acute episode of psychosis. In light of these issues, we set aboutdetermining the feasibility of conducting a blind randomisedcontrolled trial of ACT for emotional dysfunction followingpsychosis. The PICO framework (Oxman, Sackett, & Guyatt, 1993;Richardson, Wilson, Nishikawa, & Hayward, 1995) was used tospecify the parameters of the study aims and objectives:   Population: Could appropriate individuals be identi 󿬁 ed andrecruited to a trial of ACT for emotional dysfunction followingpsychosis?   Intervention: Would ACT be an acceptable intervention forindividuals diagnosed with a psychotic disorder? Would theyrate the extent to which they were able to identify goals fortherapy, work towards tasks and form a therapeutic bond withthe therapist be rated favourably?   Comparison: Could an appropriate group of participants berecruited to facilitate comparison with the ACT intervention?   Outcomes:Whatmeasureswouldbeimportantforassessingtheimpact of ACTon emotional dysfunction following psychosis? Method Design PACT was a 12-month  P rospective  R  andomised  O pen  B lind E valuation (PROBE) clinical trial exploring the feasibility of usingAcceptance and Commitment Therapy to facilitate emotionalrecovery following psychosis. Participants Participants were consecutively recruited, assessed and rando-misedfrommentalhealthservicesacrossNHSGreaterGlasgowandClydeincludingcommunitymentalhealthteams,earlyinterventionservices for psychosis, a medium-secure forensic service, andpsychiatric rehabilitation services. Participants all met ICD-10(WHO, 1992) criteria for a psychotic disorder (i.e., schizophrenia,schizoaffective disorder, schizophreniform disorder, delusionaldisorder, brief psychotic disorder, psychotic disorder NOS, bipolardisorder with psychotic features, or depressive disorder withpsychotic symptoms). Diagnoses were determined by case- 󿬁 lereview. Participants were excluded if there was a (1) diagnosis of learning disability; (2) inability to participate in psychotherapy/research due to acute medical condition or acute psychosis (asde 󿬁 ned by a score  5 on an item of the PANSS Positive Syndromesubscale); (3) psychotic symptoms due to a general medical condi-tion; (4) systematic psychological therapy being delivered at thepoint of referral. MeasuresGeneral outcome measures The  Hospital Anxiety and Depression Scale  (HADS; Zigmond &Snaith, 1983) is a widely used self-report instrument designedas a brief assessment tool of the distinct dimensions of anxietyand depression in non-psychiatric populations. Bjelland, Dahl,Haug, and Neckelmann (2002) noted that the psychometricproperties of the HADS are such that it can be used with con 󿬁 -dence clinically.The  Positive and Negative Syndrome Scale  (PANSS; Kay, Fiszbein,& Opler,1987): The PANSS is a 30-item observer rated scale used toassess the presence and severity of positive (e.g. delusions, hallu-cinatory behaviour) and negative (e.g. blunted affect, emotionalwith-drawl) symptoms. Derived scores include  ‘ positive ’  and ‘ negative ’  syndrome scores. Psychometric studies have reportedgood inter-rater reliability and satisfactory internal consistency,construct validity and concurrent validity in relation to othermeasures of psychopathology (Kay, Opler, & Lindenmayer, 1988;Kay, Opler, & Lindenmayer, 1989). Therapy-speci  󿬁 c measures Acceptance and Action Questionnaire  e  II (AAQ-II; Bond et al.,2011, in press): developed speci 󿬁 cally for assessing ACT outcomesit measures acceptance and experiential avoidance. The total scoreprovides an indication of psychological  󿬂 exibility. Consequently,lower scores on the AAQ-II are indicative of higher levels of expe-riential avoidance. Kentucky Inventory of Mindfulness Skills  (KIMS; Baer, Smith, &Allen, 2004): is a self-report inventory for the assessment of mindfulness skills. It assesses four mindfulness skills: observing,describing, acting with awareness, and accepting without judge-ment. Analyses have shown that the KIMS has good internalconsistency and test e retest reliability. Therapeutic alliance measure The Working Alliance Inventory (Short Form Revised; WAI-SR;Hatcher & Gillaspy, 2006) is a 12-item self-report measure of therapeutic alliance. It assesses 3 aspects of the therapeutic alli-ance: (a) agreement between client and therapist on the goals of therapy, (b) agreement between client and therapiston the tasks of therapy, and (c) the quality of the interpersonal bond betweenclient and therapist. Hatcher and Gillaspy (2006) reported that theinternal consistency coef  󿬁 cient alphas and convergent validitywere suf  󿬁 ciently high.  Arms of the study Treatment as usual (TAU) only: In the current study, TAUconsisted of psychopharmacology, case management, and/orpsychotherapy. This included review with the Consultant Psychi-atrist and contact with a designated key-worker (i.e. a CommunityPsychiatric Nurse or Occupational Therapist). In some cases, TAUalso included contact with a Social Worker and/or ClinicalPsychologist.Acceptance and Commitment Therapy þ TAU (which, to avoidconfusion with the other condition, is simply labelled  ACT  throughout the rest of this article): The ACT intervention wasdelivered by the lead author (RW) for up to 10 sessions in a one-to-one format. The ACT protocol was developed speci 󿬁 cally for thetrial and was based on the work of  Polk, Hambright, and Webster(2009). The ACT sessions incorporated work focussing on thefollowing themes: (1) Distinguishing between different types of experience: internal experience vs. 5-sense experience; (2) Rec-ognising how we get caught up struggling to move away from R. White et al. / Behaviour Research and Therapy xxx (2011) 1 e 7  2 Please cite this article in press as: White, R., et al., A feasibility study of Acceptance and Commitment Therapy for emotional dysfunctionfollowing psychosis,  Behaviour Research and Therapy  (2011), doi:10.1016/j.brat.2011.09.003  suffering; (3) Moving towards our values; (4) Getting distancebetween us and our  ‘ life stories ’ ; (5) Exploring how trying tocontrol dif  󿬁 cult mental experiences can be part of the problemrather than the solution; (6) Noticing that we can notice: focussingon the context in which mental experiences occur rather on thecontent of these experiences; (7) Exploring worry thoughts asso-ciated with psychosis. ACT has a strong mindfulness component. Amindfulness of breathing exercise used by Chadwick et al. (2005)was incorporated into the treatment. All therapy sessions wererecorded and competence and  󿬁 delity assessed by an expert inACT (GM). All participants receiving the ACT intervention werealso free to receive whatever psychopharmacology, case manage-ment, and/or additional psychotherapy that the clinical teamdeemed necessary. Procedure The research procedures were approved by the West of Scotland NHS Research Ethics Committee No. 3 (ref: 09/S0701/74), and R & D approval (ref: PN09CP213) was granted from NHSGreater Glasgow and Clyde. The research team met with sevenpsychiatric services from across NHS Greater Glasgow and Clydeto present the proposed research. Clinical vignettes were used tohighlight how emotional dysfunction following psychosis mightpresent. Referrals were invited to the study. A member of theresearch team then met with the individual to assess theirappropriateness for the study. Informed consent was thensought. Once consented, participants completed assessmentmeasures with a Research Assistant. Once baseline assessmentshad been completed, participant details were passed to AG whoundertook computerised randomisation using a predeterminedschedule of permuted blocks of random size. The research ther-apist then communicated the outcome of randomisation to eachparticipant.Participants met with a researcher (JMcT, LR, and DMcC) ona monthly basis to complete the self-report general outcome andtherapy-speci 󿬁 c measures. The assessors were all blind to treat-ment allocation. Two researchers (JMcT and LR) also completed allof the PANSS assessments. They both had extensive experience of using the PANSS in clinical and research settings. Inter-rateragreement was determined to be over 80%; consistent withrecommendations for reliability (Kay et al., 1991; Norman et al.,1996). One participant missed appointments at 3 month follow-up to complete the therapy-speci 󿬁 c measures. Their scores onthese measures at 2 month post-baseline follow-up were usedinstead. Overall, blindness was breached on 9 occasions ( n ¼ 7 forthe ACT arm;  n ¼ 2 for the TAU arm) during the trial. For all buttwo of these individuals, further follow-up assessments werecompleted by another researcher who remained blind toallocation.RGWdeliveredallACTsessions.Eachsessionwasapproximatelyonehourinlength.TheWAI-SRwasadministeredtoparticipantsinthe ACT arm of the study following session 5 of the ACT interven-tion. It was posted out to participant ’ s home address with a stam-ped addressed envelope for the form to be returned oncecompleted.Theletteraccompanyingthequestionnairewassentoutby AG and was returned to him.Following the completion of the trial JMcT reviewed thePersonal Information Management System (PiMS). This is an elec-tronic database used by NHS Greater Glasgow and Clyde to recordcontact between NHS Greater Glasgow and Clyde staff membersandpatients. The nature of these contacts (i.e.planned, crisisetc) isalso recorded. We were speci 󿬁 cally interested in determining thenumber of crisis contacts that participants had with mental healthprofessional over the duration of the trial. Statistical analyses Likelihood ratio  c 2 analyses were used to test whether thefrequency distribution of particular events were consistent acrossthe sample. Kolmogorov e Smirnov analyses were conducted todetermine if variables were normally distributed. Independentgroup  t  -tests were then used to compare between-group differ-ences between the TAU and ACT arms of the study at baseline fornormally distributed variables. The nonparametric equivalent of anindependent  t  -test (Mann e Whitney  U  ) was used to assess differ-ences between the TAU and ACT arms of the study at baselinevariables that were not normally distributed. Independent group t  -tests wereused tocompare between change scores (calculatedbysubtracting 3 month post-baseline scores from the baseline scores)for the TAU and ACT arms of the study at baseline for normallydistributed variables. The nonparametric equivalent of an inde-pendent  t  -test (i.e. Mann e Whitney  U   tests) was used to comparebetween change scores for the TAU and ACT arms of the study forvariables that were not normally distributed.Spearman ’ s  r  correlations (two-tailed) were used to test asso-ciations between change scores for the general outcome measuresand the therapy-speci 󿬁 c measures for the individuals in the ACTarm of the trial. Results Recruitment to the trial Fig. 1 provides information about the numbers of individualsreferred to the trial and how this translated into the number of participantsthatwererandomisedintothetrial.Overthe6monthsof the trial recruitment period 43 referrals were received (7.2referrals/month). A total of 35 individuals consented to participate Fig. 1.  CONSORT diagram of participants through the trial. R. White et al. / Behaviour Research and Therapy xxx (2011) 1 e 7   3 Please cite this article in press as: White, R., et al., A feasibility study of Acceptance and Commitment Therapy for emotional dysfunctionfollowing psychosis,  Behaviour Research and Therapy  (2011), doi:10.1016/j.brat.2011.09.003  (consentrateof81.39%).Atotalof27participantswererandomisedinto the study: 13 to treatment as usual (TAU) and 14 to ACT. Meanvalues for the participants randomised into the study were calcu-lated for the general outcome measures used in the study PANSSPositive Syndrome Subscale (12.00, SD ¼ 3.42), PANSS NegativeSyndrome Subscale (15.04, SD ¼ 5.50), HADS Depression Subscale(8.15, SD ¼ 4.36) and the HADS Anxiety Subscale (10.11, SD ¼ 5.18).Demographic details for participants are provided in Table 1.Likelihoodratio c 2 analysesindicatedthattherewerenosigni 󿬁 cantdifferences between the TAU and ACT groups on these variables.  Acceptability of the treatment  At 3 months post-baseline follow-up, 3 of the individualsrandomised to TAU had withdrawn from the study compared tonone of the participants randomised to the ACT intervention. Thisdifference was signi 󿬁 cant (likelihood ratio  c 2 ¼ 4.79,  p < 0.05). Theindividuals that withdrew from the study did not differ signi 󿬁 -cantly from the individuals that did not withdraw on any of themeasures. There were no suspected unexpected serious adversereactions over the course of the trial.AlloftheparticipantsreceivingtheACTinterventionreceived10sessions of ACT. The median number of appointments offered tocomplete the 10 sessions was 11 (IQR  ¼ 10.00 e 13.25). The mediannumber of DNAs was 1 (IQR  ¼ 0.00 e 2.00) and the median numberof therapy sessions cancelled was 0 (IQR  ¼ 0.00 e 1.25). The ACTintervention took a median of 11.50 (IQR  ¼ 10.00 e 17.25) weeks todeliver.The mean Working Alliance Inventory (short form; WAI-SF)ratings provided by participants for the Goal, Task and Bondsubscales were 17.50 (SD ¼ 2.12), 15.18 (SD ¼ 2.89) and 17.20(SD ¼ 2.70) respectively. The maximum possible score on each of the WAI-SFR subscales was 20. Comparisons of change in measures between the ACT and TAU  groups Table 2 provides details of the between-group comparisons of change scores between the TAU and ACT groups. There was nosigni 󿬁 cant difference in the change scores for the PANSS PositiveSyndromesubscale( t  ¼ 0.24,df  ¼ 19,  p > 0.05).Therewas,however,a signi 󿬁 cant difference between the ACT and TAU groups for thechange score of the PANSS Negative Syndrome subscale ( t  ¼ 2.36,df  ¼ 19,  p < 0.05). Therewasalsoatrendon thelimitofsigni 󿬁 cancefor differences between the groups in the change scores on theDepression subscale of the HADS ( t  ¼ 2.09, df  ¼ 19,  p ¼ 0.051).There was no signi 󿬁 cant difference between the groups on theAnxiety subscale of the HADS ( t  ¼ 0.12, df  ¼ 20,  p > 0.05).In terms of differences in the change scores of the therapy-speci 󿬁 c measures, there was a signi 󿬁 cant difference for the KIMSTotal score ( t  ¼ 2.66, df  ¼ 21,  p < 0.05). There were trendsapproaching signi 󿬁 cance regarding between-group differences forthe change scores for the KIMS Description ( t  ¼ 2.06, df  ¼ 21,  p ¼ 0.052), and the KIMS accepting without judgement ( t  ¼ 1.99,df  ¼ 21,  p ¼ 0.059) subscales. There was no signi 󿬁 cance differencebetween the two groups in the change scores for the AAQ-II( t  ¼ 0.60, df  ¼ 21,  p > 0.05).  Associations between change scores in general outcomeand therapy-speci  󿬁 c measures The change in Depression subscale of the HADS had signi 󿬁 cantcorrelations with the change scores for the: KIMS  Total  Score( r ¼ 0.66,  p < 0.05);  Describing   ( r ¼ 0.70,  p < 0.05) and KIMS  Acting with awareness  ( r ¼ 0.72,  p < 0.01) subscales. There wasatrendapproachingsigni 󿬁 canceregardingthenegativecorrelationbetween the change score for the Depression Subscale of the HADSand change in experiential avoidance as assessed by the changescoreoftheAAQ-II( r ¼ 0.57,  p ¼ 0.051).Thechangeinscoresfrombaseline to 3 month post-baseline of the PANSS Negative subscalescores did not correlate signi 󿬁 cantly with any of the therapy-speci 󿬁 c measures. Crisis contacts The ACT arm of the study, relative to the TAU arm, had a signif-icantlylowerproportionofindividualswhohadcrisiscontactsoverthe duration of the trial (likelihood ratio  c 2 ¼ 5.75,  p ¼ 0.016). TheACT arm also had a signi 󿬁 cantly lower number of crisis contacts(  Z  ¼ 2.24,  p < 0.05). Post-hoc analyses Post-hoc analyses were conducted on the basis of whetherparticipants met caseness on the HADS at baseline for depressionand anxiety. The Bjelland et al. (2002) criteria of a score  8 on the  Table 1 Demographic information about participants.Participantsrandomised toACT ( N  ¼ 14)Participantsrandomised toTAU ( N  ¼ 13)GenderMale 10 (71.40%) 11 (84.60%)Female 4 (28.60%) 2 (15.40%)Mean age (std) 33.57 (8.63) 34.54 (10.97)Marital statusSingle 13 (92.90%) 10 (76.90%)In a relationship 0 (0.00%) 2 (15.40%)Married 1 (7.10%) 0 (0.00%)Divorced 0 (0.00%) 0 (0.00%)Separated 0 (0.00%) 1 (7.70%)EducationLeft school < 16 yrs 5 (35.70%) 3 (23.10%)Left school at 16 yrs 4 (28.60%) 4 (30.80%)Left school at 17/18 yrs 2 (14.30%) 3 (23.10%)Completed/completing college course 3 (21.40%) 0 (0.00%)Completed university degree course 0 (0.00%) 1 (7.70%)Employment statusFull-time paid 0 (0.00%) 1 (7.70%)Part-time paid 0 (0.00%) 1 (7.70%)Student 1 (7.10%) 0 (0.00%)Unemployed (bene 󿬁 ts) 12 (85.70%) 9 (69.00%)Unemployed (no bene 󿬁 ts) 1 (7.10%) 2 (15.40%)EthnicityWhite British 14 (100.00%) 12 (92.30%)White other 0 (0.00%) 0 (0.00%)Pakistani 0 (0.00%) 0 (0.00%)African 0 (0.00%) 0 (0.00%)Other 0 (0.00%) 1 (7.70%)Not provided 0 (0.00%) 0 (0.00%)DiagnosisSchizophrenia (F20) 7 (50.00%) 6 (46.20%)Unspeci 󿬁 ed non-organicpsychosis (F29)4 (28.60%) 3 (23.10%)Schizo-affective disorder manic type(F25.0)0 (0.00%) 1 (7.70%)Schizo-affective disorder notspeci 󿬁 ed (F25.9)1 (7.10%) 1 (7.70%)Bipolar disorder mania andpsychosis (F31.2)1 (7.10%) 0 (0.00%)Bipolar disorder depression andpsychosis (F31.5)1 (7.10%) 1 (7.70%) R. White et al. / Behaviour Research and Therapy xxx (2011) 1 e 7  4 Please cite this article in press as: White, R., et al., A feasibility study of Acceptance and Commitment Therapy for emotional dysfunctionfollowing psychosis,  Behaviour Research and Therapy  (2011), doi:10.1016/j.brat.2011.09.003  Depression and Anxiety subscales of the HADS were used toascertain respective caseness. Fourteen individuals met casenessfor depression at entry into the trial (6 were subsequently rando-mised to TAU and 8 to ACT). A chi-square analysis, selecting onlythose individuals who were depressed at baseline, indicated thata signi 󿬁 cantly smaller proportion of individuals in the ACT arm of the study ( N  ¼ 2) met caseness for depression at 3 month post-baseline follow-up relative to the TAU arm ( N  ¼ 6) (likelihoodratio  c 2 ¼ 5.00,  p < 0.05). Eighteen individuals met caseness foranxiety at entry into the trial (9 were subsequently randomised toTAU and 9 to the ACT arm of the study). A chi-square analysis didnot  󿬁 nd any signi 󿬁 cant differences in the proportion of individualsmeetingcasenessforanxietyat3monthpost-baselinefollow-upinthe ACT arm of the study ( N  ¼ 6) and the TAU arm ( N  ¼ 7) (likeli-hood ratio  c 2 ¼ 1.01,  p ¼ 0.314). Discussion ThisfeasibilitystudyofACTforemotionaldysfunctionfollowingpsychosis is the  󿬁 rst  blind-rated  randomised controlled trial of ACTfor individuals with psychosis. The trial evidenced that referralpathways could be successfully established to identify individualspresenting with emotional dysfunction following psychosis. Averyhigh proportion of these referrals consented to participate in theresearch. The mean HADS Depression and Anxiety subscale scoresfor the sample were above the established cut-off point for clinicalcaseness (i.e.   8). The mean HADS Depression subscale score forthe sample corresponded to the 90th and 94th percentile ranks forScottishfemalesandmalesrespectively(Crawford,Henry,Crombie,& Taylor, 2001). Similarly, the mean Anxiety score for the samplecorresponded to the 84th and 91st percentile rank for Scottishfemales and males respectively (Crawford et al., 2001). It seemstherefore that the study succeeded in recruiting individuals expe-riencing elevated levels of emotional dysfunction.All of the participants receiving the ACT intervention completedthe treatment. The median number of DNA and cancelledappointments was low. The participants receiving ACT rated theextent to which they could work collaboratively with the therapistto identify goals for therapy; work through tasks during therapy;and form a close bond with the therapist, very favourably. Over thecourse of the treatment none of the participants receiving ACTexperienced any suspected unexpected serious adverse reactions.Consequently, it would appear that ACT as a treatment was highlyacceptable to this group of individuals.Measures were included in the current study to assesspurported treatment targets of ACT i.e. experiential avoidanceand mindfulness. Neither of the two previous studies investigatingACT for psychosis (Bach & Hayes, 2002; Gaudiano & Herbert,2006) employed these measures. Relative to the TAU arm of thestudy, participants receiving ACT had a signi 󿬁 cantly greaterchange in mindfulness skills. Our  󿬁 ndings add to previousresearch (Abba, Chadwick, & Stevenson, 2008; Chadwick, Hughes,Russell, Russell, & Dagnan, 2009) showing that individuals withpsychosis can tolerate mindfulness exercises and develop mind-fulness skills over time. In the current trial, there was no signi 󿬁 -cant difference between the two arms of in the degree of changein experiential avoidance across time. It could be that a longerperiod of follow-up is required for signi 󿬁 cant differences inexperiential avoidance to emerge. The AAQ-II (Bond et al., 2011, inpress) is a general measure of experiential avoidance and it maybe that other measures need to be developed to assess speci 󿬁 cforms of avoidance relevant to the experience of psychosis. Thiswork has already begun with the development of the VoicesAction and Acceptance Questionnaire (Shawyer et al., 2007) forassessing acceptance-based attitudes/actions in relation to audi-tory and command hallucinations.Asigni 󿬁 cantlygreaterproportionofpeopleintheACTarmofthestudy, relative toTAU, changed from being depressed at baseline tonot being depressed at 3 month follow-up. There was also a trendapproaching signi 󿬁 cance suggesting that the change in depressionscores intheACTarm of thestudywas greater than inthe TAU arm.These results are consistent with those of  Gaudiano and Herbert(2006) who found a marginally signi 󿬁 cant impact of ACT, relativeto enhanced TAU, on mood as assessed by the BPRS affect subscore.In the ACT arm of the current study, a signi 󿬁 cant association wasfound between changes in depression and changes in mindfulnessskills. To date methodologically rigorous research has failed toevidence the effectiveness of CBTp for treating depression inpsychosis (Wykes et al., 2008). This is all the more concerning in light of  Saarni et al. ’ s (2010)  󿬁 nding that depressive symptoms are  Table 2 Means (SD) on the measures for the ACT and TAU participants at baseline and 3 month follow-up.Baseline 3 Months post-baseline Change: baseline  3 months Between-group differencesin change scoresACT TAU ACT TAU ACT TAU ACT  TAU t  -Test/Mann e Whitneydf   p  Effectsize General outcome measures Positive syndromesubscale PANSS11.36 (2.62) 12.75 (4.16) 9.75 (3.60) 11.70 (4.72) 0.92 (3.99) 1.33 (3.71)   0.41  t  ¼ 0.24 19 0.810 0.05Negative syndromesubscale PANSS16.29 (4.87) 13.58 (6.02) 12.25 (5.03) 14.30 (4.67) 3.50 (3.78)   0.89 (4.76) 4.39  t  ¼ 2.36 19 0.029* 0.47HADS depression 8.62 (4.84) 7.92 (4.07) 4.00 (3.06) 6.22 (3.73) 4.62 (4.33) 1.63 (2.20) 2.99  t  ¼ 2.09 19 0.051 0.43HADS anxiety 8.57 (4.77) 11.77 (5.28) 6.08 (3.71) 10.70 (4.62) 2.69 (4.07) 2.88 (3.06)   0.19  t  ¼ 0.12 20 0.904 0.03 Therapy-speci  󿬁 c measures KIMS observation 33.08 (8.89) 31.62 (10.70) 38.31 (6.98) 34.90 (8.91)   5.23 (6.13)   2.00 (6.67)   3.23  t  ¼ 1.21 21 0.241 0.26KIMS description 24.15 (6.48) 26.00 (6.10) 29.54 (5.65) 24.80 (7.39)   5.38 (6.27) 0.90 (8.39)   6.28  t  ¼ 2.06 21 0.052 0.41KIMS awareness 28.92 (7.48) 30.54 (8.71) 30.77 (5.02) 28.50 (6.88)   1.85 (4.49) 1.30 (6.85)   3.15  t  ¼ 1.33 21 0.198 0.28KIMS acceptancewithout judgement25.46 (8.72) 28.77 (9.40) 29.15 (4.47) 25.10 (11.12)   3.69 (8.96) 3.70 (8.62)   6.69  t  ¼ 1.99 21 0.059 0.40KIMS total 111.61 (10.16) 116.92 (16.71) 127.77 (8.63) 113.30 (20.27)   16.15 (3.90) 3.90 (22.37)   20.05  t  ¼ 2.66 21 0.015* 0.50AAQ-II 40.15 (11.04) 39.23 (15.35) 47.77 (11.39) 42.10 (15.49)   7.62 (13.53)   4.20 (13.67)   3.42  t  ¼ 0.60 21 0.557 0.13*  p < 0.05.PANSS ¼ Positive And Negative Syndrome Scale; HADS ¼ Hospital Anxiety and Depression Scale; KIMS ¼ Kentucky Inventory of Mindfulness Skills; AAQ-II ¼ Acceptance andAction Questionnaire. R. White et al. / Behaviour Research and Therapy xxx (2011) 1 e 7   5 Please cite this article in press as: White, R., et al., A feasibility study of Acceptance and Commitment Therapy for emotional dysfunctionfollowing psychosis,  Behaviour Research and Therapy  (2011), doi:10.1016/j.brat.2011.09.003
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks