Journal of Affective Disorders 120 (2010) 272–276

Contents lists available at ScienceDirect

Journal of Affective Disorders

j ourna l homepage: www.e lsev ie r.com/ locate / j ad

Brief report

A brief dyadic group based psychoeducation program improves relapse ratesin recently remitted bipolar disorder: A pilot randomised controlled trial

Russell D'Souza a,b, Danijela Piskulic a,b, Suresh Sundram a,b,c,⁎a Northern Psychiatry Research Centre, Mental Health Research Institute, Melbourne, Australiab Department of Psychiatry University of Melbourne, Mental Health Research Institute, Melbourne, Australiac Molecular Psychopharmacology, Mental Health Research Institute, Melbourne, Australia

a r t i c l e i n f o

⁎ Corresponding author. Molecular PsychopharmacResearch Institute, Locked Bag 11, Parkville, VictTel.: +613 9388 1633; fax: +613 9387 5061.

E-mail address: ssundram@mhri.edu.au (S. Sundra

0165-0327/$ – see front matter © 2009 Elsevier B.V.doi:10.1016/j.jad.2009.03.018

a b s t r a c t

Article history:Received 11 November 2008Received in revised form 27 March 2009Accepted 27 March 2009Available online 9 May 2009

Background: Various adjunctive psychotherapies assist in decreasing relapse and improvingoutcomes for people with bipolar disorder (BD). Psychoeducation programs involving patient-only or caregiver-only groups have demonstrated some efficacy. We tested in recently remittedBD if a combined group based psychoeducation program involving patient–companion dyadsdecreased relapse.Method: 58 recently remitted BD out-patients were randomised to receive either treatment asusual (TAU, n=31) or 12×90 minute psychoeducation sessions delivered weekly in a groupprogram to the patient and companion (SIMSEP, n=27). After 12weeks SIMSEP patients revertedtoTAU and all patientswere followeduntil week 60 or relapse. The primary outcomemeasurewasrelapse requiring hospital or intensive community intervention.Results: 45 patients completed the study. 29 patients remained well at week 60 (SIMSEP n=17,TAU n=12), whilst 16 had relapsed (SIMSEP n=3, TAU n=13). The SIMSEP group were lesslikely to relapse (Fisher's exact test p=0.013; OR=0.16; 95% CI 0.04–0.70) and had an 11 weeklonger time to relapse compared to the TAU group (chi-square (1)=8.48, pb0.01). At studycompletion SIMSEP compared toTAUpatients had lower YoungMania Rating Scale scores (Mann–Whitney U=255, pb0.01).Limitations: The study was limited by a small sample size.Conclusion: A brief group psychoeducation program with recently remitted BD patients and theircompanions resulted in a decreased relapse rate, longer time to relapse, decreasedmanic symptomsand improved medication adherence suggesting utility in the adjunctive psychotherapeutictreatment of BD.

© 2009 Elsevier B.V. All rights reserved.

Keywords:Bipolar 1 disorderGroup psychoeducationClinical trial

1. Introduction

The chronic relapsing nature of bipolar disorder inconjunction with a lifetime prevalence of 1–2% results in itsglobal ranking as the 10th leading cause of years lived withdisability (Mathers et al., 2003) despite the widespreadavailability of efficacious pharmacotherapy. Closing the gap

ology, Mental Healthoria, 3052 Australia

m).

All rights reserved.

.

between effective treatment and poor long-term outcomesrequires interventions that address symptom persistence andfrequency of relapse. In this regard the use of adjunctivepsychosocial strategies aimed at ameliorating the deleteriouseffects of stressful life events, interpersonal and family con-flict and improving early recognition of relapse and medica-tion adherence have been trialled (Colom et al., 2003; Lamet al., 2005, 2003; Miklowitz et al., 2003; Reinares et al., 2008;Scott et al., 2006; Simon et al., 2005). A number of recentliterature reviews and meta-analyses have cautiously sup-ported the use of adjunctive psychotherapies in relapseprevention (Beynon et al., 2008; Miklowitz, 2006; Scottet al., 2007). However, the evidence base for psychoeducation

Table 1Description of the sessions comprising the group based psychoeducationprogram involving patients with bipolar 1 disorder and their nominatedcompanion.

Week 1 Introduction; patient and companion perspectiveson illness

Week 2 Program familiarisation; knowledge of the disorderWeek 3 Symptoms—mania, depression, mixed states, psychosisWeek 4 Symptoms—biological rhythms, episodic, stressWeek 5 Pharmacotherapy—role of medications, types of

medicinesWeek 6 Pharmacotherapy—adherence, ambivalence,

combinations, monitoringWeek 7 Psychotherapy—illness models, stressors, drugs and

alcohol, sexualityWeek 8 Psychotherapy—life charts, personal learning, risks,

coping strategiesWeek 9 Signals—relapse signature, emergency planning, help

optionsWeek 10 Signals—work, study, legal, financial, relationship

issues and strategiesWeek 11 Fine tuning—further resources and groups, mood charts,

residual symptomsWeek 12 Review and assessment—warning signs, help actions,

management plan, feedback, “graduation”

273R. D'Souza et al. / Journal of Affective Disorders 120 (2010) 272–276

programs is small and it has not been possible to identify thecomponents of these programs relevant to improved out-comes in bipolar disorder (Beynon et al., 2008).

A large psychoeducation trial (Colom et al., 2003)incorporated components on illness awareness, early detec-tion of prodromal symptoms and recurrences, enhancementof treatment compliance and induction of lifestyle regularityadministered in patient groups over 31.5 h. During the followup approximately 50% of patients had relapsed by 12 monthsand 67% by 24 months in the psychoeducation group andalthough this was highly significantly better than the controlgroup it indicates possible scope for improvement. The sameinvestigators recently examined if psychoeducation directedat caregivers, 12 weekly sessions of 90 minutes each, reducedrates of relapse in people with bipolar disorder (Reinareset al., 2008). They found the group intervention decreased theoverall rate of relapse in 12 months and this was due todecreased episodes of hypomania/mania but not depressiveor mixed episodes. Using a combination of these twoapproaches we investigated if a companion (spouse, careror significant other) and patient dyad in a group intervention(18 h) founded on similar psychoeducation principles couldimprove relapse rates. This was based on the premise that asloss of insight predicts poorer outcome in bipolar disorder(Yen et al., 2007, 2008) the ability of a companion to recogniserelapse and intervene early when the patient is unable mayfurther decrease the relapse rate. Couple/marital and familybased interventions attest to this improved outcome but havebeen considerably longer to administer and delivered singlyto couples or families (Clarkin et al., 1998; Miklowitz et al.,2003). We developed a relatively brief program, the Systema-tic Illness Management Skills Enhancement Programme-Bipolar Disorder (SIMSEP-BD) (D'Souza and Rich, 2002) tobe administered to companion–patient dyads in a groupsetting, and report here on the outcome of a pilot studyevaluating its efficacy.

2. Patients and methods

The study was approved by the local hospital humanresearch ethics committee and written informed consent ofeach participant was obtained. Recently remitted patients(Young Mania Rating Scale (YMRS) score b10 and Montgom-ery–Asberg Depression Rating Scale (MADRS) score b8) wererecruited within 1month following discharge from hospital forrelapse of bipolar disorder. Diagnosis was established using theMINI and patients excluded if they had a comorbid axis 1disorder or a current substance dependence disorder. Patientswere randomly assigned in a non-stratified manner to receiveSIMSEP-BD or treatment as usual (TAU). The SIMSEP-BD wasadministered in a group setting as 12weekly sessions of 90minby four mental health clinicians led and trained by one of theauthors (RD'S). Supervision andmonitoring of the trainingwasconducted three-weekly by RD'S. The components of theSIMSEP-BD are listed in Table 1. Patients and their companionsattended all sessions and were excluded if more than 2consecutive or four in total sessions were missed. Aftercompletion of the program the SIMSEP-BD patients revertedto TAU. TAU was a community based case management modelinvolving a trained mental health clinician review weekly withthe patient for approximately 45 min and a medical review

monthly. These interactions were not controlled. All patientswere followed up for 60 weeks from baseline or until relapse.Relapse was defined as a recurrence of bipolar symptomsrequiring either hospital admission or intensive communitypsychiatric intervention by a psychiatric crisis team. Patientswere ratedbyexternal raters blind to treatment groupusing theYoung Mania Rating Scale (YMRS), Montgomery–AsbergDepression Rating Scale (MADRS) and a rater assessedmedication adherence scale (ARS). The ARS was scored 0 fornon-adherence,1 for partial adherence and 2 for full adherencebasedonpill count andneed for repeatmedicationprescription.All rating scales were administered weekly for the first12 weeks, then at 16, 20, 24, 32, 40, 48 and 60 weeks or untilrelapse or withdrawal.

Data were initially tested for normality and an intentionto treat analysis was used. Demographic and symptom ra-tings were compared between groups using the Mann–Whitney U test with last observation carried forward usedfor missing data. The primary outcome variable, relapse,was compared between the two groups using Kaplan–Meierand Cox regression survival analyses. Two separate Kaplan–Meier survival analyses were performed to ascertain groupdifferences in survival time, following the SIMSEP programor TAU weeks 1–12 and post intervention weeks 13–60.Likewise, two separate Cox regression survival analyseswere performed for the time intervals mentioned above, toassess the effectiveness of SIMSEP after adjusting for twocovariates which were found to be significantly differentbetween the two groups: medication type and medicationadherence. Alpha levels below 0.05 were viewed as sta-tistically significant and all analyses were performed usingSPSS, Version 13.

3. Results

58 consecutive patients were randomised, 27 to theSIMSEP arm and 31 to the TAU arm and the summarised

Fig. 1. Survival curves for patients receiving a group based psychoeducationprogram (SIMSEP-Yes) or treatment as usual (SIMSEP-No).

274 R. D'Souza et al. / Journal of Affective Disorders 120 (2010) 272–276

demographic and clinical data are presented in Table 2. Therewere no significant differences between the two groups inage, gender, duration of illness, last illness episode, presenceor absence of companion or baseline YMRS or MADRSsymptom ratings. Companions in the SIMSEP group werespouse/partner (n=20), child (n=3), friend (n=2), parent(n=1) and sibling (n=1). Although all patients receivedmood stabilisers, more patients in the SIMSEP group receivedantidepressants (SIMSEP n=6, TAU n=2) and benzodiaze-pines (n=15, n=5), although this was only significant forbenzodiazepines.

Forty-five (78%) participants completed the study with 29(50.0%) remaining well until week 60. These were 17 subjectsin the SIMSEP group and 12 in the TAU whilst three SIMSEPand 13 TAU subjects relapsed. Of the 13 withdrawn cases, 1 inthe SIMSEP and 4 in the TAU groups moved away from theregion (Fisher's exact test: pN0.1) and 6 in the SIMSEP and 2in the TAU groups withdrew from the study (Fisher's exacttest: pN0.1). Cases that moved out of region were excludedfrom the final analysis, and the outcomes for the 53 remainingsubjects are displayed in Fig. 1. The rate of relapse wassignificantly different between the two groups (Fisher's exacttest p=0.013) with the risk of relapse being significantly lessin the SIMSEP group (OR=0.16; 95% CI 0.04–0.70). Moreover,the SIMSEP group was likely to have significantly longer timeto relapse then the TAU group (hazards ratio of 0.19; 95% CI0.05–0.68; G2 (1)=6.48, pb .05) and this was on average11 weeks longer in the SIMSEP group (chi-square (1)=8.48,pb0.01). This difference in relapse was reflected in themeasure of manic symptoms where the SIMSEP group(M=3.3, SD=5.3) rated significantly less than the TAUsubjects (M=9.4, SD=9.5; U=255, p=0.009). There washowever, no difference on the MADRS scores (U=390.5,pN0.05) between the SIMSEP (M=2.7, SD=5.6) and the TAUgroup (M=3.6, SD=6.5). Medication adherence was sig-nificantly better in the SIMSEP group (M=1.2, SD=1.0)compared to the TAU group (M=0.4, SD=0.7; U=233,p=0.001).

Given that increased use of benzodiazepines and / orimproved medication adherence could explain the decreasedrelapse rate observed in the SIMSEP group, we conducted Coxsurvival regression analyses with these two factors as

Table 2Demographic and clinical characteristics of patients at baseline assigned toeither group based psychoeducation program (SIMSEP) or treatment as usua(TAU).

SIMSEP(n=27)

TAU(n=31)

P value

Gender (M/F) 13/14 15/16 NS b

Age (years) mean±SD 40.7±8.4 39.5±10.7 NSBipolar disorder type I/II 21/6 29/2 NSDuration of illness (years)mean±SD

6.4±4.3 6.0±5.4 NS

Last illness episodemanic/depressed/mixed

14/11/2 20/11/0 NS

Medications AP/AD/BZ a 5/6/15 6/2/5 NS/NS/b0.01YMRS mean±SD 5.4±4.6 3.8±4.3 NSMADRS mean±SD 4.3±3.8 2.9±3.1 NS

a AP=antipsychotic/AD=antidepressant/BZ=benzodiazepines.b NS=not significant.

l

covariates. After adjusting for the covariates; benzodiazepinetreatment ceased to be significant (G2 (1)=0.124, p=0.76).Medication adherence did however, significantly predictsurvival time (G2 (1)=3.12, pb0.001).

4. Discussion

Meta-analytic findings support the utility of adjunctivepsychotherapy (AP) programs for relapse prevention in bi-polar disorder (Beynon et al., 2008; Miklowitz, 2006; Scottet al., 2007; Soares-Weiser et al., 2007). However, thecharacteristics and components of these programs thatunderlie their effectiveness are not fully elucidated (Beynonet al., 2008). We report a pilot study demonstrating that arelatively brief dyadic psychoeducation program is effective inpreventing relapse of bipolar disorder, in part, through thepromotion of medication adherence.

The majority of AP programs have been either individuallyadministered or administered in a group setting to patients orto care-givers alone. The SIMSEP program was based on thereasoning that as impaired insight is both a feature ofeuthymic BD patients and correlates with poor outcome(Yen et al., 2007, 2008) then involving a close companion toassist in monitoring and promoting medication adherencemay improve outcomes; the emphasis being on a sharedresponsibility to maintain wellness. Thus the program wasdeveloped to be delivered in a group setting to a patient andhis/her nominated companion. The requisites of the compa-nion were to spend time with the patient at least weekly andto know the patient well enough to both recognise deteriora-tion in his/her mental state and to initiate an interventionsuch as monitoring of medication adherence or early re-assessment by the psychiatric team. Hence, the SIMSEPprogram fostered a collaborative approach to the manage-ment of the illness by empowering a trusted figure to inter-vene early when symptoms may have compromised the

275R. D'Souza et al. / Journal of Affective Disorders 120 (2010) 272–276

subject's capacity to do so. The active involvement of thecompanion appeared relevant rather than simply their pre-sence as all but one participant in the TAU group alsoidentified a companion with the above qualities.

The difference in relapse rates was not attributable tobaseline differences in either demographic or clinical char-acteristics. Both groups had some but relatively low symptomlevels at baseline consonant with having been recentlydischarged from hospital. This also permitted individuals tofully participate in the psychoeducation program. However,there was no minimum period of euthmyia in contrast toother studies (Colom et al., 2003; Reinares et al., 2008)possibly more closely approximating clinical practice. Theeffectiveness of the SIMSEP program contrasts with the Scottet al. (2007) meta-analysis where only euthymia greater than12 months was associated with a decreased relapse rate inindividuals receiving AP. As such the SIMSEP program mayoffer promise as an effective therapy earlier in the recoveryprocess. However, possible clustering and regression to themean in the SIMSEP group may have partly contributed to thedifference observed.

It should be noted that number of previous illnessepisodes was not recorded; hence, although length of illnesswas not different between the groups it is possible that onegroup may have been more prone to relapse because of moreprevious episodes. This is of relevance as individuals withmore than 12 previous mood episodes have previously beenshown to not respond well to adjunctive therapy (Scott andColom, 2005). Therefore in any larger scale replication thisvariable should be controlled. One difference within treat-ment was the significantly greater use of adjunctive benzo-diazepines in the SIMSEP group. Multivariate regression didnot identify this as a significant factor in predicting relapse;however, it remains possible that it may indicate anunidentified factor that may have contributed to the differ-ence in relapse rates.

Manic but not depressive symptoms were significantlydifferent between the SIMSEP and TAU groups at the finaltime point. These findings are consistent with a recent studyof caregiver only psychoeducation reducing hypomanic/manic but not depressive or mixed relapse (Reinares et al.,2008). It may indicate a relative ease in detecting manicsymptoms by companions which is less obvious withdepressive symptoms (Jackson et al., 2003) or it may bethat mania is more responsive to caregiver initiated inter-vention than depression (Cousins and Young, 2007).

A significant factor in explaining the difference in relapserates was the improved medication adherence in the SIMSEPgroup. This was previously noted in a couple-based maritaltherapy intervention (Clarkin et al., 1998) but not in acaregiver only program (Reinares et al., 2008). An emphasisin the SIMSEP-BD program was the acceptance of mutualresponsibility and the permission given to the companion bythe patient to intervene should relapse be imminent. Thusthis may be a key emergent property of a dyad based programfor relapse prevention in bipolar disorder.

In conclusion, a relatively brief dyadic group based out-patient psychoeducation program decreased relapse inrecently remitted bipolar disorder patients, which was inpart through improved medication adherence. This pilotstudy needs to be replicated in a larger sample but supports

the validity of this approach in improving outcomes forpeople with bipolar disorder and their families.

Role of funding sourceFunding for this study was provided by internal, departmental sources,

which had no further role in study design. There were no additional externalsponsors for this study.

Conflict of interestRD'S has received, unrelated to this study, grant and travel funding from

and acted as consultant/speaker for Astra Zeneca, Bristol-Myers Squibb, EliLilly & Co., Janssen-Cilag, Lundbeck, Organon, Pfizer and Sanofi-Synthelabo.

SS has received, unrelated to this study, grant and travel funding fromand acted as consultant/speaker for Astra Zeneca, Bristol-Myers Squibb, EliLilly & Co., Glaxo Smith Kline, Organon and Pfizer.

DP has no reported conflict of interest.

Acknowledgement

We thank Mr. Michael Tonso for his expertise in assess-ment and rating of symptoms.

References

Beynon, S., Soares-Weiser, K., Woolacott, N., Duffy, S., Geddes, J.R., 2008.Psychosocial interventions for the prevention of relapse in bipolar dis-order: systematic review of controlled trials. Br. J. Psychiatry 192, 5–11.

Clarkin, J.F., Carpenter, D., Hull, J., Wilner, P., Glick, I., 1998. Effects ofpsychoeducational intervention for married patients with bipolar dis-order and their spouses. Psychiatr. Serv. (Washington, D.C.) 49, 531–533.

Colom, F., Vieta, E., Martinez-Aran, A., Reinares, M., Goikolea, J.M., Benabarre,A., Torrent, C., Comes, M., Corbella, B., Parramon, G., Corominas, J., 2003.A randomized trial on the efficacy of group psychoeducation in theprophylaxis of recurrences in bipolar patients whose disease is in re-mission. Arch. Gen. Psychiatry 60, 402–407.

Cousins, D.A., Young, A.H., 2007. The armamentarium of treatments forbipolar disorder: a review of the literature. Int. J. Neuropsychopharmacol.10, 411–431.

D'Souza, R., Rich, D., 2002. A case-control study in the use of ‘IllnessManagement Skills Enhancement Programme’ for treatment adherencein patients with a bipolar disorder. Bipolar Disord. 4, 121.

Jackson, A., Cavanagh, J., Scott, J., 2003. A systematic review of manic anddepressive prodromes. J. Affect. Disord. 74, 209–217.

Lam, D.H., Hayward, P., Watkins, E.R., Wright, K., Sham, P., 2005. Relapseprevention in patients with bipolar disorder: cognitive therapy outcomeafter 2 years. Am. J. Psychiatry 162, 324–329.

Lam, D.H., Watkins, E.R., Hayward, P., Bright, J., Wright, K., Kerr, N., Parr-Davis,G., Sham, P., 2003. A randomized controlled study of cognitive therapy forrelapse prevention for bipolar affective disorder: outcome of the firstyear. Arch. Gen. Psychiatry 60, 145–152.

Mathers, C.D., Bernard, C., Iburg, K.M., Inoue, M., Fat, D.M., Shibuya, K., Stein,C., Tomijima, N., Xu, H., 2003. Global Burden of Disease in 2002: datasources, methods and results. Global Programme on Evidence for HealthPolicy World Health Organization, pp. 1–118.

Miklowitz, D.J., 2006. A review of evidence-based psychosocial interventionsfor bipolar disorder. J. Clin. Psychiatry 67 (Suppl 11), 28–33.

Miklowitz, D.J., George, E.L., Richards, J.A., Simoneau, T.L., Suddath, R.L., 2003.A randomized study of family-focused psychoeducation and pharma-cotherapy in the outpatient management of bipolar disorder. Arch. Gen.Psychiatry 60, 904–912.

Reinares, M., Colom, F., Sanchez-Moreno, J., Torrent, C., Martinez-Aran, A.,Comes, M., Goikolea, J.M., Benabarre, A., Salamero, M., Vieta, E., 2008.Impact of caregiver group psychoeducation on the course and outcomeof bipolar patients in remission: a randomized controlled trial. BipolarDisord. 10, 511–519.

Scott, J., Colom, F., 2005. Psychosocial treatments for bipolar disorders.Psychiatr. Clin. North Am. 28, 371–384.

Scott, J., Paykel, E., Morriss, R., Bentall, R., Kinderman, P., Johnson, T., Abbott,R., Hayhurst, H., 2006. Cognitive–behavioural therapy for severe andrecurrent bipolar disorders: randomised controlled trial. Br. J. Psychiatry188, 313–320.

Scott, J., Colom, F., Vieta, E., 2007. A meta-analysis of relapse rates withadjunctive psychological therapies compared to usual psychiatric treat-ment for bipolar disorders. Int. J. Neuropsychopharmacol. 10, 123–129.

276 R. D'Souza et al. / Journal of Affective Disorders 120 (2010) 272–276

Simon, G.E., Ludman, E.J., Unutzer, J., Bauer, M.S., Operskalski, B., Rutter, C.,2005. Randomized trial of a population-based care program for peoplewith bipolar disorder. Psychol. Med. 35, 13–24.

Soares-Weiser, K., Bravo Vergel, Y., Beynon, S., Dunn, G., Barbieri, M., Duffy, S.,Geddes, J., Gilbody, S., Palmer, S., Woolacott, N., 2007. A systematic reviewand economic model of the clinical effectiveness and cost-effectivenessof interventions for preventing relapse in people with bipolar disorder.Health technology assessment (Winchester, England) 11, iii–iv, ix–206.

Yen, C.F., Chen, C.S., Yang, S.J., Ko, C.H., Yen, J.Y., Huang, C.F., 2007. Re-lationships between insight and psychosocial adjustment in patientswith bipolar I disorder. Bipolar Disord. 9, 737–742.

Yen, C.F., Chen, C.S., Yen, J.Y., Ko, C.H., 2008. The predictive effect of insight onadverse clinical outcomes in bipolar I disorder: a two-year prospectivestudy. J. Affect. Disord. 108, 121–127.