A psychosociAl needs Assessment of communities in 14 conflict-Affected districts in Aceh

2007

tAble of contents

i foreword from the ministry of health of indonesia ii foreword from the World bank iii foreword from harvard medical school 1 Acknowledgements 2 executive summary 2 project design 3 Key findings 7 recommendations 5 introduction 10 research design and methodology 10 Goals of the research 10 study design 10 TheInterviewStructure 12 PhasesofFieldworkandGeographicCoverage 12 ResearchersandFieldworkTeams 13 SelectionofVillages–DesignandImplicationsofPNASamplingStrategies 14 SelectionofRespondents 14 InterviewMechanics 15 DataEntryandAnalysis 15 reporting by six Geographic regions 15 NorthCoast 15 EastCoast 16 CentralHighlands 16 SoutheastHighlands 16 SouthwestCoast 17 AcehBesar 17 reporting results18 sample demographics: comparing pnA1 and pnA222 traumatic events 22 traumatic events during the conflict 22 Variation in traumatic events by region 22 HighPrevalenceRegions 22 HotSpots:LocalSpecificityWithinRegionsandDistricts 26 CaseStudy:TheKluetRiverValley 30 LowPrevalenceRegions 30 Gendered trauma 30 SexualViolence 31 HeadTrauma 31 forced evacuation and other displacement experience 36 post-conflict stress 39 Villages designated as high conflict by iom’s mGKd programme 39 PastTraumaticEventsinMGKDVillages 39 Post-conflictStressinMGKDVillages 39 summary: conflict trauma and collective memory in Aceh46 the psychosocial and psychological remainders of Violence: depression, Anxiety and traumatic stress disorders 46 methods 46 MeasuresofPsychologicalDistressandNeuropsychiatricDisorders 46 AnalysesofPsychologicalSymptomsandPsychiatricDiagnoses 48 findings 48 SelfAssessmentofEmotionalDistress 49 PsychologicalandPsychiatricDiagnosesUsingFormalInstruments 54 TheDistributionofRisk:TheEffectsofTraumaticExperiencesonPsychologicalDistress 55 TheDistributionofRisk:GroupsatHighRisk 57 TheDistributionofRisk:HeadTrauma 58 SocialFunctioning

60 ReducedLevelsofPsychologicalSymptoms:AcehneseResilience,ContinuedNeedforServices,andQuestions forFurtherResearch

66 community mental and psychosocial health 66 Available resources in communities 67 perceptions of conflict-related mental disability in the community 72 perceptions of nGos and public health services73 conflict-related clinical illnesses in these communities: findings from a mental health outreach project74 recommendations

list of tAbles

13 Chart1 ImplementationofPNA1andPNA218 Table1.1 DistrictLocationsandSampleSizeofPNA1andPNA2Respondents19 Table1.2 SampelCharacteristics:Gender,AgeandEthnicitybyRegion20 Table1.3 PNA1andPNA2RespondentCharacteristicsbySex,Age,MaritalStatus,EducationandHousing21 Table1.4 RespondentsbyPresenceorAbsenceofPost-ConflictCommunitiesReintegraionProgramme(MGKD)

byPNAandPNA221 Table1.5 RespondentsbyPresenceorAbsenceofMGKDProgrammebyRegionandDistrict24 Table2.1 PastTraumaEventsbyPNAStudyandRegion27 Table2.2 PastTraumaEventsintheKluetRiverValley32 Table2.3 PastTraumaEventsExperiencedbyMen,byRegion34 Table2.4 PastTraumaEventsExperiencedbyWomen,byRegion36 Table2.5 HeadTrauma/PotentialBrainInjury:Men36 Table2.6 HeadTrauma/PotentialBrainInjury:Women38 Table2.7 ForcedEvacuationsandOtherDisplacementExperience38 Table2.8 ConflictIDPDestinations,byRegion(PNA2DataOnly)40 Table2.9 CurrentStressfulEventsExperiencedbyRespondents42 Table2.10 PastTraumaEventsbyPNAandMGKDVariable44 Table2.11 CurrentStressfulEventsExperiencedbyRespondents,byPNAandMGKDDesignation49 Table3.1 HSCLCoreDiagnosticDepressionSymptoms50 Table3.2 HarvardTraumaQuestionaire(HTQ)CoreDiagnosticSymptoms50 Table3.3 GeneralPsychologicalDistressbyRegions52 Table3.4 Depression,PTSD,andAnxietySymptomsandDiagnosesbyRegions52 Table3.5 Depression,PTSD,andAnxietySymptomsandDiagnosesbyGender54 Table3.6 Depression,PTSD,andAnxietySymptomsandDiagnosesbyMGKDDesignation56 Table3.7 AdjustedOddsRations*:MentalHealthProblemsbyPastTraumaticEventsforPNA2Respondents57 Table3.8 Adjusted Odds Rations*: Mental Health Problems by Numbers Current Stressful Events, for PNA2

Respondents58 Table3.9 AdjustedOddsRations:MentalHealthProblemsbyStudyandbyGender59 Table3.10A AdjustedOddsRations:MentalHealthProblemsbyAge(PNA1)59 Table3.10B AdjustedOddsRations:MentalHealthProblemsbyAge(PNA2)61 Table3.11 HeadTraumabyGenderandAgeforAcehTimur61 Table3.12 HeadTraumabyGenderandAgeforAcehSelatan62 Table3.13 SocialFunctioningbyDepression66 Table4.1 HelpSeekingBehaviourDuringthePastSixMonths68 Table4.2 RespondentPerceptionsofMentalIllnessintheCommunityandatHome68 Table4.3 Respondent Selection of Groups in Their Community Suffering Most from Conflict-Related Stress or

Trauma70 Table4.4 OpinionsAboutNGOMentalHealthServicesandImplementingPartners70 Table4.5 OpinionsAboutNGOMentalHealthServicesandImplementingPartners,byGender

“A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 20074 “A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007 i

foreWord

TheIndonesianProvinceofNanggroeAcehDarusalam(NAD)isaregionwhichisfacingauniquesetofproblems,includingtheconflictendedbythe2005peacedeal,andthe2004tsunami.

These events have generated a widespread impact on the lives of the communities. One of the most crucial issues to beaddressedasidefromlegal,security,socialandeconomicproblemsisthematterofhealth,includingmentalhealth.

Weare joyous thatwehave left the thesedifficult times, and it isnowourobligation to restoreaspectsof life thatwouldotherwisebringadverseeffectonthepeople,includingthelingeringeffectsofsuchevents.

In regards to health issues, comprehensive steps have been formulated into various short-, medium-, and long-termprogrammes.

Specifically on mental health issue, whose impact is quite significant, the Indonesian Ministry of Health has collaboratedwith the NAD government and national as well international NGOs. With this aim in mind, a comprehensive mentalhealthcare model has been designed and commenced, targeting not only regions affected by the tsunami, but also otherprovincesinwhichthismodelmayserveasreferenceindevelopingmentalhealth.

Therefore, we are very happy to see the organizing of this psychosocial needs assessment in 14 high conflict districts,under a cooperation between the International Organization for Migration (IOM), the Department of Social MedicinefromHarvardMedicalSchoolandtheSyiahKualaUniversity(SKU).

Iamconvincedthattheoutcomeofthisassessmentisinlinewithandsignificantlycontributetotheprogrammesthatwearecurrentlydeveloping,suchasthecapacitybuildingprojectintheformoftrainingsforcommunityhealthcenterandhospitalphysiciansatthedistrictlevel,aswellasthedevelopmentoftheCommunityMentalHealthNursing(CMHN)concept.Itishopedthatthispartnershipwillbefollowedbyotherprogrammes.

Toallthepartieswhohavemadethisundertakingareality,Iexpressmyhighestappreciation.

LetushopethatitwillbringgreatbenefittotheAcehneseinparticular,andtheentireIndonesianpeopleingeneral.

MayGodtheAlmightygrantHisblessinguponusall.

foreWord from the ministry of heAlth of indonesiA

dr. dr. siti fadilah supari, sp.Jp (K)MinisterofHealthoftheRepublicofIndonesia

“A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007ii “A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007 iii

This project, the PNA2, represents a collaboration between IOM and members of the Department of Social Medicine,HarvardMedicalSchool.ItissupportedbyacontractfromIOMtotheDepartmentofSocialMedicine,enablingmembersofthefacultyandstaffoftheDepartmenttoprovidetechnicalassistancetoIOMprojectsandtocollaborateindevelopmentofprojectsofcommunityandmentalhealthforpost-tsunamiandpost-conflictAceh.ItrepresentsalargercommitmentoftheDepartmentofSocialMedicinetoprogrammesofglobalhealthandhealthandhumanrights,andmorespecificallyacommitment to address health and mental health problems through science and the development of clinical and publichealthinterventions.IthasbeenourpleasuretocollaboratewithIOMinthissharedmission.

Thisprojectwasdesignedspecificallyasa“psychosocialneedsassessment.”Althoughtheprojecthasobvioushumanrightsimplications, the focus of the project was on mapping out levels of violence, traumatic experiences, and psychologicalsymptoms in the communities of rural Aceh in order to identify levels of need for mental health and psychosocialservices,establishpriorities forsuchservices,andprovidebase linedata for thosewhomightcarryoutprojectsor furtherresearch. Taken together, the PNA1 and PNA2 represent the most complete mapping of traumatic experience in highconflictdistrictsandsubdistrictsofAceh.WehopethesedatawillbeusedbythegovernmentofAcehandthegovernmentofIndonesia,aswellasbythedonorcommunity,tosupportthedevelopmentofprogrammestoaddresstheneedsformentalhealthservicesthatfollowyearsofsustainedviolenceenactedagainstvillagecommunitiesinAceh.

The PNA1 and PNA2 studies document widespread experiences of conflict across Aceh, as well as remarkably high levelsof systematic violence against civilian populations in particular districts and subdistricts. These experiences will not easilybe forgotten, and the mental health problems remaining – depression, PTSD, communal anxiety, head trauma – will belong-lasting.These“remaindersofviolence”deserveseriousattention.Theyrequirethedevelopmentofcommunitymentalhealth services forallofAceh.Theyrequireexplicitprogrammesaimedatproviding services for themosthighlyaffectedcommunities, many of which are relatively isolated and outside of access to routine care. And the memories of violencewillrequireanon-goingcommitmenttoprogrammesofreconciliationand‘traumahealing’atmanylevels.

Wearedelighted tobeable toworkwith IOM indevelopingmentalhealthoutreach services to someof themosthighlyaffectedcommunities,andcollaboratewithIOMandmanyotherstocontributetowiderprogrammesaimedataddressingthe remainders of violence. We hope these studies will bring attention to the continuing needs of those who haveenduredremarkablelossesandpersonalviolenceduringtheconflictandwillcontributetothelargergoalsofpeace-buildinginAceh.

foreWord from the depArtment of sociAl medicine, hArVArd medicAl school

mary-Jo delVecchio GoodProfessorofSocialMedicineDepartmentofSocialMedicineHarvardMedicalSchool

byron J. Good ProfessorMedicalAnthropologyDepartmentofSocialMedicineHarvardMedicalSchool

The Decentralization Support Facility (DSF) and the World Bank were pleased to support the important work ofthis Psychosocial Needs Assessment second phase, conducted by IOM, the Department of Social Medicine from HarvardMedical School, and researchers from Syiah Kuala University. The assessment highlights the importance of developingmental health and psychosocial services to support communities’ efforts at full recovery following conflict situations.The study highlights the importance of understanding and responding to the psychological and acute mental healthproblemsresultingfromyearsofconflictandprovidesimportantlessonsformentalhealthprogrammingnotonlyinAcehprovincebutinotherpost-conflictenvironmentsaswell.

Based upon the findings from these assessments, the DSF and the World Bank have been supporting IOM’s pilot mentalhealth program in 2006-2007 to reach civilian populations in high conflict villages in Aceh and respond to acute mentalhealth needs. This assistance forms part of our institutions’ broader support to post-conflict reconstruction andsustainablepeace-buildingwithinAceh.

It is with great pleasure therefore that we congratulate our partners on this valuable report and their contributiontowardsadeeperunderstandingofpsychosocialandmentalhealthproblemsinAceh.

foreWord from the decentrAliZAtion support fAcility And the World bAnK

susan WongActingSectorCoordinatorWorldBankSocialDevelopmentUnitandonbehalfofDSF

“A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007iv “A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007 1

AcKnoWledGements

1. FirstandforemostwewouldliketothankthegraciousandbraveAcehnesewhoagreedtotheoftendifficultprocess of undertaking our questionnaire, as well asthekeyinformantswhoprovidedadditionalqualitativedata. Dozens of village heads and other local leaderswelcomed our research teams and facilitated thisprocess.

2. There were five team leaders and 30 interviewingstaffspreadoutalloverAcehprovinceformorethan13 days of intensive and exhausting fieldwork informer conflict areas. They came from diverse buttopically related backgrounds such as nursing,counselling, public health, and education. Althoughtoo many to list here, we thank them for their hardworknevertheless.

3. The Ministry of Health supported IOM’s psychosocialresearch in Aceh, especially the Provincial HealthOffice in Banda Aceh andthe district health officesthroughout the province. Thepsychiatric hospitalin Banda Aceh providedninepsychiatric nurses tojoin the research staff, and several district healthofficesprovidedsupportwiththeircommunitymentalhealthnurses.Thenursecontributionstothisresearchwereoutstanding.

4. IOM’s research partners at Harvard Medical Schoolare responsible for the overall study design,questionnairedevelopment, database management,statistical analyses and this report. Professor ByronGood and Professors Mary-Jo DelVecchio Good werethe primary investigators of this project. MathewLakoma delivered efficient and creative statisticalanalyses.

5. IOM’s Post-Conflict Reintegration Programme in Acehhas been an exceptional forum for planning anddiscussing the material contained in this report.Jesse Grayman was overall project manager andcoordinator for this study. The Programme’s medicalteam, under the leadership of Dr. Teuku Arief Dianand Dr. Dara Juliana, hasbeen a reliable sourceof programming and material support. Thanksare most especially dueto Hayatullah, who didanoutstandingjoboforganisinglogisticsforthefieldresearch, supervised data entry, translated Acehneselanguage interviews, and always providedreliableandinvaluableassistanceunderpressure.

We would like to thank Dr. Ibrahim Puteh for hispsychiatric expertise and for helping us navigate

the health and university systems in BandaAceh. Saleh Amin carefully read and coded severalbatches of qualitative data. Su Lin Lewis providedcrucialinsightandbackgrounddatafordesigningthesamplingmethodology.

Mental health is a novel feature in post conflictprogramming and we thank Programme DirectorMark Knight and Deputy Programme Director LucChounet-Cambas for recognising its importance.Marianne Kearney graciously undertook theediting ofthis report, Sandanila Fatharani wholaid out this report, and Adi Nugroho provided anexcellent translation for the Indonesian languageversion. Finally Dr. Nenette Motus wrote the originalproposal to carry out this work and started theresearchprocesswithaphonecalltoHarvardMedicalSchool in October 2005. We thank Dr. Nenette forherpatienceandconfidenceinourwork.

6. Professor Bahrein Sugihen and his team ofresearchers at the Centre for the Developmentof Regional Studies at Syiah Kuala Universitywere key consultants on the study design andfieldwork for this research. In particular we thankIbu Rosnani from the Centre who provided helpfulassistancetotheHarvardteaminAceh.

7. The World Bank and the Decentralization SupportFacility (DSF), a multi-donor fund, provided crucialsupport and funding for the field research in tendistricts in Aceh. The Harvard Medical School, incontract with IOM, provided funding for the fieldresearch in Aceh Besar, and the two organisationsalso funded the data analysis and writing of thereport.

8. Results from the first phase of this research broughtabout the implementation of a pilot psychosocialand medical outreach programme to villagers in25 high conflict villages in Bireuen district. Wewould like to thank everyone on the outreach team:Dr. Ira Aini Dania, Dr. Karliansyah, Dr. Riska Sofia,Dr. Enny Bahari, and the three nurses; NifantoroWau, Hafidh, and Bukhari. And we especiallyacknowledge the crucial support from andcollaborationwithDr.AmrenRahimandDr.MursyidahattheBireuenDistrictHealthOffice.

9. The authors of this report are: Mary-Jo DelVecchioGood, Byron Good, Jesse Grayman and MathewLakoma.

Photo:Dr.TeukuAriefDian

“A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 20072 “A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007 �

executiVe summAry

1 Good,B.,M.-J.D.Good,J.Grayman,andM.Lakoma.2006.Psychosocial Needs Assessment of Communities Affected by the Conflict in the Districts of Pidie,

Bireuen, and Aceh Utara.InternationalOrganizationforMigration.

Between December 2005 and November 2006, a team of researchers from the International Organization for Migration(IOM)andtheDepartmentofSocialMedicinefromHarvardMedicalSchool,carriedoutaPsychosocialNeedsAssessment(PNA) in high conflict sub-districts across Aceh, in two phases. Phase 2, or Psychosocial Needs Assessment 2 (PNA2)conducted research in 75 high conflict villages in 11 districts throughout Aceh. The PNA2 report is an extension of theresearch for Psychosocial Needs Assessment 1 (PNA1), which was conducted in high conflict sub districts in Aceh Utara,BireuenandPidie,Aceh inFebruary2006.Research for this second studywas conducted in10districts in July2006withfundingfromtheWorldBank,DecentralizationSupportFacility(DSF),IOM,andtheHarvardMedicalSchool,andinAcehBesardistrict inNovember2006, fundedby IOMand theHarvardMedicalSchool.Theprimary focusof this report is toprovide findings from the PNA2 data and to compare these data with data previously analysed and published in the firstPsychosocial Needs Assessment (PNA1) report.1 Research for PNA1 was funded by the Canadian Department of ForeignAffairsandInternationalTrade,IOMandHarvardMedicalSchool.

Thebasicgoaloftheoverallprojectwastoevaluatethepsychosocialandmentalhealthneedsincommunitieswhichhavebeen deeply affected by the years of conflict between armed forces of the Republic of Indonesia and the Free AcehMovement (GAM), given the cessation of violence after the signing of the August 2005 Memorandum of Understanding.This report focuses on past traumatic experiences and current psychosocial and mental health needs in high conflictareas throughout Aceh. Although the peace agreement ended almost three decades of violence most of the traumaticexperiencesreporteddatefromtheearly1990’suntilAugust2005.Thereportdeliberatelyrefrainsfromidentifyinggroupsorindividualsinstrumentalintheviolencevisiteduponthesecommunities.

proJect desiGnTheprojectwasdesigned toprovidescientifically-derived,empiricaldatawhichcanserveasabasis fordevelopingmentalhealth and psychosocial services to support these communities’ efforts at recovery. Specifically, the two studies sought todeterminethelevelofconflict-relatedtraumaticexperiencessufferedbymembersofthesecommunities,tomapdifferencesin these experiences across regions and communities, to assess levels of psychosocial and mental health problems,identify high risk subgroups in the population, to identify patterns of resilience and resources drawn on by communitiesin managing mental health problems, and to assess the urgency for particular forms of mental health interventions inareasaffectedbydecadesofviolence.Giventhefindingsofextraordinarilyhighlevelsoftraumaticexperienceandmentalhealth symptoms in the initialPNA1study inAcehUtara,Bireuen,andPidie,PNA2wasdesignedexplicitly toextend theneeds assessment to all high conflict districts in Aceh, and to compare the findings in other parts of Aceh with those intheindexcommunitiesstudiesinPNA1.

Thestudywasdesignedbyseniorresearchers fromHarvardMedicalSchool, ledbyProfs.ByronGoodandMary-JoGood,and by Jesse Grayman, Ph.D. candidate at Harvard and IOM staff member. It included two components: a qualitative,key-informant study designed to explore how the conflict has affected particular communities and what communityleaders feelshouldbethepriorities forrespondingtothepsychosocialeffectsof theconflict;andsecond,a formalsurveyof adultmembersof selectedcommunitiesdesigned tomeasure levelsof experienceof traumaevents associatedwith theviolence, levels of psychological distress associated with these experiences, and perceived priorities for services. The fieldsurveyforPNA2wascarriedoutbya teamofresearchershiredanddirectedbytheIOMfieldstaff inBandaAceh, ledby

Jesse Grayman. Data were analysed at Harvard Medical School by Prof. Mary-Jo Good and Matthew Lakoma. This reportwasauthoredbyMary-JoGood,ByronGood,JesseGraymanandMatthewLakoma.

ThesampleforthequantitativesurveyforPNA2consistedof1,376adult,aged17orolder(andforPNA1thesampleconsistedof596adultrespondents)randomlyselectedfromruralcommunities thatexperiencedthehighestconflictsincetheearly1990’s. Sampling procedures produced a well distributed and representative sample of adult men and women in thesecommunities. In addition, key informants, consisting of leaders in the selected communities, were interviewed in allparticipatingvillages.Thepresentreportfocusesonanalysisofthequantitativesurvey.

Forpurposesofanalysis,thecommunitiesinthesurveyaredividedinto6regions,groupingdistrictsthatsharegeographicalcontiguity,culturalsimilarityandcommonconflicthistories.Thesixregionsare;theNorthCoast(AcehUtara,Bireuen,Pidie)AcehBesar,theEastCoast(AcehTamiang,AcehTimur)theCentralHighlands(AcehTengah,BenarMeriahandGayoLues),SoutheastHighlands(AcehTenggara)andSouthwestCoast(AcehBarat,NaganRaya,AcehBaratDayaandAcehSelatan).Findingsmaybegeneralizedonlytohighconflictcommunitiesinthesesixregions.

Key findinGs1. We expected to find that communities surveyed for PNA2 would have lower levels of violence than those surveyed

for PNA1, i.e. North Coast, given the centrality of this district in the history of GAM and the conflict. But the firstoverwhelmingfindingwas that two regions– theEastCoast and theSouthwestCoast – suffered terrible violenceandtraumaticeventsat a levelequivalent toorevenhigher than that in theNorthCoast. In theEastCoast communities,for example, 80% of the respondents reported having lived through combat experiences, 45% experienced havingto flee from burning buildings and 61% having to flee from danger. 7% of women have had their husband killed inthe conflict, 50% of respondents report having had a family member or friend killed, and 45% reported having afamily member or friend kidnapped or disappear. Nearly half, or 47% reported having their property confiscated ordestroyed, and 31% experienced extortion or robbery. Persons in the Southwest Coast region, where the violencewasofmuchshorterduration,reportedexperiencesofviolenceatnearlyequivalentrates.

2. While the sampling employed for the PNA had in mind a broad stroke mapping of conflict events and psychologicalsymptomsacrossAceh,usingthePNAdataitispossibletopinpointsomeexemplarymicro-localities,suchastheKluetRiver valley case study from Aceh Selatan, which can not be captured by the regional aggregate data alone. Theseso-calledhotspotswithhistoriesofintense,complex,andsustainedburstsofviolencemaysuggestprioritiesforthetailoreddesignanddeliveryofpsychosocialandotherpost-conflictreintegrationservices.

3. Given the extraordinarily high levels of traumatic events reported for PNA1, we were not surprised that our PNA2survey found significantly lower levels of traumatic events, particularly in the Southeast Highlands and in AcehBesar. But an equally important finding is just how many persons even in these relatively low conflict communitieswere deeply affected by the conflict and violence. 69% of respondents in the Southeast Highlands, and 68% ofrespondents in Aceh Besar and 62% in the Central Highlands reported experiencing combat. Even in the SoutheastHighlands, 12%reportedhavinga familymemberor friendkilled,while31%inAcehBesarand39%in theCentralHighlands had lost a family member or friend to the violence. A quarter or 25% of men in the Central Highlands

“A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 20074 “A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007 �

and 32% of men in Aceh Besar reported being beaten on the body. Thus, while violence was localized in particularregions and villages, it was also widespread, affecting a remarkably high number of persons throughout these ruralcommunities in Aceh and contributing to a consensus of collective memories of terrifying and sustained violenceperpetratedagainstordinaryruralciviliansacrosstheentireprovince.

4. Both men and women experienced extraordinary levels of violence, but the level and type of traumatic eventsexperienced as part of the conflict varied by gender. Men reported significantly greater physical violence thanwomen.AcrossthePNA2sample,38%ofmenreporthavingbeenbeaten(9%ofwomen),19%reportbeingattackedby a gun or knife (8% of women), 16% of men report being tortured (3% of women), 15% of men reported beingbeen taken captive (3% of women), and 44% of men (and 35% of women) report witnessing physical violenceagainst others. Nonetheless, women suffered enormously. For example, in the Southwest Coast communities, 79%of women experienced combat and 56% had to flee danger, 52% were forced to witness physical punishment,36% had a family member or friend killed, 43% had property confiscated or destroyed, and 32% were forced tosearch for GAM members in the forest. Although rates of reported sexual violence toward women are relativelylow (1% raped, 4% other sexual assault), owing in part to stigma, stories make it clear that sexual violence by malecombatantstowardwomenwasnotuncommon.

5. Rates of head trauma and potential brain injury, suffered through beatings, strangulation, near drownings and otherforms of torture or violence, were extraordinarily high and deserve clinical interventions and further research.Men, particularly those in the highest conflict regions, were at the highest risk. Remarkably, 43% of all men in theEast Coast region and 41% in the Southwest Coast, report having suffered head trauma – rates equivalent to thosefoundinPNA1fortheNorthCoast.CliniciansintheIOMmentalhealthoutreachprogrammetohighconflictvillagesinBireuen are finding that head trauma is indeed a common problem, with nearly one fourth of all patientswhohaveclinicallysignificantlevelsofmentalhealthproblems,displayingheadtraumasymptoms.

6. In nearly all of the high conflict areas sampled, between one-third and two-thirds of all respondents were displaced,usually by force, during the conflict. Almost every respondent who told us about their displacement experienceduring the conflict should be considered as a returned IDP. But there are still thousands of Javanese transmigrantIDP families still living in North Sumatra province just south of Aceh, and several thousand Acehnese refugeesstillinMalaysia,whichthePNAdatathereforedoesnotcapture.ThecommonexperienceforIDP’swasthattheyreturnedtofindtheirwholevillageburnt,fieldsdestroyed,housesransackedandanimalskilledandthesevulnerableeconomicconditionshavesignificantbearingontheirpsychosocialcondition.

7. Thesetofquestionsoncurrentstressfulortraumaticeventsoccurringsincethesigningofthepeacedeal,werereportedat a much lower rate in PNA2 than in PNA1. The pattern holds, even in the regions where the conflict was greatest,andisparticularlytrueofreportsofseeingperpetrators(47%inPNA1,7%inPNA2),ofexperiencingassault(31%vs.1%), experiencing robbery (21% vs. 1%), violence toward women (4% vs. 1%) and violence toward children (7% vs.1%).ThesefindingssuggestaverysignificantchangebetweenFebruary2006,whenthePNA1surveywasconducted,andwhen trust in the peace deal was limited and non-locally based Indonesian troops had only recently left Aceh, andJuly2006,whenthePNA2surveywasconducted.ThePNA2findingssuggestthatbyJulyonwardsresidentsexperiencedfarlesscontactwithperpetratorsoftheviolenceandfargreatersecurity.

ThepatternoflowercurrentstressfuleventsscoresfoundinPNA2,holdstrueforbasiclivingconditions(lackofadequatehousing59%vs.38%;water/sanitation75%vs.55%;and foodsecurity72%vs.63%)and livelihood issues(difficultyproviding for family 85% vs. 72%; finding work 89% vs. 75%; and starting a livelihood 71% vs. 56%). However, eventhough they are significantly reduced, these remained extremely high in the PNA2 survey. The qualitative interviewssuggestthatthesecurrentstressorsreflectthedevastationofthevillageeconomiesfromthreedecadesofeffortstodestroythematerialfoundationssupportingGAM.Thus,recoverywillrequireboththattheterribletraumaticeventssufferedbythesecommunitiesandthebrokeneconomyanddestroyedcommunityresourcesbedealtwithinatimelyfashion.

8. Levels of past traumatic events were substantially and significantly higher in villages which received IOM’s Post-Conflict Community Reintegration Programme or MGKD programme, for both PNA1 and PNA2. Rates of somespecific past traumatic events were 50% or more higher in MGKD communities in both samples (for eventscommonly experienced, these achieve high levels of statistical significance). These findings validate the designation

by IOM and the World Bank of these villages as high conflict communities requiring special assistance. On thequestion of current stressful events (post-peace agreement) PNA2 respondents reported current stressful events atrates significantly lower than in PNA1, with MGKD villages reporting higher rates than non-MGKD villages for only4of18events.TherecouldbeasmallsamplingbiastowardslesshighconflictvillagesinthedistrictsthatdidnothaveaMGKDvillageprogramme(BenerMeriah,AcehTengah,GayoLuesandAcehBesardistricts).

9. RespondentsfromPNA2suffermentalhealthproblemsassociatedwiththeviolenceatasignificantlevel.Internationallyaccepted protocols for determining who suffers from psychological symptoms, indicate that 35% of the totalPNA2 sample ranked high on depression symptoms, 39% on anxiety symptoms, and 10% on Post Traumatic StressDisorder (PTSD) symptoms. Rates are substantially higher for respondents from the Southwest Coast, East Coast,and Central Highlands. For example, 41% of respondents from the Southwest coast suffered depressive symptomsaboveinternationallyrecognizedcutofflevels,43%anxietysymptomsand14%PTSDsymptomsatsuchlevels.

At the same time, a highly significant finding of PNA2 is that respondents in all districts surveyed in July andDecember 2006 reported substantially lower rates of major depression, anxiety disorders, and Post Traumatic StressDisorder (PTSD) than those in PNA2. Given that the reported rates of psychological symptoms among the PNA1(North Coast) respondents, were at some of the highest levels reported for post-conflict settings worldwide, wewerenotsurprisedtofindthatthepercentageofPNA2respondentswhosufferhighlevelsofsymptomsfordepression,anxiety, and PTSD (35%, 39%, 10% respectively) are much lower than comparable rates reported from PNA1(65%, 69% and 34%). Because symptom levels, using these standard cutoffs were so high in PNA1, our studydeveloped higher cutoffs to identify those at risk for the most severe forms of depression, anxiety and PTSD. WhiletheseratesarealsoveryhighinPNA1,theyaresubstantiallyandsignificantlylowerforPNA2respondents.

This report documents continuing high levels of conflict-related psychological symptoms in all parts of Aceh,and the urgent need for mental health services to be provided as a part of the peace-building and post-conflictrecovery process. It suggests that priority should be given to the North Coast, the Southwest Coast, the East Coast,and the Central Highlands, in this order. Significantly many respondents suffer the effects of complex trauma– many years of repeated experiences of violence and insecurity, not just a single episode of trauma, followed by areturn to a situation of safety and security. In some cases this will require sustained mental health services. At thesame time, this report finds substantially lower rates of psychological symptoms in PNA2 than in PNA1, which raisesimportantquestionsaboutresilienceandrecovery.

10. Although symptoms for depression and PTSD found in PNA2 are lower than for those found PNA1 symptoms, thetotal amount of traumatic events a person lives through is still the most predictive variable for suffering symptoms ofdepression and PTSD. The odds analysis (a statistical predictor of risk) shown in this report proves the association.AlthoughoddsratiosarelowerinPNA2thaninPNA1,highernumbersofreportedexperiencesofconflict-relatedeventsgreatlyincreasethelikelihoodthatpersonswillsuffersymptomsofdepressionandPTSDordiagnosablementalillnesses.Women have greater odds than men for suffering depression and anxiety. Men and women have almost equivalentoddsofsufferingPTSD.Andageeffectsshownoclearpatterns.Theoldestrespondents(aged54-82)areatgreaterriskfordepressionandgeneralanxiety.SomeoftheyoungergroupswhohadparticularlyhighriskforpsychiatricsymptomsinPNA1donotshowexcessriskinPNA2.

11. In an attempt to measure effects of psychological symptoms on social functioning, this study added a socialfunctioning scale, and found that overall rates of expression of difficulty in carrying out basic daily routines,such as manual labour, housework, were extremely low. However, even with this limited instrument, analysisdemonstrated highly significant relationships between social functioning and depression, anxiety and PTSD. Moststriking were relationships between psychological symptoms and expressed difficulty in basic livelihood activities –earning money, manual labour, farming and fishing. Although causal directionality cannot be established, this studyprovides evidence that persons who are depressed or chronically anxious find it particularly difficult to maintainlivelihood activities. Mental health problems have real consequences for social functioning, and mental health andlivelihoodinterventionsneedtogohandinhand.

12.The finding that psychological symptoms were substantially lower for PNA2 respondents, even those in areas thatsuffered quite high conflict, in comparison with PNA1 respondents was unexpected and raises a series of questions.

“A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007� “A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007 7

First,does this simply reflect lower ratesof traumaamong those respondents?Ourfindings show this isnot thecase.Symptoms are lower for those districts in which trauma levels are equal to those in PNA1. Do the differences reflectexposure to violence for a shorterperiodof time for those in thePNA2districts?Again, this isnot supportedby thestudy.PNA2foundsymptoms tobe lowerboth inAcehTimur,where theconflictwasas longas inBireuenandAcehUtara (PNA1), and in Aceh Selatan, where it was more recent. Do the differences that something changed betweenFebruary and Julyof 2006, and that symptom levels reflect this?Clearly, ourdata show that current levelsof stressorsweregreatlyreducedbyJuly.Atthattime,reportedratesofcurrentviolence(experiencingassaultandrobbery,seeingformer perpetrators) were much, much lower for all PNA2 regions than they had been for the PNA1 districts at thetimeofthatstudy.

This study thus suggests that with increased security and reduced levels of current stressors, general psychologicalsymptomsandcollectiveanxietywerereducedsignificantly.Clearly, individuals inthesecommunitiesareresilient,andcollective processes of recovery can move forward when there is security. At the same time, anxiety, depression, andPTSDremainedquitehighinthePNA2communities.Theresearchshowedthisiscloselyrelatedtopastexperiencesofviolence.Andourclinicalworkhasshownthatmentalhealthproblemsrelatedtotheconflictremainverysignificantinthesecommunities,andthattheseproblemsaretreatableusinginnovativeapproachestoprovidingmedicalcaretothesecommunities.

13. Despite the history of terrible violence levied against village populations in high conflict regions of Aceh, thesecommunities and most individuals within them remain remarkably strong and highly resilient. The PNA2 studysurveyed some of the local religious, cultural, and community resources, which people draw on to overcome conflictexperiences.91%reportusingprayerand54%reportconsultingareligiousspecialistforthispurpose.68%reporttalkingwitha friendor familymember,and56%report trying to forgetabout theexperience.At the sametime,33%reportlookingformedicalhelpspecificallyforthesepurposes.RatesformanyoftheseactivitieswerehighestintheEastCoastregion,whereviolencewasparticularlyintense.Theseareonlysmallindicatorsofthelocalresourcesandpsychologicalprocesses used in efforts to recover from the violence, larger scale political processes are almost certainly equallyimportantinthelongtermeffortsforcommunityandpersonalrecovery.

14. Althoughmistrustofgovernmentservicesremainsquitesignificant,respondentsexpressedahighlevelofwillingnesstoacceptmentalhealthassistance,whetheradministeredthroughGAMortheIndonesiangovernment.Thisdistinctionisof less importance today, where the governor and heads of many district governments are former GAM members.ThePNA2surveyfoundthatthereiswidespreadagreementthroughouthighconflictareasthatmentalhealthproblems,affectbothrespondentsandtheirfamilies,andhighlevelofrecognitionofproblemsassociatedwithstressandtraumainthesecommunities(referredtointhiswayinbothAcehneseandIndonesian).

15.Clinicians from IOM’s pilot mental health outreach programme in Bireuen, are finding that problems identified inthe PNA surveys are extremely prominent among persons diagnosed as suffering mental health problems. PTSDsymptomsareubiquitous,withnearlyaquarter(23%)ofallpatientsmeetingcriteriaforPTSD,and42%ofallpersonstreated for diagnosable mental health problems saying that their illnesses are related to conflict-related traumaexperience.Headtraumaappearsasanimportantclinicalphenomenon,withpatientsreportingcontinuingsymptomsdating back to being beaten or tortured. Clinical depression is common and is often associated with traumaticmemoriesoftheconflict,withsocialisolationandcommunitydivisiveness,andwiththelossesofpropertyandeconomicdevastationinflictedonthesecommunities.

Although the PNA2 study found reduced levels of symptoms in comparison with PNA1, treatable mental healthproblemsdirectlyassociatedwiththeconflictareveryrealinthesecommunities,andacontinuedurgencyshouldbefeltto provide care for these populations. This Norwegian Embassy funded pilot mental health project has demonstratedsuccessfullyoneapproachwhichiseffectiveintreatingthementalhealthproblemsidentifiedinthisstudy.

recommendAtions1. All programmes undertaken in rural Aceh should take account of the ubiquity and complexity of violence and its

psychological and social remainders in the affected communities.The legacy of accumulated traumatic events allacross Aceh as shown by the PNA data poses unique challenges even for programmes such as housing and schoolreconstructionthatarenotspecificallydesignedforpsychosocialassistance.Consultativeprocessesandkey informantinterviews,earlyandoften,andatthemostlocallevel,shouldinformthedevelopmentofanyintervention.Fromthese,anunderstandingofthehistoricalexperienceandcurrentsocialdynamics insitesofpostconflictassistancewillhelprefineprioritiesfortheprogrammeandensuresmoothimplementation.

2. The international community should recognize the continued urgency to provide mental health services to thecommunities most affected by the conflict. This report documents remarkable levels of traumatic violence enactedagainstordinarycivilianpopulationsinruralAceh,particularlyinthehighestconflictdistricts,sub-districts,andvillages.Thereport,aswellasfindingsfromthepilotmentalhealth interventionundertakenbyIOM,showthat thisviolenceis closely associated with high levels of depression, anxiety, PTSD, and neuropsychiatric conditions in thesecommunities. These problems have not gone away. Nearly two years after the signing of the MoU, which ended themilitaryviolenceinAceh,acutementalhealthproblemsremainacriticallegacyoftheviolence.Thereisurgentneedtoprovidemedically-basedmentalhealthresponses,aswellaspsychosocialandlivelihoodprogrammes,forvictimsinthesecommunities.

3. Provisionofmentalhealthserviceswillrequiresustainedinvestment inthe long-termdevelopmentof thehealthandmentalhealthsystemofAceh.Acehhasover4millionpeople.Ithasthreepsychiatrists.BuildingamentalhealthsystemthatwillreachthewidelydispersedcommunitiesofAcehshouldberecognizedasanimmediateandurgentneedandas a domain requiring sustained, long term investment. The mental health needs in these communities can only bemet through thedevelopmentofacompetentandeffectivehealth systemthatgives specialpriority tomentalhealthcare. International, national and provincial agencies should collaborate in strengthening the capacity of the publichealth system in general, and specifically in developing innovative solutions to the difficult task of providingcommunity-basedmentalhealthservices.

4. Specializedoutreachservicesshouldbesupportedtomeetthemosturgentmentalhealthneedsinhighconflictareasof Aceh. While the long term needs for mental health care in Aceh can only be addressed through investment inimproving the public mental health system, persons suffering the mental health consequences of violence, torture,anddisplacement shouldnotbemade towait. Specializedprogrammes thatprovidementalhealthandpsychosocialservices to the victims of the conflict should be given immediate support. The Norwegian Embassy funded MentalHealthOutreachprogrammedevelopedjointlybyIOMandHarvardMedicalSchoolhasbeenshowntobeoneeffectivemechanism for addressing acute and urgent needs in relatively isolated communities. Serious investment should bemadeinprogrammesthatbringservicesdirectlytothesecommunities.

5. Focusedefforts to treatpersons suffering theeffectsof complex trauma shouldbeundertaken in the contextof thedevelopment of specialized mental health and psychosocial programmes. In communities in which 15-18% of thetotal population and 25% of all men report being tortured, in which 50-70% of young men report being beaten onthehead,suffocated,orsubmittedtoneardrowning,inwhich50-65%ofallmenand15-20%ofwomenreportbeingbeaten, complex trauma is a common and important remainder of the violence. Managing mental health andpsychosocial problems associated with PTSD and complex trauma in relatively isolated settings with limited accessto mental health care is extremely challenging. It should be explicitly recognized that there is no single therapeuticmodality which is certain to be effective and sustainable. Instead, a commitment should be made to developinginnovativetherapeuticprogrammesinselectedsettings,todocumentingeachprogramme,andtocarefulevaluationoftheefficacyoftherapeuticapproaches.

6. Specialattentionneedstobegiventotheproblemofheadtrauma,braininjury,andlong-lastingdisabilityresultingfromtortureandviolenceassociatedwith theconflict.Both this reportand thePNA1documentedremarkablyhigh levelsofheadtrauma–beatingstothehead,strangulationandsuffocation,andneardrowning–thatwerearoutinepartoftorture,particularlyofmen,inhighconflictcommunities.Headtraumacancausebraininjuryandanoxia(lackofoxygentothebrain)thatcancauselonglastingemotional,cognitive,andbehaviouraleffects.Thesemayincludereducedabilitytoconcentrateandparticipateinlivelihoodtraining,impairedjudgmentleadingtowhatmayappeartoberoutineacting

“A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007� “A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007 �

out or even criminal behaviour, as well as personal suffering. Research should be undertaken to determine whetherspecializedprogrammesareneededtorespondtotheseproblems.Themedical,legal,andeducationalsystemsshouldbemadeawareoftheimportanceoftheseissuesforpersonswhohavesufferedtraumaticviolenceintheconflict.

7. Specialattentionneeds tobegiven to thementalhealthproblemsofolderpersons in thehighconflictareas.Whileyoungpeoplewere submitted toparticularviolenceduring theconflictandrightlydeserve specializedattention, thisreport suggests thatoldermenandwomenmaycontinue toexperience thehighest ratesofmentalhealthproblemsin these communities. Little attention has been directed to the effects of the conflict on the elderly. This findingsuggests the need for further research and the development of programmes to address the mental health andpsychosocialneedsofoldermenandwomeninthesecommunities.

8. Those districts and villages that suffered particularly egregious violence should be provided special attention in thedevelopmentofmentalhealthandpsychosocialservices.Exposuretotraumaticeventsduringtheconflictisthesinglelargest predictor of current mental health disability in both PNA1 and PNA2. A mapping of conflict events acrossAcehshowswheretheprioritiesareformentalhealthandpsychosocialservices.TheseincludethenorthandeastcoastdistrictsofBireuen,AcehUtaraandAcehTimur,aswellas theSouthwestcoastdistrictofAcehSelatan.Additionally,the mapping of conflict events can be taken down to the sub-district and village level, revealing the micro-localitieswhereserviceprovidersaremostlikelytofindthehighestconflict-relatedmentalhealthburden.

9. National and international agencies should recognize the continued need for livelihood interventions in highconflict areas. These should be linked specifically with mental health and psychosocial programmes. Damaged orlost livelihoods are more than just an unfortunate by-product of the conflict. In most cases, there was a deliberateand systematic attempt to destroy local economies that were seen by military forces as a strategic material basefor continued rebellion. These are devastating losses for the civilians in these communities and their recovery isinvariablyidentifiedbyrespondentsasafirstpriority.Buttheworstaffectedcommunitiesinneedoflivelihoodrecoveryarealsothecommunitieswiththehighestmentalhealthburden,whichmayhinderthesuccessofprogrammesdesignedfor material recovery. Transitional assistance for the rehabilitating destroyed fields and forest gardens, capital inputsfor restarting local business, livelihood training and the development of small trade cooperatives can all be seen aspsychosocial interventionson theirown,but thosewith themostdisabilitywillneedexplicitmentalhealthassistancetoaccompanytheirlivelihoodsupport.

10. The development of programmes for rural Aceh should include a systematic awareness of the long term effects ofdisplacementinthehighconflictcommunities.Nearlyhalfthesamplereporteddisplacementduetoconflict.Inmanyvillages,thisfigureisbetween90%and100%.ThepeoplelivinginformerhighconflictareasmustberecognizedasIDPswith all the vulnerabilities and needs that accompany their recent displacement experience. Displacement recoveryprogrammes–inparticularthereconstructionofdamagedanddestroyedhouses,schools,roadsandotherinfrastructure,and the recovery of lost livelihoods – are a prerequisite for any kind of broad psychosocial recovery in thesecommunities.

11. There is an enormous and lasting reservoir of memories of torture, violence, and displacement enacted againstcommunities and individuals in Aceh. Profound loss and a potent sense of injustice are remainders of the violence.Careful consideration should be given to specific efforts to work through these memories as a part of the on-goingpeace process in the context of rebuildingAceh.These effortswill in turnhave consequences for the larger goal oftrauma healing for individuals and communities. This report documents remarkable levels of violence enactedagainst civilian communities in Aceh. As a psychosocial needs assessment, the report focuses on specific clinical andmental health problems associated with this violence. However, the ubiquity of violence documented in this reporthas broader social and political implications which are critical to the larger goal of trauma healing for the peopleof Aceh. Special consideration should be given to finding mechanisms for commemoration, for working throughpainful and contested memories, for dealing with loss, and for reconciliation. These efforts have the potential tocontribute to traumahealingandultimately toaddressing thepainful remnantsof violence in thecommunitieswhocontributedtothisreport.

introduction

2 TheDepartmentofSocialMedicine,HarvardMedicalSchool,hasenteredacollaborativeagreementwithIOMtoprovidetechnicalconsultation

onmentalhealthandcommunityandenvironmentalhealthprojectsaimedatsupportingrecoveryfromthetsunamiandtheconflict inAceh,

andatinvestinginhumanresourcedevelopmentforhealthandmentalhealthinN.A.D.JesseGraymanistheIOMstaffscientistresponsibleforthe

PNAproject.ProfessorByronGoodandProfessorMary-JoDelVecchioGood,DepartmentofSocialMedicine,HarvardMedicalSchool,aresenior

scientistsontheproject,andMatthewLakomabiostatisticiananddataanalyst.

3 Good, B., M.-J. D. Good, J. Grayman, and M. Lakoma. 2006. Psychosocial Needs Assessment of Communities Affected by the Conflict in the

Districts of Pidie, Bireuen and Aceh Utara.InternationalOrganizationforMigration.

Between December 2005 and November 2006, a teamof researchers from the International Organization forMigration(IOM)and theDepartmentofSocialMedicine,Harvard Medical School, carried out a PsychosocialNeeds Assessment (PNA) in high conflict sub-districts in14districtsacrossAceh.Thestudywasdesignedtosupportthe work of IOM in responding to the psychosocialand mental health needs of individuals, families, andcommunities deeply affected by years of violence Aceh.More specifically, it was designed to provide highquality empirical data to determine levels oftraumatic events and mental health problems inhigh conflict communities throughout Aceh andto assess priorities for the development of mental healthservicesandpsychosocialinterventions.

The PNA study was undertaken in two phases. PNA1 (thePsychosocial Needs Assessment Part 1) was conducted inAceh Utara, Bireuen and Pidie funded by the CanadianDepartmentofForeignAffairsandTrade,IOMandHarvardMedicalSchool.2PNA2(thePsychosocialNeedsAssessmentPart 2) was the extension of the PNA1 project to highconflict sub-districts in 11 further districts throughoutAceh. Interviews were conducted in 10 districts in

July 2006, with funding from the World Bank, the multi-donor funded Decentralization Support Facility (DSF),IOM and the Harvard Medical School, and in Aceh Besarin November 2006, funded by IOM and the HarvardMedicalSchool.Theprimarypurposeofthecurrentreportis to provide analysis of findings from the PNA2 data;these data are compared throughout the report withdata previously analysed and published in the report onPNA1.3

The basic goal of the overall project was to evaluate thepsychosocial and mental health needs in communitiesdeeplyaffectedbytheyearsofconflictbetweenarmedforcesoftheRepublicofIndonesiaandtheFreeAcehMovement(Gerakan Aceh Merdeka, GAM), given the cessation ofviolence following the signing of the Memorandum ofUnderstanding of August 15, 2005, and to determinepriorities for developing programmes to respond to theseneeds. This report focuses on mapping out levels of pasttraumaticexperiences,psychologicalsymptoms,andmentalhealthneedsinthespecificregionsofAceh,eachofwhichhad quite different histories of violence. The reportdeliberately refrains from identifying groups orindividuals instrumental in theviolencevisiteduponthesecommunities.

“A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 200710 “A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007 11

reseArch desiGn And methodoloGy

GoAls of the reseArchThestudysetforthninegoals:1. TounderstandhowspecificcommunitiesanddistinctregionsinAcehhavebeenaffectedbytheconflict.2. Todeterminelevelsofspecifictraumaticeventssufferedbythegeneralpopulationandbyspecificsocialgroupswithin

eachoftheseregions.3. Todeterminelevelsofspecifictypesofsocialandpsychologicalproblemsresultingfromtheconflictwithineachofthese

regions.4. To observe and document the way community members speak about the conflict and about the demilitarization and

reintegrationprocesses.5. TomapoutandcomparelevelsofpsychologicalsymptomsandmentalhealthproblemsinthedistinctregionsofAceh,as

ameanstoestablishprioritiesforthedevelopmentofpost-conflictinterventions.6. To determine the priorities of community members and leaders concerning which psychosocial and mental health

problemsareregardedasrequiringthemosturgentresponse.7. Todeterminelevelsofdisplacementexperiencedbycommunitieswithinthesehighconflictregions.8. To determine what groups are at special risk for traumatic experiences, psychological symptoms and mental health

problems,andtoassesstheneedfortheprovisionofcommunitybasedmentalhealthservices.9. To identify resources in the community that may be useful for collaboration in developing particular psychosocial

interventions.

study desiGnThis PNA 2 study is a quantitativeand qualitative survey conducted invillages in high conflict subdistrictsacross 11 districts of Aceh. Villageswere selected using a randomstratified procedure, and individualswere selected using a randomizingprocedurewithintheselectedvillages.The survey used bothstandardized instruments todetermine individual respondents’experiences of violence and currentlevels of psychological symptoms, aswellasopen-endedquestionsdesignedtoelicitinformationaboutcommunityhistoryandindividualandcommunitypriorities for developing services.Here we describe the structureof the interviews, the sampling ofcommunities and respondents, the

ended questions, designed forAcehnese populations who haveexperienced decades of conflictand a tsunami, and widely usedvalidated scales allowing forcomparabilitywithprevious studiesof psychosocial needs of conflictand post-conflict populations inotherpartsoftheworld.

The interview began with basicdemographic questions followedby open questions. Respondentswereaskedif theywereaffectedbythe tsunami, whether the conflictaffected their lifeand thatof theirfamily, and whether anyone in thefamily, including the respondent,was a victim of the conflict. Theseopen questions were followed byquantitative measures drawn fromthe validated Harvard TraumaEvents scales, adapted specificallyto represent typical forms oftraumatic events common in thecommunities being surveyed.Scales includedayes/nochecklistof traumatic events experiencedduring the conflict and a yes/ nochecklist of experiences of currentstresses and traumatic events inthe post-conflict period. Levelsof emotional and psychologicaldistress were assessed with a set ofgeneralself-assessmentquestions.

These elementary questions werefollowed by a 25 item version ofthe Hopkins Symptom Checklistfor Depression and Anxiety, ascale used widely in disaster andtrauma community assessments ofemotional distress, and the

42 item Harvard TraumaQuestionnaire(HTQ).TheHTQisabroadmeasureof trauma-relatedsymptoms,whichincludes16itemswhich can be used to assess PostTraumatic Stress Disorder (PTSD).In addition, items designed tocapture popular discourses aboutdisturbing post-tsunami andpost-conflict experiences wereintegrated into the quantitativemeasures to elicit experiences ofnightmares, ghosts, spirits andhearing voices of people who haddied.

A four item measure was includedfrom the Harvard TraumaQuestionnaire to assess presenceand severity of head traumathat might have produced braininjury; the questionnaire asksspecifically about beatings to thehead, suffocation or strangulation,near drowning and other physicalinjuriestothehead.

The survey concluded with closedand open questions regarding therespondent’s perceptions of whatcommunity mental health servicesare most needed, their opinionsabout which groups suffered mosttrauma due to the conflict orare at the greatest mental healthrisk, assessments of who providescare and to whom communitymembers can turn to overcomebad experiences that remain fromthe conflict, attitudes about thepublic health care services, andcomments and suggestions aboutthepost-conflictpeaceprocessand

communityrebuilding.

Thesurveywasdesignedtofacilitatecomparabilitywithother studiesofconflict-affected populations withthe intention of drawing lessonsconcerning useful mental healthinterventions from previous cases.Asignificantpartof thesurveywasalso devoted to open questionsallowing for the specificity ofAcehneseexperiencestodeterminethe interpretation and meaning ofcomparativeanalysesandlessons.

Lessons learned from the PNA1researchinPidie,BireuenandAcehUtaraallowedforsomerevisionsinthequestionnairebeforethePNA2research. Two sections were addedto the formal questionnaire. Thefirst was a social functioning scaleadaptedtoruralAceh,whichitwashoped might help assess to whatextent psychological symptoms areaccompaniedbyactualpsychosocialdysfunction in the community.The second was devoted entirelyto conflict-related displacementexperience of communitiesand individuals, an issue whichemergedinthePNA1research,andwhich is important to the largermissionofIOM.

The survey was designed as amental health needs assessment.The questionnaires did not focuson determining who was primarilyresponsibleforcommittingviolenceagainst Acehnese communitiesand individuals.The resultsof thisstudythusdonotmeetthespecific

research teams, and the regions thatwere included in the study. PNA2interviewsusedthesame instrumentsas used in PNA1, with somequestionsadded toaddressquestionsraised by PNA1. PNA2 samplingprocedures followed those of PNA1to the extent possible, allowing forcomparisons across high conflictregionsofAceh.

• the interview structure Thestudyincludedtwocomponents:

key informant interviews and aformalsurveyofrandomlyselectedadultsaged17yearsandolder.

The key informant qualitativeinterviewsweredesignedtoexplorethe historical context of theconflict in specific regions and

communities, to discuss how theconflict affected communitiesover time and whether certainsegments of the population weremore vulnerable than others.Community leaders were asked toassess their community’s prioritiesforpsychosocialandmentalhealthservices,andtodescribetheirviewsof the best ways to respond to theeffectsoftheconflict.

The formal survey interview wasdesigned to measure levels of pastexperiences of traumatic eventsassociatedwiththeconflict,toassessexperiences of current stressorevents, and to identify levels ofcurrent psychological distressassociated with these experiences.The survey combined open-

“A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 200712 “A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007 1�

criteria usually required in humanrights investigations. Instead, theresearch attempted to connectpast traumatic experience withcurrenthealthneedsasameanstoinforming the development ofmental and psychosocial healthservices in conflict affectedcommunitiesacrossAceh.

• phases of fieldwork and Geographic coverage

The planning and implementationofthePNAresearchspannedalloftheyear2006.PhaseOneresearch(PNA1) focused on three districtsonthenorthcoastofAceh–Pidie,BireuenandAcehUtara–becausethey were known to be central tothe historical development of theFree Aceh Movement and a siteof some of the longest and mostintensive conflict. Planning forPhase One began in December2005, led by teams from HarvardMedical School and IOM, andthe field research was conductedduring February 2006. SyiahKuala University’s Centre for theDevelopment of Regional Studies,inBandaAcehwastheimplementingpartnerforthefieldsurvey.

Phase Two research – PNA2 – wasdesigned to extend the survey toall of the remaining districts ofAcehwhichexperiencedsignificantlevelsofconflictandinwhichfieldresearch was feasible. The primaryPNA2fieldinterviewingwascarriedout in July 2006 in ten districtsthroughout Aceh (see SampleDescription by Site and Dates,Chart 1). In November 2006,interviewing was conducted inAcehBesar,inordertoimprovethegeographic scope of the research.Thus,excludingtheislanddistrictsand urban municipalities of Acehprovince, the only mainlanddistricts not included in the PNAresearch are Aceh Jaya (which stillposes transportation challenges

healers. Harvard and Syiah Kualateams held group discussions inseveral communities, particularlyamong women, and together withthe IOM coordinator convened afocus group discussion with GAMmembers, including commanders,ex-combatants, amnestied prisonersand civilians. The PNA2 studyincludedasupplementalqualitativecase study of the history of theconflict in Aceh Selatan andthe enduring psychosocial andeconomic consequences of theviolence.

• selection of Villages – design and implications of pnA sampling strategies

A totalof75villages in11districtswere surveyed during the PNA 2research. 30 villages were surveyedduring PNA1, 65 villages duringtheprimaryphaseofPNA2and10more during the survey in Aceh

duetotsunamidamage)andAcehSingkil (which was not a highintensityconflictarea).Dataanalysisandreportingforallthreephasesofthe psychosocial needs assessmentareconductedandpaid forby theHarvard Medical School and byIOM’s contract with the HarvardMedicalSchool.

• researchers and fieldwork teams

TheoverallPNAwas conductedasa collaborative project betweenIOM and members of theDepartment of Social Medicine,Harvard Medical School. Inaddition, PNA1 research featuredsignificant collaboration withfaculty, staff, and field researchersfrom Syiah Kuala University’sCenter for the Development ofRegionalStudies,undersupervisionfrom sociologist Professor BahreinSugihen. IOM and its staffcoordinatedthestudyandprovideda project director. The HarvardMedicalSchoolteamwasresponsiblefor overall scientific design of theproject and for analysis of thequantitativedata.TheHarvardandIOM teams share responsibility forthewritingofprojectreports.

Permission to carry out socialscienceandpublichealth researchin Acehnese villages, was securedfrom the Community Protectionand State Unity Board at Aceh’sGovernor’s Office. Subsequentphases were supported by MoUagreements between IOM andthe Ministry of Health, and alsobetween IOMand theDepartmentofSocialWelfareforIOM’sworkinPost-ConflictactivitiesinAceh.Theprojectwasreviewedandapprovedby the Institutional Review BoardofHarvardUniversity.

The Harvard team and the IOMfield coordinator developed thedesign of the overall study and

the survey in December 2005.Questionnaires were translatedand back translated, then pre-tested in Banda Aceh in January2006.Fieldstafftrainingsprecededeach phase of the fieldwork,presenting opportunities for morequestionnairetestingandrevisions.Revisions of the questionnairefor the PNA2 study included theaddition of a social functioningscaleandquestionsondisplacementand internal migration, as well asminor changes aimed at clarifyinga few questions. Village selectionstrategies were also tailored to thewider reach of the second phaseof the study and are explainedbelow. Field research for PNA1was conducted by a team ofresearchers under the direction ofthe Center for the Developmentof Regional Studies at Syiah KualaUniversity, supervised by the IOMproject coordinator. Field researchfor PNA2 was conducted by ateam hired directly by IOM andsupervised by the IOM projectcoordinator.

For both PNA1 and PNA2, everyvillage was visited by one team,which during the course of oneday conducted key informantinterviews and survey interviews,mostly resulting in 18 interviewswith randomly selected villagers.Each team consisted of six surveyinterviewers and a team leaderresponsibleformanagingthesurvey,liaising with local governmentand community leaders to ensuresmooth entry into the villages,and conducting the key informantinterviews.Additionalkeyinformantinterviews were conducted by theHarvardandSyiahKualateams,theIOMcoordinatorandIOMtechnicalassistants. The IOM coordinatorinterviewed doctors, nurses, and/ormidwivesfromthenearestpublichealthclinicswhereresearchteamswere visiting and also traditional

Besar.Thesamplingprocedurewasdesigned to select a representativesampleofvillagesfromhighconflictareas ineachdistrict surveyed,nota random sample of all villages ineachdistrict.Selectionmethodologyin four districts surveyed duringPNA2 was somewhat differentfrom the selection procedure usedintheothertendistricts,aswillbedescribedbelow.

Selection criteria depended uponthe identification of high conflictsub-districts(kecamatan)andvillageswithineachdistrict.BothPNA1andPNA2 based the sampling of highconflictsub-districtsandvillagesonthemethodologypreviouslyusedbyIOMtodeterminerecipientvillagesfor its European Commission-funded Post-Conflict CommunityReintegration Programme(Makmue Gampong KareunaDamai – MGKD), which made

use of the World Bank KecamatanDevelopment Programme (KDP)staff as facilitators. The MGKDprogramme determined highconflict sub-districts based upon aconflictstressassessmentpreviouslyconducted by the World Bank,supplementedbyanecdotalreportsfrom sub-district governmentofficials, local NGOs, local GAMleaders and the IOM staff workingfor the MGKD programme in thearea.Afteridentifyinghigh-conflictsub-districts in this way, the Post-Conflict Community ReintegrationProgramme then identifiedbetween five and ten high-conflictvillages within these high-conflictsub-districts.

ForthethreedistrictsinPNA1andthesevendistrictsinPNA2thathadMGKD programmes, 50% of thevillagesselectedforthesurveywererandomly sampled from the short

chArt 1: implementAtion of pnA1 And pnA2

phase districts fieldwork dates implementing Groups

donor Analysis & reporting

pnA1 AcehUtaraBireuenPidie

February2006 IOM&SyiahKualaUniversity

Canada&HMS-IOMContract

HMS&IOM

pnA2 (A) AcehBaratAcehBaratDayaAcehSelatanAcehTamiangAcehTenggaraAcehTengahAcehTimurBenerMeriahGayoLuesNaganRaya

July2006 IOM WorldBank,theDecentralizationSupportFacility(DSF),HMS&IOM

HMS&IOM

pnA2 (b) AcehBesar November2006 IOM HMS-IOMContract

HMS&IOM

mainland districts not covered AcehJayaAcehSingkil

municipalities and island districts not covered

BandaAcehLangsaLhokseumaweSabangSimeuleu

“A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 200714 “A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007 1�

list of high conflict villages thatparticipated in the programme.The other 50% of the villages inthesamplewererandomlyselectedfrom the remaining villages inthese same high conflict sub-districts that did not receive theprogramme benefits. (Analysis oftraumaexperiencesbyrespondentsin MGKD and non-MGKD villagesthus provides an opportunity toevaluate the validity of the WorldBank and IOM designation ofthese villages as high conflict).However, four districts in thePNA2 study – Bener Meriah, AcehTengah,GayoLuesandAcehBesar–werenotrecipientsoftheMGKDprogramme, so another villageselection methodology had to beemployed.Qualitativeinformationfrom IOM staff working in thesedistricts was collected and a shortlist of high conflict sub-districts inthese four districts was generated.Villages were then randomlyselected from these sub-districtsfor inclusion in the PNA2 survey.This difference in samplingprocedure leads to the chance ofa small sampling bias toward lesshigh conflict villages in these fourdistricts.

The village selection methodologyand the phased administration ofthesurveyhaveseveralimplications,whichshouldbecleartothereaderofthisreport.First, thefindingsofthePNAstudycannotbegeneralizedto the whole populations of thedistricts in our sample or to Acehas a whole. The findings arerepresentative of the high conflictsub-districts of these 14 districtsonly. This survey was intentionallydesigned to assess psychosocialneeds of members of highconflict communities, not of allcommunities of Aceh. Second, thephasedadministrationofthissurveymeans that findings of differencesbetween PNA1 and PNA2 districts

interview were prepared daily bythe team leaders. Topics coveredlocal conflict history, localunderstandings of trauma andmental illness, stories of stress,trauma and mental illness in theircommunity related to the conflict,local resources and priorities formanaging such conditions andopinionsaboutthepeaceprocess.

• data entry and Analysis After completion of each phase of

the research, data from the surveyinstruments were entered into adatabase in Banda Aceh and thentransferred to the teamatHarvardMedical School for cleaning,development of variables, initialdescriptive analyses and morecomplex statistical analyses (usingSAS). All quantitative data analysishasbeenconductedbytheHarvardresearch team, as part of theHarvard - IOM collaboration.Analysesaredesigned inparticularto identify levels of traumaticexperiences, psychological distressandpsychiatricdisorders,riskfactorsassociatedwiththesedisordersandpriorities in the community formental health and psychosocialinterventions.

Open-ended, qualitative responseson the interview forms were alsoentered into the database, sortedby district and gender, coded foremergent themes and used formore culturally sensitive analyses.The team leaders wrote extensivenotes (in Bahasa Indonesia) abouteach key informant interview, aswell as summary analyses abouteach village. These data have alsobeen important for the analyses

mayreflectnotonlybydifferencesamong districts but also bydifferences across time. Althoughit is not obvious that there wouldbe a strong effect of interviewingpersons in February versus July2006,itshouldberememberedthatFebruary 2006 was only 6 monthsaftersigningoftheMoU(15August2005)whenmanypeoplemayhavedoubtsthatthepeaceprocesswouldhold.

The peace process continuedactivelythroughthisperiod,leadingtoaquitedifferentlevelofsecurityin many regions by July 2006.Findings of this research thusrequire quite nuancedinterpretation. Third, the lackof MGKD programmes in fourdistricts,includingthreeintheAcehhighlands,introducesthepossibilityof a small bias toward less highconflict villages, as noted above.Fourth, some districts had verywidespread violence, while othershad intensive violence but only inveryspecific“hotspots.”Inthelattercase,samplingproceduresbasedonsub-districts may not account forsuch localization of violence,allowing for difficulties ingeneralizing the findings of theresearch in these districts. Andfinally, to make sense of the datafrom 14 districts, it has beennecessary to group districts intoregions,aswillbedescribedbelow.Such groupings allow for overallcomparisons and the establishingof priorities for mental healthservices,buttheymaynothighlightthe presence of smaller areas ofintensive need for trauma-relatedservices.

• selection of respondents Upon arrival in a selected village,

teams would report first to thevillageheadorthevillagesecretaryto explain the reason for theirresearch visit. Team leaders

would work together with thesecommunity leaders to generate arandom sample of 18 householdsin the village. Most villages keep arecord of the households, and sorandom selection was typically aneasy process. Each surveyor thusconducted three interviews onaverageinavillage.Uponarrivalatahousehold,surveyorswouldselecta random respondent from theresidentsaged17andolder.Therewerenostratificationcriteriaatthevillagelevel.

• interview mechanics If the selected respondent agreed

to the invitation to participate inthe survey, a consent form with adescriptionof theprojectwasreadtogether with the respondent,covering procedures, risks andbenefits, questions or concerns,confidentiality, and the voluntarynature of participation, usingAcehnese language or anotherdialect when preferred. The formwas then signed and dated bythe interviewer only and a copywas provided to the intervieweeincluding a list of organizationshelpfulindealingwithpsychosocialproblems. Each questionnaire wasassigned a numerical code leavingno personal identifiers in orderto ensure the anonymity of allrespondents. Researchers followedstandard consent protocols thatwere approved by the HarvardUniversity Faculty of Arts andSciences’ Institutional ReviewBoardfortheProtectionofHumanSubjects.

Team leaders’ interviews withkey informants were less formal.Consent was obtained, and teamleaders held conversations, usuallyin themeunasah– the communitycenter used primarily by the menof the community but also bywomen when receiving outsideresearchteams.Fieldnotesoneach

outlined in the following pages ofthisreport.

reportinG by six GeoGrAphic reGionsAnalyses of the survey data arereported by gender, age groupings,and by PNA1 versus PNA2. However,to make sense of data collected in105 villages in 14 districts of Acehfor both PNA1 and PNA2, we havegrouped districts into six meaningfulgeographic regionsbasedonconflicthistory, geography, ethnicity, andthe methodological differences intime and village selection describedabove. Here we describe the sixgeographic regions that serve asthebasisformanyofthecomparativeanalysesinthisstudy.• north coast Thehistoricalandideologicalroots

of the conflict in Aceh are foundalong the North Coast districts –Pidie, Bireuen, Aceh Utara (NorthAceh), the original districtsincluded in Phase One of thepsychosocial assessment (PNA1).Tiro, a sub-district in Pidie, isparticularly historically important,asitwasfromthisarea,thatGAM’sleader, Hasan Di Tiro, launchedthe Aceh Free Movement (GAM)in1976.

The North Coast districts are also

the most densely populated anddeveloped, with Aceh’s mainhighway from Banda Acehrunning along the coastlinethrough these districts to Medanin North Sumatra province, theeconomic hub and largest city inSumatra. Lhokseumawe, in Aceh

Utara, home to Exxon-Mobil’shugely profitable oil and naturalgas facilities, and a source ofrevenueforJakartathroughoutthe1980sand1990s,becamea symbolof center-periphery inequalitiesbetween Jakarta and Aceh and aflashpointforrebelactivity.4

After dozens of GAM troops weretrainedinLibyafrom1986to1989,the organization consolidated itscommand structure in thesedistricts. Starting in 1989, theIndonesian government declaredAceh a Military Operations Zone(DaerahOperasiMiliter,DOM)anduntil the resignation of PresidentSoeharto in 1998, the Indonesianmilitary’s counter-insurgency effort– in which thousands of civilianswere killed – was concentratedalongtheseNorthCoastdistricts.56

• east coast HeadingSoutheast fromthenorth

coast, the East Coast districts ofAcehTimur(EastAceh)andAcehTamiangonthewaytowardMedanshare the same historical andgeographic features as the northcoast districts in our sample, butthe population density decreasesto less than a third of the northcoast region. Aceh Timur inparticular spans a long stretch ofthe Sumatran coast and reachesdeep into a vast mountainousinteriortomeetthebordersofthemountaindistrictsofBenerMeriahandGayoLues.Travelling throughthe rural sub-districts of AcehTimur between Lhokseumawe andLangsa municipalities, one passesthe well-known conflict regions ofIdiandPeureulak.

4 Kell,T.1995.The Roots of Acehnese Rebellion,1989-1992.(Publication no. 74).Ithaca,N.Y.:CornellModernIndonesiaProjectSoutheastAsiaProgram

CornellUniversity.5 Ibid.6 Schulze, K. E. 2004. The Free Aceh Movement (GAM): Anatomy of a Separatist Organization. Policy Studies, 2. Washington, DC: East-West Center

Washington.

“A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 20071� “A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007 17

Villages far off the main highway,whether heading towards coastalfishing villages or towardsthe mountains, and palm oilplantations formerly settled byJavanese transmigrants, still retainthe qualities of a rogue, isolatedfrontierlands, not least becauseAcehTimurissofarawayfromthereach of provincial governmentin Banda Aceh. During theconflict, whole villages of Javanesetransmigrants were emptied andthe local economies of entire sub-districts have been crippled by themassexodus.

To date, most transmigrants havenot returned, and many formertransmigrants have chosen to selltheir land in Aceh Timur ratherthanreturninspiteoftheimprovedsecurity environment. PopulationdensityinAcehTamiang,arecentlyformed district which split fromAceh Timur, increases as oneapproaches Medan in NorthSumatra and ethnic identificationsamong the population shift fromAcehnese to Melayu. Conflictactivity in Aceh Tamiang was notinsignificant along the provincialborder areas here, but reduced incomparisonwithallcoastaldistrictsto the west. One important reasonwhy thePNAanalysis separates theeast coast districts from the northcoast districts is because thesedistricts were surveyed as partof PNA2 in July 2006, nearly sixmonthsafterthePNA1researchinthenorthcoastdistricts.

• central highlands The mountainous Central

Highlands of Aceh separating theNorth from Southwest coasts have

Indonesian forces to GAM’sexpansion along the Southwestcoast districts from 2000 until thepeace agreement was swift, brutalandspectacular.Thisregion’smorerecent and chaotic conflict historycreated particularly terrifyingsecurity conditions and paralyzingmutual suspicion withincommunities.7

• Aceh besar Finally, Aceh Besar district has a

unique geographic orientation,being the district at the northtip of Aceh province (and thewholeofSumatra island),andalsosurrounding the provincial capitalof Banda Aceh, arguably the mostcosmopolitanareaoftheprovince.

Aceh Besar is distinguished inthe PNA2 study because it was thelast district to be included in thestudy and the only district to beinterviewed in November 2006,overayearafterthesigningof thepeace agreement. Aceh Besar is

the most ethnic diversity. A greatmajority of respondents from thedistricts of Bener Meriah, AcehTengah (Central Aceh) and GayoLues self-identify as Gayo, thelargest highland ethnic groupin Aceh with a history of mixedsupport for GAM or Indonesianforcesduringtheconflict.

Many respondents from thecentral highlands self-identifyas Javanese, an importanttransmigrant demographic citedbyGAMasevidenceofIndonesia’scolonial expansion into Acehand a justification for recruitingmembership in the mountainsduring a period of rapid GAMexpansion after DOM. Vulnerabletransmigrant groups and manyGayo were easily recruited bythe Indonesian military to formpro-Indonesia militia groups inthese districts which introduced aunique and complicating conflictdynamic in the central highlandsthat differed significantly from thecoastal regions. All three districtsdefined as “central highlands” inthis research used the alternativevillage selection methodologydescribedaboveduetotheabsenceof a Post-Conflict CommunityReintegration Programme at the

timeofthefieldwork.

• southeast highlands The majority of respondents from

the mountain district of AcehTenggara, (Southeast Aceh), self-identifyaseitherGayoorAlas, thesecondlargest indigenousminoritygroup in Aceh. Southeast Acehexperienced the least violencerelative to the other five regionsdescribed in this report. Many

conflict IDPs from other parts ofAceh found temporary safe havenin this region. For example, lessthan 10% of respondents in AcehTenggara described themselves orsomeoneintheirfamilyasavictimoftheconflict,whereasinallotherdistrictsatleast40%answeredyestothesamequestion.Forthisreason,SoutheastAcehisausefulreferencepoint of comparison againstother regions that experiencedsignificantly higher conflictintensity.

• southwest coast All 461 respondents from the

Southwest Coast districts of AcehBarat (West Aceh), Nagan Raya,AcehBaratDaya(SouthwestAceh)and Aceh Selatan (South Aceh)self-identify as Acehnese exceptone Javanese respondent in AcehBarat.

These districts were targetedfor heavy recruitment by GAMduring two separate cease fireagreements after 1999, mostprobably due to shared ethnicidentification with the northand east coast GAM strongholds.Membership increase here,sometimes under duress, occurredtoo rapidly for new recruits todevelop strong ideologicalidentification with GAM’s struggle,and often attracted a thuggishcriminal element, among otheropportunists, who were easilyconvinced to defect or becomeinformants for the Indonesianmilitary or police intelligenceunits.

All the key informant interviewsdetail how the response by

7 The authors acknowledge the important contributions to our PNA research by anthropologist John MacDougall during two weeks of

fieldwork in Aceh Barat Daya and Aceh Selatan in July 2006. His work helped us understand, in graphic detail, how this region’s messy and

recent onset conflict history created particularly terrifying security conditions throughout this region and mutual suspicion within

communities.

also the fourth district in thesample that used the alternativesampling methodology describedabove due to the lack of a Post-ConflictCommunitiesReintegration(MGKD) Programme there.Inclusion of Aceh Besar ensuresnearly complete geographicrepresentation across mainlandAcehforthisresearch.

reportinG resultsThe overall strategy for analyzingresultsofthePNA2dataisasfollows.First, as with the PNA1 report, thefocus is indeed on assessing needsfor mental health and psychosocialservices.Afterademographicoverviewof the sample, levels of traumaticevents experienced by personsin high conflict communities arereported, thenlevelsofpsychologicalsymptoms and mental illnesses,evidence suggesting which groupsareatparticularrisk,localperceptionsof needs for services, and localresources used for coping with the

remainders of traumatic violence.Second, running throughout is acomparison of PNA2 results fromthose of PNA1. Differences bothin levels of traumatic events and inpsychological symptoms allow us toaskwhyoverallPNA2findslowerlevelof psychological symptoms thanPNA1, and to raise questions aboutthe implications for the need forcontinued investment in mentalhealthservices.

Third,resultsofPNA1andPNA2arejoined to ask questions about whichregions should be given specialpriority in the development ofmentalhealthservicesinpost-conflictregions of Aceh. Although there area number of scientific and academicquestions that can be pursuedthrough analysis of these data, theprimary focus here remains onthe need for mental health andpsychosocial services to respond tothe mental health consequencesof traumatic violence inflicted oncivilianpopulationsofAceh.

“A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 20071� “A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007 1�

sAmple demoGrAphics: compArinG pnA1 And pnA2

The two Psychosocial NeedsAssessmentsurveys(PNA1andPNA2)included 1972 adults over the age17 from105villages in14districts inAceh.Tables1.1through1.6presentthe demographic characteristics ofthe PNA1 and PNA2 sample. Table1.1indicatesthegroupingsofdistrictsintoregionsforPNA1andPNA2andsamplesizefordistrictsandregions.

Table 1.2 provides gender and agebreakdown by region for the PNA2

sample.Theoverallsamplewasequallydivided between men and women.There was some gender variation byregion, with a slight over samplingof men, except in Aceh Besar where63% of respondents were women.Given that men and women oftendiffer in types of traumatic eventssufferedandlevelsofsymptoms,manysubsequent tables include gender asa critical distinguishing variable. Theoverall gender and age distributionshows that the sampling procedure

produced a good, representativesampleofadultsinthesehighconflictcommunities for both PNA1 andPNA2, and that overall, the samplesarelargelycomparable,asisindicatedfurtherbyTable1.3.

ThePNA2sampleincludesasomewhatyounger population, with 68 percentof respondents age 40 and under, incontrast to 56 percent of the sampleinPNA1. Otherwisethetwosamplesare remarkable similar in terms of

tAble 1.1 district locAtions And sAmple siZe of pnA1 And pnA2 respondents

location%

pnA 1 n = 596

% pnA 2

n = 1,376

% pnA total n= 1,972

north coast districts (pnA1) 100 - 30.23

AcehUtara(NorthAceh) 30 - 9.08

Bireuen 30 - 9.13

Pidie 40 - 12.02

Aceh besar (pnA2) - 13 9.13

AcehBesar 13 9.13

east coast districts (pnA2) - 18 12.48

AcehTamiang 3 1.83

AcehTimur - 15 10.65

central highlands districts (pnA2) - 23 16.57

AcehTengah - 7 4.61

BenarMeriah - 10 7.40

GayoLues - 6 4.56

southeast highlands districts (pnA2) - 12 8.22

AcehTenggara - 12 8.22

southwest coast districts (pnA2) - 34 23.99

AcehBarat(WestAceh) - 14 9.69

AcehBaratDaya(SouthwestAceh) - 4 2.74

AcehSelatan(SouthAceh) - 13 9.13

NaganRaya - 3 1.83

tAble 1.2 sAmple chArActeristics: Gender, AGe And ethnicity by reGion

%Aceh besar

(n=180)

%east coast(n=246)

%central

highlands(n=327)

%southeasthighlands(n=162)

%southwest

coast(n=461)

%pnA2:

overall totaln=1376

%pnA1: north coastn=596

%pnA1+pnA2:

overall totaln=1972

Gender

Male 37 51 55 58 49 50 53 51

Female 63 49 45 42 51 50 47 49

Age

17-29 38 36 37 36 33 35 25 32

20-40 27 28 33 35 36 33 31 32

41-53 16 23 17 16 18 18 24 20

54-109 19 13 13 14 13 14 20 16

ethnicity

Acehnese 99 80 9 8 100 64 99 75

Gayo 0 0 80 30 0 23 0 16

Alas 0 0 <1 54 0 7 0 5

Javanese 0 15 10 4 <1 6 <1 4

Melayu <1 4 <1 2 0 <1 0 <1

Batak 0 <1 <1 1 0 <1 <1 <1

maritalstatusandhousingownership.A greater percentage of PNA2respondents have achieved middleandhighschooleducation,indicativeof the larger number of youngerrespondents; yet the percentage ofrespondents with no schooling (8%PNA1/ 9%PNA2) or with advancededucation (5% PNA1/ 6% PNA2) issimilar.

The ethnicity data reported inTable 1.2 are self-identifications bythe respondents in the sample. Forexample,mixedethnicityrespondentsfromAcehneseandJavaneseparents,particularlybyparentswhoaresecond

and third generation Javanese inAceh, may very well self identify asAcehnese. Likewise, the diversity ofethnic minorities and speakers ofdifferent local languages inAceharenot always reflected in respondents’self-identifiedethnicities.ResidentsinAcehTamiang,forexample,mayself-identifyaseitherMelayuorAcehnese,but speak Bahasa Tamiang. This isespecially evident in the data fromAcehSelatanAcehSelatanwherethemajorityofrespondentsdonotspeakAcehnese as their first language.Most residentsofAcehSelatan speakeither Bahasa Aneuk Jame, BahasaKluet, or Bahasa Singkil as their

first language, but everyone in thesample self-identified as Acehnese,thus highlighting the especiallycomplicated relationship betweenlanguage and ethnicity in theconstitution of self-identity in acountry as vast and diverse asIndonesia. In the highlands of Aceh,on the other hand, ethnic minoritygroups there more readily identifythemselvesasGayoandAlas.Javanesetransmigrant populations are heavilyrepresented in the east coast andcentral highland districts in thesample. The inconsistentdistributionof ethnic groups across Aceh wereone of many complicating factors

“A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 200720 “A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007 21

affecting the patterns of violenceduringtheconflict.

As described above, one of themechanisms for sampling highconflict villages was to draw on thedata from IOM’s EC-funded Post-Conflict Community ReintegrationProgramme (Makmue Gampong Kareuna Damai – MGKD). In thosedistricts with a MGKD programme,approximately 50% of the villageswere drawn from those includedin the MGKD programme sample,and approximately 50% were drawnrandomly from other villages in the

high conflict sub-districts. However,four districts – Aceh Besar and theCentral Highlands districts – did nothaveMGKDprogrammes.

Although all villages were sampledfrom high conflict sub-districts, andother means were utilized to identifyhigh conflict villages in the fourdistricts with no MGKD programme,it is possible that this difference insampling procedure might have ledto a bias toward less high conflictvillagesinAcehBesarandtheCentralHighlands, and thus for the PNA2sampleoverall.

Tables1.4and1.5showthedistributionof respondents living in villagesselected for IOM’s MGKDprogramme. Forty-six percent of thePNA1 respondents resided in villageswith a Post-Conflict CommunityReintegration Programme (MGKD),whereas only 33 percent of PNA2respondents lived in such villages. Acareful examination of respondentsliving in villages with MGKDprogrammes in each district showsthat the sampling strategy did notlead uniformly to 50% of allrespondents living in villages withandwithoutMGKDprogrammes.

tAble 1.3. pnA1 And pnA2 respondent chArActeristics by sex, AGe, mAritAl stAtus, educAtion And housinG

demographicspnA1: n=596

pnA2:n=1376

pnA1+pnA2:n=1972

Gender

Male 53 50 51

Female 47 50 49

Age

17-29 25 35 32

20-40 31 33 32

41-53 24 18 20

54-109 20 14 16

marriage status male female total male female total male female total

NeverMarried 20 12 16 21 14 17 20 13 17

CurrentlyMarried 77 70 74 77 70 73 77 70 73

DivorcedorSeparated 2 3 2 <0 3 2 1 3 2

Widowed 2 16 9 2 13 8 2 14 8

schooling male female total male female total male female total

NoSchooling 6 11 8 7 12 9 7 12 9

PrimarySchool 48 48 48 42 41 41 44 43 43

MiddleSchool 23 21 22 25 25 25 24 24 24

SecondarySchool 20 13 17 21 16 19 21 15 18

AssociatesDegreeorProfessional 2 5 3 2 3 3 2 3 3

UniversityEducation 2 2 2 3 3 3 2 3 3

housing male female total male female total male female total

LiveinOwnHome 84 87 85 89 90 89 87 89 88

LivewithFriendorRelative 8 3 6 5 6 6 6 5 6

LiveinAbandoned/DestroyedHome 4 2 3 2 1 1 3 2 2

RentHousing 2 5 3 3 2 3 3 3 3

LiveinBarracksorTent 2 3 3 <0 <0 <0 1 1 1

tAble 1.4 respondents by presence or Absence of post-conflict communities reinteGrAtion proGrAmme (mGKd) by pnA1 And pnA2

iom dataset (n) mGKd

nrow percentage

%mGKd+

%mGKd-

pnA1 (n=596) n=27546%

n=32154%

pnA2 (n=1376) n=45433%

n=92267%

total (n=1972) n=72937%

n=124363%

tAble 1.5 respondents by presence or Absence of mGKd proGrAmme by reGion And district

location%

pnA1: n=596

%pnA2:

n=1,376

%total

n=1,972

mGKd+ mGKd- mGKd+ mGKd- mGKd+ mGKd-

north coast (pnA1)Aceh utara bireuenpidie

46181513

54121527

----

----

13.945.584.563.80

16.283.504.568.22

Aceh besar (pnA2) - - - 13 - 9.13

Aceh besar - - - 13 - 9.13

east coast (pnA2) - - 8 10 5.78 6.70

Aceh tamiangAceh timur

--

--

-8

37

-5.78

1.834.87

central highlands (pnA2) - - - 24 - 16.57

Aceh tengahbenar meriahGayo lues

---

---

---

7116

---

4.617.404.56

southeast highlands (pnA2) - - 7 5 4.56 3.65

Aceh tenggara - - 7 5 4.56 3.65

southwest coast (pnA2) - - 18 15 12.68 10.70

Aceh baratAceh barat dayaAceh selatannagan raya

----

----

8181

6351

5.380.915.480.91

4.311.833.650.91

“A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 200722 “A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007 2�

trAumAtic eVents

trAumAtic eVents durinG the conflict This survey found that respondentsfromPNA2inthe11districtssufferedremarkably high levels of traumaticevents.Howevertheseweresomewhatlower than the extraordinary highlevelsoftraumaticeventsexperiencedin PNA1. In PNA2, 73% of thesample report experiencing combat(compared with 78% of PNA1respondents), 33% forced to fleeburning buildings (38% PNA1),and 45% forced to flee danger(47% PNA1). Nine percent of thePNA2 sample were forced to hide(16% PNA1); 23% suffered beatingstothebody(39%PNA1);tenpercentwere tortured (18% PNA1); 37%hadafamilymemberorfriendkilled(41% PNA1); two percent had aspouse killed (4% PNA1); and threepercentachildkilled(5%PNA1).

Table 2.1 summarizes past traumaticeventsexperiencedbyrespondentsinthe total sample, comparingbetweenbothPNA1andPNA2studiesandthesixgeographicregionsdefinedintheResearch Design and Methodologysectionabove.

VAriAtion in trAumAtic eVents by reGionThe survey was designed to mapdifferences across regions andcommunities in their experienceswith violence. For nearly everytraumaticeventreportedinTable2.1thetotalPNA2samplereportsslightlyfewerexperiencesoftraumaticeventsthan the PNA1 sample despite thewidespread experience of combatacross all of Aceh. The footnote

beneathTable2.1(“*”)showsthatthedifferencesbetweenPNA1andPNA2are statistically significant for nearlyeveryitematp<.0001.Howeverwhenone examines reported traumaticeventsbygeographicregion,locationemerges as a critical element inunderstanding which respondentswere most affected by the conflict.The sections below report some ofthe outstanding regional variationsin the sample, puts these variationsinto context, and then highlightshow a macro-regional approach todocumenting traumatic experiencecanobscuretheintensityoftraumaticeventsinveryparticularlocalsettings.

• high prevalence regions It was anticipated that PNA1 had

surveyed communities with thehighest levels of violence over thelongest periods of time, given thecentrality of the North Coast inthehistoryofGAMandtheconflict,and that overall, other regionswould be found to have sufferedlower levels of violence duringthe conflict. Somewhat contraryto expectation, then, was theoverwhelming finding that tworegions – the East Coast andthe Southwest Coast – sufferedterrible violence and traumaticevents at a level equivalent toor even higher than that in theNorth Coast. In the East Coastcommunities, for example, 80%of the respondents reportedhaving lived through combatexperiences, 45% experiencedhaving to flee from burningbuildings in their communityand 61% having to flee from

danger. Seven percent ofwomen have had their husbandkilled in the conflict, 50% ofrespondents report having had afamily member or friend killed,and 45% reported having a familymember or friend kidnapped ordisappear. Almost half, or 47%reported having their propertyconfiscated or destroyed and31% experienced extortion orrobbery. Persons in the SouthwestCoastregion,wheretheviolencewasofmuchshorterduration,reportedexperiences of violence at nearlyequivalentrates.

The distinctive experiences byregion are highly significant atp<.0001formostevents,withlesserlevelsofsignificanceforothers<.01to <.05 (see footnote “†” beneathTable 2.1). Events that show nosignificant difference betweenregionsarethosewithlowreportingin both the PNA1 and PNA2, andacross all regions. Qualitativeresearch,asmallportionofwhichisdescribedinthefollowingsections,supports the quantitative data infindingahistoryofextremeviolencetowardcivilianpopulationsinmanyvillagesintheseregions.

• hot spots: local specificity Within regions and districts

One of the most puzzling resultsfromthePNA1datawasthestrikingdifference in levels of traumaticexperience (and psychologicalsymptoms) between Pidie districton theonehandandBireuenandAceh Utara districts on the other.

Respondents from Bireuen andAceh Utara reported a prevalencetwo times as high as Pidie fornearly all significant items on thetraumatic events checklist. Toeconomizeinthereportingofdatafrom all across Aceh province, all14 sampled districts have beengrouped into the six regions asdescribed in the Research DesignandMethodologysectionabove.

The PNA1 data from Pidie,Bireuen, and Aceh Utara in thisreport, for example, is aggregatedinto the North Coast region.The data here reported from theNorth Coast does not capture theremarkable variability betweenthe three districts. Herein liesan important shortcoming inpresenting the results of this PNAresearch at a macro-geographiclevel that describes data collectedfrom 105 villages across Aceh—eachwithitsownparticularhistoryof conflict—aggregated into sixbroadlydefinedregions.

Findings from the qualitativeresearch suggest a wide rangeof factors that may contribute tointense variations among villageswithin the regions defined in thisreport, within districts (kabupaten),and even within sub districts(kecamatan). In several coastaldistricts,forexample,thereisbroadconsensus that as one traveledaway from the main highway,either toward the mountains ortowardthesea,thesecuritysituationduring the conflict increasedin danger. In Aceh Selatan the

chances that someone evacuatedto safety during the conflictincreases if his or her village iscloser to the mountains. Residentsof certain sub-districts in AcehSelatan can describe patterns ofpopulationdisplacementduringtheconflict simply as, “everyone fromvillageXandall thevillagespast itup into the mountains evacuatedduring the conflict.” The reasonsfor evacuation are described asan often intolerable litany oftraumatic experiences, similar tothe itemized events in the PNAquestionnaire, which left no otherchoicebuttoleave.

But why do some sub-districts inAceh Selatan have patterns likethis while others that featuresimilar geography do not? Militaryintervention by Indonesianforces typically followed GAM’srecruitment activities along theSouthwest Coast, but recruitmentwas hardly uniform across sucha vast and relatively sparselypopulatedregion.Uponarrival,thevarious regiments of Indonesianmilitary and police units importedfrom outside of Aceh developedreputations for their unique stylesof intervention. Some batallionswere known for their paternalisticand sympathetic kindness towardcivilians,whileotherswereinfamousfortheirspectacularandinhumanebrutality. GAM operations couldbe equally inconsistent in theirtreatment toward civilians,especially among the districts oftheSouthwestCoast,theEastCoastand the Central Highlands. Entire

villages of Javanese transmigrantswereemptiedbyGAM forces, thushighlighting ethnicity as anotherfactor that determined specificpatterns of violence within theregions.

Levels of violence could alsovary by economic interests innatural resources by both sides.Competition over small scale goldmining operations in the upperreaches of the Kluet River valleyin Aceh Selatan may explain whyKluet villages suffered through amore highly concentrated blast ofviolence and terror from 2001through2003.Butliketheinternalvariations in the North Coastdistricts, stories with such localspecificity that vary within a singledistrict such as Aceh Selatan, cannot be reported as a SouthwestCoastphenomenon.

While the sampling employedfor the PNA had in mind a broadstroke mapping of conflict eventsandpsychological symptomsacrossAceh, an understanding of thelocal specificities that determineconflict traumatic experiencessuch as those described aboveis essential for the developmentof programmes that target areaswith greatest need. The PNA data,dependent on a sample drawnfrom105villagesacross14districts,can not map all of the micro-localhot spots in Aceh’s long historyof conflict, but can be used topinpoint someexemplary localitiesnot described by the regionalaggregatedata.

“A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 200724 “A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007 2�

tAble 2.1 pAst trAumA eVents by pnA study And reGion

traumatic events

*pnA1 data †pnA2 datapnA1 + pnA2

data

% north coast

(n=596)

%Aceh besar

(n=180)

%east coast(n=246)

%central

highlands(n=327)

%southeast highlands(n=162)

%southwest

coast(n=461)

% total pnA2

sample(n=1,376)

%total

sample(n=1,972)

experienced combat (bombing, fire fights)forced to flee burning buildingsforced to flee dangerforced to hide

78384716

6813333

8045617

6237408

696

122

79425717

7333459

74354611

beating to the bodyAttacked by knife or guntorturedserious physical injury from combatWitnessed physical punishmenthumiliated or shamed in public

392618135417

16634

354

421915165112

15856

223

332393

302115135718

231410103910

281712114412

rapeforced to rape a family memberother sexual assault

1<13

002

114

114

001

113

113

1<1

3

spouse killedspouse disappeared, kidnappedchild killedchild disappeared, kidnappedfamily member or friend killedfamily member or friend disappeared

5353

4133

4221

3122

4312

5045

1344

3938

1010

1211

2242

4028

2232

3731

3242

3831

Kidnappedcaptured, held by tni/polri or GAm

512

35

714

35

14

412

49

410

sent to prisonforced separation from familyforced isolation

398

181

573

344

221

51310

485

386

confiscation, destruction of propertyextortion, robberyforced laborforced to give food, shelter to tni or GAmforced to fight against tni or GAmpunished for not fighting against tni or GAmforced to search for corpsesnot allowed to provide muslim burial

453329272312126

2617191214552

4731211520156

11

3313221073

113

156353330

472730252618102

37212215161084

402424191811

94

forced to injure family memberforced to injure non-family memberforced to destroy someone’s propertyforced to betray/endanger family memberforced to betray/endanger non-familysomeone forced to betray/endanger youforced to humiliate another personforced to search for GAm member in forest

6737778

35

0013531

14

1336473

29

1152143

12

00100001

45466

117

37

2234364

22

3435465

26

lack of shelter because of conflict 24 21 29 34 7 31 27 26

lack of food, water because of conflict 82 56 75 85 65 79 75 77

sick, lack of access to health care 60 46 60 69 39 71 62 61

* PNA1VS.PNA2districtlocations:Statisticallysignificantlydifferentatp<0.0001[AllEvents]

Except…. Rape NS Forcedtorapefamilymember NS Othersexualabuse NS Senttoprison p<0.01 Kidnapped p<0.01 Forcedtodestroys.o.’sproperty p<0.05 Spousekilled p<0.05 Childkilled p<0.01 Spousekidnappedordisappeared NS Childkidnappedordisappeared p<0.01

† PNA2districtlocationscomparison:Statisticallysignificantlydifferentatp<0.0001[AllEvents]

Except…. Rape NS Forcedtorapefamilymember NS Othersexualabuse NS Senttoprison p<0.01 Kidnapped p<0.01 Forcedtosearchforcorpses p<0.01 Forcedtoinjurefamilymember p<0.01 Forcedtodestroys.o.’sproperty p<0.05 Spousekilled NS Childkilled p<0.05 Spousekidnappedordisappeared NS Childkidnappedordisappeared p<0.01

“A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 20072� “A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007 27

Travelling southboundoutofAcehBaratDaya(SouthwestAceh) district into the district of Aceh Selatan (SouthAceh), the main highway hugs the shoreline on theapproach to the sleepy district capital of Tapaktuan. Thefoothills of the mountainous interior fall flat immediatelyto the left of the highway, leaving a narrow margin forcoastal farmingandfishingcommunities. InAcehSelatan,all respondents in the PNA sample self-identified theirethnicity as Acehnese. But in fact, in this district far fromthe provincial capital of Banda Aceh and the denselypopulated north coast districts classically identified as thecentre of Acehnese culture, the number of residents herewho speak Acehnese as their first language are theminority. Coastal communities in Aceh Selatan speakBahasa Aneuk Jame, a language more closely related tothe languages spoken in Sumatra Barat (West Sumatra)province further south down the west coast of the island.Highlandgroups,suchastheKluetdescribedbelow,speaktheirownlanguagesaswell.

Tapaktuan is not much more than a narrow strip ofshophouses and a harbour between the foothills and theIndian Ocean. The margin between the foothills and theocean ends at Tapaktuan’s southern city limits, and thehighwaythensnakesupthroughamountainpasswithstarkcliffs down to the ocean below with fabulous views alongthe way. A rest stop at the top of a majestic outcroppingmarks the spot where Indonesia’s first Vice-President,MohammadHatta, stopped in1953,duringhisdiplomatictour of the province to end the Darul Islam Rebellion- a different conflict against the newly independentIndonesian republic that preceded the government’sconflictwithGAMbytwodecades.

Upon descent to the south side of the mountain pass thehighway turns inland from the coast a bit as the flat landbetween the hills and the ocean widens out by severalkilometres width. After passing a prison and a stateagricultural training institute, both of which were burneddownduringtheconflict,afewdozenkilometressouththehighwaymeetstheintersectionatKotaFajar,amarkettownon thenorth sideof theKluetRiver thatempties into theIndian Ocean. A left turn at Kota Fajar follows a road in

disrepair that parallels the Kluet River into themountainous interior. Residents of the Kluet River valleyspeak Bahasa Kluet, similar to Bahasa Alas spoken in themountainsofAcehTenggara(SoutheastAceh)anddistantlyrelated to Bahasa Karo spoken by the Karo Batak in themountainous interior of Sumatra Utara (North Sumatra)province.

The formal economy of the Kluet River valley thrived inthe 1990s on patchouli (nilam), a cash crop whose leavesare boiled down to extract patchouli oil, a raw materialusedasabaseforawiderangeofproductsintheperfumeindustry. The informal economy of the Kluet River valleythrives on illegal logging and the aforementioned goldmining interests. Logging and mining provided strategicsources of income for whoever could control it, thuscreating a lucrative and devastating contest betweenGAM and Indonesian forces, particularly from 2001 to2003, when the Indonesian military and police forcesarrived in the region in order to neutralize the suddenandmassiveGAMrecruitment.

In the random sample of villages selected for the team ofresearchers that surveyed Aceh Selatan and Aceh BaratDaya, 50% were randomly selected from a short list ofknown high conflict villages in high conflict sub-districts that received IOM’s Post-Conflict CommunityReintegration Programme (see “Selection of Villages” inthe Research Design and Methodology section above).The remaining 50% of villages sampled in these districtswere randomly selected from villages in the same sub-districts that did not receive the benefit of IOM’s Post-Conflict Community Reintegration Programme. Five outof the 13 villages visited by the research team were in theKluet River valley where the residents primarily speakBahasaKluetastheirfirstlanguage.

The high number of Kluet villages in the PNA sample islikelydue to the fact thatmany villages in theKluetRivervalley met criteria for inclusion in IOM’s Post-ConflictCommunityReintegrationProgramme,i.e.anintensivehistoryofconflict,withlargenumbersofex-combatantsandformerpoliticalprisonersreturninghometothesevillagesafterthe

hot spot cAse study: the Kluet riVer VAlley

* AllelevenrespondentsinAcehSelatanwhoreported“othersexualassault”liveintheKluetRiverValley.**AlltenrespondentsinAcehSelatanwhoreportedbeingforcedtoinjureamemberoftheirfamilyliveintheKluetRiver

Valley.

tAble 2.2 pAst trAumA eVents in the Kluet riVer VAlley

traumatic events%

southwest coast districts(n=461)

%Aceh selatan district

(n=180)

%Kluet river Valley

(n=90)

%total sample

(n=1972)

mGKd+ mGKd- mGKd+ mGKd-

experienced combat (bombing, fire fights)forced to flee burning buildingsforced to flee dangerforced to hide

79425717

91537927

92629330

74354611

beating to the bodyAttacked by knife or guntorturedserious physical injury from combatWitnessed physical punishmenthumiliated or shamed in public

302115135718

433120237633

463825328540

281712114412

rapeforced to rape a family memberother sexual assault

113

216

32

12*

1<13

spouse killedspouse disappeared, kidnappedchild killedchild disappeared, kidnappedfamily member or friend killedfamily member or friend disappeared

2242

4028

32323934

42504038

32423831

Kidnappedcaptured, held by tni/polri or GAmsent to prisonforced separation from familyforced isolation

4125

1310

61692416

71782922

410386

confiscation, destruction of propertyextortion, robbery

4727

6235

7745

4024

forced labourforced to give food, shelter to tni or GAmforced to fight against tni or GAmpunished for not fighting against tni or GAmforced to search for corpsesnot allowed to provide muslim burial

30252618102

49304126103

61444629177

2419181194

forced to injure family memberforced to injure non-family memberforced to destroy someone’s propertyforced to betray/endanger family memberforced to betray/endanger non-familysomeone forced to betray/endanger youforced to humiliate another personforced to search for GAm member in forest

45466

117

37

68555111165

12**13798131972

343546526

lack of shelter because of conflictlack of food, water because of conflictsick, lack of access to health care

317971

448684

579291

267761

“A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 20072� “A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007 2�

peaceagreement.Table2.2comparesthelevelsoftraumaticevents reportedby respondentsduring theconflict infirsttheSouthwestCoastregion,thenAcehSelatanatthedistrictlevel,andthenfocusedonthefivevillagesintheKluetRivervalleyofAcehSelatan.

In a region already remarkable for the chaotic andcruel violence inflicted upon civilians up and down theSouthwest coast of Aceh during the latter years of theconflict,wheremorethanhalfofrespondents(57%)fromfour districts (Aceh Barat, Nagan Raya, Aceh Barat Dayaand Aceh Selatan) report being forced to flee danger,a focus upon the events reported from villages in theKluet River valley stands out for nearly every item onthe list. Comparing the numbers down from region,to district, to the Kluet River valley, it comes as littlesurprise that 93% of the Kluet respondents wereforcedtofleeundersuchintolerableconditions.

The astonishingly high levels of violence (beatings,torture, attacked with a weapon, destruction of property,combat and combat injuries) give pause, the percentagesforexperiencesoftraumaticeventsintheKluetRivervalleyare typically twice as high or greater than the averagepercentage reported by the total sample (PNA1 &PNA2).

In one of the villages, during three days of nearlycontinuous torture and interrogation of every adultman in the village, an elderly farmer was hacked with amachete down the side of his torso because he could notkeep up with the forced labour and “exercises” imposedby security forces. Today the wound is thickly scarred,often turns red accompanied with throbbing pain andpus, and has permanently restricted movement of hisright shoulder so that he can not raise his arm. Unableto even work in his own garden to harvest areca nuts, heis so demoralized that his only activity is “waiting to die.”

During those three days of torture, when security forcesfirstoccupied the village, thismanalongwith three other

men whose sons were suspected combatants, were forcedto spend each night submerged in sewage and irrigationcanals. One father of two teenage combatants was beatensobadlyon theheadand thenheld forcefullyunderwaterin the sewage canal until he lost consciousness; hesuffered severe anoxic brain injury as a result and sincethenhisbodyhasgraduallylostnearlyallmotorfunction.

Although he is alert, answering questions lucidly andretaining a sense of touch, his emaciated body now liesalmost completely spastic. He is unable to sit up, hismuscles are contracted, his joints in permanent flexion,he’s incontinent, and his ability to speak is increasinglycompromised. His family keeps him clean and cares forhimattentively,buthewillneverrecoverfromthisterribleinjury. The family needs additional education aboutproviding physical therapy, hygienic care for his body,and giving him medication to ease the discomfort andspasticitythathewillsufferfortherestofhislife.

What makes the Kluet River valley’s conflict experiencestand out even more than the violence are the levels ofinsult and humiliation suffered by this population. Thenumbers for traumatic events that deliberately inflictpsychological injury and shame upon respondents aremostly between three and four times higher than theaverage numbers reported by the total sample. 85% ofKluet respondents witnessed physical punishment,40% were humiliated in public, 19% were forced tohumiliate others, 22% were forced into isolation, and29% were forced apart from their families. 12% of Kluetrespondents were the only respondents in all of AcehSelatanwhowere forced to injureamemberof theirownfamily. This is four times higher than the total sampleaverage.

The Gendered Trauma section below describes thedifficulty of quantifying sexual violence, given such lowreporting, but it is worth noting that the percentage ofKluet respondents admitting rape is three times higherthan the total sample average. The percentage of Kluet

respondentsadmittingtheywereforcedtorapeamemberof their own family is twice as high as the total sampleaverage, and the percentage of Kluet respondentsreporting“othersexualassault”(12%)isfourtimeshigherthanthetotalsampleaverage.

Upon hearing the stories told by key informants andquestionnaire respondents in the Kluet River valley,senior researchers from Harvard, IOM staff, and theAcehnese interviewing staff were all rendered speechlessandexhaustedbywhat they learned.Theresidentsofonevillage in the sample wished they had evacuated like theneighbouring communities had done before the violenceand torture reached its peak. A young man describedbeing forced to stand inacirclewithhis friends, andoneby one the men were ordered to punch the man to hisrightwithgenuinestrengthandforce,aroundandaroundthe circle. Half-hearted punches were met with muchworse blows by the soldiers who ordered the men to hurteach other. When asked how he lives with the memory ofinjuring the friends and neighbours he has known all hislife, he answered, “the situation was so absurd, we couldonlylaughaboutit.

Yes, we really did punch each other as hard as we could,we had no choice! But we laughed our way throughit, that is all we could do. How can we hold a grudgeagainst each other under such ridiculous conditions?We don’t feel hurt or vengeful about what we did toeach other; we can only laugh if we remember it. Thecapacity to find humour in the absurd conditionsof senseless violence and terror may have been animportant mode of psychosocial resilience for the menand women who lived through such intensive andconcentrated periods of conflict, as well as an indicationof their psychological strength in being able to position

themselvesoutsideofthetraumaticsituation.

Other examples of humiliating trauma perpetrated in theKluetRivervalleyneednotbereproducedinthisreportinorder to understand the point of this particular examplefrom Aceh Selatan. While the PNA data finds conflictevents widely distributed across all of Aceh, leaving nearlyno one unaffected, it is possible to identify, with minimaladvance quantitative and ethnographic assessment, theso-called hot spots or micro-localities where intense,complex, and sustained bursts of conflict violence maysuggest the priorities for the tailored design and deliveryof psychosocial and other post-conflict reintegrationservices.

Today nearly all of the residents of the Kluet River valleyhave returned home from their IDP camps in Tapaktuan,where entire villages lived for more than a year until thepeace agreement was signed. Fields and forest gardenswere destroyed or have gone fallow, along with manyhomes, and the market price of patchouli oil hasplummeted to less than a third of its value in the late1990s, before the conflict began. Returned IDPs in theKluet villages are living in makeshift homes and growingquick but low profit cash crops such as chili peppers andcassava. There are also unverified reports of continuedextortion and intimidation of civilians by armed groupsinthearea.Thereisadireneedfortransitionallivelihoodsupport in order for Kluet villagers to rehabilitate theirfields.Communityorganizingand training to improve thequality, value, and marketing of patchouli as their localcash crop is also recommended. Psychosocial screeningand support for conflict victims suffering from depressionor trauma should be coupled with post-conflict economicsupportforthesecommunitiesinorderforallassistancetobesuccessfulandsustainable.

“A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007�0 “A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007 �1

• low prevalence regions The PNA2 survey did indeed

find significantly lower levels oftraumatic events in other regions,particularly in the SoutheastHighlands and in Aceh Besar.But the preceding discussionabout the remarkably localspecificities of conflict experience,such as in the Kluet Valley, shouldserve as a cautionary note beforediscounting entire regions. Thereal finding of this study is justhow many persons even in theseso-called low prevalencecommunities were deeply affectedbytheconflictandviolenceaswell.

In the Southeast Highlands 69 %and in Aceh Besar 68% ofrespondentsreportedexperiencingcombat. Even in the SoutheastHighlands, usually cited as theleast conflict-affected region ofthe province, 12% of respondentsthere reported having a familymember or friend killed. In AcehBesar,32%ofmen(seeTable2.3)reportedbeingbeatenonthebody.Thus, while violence was certainlylocalized in particular regions andvillages, it was also widespread,affecting a remarkably highnumber of persons throughouthundreds,ifnotthousands,ofruralcommunitiesthroughoutAceh.

Gendered trAumATable 2.3 below reports pasttraumatic events experiencedby men in the different geographicregions. Table 2.4 reports the sameinformationforwomen.Bothmenandwomen experienced extraordinarylevels of violence, but the level andtype of traumatic events experiencedas part of the conflict varied bygender. Men reported significantlygreater physical violence thanwomen. Across the full sample,44% of men report having beenbeaten(12%ofwomen),24%reportbeing attacked with a gun or knife

(10%ofwomen),19%ofmenreportbeing tortured (5% of women),16% of men reported being beentaken captive (4% of women), and49% of men (and 38% of women)reported witnessing physical violenceagainst others. Nonetheless, womensuffered enormously. For example,in the Southwest Coast communities,79% of women experienced combatand 56% had to flee danger, 52%were forced to witness physicalpunishment, 36% had a familymember or friend killed, 43% hadproperty confiscated or destroyedand 32% were forced to search forGAMmembersintheforest.

• sexual Violence Asnoted in thePNA1report,rape

and other sexual violence is rarelyreportedinthePNAquestionnairedata, although human rightsadvocates have identified sexualviolence as an important featureof conflict violence in Aceh.Interviewers did not challengerespondents’ choice of notreporting these stigmatizingevents. It is worth noting that onthe traumatic events checklist,respondents could answer non-specific items about beinghumiliated or shamed in public,or being forced to humiliate orshame others. For example, twelvepercentofEastCoast respondents,17% of North Coast respondentsand 18% of Southwest Coastrespondents report havingexperienced public humiliationduring the conflict. Genderdifferences are significant indistinguishing who reported mosthumiliation.

InbothPNA1andPNA2,menweremore likely than women to reportbeing humiliated in public orforced to humiliate another. Fourpercentofwomeninbothsamplesreported other sexual assault andone percent reported rape. Insimplenumbersthismeansthat50

out of the 965 women in the totalsamplereportedsexualviolenceofsomekindaspartoftheirtraumaticexperienceduringtheconflict.

Key informant interviews andinformalgroupdiscussionsintheseregions, during both PNA1 andPNA2 fieldwork, recount acts ofsexualabuseandbodilyhumiliationin ways that suggest that therewas more of this kind of violencecommitted than the reportednumbers tell us. Men and womenwere often forced by combatantgroups to lineupand stripnaked.Children were forced to touchthe genitals of their parents ofthe opposite sex. Many women insomepartsofAcehSelatandistrictreportedhavingtheirheadsshavedbaldevery twoweeksbecause theirhusbands or sons were suspectedcombatants.

Those who describe finding thecorpses of conflict victims, usuallydumpedinpublicplacessuchasroadsides and river banks, often depictthe genital mutilation inflicteduponthebodiesasbutoneofmanysignsof torture.Manyrespondentsspoke about how young womenfrom their communities were senttolivewithrelativesinsafer,urbanareas rather than riskhaving themsubjectedtosexualabusebycombatgroups stationed in their village.Thesesamerespondentsrecallwell-known “posts” or empty buildingsinorneartheirvillagewhereyoungwomen suffered various forms ofsexualviolence.

Finally, women’s narratives revealthat the common experience ofhaving one’s house ransackedor destroyed was experienced asan especially powerful attack onwomen, bearing in mind that inrural Aceh women typically owntheir own homes and a husbandmoves into his wife’s home uponmarriage. In fact, a man’s wife in

Acehisoftenreferredtoaspo rumoh,theownerofthehouse.

• head trauma A unique form of physical trauma

widely experienced during theconflict, particularly among men,thatcanhaveterribleconsequencesfor an individual’s organic mentalhealth is head injury. In the worstcases,severeheadinjuriescanresultin total loss of motor control suchas the example described in theKluetRivervalleycasestudyabove.In other cases, people lose theirability to remember, experiencenumbness and other sensorydisabilities such as vision loss, orhaverapidandunexplainablemoodswings.

Ratesofheadtraumaandpotentialbrain injury, suffered throughbeatings, strangulation, neardrownings,andotherformsoftortureor violence, were extraordinarilyhighinthePNAsampleanddeserveclinicalinterventionsandadditionalresearch. Men, particularly youngmen in the highest conflict areas,wereatthehighestrisk.43%ofallmen in the East Coast region and41%intheSouthwestCoastreporthavingsufferedheadtrauma—ratesequivalenttothosefoundinPNA1fortheNorthCoast.Tables2.5and2.6 present the data on physicalhead injuries sustained due toconflictviolencebymenandwomenrespectively.

forced eVAcuAtion And other displAcement experienceTable 2.7 reproduces two items fromthe Traumatic Events checklist thatconcern evacuation due to conflictviolence.AfterseeingthesefiguresinPNA1,andgivenIOM’scoremandateto deal with migration, additionalquestions about displacementexperience were included in thePNA2 questionnaire, including the

information at the bottom of Table2.7 about forced and voluntaryevacuation.

In nearly all of the high conflictareas sampled, between one-thirdand two-thirds of all respondentswere displaced, usually by force,during the conflict. Respondentswho answered yes to the questionabout voluntary or forced evacuationwere then given an opportunity totell interviewers about theirdisplacement experience, includingtheir main reason for leaving theirhome of origin during the conflict.The following unranked listsummarizes the main themes thatemerge from the answers to thisquestion:

· Combat operations in thecommunity(kontak senjata)

· Toomuchextortionandrobbery· Ordered to leave by either GAM,

TNI,or“unidentifiablegroups”· Respondentshomesweredestroyed· Threatenedby,orfearfulof,either

TNIorGAM· Too much violence and torture

againstcivilians· Notpermittedtobuyessentialgoods

by armed groups occupying thevillage

· Respondents, or respondents’relative(s),involvedinGAM

· Accused of providing logisticalsupporttoarmedgroups

· Respondents heard that TNI orGAMweresearchingforthem

· Everyone else in the village wasevacuating

Almosteveryrespondentwhotoldusabout their displacement experienceduring the conflict should beconsidered as returned IDPs, or forthose living very close to their homeof origin prior to displacement, ascurrentIDPs.TherearestillthousandsofJavanesetransmigrantIDPfamiliesstill living in Sumatra Utara (NorthSumatra) province just south ofAceh,andseveralthousandAcehnese

refugees still in Malaysia. Indeed,most Javanese transmigrants havechosen to sell their land rather thanto risk returning and facing moreviolence. The PNA data thereforedoes not capture the experience ofthe current IDPs still living outsideAceh. Bearing this in mind, thefollowing Table 2.8 provides dataon where the returned IDPs in thePNA2 sample went during theirdisplacement.

Following the numbers diagonallyfrom top left to bottom right, itis overwhelmingly clear that thereturned IDPs in the PNA2sample did not go very far duringtheir displacement. Most of thequalitativedatadescribesdisplacementexperiences as close as theneighbouring village, the sub-districtoffice/ school/ mosque, the districtcapital, the neighbouring district, orthe nearest municipality. Countingthe diagonal numbers, 545 out ofthe 595 (95%) respondents whotold us where they went duringtheir evacuation stayed in the samegeographic region. Not shown inTable2.8,acloseranalysisshowsthat505oftheserespondents(85%)stayedwithin their home district. Localiseddisplacement is the overwhelmingfinding among returned IDPs in thePNA2sample.

Duration of displacement is varied.Respondents report being away fromtheir villages for as brief as a week,andforaslongastwoyears.(CurrentIDPsnotinthesamplehavebeenawayforfiveyearsormuchlonger)Butthecommonrefrain,nomatterhowshortorlongthedurationofdisplacement,is that returning IDPs camehome tonothing. Itonly tookadayor two topillage and burn entire villages—homes, fields, livestock, gardens,schools,mosques.“Startingfromzero”isthecommonexperienceofreturningIDPs, and vulnerable economicconditions have significant bearingontheirpsychosocialcondition.

“A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007�2 “A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007 ��

tAble 2.3 pAst trAumA eVents experienced by men, by reGion

traumatic events

*pnA1district location

%total sample

(n=1,006)

†pnA2

%north coast

(n=315)

%Aceh besar

(n=66)

%east coast(n=125)

%central highlands

(n=180)

%southeast highlands

(n=94)

%southwest coast

(n=226)

experienced combat (bombing, fire fights)forced to flee burning buildingsforced to flee dangerforced to hide

83435220

6212336

83446110

6133368

789183

80465821

77374714

beating to the bodyAttacked by knife or guntorturedserious physical injury from combatWitnessed physical punishmenthumiliated or shamed in public

563625196122

329911358

662824256215

25988254

5534154

483025236326

442419174916

rapeforced to rape a family memberother sexual assault

1<13

005

222

211

001

0<13

112

spouse killedspouse disappeared, kidnappedchild killedchild disappeared, kidnappedfamily member or friend killedfamily member or friend disappeared

22524936

22222626

01115951

12543739

10101815

2141

4535

12424336

Kidnappedcaptured, held by tni/ polri or GAm

819

611

1123

37

15

721

616

sent to prisonforced separation from familyforced isolation

41110

292

783

355

231

101814

5108

confiscation, destruction of propertyextortion, robbery

4936

2117

5236

3514

226

5131

4327

forced labourforced to give food, shelter to tni or GAmforced to fight against tni or GAmpunished for not fighting against tni or GAmforced to search for corpsesnot allowed to provide muslim burial

44292817157

311417982

30162519716

30883134

555450

45313525142

36212315125

forced to injure family memberforced to injure non-family memberforced to destroy someone’s propertyforced to betray/ endanger family memberforced to betray/ endanger non-familysomeone forced to betray/ endanger youforced to humiliate another personforced to search for GAm member in forest

101161010111146

000888019

2678511440

214316315

00100001

57798

161044

6657610733

lack of shelter because of conflictlack of food, water because of conflictsick, lack of access to health care

228664

115538

277765

338469

137042

328073

258063

mAles *†

past trauma events experienced by male informant

pnA1 Vs. pnA2 district locations

comparison

pnA2 district locations

comparison

experienced combat (bombing, fire fights)forced to flee burning buildingsforced to flee dangerforced to hide

p<0.0001p<0.0001p<0.0001p<0.0001

p<0.0001p<0.0001p<0.0001p<0.0001

beating to the bodyAttacked by knife or guntorturedserious physical injury from combatWitnessed physical punishmenthumiliated or shamed in public

p<0.0001p<0.0001p<0.0001p<0.0001p<0.0001p<0.0001

p<0.0001p<0.0001p<0.0001p<0.0001p<0.0001p<0.0001

rapeforced to rape a family memberother sexual assault

NSNSNS

NSNSNS

spouse killedspouse disappeared, kidnappedchild killedchild disappeared, kidnappedfamily member or friend killedfamily member or friend disappeared

NSNSNSNS

p<0.0001p<0.0001

NSNSNSNS

p<0.0001p<0.0001

Kidnappedcaptured, held by tni/polri or GAmsent to prisonforced separation from familyforced isolation

p<0.01p<0.0001p<0.01

p<0.0001p<0.0001

p<0.01p<0.0001p<0.01

p<0.0001p<0.0001

confiscation, destruction of propertyextortion, robbery

p<0.0001p<0.0001

p<0.0001p<0.0001

forced laborforced to give food, shelter to tni or GAmforced to fight against tni or GAmpunished for not fighting against tni or GAmforced to search for corpsesnot allowed to provide muslim burial

p<0.0001p<0.0001p<0.0001p<0.0001p<0.05

p<0.0001

p<0.0001p<0.0001p<0.0001p<0.0001

NSp<0.0001

forced to injure family memberforced to injure non-family memberforced to destroy someone’s propertyforced to betray/endanger family memberforced to betray/endanger non-familysomeone forced to betray/endanger youforced to humiliate another personforced to search for GAm member in forest

p<0.0001p<0.0001p<0.05p<0.001

p<0.0001p<0.0001p<0.0001p<0.0001

p<0.05p<0.0001p<0.05p<0.001p<0.0001p<0.0001p<0.0001p<0.0001

lack of shelter because of conflictlack of food, water because of conflictsick, lack of access to health care

p<0.0001p<0.0001p<0.0001

p<0.0001p<0.0001p<0.0001

“A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007�4 “A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007 ��

femAles *†

past trauma events experienced by male informant

pnA1 Vs. pnA2 district locations

comparison

pnA2 district locations

comparison

experienced combat (bombing, fire fights)forced to flee burning buildingsforced to flee dangerforced to hide

p<0.001p<0.0001p<0.0001p<0.0001

p<0.001p<0.0001p<0.0001p<0.0001

beating to the bodyAttacked by knife or guntorturedserious physical injury from combatWitnessed physical punishmenthumiliated or shamed in public

p<0.0001p<0.0001p<0.0001p<0.01

p<0.0001p<0.0001

p<0.0001p<0.001P<0.001p<0.01

p<0.0001p<0.0001

rapeforced to rape a family memberother sexual assault

NSNS

p<0.01

NSNS

p<0.01

spouse killedspouse disappeared, kidnappedchild killedchild disappeared, kidnappedfamily member or friend killedfamily member or friend disappeared

p<0.01NSNS

p<0.05p<0.0001p<0.0001

p<0.05NSNS

p<0.05p<0.0001p<0.0001

Kidnappedcaptured, held by tni/polri or GAmsent to prisonforced separation from familyforced isolation

NSNSNS

p<0.01p<0.01

p<0.05NSNS

p<0.01p<.01

confiscation, destruction of propertyextortion, robbery

p<0.0001p<0.0001

p<0.0001p<0.0001

forced labourforced to give food, shelter to tni or GAmforced to fight against tni or GAmpunished for not fighting against tni or GAmforced to search for corpsesnot allowed to provide muslim burial

p<0.01p<0.0001p<0.0001p<0.0001p<0.01p<0.05

p<0.001p<0.01

p<0.0001p<0.0001p=0.01p<0.05

forced to injure family memberforced to injure non-family memberforced to destroy someone’s propertyforced to betray/endanger family memberforced to betray/endanger non-familysomeone forced to betray/endanger youforced to humiliate another personforced to search for GAm member in forest

NSNS

p<0.01NSNS

p<0.01NS

p<0.0001

NSNS

p<0.05NSNS

p<0.01NS

p<0.0001

lack of shelter because of conflictlack of food, water because of conflictsick, lack of access to health care

p<0.0001p<0.0001p<0.0001

p<0.0001p<0.0001p<0.0001

tAble 2.4 pAst trAumA eVents experienced by Women, by reGion

traumatic events

*pnA1district location

%total sample

(n=966)

†pnA2

%north coast

(n=281)

%Aceh besar

(n=114)

%east coast(n=121)

%central highlands

(n=147)

%southeast highlands

(n=68)

%southwest coast

(n=235)

experienced combat (bombing, fire fights)forced to flee burning buildingsforced to flee dangerforced to hide

73334212

7114331

7746615

6342458

56340

79395613

7233449

beating to the bodyAttacked by knife or guntorturedserious physical injury from combatWitnessed physical punishmenthumiliated or shamed in public

20141164511

7400342

16957419

3714193

000020

121364

5211

121054388

rapeforced to rape a family memberother sexual assault

204

001

106

008

000

213

1<14

spouse killedspouse disappeared, kidnappedchild killedchild disappeared, kidnappedfamily member or friend killedfamily member or friend disappeared

83543130

52203320

75234138

23344037

000034

3452

3622

53433327

Kidnappedcaptured, held by tni/polri or GAm

25

12

45

33

02

<13

24

sent to prisonforced separation from familyforced isolation

274

070

362

222

200

196

263

confiscation, destruction of propertyextortion, robbery

4028

2818

4226

3111

64

4324

3621

forced labourforced to give food, shelter to tni or GAmforced to fight against tni or GAmpunished for not fighting against tni or GAmforced to search for corpsesnot allowed to provide muslim burial

112516584

121112333

1213141247

13116291

040000

1520171162

121713663

forced to injure family memberforced to injure non-family memberforced to destroy someone’s propertyforced to betray/endanger family memberforced to betray/endanger non-familysomeone forced to betray/endanger youforced to humiliate another personforced to search for GAm member in forest

22<1343524

002130111

000334317

01511128

00000000

2224364

32

112233319

lack of shelter because of conflictlack of food, water because of conflictsick, lack of access to health care

257755

265650

317355

368670

05935

307769

277459

“A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007�� “A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007 �7

post-conflict stressWhereaspasttraumaticeventsduetoconflictvarieddirectlywitharegion’shistory of violence in the PNA2regions, “current stress” - defined asthose conditions that prevail sincethe signing of the peace agreement- follow a very different pattern.Overall, current stressful events werereportedatamuchlowerrateinPNA2thaninPNA1.Althoughthereissomevariationamongtheregionssurveyedin PNA2, the pattern of reportingmuch lower rates of current eventsis true even in the regions wherethe conflict was greatest. This isparticularly true of reports of seeingperpetrators (47% in PNA1, 7% inPNA2),ofexperiencingassault (31%vs. 1%), experiencing robbery (21%vs. 1%), and even violence towardwomen (4% vs. 1%) and violencetoward children (7% vs. 1%). Postconflict stress events among thegeographic regions and between thePNA1andPNA2dataaresummarizedinTable2.9.

These findings suggest a verysignificant change between February2006,whenthePNA1wasconducted,andJuly2006,whenthePNA2surveywas conducted (November 2006 forAceh Besar). Although the peaceagreement, or Memorandum ofUnderstanding (MoU) was signed inAugust2005,onepossibleexplanationfor thehighratesofcurrentstress inFebruary2006 is that the Indonesianmilitary and police troops were stillbased in their village posts all overAcehuntilaslateasDecember2005.

InFebruary2006,securitylevelswerestill uncertain and trust in the peaceprocess still very limited, becausealthough the MoU called for thereturn of 23,000 inorganic troops,i.e. Indonesian forces imported toAceh from other parts of Indonesia,theirdeparturefromwastimedalongwith GAM’s lengthy phased processof surrenduring its weapons. Thusmany villages were still occupied by

* Fromthefourdifferenttypesofheadinjury,ifarespondentanswersyestooneor more of those four questions, then the answer is yes for the new variable(“Any type of head trauma”), which will then tell us how many respondentsexperiencedphysicalheadtraumaofanykindatall.

‡ pnA2 district locations comparison: Anytypeofheadtrauma- p<0.0001 Beatenonthehead- p<0.0001

Suffocationorstrangulation- p<0.0001 Neardrowning- p<0.01 Otherheadtrauma- p<0.01

males† pnA1 Vs. pnA2 district locations: statistically significantly

different at p< 0.0001 [All measurements] Anytypeofheadtrauma- p<0.0001 Beatenonthehead- p<0.0001 Suffocationorstrangulation- p<0.0001 Neardrowning- p<0.01 Otherheadtrauma- p<0.0001

tAble 2.5 heAd trAumA/ potentiAl brAin inJury: men

head trauma

†pnA1 datadistrict location

pnA1 + pnA2†‡pnA2 data

%north coast

(n=315)

%Aceh besar

(n=66)

%east coast(n=125)

%central highlands

(n=180)

%southeast highlands

(n=94)

%southwest coast

(n=226)

%total sample

(n=1006)

*Any type of head trauma 41 17 43 16 5 41 32

specific type

beaten on the head suffocation or strangulation near drowning other head trauma

361979

14552

361443

9752

3130

3218137

261375

* Fromthefourdifferent typesofheadinjury, ifarespondentanswersyes tooneor more of those four questions, then the answer is yes for the new variable(“Any type of head trauma”), which will then tell us how many respondentsexperiencedphysicalheadtraumaofanykindatall.

‡ pnA2 district locations comparison: Anytypeofheadtrauma- NS Beatenonthehead- p<0.01 Suffocationorstrangulation- p<0.01 Neardrowning- NS Otherheadtrauma- NS

females† pnA1 Vs. pnA2 district locations: statistically significantly

different at p< 0.0001 [All measurements] Anytypeofheadtrauma- p<0.05 Beatenonthehead- p=0.0001 Suffocationorstrangulation- p<0.001 Neardrowning- NS Otherheadtrauma- p<0.05

tAble 2.6 heAd trAumA/ potentiAl brAin inJury: Women

head trauma

†pnA1 datadistrict location

pnA1 + pnA2†‡pnA2 data

%north coast

(n=281)

%Aceh besar

(n=114)

%east coast(n=121)

%central highlands

(n=147)

%southeast highlands

(n=68)

%southwest coast

(n=235)

%total sample

(n=966)

*Any type of head trauma 12 4 10 5 6 7 8

specific type

beaten on the head suffocation or strangulation near drowning other head trauma

77<12

1120

4801

0410

0600

4130

4511

“A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007�� “A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007 ��

combat groups until several monthsaftertheMoUwassigned.Therecentmemory of village occupation inFebruary2006maybewhatconstitutessuch high levels of current stress forrespondents in PNA1. Whereas byJulyandNovemberthepeaceprocesshadadvancedsignificantly,withruralcommunitiesexperiencingatleastsixmonths of legitimate peace, withoutthe intimidating and sometimesharrassing presence of combatgroups in their midst. The PNA2findingssuggestthatbyJuly,residentsexperienced far less contact withperpetrators of past violence and fargreatersecurity.

The pattern of lower PNA2 scoresholds true also for basic livingconditions(lackofadequatehousing59% vs. 38%; water/ sanitation 75%vs. 55%; and food security 72% vs.63%)and livelihood issues (difficultyproviding for family 85% vs. 72%;findingwork89%vs.75%;andstartinga livelihood 71% vs. 56%). However,even though they are significantlyreduced, the figures from PNA2are still extremely high. Qualitativeinterviews suggest that these currentstressorsreflectthedevastationofthe

villageeconomiesafteryearsofeffortsto destroy the material foundationssupportingGAM.Theconflictclearlywrecked havoc on local economies,destroying trade networks, wreckingtheir houses, killing their animalsandpreventingfarmersfromworkingtheir land and young people fromentering into the labour economy.Thus, recovery will require both thatthe terrible traumaticevents sufferedbythesecommunitiesandthebrokeneconomy and destroyed communityresources be dealt with in a timelyfashion.

Manymonthsaftercompletionofthe

PNA fieldwork and nearly half wayinto 2007 at the time of writing thisreport,itisworthnotingthatinsomeareas of Aceh the security situationremains unstable, particularly inthe districts of Bireuen, Aceh Utara,and Aceh Timur where emergingreports of extortion, robbery, bombthreats, knife attacks, and turf warsare unsettling local communities.The PNA data does not measureevents or symptoms prospectively/longitudinally,soitisunclearwhetherlevels of current stress, along withpsychological symptoms, would befluctuatingwiththechangingsecurityenvironment.

VillAGes desiGnAted As hiGh conflict by iom’s mGKd proGrAmmeBoth the PNA1 and PNA2 based thesamplingofhighconflictsub-districtsand villages in part on previousresearch by the World Bank andIOM’s EC-funded Post-ConflictCommunityReintegrationProgramme,known locally as Makmue Gampong Kareuna Dame (MGKD) programme.The MGKD programme implementscommunity-driven quick impactpeace dividend projects in villageswith intensivehistoryofconflict,anda large number of returning formerGAM combatants and amnestiedprisoners.

TheMGKDprogramme’sselectionofhighconflictsub-districtsandvillageswas an important component of thePNA sampling methodology if thedistrict covered by the PNA featuredMGKD programming. On average,wherever the MGKD programme waspresent, half of the villages sampledin the PNA were randomly selectedfrom the list of MGKD recipientvillages. The remaining villages wererandomly selected from the samesub-districts identified by the MGKDprogramme as high conflict, butthat did not end up as recipients ofthe peace dividend project. 46% ofPNA1respondentsand33%ofPNA2respondents resided in villages withMGKD programmes. The overall

percentage is lower for the PNA2sample because four districts - GayoLues,BenerMeriah,AcehTengah,andAcehBesar-didnothavetheMGKDprogramme and thus employed adifferentsamplingmethodology.

Thedifferenceinsamplingprocedureleads to the chance of a samplingbias toward less high conflict villagesin these four districts. Nevertheless,analysis of traumatic experiencesby respondents in MGKD andnon-MGKD villages provides anopportunity to evaluate theeffectiveness of the World Bankand IOM designation of thesecommunities as high conflict. Onemight hypothesize that if the highconflict designation turns out to behighly robust in terms of traumaticevents experienced by respondentsin MGKD villages, then the PNAsampling bias in the districts withoutan MGKD programme increases aswell.

• past traumatic events in mGKd Villages

Levels of past traumatic eventswere substantially and significantlyhigher in the MGKD villages inboth PNA1 and PNA2. Rates ofsomespecificpast traumaticeventswere50%ormorehigherinMGKDcommunities in both samples (forevents commonly experienced,these achieve high levels of

* InclusiveofrespondentswhomentionedLhokseumawemunicipalityastheirdestination.** InclusiveofrespondentswhomentionedBandaAcehmunicipalityastheirdestination.*** InclusiveofrespondentswhomentionedLangsamunicipalityastheirdestination.**** InclusiveofrespondentswhomentionedAcehJaya,AcehSingkil,andSimeuleudistrictsasdestinations***** Mostly Sumatra Utara (North Sumatra) province, but Riau province, Sumatra Selatan (South Sumatra) province, Jakarta, and

Malaysiawereeachmentionedonceonly.

tAble 2.8 conflict idp destinAtions, by reGions (pnA2 dAtA only)

conflict idp destinations

†pnA2 data

%Acehbesar

(n=63)

%east

coast(n=143)

%central

highlands(n=124)

%southeasthighlands

(n=9)

%southwest

coast(n=256)

% total pnA2

sample(n=595)

north coast *Aceh besar **east coast ***central highlandssoutheast highlandssouthwest coast ****outside Aceh *****

010000000

2<18700

<110

320

89024

0001167220

<14

<10095<1

21321191

424

statistical significance). Thesefindings validate the designationby IOM, based in part on priorWorld Bank assessment data, ofthese villages as high conflictcommunities requiring specialassistance.

• post-conflict stress in mGKd Villages

Levelsofcurrentpost-conflictstressfollow a different pattern. PNA1respondents in MGKD villagesreported statistically significanthigher rates of current stressfulevents for 13 of 18 events on thelist. PNA2 respondents overall,as noted above, reported currentpost-conflict stressful events atrates significantly lower than inPNA1, and respondents in MGKDvillages reported higher rates foronly four of the 18 events on thelist.

summAry: conflict trAumA And collectiVe memory in AcehThe broad finding of the PNA2mappingofconflicteventsacrossAcehis that the East Coast and SouthwestCoast regions of the provinceexperiencedlevelsofviolenceashighas, and sometimes higher than, thereference communities of the NorthCoast sampled during PNA1. The

tAble 2.7 forced eVAcuAtions And other displAcement experience

displacement experiences

*pnA1 data †pnA2 data pnA1 + pnA2 data

% north coast

(n=596)

%Aceh besar

(n=180)

%east coast(n=246)

%central highlands

(n=327)

%southeast highlands

(n=162)

%southwest coast

(n=461)

% total pnA2 sample

(n=1,376)

%total sample

(n=1,972)

forced to flee burning buildingsforced to flee danger

3847

1333

4561

3740

612

4257

3345

3546

evacuation by force (pnA2 only)evacuation by choice (pnA2 only)

--

306

557

365

61

4317

389

--

“A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 200740 “A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007 41

tAble 2.9 current stressful eVents experienced by respondents

current stressors

*pnA1 data †pnA2 datapnA1 + pnA2

data

%north coast

(n=596)

%Aceh besar

(n=180)

%east

coast(n=246)

%central

highlands(n=327)

%southeast highlands(n=162)

%southwest

coast(n=461)

%total pnA2

sample(n=1376)

%total

sample(n=1972)

lack of proper place to livelack of water, sanitation facilitieshungry or lack of food

597572

323133

305642

446559

226256

455463

385554

446159

difficulty providing for your familydifficulty finding workdifficulty starting a livelihood

858971

605736

657555

797964

677052

778261

727556

768061

returned to find home destroyedlearned of death of family member, friendnot know what happened to family/friend

214514

10272

18297

183711

6283

203413

16329

183610

seeing perpetrators 47 3 16 10 0 5 7 19

rejection by family, communityfear of living with family, community

318

02

23

111

04

18

17

210

experienced assaultexperienced robbery

3121

00

33

12

00

11

11

107

change in religious valueschange in community values

1120

2116

5851

2133

1920

1511

2625

2124

Violence toward womenViolence toward children

47

01

3<0

32

00

11

11

23

* PNA1VS.PNA2districtlocations:Statisticallysignificantlydifferentatp<0.0001[AllEvents]. Except…. Rejectionbyfamily,community- p<0.05

† PNA2districtlocationscomparison: Lackofproperplacetolive p<0.0001 Lackofwater,sanitationfacilities p<0.0001 Hungryorlackoffood p<0.0001 Difficultyprovidingforyourfamily p<0.0001

Difficultyfindingwork p<0.0001 Difficultystartingalivelihood p<0.0001 Returnedtofindhomedestroyed p<0.0001 Learnedofdeathoffamilymember,friend NS Notknowwhathappenedtofamily/friend p<0.0001 Seeingperpetrators p<0.0001

Rejectionbyfamily,community NS Fearoflivingwithfamily,community p<0.0001 Experiencedassault p<0.05 Experiencedrobbery p<0.01 Changeinreligiousvalues p<0.0001 Changeincommunityvalues p<0.0001 Violencetowardwomen p<0.01 Violencetowardchildren NS

entireintegrateddatasetisusefulfordocumentingthesemacro-geographictrends, but in many cases can alsobeused topinpointmanyof thehotspot micro-localities at the districtand village levels, even in regionswith comparably lower levels in theirhistory of violence. Supplementedwith qualitative data, the stories ofwhat entire villages and theirindividual residents lived throughduring the conflict complementsthe quantitative data and fills in thedetails that the macro indicators donotcapture.

The PNA data on conflict traumaevents suggests nothing less thanan enormous reservoir of collectivememories of terrifying and sustainedviolence perpetrated againstordinary rural civilians in Aceh.Even in so-called low prevalenceregions, a consensus of memoryacknowledges the devastationinflicted upon the entire province.Fromindividuals,tovillages,districts,regions,andtheprovinceasawhole,the deep memory and real injuriesof Aceh’s violent history feed intocollective perceptions of the currentsecurity situationandanxietiesaboutthe future. In general psychosocialterms, the history of violenceinforms every person’s position andeverydaypracticeinAceh,andshouldtherefore properly inform and beaccounted for in the developmentof post-conflict and all otherhumanitarian interventions in Aceh.

In specific psychosocial terms, thehistory of violence in Aceh hasirrevocably altered thousands oflives, both in terms of tangibleeconomicsandpsychologicalsecurity,individual and collective, as will beamply demonstrated in the sectionsthatfollow.

“A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 200742 “A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007 4�

tAble 2.10 pAst trAumA eVents by pnA And mGKd VAriAble

traumatic events

*†pnA1 data *‡pnA2 data pnA1+pnA2

mGKd+(n=275)

mGKd-(n=321)

total (n=596)

mGKd+(n=454)

mGKd-(n=922)

total (n=1376)

total(n=1972)

experienced combat (bombing, fire fights)forced to flee burning buildingsforced to flee dangerforced to hide

84476021

73313612

78384716

82436013

6829387

7333459

74354611

beating to the bodyAttacked by knife or guntorturedserious physical injury from combatWitnessed physical punishmenthumiliated or shamed in public

483424166423

311913104512

392618135417

312217155218

191067336

231410103910

281712114412

rapeforced to rape a family memberother sexual assault

2<15

102

1<13

115

1<12

113

1<13

spouse killedspouse disappeared, kidnappedchild killedchild disappeared, kidnappedfamily member or friend killedfamily member or friend disappeared

63844841

42333426

53534133

3231

4132

2232

3530

22323731

3242

3831

Kidnappedcaptured, held by tni/ polri or GAm

816

39

512

414

36

49

410

sent to prisonforced separation from familyforced isolation

41411

264

398

6129

363

485

386

confiscation, destruction of propertyextortion, robbery

5541

3625

4533

5027

3117

3721

4024

forced labourforced to give food, shelter to tni or GAmforced to fight against tni or GAmpunished for not fighting against tni or GAmforced to search for corpsesnot allowed to provide muslim burial

36322915199

232317862

29272312126

2722241795

191212773

2215161084

2419181194

forced to injure family memberforced to injure non-family memberforced to destroy someone’s propertyforced to betray/ endanger family memberforced to betray/ endanger non-familysomeone forced to betray/ endanger youforced to humiliate another personforced to search for GAm member in forest

99499111147

453554526

673777835

4557610732

1132252

18

2234364

22

3435465

26

lack of shelter because of conflictlack of food, water because of conflictsick, lack of access to health care

279172

217449

248260

337766

257459

277562

267761

past trauma events experienced by informant† pnA1/mGKd+

vs. pnA1/mGKd-

‡ pnA2/mGKd+ vs.

pnA2/mGKd-

experienced combat (bombing, fire fights)forced to flee burning buildingsforced to flee dangerforced to hide

p<0.01p=0.0001p<0.0001p<0.01

p<0.0001p<0.0001p<0.0001p<0.001

beating to the bodyAttacked by knife or guntorturedserious physical injury from combatWitnessed physical punishmenthumiliated or shamed in public

p<0.0001p<0.0001p<0.001

NSp<0.0001p<0.001

p<0.0001p<0.0001p<0.0001p<0.0001p<0.0001p<0.0001

rapeforced to rape a family memberother sexual assault

NSNS

p<0.05

NSNS

p<0.05

spouse killedspouse disappeared, kidnappedchild killedchild disappeared, kidnappedfamily member or friend killedfamily member or friend disappeared

NSNS

p<0.05NS

p<0.001p<0.001

NSNSNS

p<0.05p<0.05

NS

Kidnappedcaptured, held by tni/ polri or GAmsent to prisonforced separation from familyforced isolation

p<0.01p<0.05

NSp<0.001p<0.01

NSp<0.0001p<0.01

p<0.0001p<0.0001

confiscation, destruction of propertyextortion, robbery

p<0.0001p<0.0001

p<0.0001p<0.0001

forced labourforced to give food, shelter to tni or GAmforced to fight against tni or GAmpunished for not fighting against tni or GAmforced to search for corpsesnot allowed to provide muslim burial

p<0.001p<0.05p<0.01p<0.01

p<0.0001p=0.0001

p<0.001p<0.0001p<0.0001p<0.0001

NSNS

forced to injure family memberforced to injure non-family memberforced to destroy someone’s propertyforced to betray/ endanger family memberforced to betray/ endanger non-familysomeone forced to betray/ endanger youforced to humiliate another personforced to search for GAm member in forest

p<0.05p<0.05

NSp<0.05p<0.05p<0.01p<0.05

p<0.0001

p<0.001p=0.0001

NSp<0.0001p<0.001p<0.001p<0.0001p<0.0001

lack of shelter because of conflictlack of food, water because of conflictsick, lack of access to health care

NSp<0.0001p<0.0001

p<0.01NS

p<0.05

“A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 200744 “A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007 4�

NOTE: MGKDvariableanswers“yes”or“no”ifIOMhasacommunitypeacedividendproject (called MGKD) in the village where therespondentlives.

* PNA1/MGKD+,PNA1/MGKD-,PNA2/MGKD+,PNA2/MGKD-comparison- Statisticallysignificantlydifferentatp<0.0001[Allmeasurements] Except…. LackofShelter p<0.01 Rape NS Forcedtorapefamilymember NS Othersexualabuse p<0.05 Senttoprison p<0.05 Kidnapped p<0.01 NoMuslimburial p=0.0001 ForcedtodestroyS.O.property NS Spousekilled p<0.05 Childkilled p=0.01 Familymemberorfriendkilled p<0.001 Spousekidnappedordisappeared NS Childkidnappedordisappeared NS Familymember/friendkidnapped/disappeared p<0.01

tAble 2.10 pAst trAumA eVents by pnA And mGKd VAriAble (continue from page 43)

tAble 2.11 current stressful eVents experienced by respondents, by pnA And mGKd desiGnAtion

current stressors

*†pnA1 data *‡pnA2 data pnA1+pnA2

mGKd+(n=275)

mGKd-(n=321)

total (n=596)

mGKd+(n=454)

mGKd-(n=922)

total (n=1,376)

total(n=1,972)

lack of proper place to livelack of water, sanitation facilitieshungry or lack of food

668482

526864

597572

416159

365251

385554

446159

difficulty providing for your familydifficulty finding workdifficulty starting a livelihood

909477

828666

858971

727859

727455

727556

768061

returned to find home destroyedlearned of death of family member, friendnot know what happened to family/friend

2850

15

1541

13

2145

14

2334

11

1332

7

1632

9

1836

10

seeing perpetrators 58 38 47 7 7 7 19

rejection by family, communityfear of living with family, community

321

215

318

27

16

17

210

experienced assaultexperienced robbery

3827

2416

3121

12

11

11

107

change in religious valueschange in community values

1324

1018

1120

2824

2426

2625

2124

Violence toward womenViolence toward children

69

35

47

21

11

11

23

† current (post-conflict) stressful events experienced by informant

pnA1/ mGKd+ vs. pnA1/ mGKd-

lack of proper place to livelack of water, sanitation facilitieshungry or lack of food

p<0.001p<0.0001p<0.0001

difficulty providing for your familydifficulty finding workdifficulty starting a livelihood

p<0.01p<0.01p<0.01

returned to find home destroyedlearned of death of family member, friendnot know what happened to family/friend

p<0.0001p<0.05

NS

seeing perpetrators p<0.0001

rejection by family, communityfear of living with family, community

NSp<0.05

experienced assaultexperienced robbery

p<0.001p<0.001

change in religious valueschange in community values

NSNS

Violence toward womenViolence toward children

p<0.05NS

† current (post-conflict) stressful events experienced by informant

pnA2/mGKd+ vs. pnA2/mGKd-

lack of proper place to livelack of water, sanitation facilitieshungry or lack of food

NSp<0.01p<0.01

difficulty providing for your familydifficulty finding workdifficulty starting a livelihood

NSNSNS

returned to find home destroyedlearned of death of family member, friendnot know what happened to family/ friend

p<0.0001NS

p<0.05

seeing perpetrators NS

rejection by family, communityfear of living with family, community

NSNS

experienced assaultexperienced robbery

NSNS

change in religious valueschange in community values

NSNS

Violence toward womenViolence toward children

NSNS

NOTE: MGKD variable answers “yes” or“no” of IOM has a communitypeace dividend project (calledMGKD) in the village where therespondentlives.

* PNA1/MGKD+,PNA1/MGKD-,PNA2/MGKD+,PNA2/MGKD-comparison-

Statisticallysignificantlydifferentat p<0.0001[Allmeasurements] Except…

Notknowwhathappenedtofamilymember/friendp<0.001

Rejectionbyfamily/communityp<0.05

“A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 20074� “A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007 47

the psychosociAl And psycholoGicAl remAinders of Violence: depression, Anxiety And trAumAtic stress disorders

methods• measures of psychological

distress and neuropsychiatric disorders

The survey questionnaire openswith a very general, open-endedset of questions about the effectsof the conflict on individuals andtheir lives; “were you or anyone inyourfamilyavictimoftheconflict?If so, who? Can you describe whathappened?”

Thisisfollowedbyaskingaboutthelist of traumatic events, discussedabove. Similarly, the section onpsychological distress beginswith very open-ended questions;”the conflict has brought uniquepressuresupontheAcehnesepeopleduring the past number of years.Have thesepressureshadaneffecton your feelings, energy, or yourhealth in your daily life? Can youexplainwhatthiseffecthasbeen?”

This is then followed by a verygeneral self-assessment, with aquantitative response; “in the pastyear, have you ever had difficultieswithyourinnerfeelingsorthewayyoufeel(forexample,feltdepressedor often sad, anxious, fearful,or not being able to control youranger)?” “If yes, how serious wasthis?” (This was measured by a1-4 scale, from ‘not serious’ to‘extremely serious’). “If yes, inyouropinionwere thesecausedbystress or trauma connected to theconflict?”

Thisgeneralquestionwasfollowed

by asking respondents to reporton psychological symptoms orproblems they experienced inthe past week, using a 25 itemversion of the Hopkins SymptomCheckList (HSCL) forDepressionand Anxiety. Fifteen symptomsassociated with depression and10 symptoms associated withanxiety were asked, andrespondentswereaskedtodescribewhether they had experiencedtheseduringthepastweek‘notatall,’‘a little,’ ‘sometimes,’ and ‘often.’This scale is incorporated into theHarvardTraumaQuestionnaireandhasbeenusedwidelyindisasterandtrauma community assessments ofemotionaldistress(Mollica2004).

In addition, respondents wereasked to tell the interviewer, usingthesameformat,whethertheyhadexperiencedsymptomsorproblemswhicharelistedaspartofthe42itemHarvard Trauma Questionnaire(HTQ), developed by Mollica andhis team for use in conflict areas.The HTQ is a broad measure ofsymptoms associated with traumaand dissociation, which includesa 16 item core used to assess PostTraumaticStressDisorder(PTSD).

Care was taken to incorporatecommon ways of expressingpsychologicaldistress in Indonesia,andspecifically inAceh, intothesequestions. The full questionnaireiswritten inBahasa Indonesia,notin Bahasa Aceh and interviewerssometimes explained particularquestions in Acehnese or another

local language in Aceh, whenneeded. Items on the HSCLand HTQ were translated usingcommon Indonesian terms, whenavailable–suchasbingung(feelingconfused), melamun (day-dreamingor ‘spacing out’), and pusing (acombination of feeling dizzy andhavingaheadache).

In addition, items designed tocapture popular discourses aboutdisturbingexperiencespost-tsunamiand post-conflict were includedin the questionnaire to elicitexperiences of nightmares, ghosts,spirits,andhearingvoicesofpeoplewhohaddied,experiencesweknewto be important from previousethnographicresearch.

A four item measure was includedfrom the Harvard TraumaQuestionnaire to assess presenceand severity of events that mighthave produced head trauma orbrain injury, including beatingsto the head, suffocation orstrangulation, near drowning, andotherphysicalinjuriestothehead.

• Analyses of psychological symptoms and psychiatric diagnoses

Psychological distress can beconceptualized in two ways: asa ‘continuous variable,’ i.e., asa level of distress or symptoms,such as depression or anxiety,ranging continuously from verylow levels to very high levels; andas a ‘dichotomous variable,’ i.e.,as being either high or low, as

beinga‘case’ornot(forexample,a case of depression or anxiety,or a case requiring treatment) oras meeting criteria for a clinicaldiagnosis (for example, of majordepressivedisorder,panicdisorder,or Post Traumatic Stress Disorder)or not meeting criteria for adiagnosis.

Psychological symptom checklistsare designed primarily to be usedas continuous variables in clinicalwork or research – to answer suchquestions as ‘is this patient feelingbetter than he or she did onemonth ago?’ or ‘are psychologicalsymptoms especially high insome risk groups?’ or ‘are levelsof psychological distress highlycorrelated with levels of stressor numbers of traumatic eventsexperienced?’

On the other hand, questionssuch as ‘what percentage ofpersons in this village suffer majordepressive disorder?’ requiremaking dichotomous ratings,determining whether someone isor is not a ‘case’ of depression ordoes or does not meet diagnosticcriteria for Post Traumatic StressDisorder (PTSD). Such variablescan then be analysed using oddsanalysis, indicating elevated riskfor suffering a particular illness

among persons with particularcharacteristics in comparison withothers.

Inmentalhealthsurveys,therearetwomethodsusedfortransforminga continuous variable into adichotomous variable. First, onecanmakeadeterminationthatanyrespondent who reports symptomsabove a particular level will bejudgedtobea‘case’–forexample,defining someone who hassymptoms for a depression abovea certain level, and therefore inneed of mental health services.The level the analyst sets for thecut-off point, along with the levelof symptoms in the community,will determine what percentageof persons are considered to be‘cases’.

Second, one can use a diagnosticalgorithm, based on currentpsychiatric diagnostic practices. Ifa respondent indicates that he orshe has experienced a particularcombination of symptoms thatserve as criteria for a particulardiagnosis(majordepressivedisorderor post-traumatic stress disorder,for example), that person may berated as ‘meeting criteria’ for thatdisorder.

In what follows, we report our

findings in four ways. First, wefollow the standard procedurerecommended by Mollica et al touse a mean of 1.75 on depressionitems on the HSCL 15-itemdepression scale and 10-itemanxiety scale as cut-off points, and2.50 on the 42 trauma symptomson the HTQ, to identify a personas suffering depression or a post-traumaticdisorder.8

Using this method allows us tocompare findings for the Acehsample with similar samplesfrom high conflict areassuch as Bosnia or Cambodia.Second,forsomeanalyses,weusedmoreconservativeorhighthresholdcut-offs, 3.0 on depression andanxietyitemsontheHSCLand3.0onthetraumasymptomsontheHTQ.Raising the cut-off levels identifiesa smallergroupof individualswhoarecurrentlysufferingmoreseveresymptoms,andallowsustoaskwhatgroupsofpersonsorwhatformsoftraumatic experience place anindividual at particularly high riskfor suffering the most severepsychiatric distress. We usedthis method in PNA1 analysesbecause symptom levels wereso high that it was necessaryto identify a smaller group ofpersons with the highest level ofsymptoms.

8 Mollica,RichardF.,LauraS.MadDonald,MichaelMassagli, andDerrickM.Silove.2004.MeasuringTrauma,MeasuringTorture.Instructions

and Guidance on the Utilization of the Harvard Program in Refugee Traumas Versions of The Hopkins Symptom Checklist-25 (HSCL-25) &

TheHarvardTraumaQuestionnaire(HTQ).Cambridge,MA:HarvardPrograminRefugeeTrauma.

“A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 20074� “A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007 4�

Third, we followed the algorithmdevised by Mollica et al todetermine whether individualssuffer particular constellations ofsymptoms associated withdepressive illness or PTSD,according to the AmericanPsychiatric Association’s Diagnosticand Statistical Manual 4th edition(DSM-IV). Because this algorithmis based on symptoms from asymptom checklist rather than apsychological interview designedexplicitly to determine a clinicaldiagnosis, and because they askabout the presence of thesymptoms during the past weekbut not the duration of thesymptoms, these ratings can beconsidered approximations only.Theydo,however,indicatelevelsofdepression and trauma-relatedillnesses in these communities,suggesting levels of need forservices.

Atotalof14depressionitemsfromthe HSCL were included withinthe depression algorithm (seeTable 3.1). Individuals wereconsidered to be suffering aparticular symptom if they ratedthemselves 3 or 4 on a particularitem. In order to be classified assymptomatic for depression, asubject initially needed a positiveresponse on any of the depressedmood or decreased interest/pleasure items Additionally, apositive score on 4 out of the 6DSM-IV Criterion A symptoms wasrequired for positive classification.

A positive score on 3 out of the 6DSM-IV Criterion A symptoms wasrequired when positive responsesfor both depressed mood anddecreased interest/ pleasure werepresent.91011

A more conservative or high

threshold algorithm was alsoexamined. In this case, questionswere ‘checklist positive’ only ifratings were at the highest level,4. All other steps in the primarydepression algorithm remainedthe same. Again, this method wasused in PNA1 because symptomlevels were so high. We providesimilar analyses in PNA2 forpurposesofcomparison.

A total of 16 Harvard TraumaQuestionnaire (HTQ) itemswere included within the PTSDalgorithm. Individuals wereconsidered to be suffering aparticular symptom if they ratedthemselves 3 or 4 on a particularitem. In order to be classified assymptomatic for PTSD or meetingdiagnostic criteria for PTSD, asubject needed a positive responseon 1 or more re-experiencingsymptoms,3ormoreavoidanceandnumbingsymptoms,and2ormorearousal symptoms. (See Table 3.2)Subject exposure to a traumaticevent (Criterion A) has beenassumedforallrespondents.

Onceagain,amoreconservativeorhigh threshold algorithm was alsoexamined. In this case, questions

were checklist positive if ratingswere 4 only. All other steps in theprimaryPTSDalgorithmremainedthesame.

findinGsTables 3.3 through 3.6 providefindings concerning self-perceivedlevels of general emotional distress,as well as symptoms and diagnosesof depression anxiety and PTSD, bygender, study (PNA1 vs PNA2), thesix geographical categories, and bysampledesignationasMGKDpresentor absent. The findings compare theextremelyhighlevelsofpsychologicaldistressamongrespondentsinthefirststudy PNA1 (interviewed in February2005),withdistresslevelsexhibitedinthenewlysurveyedpopulation,PNA2,undertaken indifferent geographicallocations and 5 months after thePNA1study.

• self - Assessment of emotional distress

Table3.3reportsfindingsfromthethree general questions designedtoassessrespondents’generalsenseof emotional distress over the pastyear.Inansweringthesequestions,78%ofrespondents inPNA1fromNorth Coast reported they hadexperienced general psychologicaldistressinthepastyear,incontrastwith 51% of PNA2 respondents.While 97% of PNA1 respondentsfelt the distress was caused by theconflict and rated their level ofdistress as 2.94 on degree ofseriousness(1-4scale),81%ofPNA

2respondentsfelttheirdistresswascausedby theconsequencesof theconflict and rated their distresslevels at 2.54 on degree ofseriousness(1-4scale).

The district variations for PNA2clearly mirror levels of traumaticevents experienced. East Coast(68%)andSouthwestCoast(60%)regions reported high levelsof experiencing psychologicaldistress in the past year at ratesonly somewhat lower than PNA1North Coast respondents (78%),while Aceh Besar (22%) andSoutheast Highlands (33%)respondents reported much lowerrates of such experiences. Forthose who reported distress,respondents from the East Coast(91%), Central Highlands (87%),and Southwest Coast (81%) weremost likely to report distress asresulting from the conflict andrated their distress as mostsevere (with the East Coast levelscomparable to PNA1 North Coast

findings). Differences betweenthe PNA1 and PNA2 and amongPNA2 regions are statisticallysignificantatp<0.0001.

These findings suggest that whenasked to report on distress in thepast year, respondents in thehighest conflict regions of PNA2report levels of distress similar to(though somewhat lower than)PNA1 respondents, while PNA2respondents in the lower conflictregions report significantly lowerpsychologicaldistress.

• psychological and psychiatric diagnoses using formal instruments

Theprimarymethodofmeasuringpsychological distress experiencedby individuals in the sampledhighconflict villages was through theuse of standardized psychologicalsymptomquestionnaires.

These ask individuals to reportwhether they have experienced

particularsymptomsinthepastoneweek and if so at what level, on a1-4 scale. Usingcut-off scores anddiagnostic algorithms, these havebeen converted into categoricalvariables,reportedaspercentageofpersons who are “symptomatic” or“highsymptomatic”andwho“meetcriteria” for depression or PTSDor “meet high threshold criteria”for that disorder. These questionsthus rate respondents’ experienceat the time of the interview (overthe past week), rather than askingthem to describe experiences overthepastyear.

Tables3.4and3.5examinesymptomlevelsanddiagnosesfordepression,PTSD, and anxiety by region andgender,forbothPNA1andPNA2.

Respondents from PNA2 suffermental health problems associatedwith the violence at a significantlevel. Using exactly the samemethodsusedinPNA1,35%ofthetotal PNA2 sample ranked high

tAble 3.1 hscl core diAGnostic depression symptoms

depressed mood•Cryingeasily•Feelinghopelessaboutthefuture•Feelingblue•Feelinglonely

diminished interest/pleasure•Feelingnointerestinthings•Lossofsexualinterestorpleasure

dsm-iV criterion A symptoms•Poorappetite•Difficultyfallingasleeporstayingasleep•Feelinglowinenergyand/orFeelingeverythingisaneffort•Blamingyourselfforthings•Worryingtoomuchaboutthingsand/orFeelingsofworthlessness•Thoughtsofendingyourlife

Each symptom was rated as having bothered or distressed the respondent “Not at all”, “A little”, “Quite a bit”, or “Extremely often”(1-4respectively)duringthepastweek.

9 Mollica et al. “Disability Associated with Psychiatric Comorbidity and Health Status in Bosnian Refugees Living in Croatia” in Journal of the

American Medical Association (JAMA).Volume282(5),04August1999,pp433-439.

10 Mollica et al. “Dose-effect Relationships of Trauma to Symptoms of Depression and Post-Traumatic Stress Disorder Among Cambodian

SurvivorsofMassViolence”inThe British Journal of Psychiatry.Volume173(12),December1998,pp482-488.

11 Sabin et al. “Factors Associated with Poor Mental Health Among Guatemalan Refugees Living in Mexico 20 Years After Civil Conflict” in

Journal of the American Medical Association (JAMA).Volume290(5),06August2003,pp635-642.

“A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007�0 “A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007 �1

on depression symptoms, 39% onanxiety symptoms, and 10% onPTSDsymptoms.(Usingdiagnosticcriteria, corresponding rates are23%ofrespondentsmeetingcriteriafor major depressive disorder and

surveyed in July (and November)2006 reported substantially lowerrates of major depression, anxietydisorders, and PTSD than thosein PNA1. PNA1 reported rates ofpsychologicalsymptomsamongtheNorth Coast respondents surveyedin February 2006 at some of thehighest levels reported for post-conflict settings worldwide. Thepercentage of PNA2 respondentswho suffered high levels ofsymptoms for depression, anxiety,and PTSD (35%, 39%, 10%respectively), while high, aremuch lower than comparablerates reported from PNA1 (65%,69%, and 34%). Because ratesusingthesecutoffsweresohigh inPNA1,wedevelopedanalysesusinghigher cutoffs to identify those atrisk for the most severe forms ofdepression,anxiety,andPTSD.

While rates for high thresholdmental health problems werealso found to be extremely highin PNA1, they are substantiallyand significantly lower for PNA2respondents. For example, while18% of the total population ofPNA1on theNorthCoastmet theextremely high threshold criteriafor depression, only 4% of thetotalPNA2samplemetthesemore

stringent criteria. And even in theSouthwest Coast, the East Coast,andtheCentralHighlands,amoremodest 6%, 5% and 9% of thesample met these more stringentcriteriafordepression.

Rates of high threshold PTSDare even lower in PNA2 districts,including those regions withthe highest trauma scores, incomparisonwithPNA1.

Villagesdesignatedashighconflictby the IOM and World Bankand selected for the IOM Post-Conflict Community ReintegrationProgramme(i.e.theMGKDvillages– described above in the sectionon sampling), showed substantiallyand significantly higher rates ofall psychological symptoms in thePNA1 study. (See Table 3.6; thesedata were not reported in thePNA1Report.)Strikingly,Table3.6indicatesthatthisisnottrueforthePNA2respondents.

Althoughformanyofthesymptommeasures in PNA2, MGKDrespondents report slightly higherlevels of symptoms than NON-MGKD respondents, not one ofthese differences is statisticallysignificant.

tAble 3.3 GenerAl psycholoGicAl distress by reGions

* pnA1 Vs. pnA2 district locations: Statistically significantly different atp<0.0001

† pnA2 district locations comparison: Experience general psychological

distressp<0.0001

Causedbytheconflict?p<0.0001

Seriousness(1-4scale–mean(SD))p<0.0001

General emotional distress experienced by informants

*pnA1 data †pnA2 data pnA1 + pnA2 data

% north coast

(n=596)

%Aceh besar

(n=180)

%east coast(n=246)

%central highlands

(n=327)

%southeast highlands

(n=162)

%southwest coast

(n=461)

% total pnA2 sample

(n=1,376)

%total sample

(n=1,972)

experienced general psychological distress during the past year?

78 22 68 48 33 60 51 59

if yes…caused by the conflict? 97 56 91 87 45 81 81 87

if yes…seriousness (1-4 scale – mean (sd)) 2.94(0.85)

2.10(0.88)

2.89(1.11)

2.56(0.78)

2.15(0.69)

2.44(0.91)

2.54(0.95)

2.70(0.93)

tAble 3.2 hArVArd trAumA QuestionnAire (htQ) core diAGnostic symptoms

re-experiencing symptoms (dsm-iV criterion b)•Recurrentthoughtsormemoriesofthemosthurtfulorterrifyingevents•Feelingasthoughtheeventishappeningagain•Recurrentnightmares•Suddenemotionalorphysicalreactionwhenremindedofthemosthurtfulortraumaticevents

Avoidance and numbing symptoms (dsm-iV criterion c)•Feelingdetachedorwithdrawnfrompeople•Unabletofeelemotions•Avoidingdoingthingsorgoingplacesthatremindyouofthetraumaticorhurtfulevents•Inabilitytorememberpartsofthemosttraumaticorhurtfulevents•Lessinterestindailyactivities•Feelingasifyoudon’thaveafuture•Avoidingthoughtsorfeelingsassociatedwiththetraumaticorhurtfulevents

Arousal symptoms (dsm-iV criterion d)•Feelingjumpy,easilystartled•Difficultyconcentrating•Troublesleeping•Feelingonguard•Feelingirritableorhavingoutburstsofanger

Eachsymptomwas ratedashavingbotheredordistressed therespondent “Notatall”, “A little”, “Quiteabit”,or “Extremelyoften”(1-4respectively)duringthepastweek.

12%meetingcriteriaforPTSD.)

Rates are substantially higher forrespondents from the SouthwestCoast, East Coast, and CentralHighlands – areas that suffered

the highest levels of violence. Forexample, 41% of respondentsfrom the Southwest coast suffereddepressive symptoms aboveinternationally recognised cutofflevels, 43% anxiety symptoms and

14%PTSDsymptomsatsuchlevels.This follows the pattern of PNA1,in which levels of symptoms werefound to be significantlyhigher in districts with higherlevels of traumatic events. TheCentral Highlands is somewhatexceptional. Although traumaticevents are somewhat lower in theCentralHighlandsthanintheEastCoastandSouthwestCoastdistricts,psychological symptoms are ashigh as or higher than symptomlevelsinthoseregions.

This may be linked to ethnic

differences in symptom reporting(80% of Central Highlandsrespondents are Gayo, as opposedto being ethnically Acehnese), orto differences in general livingconditions in this region. ThePNA2 thus finds significant levelsof trauma-related mental healthproblems, deserving specializedinterventions, with the highestprioritiesincludingtheNorthCoast(from PNA1), the adjoining EastCoast districts, the South Coast(particularlyAcehSelatan),andtheCentralHighlands,inthisorder.

At the same time, a highlysignificant finding of PNA2 isthat respondents in all districts

“A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007�2 “A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007 ��

tAble 3.4 depression, ptsd, And Anxiety symptoms And diAGnoses by reGions

* PNA1 VS. PNA2 district locations: Statistically significantlydifferentatp<0.0001[Allmeasurements]

† PNA2districtlocationscomparison: p<0.0001 [Depression-initialalgorithm] p<0.05 [Depression-revisedalgorithm] p<0.0001 [Depressionscore->1.75] NS [Depressionscore->3.00]

p<0.0001 [PTSD-initialalgorithm] p<0.05 [PTSD-revisedalgorithm] p<0.0001 [PTSDscore->2.50] NS [PTSDscore->3.00]

p<0.0001 [Anxietyscore->1.75] p<0.001 [Anxietyscore->3.00]

formal measures of psychological symptoms and diagnoses

*pnA1 data †pnA2 data pnA1 + pnA2 data

% north coast

total pnA1 sample(n=596)

%Aceh besar

(n=180)

%east coast(n=246)

%central highlands

(n=327)

%southeast highlands(n=162)

%southwest

coast(n=461)

% total pnA2 sample

(n=1,376)

%total sample

(n=1,972)

meet criteria for major depressive disorder

55 13 24 23 16 29 23 33

meet high threshold criteria for major depressive disorder

18 3 5 9 10 6 7 10

mean depression score “symptomatic” (>1.75)

65 20 33 43 25 41 35 44

mean depression score “high symptomatic” (>3)

17 2 5 4 7 4 4 8

meet criteria for ptsd 36 4 9 15 6 17 12 19

meet high threshold criteria for ptsd 10 1 2 5 4 2 3 5

mean ptsd score “symptomatic” (>2.5)

34 2 8 12 6 14 10 17

mean ptsd score “high symptomatic” (>3)

16 1 4 4 5 5 4 8

mean Anxiety score “symptomatic” (>1.75)

69 31 42 43 25 43 39 48

mean Anxiety score “high symptomatic” (>3)

33 3 14 9 5 10 9 16

tAble 3.5 depression, ptsd, And Anxiety symptoms And diAGnoses by Gender

* PNA1VS.PNA2gendercomparisons (PNA1Malesvs.PNA2MalesandPNA1Femalesvs.PNA2Females):

Statisticallysignificantlydifferentatp<0.0001[Allmeasurements]

† PNA1withingendercomparison: NS[Depression-initialalgorithm] NS[Depression-revisedalgorithm] NS[Depressionscore->1.75] NS[Depressionscore->3.00] NS[PTSD-initialalgorithm] NS[PTSD-revisedalgorithm] NS[PTSDscore->2.50] NS[PTSDscore->3.00] p<0.01[Anxietyscore->1.75] NS[Anxietyscore->3.00]

PNA2withingendercomparison: p<0.05[Depression-initialalgorithm] NS[Depression-revisedalgorithm] p<0.001[Depressionscore->1.75] p<0.05[Depressionscore->3.00] NS[PTSD-initialalgorithm] NS[PTSD-revisedalgorithm] p<0.05[PTSDscore->2.50] NS[PTSDscore->3.00] p<0.0001[Anxietyscore->1.75] NS[Anxietyscore->3.00]

formal measures of psychological symptoms and diagnoses

*pnA1 data †pnA2 data pnA1 + pnA2

% male

(n=315)

% female(n=281)

% total pnA1 sample

(n=596)

% male

(n=691)

% female(n=685)

% total pnA2 sample

(n=1,376)

%total sample

(n=1,972)

meet criteria for major depressive disorder

54 57 55 20 26 23 33

meet high threshold criteria for major depressive disorder

18 19 18 6 7 7 10

mean depression score “symptomatic” (>1.75)

64 67 65 31 40 35 44

mean depression score “high symptomatic” (>3)

16 18 17 3 5 4 8

meet criteria for ptsd 37 35 36 11 13 12 19

meet high threshold criteria for ptsd 11 10 10 3 3 3 5

mean ptsd score “symptomatic” (>2.5)

33 35 34 8 12 10 17

mean ptsd score “high symptomatic” (>3)

17 16 16 3 4 4 8

mean Anxiety score “symptomatic” (>1.75)

64 75 69 33 46 39 48

mean Anxiety score “high symptomatic” (>3)

30 36 33 8 10 9 16

“A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007�4 “A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007 ��

Thus, while levels of traumaticexperienceweresignificantlyhigherfor respondents in those PNA2villages selected for the MGKDprogramme, these respondentsdid not show higher levels ofpsychologicalsymptoms.Thisraisesfurtherquestionsabout the reasonfor the reduction of psychologicalsymptoms in the PNA2 study incomparisonwiththePNA1study.

• the distribution of risk: the effects of traumatic experiences on psychological distress

A major question for the overallpsychosocial assessment is thequestion of what places persons atrisk for mental health problems.Wehavebeenparticularlyinterestedin the extent to which the level of

traumatic incidents individualsexperience places them at riskfor mental health problems, andwhat gender and age groups areat particular risk for traumaticexperience and mental healthproblems.InthePNA1report,theserelationships are analysed utilizingadjustedoddsratiosfordepressionand PTSD. This is a statisticalmodelthatallowsonetodeterminehow much the risk for an illnesslike depression is increased forthose who have suffered particularpatterns of traumatic violence orotherriskfactors.

PNA1 showed that the strongestpredictor of mental healthsymptoms was the number oftraumatic experiences a personhad suffered. PNA1 showed only

modest patterns of special riskby age and gender. Here we usethe same statistical model forcalculating risk associated withincreased levels of past traumaticeventsandcurrentstressors,aswellas for particular gender and agegroups.

Table 3.7 shows the effects of pasttraumaticeventson increasingriskfor symptoms and diagnoses ofdepression, PTSD, and anxiety forthePNA2sample.Forallmeasures,an increase in numbers of pasttraumatic events is shown to placepersonsatdirectlyincreasinglevelsriskformentalhealthproblems.

Persons who experienced thehighest level of traumatic eventshave3to12timestheriskofamental

health problem when comparedwith persons who did not suffer atraumatic event, and their risk ofsuffering themore severe formsofPTSD (highest symptom cutoff orhighest threshold for diagnosis)are increased 7 to 12 times. ThistablemaybecompareddirectlywithTable9.1ofthePNA1report.12

For PNA1 respondents, becausesymptoms were reported at muchhigher levels, odds ratios are alsomuchhigher.However,thepatternpersists of a powerful relationshipbetweentraumaticeventsassociatedwith conflict and continued levelsofmentalhealthproblems.

Table 3.8 shows the pattern forinfluence of ‘current stressors’on mental health symptoms and

diagnoses. Increased levels ofcurrent stressors is associated withincreased risk for mental healthproblems in PNA2, although lessdirectlyandstronglythanfoundforthe PNA1 respondents. (CompareTable3.8herewithTable9.2inthePNA1report).

The total number of ‘currentstressors’ associated with realinsecurity and violent events –particularly thenumberofpersonswho report seeing perpetrators orexperiencingattackorrobbery–ismuch lower in PNA2 than PNA1,making it somewhat difficult tointerpret these findings for PNA2.Nonetheless, reported experiencesof post-conflict stressors andinsecurities of daily living increaserisk for mental health measures

by3to13times.

• the distribution of risk: Groups at high risk

Tables 3.9 examines adjustedodds ratios for depression, PTSD,and anxiety by study (PNA1 andPNA2)andbygender.Asindicatedin the previous analyses, PNA1respondents were four times morelikely than PNA2 respondentsto meet criteria for depression,PTSD, and 4.7 times more likelyto score high on anxiety symptommeasures. The PNA1 versusPNA2 differences are statisticallysignificant (<p.0.0001) for allpsychiatric symptom measures.Table 3.5 (above) indicates thata slightly higher percentage ofwomenthanmensufferdepression,PTSD,andanxiety.Thisisverified

tAble 3.6 depression, ptsd, And Anxiety symptoms And diAGnoses by mGKd desiGnAtion NOTE: MGKD- answers “yes” or “no” if IOM/PIKR has a community peace dividend project(calledMGKD)inthevillagewheretherespondentlives.

* PNA1/MGKD+,PNA1/MGKD-,PNA2/MGKD+,PNA2/MGKD-comparison- Statisticallysignificantlydifferentatp<0.0001[Allmeasurements]

† PNA1/MGKD+,PNA1/MGKD-comparison: p<0.0001 [Depression-initialalgorithm] p<0.01 [Depression-revisedalgorithm] p<0.0001 [Depressionscore->1.75] p<0.01 [Depressionscore->3.00] p<0.01 [PTSD-initialalgorithm] p<0.05 [PTSD-revisedalgorithm] p<0.001 [PTSDscore->2.50] p<0.05 [PTSDscore->3.00] p<0.01 [Anxietyscore->1.75] p<0.001 [Anxietyscore->3.00]

‡ PNA2/MGKD+,PNA2/MGKD-comparison: NS [Depression-initialalgorithm] NS [Depression-revisedalgorithm] NS [Depressionscore->1.75] [Depressionscore->3.00] NS [PTSD-initialalgorithm] NS [PTSD-revisedalgorithm] NS [PTSDscore->2.50] NS [PTSDscore->3.00] NS [Anxietyscore->1.75] NS [Anxietyscore->3.00]

formal measures of psychological symptoms and diagnoses

*pnA1 data †pnA2 data pnA1 + pnA2

mGKd+(n=275)

mGKd-(n=321)

total(n=596)

mGKd+(n=454)

mGKd-(n=922)

total(n=1376)

total (n=1972)

meet criteria for major depressive disorder

64 48 55 25 22 23 33

meet high threshold criteria for major depressive disorder

23 14 18 8 6 7 10

mean depression score “symptomatic” (>1.75)

74 58 65 36 35 35 44

mean depression score “high symptomatic” (>3)

22 13 17 5 4 4 8

meet criteria for ptsd 43 31 36 13 11 12 19

meet high threshold criteria for ptsd 14 8 10 3 3 3 5

mean ptsd score “symptomatic” (>2.5)

41 27 34 11 9 10 17

mean ptsd score “high symptomatic” (>3)

20 13 16 5 3 4 8

mean Anxiety score “symptomatic” (>1.75)

75 65 69 42 38 39 48

mean Anxiety score “high symptomatic” (>3)

40 27 33 10 9 9 16

12Good,B.,M.-J.D.Good,J.Grayman,andM.Lakoma.2006.Psychosocial Needs Assessment of Communities Affected by the Conflict in the Districts of Pidie,

Bireuen, and Aceh Utara.InternationalOrganizationforMigration.

“A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007�� “A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007 �7

by odds analyses, where men areat lower risk for all categories ofsymptoms.

However, these differences are notstatistically significant, except fordepression and anxiety symptomsatthe“Symptomatic”level(womenscoring higher than men p<.001and p<.0001), and depression,that is theymeetcriteria forMajorDepressive Disorder (MDD),PTSD “Symptomatic”, and Anxiety“High Symptomatic,” which are ofborderline significance (p<.05). Itshouldbenotedthatthedifferencesbetween women and men are less

than in many normal populations,where women often show muchhigher levels of depression thanmen, suggesting that men maybe somewhat more affected byconflict related symptoms thanwomen.

Table3.10examinestherelationshipbetween age and mental healthproblemsforbothPNA1andPNA2,adjusted for gender and MGKDprogramme. What is interestingto note in these two tables is theapparentshiftinriskforagegroupsfrom PNA1 to PNA2. In PNA1,young adults (age 17-29) were at

greaterriskfornearlyallcategoriesof mental health problems thanall other age groups, though mostoften at statistically non-significantlevels.

For PNA2, the pattern is quitedifferent. Young adults have thelowest risk for mental healthproblems, whereas other groups,particularly the oldest group(54 years and older), are at thehighest risk, particularly fordepression and anxiety. Forexample, older PNA2 respondentsare over 2.5 times more likely tomeet criteria for depression than

OddsRatios:CI95%Note:“0-3”Events=referencegroup.*Adjustedfordistrictlocations. DistrictLocationsinclude -AcehBesar (N=180) -EastCoastdistricts (N=246) -CentralHighlands (N=327) -SouthwestHighlands (N=162) -SouthwestCoast (N=461)

formal measures of psychological symptoms and diagnoses for pnA2 (n=1376)

number of past traumatic events

0-3 4-7 8-10 >11

meet criteria for major depressive disorder 1.00 1.43(0.99-2.09)

§2.25(1.46-3.48)

||3.56(2.43-5.22)

meet high threshold criteria for major depressive disorder 1.00 1.43(0.72-2.84)

1.87(0.80-4.38)

||5.98(3.04-11.75)

mean depression score “symptomatic” (>1.75) 1.00 †1.39(1.01-1.90)

||2.11(1.44-3.08)

||3.53(2.52-4.96)

mean depression score “high symptomatic” (>3) 1.00 0.62(0.24-1.58)

0.90(0.28-2.94)

||4.73(2.16-10.35)

meet criteria for ptsd 1.00 0.88(0.53-1.48)

1.02(0.54-1.92)

||3.73(2.32-5.99)

meet high threshold criteria for ptsd 1.00 0.83(0.23-3.04)

3.07(0.84-11.25)

||12.12(4.15-35.37)

mean ptsd score “symptomatic” (>2.5) 1.00 1.13(0.63-2.04)

1.57(0.80-3.10)

||4.34(2.51-7.49)

mean ptsd score “high symptomatic” (>3) 1.00 0.68(0.22-2.09)

2.01(0.65-6.25)

||7.16(2.91-17.60)

mean Anxiety score “symptomatic” (>1.75) 1.00 †1.42(1.05-1.91)

||2.48(1.72-3.58)

||3.30(2.37-4.59)

mean Anxiety score “high symptomatic” (>3) 1.00 1.52(0.84-2.73)

1.65(0.83-3.30)

||3.62(2.05-6.41)

tAble 3.7 AdJusted odds rAtios*: mentAl heAlth problems by pAst trAumAtic eVents for pnA2 respondents

† p<0.05‡ p<0.01§ p<0.001|| p<0.

17-29 year olds. This suggests aninteresting pattern of resilienceamong the young adults in thispopulation, as well as a specialriskamongelders.

• the distribution of risk: head trauma

One form of trauma isparticularly noteworthy amongPNA2 respondents, as it was forrespondents in PNA1. Rates ofhead trauma and potential braininjury, suffered through beatings,strangulation,neardrownings,andother formsof tortureor violence,used by the Indonesian forces

to gather information or punishvillages for perceived support forGAM,wereextraordinarilyhigh.

As our clinical work in Bireuenhas shown, head trauma may havelong term effects, both physicallyand psychologically, and deserveclinical interventions and furtherresearch.

Overall, 17% of the entire PNA2sample suffered from conflictrelated head trauma; these areextremely high rates, consideringthatthisincludeshighandrelativelylow conflict areas, and may be

compared with the rate of 27%forthePNA1sample.InthePNA2districts rating particularly highfor head trauma, rates werecomparable with those foundin the PNA1 study on the NorthCoast.InPNA2,27%ofEastCoastrespondentsand24%ofSouthwestCoast respondents experiencedconflict related head trauma,including beating on the head,suffocation, and near drowning.As mentioned above, while bothmen and women suffered headtrauma, men, particularly those inthehighestconflictregions,wereatthehighestrisk.

OddsRatios:CI95%Note:“0-3”Events=referencegroup.*AdjustedfordistrictlocationsDistrictLocationsinclude -AcehBesar (N=180) -EastCoastdistricts (N=246) -CentralHighlands (N=327) -SouthwestHighlands (N=162) -SouthwestCoast (N=461)

formal measures of psychological symptoms and diagnoses for pnA2 (n=1376)

number of current traumatic events

0-3 4-7 8-10 >11

meet criteria for major depressive disorder 1.00 ||3.28(2.26-4.76)

||4.34(2.72-6.91)

||3.45(2.21-5.40)

meet high threshold criteria for major depressive disorder 1.00 ||8.69(3.10-24.36)

||13.49(4.48-40.67)

||10.08(3.36-30.26)

mean depression score “symptomatic” (>1.75) 1.00 ||3.69(2.69-5.06)

||5.54(3.66-8.37)

||3.37(2.28-4.99)

mean depression score “high symptomatic” (>3) 1.00 ‡3.68(1.40-9.73)

§7.01(2.39-20.50)

‡5.91(2.05-17.02)

meet criteria for ptsd 1.00 ‡2.17(1.31-3.58)

||4.29(2.40-7.69)

||3.34(1.90-5.87)

meet high threshold criteria for ptsd 1.00 †5.36(1.23-23.41)

‡10.97(2.32-51.96)

‡12.11(2.66-55.15)

mean ptsd score “symptomatic” (>2.5) 1.00 §2.72(1.52-4.89)

||4.97(2.55-9.68)

||3.75(1.96-7.21)

mean ptsd score “high symptomatic” (>3) 1.00 2.15(0.85-5.43)

‡4.23(1.49-11.97)

§5.34(2.04-14.02)

mean Anxiety score “symptomatic” (>1.75) 1.00 ||2.48(1.87-3.29)

||3.05(2.07-4.49)

||2.23(1.55-3.21)

mean Anxiety score “high symptomatic” (>3) 1.00 §2.99(1.64-5.45)

||4.64(2.33-9.25)

||3.84(1.93-7.64)

tAble 3.8 AdJusted odds rAtios*: mentAl heAlth problems by numbers current stressful eVents, for pnA2 respondents

† p<0.05‡ p<0.01§ p<0.001|| p<0.0001

“A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007�� “A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007 ��

tAble 3.9 AdJusted odds rAtios: mentAl heAlth problems by study And by Gender

formal measures of psychological symptoms and diagnoses for pnA2 (n=1376)

*dataset:pnA1 Vs.

pnA2or (95% ci)

‡Gendermale Vs. female

or (95% ci)

meet criteria for major depressive disorder 4.01(3.26-4.93)

0.77(0.63-0.94)

meet high threshold criteria for major depressive disorder

2.98(2.20-4.02)

0.93(0.69-1.25)

mean depression score “symptomatic” (>1.75)

3.43(2.79-4.20)

0.71(0.59-0.86)

mean depression score “high symptomatic” (>3)

4.51(3.19-6.38) 0.72(0.51-1.00)

meet criteria for ptsd 4.04(3.19-5.11)

0.94(0.75-1.19)

meet high threshold criteria for ptsd 3.73(2.47-5.64)

0.92(0.61-1.38)

mean ptsd score “symptomatic” (>2.5)

4.50(3.51-5.77)

0.78(0.61-1.00)

mean ptsd score “high symptomatic” (>3)

4.76(3.34-6.78)

0.88(0.63-1.24)

mean Anxiety score “symptomatic” (>1.75)

3.47(2.81-4.27)

0.59(0.49-0.71)

mean Anxiety score “high symptomatic” (>3)

4.71(3.66-6.07)

0.78(0.60-1.00)

* dataset: 0=PNA2(reference) p<0.0001forallpsychmeasures

‡ Gender: 0=female (reference), ns for all psych measures, with the following exceptions:

Depression “Symptomatic” (p<0.001),Anxiety“Symptomatic” (p<0.0001),

and borderline significance: Depression (meet criteria for MDD),

PTSD“Symptomatic”,and Anxiety“HighSymptomatic”(p<0.05)

Remarkably,43%ofallmenintheEast Coast region and 41% in theSouthwest Coast, report havingsuffered head trauma. These ratesare equivalent to those found inPNA1 for the North Coast. (SeeTables2.5and2.6above.)Itmightbe noted that women experiencedhigher rates of suffocation andstrangulation than other formsof head trauma, including beingbeaten to the head. In our PNA1report, we further broke thesefindings down by district and byagegroup.

This produced the remarkablefinding that 68% of young men(aged17-29)inBireuenand67%of

youngmeninAcehUtarareportedexperiencing conflict related headtrauma.13

Tables 3.11 and 3.12 showcomparablefiguresforAcehTimurand Aceh Selatan, the highestconflict districts within the EastCoastandSouthwestCoastregions.Again, the findings are strikinglyhighforyoungmen;50%forAcehTimur,38%forAcehSelatan.

However, the distribution in theseareas is somewhat different thanin the PNA1 districts. In AcehTimur, 50% of men aged 30-40,which is equivalent to the rateamong young men, suffered head

trauma. In Aceh Selatan, 57% ofmen aged 30-40 and 52% of menaged 41-53, higher than the ratesfor younger men, suffered headtrauma.

• social functioning The PNA2 protocol added a

simple social functioning scale tothe questionnaire, to attempt tomeasure the actual effects ofpsychological symptoms on socialfunctioning. The questionnaireaskedpersonstoratedifficultiestheyfeel in carrying out routine dailyactivities – washing and dressingthemselves, raising their children,earning money, doing manuallabour or cooking and cleaning

tAble 3.10A AdJusted odds rAtios: mentAl heAlth problems by AGe (pnA1)

formal measures of psychological symptoms and diagnoses:pnA1 (n=589, 7 missing)

Age

17-29(n=145)

25%

30-40(n=185)

31%

41-53(n=139)

24%

54-82(n=120)

20%

or (95% ci)

meet criteria for major depressive disorder 1.00 0.97(0.62-1.51)

1.60(0.99-2.61)

1.02(0.61-1.68)

meet high threshold criteria for major depressive disorder

1.00 †0.46(0.25-0.82)

0.75(0.42-1.35)

1.03(0.57-1.88)

mean depression score “symptomatic” (>1.75)

1.00 0.94(0.59-1.50)

1.45(0.87-2.43)

0.83(0.49-1.39)

mean depression score “high symptomatic” (>3)

1.00 0.60(0.33-1.09)

0.72(0.38-1.36)

1.41(0.76-2.61)

meet criteria for ptsd 1.00 0.82(0.52-1.28)

0.83(0.51-1.34)

0.71(0.42-1.19)

meet high threshold criteria for ptsd 1.00 *0.46(0.22-0.96)

0.58(0.27-1.24)

0.96(0.46-2.00)

mean ptsd score “symptomatic” (>2.5)

1.00 0.66(0.42-1.05)

0.72(0.44-1.18)

0.59(0.35-1.02)

mean ptsd score “high symptomatic” (>3)

1.00 †0.44(0.24-0.81)

0.55(0.29-1.03)

0.88(0.47-1.63)

mean Anxiety score “symptomatic” (>1.75)

1.00 0.95(0.59-1.55)

1.07(0.63-1.80)

0.87(0.51-1.48)

mean Anxiety score “high symptomatic” (>3)

1.00 0.89(0.55-1.43)

1.35(0.82-2.22)

1.09(0.63-1.86)

Note:OddratiosareadjustedforgenderandMGKD.*p<0.05†p<0.01

tAble 3.10b AdJusted odds rAtios: mentAl heAlth problems by AGe (pnA2)

formal measures of psychological symptoms and diagnoses:pnA1 (n=1374, 2 missing)

Age

17-29(n=489)

35%

30-40(n=448)

33%

41-53(n=250)

18%

54-82(n=187)

14%

or (95% ci)

meet criteria for major depressive disorder 1.00 *1.47(1.07-2.03)

†1.69(1.17-2.45)

¶2.59(1.75-3.82)

meet high threshold criteria for major depressive disorder

1.00 1.09(0.64-1.86)

1.02(0.53-1.94)

1.75(0.94-3.25)

mean depression score “symptomatic” (>1.75)

1.00 †1.56(1.19-2.06)

*1.50(1.08-2.08)

‡1.87(1.31-2.68)

mean depression score “high symptomatic” (>3)

1.00 1.63(0.87-3.06)

0.96(0.41-2.26)

1.05(0.41-2.74)

meet criteria for ptsd 1.00 1.32(0.89-1.97)

1.32(0.83-2.11)

1.07(0.62-1.86)

meet high threshold criteria for ptsd 1.00 1.72(0.85-3.50)

0.45(0.13-1.61)

0.83(0.26-2.58)

mean ptsd score “symptomatic” (>2.5)

1.00 1.35(0.86-2.13)

1.49(0.89-2.51)

1.75(1.00-3.07)

mean ptsd score “high symptomatic” (>3)

1.00 1.52(0.79-2.93)

1.00(0.42-2.37)

1.05(0.40-2.76)

mean Anxiety score “symptomatic” (>1.75)

1.00 †1.55(1.18-2.03)

‡1.76(1.28-2.42)

¶2.71(1.90-3.86)

mean Anxiety score “high symptomatic” (>3)

1.00 †1.94(1.17-3.21)

†2.22(1.26-3.89)

¶3.37(1.90-5.96)

Note:Odd ratios are adjusted forgenderandMGKD.

MGKD-answers“yes”or“no”if IOM has a communitypeace dividend project(calledMGKD)inthevillagewheretherespondentlives.

*p<0.05†p<0.01‡p<0.001¶p<0.0001

13Ibid.Seetables9.5and9.6

“A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007�0 “A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007 �1

the house, engaging in farming orfishing,communicatingwithothers,participating in community events,doingtheirprayers.

Overall rates of expression ofdifficultyincarryingoutsuchbasiclife routines were extremely low.Whileusingadisability instrumentwith items that focus primarily onsevere disability (e.g., “difficultywithwashinganddressingoneself”)alwaysworksfarbetterforthemostseverely disabled persons than forthose with more mild disability,this causes serious statisticaldifficulties for analysis. (Thesevariables lack normality indistribution [median=0]; thereforemean tests are unreliable[requiring non parametric tests]).Furthermore, expressed difficultieswere highest for earning money,manuallabour(men),andfarmingor fishing; issues which reflecteconomic conditions as well asdisability.

However, even with this limitedinstrument, analysis demonstratedsignificant relationships betweensum scores for social functioningand depression, anxiety, andPTSD. (See Table 3.13 foranalysis of the depression data).Most striking were relationshipsbetween psychological symptomsand expressed difficulty in basiclivelihoodactivities,suchasearningmoney, manual labour, farmingandfishing.

Although causal directionalitycannot be established, non-significant trends indicate thatthose respondents with depressionfind greater difficulty withlivelihood issues. For example,men with the highest scores ondepression rated their ability toearn money, do manual labour,or engage in farming or fishing

as “difficult” (2), on average,compared with all other personswho on average rated these samelivelihood issues as “somewhatdifficult”(1).Mentalhealthproblemshave real consequences for socialfunctioning, and mental healthand livelihood interventions needtogohandinhand.

• reduced levels of psychological symptoms Acehnese resilience, continued need for services, and Questions for further research

The finding that psychologicalsymptoms were substantially lowerfor PNA2 respondents, even thosein areas that suffered quite highconflict, in comparison with PNA1respondents was unexpected andraises a series of questions. First,thePNA2researchwasdoneacrossAceh in areas with quite variedexperiences of violence. Do thelower levelsof symptoms forPNA2respondents simply reflect lowerrates of trauma among thoserespondents? Second, we knowthat the violence came to someregions (such as Aceh Selatanand more generally the SouthwestCoast) much later than to otherareas (particularly the east coastregions).

Do the differences in levels ofpsychological symptoms reflectexposure to violence for a shorterperiodoftimeforthoseinthePNA2districts?Third,thePNA2researchwasconductedfivemonthsafterthePNA1 research. Is there a reasonwhy this difference in the time ofthe research may have influencedthe reporting of symptoms amongtherespondents toPNA2? Does itreflectresilienceinthispopulationand recovery for many persons?Do the differences in levels ofsecurity,asthepeaceprocessmoved

tAble 3.11 heAd trAumA by Gender And AGe for Aceh timur

head trauma/ potential brain injury

Aceh timur

male(n = 101 - 106)

Age17 - 29

%(n = 41 - 42)

Age30 - 40

%(n = 19 - 20)

Age41 - 53

%(n = 24 – 26)

Age54 +%

(n = 17 - 18)

*Any type of head trauma 50 50 42 44

specific type

beaten on the headsuffocation or strangulationnear drowningother head trauma

4510102

502105

351300

2828011

*Fromthefourdifferenttypesofheadinjury, ifarespondentanswersyestooneormoreofthosefourquestions,thentheanswerisyesfor the new variable (“Any type of head trauma”), which will then tell us how many respondents experienced physical head trauma ofanykindatall.

forwardbetweenFebruaryandJuly,contributetothisfinding?Becausethe research was conducted indifferent districts at somewhatdifferent times, it is impossible todetermine whether time or placeismostcritical. However,ourdatasuggestanswerstothesequestions.

First, one might hypothesize thatthe differences in psychologicalsymptoms among the PNA1 andPNA2 populations simply reflectdifferent rates of violence in thebroad region covered by PNA2in contrast with the PNA1, withsome PNA2 regions experiencingconsiderablylessviolence.

This hypothesis is based on thegeneral assumption that higherlevelsofacutetraumaticeventswillproduce higher levels of enduringpsychological symptoms. Thishypothesiswasdemonstratedbyourresearch.Calculatingadjustedoddsratios,bothPNA1andPNA2foundthat persons with higher levels ofpast traumatic experiences were atrisk for significantly higher levelsof depression, anxiety, and PTSD,although these ratios were muchhigher forPNA1. When lookedatby region, the PNA2 districts thatsuffered highest rates of traumaticevents also suffered higher levelsof symptoms and diagnoses,when compared with regions thatexperienced the lowest levels ofviolence. This led us to predictthatthedistrictsinthePNA2studywithlevelsofviolenceequivalenttothoseinthePNA1studywouldshowequivalent levels of symptoms aswell.

Butourfindingsdidnotsupportthisexpectation.ThePNA2EastCoastand Southwest Coast respondentsexperienced violent and traumaticevents at levels equivalent to (orforsomeevents,higherthan)those

tAble 3.12 heAd trAumA by Gender And AGe for Aceh selAtAn

head trauma/ potential brain injury

Aceh selatan

male(n=100-112)

Age17 - 29

%(n = 29)

Age30 - 40

%(n = 30)

Age41 - 53

%(n = 21)

Age54 +%

(n = 14)

*Any type of head trauma 38 57 52 21

specific type

beaten on the headsuffocation or strangulationnear drowningother head trauma

3424317

53272723

4810014

1421147

*Fromthefourdifferenttypesofheadinjury, ifarespondentanswersyestooneormoreofthosefourquestions,thentheanswerisyesfor the new variable (“Any type of head trauma”), which will then tell us how many respondents experienced physical head trauma ofanykindatall.

ofPNA1NorthCoast respondents.However, their ratesofdepression,anxiety, and PTSD symptoms aremuch lower. Residents of PNA2regions with both the highest andlowest rates of violence reportedsignificantly lower rates of current

psychological distress than didresidents of PNA1 communities.Furthermore, MGKD respondents(those from villages identified ashaving had particularly high levelsof violence) in the PNA1 surveyreported significantly higher levels

of symptoms than non-MGKDrespondents,whilethisrelationshipdoesnotholdforthePNA2survey.Thus, this initial hypothesis is notsupported.

The second question to ask, is

“A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007�2 “A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007 ��

tAble 3.13 sociAl functioninG by depression

†men:how difficult does it feel for you when you

conduct the following daily activities:

pnA2 data% total sample

(n~691)meet criteria for major depressive disordermeet high threshold criteria for major

depressive disordermean depression score “symptomatic” (>1.75)

mean depression score “high symptomatic” (>3)

yesn=139

non=525

yesn=44

non=620

yesn=206

non=458

yesn=20

non=644

33

sum score [range: 0-27] 6.06(4.40) 3.03(3.00) 8.18(4.58) 3.35(3.25) 5.75(4.01) 2.73(2.88) 9.45(6.35) 3.49(3.29) 3.67(3.56)

mean score [range: 0-3] 0.69(0.49) 0.35(0.34) 0.92(0.51) 0.38(0.37) 0.65(0.45) 0.31(0.33) 1.06(0.70) 0.40(0.37) 0.42(0.40)

‡Women:how difficult does it feel for you when you

conduct the following daily activities:

pnA2 data% total sample

(n~685)meet criteria for major depressive disordermeet high threshold criteria for major

depressive disordermean depression score “symptomatic” (>1.75)

mean depression score “high symptomatic” (>3)

yesn=169

non=492

yesn=45

non=616

yesn=264

non=397

yesn=35

non=626

sum score [range: 0-18] 4.40(3.80) 2.56(2.94) 5.69(3.95) 2.84(3.14) 4.44(3.67) 2.10(2.60) 5.71(4.10) 2.88(3.16) 3.03(3.28)

mean score [range: 0-2.29] 0.52(0.45) 0.30(0.34) 0.67(0.45) 0.33(0.37) 0.52(0.43) 0.24(0.30) 0.67(0.47) 0.33(0.37) 0.35(0.38)

*Mean(StandardDeviation).FunctioningScale(0-3):0-NoDifficulty,1-SomeDifficulty,2-Difficult,3-Oftencannotcompletetask.

*pnA2 report: sociAl functioninG by ptsd

†men:how difficult does it feel for you when you

conduct the following daily activities:

pnA2 data% total sample

(n~691)ptsd symptoms “meet criteria” initial Algorithm

ptsd symptoms “meet criteria”revised Algorithm

mean ptsd score (>2.5) “symptomatic” mean ptsd score (>3) “high symptomatic”

yesn=76

non=588

yesn=18

non=646

yesn=55

non=609

yesn=22

non=642

sum score [range: 0-27] 6.25(4.90) 3.33(3.21) 9.67(5.47) 3.50(3.35) 6.51(4.80) 3.41(3.31) 8.91(5.65) 3.49(3.33) 3.67(3.56)

mean score [range: 0-3] 0.71(0.55) 0.38(0.36) 1.08(0.61) 0.40(0.38) 0.75(0.53) 0.39(0.38) 1.00(0.63) 0.40(0.38) 0.42(0.40)

‡Women:how difficult does it feel for you when you

conduct the following daily activities:

pnA2 data% total sample

(n~685)ptsd symptoms “meet criteria”initial Algorithm

ptsd symptoms “meet criteria”revised Algorithm

mean ptsd score (>2.5) “symptomatic” mean ptsd score (>3) “high symptomatic”

yesn=85

non=576

yesn=22

non=639

yesn=77

non=584

yesn=30

non=631

sum score [range: 0-18] 4.88(3.77) 2.76(3.11) 5.91(3.15) 2.93(3.24) 5.29(3.90) 2.74(3.07) 5.77(2.98) 2.90(3.23) 3.03(3.28)

mean score [range: 0-2.29] 0.57(0.43) 0.32(0.36) 0.70(0.38) 0.34(0.38) 0.62(0.46) 0.32(0.35) 0.69(0.35) 0.34(0.38) 0.35(0.38)

*Mean(StandardDeviation).FunctioningScale(0-3):0-NoDifficulty,1-SomeDifficulty,2-Difficult,3-Oftencannotcompletetask.

*pnA2 report: sociAl functioninG by Anxiety

†men:how difficult does it feel for you

when you conduct the following daily activities:

pnA2 data% total sample

(n~691)mean anxiety score (>1.75) mean anxiety score (>3)

yesn=224

non=440

yesn=57

non=607

sum score [range: 0-27] 5.24(4.12) 2.86(2.93) 6.11(4.82) 3.44(3.33) 3.67(3.56)

mean score [range: 0-3] 0.60(0.46) 0.33(0.33) 0.69(0.54) 0.39(0.38) 0.42(0.40)

‡Women:how difficult does it feel for you

when you conduct the following daily activities:

pnA2 data% total sample

(n~685)mean anxiety score (>1.75) mean anxiety score (>3)

yesn=308

non=353

yesn=67

non=594

sum score [range: 0-18] 4.21(3.67) 2.00(2.47) 4.97(3.96) 2.81(3.12) 3.03(3.28)

mean score [range: 0-3] 0.49(0.43) 0.23(0.28) 0.58(0.46) 0.33(0.36) 0.35(0.38)

*Mean(StandardDeviation).FunctioningScale(0-3):0-NoDifficulty,1-SomeDifficulty,2-Difficult,3-Oftencannotcompletetask.

“A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007�4 “A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007 ��

whether the length of time thata particular area was exposed toviolence accounts for levels ofsymptoms. Stated as a hypothesis,this would suggest that a durationeffect - that persons who sufferrepeatedtraumaoverlongerperiodsof time are more likely to suffercontinued psychological symptomsthan persons suffering acutetrauma for a short time - couldaccountforthisfinding.Historically,the North Coast communities ofPNA1 and the East CoastcommunitiesofPNA2experiencedsevere conflict between TNI andGAM forces dating back to theinitial DOM period of the early1990’sandbefore.Bycontrast, theconflict came to the SouthwestCoast and the Central Highlandsmuchlater,usuallyafter2000.

This hypothesis would predictsimilar levels of symptoms forNorth Coast and East Coast,lower levels for Southwest Coastand Central Highlands. Again,this hypothesis is not supported.

Levels of psychological symptomsreported in the East Coast regionare not only much lower thanin the North Coast, but they arelower than in the PNA2 SouthwestCoastandCentralHighlands.

While this second hypothesis isnot supported, it should be notedthatwedonothavedataaboutthelength of exposure to violence byindividual respondents, only aboutthe history of violence in regions,and none of the high conflictvillages experienced truly shortexposure to violence (such as asingleevent).Nearlyallexperiencedat least 3-5 years of violence priortothepeaceagreement.Itremainsquite possible that many personsmay have experienced conflictrelatedtraumaticeventsrepeatedlyoverlongperiodsoftime,andthathis has led to the persistence ofhigh levels of symptoms andincreased risk for diagnosablemental illness. It is possible thatthe duration effect is important atthe individual and clinical level.

However, at the aggregate level,thedurationofconflictinaregiondoes not explain reduced levelsofsymptomsinthePNA2regions.

The third question to ask, iswhetherbeingat agreater removein time from the violence, that isby five months, combined withincreased security by July 2006comparedtoFebruary,areenoughto lead to a lowering of symptomrates.Thissuggeststwohypotheses,a resilience hypothesis and asecurityhypothesis,which togetherargue that the primary differencebetween PNA1 and PNA2 levels ofpsychological symptoms is due tothe passage of time and increasedlevelsofsecurity.

It suggests that while a verysignificant number of personscontinue to suffer symptoms ofdepression and PTSD at a levelmeeting criteria for psychiatricdiagnoses, and some personscontinue to sufferespecially severeconditions, the extraordinarily

high level of symptoms of anxietyand depression found amongPNA1 respondents declined overthisperiodoftime.

Although the PNA data are notlongitudinalandwecannotdirectlytest these hypotheses, the data areconsistent with both the resilienceand security explanations. PNA2was conducted at a time whenreported rates of current violence(experiencing assault and robbery,seeing former perpetrators) weremuch, much lower for all PNA2regions than they had been forthe PNA1 districts at the time ofthat study. (See Table 2.9.) It wasconducted when the inorganicIndonesian military troops hadbeen gone long enough to giveAcehnesecommunitiesasensethatthe peace process might actuallysucceed.

BythetimeofthePNA2interviews,our data show that psychologicalsymptomswerelowerinallregions

interviewed, both those whichexperienced trauma at the samelevel as PNA1 districts and thosethatexperiencedmuchlowerlevelsoftrauma.

While odds analysis shows thatindividual levels of experience oftraumatic events increased therisk for psychological symptomsinbothPNA1andPNA2,theseoddsratios are much lower in PNA2.

Allof thesesupport thehypothesisthat as security increased fromFebruary to July of 2006, a broadprocessofcommunityrecoverywasunderway.Itsupportsourbroaderobservations that the people ofAceh are remarkably resilient. Italso suggests that the symptomchecklists used in this study notonly measure psychopathology,but that they are also sensitivebarometersofcollectiveanxiety.

The PNA data overall suggest thatpsychological symptoms reflect

both severe traumatic violenceexperiencedinthepastandcurrentlevels of security and collectiveanxiety. Our research suggests onthe one hand that carrying thepeaceprocessforward,maintainingand enhancing security, is utterlycritical to the mental health ofpersons in high conflict regionsofAceh. It alsoprovides evidencethat while this is an extremelyresilient population, trauma-related symptoms and treatablepsychiatric conditions remain verycommoninthesecommunities.

Continued interventions areneeded to address the acute andpersistent psychiatric conditionsassociated with the conflict, evenas the peace process movesforward. And as programmes aredevelopedtorespondtotheeffectsof violence, research is needed tounderstand natural processes ofresilience and recovery in Acehand to assure that interventionswillsupporttheseprocesses.

† t-test AnAlyses sociAl functioninG sum And meAn scores by psych. meAsures: mAles

General emotional distress experienced by informants (pnA2)sum score

(0-27)mean score

(0-3)

depression symptoms “meet criteria” initial Algorithm p<0.0001 p<0.0001

depression symptoms “meet criteria” revised Algorithm p<0.0001 p<0.0001

mean depression score (>1.75) “symptomatic” p<0.0001 p<0.0001

mean depression score (>3) “symptomatic” p<0.001 p<0.001

ptsd symptoms “meet criteria” initial Algorithm p<0.0001 p<0.0001

ptsd symptoms “meet criteria” revised Algorithm p<0.001 p<0.001

mean ptsd score (>2.5) “symptomatic” p<0.0001 p<0.0001

mean ptsd score (>3) “symptomatic” p<0.001 p<0.001

mean anxiety score (>1.75) p<0.0001 p<0.0001

mean anxiety score (>3) p<0.0001 p<0.0001

† t-test AnAlyses sociAl functioninG sum And meAn scores by psych. meAsures: femAles

General emotional distress experienced by informants (pnA2)sum score

(0-18)mean score

(0-2.29)

depression symptoms “meet criteria” initial Algorithm p<0.0001 p<0.0001

depression symptoms “meet criteria” revised Algorithm p<0.0001 p<0.0001

mean depression score (>1.75) “symptomatic” p<0.0001 p<0.0001

mean depression score (>3) “symptomatic” p<0.001 p<0.001

ptsd symptoms “meet criteria” initial Algorithm p<0.0001 p<0.0001

ptsd symptoms “meet criteria” revised Algorithm p<0.0001 p<0.0001

mean ptsd score (>2.5) “symptomatic” p<0.0001 p<0.0001

mean ptsd score (>3) “symptomatic” p<0.001 p<0.001

mean anxiety score (>1.75) p<0.0001 p<0.0001

mean anxiety score (>3) p<0.0001 p<0.0001

“A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007�� “A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007 �7

community mentAl And psychosociAl heAlth

AVAilAble resources in communitiesA series of questions were askedaboutwhat,orwho,peopleturntointimes of stress. Interviewers asked allrespondents; “in thepast sixmonths,have you done any of the followingthings to overcome bad experiencesrelated to the conflict?” The list ofpossible responses is shown in theleftcolumnofTable4.1.Respondentswere free tochooseasmanyof theseitemsastheywanted;eachrowinthetablerepresentsthepercentofpeople

who said “yes” to that category, butnot at the expense of others. Thepercentagesineachcolumnthereforedonotsumto100%.

The remarkable difference betweenPNA1 and PNA2 in the responsesto each item on this checklist ofactivities is due to a change inthe format of the questionnaire.In PNA1, the response rate is lowbecause each choice was notread out loud to the respondent.Knowing that much more than

three percent of residents livingin rural areas of Aceh make use oftraditional healers, the researchersspeculated that respondents werereluctant to offer that up on theirown to interviewers, many of whomwere medical nurses from thepsychiatric hospital in Banda Acehand the rest of whom were educatedcityresidentsaswell.

In PNA2, interviewers went througheach item on the checklist, askingout loud whether or not in the past

tAble 4.1 help seeKinG behAViour durinG the pAst six months

in the past 6 months, have you done any of the following things to overcome bad

experiences related to the conflict?

*pnA1 data †pnA2 datapnA1 + pnA2

data

% north coast

(n=596)

%Aceh besar

(n=180)

%east coast(n=246)

%central highlands

(n=327)

%southeast highlands

(n=162)

%southwest coast

(n=461)

% pnA2 sample

(n=1,376)

%total sample

(n=1,972)

talk about it with friend or family 35 65 87 71 62 63 68 58

Visit a traditional healer or take traditional medicine

3 6 10 24 26 17 17 13

look for medical help (Puskesmas, hospital, village midwife)

15 23 48 37 29 31 33 28

consult a mental health specialist (psychologist/psychiatrist/ community mental health nurse)

1 2 1 4 1 4 3 2

consult a religious specialist (imam ustad, ulama)

17 34 67 46 53 65 54 42

prayer 71 90 98 92 74 93 91 85

sport / exercise 2 9 26 32 31 23 24 18

try to forget about the experience 16 24 85 70 39 54 56 44

move somewhere else 3 2 21 19 6 9 12 9

do nothing 6 0 11 6 4 7 6 6

other 2 0 2 7 0 <1 2 2

don’t know / refuse / no opinion 0 0 1 <1 0 1 1 <1

sixmonths therespondentmadeuseof a healer, a religious leader, andso on. Instead of thinking of theirways of coping with stress on theirown and then matching to a list ofanswers, respondents had a chanceto affirm or deny their use of eachitem shown on the table above, andthis probably accounts for the jumpin affirmative answers for each itemonthelist.

Nevertheless the pattern establishedby the answers of the PNA1 sample

remainsmoreorlessthesame.Nearlyall respondents in the PNA2 sample(91%) make use of prayer in timesof stress, followed by talking withfriendsandfamily(68%)asadistantsecond. Consulting a religiousspecialist (54%) and trying to forgetwhat happened (56%) are roughlytied for third place ranking. Medicalcare (33%), sport and exercise(24%) and traditional healing care(17%) are all noteworthy sources ofsupport to overcome bad memoriesof conflict experiences. Rates for

many of these activities were highestin the East Coast region, whereviolence was particularly intense.These are, however, only smallindicators of the local resourcesand psychological processes used inrecovery and efforts to deal withthe memories of violence. Largerscale political processes are almostcertainly equally important in thelong termefforts for community andpersonalrecovery.

NotshowninTable4.1arethegenderdifferences. In the PNA2 sample,womenandmensoughttoovercomeconflictrelatedexperiencesinsimilarways, with only sports and exerciseengaged in more frequently by men(34%)thanbywomen(16%).

• perceptions of conflict-related mental disability in the community

Respondents were asked a broadseries of questions about mentalhealth problems in their owncommunities. This section of thequestionnaire began by askingwhether respondents felt thatthere are mental health problemsin their community related to theconflict and/ or the tsunami, andif thoseproblemsareaffecting therespondents or the respondents’families. The results are presentedinTable4.2.

There is a significant differencein respondent perceptions of thepresence of conflict or tsunami-related mental health problemsintheircommunity,betweenPNA1

“A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007�� “A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007 ��

andPNA2,downfrom66%inPNA1to 39% in PNA2. However giventhe significant difference betweenregions in PNA2, it might not beappropriate to compare PNA2 asa whole to PNA1, given that theycover different regions as well.Rather, it makes more sense, for

example, to compare North Coast(66%) with East Coast (50%), inwhich the numbers are closer toeachotherandagreaterpercentageofEastCoastrespondentswhosaidyes to the first question felt thatthese problems affect themselvesortheirfamilies(64%ofPNA2East

Coastvs.55%ofPNA1NorthCoastrespondents).

Overall, nearly half of all PNArespondents felt that there areconflictandtsunami-relatedmentalillnesses in their community, halfof whom feel that these problems

the PNA2 questionnaire two morechoices were added: communityleaders, and Indonesian securityforces.TheresultsarepresentedinTable4.3.

As with the results shown inTable 4.1, the response rate per

affectthemselvesorpeopleintheirfamily.

Respondents were then asked totell us which groups in theircommunitysufferedthemostfrom“stress or trauma related to theconflict.” Respondents were free

to choose as many groups as theywanted, without rank, from thefollowing groups: women, men,children, youth, former politicalprisoners, former GAM-TNAcombatants, the elderly, andconflict widows and widowers.Based on feedback from PNA1, in

tAble 4.2 respondent perceptions of mentAl illness in the community And At home

* PNA1VS.PNA2districtlocations: - Doyouthinkthereareanymentalhealthproblems

in your community related to the tsunami and/or theconflict?[p<0.0001]

- Do you feel that these problems have affected you andyourfamily?[p<0.0001]

†PNA2districtlocationscomparison: - Doyouthinkthereareanymentalhealthproblems

in your community related to the tsunami and/or theconflict?[p<0.0001]

- Do you feel that these problems have affected you andyourfamily?[p<0.0001]

*pnA1 data †pnA2 datapnA1 + pnA2

data

% north coast

(n=596)

%Aceh besar

(n=180)

%east coast(n=246)

%central highlands

(n=327)

%southeast highlands

(n=162)

%southwest coast

(n=461)

% pnA2 sample

(n=1,376)

%total sample

(n=1,972)

do you think there are any mental health problems in your community related to the tsunami and/ or the conflict? (%yes)

66 16 50 42 17 46 39 47

Those responding “yes”: (n=393) (n=29) (n=122) (n=137) (n=27) (n=214) (n=529) (n=922)

do you feel that these problems have affected you and your family? (%yes)

55 31 64 31 30 43 43 48

Note: PNA1“CommunitySuffering”questions=binary(Yes=1/No=0) PNA2“CommunitySuffering”questions=“Didnotsufferatall”=0,“Sufferedalittle”=1,“Suffered”=2,“Sufferedalot”=3 Binary(Yes=1,ifresponse=1,2or3,No=0,ifresponse=0)

tAble 4.3 respondent selection of Groups in their community sufferinG most from conflict- relAted stress or trAumA

Which of the following groups in your community suffer the

most because of stress or trauma related to the conflict? (%yes)

*pnA1 data †pnA2 datapnA1 + pnA2

data

% north coast

(n=596)

%Aceh besar

(n=180)

%east coast(n=246)

%central highlands

(n=327)

%southeast highlands

(n=162)

%southwest coast

(n=461)

% pnA2 sample

(n=1,376)

%total sample

(n=1,972)

Women 65 77 95 87 70 94 87 80

men 81 89 97 92 74 96 92 88

children 32 60 78 74 50 76 71 59

youth 71 88 96 91 69 96 90 84

former political prisoners 19 63 81 48 63 74 68 51

former GAm-tnA combatants 26 69 86 53 68 80 73 57

elderly 38 73 89 77 65 90 82 68

conflict widows/ widowers 32 68 90 81 69 86 82 66

community leaders - 72 93 72 52 88 79 79

security forces (ri) - 47 65 41 48 22 40 40

* PNA1VS.PNA2districtlocations: Women p<0.0001 Men p<0.0001 Children p<0.0001 Youth p<0.0001 Formerpoliticalprisoners p<0.0001 FormerGAM-TNAcombatants p<0.0001 Elderly p<0.0001 Conflictwidows/widowers p<0.0001 CommunityLeaders p<0.0001 SecurityForces(RI) p<0.0001

† PNA2districtlocationscomparison: Women p<0.0001 Men p<0.0001 Children p<0.0001 Youth p<0.0001 Formerpoliticalprisoners p<0.0001 FormerGAM-TNAcombatants p<0.0001 Elderly p<0.0001 Conflictwidows/widowers p<0.0001 CommunityLeaders p<0.0001 SecurityForces(RI) p<0.0001

“A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 200770 “A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007 71

tAble 4.4 opinions About nGo mentAl heAlth serVices And implementinG pArtners

* PNA1VS.PNA2districtlocations: Statistically significantly different at p<0.0001

[AllActivities]

Except…..

“Consultamentalhealthspecialist”(p<0.05) “Don’tknow/refuse/noopinion”(NS)

†PNA2districtlocationscomparison: Statisticallysignificantlydifferentatp<0.0001

[AllActivities] Except….. “Consultamentalhealthspecialist”(p<0.05) “Don’tknow/refuse/noopinion”(NS)

*pnA1 data †pnA2 datapnA1 + pnA2

data

% north coast

(n=596)

%Aceh besar

(n=180)

%east

coast(n=246)

%central

highlands(n=327)

%southeast highlands(n=162)

%southwest

coast(n=461)

% pnA2

sample(n=1,376)

%total

sample(n=1,972)

if an outside nGo offered you or your family member mental health assistance, administered through GAm, would you accept it?

Yes-60No-11DK/refuse-29

Yes-67No-10DK/refuse-23

Yes-40No-22DK/refuse-38

Yes-40No-35DK/refuse-25

Yes-67No-27DK/refuse-6

Yes-49No-23DK/refuse-28

Yes-50No-24DK/refuse-26

Yes-53No-20DK/refuse-27

if an outside nGo offered you or your family member mental health assistance, that was administered by the indonesian government, would you accept it?

Yes-51No-23DK/refuse-26

Yes-69No-10DK/refuse-21

Yes-37No-26DK/refuse-37

Yes-52No-24DK/refuse-24

Yes-69No-23DK/refuse-8

Yes-50No-25DK/refuse-25

Yes-53No-22DK/refuse-25

Yes-53No-22DK/refuse-25

tAble 4.5 opinions About nGo mentAl heAlth serVices And implementinG pArtners, by Gender

* PNA1VS.PNA2districtlocations: Statisticallysignificantlydifferentatp<0.0001[AllActivities] Except…..

“Consultamentalhealthspecialist”(p<0.05) “Don’tknow/refuse/noopinion”(NS)†PNA2districtlocationscomparison: Statisticallysignificantlydifferentatp<0.0001[AllActivities]

Except…..

“Consultamentalhealthspecialist”(p<0.05) “Don’tknow/refuse/noopinion”(NS)

*pnA1 data †pnA2 data pnA1 + pnA2

% male

(n=315)

% female(n=281)

% total pnA1 sample

(n=596)

% male

(n=691)

% female(n=685)

% total pnA2 sample

(n=1,376)

%total sample

(n=1,972)

if an outside nGo offered you or your family member mental health assistance, administered through GAm, would you accept it?

Yes-68No-6DK/refuse-26

Yes-51No-16DK/refuse-33

Yes-60No-11DK/refuse-29

Yes-60No-19DK/refuse-21

Yes-40No-30DK/refuse-30

Yes-50No-24DK/refuse-26

Yes-53No-20DK/refuse-27

if an outside nGo offered you or your family member mental health assistance, that was administered by the indonesian government, would you accept it?

Yes-51No-24DK/refuse-25

Yes-52No-21DK/refuse-27

Yes-51No-23DK/refuse-26

Yes-59No-20DK/refuse-21

Yes-47No-25DK/refuse-28

Yes-53No-22DK/refuse-25

Yes-53No-22DK/refuse-25

Additional assistance by genderpnA1 males

vs.pnA2 males

pnA1 females vs.pnA2 females

nGo offered assistance, administered by GAm p<0.0001 P<0.0001

nGo offered assistance, administered by indonesian gov. NS NS

* pnA1 Vs. pnA2 Gender compArisons

Additional assistance by genderpnA1 within

gender comparison:

pnA2 within gender

comparison:

nGo offered assistance, administered by GAm p<0.0001 P<0.0001

nGo offered assistance, administered by indonesian gov. NS P<0.0001

† pnA1 And pnA2 Within Gender compArisons (mAles Vs. femAles)

“A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 200772 “A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007 7�

The final finding described in this study is drawn notfrom the PNA2 survey but from the Norwegian Embassyfundedpilotmentalhealthoutreachprogrammecurrentlybeingundertaken inBireuen. IOM’s threemedical teams,each with a GP physician and a nurse, both with specialmentalhealthtraining,havescreenedpatientsandarenowproviding treatment to over 580 persons with diagnosablemental health problems in 25 villages, in high conflictsub-districts.

Clinicians are finding that the problems identified in thePNA surveys are extremely prominent among personsdiagnosed as suffering mental health problems. PTSDsymptoms are ubiquitous, with nearly a quarter of allpatientsmeetingcriteria forPTSDand42%ofallpersonstreated for diagnosable mental health problems sayingthat their illnesses are related to conflict-related traumaexperience. Head trauma appears as an important

conflict-relAted clinicAl illnesses in these communities: findinGs from A mentAl heAlth outreAch proJect

item increases significantly inPNA2 because of a format changein the questionnaire. In PNA2,interviewers listed each item andasked respondents whether theyfelt men, women, youth, and theother community groups listedwere heavily affected by conflictrelated trauma. In PNA1,interviewers solicited answerswithoutreadingthechoices.

Inspiteofthechangeinformatofthe question, and in spite of theleap in response rate, the generalpattern established by the resultsof PNA1 still holds in PNA2.Respondents continue to citemen (92%) and youth (90%)as suffering the most due toconflict related trauma and stress.However, respondents often saidthat everyone in the communitysufferedgreatly,andsoeverychoice(except for government securityforces) ranks higher than 65%,and most are above 70%.

Comparedwiththegenderandagedistributions of traumatic events,respondents correctly identifymenand youth as most vulnerable,butitisalsoclearthatrespondentsfelt that no one was exempt fromsuffering stress or trauma dueto conflict experience. Regionaldifferences accurately reflectdifferent intensities of conflictviolence across Aceh, but thegeneral ranking pattern holdsacrossallregionsoftheprovince.

perceptions of nGos And public heAlth serVicesIn much of the qualitative data,respondents tendtoask foroutreachand express interest in having non-governmental organizational (NGO)support for developing community-basedmentalhealthservices.

The questionnaire included twoquestions about interest in NGOservices, designed also to measurepreference in local implementingpartners, i.e. the Indonesiangovernment or GAM, which nowoperates in Aceh as a civil societyorganizationandpoliticalparty.

These questions were asked beforethe elections in December 2006,when many GAM-backed candidateswere voted into office, including theGovernor and many district-levelRegents. It would not make as muchsense to ask this question today,given that GAM has stepped intogovernment power at the provincialand district levels. In any case, thequestion is hypothetical in an effortto also measure levels of trust in thegovernmentorGAM.Theresultsaresummarized by region in Table 4.4andbygenderinTable4.5

In the PNA1 we reported that only35% of respondents in Bireuenand 36% in Aceh Utara said theywould be willing to accept mentalhealth assistance if it were offeredby the Indonesian government.

We interpreted this to mean thatthere was significant mistrust ofthe government and governmentservices by communities in thesetwo districts, and that this was animportant potential barrier toproviding mental health servicesthrough the public mental healthsystem.

Findings in PNA2 are similar. In thehighestconflictregion,theEastCoastdistricts, only 37% said they wouldaccept such services. This providessome quantitative support for theimpression gained from qualitativeinterviews that significant mistrustremained at the time of theseinterviews for services offered by thepublicorgovernmentsector.

Comparing responses from thePNA1 and PNA2 studies, fewerwomen (51% and 40%, respectively)say they would accept mental healthassistance offered by an NGOadministered through GAM thando men (68% and 60%). Fewerwomen (47%) in the PNA2 samplealso state they would accept mentalhealth assistance if administered bythe Indonesian government thanmen (59%), whereas half of menand women for PNA1 indicated theywould accept an offer. A quarterof the sample, more women thanmen,refusedtoanswerbothofthesequestions, and 20% responded noto GAM administered services and22% to government administeredservices.

clinical phenomenon, with patients reporting continuingsymptomsdatingbacktobeingbeatenortortured.Clinicaldepressioniscommon.

In these high conflict settings, depression is oftenassociated with traumatic memories of the conflict, withsocial isolation and community divisiveness, a the legacyof the violence, and with the losses of property andeconomicdevastation.

Whatever evidence there is in the PNA2 report ofreduced levels of symptoms in comparison withthe PNA1 study, mental health problems directlyassociated with the conflict are very real in thesecommunities, and a continued urgency should be felt toprovidecare for thesepopulations.The IOMpilotmentalhealthprojecthasdemonstratedsuccessfullyoneapproachwhichiseffectiveinprovidingcareinthesecommunities.

“A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 200774 “A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 2007 7�

1. All programmes undertaken in rural Aceh should take account of the ubiquity and complexity of violence and its psychological and social remainders in the affected communities.

The legacy of accumulated traumatic events allacross Aceh as shown by the PNA data poses uniquechallenges even for programmes such as housingand school reconstruction that are not specificallydesigned for psychosocial assistance. Consultativeprocesses and key informant interviews, early andoften, and at the most local level, should inform thedevelopment of any intervention. From these, anunderstandingofthehistoricalexperienceandcurrentsocial dynamics in sites of post conflict assistance willhelp refine priorities for the programme and ensuresmoothimplementation.

2. the international community should recognize the continued urgency to provide mental health services to the communities most affected by the conflict.

Thisreportdocumentsremarkable levelsof traumaticviolenceenactedagainstordinary civilianpopulationsin rural Aceh, particularly in the highest conflictdistricts, sub-districts, and villages.The report, aswellas findings from the pilot mental health interventionundertakenby IOM, show that this violence is closelyassociated with high levels of depression, anxiety,Post Traumatic Stress Disorder (PTSD), andneuropsychiatric conditions in these communities.Theseproblemshavenotgoneaway.

Nearly two years after the signing of the peace deal,whichendedthemilitaryviolenceinAceh,acutementalhealthproblemsremainacriticallegacyoftheviolence.There is urgent need to provide medically-basedmental health responses, as well as psychosocialand livelihood programmes, for victims in thesecommunities.

recommendAtions

3. provision of mental health services will require sustained investment in the long-term development of the health and mental health system of Aceh.

Aceh has over 4 million people, but just threepsychiatrists. Building a mental health system thatwill reach the widely dispersed communities of Acehshould be recognized as an immediate and urgentneed and as a domain requiring sustained, longterm investment. The mental health needs in thesecommunitiescanonlybemetthroughthedevelopmentof a competent and effective health system that givesspecial priority to mental health care. International,national, and provincial agencies should collaboratein strengthening the capacity of the public healthsystem in general, and specifically in developinginnovative solutions to the difficult task of providingcommunity-basedmentalhealthservices.

4. specialized outreach services should be supported to meet the most urgent mental health needs in high conflict areas of Aceh.

While the long term needs for mental health carein Aceh can only be addressed through investmentin improving the public mental health system,persons suffering the mental health consequences ofviolence, torture, and displacement should not bemade to wait. Specialized programmes that providemental health and psychosocial services to thevictims of the conflict should be given immediatesupport.

The Norwegian Embassy funded Mental HealthOutreach programme developed jointly by IOM andHarvard Medical School has been shown to be oneeffective mechanism for addressing acute and urgentneeds in relatively isolated communities. Seriousinvestmentshouldbemadeinprogrammesthatbringservicesdirectlytothesecommunities.

5. focused efforts to treat persons suffering the effects of complex trauma should be undertaken in the context of the development of specialized mental health and psychosocial programmes.

In communities in which 15-18% of the totalpopulationand25%ofallmenreportbeingtortured,inwhich50-70%ofyoungmenreportbeingbeatenonthe head, suffocated, or submitted to near drowning,in which 50-65% of all men and 15-20% of womenreportbeingbeaten,complextraumaisacommonandimportantremainderoftheviolence.

Managing mental health and psychosocial problemsassociatedwithPTSDandcomplextraumainrelativelyisolated settings with limited access to mental healthcare is extremely challenging. It should be explicitlyrecognizedthatthereisnosingletherapeuticmodalitywhich is certain to be effective and sustainable.Instead,acommitmentshouldbemadetodevelopinginnovative therapeutic programmes in selectedsettings, to documenting each programme, andto careful evaluation of the efficacy of therapeuticapproaches.

6. special attention needs to be given to the problem of head trauma, brain injury, and long-lasting disability resulting from torture and violence associated with the conflict.

BoththisreportandthePNA1documentedremarkablyhigh levels of head trauma – beatings to the head,strangulation and suffocation, and near drowning –that were a routine part of torture, particularly ofmen, in high conflict communities. Head traumacan cause brain injury and anoxia (lack of oxygento the brain) that can cause long lasting emotional,cognitive, andbehavioural effects.Thesemay includereduced ability to concentrate and participate inlivelihood training, impaired judgment leading towhat may appear to be routine acting out or evencriminal behaviour, as well as personal suffering.

Researchshouldbeundertakentodeterminewhetherspecialized programmes are needed to respond tothese problems. The medical, legal, and educationalsystems should be made aware of the importance ofthese issues for persons who have suffered traumaticviolenceintheconflict.

7. special attention needs to be given to the mental health problems of older persons in the high conflict areas.

While young people were submitted to particularviolence during the conflict and rightly deservespecializedattention,thisreportsuggeststhatoldermenand women may continue to experience the highestratesofmentalhealthproblemsinthesecommunities.Littleattentionhasbeendirectedtotheeffectsoftheconflictontheelderly.Thisfindingsuggeststheneedfor further research and the development ofprogrammes to address the mental health andpsychosocialneedsofoldermenandwomen in thesecommunities.

8. those districts and villages that suffered particularly egregious violence should be provided special attention in the development of mental health and psychosocial services.

Exposure to traumatic events during the conflict isthe single largest predictor of current mental healthdisability in both PNA1 and PNA2. A mapping ofconflicteventsacrossAceh showswhere theprioritiesareformentalhealthandpsychosocialservices.TheseincludetheNorthandEastCoastdistrictsofBireuen,AcehUtara,andAcehTimur,aswellastheSouthwestcoast district of Aceh Selatan. Additionally, themapping of conflict events can be taken down to thesub-district and village level, revealing the micro-localities where service providers are most likely tofind the highest conflict-related mental healthburden.

“A Psychosocial Needs Assessment of Communities in 14 Conflict-Affected Districts in Aceh”/ June 20077�

9. national and international agencies should recognize the continued need for livelihood interventions in high conflict areas, which should be linked specifically with mental health and psychosocial programmes.

Damaged or lost livelihoods are more than just anunfortunateby-productof the conflict. Inmost cases,there was a deliberate and systematic attempt todestroy local economies that were seen by militaryforces as a strategic material base for continuedrebellion. This created devastating losses for thecivilians in these communities and their recovery isinvariably identified by respondents as a first priority.But the worst affected communities in need oflivelihood recovery are also the communities withthe highest mental health burden, which may hinderthe success of programmes designed for materialrecovery.

Transitional assistance to rehabilitate destroyedfields and forest gardens, capital inputs torestart local business, livelihood training and thedevelopment of small trade cooperatives, can allbe seen as psychosocial interventions on their own,but those with the most disability will need explicitmentalhealthassistancetoaccompanytheirlivelihoodsupport.

10. the development of programmes for rural Aceh should include a systematic awareness of the long term effects of displacement in the high conflict communities.

Nearly half the sample reported displacement dueto conflict. In many villages, this figure is between90% and 100%. The people living in these highconflictareasmustberecognizedas IDPswithall thevulnerabilities and needs that accompany their

recent displacement experience. Displacementrecovery programmes, in particular thereconstruction of damaged and destroyed houses,schools, roads and other infrastructure, and therecovery of lost livelihoods, are a prerequisite forany kind of broad psychosocial recovery in thesecommunities.

11. there is an enormous and lasting reservoir of memories of torture, violence, and displacement enacted against communities and individuals in Aceh. profound loss and a potent sense of injustice are remainders of the violence. careful consideration should be given to specific efforts to work through these memories as a part of the on going peace process in the context of rebuilding Aceh.

These efforts will in turn have consequences for thelarger goal of trauma healing for individuals andcommunities. This report documents remarkablelevelsofviolenceenactedagainstciviliancommunitiesin Aceh. As a psychosocial needs assessment, thereport focuses on specific clinical and mental healthproblemsassociatedwiththisviolence.

However, the ubiquity of violence documented inthis report has broader social and politicalimplications which are critical to the larger goalof trauma healing for the people of Aceh.Special consideration should be given to findingmechanisms for commemoration, for workingthrough painful and contested memories, fordealing with loss, and for reconciliation. Theseefforts have the potential to contribute to traumahealing and ultimately to addressing the painfulremnants of violence in the communities whocontributedtothisreport.

SuryaBuilding13thFloorJl.MH.ThamrinKav.9Jakarta10350,Indonesia

desi

gnb

y:s

.fath

aran

i@gm

ail.c

om/

+62

811

161

184

Ph. +622139838529Fx. +622139838528

infoindonesia@iom.intwww.iom.or.id

INTERNATIONALORGANIZATIONFORMIGRATION(IOM)