Development and psychometric assessment of a measure of internalized HIV stigma

Sayles的艾滋病患者内化歧视量表

AIDSBehav(2008)12:748–758DOI10.1007/s10461-008-9375-3

ORIGINALPAPER

DevelopmentandPsychometricAssessmentofaMultidimensionalMeasureofInternalizedHIVStigmainaSampleofHIV-positiveAdults

JenniferN.SaylesÆRonD.HaysÆCatherineA.SarkisianÆAnishP.MahajanÆKarenL.SpritzerÆWilliamE.Cunningham

Publishedonline:4April2008

ÓSpringerScience+BusinessMedia,LLC2008

AbstractThereisaneedforapsychometricallysoundmeasureofthestigmaexperiencedbydiversepersonslivingwithHIV/AIDS(PLHA).ThegoalofthisstudywastodevelopandevaluateamultidimentionalmeasureofinternalizedHIVstigmathatcapturesstigmarelatedtotreatmentandotheraspectsofthediseaseamongsoci-odemographicallydiversePLHA.Wedevelopeda28-itemmeasureofinternalizedHIVstigmacomposedoffourscalesbasedonpreviousqualitativework.Internalcon-sistencyreliabilityestimatesinasampleof202PLHAwas0.93fortheoverallmeasure,andexceeded0.85forthreeofthefourstigmascales.Itemsdiscriminatedwellacrossscales,andcorrelationsofthescaleswithshame,socialsupport,andmentalhealthsupportedconstructvalidity.Thismeasureshouldproveusefultoinvestigatorsexam-iningintheroleofstigmainHIVtreatmentandhealth

J.N.Sayles(&)ÁR.D.HaysÁK.L.SpritzerÁW.E.Cunningham

DivisionofGeneralInternalMedicineandHealthServicesResearch,UCLA,911BroxtonAve.,LosAngeles,CA90024,USA

e-mail:jsayles@mednet.ucla.edu

R.D.Hays

RANDCorporation,SantaMonica,CA,USAC.A.Sarkisian

DivisionofGeriatrics,UCLA,LosAngeles,CA,USAA.P.Mahajan

UCLA/RobertWoodJohnsonClinicalScholarsProgram,LosAngeles,CA,USA

W.E.Cunningham

DepartmentofHealthServices,UCLASchoolofPublicHealth,LosAngeles,CA,USA

123

outcomes,andevaluatinginterventionsdesignedtomiti-gatetheimpactsofstigmaonPLHA.

KeywordsHIV-relatedstigmaÁMeasurementÁAfricanAmericanÁWomenÁHealthcare

Introduction

HIVstigmanegativelyimpactspersonslivingwithHIV/AIDS(PLHA).HighperceivedstigmaisassociatedwithmoredepressiveandotherHIV-relatedsymptoms(Sowelletal.1997;Vanableetal.2006),lowerlevelsofantiret-roviraltherapyadherence(Rintamakietal.2006;Stirrattetal.2006;Vanableetal.2006),andpooraccesstomedicalcare(Kinsleretal.2007)forPLHA.ThetwomostcommonlyusedmeasuresofstigmainPLHA(Bergeretal.2001;Sowelletal.1997)werebothdevelopedbeforetheeraofHAART.Asaresult,theydonotadequatelycapturestigmaanddisclosureconcernsthatariseinthecontextofHIVmedicationuse,orlivingwithHIVasachronicdis-ease.Additionally,thesemeasuresdonotaddressstigmainthehealthcaresetting,includingproviderattitudestowardPLHA,concernsaboutcon?dentialityofHIVstatus,andunwanteddisclosurecausedbyseekingmedicalcare.ThesedimensionsofstigmaandHIVtreatmenthavebeenidenti?edtobehighlysalienttopersonslivingwithHIV(CarrandGramling2004;Kinsleretal.2007;Rintamakietal.2006).

Theconceptofstigmaiscomplexandisrootedintheconceptofdeviancefromthevaluesandsocialnormsofacommunity(Goffman1963).ItfollowsthattheperceptionandexperienceofstigmaforaPLHAmaydifferamongsocialgroups—includingbygender,race,sexualorienta-tion,druguse,andsocioeconomicstatus(SES)(Herekand

Sayles的艾滋病患者内化歧视量表

AIDSBehav(2008)12:748–758Capitanio1998;Hereketal.1998;LinkandPhelan2001).Therefore,itisimportanttodevelopandevaluatemeasuresofstigmaamonggroupswhosedemographicsre?ectthediversityoftheHIVepidemic.Forexample,studiesofwomenandminoritieslivingwithHIVrevealthatmanymothersandfathersexperiencestigmaintheformofste-reotypesandblamedirectedtowardHIV-positiveparents(Bogartetal.2008;Murphyetal.2006;SandelowskiandBarroso2003).However,thedimensionsofstigmathatmaybeexperiencedasaparenthavebeenleftoutofmeasurestodate.Toeffectivelystudy,measure,andulti-matelyreducethestigmaforPLHA,itiscriticaltocapturethefullrangeofperceptionsandexperiencesoftheprocessofstigmainagivenculturalcontextandcommunity(Nyblade2006;ParkerandAggleton2003).

ThemostcommonlyusedstigmameasureforPLHAisa40-itemscaledevelopedandtestedin1994inasampleofalmostentirelywhitemales(Bergeretal.2001).Theinstrument(andvariousshortforms)hasbeenfoundtohavegoodinternalconsistencyreliabilityinAfricanAmericans(Raoetal.2008;Wrightetal.2007)andaruralpopulationofpredominantlywhitemenlivinginNewEngland(Bunnetal.2007).However,thedevelopmentoftheinstrumentdidnotincludedirectinputfromHIV-positivewomenandmen,suchasinterviewsorfocusgroupstoinformitemcontentorformalcognitiveinterviewing.Additionally,inbothcon?rmatoryandexploratoryfactoranalysisthemeasurehasbeenshowntolackitemdiscriminationacrossscaleswithmanyitemsloadingontomultiplescales(Bergeretal.2001;Raoetal.2008).AmeasurecapturingthemultifacetednatureofstigmaforadiversepopulationofPLHAcanaddtoourunderstandingoftherelationshipbetweenstigmaandhealthoutcomes,andassistinevalu-atingtheeffectivenessofinterventionstomitigatethenegativeeffectsofstigmaforPLHA.

Studiesandframeworksofstigmahaveoftendrawnthedistinctionbetween‘‘enacted’’stigma,suchasactsofdiscrimination,and‘‘perceived’’or‘‘internalized’’stigma,whichcapturesstigmaasperceivedandexperiencedbythepersonbeingstigmatized(Jacoby1994;Scambler1998;Simbayietal.2007).Inourconceptualizationofthesocialprocessofstigma,internalizedstigmaoccursasanindi-vidualinternalizesculturalnormsandnarrativesthatidentifyhim/herasamemberofadeviantgroup,andassumesa‘‘spoiledidentity’’asdescribedintheworkofGoffman(1963).ThegoalofthisstudywastodevelopandevaluateamultidimensionalmeasureofinternalizedHIVstigmathatadequatelycapturesstigmarelatedtotreatmentandotheraspectsofHIVamongsociodemographicallydiversePLHA.Toaccomplishthisgoalwebuiltonpre-viousfocusgroupsandinterviewswithPLHAandaliteraturereviewtoinformthedevelopmentofa28-itemmeasureofstigma.Inthisreportwedescribethe

749

developmentandpsychometricpropertiesofthenewinstrumentinasampleof202HIV-positiveadultsinwhichwomen,racial/ethnicminorities,andpeoplewithlimitedincomeandeducationwererepresented.

Methods

SurveyDevelopment

Toidentifymeaningfulcontentforameasureofinternal-izedstigmathatincludesexperiencesofdiversePLHA,weconductedfourfocusgroupswithHIV-positivewomen(n=26)andthreefocusgroupswithHIV-positivemen(n=22)in2005.Fifty-fourpercentofthefocusgroupparticipantswerefemale,56%wereAfricanAmerican,21%wereLatino/a,over70%reportedahighschooleducationorless,and50%reportedbeinghomelessatpresentorinthepast.

Focusgroupdetailsandresultsarereportedelsewhere(Saylesetal.2007).Qualitativeanalysisofthefocusgrouptranscriptsidenti?edfourhigherlevelconceptualdomainsofInternalizedHIVstigma:(1)confrontingblameandstereotypesofHIV;(2)encountering‘‘fearofcontagion;’’(3)negotiatingdisclosureofastigmatizedrole;and(4)renegotiatingsocialrelationships.Weusedtheconceptsderivedfromthesefocusgroupsandthepeer-reviewedliteraturetoconstructanewmeasureofinternalizedHIVstigma.Theitemsinthenewmeasureweredesignedtocapturethecorecontentofthefourdomainsfromourqualitativeanalysis,includingstigmarelatedtoHIVtreatmentandparenting.Wheneverpossibleweretainedtheexactlanguageparticipantsusedtoconveyperceptionsorexperiencesofstigma.

Apoolof78itemswereinitiallydraftedandthenpre-testedinEnglishbythe?rstauthorintencognitiveinter-views(Willis1994)withHIV-positivewomen(n=5)andmen(n=5).Thecognitiveinterviewswereconductedtoidentifyerrorsandtestthelogic,clarity,andacceptabilityofindividualitemsandoftheinstrumentasawhole.Interviewparticipantswereaskedaboutrelativeeaseofquestions,wordingoftheresponsecategories,andinter-pretationofspeci?cwordsandphrasestodetermineiftheirmeaningwasthesametotherespondentandtheitemdesigners.Basedontheresultsofcognitiveinterviews,themeasurewasmodi?edinaniterativefashionandcon-structedtoincludeaseriesofstatementsdescribingperceptionsandexperiencesofinternalizedHIVstigmafollowedbya5-pointcategoricalresponsescale(noneofthetime,alittlebitofthetime,someofthetime,mostofthetime,allofthetime).Poorlywordedandambiguousitemsweretheneliminated,resultingina52-itemversionofthestigmameasureusedinthesurveydatacollection

123

Sayles的艾滋病患者内化歧视量表

750describedbelow.Ofthese52items,11itemsdescribedstigmarelatedtoHIVtreatmentand7itemsdescribedstigmainthecontextofparenting.ApproximatelyhalfoftheitemsaskedparticipantsabouttheirownexperiencesandperceptionsofHIV-relatedstigma,whiletheotherhalfoftheitemsaskedaboutstigmaexperiencedbyHIV-positivepeopleingeneral.SurveyParticipants

WecollectedsurveydatafromaconveniencesampleofHIV-positiveadultsover18yearsofageinJanuarythroughMayof2007.Werecruitedparticipantsfrom?vecommunity-basedorganizationsprovidingoutreachandsocialservicestoHIV-positivewomenandmen,aswellasfromtwoHIVspecialtyclinicsinalargeU.S.city.Becauseourgoalwastotestthenewstigmameasureinasamplethatincludedapproximatelyequalnumbersofwomenandmen,andwasdiversewithrespecttorace/ethnicityandsocio-economicstatus,weintentionallyrecruitedfromseveralsitesthatprovideservicestolargenumbersofHIV-positivewomenandminorities,manyofwhomhavelimitededucationandincome.WealsorecruitedparticipantsfromonesitethatservesprimarilymaleclientswithprivateinsuranceorMedicare.Partneringwithbothcommunityorganizationsandclinicalsitesallowedustoincludeindividualsinourstudywhoaccessthehealthcaresystemandthosewhomaynot.Theinclusioncriteriaforthestudywereasfollows:ageover18years,HIV-positive,abilitytoanswerquestionsinEnglish,nothavingparticipatedintheearlierfocusgroupsweconducted,andabilitytoprovideinformedconsent.Participantswererecruitedforthestudythrough?http://wendang.chazidian.coma-nizationandclinicstaffalsoreferredinterestedparticipantstothestudy.Atrainedresearchassociatescreenedpotentialsubjects,andwheneligible,invitedthemtocompletea154-itemanonymousself-administeredquestionnairethatincludedthe52-itemstigmameasureinaprivateareaoftheorganizationorclinic.SurveyMeasures

Therevised52-itemversionofthestigmameasurewasadministeredto202HIV-positiveadults.Tomeasurepotentialdemographic,medical,andpsychosocialcorre-latesofinternalizedHIVstigma,wealsoincludedthefollowingmeasures:(1)sociodemographiccharacteristicsincludinggender,age,race/ethnicity,income,education,HIVexposurecategory(Berryetal.2002);(2)clinicaldataincludingCD4count,viralload,historyofAIDSdiagnosis,andtimesinceHIVdiagnosis(Berryetal.2002);(3)the5-itemStateShameScale(Marschalletal.1994)measuring

123

AIDSBehav(2008)12:748–758

generalpsychologicalshame;(4)5-itemsfromtheFleish-mansocialsupportscale(Fleishmanetal.2000);(5)a6-itemscalemeasuringoverallaccesstomedicalcare(Cunninghametal.1999;Cunninghametal.1995);(6)8individualitemsmeasuringutilizationofHIVcareandmentalhealthcare(Katzetal.2001;Shapiroetal.1999);(7)3individualitemsmeasuringantiretroviraltherapyuseandadherence(Cunninghametal.2000);and(8)the12-itemMedicalOutcomesStudyShortForm(SF-12),whichwasusedtocalculatephysical(PCS)andmentalhealthcomposite(MCS)scores(Haysetal.2000a;Wareetal.1996).DataAnalysis

Weconductedexploratoryfactoranalysisonthe52-itemstigmascaletoexaminehowtheunderlyingfactorstruc-turecomparedwiththefourhypothesizeddomainsfromourpriorqualitativework.Weusedeigenvaluesgreaterthan1.0fromaprincipalcomponentssolution,andthescreeplotfromaprincipalfactorsolutionwithsquaredmultiplecorrelationsascommunalities(Cattell1966)toprovideanindicationofthenumberofunderlyingfactors.Afterdeterminingthenumberoffactorstorotate,weperformedanobliquefactorrotation(PROMAX)toallowforinter-factorcorrelations.Wetheneliminated8ofthe52itemswithlowfactorloadings(standardregressioncoef-?cients\0.30).Nextweevaluatedtheextenttowhichthe46retaineditemscorrelatedmorehighlywiththeirhypothesizedscales(correctedforitemoverlap)thantheydidwithotherscalesusingmultitraitscalinganalysis(HaysandHayashi1990).Weeliminated12itemswithpooritemdiscrimination,andanadditional6itemsthatoverlappedwithitemshavinghigheritem-scalecorrelationsandbetteritemdiscrimination.Thisresultedin28-itemsde?ningfourmulti-itemscalessuggestedbythefactoranalysis.

The28-itemscaleretained6oftheoriginal11itemsdescribingstigmainthecontextofHIVtreatment(indomainsofstereotypes,disclosure,andsocialrelation-ships).Thisincludeditemsaboutstigmafrommedicalproviders(items10,18,19),disclosureconcernsintheHIVclinicsetting(items13,16),unwanteddisclosuresresultingfromphysicalchanges(lipdystrophy/lipoatrophy)duetoARTuse(item14)andthestigmaofHIVasuniquefromthestigmaofotherillnesses(item1).Additionally,3oftheoriginal7itemsonparentingwereretainedinthedomainofstereotypes,includingattitudesaboutPLHAdeservingchildren(item7),adoptingchildren(item8),andaPLHA’sabilitytobeagoodparent(item12).Thefourotheritemsfromthe52-iteminstrumentdescribingdisclosureconcernsandsocialrelationshipstigmaexperiencedbyHIV-positiveparentsdidnotdemonstratesuf?cientfactorloadinganditemdiscriminationtoberetainedinthe?nal28-itemscale.

Sayles的艾滋病患者内化歧视量表

AIDSBehav(2008)12:748–758Meanscoresforthe28-itemstigmascaleweretrans-formedlinearlytoa0–100range,withlowerscoresre?ectingfewerperceptionsandexperiencesofinternal-izedHIVstigma,andhigherscoresre?ectinggreaterlevelsofstigma.Wecalculatedthemean,standarddeviation,range,skewedness,andpercentageofparticipantsscoringtheminimum(?oor)andmaximum(ceiling)foreachstigmaitemandsubscale.Theoverall28-itemmeasurewasconstructedasanaverageofthe4scalescoresratherthanasanaverageofthe28itemsindividually,soasnottogivegreaterweighttodomainsthathavemoreitems.

Internalconsistencyreliabilitywasestimatedforeachofthe4subscalesandanoverallscaleusingCronbach’scoef?cientalpha(Cronbach1951).Inaddition,weevalu-ateditemdiscriminationacrossscales,calculatingthepercentageoftimesthatitemsinthesubscalecorrelatedsigni?cantlyhigher(atleasttwostandarderrorsofcorre-lationhigher)withthehypothesizedsubscale(correctingforoverlap)comparedwithotherscales.

Weusedanalysisofvariancetoevaluateifthestigmascaleswereassociatedwithsociodemographicandclinicalcharacteristics.Weexaminedhowmeanstigmascalescoresvariedbygender(male,female,transgender),age(18–35years,36–49years,50yearsorolder),race/eth-nicity(White,AfricanAmerican,Latino/a,otherrace/ethnicity),income(aboveorbelowfederalpovertylevel),primarylanguage(English,anotherlanguage),education(lessthanhighschool,highschooldiploma,collegediploma),relationshipstatus(marriedorinacommittedrelationshipornot),HIVriskcategory(MSM,heterosex-ual,IDU),everdiagnosedwithAIDS(Yes/No),currentCD4cellcount(\200cells/ul,[200cells/ul)andcurrentantiretroviraltherapyuse(Yes/No).

ContentandConstructValidityofStigmaMeasureContentvalidityassesseshowwellameasurerepresentstheconstructofinterest(Haysetal.2000b).Themethodologyusedinthefocusgroupsandcognitiveinterviewphaseofthisstudywasselectedtomaximizethecontentvalidityoftheitemsdevelopedforthemeasure.Toassessconstructvalidityoftheinstrumentweexaminedtheassociationsbetweenthe28-iteminternalizedstigmascaleandcon-structsweexpectedtobecorrelatedwithstigmabasedontheliteratureandourpriorwork.Wehypothesizedthatourmeasureofinternalizedstigmawouldbestronglyandpositivelycorrelated(r=0.5–0.7)withfeelingsofshame.Shameisonepsychologicalexperienceanindividualmayhaveasaresultofinternalizedstigma,andseveralstudiesintheliteraturehavedemonstratedtheseconstructstobeseparate,butrelated(Cunninghametal.2002;Duffy2005;Fortenberryetal.2002).StudiesofHIV-positiveadultslivingintheUShavealsodemonstratedthatstigmais

751

associatedwithpoorsocialsupportanddepressivesymp-toms(Bairanetal.2007;CarrandGramling2004;Sowelletal.1997;Vanableetal.2006).Thus,wehypothesizedinternalizedstigmawouldalsobemoderatelynegativelycorrelated(r=-0.3to-0.5)withsocialsupportandmentalhealth.Finally,wearenotawareofanyquantitativestudiesdirectlylinkingstigmaandphysicalhealth.Wethereforehypothesizedinternalizedstigmawouldbeweaklycorrelatedwithphysicalfunctioninginoursample.

Results

TwohundredandtwoHIV-positiveparticipantscompletedthesurvey.Refusalratesrangedfrom10%to30%dependingonthesiteofrecruitmentforthestudy.Fiftypercentofthesamplewasfemaleandthemeanagewas43years(range18–76years).Fifty-sixpercentofpartici-pantswereAfricanAmerican,28%wereWhite,10%wereLatino/aand6%wereAsianAmerican,NativeAmerican,oranotherrace/ethnicity.Overhalfthesamplereportedincomebelowfederalpovertylevel($1140permonthforfamilyoftwo),and70%reportedahighschooleducationorless.Sixty-ninepercentofparticipantsidenti?edthem-selvesasheterosexual,22%reportedintravenousdruguse(IDU)asariskfactorforHIV,andover70%reportedbeingcurrentlyonantiretroviraltherapy.Missingratesbystigmaitemsrangedfrom0%to3%,andnoneofthescaleshadmorethan25%ofitemsmissing.Theentiresurveytookparticipants25mintocompleteonaverage,withaFlesh-Kincaidreadinglevelofgrade6.4(MicrosoftWord2000,Redmond,WA)

Eigenvaluesandthescreeplotindicatedfourexistingunderlyingfactors.Afour-factorobliquerotationdemon-stratedthatthreeofthefactorsconsistedofitemsrepresentingthefourhypothesizeddomains.Itemsgener-atedfromthehypothesizeddomainof‘‘stereotypes’’consistentlyloadedonfactor1,itemsfromthedomainof‘‘disclosureconcerns’’consistentlyloadedonfactor2,anditemsfromthedomainof‘‘socialrelationships’’consis-tentlyloadedonfactor3.Itemsgeneratedfromthefourthdomainof‘‘fearofcontagion’’collapsedwith‘‘socialrelationships,’’loadingonfactor3.Finally,anew(nothypothesized)domainemergedinfactor4,consistingofitemsrelatedtolevelofpersonalcomfortwithone’sHIVdiagnosis,whichwerefertoas‘‘self-acceptance’’.Theitemsre?ectarangeofexperiencesofself-acceptance.‘‘IfeelashamedtotellotherpeoplethatIhaveHIV’’re?ectsalowlevelofself-acceptance,theitem‘‘MyfamilyiscomfortabletalkingwithmeaboutHIV’’suggestsalevelofacceptancewhereanindividualfeelscomfortablewithhis/herdiagnosisaroundatrustedgroupofpeople,while‘‘IamcomfortabletellinganyoneIknow’’representsthehigh

123

Sayles的艾滋病患者内化歧视量表

752AIDSBehav(2008)12:748–758

endoftheself-acceptancetrajectory.Theconceptofself-acceptanceiscentraltotheframeworkofstigmawehavedevelopedandpublishedfromourpreviouslydescribedqualitativedata.

InTable1wereportthestandardizedregressioncoef-?cientsinthefour-factorobliquesolutionforthe28stigmaitems.Standardizedregressioncoef?cientsrepresenttheindividualandnon-redundantcontributionthateachfactorismaking.Allitemsloadedconsistentlyononefactorandnoitemsloaded[0.30onmorethanonefactororsubscale.Twelveitemsloadedonfactor1(stereotypes),witharangeof0.50–0.79;?veitemsloadedonfactor2(disclosureconcerns)witharangeof0.60–0.71;sevenitemsloadedonfactor3(socialrelationships)witharangeof0.46–0.73;andfouritemsloadedonfactor4(self-acceptance)witharangeof0.37–0.72.ItemsaddressingtheinternalizedstigmaofHIVasachronicillness(item1),disclosureconcernsforcareseekingandHIVmedications(items13and14),andperceptions/experiencesofstigmafromhealthcareproviders(items18,19)allloadedmorehighlythananyotheritemsontheirrespectivefactors(stereotypes,disclosureconcerns,socialrelationships).Itemsdescribing

Table1Four-factorspatternmatrix(standardizedregressioncoef?cients)for28?nalinternalizedstigmascaleitemsItemsa

Mean

Score(SD)

Factor1

2

3

4

Factor1=Stereotypes1.2.3.4.5.6.7.8.9.10.11.12.13.14.15.16.17.18.19.20.21.22.23.24.25.26.27.28.

a

HIVisdifferentthanotherdiseaseslikecancerbecausepeoplewithHIVarejudgedPeopleassumeIhavedonesomethingbadtogetHIVSocietylooksdownonpeoplewhohaveHIV

PeoplethinkthatifyouhaveHIVthenyougotwhatyoudeservePeopleblamemeforhavingHIV

PeopleassumeIsleptaroundbecauseIhaveHIV

PeoplethinkthatifyouhaveHIVyoudonotdeservetohavechildrenPeopleareafraidtoletsomeonewithHIVadoptachildPeoplethinkIamabadpersonbecauseIhaveHIV

MedicalprovidersassumepeoplewithHIVsleeparoundPeoplelosetheirjobsbecausetheyhaveHIV

Peoplethinkyoucan’tbeagoodparentifyouhaveHIV

IamconcernedifIgototheHIVclinicsomeoneIknowmightseemeIamconcernedifIhavephysicalchangesfromtheHIVmedicinespeoplewillknowIhaveHIV

IamconcernedifIgotoanAIDSorganizationsomeoneIknowmightseemeIamconcernedpeoplewill?ndoutIhaveHIVbylookingatmymedicalpaperworkIamconcernedthatifIamsickpeopleIknowwill?ndoutaboutmyHIVNursesanddoctorstreatpeoplewhohaveHIVasiftheyarecontagiousNursesanddoctorsdislikecaringforpatientswithHIVIfeelabandonedbyfamilymembersbecauseIhaveHIVPeopletreatmeaslessthanhumannowthatIhaveHIVPeopleavoidmebecauseIhaveHIV

PeopleIamclosetoareafraidtheywillcatchHIVfrommeIfeellikeIamanoutsiderbecauseIhaveHIVIfeelashamedtotellotherpeoplethatIhaveHIVIamcomfortabletellingeveryoneIknowthatIhaveHIVMyfamilyiscomfortabletalkingaboutmyHIV

b

b

66(29)53(32)59(28)43(30)43(35)54(33)50(34)57(29)34(32)38(31)46(26)42(31)28(33)33(33)30(34)32(35)38(35)30(30)28(28)24(33)29(31)24(30)26(32)41(35)45(36)61(35)55(33)56(33)

0.790.700.680.680.660.620.590.560.540.530.530.500.060.060.040.080.010.040.250.060.010.030.240.070.060.030.230.18

0.040.050.120.030.050.080.120.060.030.020.240.020.710.710.700.610.600.080.240.110.080.130.050.100.040.010.030.09

0.170.050.200.010.120.010.000.040.220.160.080.110.030.130.060.150.030.730.690.620.610.600.530.460.180.140.140.11

0.010.220.200.180.150.220.180.200.180.130.220.090.200.030.250.110.210.000.060.130.250.160.100.360.720.600.400.37

Factor2=Disclosureconcerns

Factor3=Socialrelationships

Factor4=Self-acceptance

ItisimportantforapersontokeepHIVasecretfromco-workers

Responsestoeachitemareona5-pointcategoricalresponsescale(noneofthetime,alittleofthetime,someofthetime,mostofthetime,orallofthetime)

Responsescalevaluesreversedforscoring

b

123