ConnectednessandSuicidePreventionin - PDF document

ConnectednessandSuicidePreventionin Adolescents:PathwaysandImplications J ANIS W HITLOCK ,MPH,P H D,P ETER A.W YMAN ,P H D, AND S ARAH R.M OORE ,MA Adolescentsuicideisamajorpublichealthconcern.Stressingtheneed forpublichealth – basedsolutions,theCentersforDiseaseControlandPreven- tionidentiÞedÒconnectednessÓasonemeansofpursuingthisagenda.To advancethiseffortinsuicidepreventionwithadolescents,(1)consistenciesand variationintheliteratureovertlylinkingconnectednesstosuicidethoughtsand behaviors(STB)arereviewed,(2)threemorespeciÞcmechanisticpathwaysare proposedwherebyconnectednessmayinßuenceSTB,and(3)severalimplica- tionsrelatedtouseofconnectednessasapublichealthframeworkforadoles- centsuicidepreventionandinterventionareoutlined. THEISSUE:UNDERSTANDING ANDINTERVENINGIN ADOLESCENTSUICIDE Adolescentsuicideisamajorpublichealth concern(U.S.DepartmentofHealthand HumanServices[USHHS],2012).Cur- rentlythethirdleadingcauseofdeathfor youthaged1to19,accountingfor10%of alldeathsinthisagegroup(Centersfor DiseaseControlandPrevention[CDC]/ NCHS,2011),suicideratesincreaseten- foldfrompreadolescencetoearlyadulthood (Goldsmith&IOM,2002).Upsharply frompreviousyears,suicidalthoughtsand behaviors(STB)arethethirdleadingcause ofdeathamongthose10to14yearsold andthesecondleadingcauseofdeath 2011).Inadditiontotheemotionaland psychosocialmorbidity,thehighÞscaland societalcostsassociatedwithmedicalcare, lostproductivity,andsecondarydistress amongfamilymembersandothersrender adolescentSTBasigniÞcantpublichealth problem. Toreducethepublicimpactof suicide,boththeSurgeonGeneralofthe UnitedStatesandtheCDChaveidentiÞed reducingsuicideasanimportantpublic healthpriority.Stressingtheneedforpublic- healthÐbasedsolutionsemphasizingpreven- tion,earlyintervention,andpolicy,boththe 2001and2012NationalStrategyforSuicide PreventionidentiÞedenhancingÒconnected- nessÓasonemeansthroughwhichthis agendashouldbepursued(USHHS,2001; USHHS&NationalActionAlliancefor SuicidePrevention,2012).TheCDC(2008) elaboratedonthisconceptbystatingÒcon- nectednessisacommonthreadthatweaves J ANIS W HITLOCK ,BronfenbrennerCenter forTranslationalResearch,CornellUniversity, Ithaca,NY,USA,andDepartmentofHuman Development,CornellUniversity,Ithaca,NY, USA;P ETER A.W YMAN ,UniversityofRochester SchoolofMedicine&Dentistry,Rochester,NY, USA;S ARAH R.M OORE ,DepartmentofHuman Development,CornellUniversity,Ithaca,NY, USA. Fundingforthepaperwasprovidedbya grantfromtheCentersforDiseaseControland Prevention,R01CE001882-01(KennethCon- ner,PI).Dr.Connerprovidedinputwhenthe paperwasbeingconceptualizedandcommented onanearlierdraft. lock,BronfenbrennerCenterforTranslational Research(BCTR),CornellUniversity,Ithaca, NY14853;E-mail:jlw43@cornell.edu SuicideandLife-ThreateningBehavior1 © 2014TheAmericanAssociationofSuicidology DOI:10.1111/sltb.12071 togethermanyoftheinßuencesofsuicidalbehaviorandhasdirectrelevancetopreven-tionÓ(p.3).TheCDCÕsfocusonenhancingcon-nectednessasabroadorientingframeworkforsuicidepreventionstemsfromafunda-mentalunderstandingabouttheimportanceofsocialconnectioninhumanwell-beingandhaslaidthegroundworkforresearch,intervention,andpolicyeffortscapableofaccommodatingawidearrayofapproaches.Asothershavenoted(Barber&Schlu-terman,2008),althoughsuchbreadthisusefulfromapracticalperspective,opera-tionalizingandevaluatingeffortsfallingundersuchabroadumbrellawillrequireagreementabouthowtobestdeÞneandmeasureconnectednessandgreaterspeciÞc-ityabouthowconnectednessprotectsagainstSTBandpromoteswell-beingthroughoutthelifecourse.Toadvanceclarityoftheconnected-nessconceptinSTBpreventiontargetingadolescents,we(1)reviewhowÒconnected-nessÓisusedintheliteratureinrelationshiptoSTB,withparticularregardtoitsopera-tionalizationandinßuenceonSTB;(2)pro-posetwobroaddomains(subjectivestructural)andthreespeciÞcpathwayswherebyconnectednessmayinßuenceSTB;and(3)outlinetheimplicationsoftheseideasforresearchandpractice.DEFININGCONNECTEDNESSConceptionsofconnectednesscanbelinkedintheliteraturetoatleastninedis-tinctconceptualframeworksincluding,butnotlimitedto,attachmenttheory(Bowlby,1969),socialsupporttheory(Brown,Brady,Lent,Wolfert,&Hall,1987),bio-ecologicalmodelsofhumandevelopment(Bronfen-brenner&Morris,1998),resilienceframe-works(Werner&Smith,2001),stage-environmentÞttheories(Ecclesetal.,1993),socialdevelopmentandlearningtheories(Bandura,1997),andsocialcapitaltheories(Coleman,1988).Theconceptspansabroadnumberofdisciplinesandhasbeenusedtoexplaincomplexmechanismsgoverningthewaysocialinteraction,placement,andexchangeinßuenceindividuallevelexperi-ence,perception,andbehaviorwithinoracrosscontexts.ÒConnectednessÓisoftenusedinterchangeablywithcloselyrelatedconstructssuchasÒattachment,ÓÒbonding,ÓÒsocialintegration,ÓandÒsocialsupport.ÓIntheirreviewsoftheconnectednessconstruct,BarberandSchluterman(2008)andTownsendandMcWhirter(2005)high-lighttheconsiderablevariationinitsopera-tionalizationacrossempiricalstudies.Ingeneral,theyeachÞndthatdeÞnitionstendtoreferencethesubjectiveand/orstructuralfeaturesofsocialafÞliation.Withinthesub-jectivedomain,scholarsdescribeconnected-nessasasenseofinterpersonalclosenesswiththebroadersocialworldorwithindi-vidualscharacterizedbyfeelingssuchascaring,belonging,trust,value,andrespect(Barber,Stolz,&Olsen,2005;Lee&Rob-bins,1995;Resnick,Harris,&Blum,1993;Whitlock,2006),orasthedegreeofsatis-factionwithanenvironmentorrelationship(Eisenberg,Neumark-Sztainer,&Perry,2003).Withinthestructuraldomain,schol-arsfocusoncharacteristicssuchasnetworkdensity,strengthofsocialties,andsharingofresourcesbetweenindividualsand/ororganizationsandinstitutions(CDC,2008;Timpone,1998).Applicationsofconnected-nessspeciÞcallyinrelationshiptoSTBtendtospanbothstructuralandsubjectivedimensions.Forexample,sociologistEmileDurkheim(1897)arguedthatsuicideresultedfromlackofsocialconnectiontoothersandlowbehavioralcontrol.Theinterpersonaltheoryofsuicide(Joiner,2006)alsoreferstosocialconnectednessasacorecontributortoSTB,butgoesonestepfartherthanDurkheimbycodifyingconnectednessasanembodimentofboththestructuralideaofsocialintegrationandthesubjectiveexperienceofbelonging(VanOrdenetal.,2010).Notablylackinginthesereviews,however,aremorespeciÞcpostulationsconnectednessmightaffectyouthoutcomesor,morespeciÞcally,howitONNECTEDNESSAND mightconferprotectionforadolescentSTBrisk.ArticulationofthespeciÞcpathwaysofinßuenceisalogicalmeansofexploringtheutilityofconnectednessasaconstruct.HOWCONNECTEDNESSISCURRENTLYUSEDINRELATIONTOSTBInourreviewoftheliterature,weincludedstudiesthatexplicitlyidentiÞedconnectednessandsomeformofsuicide-relatedthoughtsorbehaviorsinadolescents20yearsofage)asaprincipleoutcomemeasure.Aninitialsearchforpeer-reviewedarticlesusingthekeywordsÒconnectednessÓandÒsuicideÓyielded112articlesfromPsy-chINFOandWebofSciencedatabases.Weselectedthe18thatfocusedonadolescents,clearlyidentiÞedconnectednessasacon-structofinterest,andmeasuredSTBasanoutcome.ResultsofthisreviewarepresentedinTable1.Studiesaregroupedaccordingtothethreeproximaladolescentcontexts:fam-ily,school,andpeergroups.AsevidentinTable1,operational-izedindicatorsofconnectednesstendedtobecontextspeciÞc,wereoftenusedanalo-gouslywithotherpreviouslydeÞnedcon-structs(e.g.,perceivedsocialsupport;bonding),andvariedwidelyintheproper-tiesmeasured.WhilenoneofthepapersincludedaformaldeÞnitionofconnected-ness,alloperationalizeditusingsubjectivemeasuresofafÞliation.Thetermconnected-wasmostfrequentlyoperationallydeÞnedandassessedbysubjectiveself-reportmeasuresof(1)perceivedwarmth,affection,attachment,andage-appropriatedevelopmentalsupport(suchasautonomyandsupervision);(2)perceivedbelongingorbondingwithinacollective,suchasfamily,schoolorpeergroup;and(3)perceivedavailabilityofinstrumentaloraffectiveresourcesorsupport.Althoughmoststudiesoperationalizedconnectednessasasingleindexappliedwithinonesocialcontextoracrossseveralcontextssimultaneously,morerecentstudiessuchasKaminskietal.(2010)havecapturedamultilayeredexperienceofconnectionacrosscontextsandexperiences.Overall,reviewedstudiesshowedsubstantialconsis-tencyinmeasuredassociationsbetweenindi-catorsofconnectedness,regardlessofhowitisdeÞned,andSTB.Themostfrequentlystudiedsettingwasthefamily,whereindica-torsofadolescentsÕconnectednesswereasso-ciatedwithdecreasedsuicideriskinthreelargelongitudinalstudiesandallcross-sec-tionalstudies(seeTable1),althoughnotwithoutcaveat.Forexample,ananalysisofa1-yearfollow-upoftheNationalLongitudi-nalStudyofAdolescentHealth(AddHealth)foundthatconnectednessindicatorswereinverselyassociatedwithsuicidalbehavioramongseventhtotwelfthgraders(Borowsky,Ireland,&Resnick,2001),butintheirstudyofformerpsychiatricpatients,Czyz,Liu,andKing(2012)foundeffectsonlyforsui-cideideationandnotattempts.Moreover,Kiddetal.(2006)foundthattherelationshipbetweenSTBandfamilyconnectednesswasparticularlypronouncedforboyswithposi-tiveschoolrelations,poorpeerrelations,andsuicideattempthistory.Similarly,resultsweremoderatedbysexintwoofthefamilyconnectednessstudiesreviewedbutinwaysthatfoundfamilyconnectednesstobeanSTBprotectivefactorforgirls,butnotforboysinhigh-riskcommunities(Logan,Cros-by,&Hamburger,2011)andwithsexualabusehistory(Pharris,Resnick,&Blum,SchoolwasthenextmostcommoncontextinwhichtherelationshipbetweenconnectednessandSTBwasevaluated.Moststudiesfoundthatschoolconnected-nesswasassociatedwithreducedSTB,althoughresultsvariedbymeasurementapproach.Fourstudies,whichusedAddHealthdata,foundthatperceivedcloseness,caring,belonging,satisfaction,andsafetyinschooland/orwithteacherspredictedlowerlikelihoodofsuicidalthoughtsorbehaviors(Kaminskietal.,2010;McNeely&Falci,2004;Resnicketal.,1997,1993;notethatthislaststudywasbasedonanAddHealthprecursor,theMinnesotaAdolescentHealthHITLOCKETAL TABLE1ReviewofEmpiricalLiteratureLinkingConnectednesstoSuicidalThoughtsandBehaviors(STB)OutcomesOperationalizedasSTBMeasureMethodMainSTB-relatedFamilyConnectednessContextStory,7Perry(2006)12thgradeCommunication,caring,andvaluingparentsÕopinion(fromEATsurvey)SuicideattemptsCross-Lowfamilycorrelatedwithsuicideattempts7Blum12gradePerceivedcaringbyparentsandfamily,familyunderstandingandattention,funwithfamily(AddHealth)SuicideattemptsCross-Connectednesstofamilynegativelycorrelatedwithsuicideattemptsetal.(2001)12thgradePerceivedparentalcloseness,caring,satisfactionfeelinglovedandwantedbyfamily(AddSuicideattemptsLongitudinalPerceivedparentandfamilyconnectednessispromotiveagainstsuicideattemptsONNECTEDNESSAND TABLE1OperationalizedasSTBMeasureMethodMainSTB-relatedetal.(2012)17yearsChangesintheextentfamilyrelationshipsareclose,conÞding,satisfying,andsupportive(PerceivedSupportScale;Slavin,Suicideideation(unrelentingorspeciÞcplan)andattemptsLongitudinalGreaterincreasesinconnectednesswithfamily3monthsafterpredictslessseveresuicidalideationduringa1-yearfollow-upperiodonlyforwithoutmultipleattempthistorieswasnotfoundsigniÞcantforsuicideattempts&Resnick9thand12thgradegay,lesbian,andcaring,enjoyment,andrespectfor(scalecreatedbySuicideideationandattemptsconnectednessiscorrelatedwithsuicideideationandattemptsHITLOCKETAL TABLE1OperationalizedasSTBMeasureMethodMainSTB-relatedetal.(2007)6th,9th,and12thgradeyouthwithhistoryofsexualabusecaring,enjoyment,andrespectfor(scalecreatedbySuicideideationandattemptsisnegativelycorrelatedwithsuicidebehaviorsinyouthwithahistoryofchildhoodsexualetal.(2007)12gradePerceivedcaringand(4-itemscalecreatedbyauthors)SuicideattemptandideationmediatedthebetweenperceivedsocialisolationandsuicideSocialisolationindicateslackofsupportandsenseofbelongingwithpeerswhichreducesself-esteemandincreasessuiciderisketal.(2010)12thgradeCaring,advicegiving,andhelpwithpracticalproblemsfromfamilymembers(VauxSocialSupportRecord)Suicide:nonsuicidalself-harm,suicideideation,suicideplans,andnonfatalsuicidalbehaviorcorrelatedwithsuicideideationandbehaviorsONNECTEDNESSAND TABLE1OperationalizedasSTBMeasureMethodMainSTB-relatedetal.(2006)12thgradePerceptionsofparentalcare,support,andrelationshipquality(AddHealth)SuicideattemptsLongitudinalParentalattime1predictedlowersuicideattemptratesattime2,particularlyforboyswithgoodschoolrelationswithpoorpeerrelationsandasuicideattemptetal.(2011)7thand9thgradeinfamilyandPerceptionofcarebehaviorand(adaptedfromanearlierstudybyetal.,1996)SuicideideationCross-Parentalcareandcorrelatedwithsuicideideation,particularlyforfemales;havingbothparentingfactorsincreasedmagnitudeofeffectforsuicideHITLOCKETAL TABLE1OperationalizedasSTBMeasureMethodMainSTB-relatedetal.(2011)12thgrade,PerceptionsoffamilysupportPerceptionofSupportInventory;Rosen,1981)Suicidality:suicideideationandcorrelatedwithsuicidalthoughtsandpositivelycorrelatedwithreasonsforlivingamongadolescentswithhighlevelsofdepression;noresultsreportedonrelationshipandSTBetal.(1997)6th,9th,and12th,gradeHowmuchfamilycaresabout,respectsprivacyof,andunderstandsadolescent,andextentoffunwithfamily(MinnesotaStudentSurvey)Thoughtaboutorattemptedsuicideconnectednessiscorrelatedwithsuiciderisk,althoughmoreweaklythanothertestedpredictors,self-esteem,andONNECTEDNESSAND TABLE1OperationalizedasSTBMeasureMethodMainSTB-relatedResnick,&Blum(1997)12thgradesexuallyabusedAmericanIndianPerceivedfamilyattentionandcaring(modiÞedAddHealthSuicideattemptandideationcorrelatedwithsuicidalthoughtsandbehaviorsinsexuallyabusedgirls,butnotetal.(1993)12thgradeSenseofbelongingandclosenesstofamily(AddHealth)Suicideideationandattemptscorrelatedwithsuicidalthoughtsandbehaviorsetal.(1997)12thgradeEnjoyandfeelcloseto,andcaredfor,loved,andwantedbyfamilymembers(AddHealth)thoughtsandconnectednessiscorrelatedwithsuicidalthoughtsandbehaviorsSchoolConnectednessContextetal.(2007)12gradeDegreetowhichrespondentlikesgoingtoschool(single-itemscalecreatedbySuicideattemptandideationSchoolconnectednessmediatedtherelationshipbetweensocialisolationandsuicideattemptsforboys,butnotgirlsHITLOCKETAL TABLE1OperationalizedasSTBMeasureMethodMainSTB-relatedKaminskietal.12thgradeFeelclosetopeopleatschool,partofschool,andhappyatschool(AddHealth)Suicideideation,suicideplans(andnonsuicidalSchoolconnectednessisnegativelycorrelatedwithsuicidalbehaviorsandsuicideideationetal.(2006)12thgradeFeelconnectedwithschool(i.e.,teachercaring,feelingclosetopeopleatschool)(AddHealth)SuicideattemptsLongitudinalNomaineffectsforschoolorpeerrelationsonsuicideattempts;3-wayinteractioneffect(parentsXpeerXschool)forboyswithasuicideattempthistoryandpoorpeerrelations,butnotforgirlsorboyswithnoattemptLogan(2009)7thand9thgradeyouthwithabuseFeelingapartofschool,closetopeopleatschool,andhappytobeatschoolSuicideideationCross-Connectednesstoschoolnegativelycorrelatedwithsuicidalthoughts10CONNECTEDNESSAND TABLE1OperationalizedasSTBMeasureMethodMainSTB-relatedetal.(2011)7thand9thgradeinfamilyandFeelclosetopeople,feelhappy,andbelongatschool(AddHealth)SuicideideationCross-SchoolconnectednessnegativelycorrelateswithsuicideideationSocialbelongingandteachersupport(AddHealthscale)Suicide:changesinsuicidality(thoughtsorLongitudinalTeachersupportprotectsagainstsuicideattemptsforthosestudentswhodonotreportexperiencingsuicidalthoughtsatwave1;socialbelongingdoesnotprotectagainstSTBPraiseandattentiongeneratesasenseofbelonging,whichleadstoengagementinschoolandmotivationtodowellandreducesSTBrisk;roleofsocialbelonginginriskvariesbasedonnormsofpeerreferentgroupsetal.(1997)6th,9th,and12thgradeHowmuchadolescentlikesschool(AddThoughtaboutorattemptedSchoolconnectednessnegativelycorrelatedwithsuicidalthoughtsandbehaviorsetal.(1993)12thgradeSchoolenjoymentandsenseof(AddHealth)SuicideideationandattemptsSchoolconnectednessnegativelycorrelatedwithsuicidalthoughtsandbehaviorsHITLOCKETAL TABLE1OperationalizedasSTBMeasureMethodMainSTB-relatedetal.(1997)12thgradeFairtreatmentbyteachers,feelclosetopeopleatschool,senseofbelonging(AddSuicidalthoughtsandbehaviorsConnectednesstoschoolnegativelycorrelatedwithsuicidalthoughtsandbehaviorsetal.(2011)Ages11,15,and19Perceptionsoftheschoolrelationships,pupilinvolvement,andschoolengagementSuicideideationandattemptLongitudinalWiththeexceptionofage11,schoolmeasuresof(poor)schoolconnectednesswereassociatedwithincreasedoddsofsuicideideationandattempt;themajorityofthesesassociationweresigniÞcant,withtheremaindermarginallysigniÞcantPeerConnectednessContextetal.(2012)17yearsChangeintheextentpeerrelationshipsareclose,conÞding,satisfying,andsupportive(PerceivedSupportScale;Slavin,1991)Suicideideation(unrelentingorspeciÞcplan)andattemptsLongitudinalPeerconnectednesspredictslowersuicideideationinfemales3monthsafterhospitalizationbutpredictsmoresevereideationat12monthsAninitialgaininconnectednessafterahospitalizationwithpeersmayturntofeelingsofrejectionorlosswhentheinitialgaininclosenessandsupportdecreasesovertime12CONNECTEDNESSAND TABLE1OperationalizedasSTBMeasureMethodMainSTB-relatedetal.(2010)12thgradePerceivedcaring,provisionofpracticaladvice,andhelpwithproblemsfrompeers(VauxSocialSupportSuicide:nonsuicidalself-harm,suicideideation,suicideplans,andnonfatalsuicidalbehaviorPeerconnectednesspositivelycorrelatedwithsuicidalthoughtsandbehaviorsetal.(2011)12thgrade,Perceptionsofpeersupport(modiÞedPerceptionofSupportInventory;Rosen,Suicideideationandattempts,andreasonsforlivingPeerconnectednesspositivelycorrelatedwithreasonsforliving,butstrongesteffectforstudentswithlowlevelsofdepression;noresultsreportedonrelationshipbetweenconnectednessandCommunityConnectednessContextetal.(1999)12thgradePerceivedcaringbyadults,schoolpeople,church,tribalelders(AddHealth)SuicideattemptsCross-ConnectednesstoadultsoutsideoffamilynegativelycorrelatedwithsuicideattemptsHITLOCKETAL TABLE1OperationalizedasSTBMeasureMethodMainSTB-relatedetal.(2011)12thgrade,SocialCohesionsubscaleoftheCollectiveEfÞcacyScale(Sampson,Raudenbaush,&Earls,1997)Suicideideationandattempts,andreasonsforlivingshowedtrendtowardbeingpositivelycorrelated(butwasnonsigniÞcant)withreasonsforliving,butstrongesteffectforstudentswithhighlevelsofdepression;noresultsreportedonrelationshipbetweenconnectednessand14CONNECTEDNESSAND Database);however,twoothersdidnot(Bearman&Moody,2004;Kiddetal.,2006;notethatBearman&MoodyusedAddHealthmeasuresbutdidnotlabelthisÒconnectednessÓ;theyarethusnotincludedinTable1).ThisdisparityinÞndingsmayresultfromthefactthatboththeKiddandBearmanandMoodystudiesincludemulti-plecontextsinÞnalmodelsandthatthelackofdirecteffectsforschoolconnected-nessonSTBinthesestudiesmaybeduetointeractionsbetweencontextsthatareoftenunmeasuredinconnectednessstudies.Schoolconnectednessstudiesemployingsimilarindicatorsrelatedtofeelingclosetopeople,belonging,feelinghappyatschool,andengagementwithteachersandpeersallrevealinverserelationshipstoSTB(Logan,2009;Neumark-Sztainer,Story,French,&Resnick,1997;Resnicketal.,1993;Young,Sweeting,&Ellaway,2011).Similarly,tea-chercaringandSTBwerepositivelycorre-latedintwostudies(Eisenberg,Ackard,&Resnick,2007;Eisenberg&Resnick,2006)forpopulationsofsexuallyabusedandsex-ualminorityyouth.AsinthefamilysettingÞndings,thereisevidencethatsexmoder-atestherelationshipbetweenconnectednessandSTBinschoolsettingaswell;however,aswithfamily,notinentirelyconsistentways.Onestudyfoundschoolconnected-nessprotectiveforgirlsonly(Hall-Lande,Eisenberg,Christenson,&Neumark-Sztainer,2007),whileanotherfoundaninteractioneffectwithschoolconnectednessandothervariablesforboysonly(Kiddetal.,2006).Incontrasttofamilyandschooldomains,studiesexaminingpeerconnectednessandSTBweremorerareandshowedlargerdiscrepanciesintheconnect-STBrelationship.AlthoughtwostudiesfoundthatpeersupportreducedriskofSTB(Czyzetal.,2012;Matlin,Molock,&Tebes,2011),athirdstudyfoundthatitincreasedriskofSTB(Kaminskietal.,2010).SuchÞndingsareconsistentwithresearchshowingthatassociationsbetweenpeerrelationshipsandwell-beingvarybythenatureofpeerinteractionsandsocialstructure.Ontheonehand,socialisolationfrompeergroupsisassociatedwithhigherriskofsuicidalthoughtsandbehavior(Bearman&Moody,2004;Prinstein,Boer-ges,Spirito,Little,&Grapentine,2000),andsuicidalyouthreporthigherlevelsofsocialisolation(Kaminski&Fang,2009).Ontheotherhand,overrelianceonpeersforsupportincreasesriskofexternalizingproblemsandconßictandcanalsoheightenemotionaldistress(Donald,Dower,Correa-Velez,&Jones,2006).Similarly,otherstudieshaveshownthathavingafriendmakeasuicideattemptincreasesriskofSTB(Bearman&Moody,2004)andthatadolescentsareuniquelyvulnerabletocon-tagioneffectsfollowingthesuicidedeathofsomeonetheyknow(Insel&Gould,2008).Insum,ÞndingsinthisareasuggestthatdisconnectionfrompeersmayheightenriskofSTB,connectiontopeersmayalsoheightenriskofSTBwhenpeergroupnormsaremaladaptiveorunconventionallysupportiveofSTB.StudiesofyouthconnectednesstocommunityandSTBarerare.Mostoftenoperationalizedassenseofconnectiontoandtrustwithadultsoutsidethefamilysys-tem,thesestudiestypicallyshowmildlysta-tisticallysigniÞcantortrendingtowardsigniÞcanteffectsofcommunityconnected-nessonSTB.Forexample,inastudyopera-tionalizingcommunityconnectednessasperceivedcaringbyadults,schoolpeople,church,andtribalelders,communitycon-nectednesswasnegativelycorrelatedwithSTBamongNativeAmericanyouth(Borow-sky,Resnick,Ireland,&Blum,1999).TheseÞndingsareconsistentwithasimilarstudyoftheeffectsofcommunityconnectedness(op-erationalizedassenseofbelongingincom-munity)onSTBamongNativeAmericanadultswherecommunityconnectednessshowedaweakbutsigniÞcanteffectonsui-cideideationandnosigniÞcanteffectonsui-cideattempt(Hill,2009).Similarly,inastudyofAfricanAmericanyouththatopera-tionalizedconnectednessassenseofsocialcohesioninneighborhoods,communitycon-nectednesstrendedtowardbeingpositivelycorrelated(butwasnonsigniÞcant)withtheHITLOCKETAL reasonsforlivingSTBmeasure(Matlinetal.,2011).WHATARETHEMECHANISMSBYWHICHCONNECTEDNESSINFLUENCESSTB?Thequantityandqualityofsocialtieshavebeenlinkedwithsuicideforoveracen-tury,sinceDurkheim(1897)Þrstpositedthataweakeningofthebondsthatnormallyinte-grateindividualsintothecollective(i.e.,ÒanomieÓ)isamongthechiefcausesofsui-cide.Althoughunityacrosscontemporaryconceptualizationsofconnectednessislar-gelylacking,thereisbroadagreementthatconnectednessisderivedfrominterpersonalafÞliationwithindividualsandinstitutionswhichenhancewell-beingthroughemo-tionalandinstrumentalsupportderivedfromoneormoreinterlockingsocialsystems(e.g.,families,schools,peergroups,communities).Beyondthis,fewstudieshaveexplainedthespeciÞcmechanismsorpathwaysthroughwhichconnectednessmodiÞesriskofsuicideorothermentalhealthproblemoutcomes,asisevidentinTable1.Asameansofpostulat-ingmorespeciÞcpathwaysofpotentialempiricalvalue,weexaminedoverlapbetweenthetheoreticalmodelsinwhichconnectednessisgrounded(identiÞedabove)alongsidemodelsofthefactorsthatleadtosuicide.Oneofthemostwell-developedsui-cidemodels,JoinerÕs(2006)interpersonaltheoryofsuicide(ITS),proposesthatsuicidearisesfromperceptionsofthwartedbelong-ingness(alsoreferredtoasocialisolation;whattheauthorsseeasonefacetofthehigherorderconstructofsocialconnected-ness)coupledwithperceivedburdensome-nessandcapacityforlethality.BothbelongingnessandburdensomenessarebasedonperceptionsofsocialafÞliationandexchange.Bothhaveanintrapersonaldimen-sion(individualthoughtsandemotionsthataffectperceptionandinterpretationofsocialinteractions)aswellasaninterpersonaldimension(presenceofandexchangewithindividualsandsystemsoutsideofoneself;theseexchangeshelptocreateandreinforceintrapersonalcognitionandaffect).Althoughsocialisolation(acorecomponentofthwartedbelonging)containsclearstructuralovertonesinthismodel,theauthorspostu-latethatthesubjectiveexperienceofisolationisthecoremechanismbywhichsocial(dis)connectednessinßuencesSTB.MorespeciÞ-callyandgermanetoourpostulation,VanOrdenetal.(2010)arguethattheneedforsocialbelongingisareßectionoftwopri-maryhumanneedsÞrstarticulatedbyBau-meisterandLeary(1995):theneedforregular,affectivelypositiveexperiencewithknownotherscoupledwithstableperceptionthatoneisengagedinasocialnetworkofotherscharacterizedbyreciprocalcareandconcern.Incontrast,themorerecentcon-nectednessSTBmodeloutlinedbytheCDC(2008)emphasizesthestructuraldimensionofsocialexchangebypostulatingthatthesocialsystemsinwhichyoungpeo-pleareembeddedconferprotectionandopportunitiesinwaysthatareofteninvisibletoindividualsandhencelesssubjectivelyexperienced(althoughtheyquietlyshapethesubjectiveexperience).Theydothisthroughsupplyingindividualsintheexternalenviron-mentwhomayassistinidentifyingandinter-veningwhensomeoneisobservedtobeatriskofSTBandbycoordinatedservicesaimedatreducingenvironmentalriskfactorsandenhancingenvironmentalprotectivefac-tors(CDC,2008).Insum,ourreviewsug-geststhreebroadpathwaysbywhichconnectednessmayprotectagainstSTB:(1)intrapersonalresponsesandprocesses,(2)collectiveresponsibilityandaction,and(3)positivenormsandexpectations.Webelievethateachoftheseareasisusefulinexplicat-ingempiricallytestablehypothesesandforguidinginterventionandpreventionefforts(Figure.1).IntrapersonalresponsesandprocessesTheITS(Joiner,2006;VanOrdenetal.,2010)holdsthatboththwartedbelongingnessandburdensomenessarisefromdistortedself-narrativesreinforcedby16CONNECTEDNESSAND perceptionsofsocialinteraction(VanOr-denetal.,2010).Inotherwords,inorderforsuicidetobeconsideredoractedon,onemustbelieve(cognition)andfeel(affect)thatoneisnotwantedandis,more-over,aburdenonsalientothersinoneÕssocialnetwork.Embeddedinthistheoryistheassumptionthatbothcognitions(e.g.,thoughts,memories,judgments)andaffec-tiveexperiences(e.g.,emotions,feelings)areassociatedwithsocialexchangeandexperiences.Italsosuggeststhatthesecog-nitiveandaffectivebeneÞtsmaybeaccruedthroughperceivingthatoneisofusetokeyreferentgroups(e.g.,notaburden)andthroughperceivingexistentialvalueandmeaninginbeingpartofagroup.Theseassumptionsareconsonantwithanemerg-ingbodyofevidencewhichsuggeststhatthesubjectiveexperienceofconnectednesstoothersmayliterallyshape(Perry,2002)orcontributetothephysiologicalsystemsthatgovernpositiveemotionandstressresponse(Eisenberger&Cole,2012;Liv-ingstone&Srivastava,2012).Wethushypothesizethatsimplyconnectedtooneormorecommunitiesmayreduceado-lescentSTBriskthroughtheneurophysio-logicalbeneÞtstocognition(e.g.,believingoneisofvalueandcaredfor)andemotionregulationthatresultfromsocialafÞliationandattachment.Inthissense,wearguethattheexperienceofafÞliationandthepositiveemotionsitengendershasdistinctivepro-tectivevalueforSTBoverandabovetheÒabsenceofpathologyÓandindependentofotherstructuralfactors,suchasnetworkdensityorquality.Associalconnectednessisrelatedtohowotherpeoplearerepresentedwithinoneself,theexperienceofconnectednessultimatelystemsfromoneÕssubjectiveper-ceptionofinterpersonalclosenessandvalue(Lee&Robbins,1995).Consonantwiththis,studiesexaminingtheeffectsofvaryingcategoriesofsocialsupport(typicallysocialsizeanddensity,enactedandperceivedsup-port)consistentlyÞndthatperceivedsupportisamorepowerfulindicatorofwell-beingthanothercategoriessuchasnumberofsocialties(Taylor&Lynch,2004).TheseÞndingsholdforyouthaswell,particularlyastheyage(Chu,Saucier,&Hafner,2010).Thissuggeststhattheexperienceoftheexternalsocialworld,asrepresentedbysubjectivethoughtsandemotions Figure1.ProposedmechanismdomainsthroughwhichconnectednessconfersprotectionagainstSTBinadoles-HITLOCKETAL directlymediatesanindividualÕsperceivedlevelofconnectednessandinßuencesuptakeofassistanceofferedintimesofdistressorimbalance.Whyisthis?Anincreasinglysophisti-catedbodyofsciencesuggestsadynamicinterplaybetweenperceptionsofsocialexperiences,neurobiology,andemotion(Ei-senberger&Cole,2012;Gilbert,2012;Steinberg,2010).Humanspossessacom-plexarrayofneuralcircuitrytosupportsocialafÞliationandcommunication(Eisen-berger&Cole,2012;Perry,2002).ThiscircuitryexiststocodemeaningofsocialexchangeandtopromoteafÞliationsthatenhancesurvivalandwell-being(Perry,2002).TheperceptionofemotionisoneofthewaysthatthebrainsignalsthebodytoseekafÞliationsdeemeduseful(Eisenberger&Cole,2012).Forexample,thefeelgoodchemicals,norepinephrineandserotonin,areproducedinthebodythroughpositiveafÞliationwithothers,areperceivedasÒpositiveemotion,Óandplayacentralroleintheabilitytosuccessfullyregulateemo-tioninchallengingcontexts(Insel&Win-slow,1998).Moreover,thesubjectiveexperienceofpositiveemotionleadstopositivementalstatesandbetterphysicalhealth(Gilbert,2012)andimprovessocia-bilityandcognition,speciÞcallycreativityandproblem-solving(Fredrickson,2001;Fredrickson&Joiner,2002).Foradolescents,theneurologicalpathwaysbetweenexternalexperiencesofinterpersonalexchangeandtheproductionofrewardchemicalsareparticularlypro-nouncedandenhancethesalienceofemo-tionincognitionandbehavioraloutputs(Chein,Albert,OÕBrien,Uckert,&Stein-berg,2011;Steinberg,2010).Adolescentsexhibithighphysiologicalandneurologicalsensitivitytoexternalemotionalcues(par-ticularlysocialrejectionandacceptance;Steinberg,2010)andtendtofeelallemo-tions,andparticularlynegativeemotions,moreacutelythanchildrenoradults(Gilbert,2012).Fordepressedadolescents,theexperienceofpositiveemotionisevenmorebluntedthaninnormativeadolescentpopulationsandislikelytoreinforcenega-tiveattributions(e.g.,thedeductionthatoneisnotwantedordoesnotbelong;Chorpita&Daleiden,2002).Aprolongedsenseofnegativeemotionstatesandsocialdisconnection(whicharemutuallyreinforc-ing)contributetoneurophysiologicalimbal-anceslinkedtomanyoftheknownprecursorstosuicide,suchasdepression,substanceuse,andpersistentfeelingsofhopelessness(Eisenberger&Cole,2012).Experiencingpositiveemotionsbroadensattention,positivecognition,self-regulation,andsocialafÞliationdesiresandcapacitieswhich,inturn,enhancesenseofwell-beingandreducesuiciderisk.CollectiveResponsibilityandActionTheintrapersonalprocessesdiscussedaboveareproducedonlyinrelationshiptosocialsystems,particularlythosewhichadvanceasenseofdevelopmentallyappro-priate(e.g.,notoverburdening)utilitytoothers.Beingembeddedinavarietyofinterconnectedsocialsystems(e.g.,family,school)accruesbeneÞtsformembersthroughmonitoringandsharingofresources(Bovier,Chamot,&Perneger,2004).Onesuchsystem-leveleffectistheenhancedlikelihoodthatpsychologicaldis-tressandsuicidewillbedisclosedordetectedbyothersandaddressedinawaythatreducesthepotentialforlethalout-comes.Thepotentialforcollectiveactionisgreaterwhenindividualsareengagedinmeaningfulrelationshipsandcollectiveaction,andresponsibilitywillbeenhancedbyincreasednetworkdensityanddiversity.Asacaseinpoint,severalsourcesofdatasuggestthatschoolswithstrongsocialnetworkslinkingadultswithadolescentsarelikelytoprovidemoreavenuesforhelp-seekingandhelp-givinginteractionsthanthosewithoutsuchnetworks(Pisani,Schmeelk-Cone,etal.,2012).Moreover,theprotectivevalueofsuchnetworksislikelytobeaugmentedbytheinclusionofcompetentadultswhocandetectandrespondtodistress(Wymanetal.,2008).18CONNECTEDNESSAND Thus,wepositthatconnectednesstoadultsandsystemsinwhichyouthadultrelation-shipsoccur(suchasschools)confersprotec-tionby(1)heighteningopportunitiesforsolicitingandactivatingassistance;(2)enhancingthelikelihoodthatnegativeaffectandbehavior,includingsignsofdistressormoredirectwarningsignsforsuicidalbehavior,willbenoticedandproactivelyaddressed;and(3)providingasenseofutil-ity,meaning,andpurpose.Reciprocalcommunicationpathwaysbetweenyouthandadultsareparticularlyimportantforadolescents,who,unlikeyoungchildren,exercisegreateragencyinseekinghelpandgreaterautonomyinacceptingorrejectinghelp,whetherfrominformalsources(e.g.,family,teachers)orformalsources(e.g.,medicalandmentalhealthprofessionals).Studieswithinthementalhealthservicesliteratureshowthatstrongtieswithadultsinkeysocialsettingsenhanceadolescentwillingnesstoseekhelpforemotionalproblems,ingeneral,andforsuicideconcerns,inparticular(Pisani,Schmeelk-Cone,etal.,2012).Establishingpersonalrelationshipswithadultsiscriticalbecausethepathwaytomostformsofhelpforadolescentsbeginswithexistingrela-tionshipstoadults(Boldero&Fallon,1995;Costello,Angold,March,&Fairbank,1998;Rickwood,Deane,&Wilson,2007)andbecauseadolescentsprefertodiscussemotionalproblemswithfamiliarpersonsandaremorelikelytoseekprofessionalhelpfromsourceswithinfamiliarsettings(Costelloetal.,1998;Logan&King,Adolescentswhoaremoresociallyintegratedintosystems,suchasschoolandclubs,andwhoalsofeelthattheybelongaremorelikelytoperceivesocialsupportasavailableandadultsascapableofhelpingsuicidalyouth,bothofwhichpredictadolescentsÕhelp-seekingbehaviorandintentions(Pisani,Schmeelk-Cone,etal.,2012;Pisani,Wyman,etal.,2012).Simi-larly,expectationsofstrongsocialsupport,theabilitytoapproachadultswithprob-lems,andclosenesswithadultswereallassociatedwithgreaterintentionstoseekandaccepthelpforsuicideideationincol-lege-agestudents(Yakunina,Rogers,Waeh-ler,&Werth,2010).Inastudyfollowingasuicideeducationprogram,highschoolstu-dentsreportedthattheinabilitytoapproachanddiscussproblemswithadultsandalackofclosenesswithadultswouldbekeybarriersforseekinghelpforoneselforafriendatschool(Cigularov,Chen,Thur-ber,&Stallones,2008).Thus,beyondthemereopportunityforsupportiverelation-ships,ahighdegreeofsocialintegrationmayexpandthebaseofsocialresourcesavailabletoindividualsinatimeofcrisis.Greaterperceivedengagementandstructuralintegrationatschoolandinothersettings,particularlywherecompetentandcaringadultsareavailable,providemoreopportunitiesforsuicideriskindicatorstoberecognizedbyothersasÒproblemsÓneedingattention(Costello,Swendsen,Rose,&Dierker,2008).Sinceparentsaretheprimaryfacilitatorsofprofessionalservicesandassis-tance(Logan&King,2001),theirawarenessofandwillingnesstodeÞneaproblemassucharecritical.Linkagesbetweensocialsystemsarealsoimportantbecause,forexample,parentsareoftenalertedtoanado-lescentperiodofdifÞcultythroughconcernsraisedbyschoolpersonnel(Costelloetal.,2008).Asaresult,adolescentswhoexperi-encelowconnectednessacrossmultiplesocialcontexts(e.g.,schools,families,peers,community)arelesslikelythantheirbetter-connectedpeerstobenoticedandrespondedtointimesofdistress.Finally,beingpartofacommunityofothersenhancesmotivationtobeofvalueandofusetoothers,anexperiencethatservesasanimportantsourceofperceivedmeaningandpurpose(Townsend&McWhirter,2005).Althoughchildrenandadolescentsaretypicallyregardedastherecipientsofcare,beingapartofacommu-nityallowsforthegivingofcareandsup-portwhichmayhaveunderrecognizedvalueinprotectingagainstSTB(Joiner,2006).Knowingthatoneislikedandcaredforisimportant,butconnectednessimpliesreci-HITLOCKETAL procalexchange:thereceivinggivingofcare,respect,trust,andsupport(Whitlock,2006).Assuicideishighlyassociatedwithdepressive,ruminativecognitionsthatpro-ducealowsenseofbelongingand/orsocialvalue(Durkheim,1897;Joineretal.,2005),opportunitiesformeaningfulcontributionandrecognitionlowersuicideriskandenhancesenseofpurposeandmeaning,par-ticularlyiftheyengenderregularopportuni-tiesforexperiencingpositiveemotion(Fredrickson&Joiner,2002;Livingstone&Srivastava,2012).Thisassumptionhasbeenvalidatedinsamplesofadultswhereprovid-ingemotionalandinstrumentalsupporttoothersconfersevenstrongerprotectiveandpromotivebeneÞtsthanreceivingsupportinavarietyofsuicide-linkedareas,suchascoping,purposeinlife,depression,andanx-iety(Brown,Nesse,Vinokur,&Smith,2003;Schwartz&Sendor,1999).Instudiesofyouth,frequentpositiveemotionsduringschoolwereassociatedwithhigherlevelsofstudentengagementandnegativeemotionswithlowerlevelsofengagement(Reschly,Huebner,Appleton,&Antaramian,2008).Thesamestudyshowedthatpositivebutnotnegativeemotionswereassociatedwithadaptivecopingand,throughthis,tostu-dentengagement.Studiesofeffectsofado-lescentgivingshowsimilarbeneÞtstogivers(e.g.,adolescentsaremorelikelytobeneÞtthanchildren)(Dillon&Wink,PositivenormsandexpectationsAssomeresearchershavenoted(Kir-by,2001),theinßuenceofconnectednessmaybepositiveornegativedependingonthenormsandvaluesoftheindividuals,groups,orsystemstowhichoneisattached.Thisisoneofthereasonsthatstructuralandsubjectiveconnectednesstopeergroupswithantisocialornegativethinkingnormsmayactuallyheightenriskofsuicide(Bear-man&Moody,2004;Insel&Gould,2008).Normsinßuenceawiderangeofyouthhealthbehaviorsincludingriskybehaviors,suchassubstanceabuseanddelinquency,andalsoprosocialbehaviors(Gilbert,2012;Steinberg,2010).Normsencompassperceptionsofhowtypicalagivenbehavioriswithinagroup(i.e.,descriptivenorms)aswellastheconse-quencesofengaging,ornotengaging,inthatbehavior(i.e.,injunctivenorms).Bothdimensionsinßuencedecisionmakingandbehavior(Rimal&Real,2003).NormsmostlikelytoreducethelikelihoodofSTBincludethosethatsupporthealthycopingandhelp-seekingbehaviorsandthosethatinßuencehowmembersperceivesuicidalthinkingandbehavior(i.e.,STBisnorma-tive,abnormal,requiresintervention;Pisani,Schmeelk-Cone,etal.,2012).ThetransmissionofnormsthroughafÞliationgroupslinksboththesubjectiveandstructuraldimensionsofconnectedness.Forexample,thesubjectivedimensionofconnectednessmayinßuencehowadoles-centsplacevalueonnorms,assuggestedbyalargebodyofresearchshowingthatado-lescentsÕnormsandbehaviorsarestronglyinßuencedbyfriendsandbythosetowhomtheyaspiretobemorecloselyafÞliated(Valente,2010).Someofthisinßuenceonadolescentsandyoungadultscanbetiedtoheightenedphysiologicalandneurologicalsensitivitytoexternalemotionalcues(e.g.,socialrejectionandacceptance)coupledwiththefactthatthepresenceofpeersaug-mentsadolescentrisk-takingandsensation-seeking(Steinberg,2010).Thus,subjectiveperceptionsofvaluinginßuencearehownormsarecocreatedanddisseminated.Thestructuraldimensionisevidentinthewaysinwhichnormsaredisseminatedthroughnetworktiesandareinßuencedbynetworkproximitytoothersocialgroups.Normsandpracticesspreadreadilythroughdensersocialnetworksandarereinforcedwhenthesenetworksareonlylooselyconnectedtoothernetworks,peers,oradults,orwhentheyaretightlyconnectedtosocialnet-workswithsimilarnorms(Valente,2010).At-riskadolescentswhoareisolatedorwhoafÞliateprimarilywithpeerswhosharetheirbehavioralandattitudinaltendenciestypicallypossessfeweropportunitiestoben-20CONNECTEDNESSAND eÞtfromexposuretohealthysocialnormsandattitudes,particularlyiftheirsocialgroupisnotafÞliatedwithothernetworksorwithadults.Groupnormsandattitudesthatsup-porthelp-seekingwithinoneÕsreferentgrouphavebeenlinkedinnumerousstudiestogreaterwillingnesstoseekhelpforemo-tionalproblems,includingsuicidalconcerns(Pisani,Schmeelk-Cone,etal.,2012;Rick-woodetal.,2007).TheseÞndingsholdtrueforadolescentsaswellwherestudiesshowthatperceptionsofpeersupportforhelp-seekingareassociatedwithgreaterlikelihoodofdisclosingrecentsuicidalbehaviortoanadult(Pisani,Wyman,etal.,2012)andgreaterintentiontoseekhelpfordistress(Schmeelk-Cone,Pisani,Petrova,&Wy-man,2012).Similarlyinßuentialmaybegroupperceptionsaboutsuicidalthoughtsandbehaviorsand,inparticular,therespon-sibilityofmembersforresponding.TheextenttowhichsymptomsbecomedeÞnedasproblemsneedinginterventionisaprocessthatoccurswithinsocialnetworks,includingfamiliesandpeergroups,whichhavetheirownnormsandexpectationsaboutbehavior(Pescosolido,1992).SUMMARYANDIMPLICATIONSApplicationoftheconnectednessconstructacrosscontextsinresearchexam-iningassociationswithSTBlargelyshowsconsistentresults,whetheroperationalizedasasingleindexorasacollectionofindi-ces,andpointtoseveralconclusions.First,fewexistingdeÞnitionsofconnectednessreßectboththesubjectiveandstructuraldimensionsofconnectedness.Forthisrea-son,weofferthefollowingdeÞnitionofnectedness:thedegreetowhichanindividual(orgroup)possessesasubjectivesenseofemotionalinterrelatedness(belonging,car-ing,value,andtrust)andawillingnesstosharewithandseekresourcesfromtheindividualsandcommunitiesinwhichhe/sheissociallyorgeographicallyembedded.ThisdeÞnitionreßectsbothsalientdimen-sions(subjectiveandstructural)emergentintheliteraturereviewedandencapsulatessev-eraloftheprimaryconstructsusedtomea-sureconnectedness.Second,theresultsofourreviewunderscorethecomplexityofstructuralandinterpersonalafÞliationovertimeandplace.Forexample,researchsug-geststhatpeerconnectednessconferspro-tectionagainstSTBinmanyconditionsbutcanbeariskfactorifafriendmakesasui-cideattemptorholdsSTBpromotiveatti-tudes.Lastly,ourresultsstronglyunderscorethebeneÞcialeffectsinreducingriskofSTBfromadolescentsÕtiestoadults,particularlyfamilymembers,whoareper-ceivedassupportiveandengaged.MorespeciÞcimplicationsforpracticeandresearchareofferedbelow.ImplicationsforPracticeOnthebasisoftheliteraturereviewedhere,wewouldexpectthatcon-nectedness-focusedinterventions,regardlessoftheecologicallevelinwhichtheyaretar-geted,willexertinßuenceonindividualSTBoutcomes.However,variationintheeffectofconnectednessonSTBbydemo-graphicfactorssuchassexandacrosscon-textsuggeststhatconnectedness-basedinterventioneffortsmaybeneÞtfromearlyandongoingevaluativeattentiontothewaysandvariationsinwhichtheinterven-tionisleveragingeffectandthewayinwhichitdifferentiallyimpactsboysandgirls.Lastly,thepatternofempiricalÞnd-ingstodatesuggeststhatthelocationofcontext(proximalversusdistal)matters.Family,forexample,consistentlyemergedasthemostpotentconnectednesscontextfollowedbyschools,peers,andthencom-munities.Theredoesexistsomeresearch,however,tosuggestthatnumberofcontextstowhichonefeelsconnectedmaymatterevenmorethanthespeciÞccontextinwhichconnectednessoccurs(Borowskyetal.,1999;Kaminskietal.,2010),soeffortstoenhancepositiveafÞliationinmultiplecontexts,evenifmoredistalcon-textsaretargeted,maybeasormoreeffec-HITLOCKETAL tivethanfocusingexclusivelyonone.Thisisaparticularlyimportantimplicationincommunitieswherefamilycontextsarelessavailableandmutabletargetsofchangethanothercontexts.Beyondthis,reviewofstudiestodatesuggeststhatconnectednessaffectsSTBthroughoneormoreofthefollowingroutes:(1)expandingintergenerationalsocialnetworks;(2)heighteningopportuni-tiesforsolicitingandactivatingassistancefromothersorsystems(e.g.,schools,fami-lies,orothersocialsystems);(3)enhancingthelikelihoodthatworrisomeaffectandbehavior,includingearlysignsofdistressormoredirectwarningsignsforsuicidalbehavior,willbenoticedandproactivelyaddressedbyproximalsystems(parents,peers,schools);(4)increasingexposuretopositivecopingandhelp-seekingnorms;(5)increasingpositiveemotionand,asaconse-quence,cognitiveßexibilityandemotionregulationcapacity;and(6)enhancingopportunitiesforexperiencingbelongingandutilityinacommunityofothers.Inter-ventionsintendedtobeasmultilayeredastheconstructwillincludeasmanyofthesemechanismsaspossibleand,ideally,willcrosscontext.However,smallerprojectsintendedtoisolateeffectsofoneormoreoftheseleveragepointsinonecontextwillbeofvalueinelucidatingthecontribution.ResearchImplicationsVariationinthewayconnectednessisdeÞnedandoperationalizedinrelationtoSTBhasresultedinadiversearrayofcon-structs,studies,andlevelsofanalyses,fewofwhichallowforsatisfyingsystematiccomparison.Althoughitmaybeuntenabletodesignstudiesreßectiveofalltheelementsofconnectedness,wejoinBarberandSchluterman(2008)inencouragingresearcherstobeexplicitandpreciseinlocatingthedimensionsanddomainsofconnectednesstheywishtopursuewithinthebroaderframeworkoutlinedhere.Ourreviewalsopointstoseverallim-itationsofcurrentresearchandareasforfutureinvestigation.First,becausevirtuallyallstudiestodatehavetestedonlymaineffects,thereareanumberofimportantunansweredquestionsrelatedtotherela-tionshipbetweenconnectednessandSTBdependingonfactorssuchaspeergroupnorms.Forexample,studiesexamininginteractionsandprocessesacrosslevelscomprisingconnectednessareneededtoascertainhowthebeneÞtsofconnectednesstopeergroupsonemotionalwell-beingmaybeoffsetbynormsthatmaypromotemaladaptivestrategies.Alsoofvaluewouldbestudiesthatexaminesalientdimensionsofconnectedness(subjectiveandstructural)withinandacrosscontextswithaneyetocontextualanddemographicmediators,moderators,interactions,andgroup-levelvariationsbuildingoffnascentworkinthisarea(Kaminskietal.,2010;Kiddetal.,2006).Forexample,variationinthewayconnectednessaffectsmaleandfemaleSTBriskcontainsimportantimplicationsforintervention.FindingstodatesuggestthattheremaybeimportantvariationsintheeffectsofconnectednessonSTBdependingonthecontextwithmoreproximalcontexts(e.g.,familiesandschools)exertingmoreinßuencethanmoredistalcontexts(e.g.,communities).However,lackofuniformityinresearchingtheconnectednessSTBlinkwithinandacrosscontexts(particularlycommunitycontexts)prohibitssoundunderstandingofdirectandindirecteffects.Furthercomplexityinmeasuringconnect-ednessisintroducedbythefactthatper-ceivedconnectednesstoindividualsmaybegovernedbydifferentfactorsthanperceivedconnectednesstolargersystems(e.g.,fami-lies,school,communities)andthatper-ceivedconnectednesstosystemsmayreßectaheterogeneoussetofexperienceswithsetsofindividuals(e.g.,schoolconnectednessmayreßectexperienceswithpeersandindi-vidualadults).ParticularlyinlightoftheCDCfocusonunderstandingandpromot-ingconnectednessacrosskeysocializingcontexts,clearlyspeciÞedresearchondirect,indirect,andinteractioneffectsacrosscontextismerited.22CONNECTEDNESSAND Althoughwehaveprofferedseveralpathwaystoexplaintheprotectiveeffectsofconnectedness,betterunderstandingofthemechanismslinkingconnectednessandSTBisaprimaryresearchneed.Therearemyr-iadpossibledesignsandresearchaimsthatmaycontributetoadvancingthisareaofknowledge.Asastartingpoint,weofferthefollowingfourbroadhypothesesasout-standingcandidatesfortheÞrstwaveofmorereÞnedtestingoftheconnectednessconstruct:Thesubjectiveexperienceofbelong-ingandutility(e.g.,feelingusefulandvaluedinatleastoneproximalcontext)willenhancepositiveemo-tionsandcognitionspertainingtoselfandotherinterrelatednessandwillresultinfewerorlessseveresui-cideriskbehaviors;Increasingthedensityofsocialtiesinsecondaryschools,particularlyforisolatedadolescents,willleadtoincreasedexposuretonormativesocialinßuencesthatpromotemorepositivehelp-seekingandcopingDirectexperiencescooperatingwithadultstowardsharedgoalsmayhelptofostermorefavorableattitudesaboutadulthelpandresultinincreasedadolescenthelp-seekingforoneselfandoneÕspeers;andThebeneÞtofconnectednesstopeersismediatedormoderatedbynormspertainingto,andpeerexpe-rienceswith,STB.Thisinteractionisinßuencedbythestrengthofcon-nectednessacrosscontextualdomain(e.g.,family,school,peer,commu-Usingexperimentaldesignstoincreasethecausalinferencesthatcanbemadeaboutconnectednessshouldbeanotherhighpriority.AsevidentinTable1,allconnectedness-focusedstudiestodatearecross-sectionalorlongitudinalobservationalstudies.WeidentiÞednoexperimentallydesignedstudiesthathadthespeciÞcintentionofassessingtheeffective-nessofconnectedness-basedinterventionsinreducingSTB.Asstudiesofinterventionsarethemosteffectivewaytounderstandmechanisms,researchisneededtoexaminechangesfrominterventionsusingrigorousmediationmodelsthatexamineproximalchangesinmechanismsandlongertermriskfactorsandSTB.Similarly,despitethecallforunderstandinghowitsmorestruc-turalelementsaffectSTBriskinadoles-cents,noconnectednessSTBstudyinadolescencereviewedhereusednetwork-relatedor-levelindicatorstomeasurecon-nectedness.Muchneededarestudiesthatoperationalizethemorestructuralelementsofthisconstructandthatassessthewaysinwhichplacementwithinsocialsystemsaffectsaccesstoanduptakeofsocialresources.Oneexampleofanexperimentaltestofaconnectedness-basedinterventionisthecurrentongoingrandomizedcontrolledtrialoftheSourcesofStrengthprogram,auni-versalsuicidepreventionprograminsecond-aryschools(Wymanetal.,2010).Aprevioustestofthisinterventionusingarandomizedcontrolledtrialdesignfoundthatafter4monthsofmessagingactivitiesconductedbydiversepeerleaderstrainedintheSourcesofStrengthcurriculum,school-widehelp-seek-ingacceptanceandnormsforcopingwithsuicideconcernswereincreased.Acurrent,ongoing,federallyfundedtrialisemployingsocialnetworkanalysestodeterminewhetherpeerleadermessagingincreasesthedensityofpositiveafÞliationtiesacrossado-lescentsandincreasespositivetiestoadults,particularlyamongisolatedstudents,andwhetherthosenetworkchangesincreasehelp-seekingforsuicideconcerns,morepositivecopingpracticesand,ultimately,decreaseSTB.Similarly,theCaringLettersProject(Motto&Bostrom,2001)isasuicideinterventionforadultsgroundedinsimilarconnectedness-basedprinciples;theideathatextendingcaringletterstoindividualsatheightenedriskofsuicidemayenhancefeel-ingsofconnectednessandbelonging.StudiesHITLOCKETAL todatesupportsuggestedefÞcacy(Motto&Bostrom,2001)andfeasibility(Luxton,Kinn,June,Pierre,Reger,&Gahm,2012)ofthisapproachinadultsandhavepromiseforyouth.Randomizedcontroltrialsofinter-ventionssuchasSourcesofStrengthandtheCaringLettersProjectarepoisedtosigniÞ-cantlyadvanceunderstandingofeffectsandmechanismsatworkintherelationshipbetweenconnectedness-basedconstructsandSTB(Wymanetal.,2010).Suchstudiesdesignedtoenhanceconnectednessasameansofleveragingitsinßuenceoffervalu-ableopportunitiesforunderstandingbasicmechanismsandeffectsintherelationshipbetweenconnectednessandSTB.Inclosing,theconnectednesscon-structpossessespracticalandintuitiveappealandcontinuestobequitewidelyusedasaresearchconstruct,despiteempiri-callimitations.Inlightofthis,wesuggestthatresearchersacknowledgeconnectednessasabroadmeta-conceptandthattheylocatetheirstudieswithinthisoverarchingframeworkand,mostimportantly,thattheyusetheseeffortstobetterarticulateandelu-cidatethespeciÞcmechanismsbywhichconnectednessleveragesinßuenceonSTB.Wefurthersuggestthatthisresearchbeconcentratedwithinoneormoreofthetwoconnectednessareasmostcommonlyassumedtoleverageinßuence:(1)positivesubjectivecognitiveandemotionalexperi-encesandappraisalsofrelationshipswithadults,peers,andsocialsystems(e.g.,schools)and(2)structuralinterrelatednessbetweennetworksinwhichyouthareembedded.Suchresearchshouldbestruc-turedtoshedlightoninteraction,media-tion,andmoderationeffects,ideallyusingstudydesignsthatpermittemporal,indi-rect,andinteractioneffectstesting.,D.M.,N,D.,,M.,&P,C.(2006).Parent-childconnectednessandbehavioralandemotionalhealthamongadolescents.AmericanJournalofPreventiveMedicine,59,A.(1997).Self-efÞcacy:Theexer-ciseofcontrol.NewYork:Freeman.,B.K.,&S,J.M.(2008).Connectednessinthelivesofchildrenandadolescents:AcallforgreaterconceptualJournalofAdolescentHealth43,209,B.K.,S,H.E.,&O,J.A.(2005).Parentalsupport,psychologicalcontrol,andbehavioralcontrol:Assessingrele-vanceacrosstime,culture,andmethod.graphsoftheSocietyforResearchinChild,209,R.F.,&L,M.R.(1995).Theneedtobelong:Desireforinterper-sonalattachmentsasafundamentalhumanmoti-PsychologicalBulletin,497,P.S.,&M,J.(2004).SuicideandfriendshipsamongAmericanAmericanJournalofPublicHealth,J.,&F,B.(1995).Ado-lescenthelp-seeking:Whatdotheygethelpforandfromwhom?JournalofAdolescence,193,I.W.,I,M.,&R,M.D.(2001).Adolescentsuicideattempts:Risksandprotectors.,I.W.,R,M.D.,I,M.,&B,R.W.(1999).SuicideattemptsamongAmericanIndianandAlaskanativeyouth:Riskandprotectivefactors.ofPediatricsAdolescentMedicine,573,P.A.,C,E.,&PT.V.(2004).Perceivedstress,internalresources,andsocialsupportasdeterminantsofmentalhealthamongyoungadults.QualityofLife,161,J.(1969).AttachmentandlossNewYork:BasicBooks.,U.,&M,P.A.(1998).Theecologyofdevelopmentalprocesses.InW.Damon&R.M.Lerner(Eds.),Handbookofchildpsychology,Vol.1:Theoreticalmodelsofhuman.(pp.793828).NewYork:Wiley.,S.D.,B,T.,L,R.W.,,J.,&H,S.(1987).Perceivedsocialsupportamongcollegestudents:Threestudiesofthepsychometriccharacteristicsandcounsel-ingusesofthesocialsupportinventory.ofCounselingPsychology,337,S.L.,N,R.M.,V,A.D.,&S,D.M.(2003).Providingsocial24CONNECTEDNESSAND supportmaybemorebeneÞcialthanreceivingitresultsfromaprospectivestudyofmortality.PsychologicalScience,320CentersforDiseaseControlandPreven-tion,NationalCenterforInjuryPreventionandControl.(2008).Strategicdirectionforthepreven-tionofsuicidalbehavior:Promotingindividual,fam-ily,andcommunityconnectednesstopreventsuicidal.AccessedMarch13,2013,fromhttp://CentersforDiseaseControlandPreven-tion/NCHS.(2011).Nationalvitalstatisticssystem:Mortality,2011and2010.AccessedFebruary20,2013,fromwww.cdc.gov/nchs/deaths.htm.,J.,A,D.,OÕB,L.,U,K.,&S,L.(2011).PeersincreaseadolescentrisktakingbyenhancingactivityinthebrainÕsrewardcircuitry.DevelopmentalSci-(2),1,B.F.,&D,E.L.(2002).Tripartitedimensionsofemotioninachildclinicalsample:Measurementstrategiesandimplicationsforclinicalutility.JournalofConsultingandClinicalPsychology,1150,P.S.,S,D.A.,&H,E.(2010).Meta-analysisoftherelationshipsbetweensocialsupportandwell-beinginchil-drenandadolescents.JournalofSocialandClinical,624.,K.,C,P.Y.,T,B.W.,&S,L.(2008).Whatpreventsadolescentsfromseekinghelpafterasuicideeducationprogram?SuicideandLife-Threatening,74,J.S.(1988).Socialcapitalinthecreationofhumancapital.AmericanJournalofSociology(Suppl.),S95,E.J.,A,A.,M,J.,,J.(1998).Lifeeventsandpost-trau-maticstress:Thedevelopmentofanewmeasureforchildrenandadolescents.PsychologicalMedi-,1275,D.M.,S,J.,R,J.S.,&D,L.C.(2008).Riskandprotectivefactorsassociatedwithtrajectoriesofdepressedmoodfromadolescencetoearlyadulthood.JournalofConsultingandClinicalPsychology,E.K.,L,Z.,&K,C.A.(2012).Socialconnectednessandone-yeartrajectoriesamongsuicidaladolescentsfollowingpsychiatrichospitalization.JournalofClinicalChildandAdolescentPsychology,214,M.,&W,P.(2007).Inthecourseofalifetime:Tracingreligiousbelief,practice,andchange.Berkeley:UniversityofCalifornia,M.,D,J.,CI.,&J,M.(2006).Riskandprotectivefac-torsformedicallyserioussuicideattempts:Acomparisonofhospital-basedwithpopulation-basedsamplesofyoungadults.AustralianandNewZealandJournalofPsychiatry,87,E.(1897).Lesuicide.paris:F.,J.S.,M,C.,W,A.,,C.M.,R,D.,F,C.,ETAL.(1993).Developmentduringadolescence:Theimpactofstage-environmentÞtonyoungadolescentsÕexperiencesinschoolsandinfami-TheAmericanPsychologist,90,M.E.,A,D.M.,&R,M.D.(2007).Protectivefactorsandsui-cideriskinadolescentswithahistoryofsexualTheJournalofPediatrics,482,M.E.,ND.,&P,C.L.(2003).Peerharassment,schoolconnectedness,andacademicachieve-TheJournalofSchoolHealth,311,M.E.,&R,M.D.(2006).Suicidalityamonggay,lesbianandbisex-ualyouth:Theroleofprotectivefactors.ofAdolescentHealth,662,N.I.,&C,S.W.(2012).Socialneuroscienceandhealth:Neuro-physiologicalmechanismslinkingsocialtieswithphysicalhealth.NatureNeuroscience,669,B.L.(2001).Theroleofpositiveemotionsinpositivepsychology:Thebroaden-and-buildtheoryofpositiveemotions.AmericanPsychologist,218,B.L.,&J,T.(2002).Positiveemotionstriggerupwardspiralstowardemotionalwell-being.PsychologicalScience,K.E.(2012).Theneglectedroleofpositiveemotioninadolescentpsychopathol-ClinicalPsychologyReview,467,S.K.,&InstituteofMedi-cine(U.S.),CommitteeonPathophysiology&PreventionofAdolescent&AdultSuicide.Reducingsuicide:AnationalimperativeWashington,DC:NationalAcademiesPress.,D.,T,P.H.,ZA.,&H,L.R.(1996).Therelationoffamilyfunctioningtoviolenceamonginner-cityminorityyouth.JournalofFamilyPsychology,J.A.,E,M.E.,,S.L.,&N,D.(2007).Socialisolation,psychologicalhealth,andprotectivefactorsinadolescence.,D.(2009).RelationshipbetweensenseofbelongingasconnectednessandsuicideHITLOCKETAL inAmericaIndians.ArchivesofPsychiatricNurs-,65,B.J.,&G,M.S.(2008).ImpactofmodelingonadolescentsuicidalPsychiatricClinicsofNorthAmerica,T.R.,&W,J.T.(1998).SerotoninandneuropeptidesinafÞliativebehav-BiologicalPsychiatry,207,T.E.(2006).Whypeoplediebysui-.Cambridge:HarvardUniversityPress.,T.E.,C,Y.,FK.K.,W,T.K.,S,N.B.,BM.T.,ETAL.(2005).FourstudiesonhowpastandcurrentsuicidalityrelateevenwhenÒevery-thingbutthekitchensinkÓiscovaried.JournalofAbnormalPsychology,291,J.W.,&F,X.(2009).Vic-timizationbypeersandadolescentsuicideinthreeUSsamples.TheJournalofPediatrics,J.,P,R.,H,D.,C,S.,C,A.,&O,L.(2010).Therelativeinßuenceofdifferentdomainsofsocialconnectednessonself-directedviolenceinado-JournalofYouthandAdolescence,S.,H,C.C.,BK.A.,D,L.,K,R.A.,&S,G.(2006).Thesocialcontextofadolescentsuicideattempts:Interactiveeffectsofparent,peer,andschoolsocialrelations.SuicideandLife-Threaten-ingBehavior,386,D.(2001).UnderstandingwhatworksandwhatdoesnÕtinreducingadolescentFamilyPlanningPerspectives,276,R.M.,&R,S.B.(1995).Measuringbelongingness:Thesocialconnected-nessandthesocialassurancescales.JournalofCounselingPsychology,232,K.M.,&S,S.(2012).Up-regulatingpositiveemotionsineverydaylife:Strategies,individualdifferences,andassociationswithpositiveemotionandwell-JournalofResearchinPersonality,504,J.E.(2009).Preventionfactorsforsuicideideationamongabusedpre/earlyadolescentyouths.InjuryPrevention,278,J.E.,C,A.E.,&H,M.E.(2011).Suicidalideation,friend-shipswithdelinquents,socialandparentalconnectedness,anddifferentialassociationsbysex:Findingsamonghigh-riskpre/earlyadoles-centpopulation.,299,D.E.,&K,C.A.(2001).Parentalfacilitationofadolescentmentalhealthserviceutilization:AconceptualandempiricalClinicalPsychology:ScienceandPractice,D.D.,K,J.T.,J,J.D.,,L.W.,R,M.A.,&G,G.A.(2012).CaringLettersProject:Amilitarysuicidepreventionpilotprogram.,S.L.,M,S.D.,&TJ.K.(2011).SuicidalityanddepressionamongAfricanAmericanadolescents:Theroleoffamilyandpeersupportandcommunityconnectedness.AmericanJournalofOrthopsychiatry,108,C.,&F,C.(2004).Schoolconnectednessandthetransitionintoandoutofhealth-riskbehavioramongadolescents:Acom-parisonofsocialbelongingandteachersupport.TheJournalofSchoolHealth,284,J.A.,&B,A.G.(2001).Arandomizedcontrolledtrialofpostcrisissui-cideprevention.PsychiatricServices,828,D.,S,M.,,S.A.,&R,M.D.(1997).Psy-chosocialcorrelatesofhealthcompromisingbehaviorsamongadolescents.HealthEducation,37,B.D.(2002).Childhoodexperi-enceandtheexpressionofgeneticpotential:Whatchildhoodneglecttellsusaboutnatureandnurture.BrainandMind,79,B.(1992).Beyondrationalchoice:ThesocialdynamicsofhowpeopleseekAmericanJournalofSociology,1096,M.D.,R,M.D.,&BR.W.(1997).ProtectingagainsthopelessnessandsuicidalityinsexuallyabusedAmericanIndianadolescents.JournalofAdolescentHealth,400,A.,S,K.,G,D.,P,M.,G,D.,T,X.,ETAL.(2012).AssociationsbetweensuicidalhighschoolstudentsÕhelp-seekingandtheirattitudesandperceptionsofsocialenvironment.JournalofYouthandAdolescence,1312.,A.R.,W,P.A.,PM.,S,K.,G,D.B.,XETAL.(2012).EmotionregulationdifÞculties,adultrelationships,andsuicideattemptsamonghighschoolstudentsinunderservedcom-JournalofYouthandAdolescence,M.J.,B,J.,S,A.,,T.D.,&G,W.L.(2000).Peerfunctioning,familydysfunction,andpsy-chologicalsymptomsinariskfactormodelforadolescentinpatientsÕsuicidalideationseverity.JournalofClinicalChildPsychology,392,A.L.,H,E.S.,A,J.J.,&A,S.(2008).Engage-26CONNECTEDNESSAND mentasßourishing:ThecontributionofpositiveemotionsandcopingtoadolescentsÕengagementatschoolandwithlearning.Psychologyinthe,419,M.D.,B,P.S.,BR.W.,B,K.E.,H,K.M.,J,J.,ETAL.(1997).Protectingadolescentsfromharm:Findingsfromthenationallongitudinalstudyonadolescenthealth.TheJournalofAmericanMedi-calAssociation,823,M.D.,H,L.J.,&BR.W.(1993).Theimpactofcaringandcon-nectednessonadolescenthealthandwell-being.JournalofPaediatricsandChildHealth,D.J.,D,F.P.,&W,C.J.(2007).Whenandhowdoyoungpeo-pleseekprofessionalhelpformentalhealthTheMedicalJournalofAustralia(7),35,R.N.,&R,K.(2003).Under-standingtheinßuenceofperceivednormsonCommunicationTheory,184,R.(1981).PerceptionofSupportInventory.Unpublishedmanuscript.Washing-ton,DC:ChildrenÕsNationalMedicalCenter.,R.J.,R,S.W.,&,F.(1997).Neighborhoodsandviolentcrime:AmultilevelstudyofcollectiveefÞcacy.,918Ð924.,K.,P,A.R.,P,M.,&W,P.A.(2012).ThreescalesassessinghighschoolstudentsÕattitudesandperceivednormsaboutseekingadulthelpfordistressandsuicideconcerns.SuicideandLife-ThreateningBehavior,157,C.E.,&S,R.M.(1999).Helpingothershelpsoneself:Responseshifteffectsinpeersupport.SocialScienceand,1563,L.A.(1991).ValidationstudiesofthePEPSS,ascaletomeasureemotionalsupportamongadolescents.JournalofAdolescentResearch,316Ð335.,L.(2010).Adualsystemsmodelofadolescentrisk-taking.,216,M.G.,&L,S.M.(2004).Trajectoriesofimpairment,socialsupport,anddepressivesymptomsinlaterlife.TheJournalsof,S238,R.J.(1998).Tiesthatbind:Measurement,demographics,andsocialconnect-PoliticalBehavior,53,K.,&M,B.T.(2005).Connectedness:Areviewoftheliteraturewithimplicationsforcounseling,assessment,andJournalofCounselingandDevelopment,191U.S.DepartmentofHealthandHumanServices(HHS),OfÞceoftheSurgeonGeneral,andNationalActionAllianceforSuicidePreven-tion.(2012).2012Nationalstrategyforsuicidepre-vention:GoalsandobjectivesforactionWashington,DC:Author.U.S.DepartmentofHealthandHumanServices(HSS),PublicHealthService.(2001).Nationalstrategyforsuicideprevention:Goalsandobjectivesforaction.Rockville,MD:Author.,T.W.(2010).Socialnetworksandhealth:Models,methods,andapplications.NewYork:OxfordUniversityPress.,K.A.,W,T.K.,,K.C.,B,S.R.,SE.A.,&J,T.E.(2010).Theinterpersonaltheoryofsuicide.PsychologicalReview,E.E.,&S,R.S.(2001).Journeysfromchildhoodtomidlife:Risk,resilience,andrecovery.Ithaca,NY:CornellUniversity,J.(2006).Youthperceptionsoflifeatschool:Contextualcorrelatesofschoolconnectednessinadolescence.AppliedDevelop-mentalScience,13,P.A.,B,C.H.,I,J.,,W.,S,K.,G,J.,ETAL(2008).Randomizedtrialofagatekeeperpro-gramforsuicideprevention:1-yearimpactonsecondaryschoolstaff.JournalofConsultingandClinicalPsychology,104,P.A.,B,C.H.,L,M.,S,K.,P,M.,YQ.,W,E.,&W,W.(2010).Anout-comeevaluationoftheSourcesofStrengthsui-cidepreventionprogramdeliveredbyadolescentpeerleadersinhighschools.AmericanJournalofPublicHealth,1653Ð1661.,E.S.,R,J.R.,WC.A.,&W,J.L.(2010).CollegestudentsÕintentionstoseekhelpforsuicidalideation:Accountingforthehelp-negationeffect.andLife-ThreateningBehavior,438,R.,S,H.,&EA.(2011).Doschoolsdifferinsuiciderisk?Theinßuenceofschoolandneighbourhoodonattemptedsuicide,suicidalideationandself-harmamongsecondaryschoolpupils.cineCentralPublicHealth,874.ManuscriptReceived:March13,2013RevisionAccepted:August21,2013HITLOCKETAL