View metadata, citation and similar papers at core.ac.uk brought to you by CORE provided by Nottingham Trent Institutional Repository (IRep) Psychological, social and welfare interventions for psychological health and well-being of torture survivors (Protocol) Patel N, Kellezi B, Williams ACDC ThisisareprintofaCochraneprotocol,preparedandmaintainedbyTheCochraneCollaborationandpublishedinTheCochrane Library2011,Issue10 http://www.thecochranelibrary.com Psychological,socialandwelfareinterventionsforpsychologicalhealthandwell-beingoftorturesurvivors(Protocol) Copyright©2011TheCochraneCollaboration.PublishedbyJohnWiley&Sons,Ltd. TABLE OF CONTENTS HEADER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 BACKGROUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 OBJECTIVES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 METHODS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 APPENDICES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 HISTORY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 CONTRIBUTIONSOFAUTHORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 DECLARATIONSOFINTEREST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 SOURCESOFSUPPORT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 Psychological,socialandwelfareinterventionsforpsychologicalhealthandwell-beingoftorturesurvivors(Protocol) i Copyright©2011TheCochraneCollaboration.PublishedbyJohnWiley&Sons,Ltd. [InterventionProtocol] Psychological, social and welfare interventions for psychological health and well-being of torture survivors NimishaPatel1,2,BlerinaKellezi1,3,AmandaCdeCWilliams4 1AuditandEvaluation,FreedomfromTorture,London,UK.2SchoolofPsychology,UniversityofEastLondon,London,UK.3Centre forCriminology,UniveristyofOxford,Oxford,UK.4ResearchDepartmentofClinical,Educational&HealthPsychology,University CollegeLondon,London,UK Contactaddress:AmandaCdeCWilliams,ResearchDepartmentofClinical,Educational&HealthPsychology,UniversityCollege London,GowerStreet,London,WC1E6BT,[email protected]@ucl.ac.uk. Editorialgroup:CochraneDepression,AnxietyandNeurosisGroup. Publicationstatusanddate:New,publishedinIssue10,2011. Citation: PatelN,KelleziB,WilliamsACDC.Psychological,socialandwelfareinterventionsforpsychologicalhealthandwell-beingof torturesurvivors.CochraneDatabaseofSystematicReviews2011,Issue10.Art.No.:CD009317.DOI:10.1002/14651858.CD009317. Copyright©2011TheCochraneCollaboration.PublishedbyJohnWiley&Sons,Ltd. ABSTRACT Thisistheprotocolforareviewandthereisnoabstract.Theobjectivesareasfollows: Primaryobjective 1.Toassessbeneficialandadverseeffectsofpsychological,socialandwelfareinterventionsversusnotreatmentforthereductionof psychologicaldistressintorturesurvivors. Secondaryobjectives 2.Todescribethequalityandgeneralisabilityofthestudiesevaluatingtheeffectsofthesetreatmentapproachesontorturesurvivors, andspecifically: • toprovideanobjectiveassessmentofriskofbiasinthesestudies; • todescribethespecificpopulationsevaluatedinstudiesoftorturesurvivors(includingdemographics,tortureexperiencesand psychologicalstatus); • todescribethevarietyofinterventionsthathavebeenevaluatedinthesepopulations;and • todescribetheoutcomesevaluatedintheseinterventionstudies. BACKGROUND 400,000torturesurvivorsliveintheEuropeanUnionalone,with similarestimatesintheUSA(Jaranson,1995). Reports of torture and other ill-treatment come from over 150 countries (AI 2010). The International Rehabilitation Con- Unlikemanyotherclientgroups,thehealthconcernsoftorture sortium for Torture Survivors (IRCT 2010) estimates around survivors are addressed in theliterature and in clinical practice, Psychological,socialandwelfareinterventionsforpsychologicalhealthandwell-beingoftorturesurvivors(Protocol) 1 Copyright©2011TheCochraneCollaboration.PublishedbyJohnWiley&Sons,Ltd. notspecificallyorsolelyintermsofcommonhealthproblems,but 2. Socialinterventionsaimtoincreasesocialinvolvementor bytheirexperienceoftortureandotherill-treatmentasdefinedby participationtoimprovesurvivors’healthandwell-being,and theUnitedNationsConventionagainstTortureandOtherCruel, maybedeliveredatindividualorgrouplevels.Thefocusofthe Inhuman or Degrading Treatment or Punishment (CAT) (UN interventionmayvary,fromeducationalinitiativestoredress. 1984),Article1. 3. Welfareinterventionsaimtobenefitsurvivors’healthand well-beingthroughimprovingsocialconditionssuchashousing Tortureisadeliberateassaultuponthebody,thepsyche,theiden- oraccesstohealthservices,andareusuallydeliveredatthelevel tity and the integrity of theperson, aiming to dehumanise, de- ofthecommunity. grade,destroyordebilitateandrendertheindividualhelpless.Its impact on psychological health can be short term and/or long- lastingandmayaffecttheindividual’sinterpersonalrelationships. Howtheinterventionmightwork Theimpactof torturecanextendbeyondtheindividual tohis/ herfamily,community andsocietybyperpetuatinguncertainty, Psychologicalinterventions,withorwithoutculturaladaptation, mistrust, suspicion and terror. Individual problems in physical, may target a specificproblemsuch asflashbacks to thetrauma, psychologicalandsocialhealthcanresultfrombothphysicaland orabroadspectrumofmooddisorders;interventionscaneffect psychological methods of torture, usually used in combination. change via a number of mechanisms, including exposure and/ The physical,psychological andother dimensions of well-being oremotionalprocessingwith/orwithoutcognitiverestructuring, interactandmanifestincomplexanddiverseways,impactingon cognitiveprocessing,meaningmaking,and/orinterpretation.Psy- the capacity of individuals to function in daily life within their chologicalinterventionsmayadditionallydrawonpolitical(asin familiesandcommunities.Additionally,torturesurvivorsincoun- documentation)oreducationalprinciples. triesofexilecanexperiencemanyadditionaldifficulties,impact- Socialinterventionscaneffectchangethroughreframingorrein- ingontheirwell-being(e.g.legalproceedings,racism,inadequate terpreting the relationships of the survivor with their family or housingorhomelessness). community.Theymayimprovesocialfunctioning,socialintegra- tionandparticipation,andrestoretrust.Socialinterventionsmay Physicalhealthproblemsrelatedtotorturehavebeenwidelydoc- alsodrawonpolitical(asinpublicrecognitionofwrong)anded- umented(Jacobs2001;Moreno2002;Norredam2005;forare- ucationalprinciples. view seeJaranson 2011; Montgomery 2011; Quiroga2005), as Welfare interventions aim to improve the material conditions havepsychologicalhealthproblems(e.g.Basoglu2001;Johnson (suchashousing)andenvironmentalandservicecontext(suchas 2008;Patel2007).Torture-relatedphysicalhealthproblemsnot healthcareandavailabilityofpurposefulactivity)whichareasso- onlycausedisabilityorrestrictedfunctioningbutcanalsoproduce ciatedwithmentalandphysicalhealthandwellbeing. additionalpsychologicalproblems,resultinginsignificantimpact onoverallsocialfunctioningandwell-beingoftorturesurvivors. Whyitisimportanttodothisreview Intheeraofevidence-basedhealthcare,thereisconsiderableem- Descriptionoftheintervention phasisonservicesprovidingtreatmentsdemonstratedtobeeffec- Thereislimitedinformationaboutthetypesofinterventionspro- tive. However,evidence for interventions with torture survivors videdtotorturesurvivors.Theavailableliterature(McIvor1995; comespredominantlyfromstudieswhichinvolveneithertorture Quiroga2005)pointstoarangeofinterventions,somepsycholog- survivorpopulations,norpopulationsdiverseincultural,ethnic, ical,somemedical,andmanyinvolvingavarietyofapproachesin- religiousandpoliticalbackgroundsandwhosefirstlanguageisnot tendedtoaddressthedifficultiesexperiencedbytorturesurvivors, English.Arguably,bothhealthcareprovidedtotorturesurvivors, includingwelfareissues,legalproceedingsrelatedtotheirasylum and the methodsand tools used to evaluate its relevance,effec- claims,housingandsocialintegration.Sincethereisnoevidenceof tivenessandimpact,arebasedonconceptsofhealthandwell-be- apsychologicalsyndromespecifictotorture(Westermeyer1998), ing,andpsychologicalmodelsandoutcometools,whicharenot the treatment literature inevitably focuses on the treatment of developedorvalidatedwithtorturesurvivors. Apreviousreview PTSD,asonediagnosis,amongsttorturesurvivors. found veryfewstudies hadbeenconducted, allwith significant 1. Psychologicalinterventionsaimtochangecognitive, limitations,includinglackofcontrolgroups,variableuseofdiag- emotionalorbehaviouraloutcomesand,inthispopulation,can nosticcriteria,lackofvalidationofthemeasuresused,andvery bedeliveredtotheindividual,family,group,orcommunity.The smallsamplesizes(Quiroga2005). interventionmaydrawonavarietyoftheoreticalandtherapeutic Mostoftheliteratureonpsychologicalandphysicalhealthdiffi- schools,butcanbegroupedintobroadmodelsincluding cultiesexperiencedbytorturesurvivors(beforeorwithouttreat- behavioural,cognitivebehavioural,integrative,humanisticand ment)isbasedonprofessionaloracademicaccounts,muchinthe psychodynamicpsychologicalinterventions. formofclinical opinions andcase studies(forreviewsseeLund Psychological,socialandwelfareinterventionsforpsychologicalhealthandwell-beingoftorturesurvivors(Protocol) 2 Copyright©2011TheCochraneCollaboration.PublishedbyJohnWiley&Sons,Ltd. 2008;Quiroga2005).Thereisrelativelylittleliteratureonout- 1.Toassessbeneficialandadverseeffectsofpsychological,social comesofhealthcareinterventionswithtorturesurvivors,andwhat andwelfareinterventionsversusnotreatmentforthereductionof existsisdominatedbycasestudiesandclinicalnarratives,withfew psychologicaldistressintorturesurvivors. cohortstudiesandfewertrials.Someofthesestudieshaveaimed toraiseawareness(andfunding)forspecificinitiativesaswellas informingthewiderfield,sopoliticalexpediencyhasoftenbeen Secondaryobjectives moreimportantthanacademicintegrity.Intheliteratureexamin- 2.Todescribethequalityandgeneralisabilityofthestudieseval- inghealthcareoutcomesfortorturesurvivors,thereiswidespread uating the effectsof these treatment approacheson torture sur- use of assessment or evaluation tools, which are frequently di- vivors,andspecifically: agnosis-based and developed in the West for Western, English- • toprovideanobjectiveassessmentofriskofbiasinthese speakingpopulationsandstandardisedonthesamepopulations. studies; Fewofthetoolstranslatedforthestudyoftorturesurvivorshave • todescribethespecificpopulationsevaluatedinstudiesof demonstratedvalidity(Bracken1995;Johnson2008;Patel2003; torturesurvivors(includingdemographics,tortureexperiences Thakker 1999), and many traditional assessment and outcome andpsychologicalstatus); measureshavebeencriticisedforfailingtodemonstratelinguistic • todescribethevarietyofinterventionsthathavebeen orsemanticequivalence,therebylackingconstructvalidityforcul- evaluatedinthesepopulations;and turallydiversepopulations(e.g.Elsass2009;Gurr2001;Hollifield • todescribetheoutcomesevaluatedintheseintervention 2002;Mahtani2003;Newlands2004;Patel2000;VanOmmeren studies. 2001)includingmanyrefugeesandasylumseekerswhoaresur- vivorsoftorture. Whilst there exists a vast body of research on the treatment of PTSD in various populations, such studies are rarely based on torturesurvivor samples(Bisson 2009;Nicholl2004).Theyare METHODS thereforeunlikelytoaddresstherangeofdifficultiesbeyondPTSD and depression (e.g. racism, destitution) experienced by torture survivors(forexamplePatel2007).Therearenumerousmethod- Criteriaforconsideringstudiesforthisreview ologicalproblemswithapplyingpsychiatricdiagnosticcriteriato thisclientgroup(Quiroga2005)andquestionsremainaboutthe validity of psychiatric diagnoses in general (Boyle 1999; Boyle Typesofstudies 2002;Kutchins1997;Pilgrim1999).Inparticular,thevalidityof Randomisedcontrolledtrials(RCTs),clusterRCTs,andquasiran- adiagnosis ofPTSDintorturesurvivorshasbeenchallengedas domisedcontrolledtrials(QRCTs)willbeincluded.QRCTswill medicalisingthesociopoliticalproblemoftorture(Bracken1995; beincludedbecause,duetothedifficultiesofconducting RCTs Bracken1998;Patel2003;Summerfield2001). inthispopulation,averysmallnumberofRCTsareexpectedto Therearenosystematicreviewsonwhichtreatmentapproaches beidentified. areeffectivewithtorturesurvivorsexperiencingarangeofpsycho- Therewillbenorestrictionsonpublicationtype,status,language logical,social,welfareandinterpersonalproblems,hencetheneed ordate.Iffulldetailscanbeobtainedfromtheauthors,conference forthissystematicreview.Inviewofthewiderangeofevidence abstractswillbeincluded,asrelevantmaterialisoftenpublishedby thatmightberelevanttothetreatmentofthispopulation,thisre- torturesurvivorcentresthemselves.Wherethereisanindication viewwillbeasinclusiveaspossible.Unlesscomparablestudiesare thattherearedatainconferencepublicationsandthecontactwith identified,thereviewwillnotattempttodirectlycomparetheef- authorsisunsuccessful,suchstudieswillbeindicatedinaseparate fectsofdifferentinterventions,andinsteadwillprovideadetailed sectionlabelled’studiesawaitingclassification’. descriptionoftheavailableevidence,summarisingthefeaturesof thesestudiesanddescribingtreatmentoutcomesforanyspecific diagnosticgroupsidentified. Typesofparticipants Formal diagnoses in these populations are frequently not avail- able. Therefore,thereview will notbe restrictedto participants meetingspecific diagnoses, butwill instead include participants OBJECTIVES who have survived any type of torture, as defined by the study authors.Torturesurvivorsmaybefoundamongrefugees,asylum seekers,warsurvivorsandsurvivorsoforganisedviolence,andin diverse settings, such as prison, detention centre, refugee camp, Primaryobjective accommodationcentre,healthcarefacility,andcommunity. Psychological,socialandwelfareinterventionsforpsychologicalhealthandwell-beingoftorturesurvivors(Protocol) 3 Copyright©2011TheCochraneCollaboration.PublishedbyJohnWiley&Sons,Ltd. Participantsofallageswillbeincludedand,asfaraspossible,stud- Rehabilitation services for torture survivors have arisen largely iesonchildrenandyoungpeople(<18)andthosetargetingadults frompsychologicalinterventions.Medicalinterventions,includ- willbeanalysedseparately.Exceptionsincludestudiesevaluating ingpharmacologicalinterventions,whereoffered,areusuallyof- familyorothersystemicinterventions.Wherenecessary,authors feredincombination with other psychosocial interventions and willbecontactedtorequestseparatedataforchildrenandadults. willthereforebeexcludedfromthereview. Typesofinterventions Typesofoutcomemeasures Interventionsprovidedinthisfieldtendtobepragmaticandrarely Oneoftheobjectivesofthisreviewistodescribetheoutcomes followtreatmentmanuals/protocolsormeetthestrictcriteriaex- evaluated in these studies. We will summarise any measures of pectedinothersettings. For thisreason our inclusion criteriais psychologicalhealthbenefitandwell-being.However,studiesof broad.However,inAppendix1wesetoutafulllistofthepsy- populations whichincludetorturesurvivors addressaverywide chotherapies,asdefinedbytheCochraneCollaborationDepres- range of outcomes, including many non-standardised measures sion,AnxietyandNeurosisGroup,sothatreadersmayunderstand specificallydevelopedforthatstudy (Jaranson 2011).Rarelydo howthesetherapiesrelatetooneanother. measuresundergotheconceptualscrutinywhichshouldprecede Wewillincludeanypsychological,socialorwelfareintervention translationandtesting(Johnson2006)andmeasuresareoftennot whichaimstoimprovehealthandwell-beingoftorturesurvivors. translatedintothefirstorfluentlanguageofresearchparticipants 1. Psychologicalinterventions.Thesewillinclude andtestedbeforeuse(onlyafewtraumainstrumentshavebeen psychodynamicandpsychoanalytictherapies,behaviouraland translatedandundergonesomevaliditytestinginsomelanguages). cognitivepsychotherapies(includingCBT,exposuretherapy), Furthermore,itisnotuncommontouseinterpreterstoassistin interpersonalpsychotherapy,narrativetherapy,cognitiveanalytic datacollection(seeVarainpressformoredetail),compromising therapy,anxiety/stressmanagementapproaches,systemic reliability(forexample,byintroducingvariationintheuseofterms psychotherapies,counselling,supportiveandexperiential andresponseoptions). psychotherapies,arttherapy,dramatherapy,dancetherapy,eye Wheredataareprovidedforspecificoutcomemeasuresthatcanbe movementdesensitisationreprocessing(EMDR),hypnotherapy. summarised,bothcontinuousorcategoricaldatawillbeincluded. Theseinterventionsmaybeundertakenwithindividualsorwith familiesorgroups. 2. Socialinterventions.Theseconsistofinvolvementin Primaryoutcomes communityactivities,reparationandjudicialactivities, 1. Reductiononascaleofpsychologicaldistress,e.g.General educationalinitiatives(suchaslearningthehostcountry HealthQuestionnaire(GHQ);ShortFormHealthSurvey(SF- language),andothers. 12)MentalHealthsubscale. 3. Welfareinterventions.Theseareexemplifiedbyprovision 2. Anyadverseeventsuchassuicideorself-harm. ofinformationaboutlegalandwelfarerightsandentitlements, findingnurseryandschoolplacesforchildren,productiveand creativeactivitiesincludingworkinitiatives,andothers. Secondaryoutcomes 1. Change(positiveornegative)inpsychologicalstatusor targetbehaviour.Thisislikelytobevariouslymeasuredas Interventionsetting changeinpsychologicalandemotionalsymptoms(suchas Interventionscanbedeliveredinanysetting,includinghealthcare depression,PTSD,anxiety),whichwillbepooled;aschangein clinicsandotherhealthcarefacilities,refugeecamps,communities, diagnosticcategory(suchasdepressionorPTSD);oras survivors’homesanddetentionfacilities. individualorgrouplevelchangeinatargetbehaviour. 2. Change(positiveornegative)inqualityoflifeorwell- being,forwhichtherearemultiplescaleseitherassessinglife Comparators qualityorglobalsatisfactionwithlife;extentofdisability. Thecontrolcomparisonwillbewaitinglist,notreatment,standard 3. Increasedparticipationandfunctioning,asmeasuredby care/accesstostandardcare,attentioncontrol. engagementineducation,training,work,orcommunityactivity. We will include trials where the intervention is added to non- 4. Changeinqualityand/orquantityoffamilyorsocial psychosocialstandardcaregiventobothgroups. relationships. Equivalencetrialswillbeincluded,butonlyifthecomparatorarm 5. Ratingsofpsychologicalfunctionmadebyothers, canreasonablybetreatedasa’treatmentasusual’.Forthisreason, includingclinicians;andforchildren:parents,orteachers non-equivalencetrials(comparisonoftwoactivetreatmentwith (ratingsbyparentsorteachersofchildren’sstatusarewidelyused thehypothesisthattheireffectsdonotdiffer)willbeexcluded. inpsychologicalinterventions.) Psychological,socialandwelfareinterventionsforpsychologicalhealthandwell-beingoftorturesurvivors(Protocol) 4 Copyright©2011TheCochraneCollaboration.PublishedbyJohnWiley&Sons,Ltd. 6. Ratingsoftheinterventionitself,suchassatisfactionwith Greyliterature intervention,ortherapeuticalliance. Inordertoidentifyrelevantgreyliterature,theRCTlibraryand OPENSIGLEwereincludedinthelistofdatabasestobesearched. Inaddition,arangeofpublicationtypeslikereports,conference papers,posters,monographsandanthologieswillbeincludedin Searchmethodsforidentificationofstudies thesearch. Searcheswillbeconductedonelectronicdatabases,websitesand thehandsearchingofreviewsandreferencelists. Referencemanagersoftware Thereferenceswillbemanagedusingthebibliographicsoftware EndNote. Electronicsearches Bibliographicdatabasesandtrialregisters: • PsycINFO(Onlinedatabaseofpsychologicalliterature) Datacollectionandanalysis • MEDLINE(Onlinedatabaseofhealthandmedical journalsandothernewssources) • EMBASE(Onlinedatabaseofhealthandmedicaljournals) Selectionofstudies • WebofScience(Onlinemultidisciplinarydatabasecovering Studyselectionhastwostages. allsciences) 1. Aninitialscreeningoftitlesandabstractsusingthe • CINAHL(Onlinedatabaseofnursingandalliedhealth inclusioncriteria,withtheaimofidentifyingstudieswhichmay literature) beeligibleandforwhichthefullpapershouldbeobtained. • TheCochraneCentralRegisterofControlledTrials Whereabstractsarenotavailableelectronicallythefullpaperwill (CENTRAL) besought. • Lilacs(Onlinedatabaseonhealthsciences,publishedin 2. Thefullpaperswillbereadandselectedagainstthe LatinAmericaandtheCaribbean) inclusioncriteriabytwooftheauthors[BK,AW]independently. • OPENSIGLE(Onlinedatabaseofreportsandothergrey Thefinallistwillbeachievedaftercomparison,and literatureproducedinEuropeuntil2005). disagreementswillberesolvedbydiscussion;wherethere • WHO:InternationalClinicalTrialsRegistryPlatform continuestobedoubtordifference,thethirdreviewer[NP]will (ICTRP) beconsultedtoachieveconsensus. • PILOTS(Onlinedatabase,PublishedInternational If full details can be obtained from the authors, conference ab- LiteratureOnTraumaticStress) stractswillbeincluded,asrelevantmaterialisoftenpublishedby torturesurvivorcentresthemselves.Wherethereisanindication thattherearedatainconferencepublicationsandthecontactwith authorsisunsuccessful,suchstudieswillbeindicatedinaseparate Searchingotherresources sectionlabelled’studiesawaitingclassification’. • OnlineLibraryoftheRehabilitationandResearchCentre forTortureVictims(RCT) • Referencelistsofreviewsemergingfromthesearches Dataextractionandmanagement • Referencelistofthefinalsetofincludedstudies Adataextractionformisintheprocessofbeingdesignedusingas • TableofContentsfromthetop10mostfrequentlycited amodeldataextractionprotocolsfromsimilarreviews. sourcesemergingfromthesearch(expectedtobejournalissues). • Studydesign • Settingofintervention • Typeofinterventions • Interventionprotocol Searchterms • Samplesizeatbaselineandoutcomeassessments Thesearchtermswillbedeliberatelybroad,asmanystudiesare • Baselinecharacteristicsofthesample(age,gender, conducted innon-Western, non-academic settings, with diverse nationality,ethnicity,typeoftortureexperienced,legalstatusif reporting structures. The following strategy will be employed refugeesandasylumseekers,livingsituation,separationfrom on the main bibliographic databases: (Population + RCT filter) closefamilymembers) OR(Population+Intervention).ThesearchtermsforPsycINFO • Baselinemeasures (asindicatedinAppendix2)willbeadaptedforeachrespective • Typeofpractitioner/therapists database. • Language/sofassessment;translation,interpretation Psychological,socialandwelfareinterventionsforpsychologicalhealthandwell-beingoftorturesurvivors(Protocol) 5 Copyright©2011TheCochraneCollaboration.PublishedbyJohnWiley&Sons,Ltd. • Propertiesofbaselinemeasures(language,translation, same,worse)willbere-categorizedintotwogroups.Wewillnot validity) calculateNNTs. • Outcomemeasuresatendofintervention(s)andatany Continuous data will be analysed using standardised mean dif- follow-upassessment ferences(SMDsoreffectsizes)usingpooledstandarddeviations • Completionrates and weighting for sample size, and calculating the 95% confi- • Adherence,participationintreatment denceinterval.Self-orother-ratingscalesriskproducingseverely • Riskofbiasofincludedstudies skeweddata,thatis,wheretheyproduceavaluebetween-1and +1whenthedifferencebetweenthescalemaximumorminimum Datawillbeextractedbytwoofthereviewers[BK,AW]indepen- and themeanisdivided by thestandard deviation. Wheredata dently,anddisagreementswillberesolvedbydiscussion. Where are severely skewed, they will be normalised where possible by therecontinuestobedoubtordifference,thethirdreviewer[NP] transforms,orifthisdoesnotproduceasatisfactorydistribution, willbeconsultedtoachieveconsensus. willbedichotomised.SMDswillthenbeinterpretedindividually withreferencetothequalityandreliabilityofthemeasurewhere available. Assessmentofriskofbiasinincludedstudies Ifauthorsarewillingtoproviderawdata,clinicalcutpoints(for Risk of bias will be assessed for each included study using the PTSDcaseness,forexample)canbeapplied.Inthemuchmore CochraneCollaboration’riskofbias’tool(Higgins2008)andis- likelyeventthatrawdataareunavailable,wewillapplythemeth- suesraisedaroundstudiesofpsychologicaltreatmentinsystematic odsdescribedbyPreston2000;inversevarianceweightsusingthe reviews(Yates2005).Thefollowingdomainswillbeconsidered: standard error.It remains likely that some severelyskewed data 1. Sequencegeneration:wastheallocationsequence willhavetobeexcludedfromanalyses. adequatelygeneratedbyamethodunrelatedtorecruitment Bothdichotomousandcontinuousdataanalyseswillbedisplayed decision? usingforestplots. 2. Allocationconcealment:wasallocationadequately concealed? 3. Sincepsychologicaltreatmentscannotblindpersonnel,and Unitofanalysisissues canrarelyblindparticipants,assessmentofbiaswillbemade using(a)equivalenceoftreatmentexpectationsofparticipants acrossarmsofthestudy;(b)presentationofthirdpartyoutcome Multiarmedtrials assessmentswherethethirdpartyisblindtotreatmentallocation? If there are two or more treatment or comparison groups, we 4. Incompleteoutcomedataforeachmainoutcomeorclassof willanalysethemseparatelybysplittingthecontrolgroupequally outcomes:wereincompleteoutcomedataadequatelyaddressed? betweenthetreatmentgroups(Higgins2008a). 5. Selectiveoutcomereporting:dotheresultspresentedmatch theassessmentsdescribed? 6. Othersourcesofbias:wasthestudyapparentlyfreeofother Clusterrandomisedcontrolledtrials problemsthatcouldputitatahighriskofbias?Additionalitems includedherearetherapistqualifications,treatmentfidelity,and Inthecaseofclusterrandomisation,wewilladjustfortheeffects researcherallegiance/conflictofinterest. ofclusteringusinganICC. The risk of bias will be assessed independently by two review authors (BK and AW), who will then reach consensus on their Dealingwithmissingdata decisions,withanyremainingdisagreementsreferredtothethird author(NP).Ifnecessary,furtherinformationwillbesoughtfrom Authorswillbecontactedtorequestmissingdata,suchasstandard studyauthors. deviations. Loss and exclusion of data will be examined to try Assessmentofriskofbiaswillusethreecategories:lowrisk,unclear to understand the reasons and implications, and if appropriate, (information not providedor effectnot clear),and highrisk. A amethodsuchaslastobservation carriedforwardmaybe used. riskofbiastablewillbeconstructedforeachstudy.Additionally, Where standard deviations are missing and unobtainable from forestplotswillbeorderedbyriskofbiastoexamineforsystematic authors,wewillcalculatewherepossiblefromF,t,orpvalues,or effectsonoutcome. fromstandarderror.Ifthisisnotpossible,thetrialwillbetreated ashavingnouseabledata. Wewillidentifyintentiontotreatanalysisasanimportantmarker Measuresoftreatmenteffect ofefforttoreducebias(seeAssessmentofriskofbiasinincluded Dichotomous outcomes (improved/not improved) will be anal- studies). ysedusingoddsratioswith95%confidenceintervals.Categori- Availablecasesanalyseswillbeincludedbutinterpretedwithcau- caloutcomeswithmorethantwocategories(suchasimproved, tion. Psychological,socialandwelfareinterventionsforpsychologicalhealthandwell-beingoftorturesurvivors(Protocol) 6 Copyright©2011TheCochraneCollaboration.PublishedbyJohnWiley&Sons,Ltd. Assessmentofheterogeneity Subgroupanalysisandinvestigationofheterogeneity High levelsofheterogeneityare likelywheretherearesufficient 1. Childandadultstudieswillbeanalysedseparately.Thisis trialsformeta-analysis.Wherethisissuggestedbytheforestplot becausemethodsandoutcomesdifferverysubstantially,asdoes (aspooroverlapofconfidenceintervalsandpresenceofoutliers) usuallythetypeoftortureexperienced. andthereisanI2statisticofover40%,itwillbeinterpretedusing 2. Iftherearesufficienttrials,wewillseparatestudies Higgins2003,withreferencetoin/consistencyinthedirectionof conductedonpopulationswhosemembersstillresideintheir effects,andwithparticularreferencetovariationbetweenstudies ownhomesvsthosewhoremainintheircountryoforiginbut intreatmentaimsandmethodswhichmightsuggestthattheset areinternallydisplacedvsthosewhoarerefugees.Thedifficulties shouldbesplit. anddangersofflightandofsettlingintoarefugeecampora countrywhereasylumhasbeensoughtcompoundandaddto theexistingproblemsandstressesonthetorturesurvivor.While Assessmentofreportingbiases thesearehardtoseparateonanindividuallevel,itmakessense Thefollowingstepswillbeundertakentoaddressreportingbiases wherepossibletorecognisethematatriallevel. which might otherwise operate, particularly inrelation to stud- iesinunder-resourcedsettingsandreportedinthegreyliterature. Searchesinarangeofdatabasesincludingthoseinlanguagesother thanEnglishandthoselistingnonpeerreviewedjournals;system- Sensitivityanalysis atic searchof thereference listsof reviews in thefieldand final includedstudies;manualsearchofthecontentpagesofthetopten Wherepossible,sensitivity analyseswillbeusedtoassesstheef- sourcesofpublicationsyieldedfromthesearch;searchofdatabases fectof the differentmethodological decisions made throughout thereviewprocess.Thesedecisionswillbetestedbysuccessively forregisteredtrialswhichcouldyieldpublishedandunpublished removing: studies;andtheinclusioninthereviewofanyeligibleunpublished aswellaspublishedstudies. 1. quasi-RCTstoleaveonlyRCTs; Foreligiblestudies,asearchwillbemadespecificallyforpublished 2. cluster-randomisedtrialstoleaveindividuallyrandomised protocols. trials; 3. trialsusingnon-ITTmethodstoleaveonlythoseanalysed usingITT(tobeconsideredITTanalysistheanalysismust Datasynthesis includeallparticipantswhoenteredtreatment,whetherornot RevMan5softwarewillbeusedtoconductmeta-analysiswhere theyprovideddataattheendoftreatment).Nuesch2009has feasible and appropriate. A random-effects model will be used foundthattrialswithintentiontotreatanalysesproducesmaller giventhevarioussourcesofdiversitydescribedabove.Wheremeta- treatmenteffectsinmeta-analyses,andthisdifferenceisgreater analysisisnotpossible,anarrativesummaryofevidencerelating inmeta-analysesinthepresenceofheterogeneity;and totheprimaryandsecondaryobjectiveswillbeprovided. 4. unpublishedtrials.Sometreatmentstudiesinthisliterature Wewillsummariseseparatelythosestudiesinvolvingdirectpsy- arepublishedinnonpeer-reviewedsources,suchaschaptersand chological interventions with individuals, couples, families or internalreportsofNGOs.Toaddressconcernsaboutdifferences groups.Wherestudiesincludebothadultandchildparticipants, inqualitybetweenthetwotypesofsources,sensitivityanalyses theywillbeanalysedtothepredominantagegroup,orincluded willbeundertaken,restrictedtothosestudiesinpeer-reviewed inbothanalysesasappropriate. journals. REFERENCES Additionalreferences oftorture. NY:KluwerAcademic/PlenumPublishers,Inc, 2001. AI2010 Bisson2009 AmnestyInternational. Reportsstatementandissuesbriefs. BissonJ,AndrewM.Psychologicaltreatmentofpost- AboutTorture. http://www.amnestyusa.org/counter-terror- traumaticstressdisorder(PTSD).CochraneDatabase with-justice/reports-statements-and-issue-briefs/about- ofSystematicReviews2009,Issue1. [DOI:10.1002/ torture/page.do?id=1031032Accessed25February2010. 14651858.CD003388.pub3] Basoglu2001 Boyle1999 BasogluM,JaransonJM,MollicaR,KastrupM.Torture BoyleM.Diagnosis. In: CNewnesGHolmesCDunn andmentalhealth: Aresearchoverview. In: GerrityE, editor(s). Thisismadness:Acriticallookatpsychiatryandthe KeaneTM,TumaFeditor(s).Thementalhealthconsequences futureofmentalhealthservices.Ross-on-Wye: PCCS,1999. Psychological,socialandwelfareinterventionsforpsychologicalhealthandwell-beingoftorturesurvivors(Protocol) 7 Copyright©2011TheCochraneCollaboration.PublishedbyJohnWiley&Sons,Ltd. Boyle2002 Jaranson,1995 BoyleM.Schizophrenia: AScientificDelusion?. Second JaransonJM.Government-sanctionedtorture: statusofthe Edition. London:Routledge,2002. rehabilitationmovement. TransculturalPsychiatricResearch Review1995;32:253–86. Bracken1995 BrackenP,GillerJ,SummerfieldD.Psychologicalresponses Johnson2006 ofwarandatrocity: Limitationsofcurrentconcepts. Social JohnsonTP.Methodsandframeworksforcross-cultural ScienceandMedicine.1995;40(8):1073–82. measurement. MedicalCare2006;44:S17–S20. Bracken1998 Johnson2008 BrackenPJ,CeliaP.Rethinkingthetraumaofwar.London: JohnsonH,ThompsonA.Thedevelopmentand SavetheChildrenwithFreeAssociationBooks,1998. maintenanceofpost-traumaticstressdisorder(PTSD)in civilianadultsurvivorsofwartraumaandtorture: areview. Elsass2009 ClinicalPsychologyReview2008;28(1):36–47. ElsassP,CarlssonJ,JespersenK,PhuntsokK.Questioning westernassessmentoftraumaamongTibetantorture Kutchins1997 survivors. Aquantitativeassessmentstudywithcomments KutchinsH,KirkS.MakingUsCrazy:DSM,ThePsychiatric fromBuddhistLamas.Torture2009;19(3):194–203. BibleandtheCreationofMentalDisorders. NewYork: Free Press,1997. Gurr2001 GurrR,QuirogaJ.ApproachestoTortureRehabilitation. Lund2008 Torture2001;11:5–35. LundM,SørensenJH,ChristensenJB,ØlholmA. MTVombehandlingogrehabiliteringafPTSD-herunder Higgins2003 traumatiseredeflygtninge. Vejle: UdgivetafRegion HigginsJPT,ThompsonSG,DeeksJJ,AltmanDA. Syddanmark[PublishedbyCenterforQuality,South Measuringinconsistencyinmeta-analyses. BritishMedical Denmark],2008. [:ISBN978–87–92217–03–5] Journal2003;327:557–560. Mahtani2003 Higgins2008 MahtaniA.Therighttoappropriatepsychologicalservices. HigginsJPT,AltmanDG(editors). Chapter8: Assessing InternationalJournalofHumanRights2003;7:40–57. riskofbiasinincludedstudies. In: HigginsJPT,Green Seditor(s). CochraneHandbookforSystematicReviews McIvor1995 ofInterventionsVersion5.0.2[updatedSeptember2009]. McIvor,RJ.Turner,SW.Assessmnetandtreatment TheCochraneCollaboration, 2008. Availablefrom approachesforsurvivorsoftorture. BritishJournalof www.cochrane-handbook.org,2008. Psychiatry1995;166:705–711. Higgins2008a Montgomery2011 HigginsJPT,DeeksJJ,AltmanDG(editors). Chapter MontgomeryE,PatelN.Torturerehabilitation: Reflections 16.5: Specialtopicsinstatistics. In: HigginsJPT,Green ontreatmentoutcomestudies.Torture2011;21:141–145. Seditor(s). CochraneHandbookforSystematicReviews Moreno2002 ofInterventionsVersion5.0.2[updatedSeptember2009]. MorenoA,GrodinMA.Tortureanditsneurological TheCochraneCollaboration, 2008. Availablefrom sequelae. SpinalCord2002;40(5):213–23. www.cochrane-handbook.org,2008. Newlands2004 Hollifield2002 NewlandsJ,PatelN.WorkingwithDiversity: Professional HollifieldM.AccurateMeasurementinCulturalPsychiatry: PracticeandEthicalConsiderations. In:TribeR,Morrissey WillwePaytheCosts?. TransculturalPsychiatry2002;39: Jeditor(s). TheHandbookofProfessionalandEthicalPractice 419–21. forPsychologists,CounsellorsandPsychotherapists.EastSussex: IRCT2010 Brunner-Routledge,2004:233–46. InternationalrehabilitationCouncilfortorturevictims. Nicholl2004 26June-InternationalDayagainstTorture. Thefight NichollC,ThompsonAR.Thepsychologicaltreatmentof againsttorture: akeypriorityfortheEU.http://europa.eu/ PTSDinadultrefugees: Areviewofthecurrentstateof rapid/pressReleasesAction.do?reference=MEMO/07/254. psychologicaltherapies. JournalofMentalHealth2004;13: Accessed25February2010. 351–362. Jacobs2001 Norredam2005 JacobsU,IacopinoV.Tortureanditsconsequences: NorredamM,CrosbyS,MunarrizR,PiwowarczykL, Achallengetoclinicalneuropsychology. Professional GrodinM.Urologiccomplicationsofsexualtraumaamong Psychology-ResearchandPractice2001;32(5):458–64. malesurvivorsoftorture. Urology2005;65(1):28–32. Jaranson2011 Nuesch2009 JaransonJM,QuirogaJ.Evaluatingtheservicesoftorture NueschE,TrelleS,ReichenbachS,RutjesAW,Burgi rehabilitationprogrammes: Historyandrecommendations. E,SchererM,etal.Theeffectsofexcludingpatients Torture2011;21:98–140. fromtheanalysisinrandomisedcontrolledtrials: meta- Psychological,socialandwelfareinterventionsforpsychologicalhealthandwell-beingoftorturesurvivors(Protocol) 8 Copyright©2011TheCochraneCollaboration.PublishedbyJohnWiley&Sons,Ltd.