0.004avav234.com

ORIGINALPAPERTheinterplayofpsychosisandvictimisationacrossthelifecourse:aprospectivestudyinthegeneralpopulationStevenHonings1MarjanDrukker1MargreettenHave2RondeGraaf2SaskiavanDorsselaer2JimvanOs1,3,4Received:29March2017/Accepted:7August2017/Publishedonline:31August2017TheAuthor(s)2017.
ThisarticleisanopenaccesspublicationAbstractPurposePsychosishasbeenassociatedwithadultvictim-isation.
However,itremainsunclearwhetherpsychosispredictsincidentadultvictimisation,orwhetheradultvictimisationpredictsincidentpsychosis.
Furthermore,amoderatingeffectofchildhoodvictimisationontheasso-ciationbetweenpsychosisandadultvictimisationhasnotbeeninvestigated.
MethodsThelongitudinalassociationbetweenbaselinepsychoticexperiencesandsix-yearincidenceofadultvic-timisationwasassessedinaprospectivegeneralpopulationcohortof6646adultsusinglogisticregressionanalysis.
Theassociationbetweenbaselineadultvictimisationandsix-yearincidenceofpsychoticexperienceswasexaminedaswell.
Furthermore,themoderatingeffectofchildhoodvictimisationonthesebidirectionalassociationswasanalysed.
ResultsPsychoticexperiencesandchildhoodvictimisationwerebothassociatedwithanincreasedriskofincidentadultvictimisation.
However,thiswasthroughcompetingpathways,assuggestedbyanegativeinteractionbetweenpsychoticexperiencesandchildhoodvictimisation.
Base-lineadultvictimisationandchildhoodvictimisationbothindependentlyincreasedtheriskofincidentpsychoticexperiences,buttherewasnointeractionbetweenadultvictimisationandchildhoodvictimisation.
ConclusionsPsychosisandvictimisationareintercon-nectedthroughoutthelifecourse.
Childhoodvictimisationisconnectedtopsychosisthroughtwopathways:onedirectandoneindirectthroughadultvictimisation.
Inindividualswithoutchildhoodvictimisation,psychosisandadultvic-timisationbidirectionallyimpactoneachother.
KeywordsPsychosisPsychoticexperienceViolenceChildhoodtraumaVictimisationIntroductionPsychosishasbeenassociatedwithanincreasedriskofviolenceperpetration[1–5].
However,contrarytothecommonstereotypethatindividualswithseverementalillnessaredangerous[6],evidenceshowsthattheseindi-vidualsaremorelikelytobevictimsofviolencethanperpetratorsofviolence[7,8].
Amongindividualswithpsychosis,victimisationisprevalent[9],bothduringchildhood[10–12]andadulthood[13–15].
Variousformsofchildhoodvictimisation,includingsexualabuse,physicalabuse,emotionalabuse[16]andbeingbullied[17],havebeenassociatedwithpsychosisintheliterature[10–12].
Childhoodvictimisationhasbeenassociatedwithbothpsychoticexperiences(PE)[10]andfull-blownpsychoticdisorder[11,12],thuscoveringthecompletespectrumoftheextendedpsychosisphenotype[18–20].
Mostresearchtodatehasfocussedonthehypothesisthatchildhoodvictimisationisariskfactorfor&JimvanOsvanosj@gmail.
com1DepartmentofPsychiatryandPsychology,SouthLimburgMentalHealthResearchandTeachingNetwork,MaastrichtUniversityMedicalCentre,Maastricht,TheNetherlands2NetherlandsInstituteofMentalHealthandAddiction,Utrecht,TheNetherlands3King'sHealthPartners,DepartmentofPsychosisStudies,InstituteofPsychiatry,King'sCollegeLondon,London,UK4DepartmentofPsychiatry,BrainCenterRudolfMagnusInstitute,UniversityMedicalCenterUtrecht,POBOX85500,3508GAUtrecht,TheNetherlands123SocPsychiatryPsychiatrEpidemiol(2017)52:1363–1374DOI10.
1007/s00127-017-1430-9thedevelopmentofpsychosis[16].
However,evidenceshowsthatPEincreasetheriskofincidentchildhoodvic-timisationaswell,therebyshowingthattheassociationbetweenchildhoodvictimisationandPEisbidirectional[10].
Recently,studiesfoundevidencethatpsychosisisassociatedwithadultvictimisationaswell.
Comparedwithgeneralpopulationindividuals,theprevalenceofcriminalandviolentvictimisationamongindividualswithpsychosisandotherseverementalillnesseswashigh[13–15].
However,thenatureofthisassociationremainsunclear,sincemoststudiestodatehadsomemethodologicallimi-tations.
First,moststudiestodateusedcross-sectionalstudydesignstoexaminetheassociationbetweenpsy-chosisandadultvictimisation[13–15].
Therefore,itisunclearwhetherpsychosisincreasestheriskofincidentadultvictimisationorviceversa.
Toourknowledge,onlyonelongitudinalstudyexaminedtheassociationbetweenadultvictimisationandpsychosis,showingthatadultadversitieswereassociatedwithanincreasedriskofinci-dentPE[21].
However,nolongitudinalstudytodateexaminedwhetherpsychosispredictsincidentadultvic-timisation.
Thus,itremainsunknownwhethertheassoci-ationbetweenpsychosisandadultvictimisationisbidirectional,similartotheassociationbetweenpsychosisandchildhoodvictimisation[10].
Second,fewstudiesexaminedtheinuenceofchildhoodvictimisationontheassociationbetweenpsychosisandadultvictimisation[22],whilechildhoodvictimisationisassociatedwithanincreasedriskofbothpsychosis[10–12]andadultvic-timisation[23–29].
PreviousstudieshaveshownthatchildhoodvictimisationandvariousenvironmentalfactorscombinesynergisticallytoincreasetheriskofPEoverandabovetheirisolatedproducts[21,30–32].
However,nopreviousstudyexaminedwhetherpsychosispredictsinci-dentadultvictimisation,whilesimultaneouslyexaminingthepotentialmoderatingeffectofchildhoodvictimisation.
Thepresentstudyaimstobridgethesegapsinthelit-erature.
Inlinewiththeresearchonchildhoodvictimisa-tionandPE,wehypothesizedthattherewouldbeabidirectionalassociationbetweenPEandadultvictimisa-tion,thatismoderatedbythepresenceofchildhoodvic-timisation(Fig.
1).
Morespecically,wehypothesizedthat:(1)PEareassociatedwithincidentadultvictimisa-tion;(2)childhoodvictimisationisassociatedwithincidentadultvictimisation;(3)theco-occurrenceofPEandchildhoodvictimisationpredictsastrongerassociationwithincidentadultvictimisationthantheproductoftheiriso-latedeffects;(4)adultvictimisationisassociatedwithincidentPE;(5)childhoodvictimisationisassociatedwithincidentPE;(6)theco-occurrenceofadultvictimisationandchildhoodvictimisationpredictsastrongerassociationwithincidentPEthantheproductoftheirisolatedeffects.
MethodsSampleThisstudyusesdatapertainingtothesecondNetherlandsMentalHealthSurveyandIncidenceStudy(NEMESIS-2),alongitudinalstudyoftheprevalence,incidence,courseandconsequencesofmentaldisordersintheDutchgeneralpopulation[33].
Participantswereselectedbasedonamultistagerandomsamplingprocedure.
Atbaseline(T0),6646personsaged18–64yearswereinterviewedwiththeCompositeInternationalDiagnosticInterview(CIDI)ver-sion3.
0,afullystructuredlay-administereddiagnosticinterviewgeneratingDSM-IVdiagnoses[34].
Atfollow-up,respectively,three(T1,N=5503)and6years(T2,N=4618)afterbaseline,subjectswerere-interviewed.
Amorecomprehensivedescriptionofthedesigncanbefoundelsewhere[33].
Inthepresentanalyses,onlyindividualswhorespondedtoallthreeassessmentswereincluded.
PsychoticexperiencesPEwereassessedatbaseline(T0)andbothfollow-upmeasurements(T1,T2)usingapsychosisadd-oninstru-mentbasedonthesectionsofpsychoticsymptomsinCIDIversions1.
0and2.
0.
Theinstrumentconsistedof20questionsregardinglifetimePE.
The20itemsincluded15delusionalexperiencesand5hallucinatoryexperiences,describedindetailelsewhere[35].
IndividualswithatleastonelifetimePEwerecontactedforare-interviewbytelephone.
Re-interviewswereconductedbyanexperi-encedclinicianatthelevelofpsychologistorpsychiatrist,within8weeksaftertheinitialinterviewusingquestionsfromtheStructuredClinicalInterviewforDSM-IV.
Find-ingsfromallre-interviewswerediscussedwithasecondclinician[35].
PEweredenedasclinicallyvalidatedwhenFig.
1Hypothesesrelatingtothebidirectionalassociationbetweenpsychoticexperiencesandvictimisation1364SocPsychiatryPsychiatrEpidemiol(2017)52:1363–1374123thepsychoticnatureoftheself-reportedPEwasconrmedatclinicalre-interview.
Theresponseratesforthere-in-terviewsatthethreetimepointswere73%(T0),84%(T1)and81%(T2),respectively.
Atbaseline,lifetimePEwereassessed.
Atbothfollow-upsurveys,participantswereaskedaboutPEsincethelastinterview.
Forthepresentanalyses,onlyclinicallyvalidatedPEwereused.
Individ-ualswithself-reportedPEwhocouldnotbereachedforre-interviewwereexcludedfromtheanalyses.
PEweredenedpresentiftheparticipanthadatleastoneclinicallyvalidatedPE.
IncidentPEwasdenedpresentifapartic-ipantreportedatleastoneclinicallyvalidatedPEatoneofthefollow-upsurveys,butreportednoPEatbaseline.
ChildhoodvictimisationAtbaseline,childhoodvictimisationwasassessedretro-spectively.
Participantswereaskedwhethertheyhadexperiencedemotionalabuse,psychologicalabuse,physi-calabuseorsexualabusebeforetheageof16years.
Consistentwithpreviousanalyses[7],childhoodabusewasdenedpresentifaparticipanthadexperiencedpsycho-logicalabuseoremotionalabuseontwoormoreoccasions,orphysicalabuse/sexualabuseononeormoreoccasion.
Inaddition,beingbulliedwasassessedatbaselinebyaskingparticipantsiftheyhadbeenbulliedregularlybeforetheageof16years.
Forthepresentstudy,childhoodvictimi-sationwasdenedpresentifaparticipantreportedchild-hoodabuseorbullyingatbaseline.
Individualswithmissingdataonchildhoodvictimisationwereexcludedfromthepresentanalyses(n=140).
AdultvictimisationAtbaseline,participantswereaskedaboutlifetimeviolentandpsychologicalvictimisationbyanintimatepartner.
Inaddition,lifetimesexualvictimisationbyanypersoningeneralsincetheageof16yearswasassessed.
Toincreasethelikelihoodoftheseformsofvictimisationbeingreported,theinterviewerdidnotmentionanytypeofvic-timisationduringtheinterview.
Instead,differentformsofvictimisationwerelistedandnumberedinabooklet.
Par-ticipantswereaskedtoprovidethenumbersofthetypeofvictimisation.
Psychologicalvictimisationincludedname-calling,offending,belittling,punishingunjustly,black-mailingandthreatening.
Physicalvictimisationincludedkicking,biting,hitting,tryingtowoundwithanobject(gun,knife,pieceofwood,pairofscissors,otherobject)orhotwater.
Sexualvictimisationincludedunwantedtouch-ing,forcedundressingandforcedsexualactivity.
Consis-tentwithpreviouswork[7,36],psychologicalvictimisationwasdenedpresentifitoccurredontwoormoreoccasions,andviolent/sexualvictimisationononeormoreoccasions.
Atbothfollow-upmeasurements,participantswereaskedaboutviolent,psychologicalandsexualvictimisationsincethelastassessmentbyanypersoningeneral,andifso,bywhom(i.
e.
partner,ex-partner,familymember,acquaintance,stranger).
Toreachconsistencywiththebaselinequestions,physicalvictimisationandpsychologi-calvictimisationatfollow-upweredenedpresentiftherespectiveformofvictimisationwasperpetratedbyanintimatepartner.
Sexualvictimisationatfollow-upwasdenedpresentifitwasperpetratedbyanypersoningeneral,conformbaselinemeasurement.
Thefrequenciesforallvictimisationoutcomesweresimilartothefre-quenciesatbaseline.
Incidentphysicalvictimisationbyapartner(hereafter:physicalvictimisation),incidentpsychologicalvictimisa-tionbyapartner(hereafter:psychologicalvictimisation)andincidentsexualvictimisation(hereafter:sexualvic-timisation)weredenedaspresentiftheparticipantreportedtherespectivetypeofvictimisationatanyofthefollow-upinterviews,whileparticipantswiththerespectivetypeofvictimisationatbaselinewereexcluded.
Inaddi-tion,asummaryvariable(anyincidentadultvictimisation)wasgeneratedtoidentifyparticipantswhoexperiencedanyformofadultvictimisationatfollow-up,buthadnovic-timisationatbaseline.
Participantswithoutapartneratbaselineandanyofthefollow-upmeasurementswereexcludedwhenanalysingphysicalvictimisation,psycho-logicalvictimisationorthesummaryvariable.
ConfoundersAge,gender,lowsocio-economicstatus,pastcriminalactivityandsubstanceusedisorderswerehypothesizedtobeconfoundersinthepresentanalyses[13,37].
Arrestwasusedasameasureofcriminalactivity.
Self-reportedarrestwasobtainedatbaselinebyaskingparticipantsiftheyhadeverbeenarrested.
Inaddition,CIDI3.
0wasusedtodenebaseline,lifetimediagnosesofanysubstanceusedisorder.
Finally,householdincomewasaproxyforsocio-economicstatus.
Thevariableincludedthreestrata,basedonmonthlyincome:low(\1500),middle(1500–3300)andhigh([3300).
StatisticalanalysesAllstatisticalanalyseswereperformedusingStataversion13[38].
Baselinecharacteristicswereassessedforthecompletesample.
Inaddition,subjectswithclinicallyval-idatedPEatbaselinewerecomparedwithsubjectswithoutclinicallyvalidatedPE,usingChi-squaretestsandinde-pendentsamplettests.
SocPsychiatryPsychiatrEpidemiol(2017)52:1363–13741365123Logisticregressionanalyseswereperformedtoanswerthesixresearchquestions.
Allregressionmodelsincludedage,gender,householdincome(dummiesofstrata),base-linesubstanceusedisordersandarrestascovariates.
ThenumbersoftheanalysescorrespondwiththenumbersinFig.
1.
1.
Logisticregressionanalyseswereperformedtoexam-inetheassociationbetweenbaselinePEandincidentadultvictimisationatfollow-upinthecompletesample.
2.
Theassociationbetweenbaselinechildhoodvictimi-sationandincidentadultvictimisationwasassessedinaseparatelogisticregressionanalysis.
3.
ToassesswhetherchildhoodvictimisationmoderatedtheassociationbetweenPEandadultvictimisation,alogisticregressionanalysiswasperformedusingincidentadultvictimisationasthedependentvariableandPE,childhoodvictimisationandtheinteractiontermPE*childhoodvictimisationastheindependentvariables.
Toanyinteractioneffect,alogisticregres-sionanalysiswasperformedusingincidentadultvictimisationasthedependentvariableandacategor-icalvariablecontainingthefollowingcategoriesastheindependentvariable,modelledasdummies:(1)NoPE,nochildhoodvictimisation(referencegroup);(2)NoPE,childhoodvictimisationpresent;(3)PEpresent,nochildhoodvictimisation;(4)PEpresent,childhoodvictimisationpresent.
Iftheinteractiontermwasbelowalpha(a=0.
10),theassociationsbetweenPE,childhoodvictimisationandadultvictimisationwereanalysedstratiedbypresenceorabsenceofPEandchildhoodvictimisation,respectively.
4.
TheassociationbetweenbaselineadultvictimisationandincidentPEatfollow-upwasexaminedinalogisticregressionanalysisinthecompletesample.
5.
TheassociationbetweenchildhoodvictimisationandincidentPEwasassessedinalogisticregressionanalysisinthecompletesample.
6.
ThehypothesizedmoderatingeffectofchildhoodvictimisationontheassociationbetweenbaselineadultvictimisationandincidentPEwasexaminedbyconductingalogisticregressionanalysisusingincidentPEasthedependentvariableandadultvictimisation,childhoodvictimisationandtheinteractiontermadultvictimisation*childhoodvictimisationastheinde-pendentvariables.
AlogisticregressionanalysiswasperformedusingincidentPEasthedependentvariableandacategoricalvariablecontainingthefollowingcategoriesastheindependentvariable:(1)Nochild-hoodvictimisation,noadultvictimisation(referencecategory);(2)childhoodvictimisationpresent,noadultvictimisation;(3)nochildhoodvictimisation,adultvictimisationpresent;(4)childhoodvictimisationpre-sent,adultvictimisationpresent.
Again,ifthepvalueoftheinteractiontermwasbelowalpha(a=0.
10),theassociationsbetweenchildhoodvictimisation,adultvictimisationandPEwereexaminedinstratiedanalyses.
ResultsBaselinecharacteristicsAtbaseline,thecompletesampleincluded6359partici-pants,afterexclusionofindividualswithself-reportedPEwhocouldnotbereachedforre-interview(n=287).
Oftheseparticipants,5.
3%(n=340)reportedclinicallyval-idatedPE.
MorewomenthanmenreportedPE(Table1).
Moreover,theproportionofindividualswithPEdifferedsignicantlybetweenthestrataofhouseholdincome.
SubjectswithPEwereoverrepresentedinthelowincomegroupandunderrepresentedinthehigh-incomegroup.
Furthermore,thebaselineprevalenceofchildhoodvic-timisation,adultvictimisation,lifetimesubstanceusedis-ordersandarrestwassignicantlyhigherinindividualswithPE(Table1).
TheassociationbetweenbaselinepsychoticexperiencesandadultvictimisationTheoddsratio(OR)oftheassociationbetweenbaselinePEandanyincidentadultvictimisationwas2.
09(95%CI0.
79–5.
56;Table2).
PEwereassociatedwithallformsofvictimisation,buttheORwasonlystatisticallysignicantfortheassociationbetweenPEandsexualvictimisation(OR=3.
51,95%CI1.
54–7.
96).
TheassociationbetweenbaselinechildhoodvictimisationandadultvictimisationChildhoodvictimisationwasassociatedwithallformsofadultvictimisation,withORsrangingfrom2.
70(95%CI1.
41–5.
16)forsexualvictimisationto5.
49(95%CI2.
26–13.
34)forphysicalvictimisation(Table2).
TheinteractionbetweenpsychoticexperiencesandchildhoodvictimisationontheoutcomeadultvictimisationTheinteractiontermPE*Childhoodvictimisationwasbelowalphafortheoutcomeanyadultvictimisation(Table2;p=0.
08).
Follow-upanalysisoftheinteractionbetweenchildhoodvictimisationandPEontheoutcome1366SocPsychiatryPsychiatrEpidemiol(2017)52:1363–1374123adultvictimisation,usingacategoricalpredictor,showedthatisolatedPE(OR=4.
49,95%CI1.
27–15.
90)andisolatedchildhoodvictimisation(OR=3.
73,95%CI2.
16–6.
47)wereassociatedwithanincreasedriskofanyadultvictimisation.
However,theco-occurrenceofPEandchildhoodvictimisationwasassociatedwithalowerriskofanyadultvictimisationthantheirisolatedeffects(OR=2.
88,95%CI0.
62–13.
39),thusindicatinganega-tiveinteraction.
AnalysesinthesubsamplestratiedbypresenceorabsenceofchildhoodvictimisationshowedthatPEwereassociatedwithadultvictimisationinthesub-samplewithoutchildhoodvictimisation(OR=4.
81,95%CI1.
34–17.
29).
However,inthesubsamplewithchildhoodvictimisationtherewasnoassociationbetweenPEandadultvictimisation.
Similarly,childhoodvictimisationwasassociatedwithadultvictimisationinthesubsamplewith-outbaselinePE(OR=3.
76,95%CI2.
17–6.
53),butwasnotassociatedwithadultvictimisationinthesubsamplewithbaselinePE.
Resultsfortheoutcomesphysicalvic-timisationandpsychologicalvictimisationweresimilartotheresultsofanyadultvictimisationandshowedatrendtowardsanegativeinteractionaswell(p=0.
16forphysicalvictimisation,p=0.
07forpsychologicalvictim-isation).
Interactioncouldnotbeexaminedinthemodelwithsexualvictimisationastheoutcome,becausenoneofthesubjectshadPEinabsenceofchildhoodvictimisation.
However,theco-occurrenceofPEandchildhoodvictimisation(OR=8.
72,95%CI3.
40–22.
32)showedastrongerassociationwithsexualvictimisationthanisolatedchildhoodvictimisation(OR=2.
24,95%CI1.
10–4.
57).
TheassociationbetweenbaselineadultvictimisationandincidentpsychoticexperiencesInthecompletesample,allformsofbaselineadultvic-timisationwereassociatedwithincidentPE,withORsrangingfrom1.
88(95%CI1.
34–2.
64)forpsychologicalvictimisationto3.
77(95%CI2.
32–6.
12)forsexualvic-timisation,afteradjustmentforconfounders(Table3).
TheassociationbetweenbaselinechildhoodvictimisationandincidentpsychoticexperiencesChildhoodvictimisationwasassociatedwithincidentPEinthecompletesample(Table3;OR=2.
64,95%CI1.
90–3.
66).
TheinteractionbetweenchildhoodvictimisationandadultvictimisationontheoutcomepsychoticexperiencesTheriskofPEinsubjectswithbothchildhoodvictimisa-tionandadultvictimisationwasnotlargerthantheproductoftheirrisksinparticipantswithisolatedadultTable1BaselinecharacteristicsofindividualswithandwithoutclinicallyvalidatedPECompletesampleSubjectswithPEaSubjectswithoutPEatv2dfpbDemographicsN63593406019––––Numberofmales(%)2852(44.
9)127(37.
4)2725(45.
3)–8.
16210.
004Age(SD)44.
4(12.
5)43.
0(13.
2)44.
4(12.
5)2.
049–63570.
041Householdincome––––32.
9152\0.
001Low1439(25.
4)119(38.
4)1320(24.
7)Middle2635(46.
5)133(42.
9)2502(46.
7)High1590(28.
1)58(18.
7)1532(28.
6)BaselinevictimisationChildhoodvictimisation,N(%)2138(34.
4)207(61.
8)1931(32.
8)–118.
1611\0.
001Physicalvictimisationbypartner,N(%)614(10.
3)73(23.
3)541(9.
6)–60.
2161\0.
001Psychologicalvictimisationbypartner,N(%)1716(28.
9)148(47.
4)1568(27.
9)–55.
1061\0.
001Sexualviolencevictimisation,N(%)279(4.
5)47(14.
1)232(4.
0)–75.
9561\0.
001Anyadultvictimisation,N(%)1940(33.
2)171(55.
0)1769(32.
0)–70.
3551\0.
001BaselineconfoundersAnylifetimesubstanceusedisorder1037(16.
3)106(31.
2)931(15.
5)–58.
1871\0.
001Everarrested1346(21.
2)88(26.
0)1258(20.
9)–4.
90610.
027aClinicallyvalidatedPEbp-valueresultingfromttestorChi-squaretestfordifferencebetweenparticipantswithvs.
withoutPESocPsychiatryPsychiatrEpidemiol(2017)52:1363–13741367123Table2Resultsoflogisticregressionanalyses(ORsand95%CI)oftheassociationbetweenbaselinepsychoticexperiences,baselinechildhoodvictimisationandincidentadultvictimisation,adjustedforage,gender,socioeconomicstatus,self-reportedarrestandlifetimesubstanceusedisorderOutcomeAnyincidentadultvictimisation(n=2296)pIncidentsexualvictimisation(n=3881)pIncidentphysicalvictimisationbypartner(n=2994)pIncidentpsychologicalvictimisationbypartner(n=2420)pSeparatemodelsPEincompletesample2.
09(0.
79–5.
56)0.
143.
51(1.
54–7.
96)\0.
011.
84(0.
42–8.
14)0.
421.
99(0.
75–5.
28)0.
17CVaincompletesample3.
15(1.
87–5.
30)\0.
012.
70(1.
41–5.
16)\0.
015.
49(2.
26–13.
34)\0.
013.
01(1.
74–5.
24)\0.
01InteractiontermPE*CVincompletesample0.
17(0.
02–1.
22)0.
08N.
A.
N.
A.
0.
12(0.
01–2.
32)0.
160.
16(0.
02–1.
12)0.
07CategoricalpredictorsNoPE,noCVReference–Reference–Reference–Reference–NoPE,CVpresent3.
73(2.
16–6.
47)\0.
012.
24(1.
10–4.
57)0.
037.
99(2.
90–22.
06)\0.
013.
47(1.
94–6.
21)\0.
01PEpresent,noCV4.
49(1.
27–15.
90)0.
02N.
A.
N.
A.
6.
32(0.
71–56.
22)0.
104.
62(1.
30–16.
36)0.
02PEpresent,CVpresent2.
88(0.
62–13.
39)0.
188.
72(3.
40–22.
32)\0.
015.
85(0.
64–53.
11)0.
122.
54(0.
55–11.
73)0.
23StratiedmodelsPEinsubsamplewithoutCV4.
81(1.
34–17.
29)0.
02––––4.
63(1.
29–16.
60)0.
02PEinsubsamplewithCV0.
73(0.
16–3.
37)0.
69––––0.
76(0.
17–3.
46)0.
72CVinsubsamplewithoutPE3.
76(2.
17–6.
53)\0.
01––––3.
49(1.
95–6.
26)\0.
01CVinsubsamplewithPE0.
53(0.
06–4.
32)0.
55––––0.
52(0.
07–4.
14)0.
54aChildhoodvictimisation1368SocPsychiatryPsychiatrEpidemiol(2017)52:1363–1374123victimisationandsubjectswithisolatedchildhoodvictim-isationseparately,indicatingthattherewasnointeractionbetweenadultvictimisationandchildhoodvictimisation(Table3).
DiscussionOverviewofresultsToourknowledge,thepresentstudyisthersttoexaminethebidirectional,longitudinalassociationsbetweenPEandadultvictimisationusingaprospective,generalpopulationsample,whilealsoassessingthemoderatingeffectofchildhoodvictimisation.
ItwashypothesizedthatPEincreasetheriskofincidentadultvictimisation,andthatadultvictimisationincreasestheriskofincidentPE.
Moreover,itwashypothesizedthatchildhoodvictimisationincreasestheriskofbothadultvictimisationandPEandthatthebidirectionalassociationsbetweenPEandadultvictimisationwouldbeincreasedbythepresenceofchildhoodvictimisation(Fig.
1).
Thepresentresultsshowedevidenceforthehypothesizedbidirectionalasso-ciationbetweenadultvictimisationandPE.
However,thehypothesisofapositiveinteractionbetweenchildhoodvictimisationandbothPEandadultvictimisationwasfalsied(Fig.
2).
TheassociationbetweenbaselinepsychoticexperiencesandincidentadultvictimisationPreviousstudiesreportedanassociationbetweenpsychoticdisorderandadultvictimisation[8,15,37,39].
However,thetemporalsequenceoftheassociationcouldnotbedeterminedinthesestudiesbecauseallstudiesusedcross-sectionalstudydesigns.
ThepresentstudyshowedthatindividualswithisolatedPEorisolatedchildhoodvictim-isationatbaselinewereatincreasedriskofanyadultvic-timisation.
However,contrarytoourhypothesisthatPEandchildhoodvictimisationwouldactsynergisticallytoincreasetheriskofadultvictimisation,resultsshowedthattheco-occurrenceofPEwithchildhoodvictimisationwasassociatedwithalowerriskofadultvictimisationthanisolatedPEorisolatedchildhoodvictimisation.
TheresultsofthestratiedanalysesrevealedthatexposuretoPEafterchildhoodvictimisationwasnotassociatedwithadultvictimisation,whilePEincreasedtheriskofadultvictim-isationinindividualswithoutchildhoodvictimisation.
Similarly,exposuretochildhoodvictimisationwasnotariskfactorforadultvictimisationinadultsexposedtoPE,whilechildhoodvictimisationdidincreasetheriskofadultvictimisationinindividualsnotexposedtoPE.
Therefore,therewasanegativeinteractioneffect,pointingtowardsTable3Resultsoflogisticregressionanalyses(ORsand95%CI)oftheassociationbetweenbaselinevictimisation,childhoodvictimisationandincidentpsychoticexperiences,adjustedforage,gender,socioeconomicstatus,self-reportedarrestandlifetimesubstanceusedisorderOutcomeAnyincidentPE(n=3691)SpecicationofadultvictimisationAnyAVpSexualvictimisationpPhysicalvictimisationbypartnerpPsychologicalvictimisationbypartnerpSeparatemodelsAVaincompletesample2.
28(1.
63–3.
20)\0.
013.
77(2.
32–6.
12)\0.
012.
20(1.
46–3.
32)\0.
011.
88(1.
34–2.
64)\0.
01CVbincompletesample2.
64(1.
90–3.
66)\0.
012.
64(1.
90–3.
66)\0.
012.
64(1.
90–3.
66)\0.
012.
64(1.
90–3.
66)\0.
01InteractiontermAV*CVincompletesample0.
95(0.
48–1.
90)0.
890.
75(0.
23–2.
48)0.
640.
86(0.
35–2.
07)0.
731.
19(0.
58–2.
42)0.
63CategoricalpredictorNoCV,noAVReference–Reference–Reference–Reference–CVpresent,noAV2.
38(1.
46–3.
89)\0.
012.
43(1.
71–3.
46)\0.
012.
54(1.
74–3.
70)\0.
012.
30(1.
47–3.
59)\0.
01NoCV,AVpresent1.
94(1.
14–3.
31)0.
023.
64(1.
23–10.
74)0.
022.
09(1.
00–4.
36)0.
051.
41(0.
80–2.
49)0.
24CVpresent,AVpresent4.
40(2.
83–6.
83\0.
016.
62(3.
77–11.
63)\0.
014.
54(2.
70–7.
64)\0.
013.
84(2.
49–5.
93)\0.
01aAdultvictimisationbChildhoodvictimisationSocPsychiatryPsychiatrEpidemiol(2017)52:1363–13741369123'parallelism'insteadofthehypothesized'synergism'[40,41],thussuggestingthatPEandchildhoodvictimi-sationactthroughcompetingpathwaysinincreasingriskforadultvictimisation.
Inotherwords,anyexcessriskforadultvictimisationwouldalreadyhavebeenconsumedafterisolatedexposuretoeitherPEorchildhoodvictimisation.
Theresultsfortheoutcomesphysicalvictimisationandpsychologicalvictimisationweresimilartotheresultsoftheoutcomeofanyadultvictimisation,showingatrendtowardsanegativeinteraction.
However,interactioncouldnotbeexaminedfortheoutcomeofsexualvictimisation.
Asopposedtotheresultsforphysicalvictimisationandpsychologicalvictimisation,availableresultsshowthatPEwereassociatedwithincidentsexualvictimisationinthesubgroupwithchildhoodvictimisation.
However,theinteractioncouldnotbeanalysed.
Therefore,itispossiblethattheassociationbetweenPEandadultsexualvictimi-sationdiffersfromtheassociationsbetweenPEandotherformsofadultvictimisation.
Thisndingwouldbeinlinewithpreviousresearchthatfoundthatchildhoodsexualabusewasdifferentiallyassociatedwithpsychosiscom-paredtoothervictimisingexperiencesduringchildhood[9,12,29].
ThemechanismsbehindtheassociationbetweenPEandincidentadultvictimisationremainunclear.
OnepossibleexplanationisthatindividualswithPEdisplaydisorderedbehaviour,forexamplearisingfromparanoiddelusions,leadingtosocialconictandvictimisation.
PreviousresearchshowedthatPEincreasetheriskofviolenceperpetration[36].
Therefore,itispossiblethatadultvic-timisationoccursinresponsetoviolenceperpetration.
AnotherpossibleexplanationisthatindividualswithPEliveinpoorersocialenvironmentswhereadultvictimisa-tionismorelikelytooccur.
However,thepresentanalyseswereadjustedforhouseholdincomeasaproxyforsocio-economicstatus.
Finally,itispossiblethatindividualswithPEaremorelikelytoreportadultvictimisation,forexamplebecauseofparanoidinterpretationsofsocialinteractions.
Toourknowledge,thereliabilityofadultvictimisationreportsinindividualswithpsychosishasnotbeenstudied.
However,ithasbeenshownthatindividualswithpsychosisareabletoprovidereliablereportsofchildhoodvictimisation[42,43].
Therefore,itisunlikelythattheassociationbetweenPEandadultvictimisationcanbefullyattributedtodifferentialreportingofadultvic-timisationbyindividualswithPE.
TheassociationbetweenbaselineadultvictimisationandincidentpsychoticexperiencesResultsofthepresentstudyshowedthatallformsofadultvictimisationwereassociatedwiththedevelopmentofincidentPE.
Thisconrmsourhypothesisthattheassoci-ationbetweenPEandadultvictimisationisbidirectional,similartothepreviouslyreportedbidirectionalassociationbetweenPEandchildhoodvictimisation[10].
Consistentwithpreviouswork[10–12,16,44],childhoodvictimisa-tionwasassociatedwithincidentPE,bothinthepresenceandabsenceofco-occurringadultvictimisation.
However,toourhypothesisofapositiveinteractionbetweenchild-hoodvictimisationandadultvictimisationfortheoutcomeofincidentPE,theriskofPEinsubjectswithbothchild-hoodvictimisationandadultvictimisationwasnotlargerthantheproductoftheisolatedrisks.
Thisndingisnotfullyconsistentwiththepreviousliterature,whichshowedpositiveinteractionsbetweenchildhoodvictimisationandadultvictimisationfortheoutcomePE[21,30,32].
However,allstudiesusedawidevarietyofdenitionsofchildhoodvictimisationandadultvictimisation,thusimpedingdirectcomparison.
Thepresentresultsshowthatchildhoodvictimisationandadultvictimisationareinde-pendent,cumulativeriskfactorsforPE.
Thisndingisrelevant,sincepoly-victimisationisprevalentamongindividualswithseverementalillness[9,15].
Moreover,thisndingshowsthattherearetwopathwaysfromchildhoodvictimisationtopsychosis:onedirectoneandoneindirectonethroughadultvictimisation.
Literatureonthemechanismsbehindtheassociationbetweenadultvictimisationandincidentpsychosisisscarce.
However,bothadultvictimisationandchildhoodvictimisationhavebeenlinkedtovariousnon-psychoticmentaldisorders[45–49].
SincePEareprevalentamongindividualswithnon-psychoticmentaldisorders[18,50]andhavebeenidentiedasanindicatorofseverityinnon-psychoticpsychopathology[50–54],thelinkbetweenadultvictimisationandincidentPEispossiblyconfoundedbythepresenceofnon-psychoticpsychopathology[50–54].
Furthermore,itislikelythatthemechanismsunderlyingtheimpactofadultvictimisationonpsychosisaresimilartothemechanismsmediatingimpactofchildhood**p<0.
01Fig.
2Resultsoftestingthehypothesesrelatingtothebidirectionalassociationbetweenpsychoticexperiencesandvictimisation1370SocPsychiatryPsychiatrEpidemiol(2017)52:1363–1374123victimisationonpsychosis[21,55].
Variousbiologicalandpsychologicalprocesseshavebeenassociatedwithbothpsychosisandchildhoodvictimisation[11,56,57].
Bio-logicalprocessesthatlinkvictimisationwithanincreaseinpsychosisriskincludehyperactivationandsensitizationofthehypothalamic–pituitary–adrenal(HPA)axis,decreasedhippocampalvolume,reducedbrain-derivedneurotrophicfactor(BDNF)[56]andincreaseddopaminerelease[11].
Furthermore,victimisationmayincreasetheriskofpsy-chosispsychologicallybycontributingtothedevelopmentofaworryingthinkingstyle,negativebeliefsabouttheselfandreasoningbiasessuchasjumpingtoconclusions[57].
However,moreresearchtoexaminethemechanismsunderlyingthelinkbetweenpsychosisandadultvictimi-sationisneeded.
StrengthsandlimitationsStrengthofthepresentstudyistheprospective,longitu-dinalstudydesignthatenablesthebidirectionalassessmentofpsychosisandvictimisationinageneralpopulationsample,whilecontrollingforvariousconfoundersidenti-edintheliterature.
Anotherstrengthistheuseofclini-callyvalidatedPEinsteadofself-reportedPE[58].
Theresultsofthestudyshouldbeinterpretedinthelightofsomelimitations.
First,thedenitionsofincidentphysicalandpsychologicalvictimisationhadtoberestrictedtovictimisationbyanintimatepartnertoremainconsistentwiththebaselinedenitionsinthedataset.
Toovercomethislimitation,individualswithoutapartneratbaselineandanyofthefollow-upmeasurementswereexcludedfromtheanalyseswhenusingphysicalvictimi-sation,psychologicalvictimisationoranyadultvictimisa-tionastheoutcome,resultinginanexclusionof1066individualsforthepresentanalysis.
Toourknowledge,nopreviousstudyexaminedwhethervictimisationbyanintimatepartnermaybedifferentlyassociatedwithpsy-chosisthanvictimisationbyanyperson.
Toexaminethis,sensitivityanalyseswereconductedexaminingtheassoci-ationbetweenbaselinePEandadultvictimisationbyanypersonatfollow-up,whileexcludingindividualswithoutapartneratbaseline.
Theresultsofthesesensitivityanalysesweresimilartotheresultsofthemainanalysesintermsofeffectsize.
However,statisticalsignicancewasgreaterasaresultofincreasedstatisticalpower.
Therefore,theassociationbetweenPEandadultvictimisationperpetratedbyanypersonappearstobesimilartotheassociationbetweenPEandadultvictimisationperpetratedbyanintimatepartner.
However,moreresearchisrequiredtoinvestigatethisissue.
Second,baselinedataonhouseholdincomewasmissingfor604individuals.
Again,sensitivityanalyseswereconductedusingthemissingdataasaseparatecategory.
Resultsoftheseanalysesweresimilartothemainresults.
Third,statisticalpowerwaslowinsomeanalyses,inparticularintheinteractionanalyses.
Thus,toscreenforpotentialinteractions,alphaforinteractioneffectswassetat0.
10.
Raisingthealphato0.
10increasestheriskoffalsepositiveresults.
Therefore,itispossiblethatsomeresultsrepresenttypeIerror.
Morestudiesarerequiredtoreplicatetheinteractionsidentiedinthisstudy.
Furthermore,dataaboutbaselinevictimisationwascollectedretrospectively.
Sinceparticipantsinthisstudywereaged18–64years,itispossiblethatdifferencesinrecallimpactedtheresults.
Previousstudiesshowedevi-denceforage-relateddifferentialrecall,withindividualsunderreportingvictimisationasageincreases[59–61].
Therefore,itispossiblethattheassociationsbetweenvictimisationandpsychosisidentiedinthisstudywouldbestrongerintheabsenceofage-relateddifferentialrecall.
Inaddition,theriskofexposuretoadultvictimisationandPEvariesbyage,suggestinganinteractioneffectbetweenage,PEandadultvictimisation.
Toexaminethis,sensi-tivityanalyseswereconductedtestingforinteractioneffectsbetweenthevariouspredictorvariablesandveagecategories.
ResultsofthisanalysisshowedthattherewasnoevidenceforinteractionbetweenanyofthepredictorvariablesandagewhenstudyingincidentPE.
Inaddition,theassociationbetweenbaselinePEandincidentAVdidnotdifferbetweenthevariousagegroupsabove25years,buttherewassomeevidenceforadecreasedassociationintheyoungestagegroup(18–25years).
However,statisticalpowerintheyoungestagegroupwasinsufcienttofurtherexplorethisnding.
Otherstudieswithmoreyoungadultsareneededtostudythishypothesis.
Finally,follow-updatawasmissingfor2028partici-pants(attrition=30.
5%).
Attritionwasassociatedwithsociodemographicfactors,butnotwithmentalhealthsta-tus,makingitunlikelythatattritionwouldinducebias[62].
Inaddition,attritionwasnotassociatedwithbaselinePEorbaselinevictimisationstatus.
Therefore,selectiveattritionisunlikelytohavebiasedtheresults.
ConclusionsandimplicationsThepresentstudydidnotndevidenceforallhypotheses(Fig.
2).
Tointegratethisupdatedevidence,Fig.
2wastransformedintoFig.
3,removingtheconnectionsthatwerenotsupportedbythepresentresults.
PEandchild-hoodvictimisationwereidentiedascompetingriskfac-torsforadultvictimisation.
Inaddition,childhoodvictimisationandadultvictimisationwereindependent,cumulativeriskfactorsforPE.
Thus,PEandadultvic-timisationarebidirectionallyassociatedinindividualsSocPsychiatryPsychiatrEpidemiol(2017)52:1363–13741371123withoutchildhoodvictimisation.
Inindividualswithchildhoodvictimisation,therearetwopathwaysfromchildhoodvictimisationtoPE:onedirectandoneindirectthroughadultvictimisation.
Inconclusion,psychosisandvictimisationarestronglyinterconnectedthroughoutthelifecourse,resultinginacomplexinterplayinwhichchildhoodvictimisationandadultvictimisationleadtobothPEandre-victimisingexperiences.
Becausevictimisationacrossthelifecoursehasbeenassociatedwithseveraladverseconsequences[49,63–66],preventionstrategiesagainstvictimisation,bothduringchildhoodandduringadulthood,areimportanttopreventindividualsfromenteringaspiralleadingtomentalillnessandre-victimi-sation.
Moreover,interventionprogramsareneededtopreventfurtherre-victimisationinindividualswhohavealreadyexperiencedvictimisation.
However,thedevelop-mentofadequatepreventionandinterventionprogramsrequiresfurtherunderstandingofthemechanismsunder-lyingtheassociationsidentiedinthepresentstudy.
Therefore,morelongitudinalresearchisrequiredtoobtainadeeperunderstandingofthecomplexinterplaybetweenpsychosisandvictimisationacrossthelifecourse.
AcknowledgementsNEMESIS-2isconductedbytheNetherlandsInstituteofMentalHealthandAddiction(TrimbosInstitute)inUtrecht.
FinancialsupporthasbeenreceivedfromtheMinistryofHealth,WelfareandSport,withsupplementarysupportfromtheNetherlandsOrganizationforHealthResearchandDevelopment(ZonMw)andtheGeneticRiskandOutcomeofPsychosis(GROUP)investigators.
FundingwasprovidedbyMinisterievanVolksge-zondheid,WelzijnenSport(GrantNo.
310253).
CompliancewithethicalstandardsEthicalstandardsNEMESIS-2wasapprovedbytheMedicalEthicsReviewCommitteeforInstitutionsonMentalHealthCare(METIGG).
Participantsprovidedwritteninformedconsenttopar-ticipateintheinterview,afterfullwrittenandverbalinformationaboutthestudywasgivenbeforeandatthestartofthebaselineassessment.
ConictofinterestOnbehalfofallauthors,thecorrespondingauthorstatesthatthereisnoconictofinterest.
OpenAccessThisarticleisdistributedunderthetermsoftheCreativeCommonsAttribution4.
0InternationalLicense(http://creativecommons.
org/licenses/by/4.
0/),whichpermitsunrestricteduse,distribution,andreproductioninanymedium,providedyougiveappropriatecredittotheoriginalauthor(s)andthesource,providealinktotheCreativeCommonslicense,andindicateifchangesweremade.
References1.
CoidJW,UllrichS,BebbingtonP,FazelS,KeersR(2016)Paranoidideationandviolence:meta-analysisofindividualsubjectdataof7populationsurveys.
SchizophrBull42(4):907–915.
doi:10.
1093/schbul/sbw0062.
DouglasKS,GuyLS,HartSD(2009)Psychosisasariskfactorforviolencetoothers:ameta-analysis.
PsycholBull135(5):679–706.
doi:10.
1037/a00163113.
FazelS,GulatiG,LinsellL,GeddesJR,GrannM(2009)Schizophreniaandviolence:systematicreviewandmeta-analy-sis.
PLoSMed6(8):e1000120.
doi:10.
1371/journal.
pmed.
10001204.
FazelS,YuR(2011)Psychoticdisordersandrepeatoffending:systematicreviewandmeta-analysis.
SchizophrBull37(4):800–810.
doi:10.
1093/schbul/sbp1355.
LargeMM,NielssenO(2011)Violenceinrst-episodepsy-chosis:asystematicreviewandmeta-analysis.
SchizophrRes125(2–3):209–220.
doi:10.
1016/j.
schres.
2010.
11.
0266.
CrispAH,GelderMG,RixS,MeltzerHI,RowlandsOJ(2000)Stigmatisationofpeoplewithmentalillnesses.
BrJPsychiatry177:4–77.
TenHaveM,deGraafR,vanWeeghelJ,vanDorsselaerS(2014)Theassociationbetweencommonmentaldisordersandviolence:towhatextentisitinuencedbypriorvictimization,negativelifeeventsandlowlevelsofsocialsupportPsycholMed44(7):1485–1498.
doi:10.
1017/S00332917130022628.
ChoeJY,TeplinLA,AbramKM(2008)Perpetrationofviolence,violentvictimization,andseverementalillness:balancingpublichealthconcerns.
PsychiatrServ59(2):153–164.
doi:10.
1176/ps.
2008.
59.
2.
1539.
BebbingtonPE,BhugraD,BrughaT,SingletonN,FarrellM,JenkinsR,LewisG,MeltzerH(2004)Psychosis,victimisationandchildhooddisadvantage:evidencefromthesecondBritishNationalSurveyofPsychiatricMorbidity.
BrJPsychiatry185(3):220–226.
doi:10.
1192/bjp.
185.
3.
22010.
KelleherI,KeeleyH,CorcoranP,RamsayH,WassermanC,CarliV,SarchiaponeM,HovenC,WassermanD,CannonM(2013)Childhoodtraumaandpsychosisinaprospectivecohortstudy:cause,effect,anddirectionality.
AmJPsychiatry170(7):734–741.
doi:10.
1176/appi.
ajp.
2012.
1209116911.
MorganC,Gayer-AndersonC(2016)Childhoodadversitiesandpsychosis:evidence,challenges,implications.
WorldPsychiatry15(2):93–102.
doi:10.
1002/wps.
2033012.
ReadJ,vanOsJ,MorrisonAP,RossCA(2005)Childhoodtrauma,psychosisandschizophrenia:aliteraturereviewwiththeoreticalandclinicalimplications.
ActaPsychiatrScand112(5):330–350.
doi:10.
1111/j.
1600-0447.
2005.
00634.
x13.
deMooijLD,KikkertM,LommerseNM,PeenJ,MeijwaardSC,TheunissenJ,DuurkoopPW,GoudriaanAE,VanHL,BeekmanAT,DekkerJJ(2015)Victimisationinadultswithseverementalillness:prevalenceandriskfactors.
BrJPsychiatry207(6):515–522.
doi:10.
1192/bjp.
bp.
113.
14337014.
HartC,deVetR,MoranP,HatchSL,DeanK(2012)AUKpopulation-basedstudyoftherelationshipbetweenmentaldis-orderandvictimisation.
SocPsychiatryPsychiatrEpidemiol47(10):1581–1590.
doi:10.
1007/s00127-011-0464-715.
KampermanAM,HenrichsJ,BogaertsS,LesaffreEM,WierdsmaAI,GhauharaliRR,SwildensW,NijssenY,vanderFig.
3Thecomplexinterplayofpsychosisandvictimisationacrossthelifecourse.
//Pointstowardsparallelism,i.
e.
childhoodvictim-isationandpsychoticexperiencesarecompetingriskfactorsforadultvictimisation1372SocPsychiatryPsychiatrEpidemiol(2017)52:1363–1374123GaagM,TheunissenJR,DelespaulPA,vanWeeghelJ,vanBusschbachJT,KroonH,TeplinLA,vandeMheenD,MulderCL(2014)Criminalvictimisationinpeoplewithseverementalillness:amulti-siteprevalenceandincidencesurveyintheNetherlands.
PLoSONE9(3):e91029.
doi:10.
1371/journal.
pone.
009102916.
VareseF,SmeetsF,DrukkerM,LieverseR,LatasterT,ViechtbauerW,ReadJ,vanOsJ,BentallRP(2012)Childhoodadversitiesincreasetheriskofpsychosis:ameta-analysisofpatient-control,prospective-andcross-sectionalcohortstudies.
SchizophrBull38(4):661–671.
doi:10.
1093/schbul/sbs05017.
WolkeD,LereyaST,FisherHL,LewisG,ZammitS(2014)Bullyinginelementaryschoolandpsychoticexperiencesat18years:alongitudinal,population-basedcohortstudy.
PsycholMed44(10):2199–2211.
doi:10.
1017/S003329171300291218.
VanOsJ,HanssenM,BijlR,RavelliA(2000)Straus(1969)revisited:apsychosiscontinuuminthegeneralpopulationSchizophrRes45(1–2):11–2019.
vanOsJ,LinscottRJ(2012)Introduction:theextendedpsychosisphenotype–relationshipwithschizophreniaandwithultrahighriskstatusforpsychosis.
SchizophrBull38(2):227–230.
doi:10.
1093/schbul/sbr18820.
vanOsJ,ReininghausU(2016)Psychosisasatransdiagnosticandextendedphenotypeinthegeneralpopulation.
WorldPsy-chiatry15(2):118–124.
doi:10.
1002/wps.
2031021.
LatasterJ,Myin-GermeysI,LiebR,WittchenHU,vanOsJ(2012)Adversityandpsychosis:a10-yearprospectivestudyinvestigatingsynergismbetweenearlyandrecentadversityinpsychosis.
ActaPsychiatrScand125(5):388–399.
doi:10.
1111/j.
1600-0447.
2011.
01805.
x22.
KhalifehH,OramS,OsbornD,HowardLM,JohnsonS(2016)Recentphysicalandsexualviolenceagainstadultswithseverementalillness:asystematicreviewandmeta-analysis.
IntRevPsychiatry28(5):433–451.
doi:10.
1080/09540261.
2016.
122360823.
WidomCS,CzajaSJ,DuttonMA(2008)Childhoodvictimiza-tionandlifetimerevictimization.
ChildAbuseNegl32(8):785–796.
doi:10.
1016/j.
chiabu.
2007.
12.
00624.
BarriosYV,GelayeB,ZhongQ,NicolaidisC,RondonMB,GarciaPJ,SanchezPA,SanchezSE,WilliamsMA(2015)Associationofchildhoodphysicalandsexualabusewithintimatepartnerviolence,poorgeneralhealthanddepressivesymptomsamongpregnantwomen.
PLoSONE10(1):e0116609.
doi:10.
1371/journal.
pone.
011660925.
BensleyL,VanEenwykJ,WynkoopSimmonsK(2003)Child-hoodfamilyviolencehistoryandwomen'sriskforintimatepartnerviolenceandpoorhealth.
AmJPrevMed25(1):38–4426.
BlomH,HogbergU,OlofssonN,DanielssonI(2014)Strongassociationbetweenearlierabuseandrevictimizationinyouth.
BMCPublHealth14:715.
doi:10.
1186/1471-2458-14-71527.
CoidJ,PetruckevitchA,FederG,ChungW,RichardsonJ,MooreyS(2001)Relationbetweenchildhoodsexualandphysicalabuseandriskofrevictimisationinwomen:across-sectionalsurvey.
Lancet358(9280):450–45428.
McIntyreJK,WidomCS(2011)Childhoodvictimizationandcrimevictimization.
JInterpersViolence26(4):640–663.
doi:10.
1177/088626051036586829.
FryD,McCoyA,SwalesD(2012)Theconsequencesofmal-treatmentonchildren'slives:asystematicreviewofdatafromtheEastAsiaandPacicRegion.
TraumaViolenceAbuse13(4):209–233.
doi:10.
1177/152483801245587330.
BebbingtonP,JonasS,KuipersE,KingM,CooperC,BrughaT,MeltzerH,McManusS,JenkinsR(2011)Childhoodsexualabuseandpsychosis:datafromacross-sectionalnationalpsy-chiatricsurveyinEngland.
BrJPsychiatry199(1):29–37.
doi:10.
1192/bjp.
bp.
110.
08364231.
MorganC,ReininghausU,FearonP,HutchinsonG,MorganK,DazzanP,BoydellJ,KirkbrideJB,DoodyGA,JonesPB,MurrayRM,CraigT(2014)Modellingtheinterplaybetweenchildhoodandadultadversityinpathwaystopsychosis:initialevidencefromtheAESOPstudy.
PsycholMed44(2):407–419.
doi:10.
1017/S003329171300076732.
MorganC,ReininghausU,ReichenbergA,FrissaS,HotopfM,HatchSL,TeamSEs(2014)Adversity,cannabisuseandpsy-choticexperiences:evidenceofcumulativeandsynergisticeffects.
BrJPsychiatry204:346–353.
doi:10.
1192/bjp.
bp.
113.
13445233.
deGraafR,TenHaveM,vanDorsselaerS(2010)TheNether-landsMentalHealthSurveyandIncidenceStudy-2(NEMESIS-2):designandmethods.
IntJMethodsPsychiatrRes19(3):125–141.
doi:10.
1002/mpr.
31734.
HaroJM,Arbabzadeh-BouchezS,BrughaTS,deGirolamoG,GuyerME,JinR,LepineJP,MazziF,RenesesB,VilagutG,SampsonNA,KesslerRC(2006)ConcordanceoftheCompositeInternationalDiagnosticInterviewVersion3.
0(CIDI3.
0)withstandardizedclinicalassessmentsintheWHOWorldMentalHealthsurveys.
IntJMethodsPsychiatrRes15(4):167–18035.
vanNieropM,vanOsJ,GuntherN,Myin-GermeysI,deGraafR,tenHaveM,vanDorsselaerS,BakM,vanWinkelR(2012)Phenotypicallycontinuouswithclinicalpsychosis,discontinuousinneedforcare:evidenceforanextendedpsychosisphenotype.
SchizophrBull38(2):231–238.
doi:10.
1093/schbul/sbr12936.
HoningsS,DrukkerM,TenHaveM,deGraafR,vanDorsselaerS,vanOsJ(2016)Psychoticexperiencesandriskofviolenceperpetrationandarrestinthegeneralpopulation:aprospectivestudy.
PLoSONE11(7):e0159023.
doi:10.
1371/journal.
pone.
015902337.
MorganVA,MorganF,GalletlyC,ValuriG,ShahS,JablenskyA(2016)Sociodemographic,clinicalandchildhoodcorrelatesofadultviolentvictimisationinalarge,nationalsurveysampleofpeoplewithpsychoticdisorders.
SocPsychiatryPsychiatrEpi-demiol51(2):269–279.
doi:10.
1007/s00127-015-1155-638.
StataCorp.
(2013)StataStatisticalSoftware:Release13.
Stata-CorpLP,CollegeStation39.
LatalovaK,KamaradovaD,PraskoJ(2014)Violentvictimiza-tionofadultpatientswithseverementalillness:asystematicreview.
NeuropsychiatrDisTreat10:1925–1939.
doi:10.
2147/NDT.
S6832140.
DarrochJ(1997)Biologicsynergismandparallelism.
AmJEpidemiol145(7):661–66841.
MurrayRM,JonesPB,SusserE,VanOsJ,CannonM(2002)TheEpidemiologyofSchizophrenia.
CambridgeUniversityPress42.
GoodmanLA,ThompsonKM,WeinfurtK,CorlS,AckerP,MueserKT,RosenbergSD(1999)Reliabilityofreportsofviolentvictimizationandposttraumaticstressdisorderamongmenandwomenwithseriousmentalillness.
JTraumaStress12(4):587–599.
doi:10.
1023/A:102470891614343.
FisherHL,CraigTK,FearonP,MorganK,DazzanP,LappinJ,HutchinsonG,DoodyGA,JonesPB,McGufnP,MurrayRM,LeffJ,MorganC(2011)Reliabilityandcomparabilityofpsy-chosispatients'retrospectivereportsofchildhoodabuse.
Schi-zophrBull37(3):546–553.
doi:10.
1093/schbul/sbp10344.
LatasterT,vanOsJ,DrukkerM,HenquetC,FeronF,GuntherN,Myin-GermeysI(2006)Childhoodvictimisationanddevelop-mentalexpressionofnon-clinicaldelusionalideationandhallu-cinatoryexperiences:victimisationandnon-clinicalpsychoticexperiences.
SocPsychiatryPsychiatrEpidemiol41(6):423–428.
doi:10.
1007/s00127-006-0060-445.
LagdonS,ArmourC,StringerM(2014)Adultexperienceofmentalhealthoutcomesasaresultofintimatepartnerviolencevictimisation:asystematicreview.
EurJPsychotraumatol.
doi:10.
3402/ejpt.
v5.
24794SocPsychiatryPsychiatrEpidemiol(2017)52:1363–1374137312346.
BurnamMA,SteinJA,GoldingJM,SiegelJM,SorensonSB,ForsytheAB,TellesCA(1988)Sexualassaultandmentaldis-ordersinacommunitypopulation.
JConsultClinPsychol56(6):843–850.
doi:10.
1037/0022-006X.
56.
6.
84347.
MueserKT,RosenbergSD,GoodmanLA,TrumbettaSL(2002)Trauma,PTSD,andthecourseofseverementalillness:aninteractivemodel.
SchizophrRes53(1–2):123–143.
doi:10.
1016/S0920-9964(01)00173-648.
LoganT(2006)Womenandvictimization:contributingfactors,interventions,andimplications.
AmericanPsychologicalAssociation49.
KesslerRC,McLaughlinKA,GreenJG,GruberMJ,SampsonNA,ZaslavskyAM,Aguilar-GaxiolaS,AlhamzawiAO,AlonsoJ,AngermeyerM,BenjetC,BrometE,ChatterjiS,deGirolamoG,DemyttenaereK,FayyadJ,FlorescuS,GalG,GurejeO,HaroJM,HuCY,KaramEG,KawakamiN,LeeS,LepineJP,OrmelJ,Posada-VillaJ,SagarR,TsangA,UstunTB,VassilevS,VianaMC,WilliamsDR(2010)Childhoodadversitiesandadultpsy-chopathologyintheWHOWorldMentalHealthSurveys.
BrJPsychiatry197(5):378–385.
doi:10.
1192/bjp.
bp.
110.
08049950.
KelleherI,KeeleyH,CorcoranP,LynchF,FitzpatrickC,DevlinN,MolloyC,RoddyS,ClarkeMC,HarleyM,ArseneaultL,WassermanC,CarliV,SarchiaponeM,HovenC,WassermanD,CannonM(2012)Clinicopathologicalsignicanceofpsychoticexperiencesinnon-psychoticyoungpeople:evidencefromfourpopulation-basedstudies.
BrJPsychiatry201(1):26–32.
doi:10.
1192/bjp.
bp.
111.
10154351.
WigmanJT,vanNieropM,VolleberghWA,LiebR,Beesdo-BaumK,WittchenHU,vanOsJ(2012)Evidencethatpsychoticsymptomsareprevalentindisordersofanxietyanddepression,impactingonillnessonset,risk,andseverity-implicationsfordiagnosisandultra-highriskresearch.
SchizophrBull38(2):247–257.
doi:10.
1093/schbul/sbr19652.
WigmanJT,vanOsJ,AbidiL,HuibersMJ,RoelofsJ,ArntzA,KelleherI,PeetersFP(2014)Subclinicalpsychoticexperiencesandbipolarspectrumfeaturesindepression:associationwithoutcomeofpsychotherapy.
PsycholMed44(2):325–336.
doi:10.
1017/S003329171300087153.
PerlisRH,UherR,OstacherM,GoldbergJF,TrivediMH,RushAJ,FavaM(2011)Associationbetweenbipolarspectrumfea-turesandtreatmentoutcomesinoutpatientswithmajordepres-sivedisorder.
ArchGenPsychiatry68(4):351–360.
doi:10.
1001/archgenpsychiatry.
2010.
17954.
GuloksuzS,vanNieropM,LiebR,vanWinkelR,WittchenHU,vanOsJ(2015)Evidencethatthepresenceofpsychosisinnon-psychoticdisorderisenvironment-dependentandmediatedbyseverityofnon-psychoticpsychopathology.
PsycholMed45(11):2389–2401.
doi:10.
1017/S003329171500038055.
GalletlyC,VanHooffM,McFarlaneA(2011)Psychoticsymptomsinyoungadultsexposedtochildhoodtrauma–a20yearfollow-upstudy.
SchizophrRes127(1–3):76–82.
doi:10.
1016/j.
schres.
2010.
12.
01056.
RubyE,PolitoS,McMahonK,GorovitzM,CorcoranC,MalaspinaD(2014)PathwaysAssociatingChildhoodTraumatotheNeurobiologyofSchizophrenia.
FrontPsycholBehavSci3(1):1–1757.
FreemanD,GaretyP(2014)Advancesinunderstandingandtreatingpersecutorydelusions:areview.
SocPsychiatryPsychiatrEpidemiol49(8):1179–1189.
doi:10.
1007/s00127-014-0928-758.
LinscottRJ,vanOsJ(2013)Anupdatedandconservativesys-tematicreviewandmeta-analysisofepidemiologicalevidenceonpsychoticexperiencesinchildrenandadults:onthepathwayfrompronenesstopersistencetodimensionalexpressionacrossmentaldisorders.
PsycholMed43(6):1133–1149.
doi:10.
1017/S003329171200162659.
YoshihamaM,GillespieBW(2002)Ageadjustmentandrecallbiasintheanalysisofdomesticviolencedata:methodologicalimprovementsthroughtheapplicationofsurvivalanalysismethods.
JournalofFamilyViolence17(3):199–221.
doi:10.
1023/a:101600121118260.
WidomCS,MorrisS(1997)Accuracyofadultrecollectionsofchildhoodvictimization,Part2:childhoodsexualabuse.
PsycholAssess9(1):3461.
WidomCS,ShepardRL(1996)Accuracyofadultrecollectionsofchildhoodvictimization:part1.
Childhoodphysicalabuse.
PsycholAssess8(4):412–421.
doi:10.
1037/1040-3590.
8.
4.
41262.
deGraafR,vanDorsselaerS,TuithofM,TenHaveM(2013)Sociodemographicandpsychiatricpredictorsofattritioninaprospectivepsychiatricepidemiologicalstudyamongthegeneralpopulation.
ResultoftheNetherlandsMentalHealthSurveyandIncidenceStudy-2.
ComprPsychiatry54(8):1131–1139.
doi:10.
1016/j.
comppsych.
2013.
05.
01263.
PalmA,DanielssonI,SkalkidouA,OlofssonN,HogbergU(2016)Violencevictimisation-awatershedforyoungwomen'smentalandphysicalhealth.
EurJPublHealth26(5):861–867.
doi:10.
1093/eurpub/ckv23464.
HodginsS,LincolnT,MakT(2009)Experiencesofvictimisationanddepressionareassociatedwithcommunityfunctioningamongmenwithschizophrenia.
SocPsychiatryPsychiatrEpi-demiol44(6):448–457.
doi:10.
1007/s00127-008-0460-865.
LysakerPH,MeyerPS,EvansJD,ClementsCA,MarksKA(2001)Childhoodsexualtraumaandpsychosocialfunctioninginadultswithschizophrenia.
PsychiatrServ52(11):1485–1488.
doi:10.
1176/appi.
ps.
52.
11.
148566.
NewmanJM,TurnbullA,BermanBA,RodriguesS,SerperMR(2010)Impactoftraumaticandviolentvictimizationexperiencesinindividualswithschizophreniaandschizoaffectivedisorder.
JNervMentDis198(10):708–714.
doi:10.
1097/NMD.
0b013e3181f49bf11374SocPsychiatryPsychiatrEpidemiol(2017)52:1363–1374123