Journal of Affective Disorders - Boston Universitysites.bu.edu › sswhrsaseminar › files › 2013 › 10 › Psychological...Journal of Affective Disorders 139 (2012) 217–229

Journal of Affective Disorders 139 (2012) 217–229

Contents lists available at ScienceDirect

Journal of Affective Disorders

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

Review

Psychological interventions for alcohol misuse among people withco-occurring depression or anxiety disorders: A systematic review

Amanda L. Baker a,⁎, Louise K. Thornton a, Sarah Hiles a, Leanne Hides b, Dan I. Lubman c

a Centre for Brain and Mental Health Research, University of Newcastle, New South Wales, Australiab Institute of Health and Biomedical Innovation (IHBI), Queensland University of Technology, Brisbane, Queensland, Australiac Turning Point Alcohol and Drug Centre, Eastern Health and Monash University, Melbourne, Victoria, Australia

a r t i c l e i n f o

⁎ Corresponding author at: Centre for Brain and Men40335690; fax: +61 2 40335692.

E-mail address: [email protected]

0165-0327/$ – see front matter © 2011 Elsevier B.V.doi:10.1016/j.jad.2011.08.004

a b s t r a c t

Article history:Received 4 March 2011Received in revised form 1 August 2011Accepted 1 August 2011Available online 3 September 2011

Objective: Depression, anxiety and alcohol misuse frequently co-occur. While there is anextensive literature reporting on the efficacy of psychological treatments that targetdepression, anxiety or alcohol misuse separately, less research has examined treatments thataddress these disorders when they co-occur. We conducted a systematic review to determinewhether psychological interventions that target alcohol misuse among people with co-occurring depressive or anxiety disorders are effective.Data sources:Wesystematically searched the PubMedand PsychINFOdatabases from inception toMarch 2010. Individual searches in alcohol, depression and anxiety were conducted, and werelimited to ‘human’ published ‘randomized controlled trials’ or ‘sequential allocation’ articleswritten in English.Study selection: We identified randomized controlled trials that compared manual guidedpsychological interventions for alcohol misuse among individuals with depressive or anxietydisorders. Of 1540 articles identified, eight met inclusion criteria for the review.Data extraction: From each study, we recorded alcohol and mental health outcomes, and otherrelevant clinical factors including age, gender ratio, follow-up length and drop-out rates. Qualityof studies was also assessed.Data synthesis:Motivational interviewing and cognitive–behavioral interventionswere associatedwith significant reductions in alcohol consumption and depressive and/or anxiety symptoms.Although brief interventions were associated with significant improvements in both mentalhealth and alcohol use variables, longer interventions produced even better outcomes.Conclusions: There is accumulating evidence for the effectiveness ofmotivational interviewing andcognitive behavior therapy for people with co-occurring alcohol and depressive or anxietydisorders.

© 2011 Elsevier B.V. All rights reserved.

Keywords:AlcoholComorbidityDepressionAnxietyTreatmentSystematic review

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2182. Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2193. Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 224

3.1. Trials of psychological interventions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 224

tal Health Research, University of Newcastle, University Drive, Callaghan NSW, Australia 2308. Tel.: +61 2

(A.L. Baker).

All rights reserved.

http://dx.doi.org/10.1016/j.jad.2011.08.004

mailto:[email protected]

http://dx.doi.org/10.1016/j.jad.2011.08.004

http://www.sciencedirect.com/science/journal/01650327

218 A.L. Baker et al. / Journal of Affective Disorders 139 (2012) 217–229

3.2. Alcohol misuse among people with mood disorders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2243.2.1. Alcohol misuse and depression. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2243.2.2. Alcohol misuse among people with dysthymia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2243.2.3. Alcohol misuse among mixed psychiatric inpatient samples . . . . . . . . . . . . . . . . . . . . . . . . . . 224

3.3. Alcohol misuse among people with anxiety disorders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2253.3.1. Alcohol misuse and social phobia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2253.3.2. Alcohol misuse and panic disorder/agoraphobia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2253.3.3. Comparison of effect sizes for changes in alcohol use and depression/anxiety symptoms immediately

following treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2253.3.4. Limitations of psychological intervention trials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226

3.4. Summary of findings of psychological intervention trials and suggestions for further research . . . . . . . . . . . . 2263.4.1. Alcohol misuse and depressive disorders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2273.4.2. Alcohol misuse and anxiety disorders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2283.4.3. Retention in treatment and follow-up . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228

4. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228Role of funding source . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228Conflict of interest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

228228228

225

226227

1. Introduction

Epidemiological surveys consistently indicate that depres-sive, anxiety and alcohol use disorders frequently co-occur(Farrell et al., 2001; Grant et al., 2004; Kessler et al., 2003).Studies conducted in the United States and Australia havefound that individuals with alcohol dependence are three tofour times more likely to have a concurrent affective or anxietydisorder compared to the general population (Degenhardt etal., 2001; Grant et al., 2004). Even higher rates of comorbiddisorders are found within treatment settings. In a largepopulation study, 32.8% of participants with alcohol usedisorders who sought treatment were found to have comorbiddepression and 33.4% were found to have a comorbid anxietydisorder (Grant et al., 2004). Such high rates are problematic asco-occurring alcohol, depressive and anxiety disorders havebeen associated with a broad range of negative outcomes,including more severe depressive and anxiety symptoms andsuicidal ideation, poorer social functioning and increasedservice utilization (Sullivan et al., 2005). In terms of treatmentoutcomes, while individuals with and without comorbidconditions improve, those with co-occurring conditions con-tinue to drink more, have poorer physical and mental healthoutcomes, and display poorer functioning following treatment(Mills et al., 2009).

Psychological treatments for unipolar depression, anxietyand alcohol use disorders have separately been shown to beeffective. Meta-analyses examining randomized controlledtrials (RCTs) of cognitive behavior therapy (CBT) for adultunipolar depression, anxiety or alcohol disorders have foundthat CBT is superior towaitlist anduntreated controls, aswell aspharmacotherapy (Dobson, 1989; Gloaguen, 1989; Hofmannand Smits, 2008;Magill and Ray, 2009; Norton and Price, 2007;Stewart and Chambless, 2009). During the last 30 years, therehas been a significant paradigm shift from the dichotomousconcept of ‘normal drinking’ versus an ‘alcohol use disorder’ tothe concept of a spectrum of hazardous to harmful drinking,delineated as ‘alcohol misuse’ (Saunders and Lee, 2000). Twometa-analytic reviews have found evidence for the efficacy of

brief (often one session) motivational interviewing (MI)interventions for alcohol misuse, with one finding a mediumeffect in non-treatment seeking populations and a small tomoderate effect in treatment seekers (Hettema et al., 2005;Moyer et al., 2002).

Although there have been numerous trials examining theeffectiveness of psychological interventions for unipolar de-pression, anxiety disorders and alcohol misuse separately,relatively few have been conducted for individuals withdepressive or anxiety disorders and comorbid alcohol misuse.Nevertheless, a number of recent reviews have demonstratedthe effectiveness of psychological interventions for co-occur-ring substance misuse and unipolar depression (assessed viadiagnostic interview or ratings/questionnaires with cut-offscores indicative of a clinical disorder), dysthymia or anxietydisorders. Hesse (2009) reported that integrated psychologicaltreatments that combine treatment for substance use disordersand co-occurring depression or dysthymia into one programhad superior outcomes in terms of the percentage of daysabstinent compared with treatment for substance use disorderalone. Similarly, Hides et al. (2010) found support for theefficacy of CBT over no treatment control conditions amongpatients with co-occurring unipolar depression or dysthymiaand substancemisuse (including alcohol). Baillie and Sannibale(2007) reviewed clinical trials for co-occurring anxiety andsubstance use disorders and concluded that standard care forsubstance use had the best outcomes for those with morethan moderate substance dependence in five of the six studiesreviewed.

No previous studies have systematically examined theefficacy of psychological interventions for patients withunipolar depression, dysthymia or anxiety disorders and co-occurring alcohol misuse specifically (rather than substancemisuse per se). This is important because treatment may bedifferentially effective according to the type of substancemisuse. Baker et al. (2009a), for example, found brief inter-ventions were effective for alcohol misuse but only somewhateffective for cannabis misuse in people with severe mentaldisorders. In this article, we systematically review the evidence

219A.L. Baker et al. / Journal of Affective Disorders 139 (2012) 217–229

from RCTs of psychological intervention for co-occurringalcohol misuse among people with unipolar depression,dysthymia or anxiety disorders and provide recommendationsfor clinical management and future research.

2. Method

The study search protocol included RCTs of psychologicalinterventions for co-occurring alcohol misuse among peoplewithmoodor anxietydisorders. Inclusion and exclusion criteriawere established prior to the literature search. Included studieswere required to employ diagnostic criteria formood (unipolardepression or dysthymia) or anxiety disorders; to utilize atreatment manual and to report data on alcohol use outcomes.Psychological interventions were operationalized as non-pharmacological treatments for either alcohol misuse alone oralcohol misuse and mood or anxiety disorders.

During March 2010, a systematic literature search wasconducted using the PubMed (ISI) and PsychINFO (CSA)databases. Individual searches in alcohol, depression andanxiety (search terms: depression, major depression, depres-sive disorder, anxiety, anxiety disorders, dysthymia, affectivedisorders, mood disorders) and treatment (search terms:treatment and therapy) were conducted, and were limited to‘human’ published ‘randomized controlled trials’(RCTs) or‘sequential allocation’ articles written in English (so as tomaximize methodological quality). Sequential allocation refersto the allocation of participants to treatment groups as theysequentially arrive in a treatment trial, allowing the trial toremain as balanced as possible throughout the recruitmentprocess (Atkinson, 2002). Itwas included in our search strategyas it was considered to be a potentially sound method of

Search Results (n=1540)

Studies matching topic opsychological treatments f

alcohol and depression oanxiety identified via abstr

(n=33)

Potentially relevant studies located by

review of reference lists (n=4)

Potentially relevant studies identified and full article reviewed

(n=34)

Studies included in literareview (n=8)

Fig. 1. Study select

treatment allocation, compatible with randomization. No datelimits were placed on the searches. The alcohol, depression,anxiety and treatment searches produced a total of 1540papers. By reviewing the titles, abstracts and reference lists, oneauthor (LT) identified 34 potentially relevant studies. Twoauthors (AB and SH) then independently reviewed thesearticles in full, and identified eight RCTs to be included in theliterature review. Inpatient studieswith samples of participantswith a variety of disorders and a larger proportion of patientswith psychotic disorders compared to affective disorders werenot included in the review. The full study selection process isshown in Fig. 1.

In addition to the narrative systematic review, effect sizes foralcohol use, depression and anxiety outcomes were comparedbetween studies and between treatment conditions. Effect sizeswere not compared for mental health outcomes due toheterogeneity in the patient samples used in the studies. Oneauthor (AB) extracted means and standard deviations of occa-sions of alcohol useperday, days of alcohol useperweek/month,days of heavy drinking, percentage of days abstinent fromdrinking, depression severity ratingsandanxiety severity ratingsfrom each article (when available), which were then indepen-dently checked by a second author (SH). Cohen's d pairedstandardized mean differences were computed comparingalcohol, depression and anxiety variables at baseline andimmediately following treatment (or equivalent timepoint forcontrol conditions) using Comprehensive Meta-Analysis. Calcu-lating apooled effect size (combining the studyeffect sizes of the8 studies), which would usually occur during a meta-analysiswas inappropriate, due to considerable heterogeneity in theclinical characteristics of participants, and the type of treatmentand measures used.

f or r act

Studies excluded because duplicate citations (n=3)

Studies excluded, with reasons (n=24): • Not a true RCT study design (n=12) • Alcohol use not a measured outcome

(n=5) • No depression or anxiety disorder

diagnosis (n=7) • Participants had additional

comorbidity (n=1) • No treatment manual (n=1)

ture

ion process.

Table 1Summary of depression and alcohol use outcomes from RCTs among samples with affective disorders and alcohol problems.

Study Sample Diagnoses (%) Entry criteriadepression/anxiety andalcohol

Design Results Clinical significanceof results

Methodologicallimitations

PEDro Scalequality rating(items v andvi blank)

Composite score(total 9)

Baker et al.(2009a,2009b)AustraliaComorbiddepressionand alcoholmisuse

N=284,OutpatientsMale=53.0%Mean age=45Participationrate: 284/433(65.6%); 149(34.4%)refused

DSMIV (SCID): 76%lifetime MD77% lifetime AUD

BDIN16At least 4 standarddrinks perday for men, 2 forwomenConcurrentmedication (61%)and substance use(13% usedcannabis at leastweekly) notexcluded

BI 1×60minutesessionalone or plus 960minutesessions of MI/CBTdepression,alcohol or integratedfocus.Psychologistsdeliveredmanual guidedintervention.Follow up 18 wksafterbaseline: 238/284(84%).Collateral reportsverifiedself-reported alcoholconsumption.

Percentage oftreatment sessionsattended: 86% for theBI; 16% none; 28% 1–4; 17% 5–8; 39% 9–10.Of those offered 10sessions, mean no. ofsessions attendedwas 5.76, with nodifferences betweenconditions.Depression:significant reductionfor all conditions, nodifferences betweenconditions; trend forintegrated focus to bemore effective thansingle focus. Alcohol:10 sessions moreeffective than BI;integrated moreeffective than singlefocused; for men,alcohol focus moreeffective thandepression focus.

Mean BDI changescore for integrated vs.single focus conditionwas 11.5 vs. 8.2.Alcohol: 10 sessionsassociated with 22.9vs. 10.8 drinks perweek reduction and1.3 vs. 0.6 reduction indrinking days.Integrated focusassociated withreduction in 1.83 vs.0.9 drinking daysreduction. Menreported meandrinking reduction of4.6 vs. 0.34 drinks perday in alcohol vs.depression focus.Thus, BI and extendedtreatment helpful fordepression. Integratedor alcohol focus bestfor men, integratedfocus best for women.

Only short-termoutcome reported.14% of sample werestill in treatment at18-week follow-up.

111?––10111 5

Kay-Lambkinet al. (2009)AustraliaComorbiddepressionand alcoholand/orcannabismisuse

N=97,OutpatientsMale=46.0%Mean age=35Participationrate: 97/116(83.6%); 19(16.4%) refused

DSMIV (SCID-RV)100.0% lifetime MDAUD Not reported

BDIN16At least 4 standarddrinks per day formen, 2 for women(n=52/97 metalcohol criteria;53.6%); or at leastweekly use ofcannabis (69/97;71.1%) Concurrentmedication notexcluded (% notreported)

BI 1×60minutesessionaloneorplus960minute sessions ofMI/CBT psychologistor computer-delivered MI/CBT(with brief weeklyinput from apsychologist).Manual guided.Follow up 12ms:41/52 (78.9%)

Percentage oftreatment sessionsattended: 87% for thetherapist-deliverygroup, mean of 9/10sessions; 76.1% for thecomputer-deliverycondition, mean of8/10 sessions.Depression:significant reductionfor all conditions;MICBT more effectivethan BI. Alcohol:Significant reductionfor all conditions, nodifferences betweenconditions.

% Improved at12 months:Depression (%BDIb17): BI (30.4%);psychologist (50%);computer (63%).Alcohol (N49%improvement inalcohol consumption):BI (53.8%);psychologist (82.4%);computer (73.3%).Thus, BI helpful fordepression andalcohol problems,with additionalimprovementfollowing longerintervention.

Use of only onetherapist.Small sample (97) ofwhom52met alcoholentry criteria.Therapy adherenceand fidelity not rated.Reliance on self-report measures.

1111––11111 9

220A.L.Baker

etal./

Journalof

Affective

Disorders

139(2012)

217–229






Markowitz etal. (2008)USAComorbiddysthymicdisorderand AUD

N=26OutpatientsMale=69.0%Mean age: 38Participationrate: Not stated.Ss were excludedfrom aconcurrent trialofpsychotherapyfor dysthymicdisorder withoutsubstance use(Markowitz etal., 2005).

DSMIV (SCID-NP)100% primarydysthymic disorder;54% current MD;77% dysthymicdisorder with earlyonset; 77% Axis IIdisorder.AUD (lifetime): 77%dependence; 23%abuse.

Primary DSMIVdysthymic disorderwith early onset(before age 21),Score N13 on HAM-D; GAFb61; andDSMIV alcoholabuse secondary todysthymic disorder.Concurrentpsychoactivemedications andsubstance abuse ordependenceexcluded except forcannabis abuse (%not reported).

16–1850-minsessions over16 wks manualguided IPT-D vs. BSPdeliveredindividually bytherapists withMSW or PhDdegrees.Post-treatmentassessment at wk16: 18/44 (40.9%).Breathalyzerverified alcoholconsumption.

43% of IPT and 42%of BSPSs failed to completetx. Mean no. ofsessions notreported.Depressionimprovedsignificantly for thesample as a whole.IPT significantlybetter than BSP onBDI at 16 wks.High% of Ss reportedabstinence in themonth before tx .

BDI: IPT: 18.9 (pre);8.9 (16 wk). BSP:25.1 (pre); 20.1(16 wk). ES:−1.38 vs−0.64.Mean% daysabstinent: IPT 40.4%(pre); 47.0% (16 wk).BSP: 32.1% (pre);49.7% (16 wk). ES:0.21 vs 0.54.

Small sample.High level of baselineabstinence.High level of txdropout.No follow-up.Therapy adherencedeemed adequate;fidelity ratings notreported.

1111––10111 8

Hulse andTait (2002)AustraliaPsychiatryinpatientswithalcoholmisuse

N=120VoluntarypsychiatryinpatientsMale=54.2%Mean age: 32Participationrate: 120/144(83.3%) of thosewho met thestudy criteriaand were askedto join the study.

DSMIV (fromhospital records)62.5% mood 15.8%anxiety 10.0%psychotic 11.7%%otherAUDIT Hazardous28.3% Harmful 42.5%Dependent 29.2%

New inpatients onpsychiatric wards.AUDIT score N7,SADQ scoreb30Concurrentsubstance abuse notreported.

45 minuteindividuallydelivered MIfollowing a templateby nurses or clinicalpsychologists vs.informationpackage.6/12 follow-up:83/120 (69.2%)

Both conditionsreduced alcoholconsumptionsignificantly.The MI conditionhad a significantlygreater change inweekly alcoholconsumption and agreater proportionwas improved.Mental healthoutcomes notreported.

The difference inalcohol consumptionper week betweenconditions was justover 3 drinks perweek, which theauthors describe asclinically meaningful.

Subjects had non-severe mentaldisorders admitted togeneral hospital unitsand includedmultipledisorders. Theintervention cannotbe separated frompsychiatric tx. Notherapy adherence orfidelity ratings. Largeloss to follow-up.Reliance on self-report measures.

11?0––10111 6

Santa Ana et al.(2007)USAPsychiatryinpatientswith SUD

N=10160 voluntary and41 non-voluntary non-psychoticpsychiatryinpatientsMale=62.4%Mean age: 37Participationrate: 101/211(47.8%) metinclusioncriteria.

DSMIV (interviewnot specified): 78.2%MD/mood 13.9%Bipolar 3%Schizoaffective 2%BorderlineAlcohol dependence67%

Current DSMIV SUDand Axis I disorderother than dementiaor psychosis

2×120 minutegroup MI sessionsfollowingmanual vs.2×120 minutetherapist attentionactivity controlgroup conducted bya psychologist.1/12 follow-up:97/101 (96%); 3/12follow-up 87/101(86%)Collateralinterviews verifiedself-report.

No differencebetween conditionsin proportionattending aftercareor attainingabstinence.MI had morefavorable drinkingoutcomes at 1- and3-months.Mental healthoutcomes notreported.

At 3-months MIgroups reportedmoreaftercare attendance(mean of 21.1 vs. 10.7occasions), fewerstandard drinks(mean 117.3) vs.control condition(262.3) and a lowerproportion bingedrinking (34.1% vs.55.8%).

Use of only onetherapist.Alternate groupassignmentfollowing initialrandomization.

1101––11011 7

Schade et al.(2005)NetherlandsComorbidagoraphobiaor

N=9665 from aninpatient clinicand 31 from anoutpatient clinicfor alcohol

DSMIV (SCID-1/P):66.6% social phobia7..3% agoraphobia26% bothAlcohol dependence100%

DSMIV diagnosis ofalcohol dependenceand comorbidagoraphobia orsocial phobiaConcurrent

Alcohol tx programalone (if aninpatient, groupsessions for 25 h aweek over 12–16 weeks, plus

Average no. ofsessions attendedwas 9. 15/47 (31.9%)used fluvoxamine.Significantimprovement in

32 wks:Improvement onFear Qr total meanfrom 40.1 to 32.7 inthe 27% and 38% inalcohol treatment vs.

Optionalfluvoxamineincluded in txcondition butunrelated tooutcome.

1110––11111 8

(continued on next page) 221A.L.Baker

etal./

Journalof

Affective

Disorders

139(2012)

217–229

Table 1 (continued)






social phobiaand alcoholdependence

dependenceMale=67.7%Mean age: 42Participationrate: 96/157(61.1%). Of 157interviewed, 31were excluded,30 refused.

substance abuseallowed, medicationexcluded.

weekly follow-upfor up to 32 weeks;if an outpatientweekly individual orgroup tx for10 weeks) vs.alcohol treatmentprogram combinedwith anxiety txcomprising manualguided CBT (12weekly 60-minindividual sessions)and optional SSRI.CBT was conductedby experiencedtherapists.Follow-up: 16 wks:84/96, (87.5%);24 wks: 74/96(77.1%); 32 wks:64/96 (66.7%).CDT levels on bloodchecked self-reportvalidity.

anxiety outcomesfor both conditions,significantly betterimprovement in theanxiety tx group.No difference inrelapse orabstinence betweeninterventionconditions.

44.1 to 21.9 inalcohol and anxietytx respectively. Daysheavy drinking (lastmth) fell from 14.8 atbaseline to 7.7 in thealcohol tx conditionand from 19.5 to 5.9in the alcohol andanxiety tx condition.32 wks: Abstinence27% and 38% inalcohol tx vs. alcoholrespectively. Daysheavy drinking (lastmth) fell from 14.8 atbaseline to 7.7 in thealcohol tx conditionand from 19.5 to 5.9in the alcohol andanxiety tx condition.

Toneatto(2005)CanadaComorbidagoraphobiaand alcoholdependence

N=14outpatientsMale=42.9%Mean age: 41Participationrate: 14/19completedtreatment(73.7%) andwere included inthe study.

DSM-IIIR (SCID):Panic disorder withagoraphobia andalcohol dependence100%Lifetime MD (35.7%)Lifetime dysthymia(35.7%) Lifetimesocial phobia(28.6%) Axis II (50%)

DSM-IIIR Panicdisorder withagoraphobia andalcohol dependenceConcurrentsubstancedependenceexcluded;medication notstated

10 individualsessions of eithercognitive therapy(addressingdysfunctionalcognitionsmediating thealcohol and anxietyproblems) vs.behavior therapy (5sessions treatingalcohol dependenceand 5 treating theanxiety problems).Manual guided.Therapists notdescribed.

Mean no. of sessionsattended notreported. 5/19dropped out beforesession 3.Both conditionsequally successful inreducing alcoholconsumption andanxiety at post-treatment and1-yearfollow-up.

5/14 (36.8%) met lowrisk drinking criteriaand 3/14 (21.4%) metmedium riskdrinking criteria atboth post-treatmentand follow-up.6/13 (46%) wereimproved on severalclinical dimensions ofanxiety at post-treatment and 4/12(33%) at follow-up.

Small sample size.Self-reportedoutcomes.

11?1––?0011 5

222A.L.Baker

etal./

Journalof

Affective

Disorders

139(2012)

217–229






Follow up 1 year:12/14 (85.7%)

Randall et al.(2001)USAComorbidsocialphobia andalcoholdependence

N=93OutpatientsMale=69.0%Mean age:38Participation rate:Of 187 screened,110 met inclusioncriteria, 17 ofthese were laterexcluded fromanalysis, leaving93/110 (84.5%).

DSMIII-R (SCID):Social phobia andalcohol dependence100%

DSMIII-R Socialphobia and alcoholdependenceConcurrent substancedependence excludedexcept for cannabis;medication notstated.

12 individual sessionsof CBT over amaximum of14 weeks focusing onalcohol problemsonly (60 mins persession) alcoholproblems plus socialphobia, ‘dual’(90 minsper session).Manual guided.Conducted by clinicalpsychologists.Follow-up at 3 mthintervals for 9 mthsafter tx.Breathalyzer andcollateral reportsverified self-reports.

Average no. ofsessions attendedwas 8.Both groupsimprovedsignificantly frombaseline on all alcoholoutcome measures,with the alcohol onlygroup significantlysuperior to the dualcondition at 3-mthfollow-up on all 3drinking variablesBoth conditionssignificantlyimproved on all socialanxietymeasures andBDI, with nodifferences betweengroups.

Authors state thatamount ofimprovement onsocial anxiety scoreswas modest (around20% from baselinescores) and averagepost-treatment scoresindicated significantimpairment.% days heavy drinkingreduced from around50% at baseline toaround 12% for thealcohol condition and25% for the dualcondition.

Social phobia plusalcohol conditionreceived moretreatment and lessreview of homeworkmaterial in order toachieve equivalenttime discussingalcohol problems.Only data of the firstfollow up werereported.

11?1––?1011 6

Physiotherapy EvidenceDatabase (PEDro) quality rating items: (i) eligibility criteriawere specified; (ii) participants allocated randomly to groups; (iii) allocation concealed; (iv) groups similar at baseline onmain prognostic signs;(vii) blinding of assessors who measured at least one key outcome; (viii) adequacy of follow-up; (ix) intent-to-treat analysis; (ix) between group statistical comparison of outcomes; (xi) study gives both point estimates andvariability for an outcome [64]. A score of 1=meets criteria, 0=does notmeet criteria, and?=unclear frommanuscriptwhether studymeets criteria or not. Two PEDro items regarding blindingof subjects (itemv) andblinding oftreatment providers (item vi) were not scored, as blinding is not feasible in this type of psychological intervention study. Composite score from PEDro quality ratings (range=0–9).PE: psychoeducation.MI: motivational interview.SI: standard psychiatric interview.IPT-D: interpersonal psychotherapy adapted for dysthymic disorder.BSP: brief supportive psychotherapy.CBT: cognitive-behavior therapy.AUD: alcohol use disorder.SUD: substance use disorder.Dep: depression.MD: major depression.ES: effect size.BDI: Beck Depression Inventory.SCID: Structured Clinical Interview for DSM.SCID-RV: Structured Clinical Interview for DSM, Research Version.SCID-NP: Structured Clinical Interview for DSM, Non-patient Version.SCID- 1/P: Structured Clinical Interview for DSM, Patient Version.Mth: month.SSRI: selective serotonin reuptake inhibitor.Wk: weeks.AUDIT: Alcohol Use Disorders Identification Test.SDQ: Severity of Alcohol Dependence Questionnaire.CDT: Carbohydrate-deficient transferring (CDT).

223A.L.Baker

etal./

Journalof

Affective

Disorders

139(2012)

217–229


The quality of the studies was assessed using the validatedPhysiotherapy Evidence Database (PEDro) scale (Centre forEvidence-Based Physiotherapy, 2009). PEDro scores are calcu-lated by assessing whether a study has (i) specified participanteligibility criteria; (ii) allocatedparticipants randomly togroups;(iii) concealed allocation; (iv) used groups similar at baseline onmain prognostic signs; (v) employed blinding of assessors whomeasured at least one key outcome; (vi) had adequate follow-ups; (vii) used intent-to-treat analysis; (viii) employedbetweengroup statistical comparison of outcomes; and (ix) given bothpoint estimates and variability for outcomes (Centre forEvidence-Based Physiotherapy, 2009). Consistent with therecent study by Spring et al. (2009), two items regardingblinding of subjects and therapists were not scored in thepresent review, as these were not feasible for the interventionsstudied. Two raters (AB and SH) independently rated the eightRCTs on the PEDro scale and reached consensus on the ratings(maximum score of 9).

3. Results

3.1. Trials of psychological interventions

Eight RCTs have reported alcohol use outcomes followingmanual-led psychological interventions for alcohol misuseamong peoplewithmood or anxiety disorders. These comprisetwo trials among samples with depression, one in a samplewith dysthymia, two among inpatient samples with mixeddiagnoses, one in a sample with social phobia, one in a samplewith social phobia or agoraphobia and one in a sample withagoraphobia or panic disorder. Details of these studies andPEDro scores are provided in Table 1, including the percentageof each sample meeting diagnostic threshold or entry criteriaon questionnaire measures.

3.2. Alcohol misuse among people with mood disorders

3.2.1. Alcohol misuse and depressionKay-Lambkin et al. (2009) reported the results of a RCT

designed to evaluate computer- versus therapist-deliveredpsychological treatment among 97 people with comorbiddepression and substance misuse, over half of whom hadalcohol misuse. All participants received an initial integratedsession comprising MI and case formulation for depressivesymptoms and substance use problems, followed by randomassignment to one of three treatments: no further treatment(brief intervention); nine further sessions of MI and CBTdelivered by a psychologist (therapist condition); or ninefurther sessions of MI/CBT therapy delivered by a computer(with brief 10–15 minuteweekly input fromapsychologist). Asdetailed in Table 1, all treatment conditions were associatedwith a significant reduction in alcohol consumption as well assymptomsof depression,withgreater benefits observedamongthe longer treatment conditions. Conclusions that can bedrawnfrom this study are limited by its small sample size and absenceof therapy adherence and fidelity ratings.

Baker et al. (2009b) extended this work to compare theeffectiveness of integrated brief intervention to single-focused(depression versus alcohol) and integrated MI/CBT among 284people with co-occurring depression and alcohol misuse. Asseen in Table 1, superior alcohol use outcomes for CBT relative

to brief interventions were found, but depression and globalfunctioning outcomes were equivalent at 18 weeks follow up.Gender differences between alcohol- and depression-focusedtreatments were reported, with males responding better toalcohol-focused and females better to depression-focusedtreatment over the short-term. Both genders responded tointegrated intervention which was found to be superior tosingle focused treatment in terms of depression and daysdrinking. The authors suggested that stepped care approachesare worthy of further investigation. In stepped-care ap-proaches, all patients receive low intensity treatment (e.g.,brief integrated interventions) as a first step, progress ismonitored, and patients who do not respond sufficiently arestepped-up to receive a treatment of higher intensity and/orlonger duration. The lack of long-term follow-up in this study isa limitation (but is currently underway) and only partialrecovery (as seen in Table 1) was achieved by manyparticipants, indicating that a stepped care approach in whichtreatment is delivered until an improvement threshold isreached, may be helpful.

3.2.2. Alcohol misuse among people with dysthymiaMarkowitz et al. (2008) conducted a small pilot study

(N=26) comparing 16 weeks of interpersonal psychotherapy(IPT) with 16 weeks of brief supportive psychotherapy amongpeoplewith co-occurringdysthymicdisorder and alcohol abuseor dependence. While depressive symptoms improved signif-icantly within each condition at 6 months follow up, IPTachieved significantly better outcomes than brief supportivepsychotherapy on this measure, with a large versus moderateeffect size. Conversely, brief supportive psychotherapy and IPThad moderate and small effect sizes respectively, for thepercentage of days abstinent from alcohol. Results were thusnot encouraging regarding the effectiveness of either IPT orbrief supportive psychotherapy for co-occurring dysthymia andalcohol misuse. There are numerous limitations to this study,including the small sample size, a high level of abstinence atbaseline, high treatment dropout and the absence of longerterm follow-up.

3.2.3. Alcohol misuse among mixed psychiatric inpatient samplesHulse and Tait (2002) assessed the effectiveness of a 45-

minute single session template-guided MI compared to aninformation package control (safer alcohol consumptionpatterns and normative feedback) among 120 hospitalizedpsychiatric patients (mainly mood and anxiety disorders) with(non-dependent) alcohol misuse. Both groups reported signif-icant reductions in alcohol use at 6 months. The MI conditionwas significantly better than the information condition in termsof lowering weekly alcohol consumption, as well as theproportion of drinkers who improved. The authors concludedthat brief interventions, particularly MI, are effective inreducing alcohol consumption among people with mentaldisorders. There were a number of limitations to this study,listed in Table 1, themost serious being the large loss to follow-up. The results reported were limited to the 69% of subjectsretained at 6 months and planned analyses of the 12 monthdata were abandoned due to the high level of attrition (53%were followed up at 1 year).

Santa Ana et al. (2007) compared the effectiveness of agroup MI consisting of two 2-hour sessions with an attention


control group condition among 101 non-psychotic inpatients(over three quarters with depression) in a psychiatric hospital.Outcomes assessed included the level of aftercare attendanceand alcohol consumption at 1 month follow up. There were nodifferences between conditions in terms of the proportion ofpeople attending aftercare or the rate of abstinence (50.0% forMI vs. 34.9% control). However, there were benefits of MI interms of number of aftercare attendances, number of standarddrinks consumed and fewer participants reporting bingedrinking (see Table 1). The main limitations of the study werethe use of one therapist (possible confounding due to therapisteffects), the lack of treatment fidelity ratings as well as thealternate randomization of groups following the initial ran-domization. The authors concluded that the study providespreliminary evidence of the effectiveness of MI in enhancingaftercare attendance and reducing drinking.

3.3. Alcohol misuse among people with anxiety disorders

3.3.1. Alcohol misuse and social phobiaRandall et al. (2001) randomly assigned 93 people with

comorbid social phobia and alcohol dependence to either 12individual sessions of CBT focusing on alcohol or bothconditions (‘dual’ condition). As described in Table 1, the lattercondition received more time in therapy and less homeworkthan the alcohol focused condition. Both conditions wereassociatedwith significant reductions in alcohol and significant,albeit modest, improvements in anxiety at 3 months posttreatment. The alcohol condition was associated with betteroutcomes on three alcohol indices compared to the dualcondition. The short-term nature of the follow-up limits theconclusions that can be drawn from the study, as it is possiblethat there may have been delayed improvement in the ‘dual’condition. The authors suggested that consideration be given tothe staging of treatments, rather than simply treating bothdisorders (Randall et al., 2001). They recommended that whileit is important to treat comorbid disorders, the order in whichthis should be done and the degree of integration (versusadjunctive or parallel approaches) should be the subject offurther study.

Schade et al. (2005) randomly assigned 96 abstinentindividuals with alcohol dependence and comorbid agorapho-bia or social phobia to either inpatient or outpatient treatmentas usual for alcohol use, or to usual alcohol treatment plusparallel CBT for anxiety (12weekly 60-min individual sessions)and optional pharmacotherapy (a selective serotonin reuptakeinhibitor [SSRI]). The additional therapywas significantly betterin terms of improving anxiety symptoms, but no differencebetween conditions on alcohol use outcomes were found atfollow up. Limitations of the study include the inclusion ofinpatients and outpatients with mixed phobias, as well as theinclusion of optional SSRIs in treatment.

3.3.2. Alcohol misuse and panic disorder/agoraphobiaToneatto (2005) conducted a small pilot trial in which 14

people with comorbid alcohol dependence and agoraphobiawere randomly assigned to either behavior therapyor cognitivetherapy. Both treatments consisted of 10 individually admin-istered sessions of cognitive therapy focused on dysfunctionalcognitions mediating the alcohol and anxiety conditions, orsessions of behavior therapy focused on the treatment of

alcohol dependence in the first five sessions and then anxietyfor the remainder of treatment. Both interventions wereequally effective in reducing drinking and anxiety symptomsat 12 months follow up. Interestingly, anxiety symptomssignificantly improved during the five alcohol-focused sessionsof the behavior therapy condition, leading Toneatto (2005) toconclude that brief behavior therapy might be an effectivetreatment for this specific comorbidity. Although this studywaslimited by its small sample size and reliance on self-reportmeasures, the integrated nature of the treatments (addressingboth alcohol and anxiety problems) and the initial focus onalcohol in the behavioral condition supports a staged approach.

3.3.3. Comparison of effect sizes for changes in alcohol use anddepression/anxiety symptoms immediately following treatment

Hulse and Tait (2002) did not measure outcome variablesimmediately after treatment and were excluded from theanalysis. The seven remaining studies for alcohol use outcomesare compared in Fig. 2. All conditionsdemonstrate standardizedmean differences in the expected direction (decrease foroccasions of use/average number of drinking days/heavydrinking days and increase in percentage days abstinentfollowing treatment). Several conditions resulted in largechanges of at least one standard deviation, including thetherapist-delivered MI/CBT for people with depression (Kay-Lambkin et al., 2009), group MI for mixed psychiatric samples(Santa Ana et al., 2007), parallel CBT for alcohol and anxiety(Schade et al., 2005) and targeted alcohol focused CBT (Randallet al., 2001) for people with anxiety disorders. These largechanges in alcohol outcome were found despite the likelihoodof the presented effect sizes being underestimates of the truepopulation effect size. Most other treatment conditionsgenerated approximately half a standard deviation of change,including the brief interventions. Standardized mean differ-ences for depression and anxiety severity scores, whenreported, are compared in Fig. 3. There were large decreasesof over one standard deviation in depression severity scores inseveral studies. Decreases in anxiety severity scores weresmaller, although tended to show decreases in similarmagnitude to the alcohol use variables.

3.3.4. Limitations of psychological intervention trialsPEDro scale scores ranged from 5 to amaximum of 9 for the

studies analyzed, with the most common limitations (4/8studies) being: failure to obtain more than 85% of subjectsinitially allocated to groups at follow-up; lack of intention totreat analysis (3/8 studies); lack of clarity regarding conceal-ment of allocation (3/8 studies); and dissimilar baselinecharacteristics (3/8 studies). In addition, as Table 1 shows, thestudies suffered from a number of other methodologicallimitations including: short-term follow-up (4/8 studies);recruitment of heterogeneous samples with different mentaldisorders (3/8 studies, one of which recruited both inpatientsand outpatients); reliance on self-reported alcohol consump-tion (3/8 studies); and small sample sizes (2/8 studies). Amongthe mood studies, there was variability in the use of diagnosticinterviews versus symptom severity ratings with diagnosticcut-off scores to determine eligibility for inclusion in the study.While the anxiety studies all included diagnostic assessmentsof both anxiety and alcohol misuse, samples included mixedanxiety diagnoses.Nevertheless, theeightRCTs reviewedabove

Occasions of use per day

*Baker et al depression focused MI/CBT

*Baker et al alcohol focused MI/CBT

*Baker et al integrated alcohol/depression MI/CBT

*Baker et al brief intervention (control)

*Kay-Lambkin et al therapist delivered integrated alcohol/depression MI/CBT

*Kay-Lambkin et al computer delivered integrated alcohol/depression MI/CBT

*Kay-Lambkin brief intervention (control)

Average number of days drinking





Santa Ana et al alcohol focused group MI

Santa Ana et al therapist attention activity (control)

Heavy drinking days (> 5 drinks)

Santa Ana et al alcohol focused group MI

Santa Ana et al therapist attention activity (control)

^Schade et al alcohol focused CBT

^Schade et al integrated alcohol/anxiety CBT

Percentage days abstinent

*Markowitz dysthymia focused interpersonal psychotherapy

*Markowitz brief supportive therapy (control)

^Randall et al alcohol focused CBT

^Randall et al integrated alcohol/anxiety CBT

^Toneatto sequence of alcohol then anxiety behavioural therapy

^Toneatto integrated alcohol/anxiety cognitive therapy

nameStudy

0-3 -2 -1 1 2 3

Standardized mean difference (d, 95% CI)

Fig. 2. Paired standardizedmean difference (Cohen's d, ±95% CI) comparing baseline vs. immediate post treatment alcohol outcome variables, stratified by alcoholoutcome measure. A negative mean difference represents a decrease in scores from pre- to post-treatment. A positive mean difference represents an increase inscores from pre- to post-treatment. * denotes study examining comorbid alcohol/depressive disorder sample. ^ denotes study examining comorbidalcohol/anxiety disorder sample. Unmarked studies involved mixed affective disorder samples. CBT: cognitive behavioral therapy. MI: motivational interviewing.


provide unique information about people with comorbidmood/anxiety disorders and alcohol misuse and the effects ofmanual guided treatment.

3.4. Summary of findings of psychological intervention trialsand suggestions for further research

3.4.1. Alcohol misuse and depressive disordersOverall, two studies suggest that co-occurring depression

and alcohol misuse are responsive to psychological treatment,including brief integratedMI interventions and longer durationCBTof up to ten sessions (Baker et al., 2009b;Kay-Lambkin et al.,2009). Additional benefits of the longer over brief integratedinterventions were seen on both depression and alcoholoutcomes, with both studies suggesting that stepped care isworthy of further investigation. When stepping up to a longerintervention, integrated CBT interventions appear to be suitablefor both men and women (Baker et al., 2009b) and can be

delivered effectively by a therapist or a computer (Kay-Lambkinet al., 2009). IPT and brief supportive psychotherapy, based onthe results of one study, are yet to demonstrate effectiveness inboth mood and alcohol use domains. The brief integrated MIintervention and CBT in the trials conducted by Kay-Lambkin etal. (2009) and Baker et al. (2009b) were delivered bypsychologists in research clinics, and further research isrequired to determine if these interventions can effectively betranslated into practice in real world clinical settings. Heather(1995; 1996) has pointed out that the effectiveness of briefinterventions among treatment seekers, often delivered bycounselors, may not generalize to non-treatment seekers whomay receive brief interventions in a general health care setting.In addition, as the sample included in Kay-Lambkin et al.'s(2009) study comprised of cannabis users as well as problemdrinkers, a direct comparison of brief integrated interventionand computer therapy among participants with comorbiddepression and alcohol misuse is required.

Depressive symptoms (BDI)





*Kay-Lambkin et al therapist delivered integrated alcohol/depression MI/CBT

*Kay-Lambkin et al computer delivered integrated alcohol/depression MI/CBT

*Kay-Lambkin brief intervention (control)

*Markowitz dysthymia focused interpersonal psychotherapy

*Markowitz brief supportive therapy (control)

Anxiety behaviors, cognitions and somatic symptoms (SPAI)

^Randall et al alcohol focused CBT

^Randall et al integrated alcohol/anxiety CBT

Anxiety behaviors (Mobility Inventory - Alone)



Anxiety cognitions (ACQ)



Anxiety somatic symptoms (BSQ)



nameStudy

0-4 -3 -2 -1 1 2 3Standardized mean difference (d, 95% CI)

Fig. 3. Paired standardized mean difference (Cohen's d, ±95% CI) comparing baseline vs. immediate post treatment depression and anxiety outcome variables. Anegative mean difference represents a decrease in scores from pre- to post-treatment.* denotes study examining comorbid alcohol/depressive disorder sample.^ denotes study examining comorbid alcohol/anxiety disorder sample. ACQ: Agoraphobic Cognitions Questionnaire; BSQ: Body Sensations Questionnaire(Chambless et al., 1984). BDI: Beck Depression Inventory (Beck, 1993). CBT: cognitive behavioral therapy. Mobility Inventory (Chambless et al., 1985). MI:motivational interviewing. SPAI: Social Phobia and Anxiety Inventory (Turner et al., 1989).


Individual (Hulse and Tait, 2002) and group (Santa Anaet al., 2007) MI among psychiatric hospital inpatients,primarily with depression, have been shown to be effectivein reducing alcohol consumption at follow-up. However, theresults were modest in the study by Hulse and Tait (2002)and replication is needed. It is possible that inclusion ofnurses as therapists may have weakened the results andfurther investigation of the influence of therapist character-istics in the effectiveness of MI in psychiatric settings iswarranted. The Santa Ana et al. (2007) study had goodoutcomes with the inclusion of inpatients with alcoholdependence, suggesting that the exclusion of inpatientswith alcohol dependence in the study by Hulse and Tait(2002) is not indicated and indeed, may have resulted in afloor effect (i.e., reduced the likelihood of potentially greaterimprovements in drinking being observed).

3.4.2. Alcohol misuse and anxiety disordersAs with comorbid depression and alcohol misuse, existing

studies suggest that co-occurring anxiety disorders and alcoholmisuse are responsive to psychological treatment, including

brief behavioral interventions focusing on alcohol (Toneatto,2005). Two of the three studies recommended a stepped orstaged approach to treatment (Randall et al., 2001; Toneatto,2005), indicating a need for future studies on stepped care.While the study by Schade et al. (2005) found a significantreduction in anxiety symptoms among a mixed sample ofparticipants with social phobia and agoraphobia, the level ofanxiety reduction in the social phobia group was modest incomparison to studies comprised mostly of participants withagoraphobia (Schade et al., 2005; Toneatto, 2005). Thissuggests that replication of these studies with larger unmixedsamples of subjects is needed. It is possible that differentanxiety disorders respond differentially to alcohol- versusanxiety-focused or integrated interventions. As no studies haveyet compared the effectiveness of anxiety, alcohol andintegrated focused interventions, research targeting groupswith specific anxiety disorders would clearly be informative.Such studies could potentially allow an evaluation of a steppedapproach to further intervention among those who requiretreatment in the anxiety and/or alcohol domain. The possibilityof gender differences in treatmentoutcomes among individuals


with comorbid anxiety and alcohol misuse is also worthy ofinvestigation, as differential treatment effects by gender werefound in one study of co-occurring depression and alcoholmisuse (Baker et al., 2009b). As pointed out by Randall et al.(2001), there is also a need to investigate the effectiveness ofdifferent levels of integrated treatment, for example, compar-ing the effectiveness of one practitioner versus more than onepractitionerwithin a service providing integrated treatment forco-existing disorders. Given its effectiveness for co-occurringdepression and alcohol misuse, the role of computer-deliveredtreatment for co-occurring anxiety and alcohol problems is alsoworthy of investigation.

3.4.3. Retention in treatment and follow-upAs seen in Table 1, completion of available treatment

sessions is uncommon and follow-up attendance diminishesmarkedly with time. It is possible that participants discontinuetreatment when they reach a desired level of improvement,hence adopting an informal stepped care approach. Ongoingmeasurement of mental health symptoms and alcohol con-sumption across the study period, as well as participants'expectations of treatment, would thus be informative, andshould include both quantitative and qualitative data. Inaddition to MI for behavior change, MI focused on increasingtreatment attendance may also be helpful for some. On theother hand, it is possible that among sampleswith co-occurringproblems, a return to a strictly defined ‘non-clinical’ level ofdepression, dysthymia, anxiety or alcohol consumption mightbe an unrealistic expectation during the typical timeframe ofclinical trials. The common clinical picture of an isolated,unemployed individual, with few resourcesmaymean that anystepped care approach may be more successful over the longerterm, with later improvement evident as social, vocational andother functioning improves.

Together, these findings indicate that MI/CBT for co-occurring alcohol misuse among people with depressive oranxiety disorders has strong effectiveness. However, thereremains room for improvement in the magnitude of changeachieved. Contingency management, where patients arerewarded for desirable behavior (e.g., abstinence), has demon-strated effectiveness among populations with co-occurringdisorders (Tidey et al., 2002; Tracy et al., 2007), and providesonepotential avenue for enhancing theeffectiveness of CBTandimproving attendance at treatment and follow-up.

4. Conclusions

This review highlights the limited research available toinform psychological treatment approaches for co-occurringalcohol misuse and depression or anxiety disorders. While thisreview may have been improved by conducting a morecomplete search of the literature (e.g., by accessing unpub-lished and/or non-English language studies), there is evidencethat psychological interventions (MI/CBT) are effective fortreating co-occurring mood or anxiety disorders and alcoholmisuse. Even brief interventions appear to be effective,although longer interventions are associated with greaterimprovements in mood and alcohol use outcomes. Furtherresearch should include larger, more homogeneous sampleswithmore frequent follow-up assessments over longer periodsof time.

Role of funding sourceAmanda Baker is supported by a NHMRC Fellowship (510702). Leanne

Hides is supported by a QUT Vice Chancellor's Senior Research Fellowship,while Louise Thornton and Sarah Hiles are supported by AustralianPostgraduate Awards. The funding sources provided no further role instudy design; in the collection, analysis and interpretation of data; in thewriting of the report; and in the decision to submit the paper for publication.

Conflict of interestDan Lubman has received speaker honoraria from Astra Zeneca, Bristol

Myers Squibb, Eli Lilly, Janssen, and Pfizer. All other authors declare that theyhave no conflicts of interest.

AcknowledgmentsNone.

References

Atkinson, A.C., 2002. The comparison of designs for sequential clinical trialswith covariate information. J. R. Stat. Soc. 165, 349–373.

Baker, A., Turner, A., Kay-Lambkin, F.J., Lewin, T.J., 2009a. The long and theshort of treatments for alcohol or cannabis misuse among people withsevere mental disorders. Addict. Behav. 34, 852–858.

Baker, A.L., Kavanagh, D.J., Kay-Lambkin, F.J., Hunt, S.A., Lewin, T.J., Carr, V.J.,Connolly, J., 2009b. Randomized controlled trial of cognitive–beha-vioural therapy for coexisting depression and alcohol problems: short-term outcome. Addiction 105, 87–99.

Baillie, A., Sannibale, C., 2007. Anxiety and drug and alcohol problems. In:Baker, A., Velleman, R. (Eds.), Clinical Handbook of Co-existing MentalHealth and Drug and Alcohol Problems. Routledge, London, pp. 197–217.

Beck, A.T., 1993. Beck Depression Inventory. Psychological Corporation, SanAntonio, TX.

Centre for Evidence-Based Physiotherapy, 2009. PEDro Scale. Centre forEvidence-Based Physiotherapy, Sydney.

Chambless, D.L., Caputo, G.C., Bright, P., Gallagher, R., 1984. Assessmentof fear inagoraphobia: the body sensations questionnaire and the agoraphobiccognitions questionnaire. J. Consult. Clin. Psychol. 61, 248–260.

Chambless, D.L., Caputo, G.C., Jasin, S.E., Gracely, E.J., Williams, C., 1985. TheMobility Inventory for agoraphobia. Behav. Res. Ther. 23, 35–44.

Degenhardt, L., Hall, W., Lynskey, M., 2001. Alcohol, cannabis and tobacco useamong Australians: a comparison of their associations with other druguse and use disorders, affective and anxiety disorders, and psychosis.Addiction 96, 1603–1614.

Dobson, K.S., 1989. A meta-analysis of the efficacy of cognitive therapy fordepression. J. Consult. Clin. Psychol. 57, 414–419.

Farrell, M., Howes, S., Bebbington, P., Brugha, T., Jenkins, R., Lewis, G.,Marsden, J., Taylor, C., Meltzer, H., 2001. Nicotine, alcohol and drugdependence and psychiatric comorbidity: results of a national householdsurvey. Br. J. Psychiatry 179, 432–437.

Gloaguen, V., 1989. A meta-analysis of the effects of cognitive therapy indepressed patients. J. Affect. Disord. 49, 59–72.

Grant, B., Stinson, F.S., Dawson, D.A., Chou, P., Dufour, M.C., Compton, W.,Lamb, R., Fletcher, B., Pickering, R.P., Kaplan, K., 2004. Prevalence and co-occurrence of substance use disorders and independent mood andanxiety disorders. Arch. Gen. Psychiatry 61, 807–816.

Heather, N., 1995. Interpreting the evidence on brief interventions forexcessive drinkers: the need for caution. Alcohol Alcohol. 30, 287–296.

Heather, N., 1996. The public health and brief interventions for excessivealcohol consumption: the British experience. Addict. Behav. 21, 857–868.

Hesse, M., 2009. Integrated psychological treatment for substance use and co-morbid anxiety or depression vs. treatment for substance use alone. Asystematic reviewof the published literature. Biomed. Cent. Psychiatry 9, 6.

Hettema, J., Steele, J., Miller, W.R., 2005. A meta-analysis of research onmotivational interviewing treatment effectiveness (MARMITE). Annu.Rev. Clin. Psychol. 1, 91–111.

Hides, L., Samet, S., Lubman, D.I., 2010. Cognitive behaviour therapy (CBT) forthe treatment of co-occurring depression and substance use: currentevidence anddirections for future research. DrugAlcohol Rev. 29, 508–517.

Hofmann, S.G., Smits, J.A., 2008. Cognitive–behavioral therapy for adultanxiety disorders: a meta-analysis of randomized placebo-controlledtrials. J. Clin. Psychiatry 69, 621–632.

Hulse, G.K., Tait, R.J., 2002. Six-month outcomes associated with a briefalcohol intervention for adult in-patients with psychiatric disorders.Drug Alcohol Rev. 21, 105–112.

Kay-Lambkin, F.J., Baker, A., Lewin, T.J., 2009. Computer-based psychologicaltreatment for comorbid depression and substance use problems: arandomised controlled trial of clinical efficacy. Addiction 104, 378–388.


Kessler, R.C., Berglund, P., Demler, O.V., Jin, R., Koretz, D., Merikangas, K.E.,Rush, J.A., Walters, E.E., Wang, P.S., 2003. The epidemiology of majordepressive disorder: results from the national comorbidity surveyreplication (NCS-R). JAMA 289, 3095–3105.

Magill, M., Ray, L.A., 2009. Cognitive–behavioral treatment with adult alcoholand illicit drug users: a meta-analysis of randomized controlled trials. J.Stud. Alcohol Drugs 70, 516–527.

Markowitz, J.C., Kocsis, J.H., Bleiberg, K.L., Christos, P.J., Sacks, M., 2005. Acomparative trial of psychotherapy and pharmacotherapy for “pure”dysthymic patients. J. Affect. Disord. 89, 167–175.

Markowitz, J.C., Kocsis, J.H., Christos, P., Bleiberg, K., Carlin, A., 2008. Pilotstudy of interpersonal psychotherapy versus supportive psychotherapyfor dysthymic patients with secondary alcohol abuse or dependence.J. Nerv. Ment. Dis. 196, 468–474. doi:10.1097/NMD.0b013e31817738f1.

Mills, K.L., Deady,M., Proudfoot, H., Sannibale, C., Teesson,M.,Mattick, R., Burns,L., 2009. Guidelines on the management of co-occurring alcohol and otherdrug and mental health conditions in alcohol and other drug treatmentsettings. National Drug and Alcohol Research Centre, University of Sydney,Sydney.

Moyer, A., Finney, J.W., Swearingen, C.E., Vergun, P., 2002. Brief interventionsfor alcohol problems: a meta-analytic review of controlled investigationsin treatment-seeking and non-treatment-seeking populations. Addiction97, 279–292.

Norton, P.J., Price, E.C., 2007. A meta-analytic review of adult cognitive–behavioral treatment outcome across the anxiety disorders. J. Nerv.Ment. Dis. 195, 521–531.

Randall, C.L., Thomas, S., Thevos, A.K., 2001. Concurrent alcoholism and socialanxiety disorder: a first step toward developing effective treatments.Alcohol. Clin. Exp. Res. 25, 210–220.

Santa Ana, E.J., Wulfert, E., Nietert, P.J., 2007. Efficacy of group motivationalinterviewing (GMI) for psychiatric inpatients with chemical depen-dence. J. Consult. Clin. Psychol. 75, 816–822.

Saunders, J.B., Lee, N.K., 2000. Hazardous alcohol use: its delineation as asubthreshold disorder and approaches to its diagnosis and management.Compr. Psychiatry 41, 95–103.

Schade, A., Marquenie, L., Van Balkom, A., Koeter, M., De Beurs, E., Brink, W.,Van Dyck, R., 2005. The effectiveness of anxiety treatment on alcohol-dependent patients with a comorbid phobic disorder: a randomizedcontrolled trial. Alcohol. Clin. Exp. Res. 29, 794–800.

Spring, B., Howe, D., Berendsen, M., Mcfadden, H.G., Hitchcock, K., Rade-maker, A.W., Hitsman, B., 2009. Behavioral intervention to promotesmoking cessation and prevent weight gain: a systematic review andmeta-analysis. Addiction 104, 1472–1486.

Stewart, R.E., Chambless, D.L., 2009. Cognitive–behavioral therapy for adultanxiety disorders in clinical practice: a meta-analysis of effectivenessstudies. J. Consult. Clin. Psychol. 77, 595–606.

Sullivan, L.E., Fiellin, D.A., O'connor, P., 2005. The prevalence and impact ofalcohol problem in major depression: a systematic review. Am. J. Med.118, 330–341.

Tidey, J.W., O'neill, S.C., Higgins, S.T., 2002. Contingentmonetary reinforcementof smoking reductions, with and without transdermal nicotine, inoutpatients with schizophrenia. Exp. Clin. Psychopharmacol. 10, 241–247.

Toneatto, T., 2005. Cognitive versus behavioral treatment of concurrent alcoholdependence and agoraphobia: a pilot study. Addict. Behav. 30, 115–125.

Tracy, K., Babuscio, T., Nich, C., Kiluk, B., Carroll, K.M., Petry, N.M., Rounsaville,B.J., 2007. Contingency management to reduce substance use inindividuals who are homeless with co-occurring psychiatric disorders.Am. J. Drug Alcohol Abuse 33, 253–258.

Turner, S., Beidel, D., Dancu, C., Stanley, M., 1989. An empirically derivedinventory to measure social fears and anxiety: the Social Phobia andAnxiety Inventory. Psychol. Assess. 1, 35–40.

http://dx.doi.org/10.1097/NMD.0b013e31817738f1

Journal of Affective Disorders - Boston Universitysites.bu.edu › sswhrsaseminar › files › 2013 › 10 › Psychological...Journal of Affective Disorders 139 (2012) 217–229

Documents