Eventhestrongestofindividualriskfactorsarepoorpredictorsofsuicideattempts.Manyriskfactorsrelyon subjectiveorself-reportedmeasuresandunderreportingcontributestopoorpredictionaccuracy.Inthemidst oftheongoingVeteransuicidecrisis,weclearlyneednew,objectivepredictorsofriskthatcanbeimplemented acrosstheVAsystem.Previousfindingsindicatethatbehavioralandneuroimagingcorrelatesofdecision- makingandcognitivecontrolarepromising,objectivemarkersofsuicidalrisk.Thoughintriguing,thesefindings arefromresearchconductedprimarilyindepressedcivilians.Itisunknownifthesepotentialmarkersofsuicide riskwillgeneralizetoatransdiagnosticsampleofVeterans.TheprimaryresearchobjectiveofthisCDA-2isto evaluatewhetherstructuralneuroimaging,andfunctionalneuroimagingduringdecision-makingandcognitive controltaskscanbeusedtoidentifybrain-behaviorcorrelatesofsuicidalthoughtsandbehaviorsinVeterans. Suicidalthoughtsandbehaviorsareassociatedwithshort-sighted,maladaptive,decision-makingthatis accompaniedbydifferencesinstriatal-prefrontalrewardcircuitry.However,studiesthathavecompared decision-makingtaskperformancebetweenindividualswhohavemademoreorganized,lethalattemptsat self-harmvs.thosemakinglessseriousattemptshaveidentifieddifferencesintheirbehaviorthatmaysupply morenuancedinformationaboutsuicidalseverity.Interestingly,thedecision-makingofseriousattemptersis actuallylessimpulsive.Infact,decisionmakinginseriousattemptersislessimpulsiveevenrelativetohealthy controls.Thisimpliesthatagreatercapacityforcognitivecontrolmayfacilitate,ratherthanprotectagainst, harminsuicidalindividuals.Thesepotentialobjectivemarkersofsuicidalthoughtsandbehaviorsandsuicidal riskseverityarepromisinganddemandfurtherevaluation.Theproposedbehavioralandneuroimaging methodologies(diffusionimaging,morphometryanalyses,univariateandfunctionalconnectivityneuroimaging) includedinthisproposalhavebeensuccessfullyappliedbythecandidate,acognitiveneuroscientist,in previouspublishedwork.Thecandidate?sexploratoryresearchobjectiveistoapplymachinelearning algorithmstoablendeddatasetofparticipants?electronichealthdataandneurocognitivemarkersina preliminarypredictivemodelingproject.Thisexploratoryobjectivewillprovidefoundationaldataforfuture Merit-fundedworkfocusedonthedevelopmentofclinicallyimplementablesuicideriskpredictiontools.Though theproposedneurocognitiveresearchhasthepotentialtoimprovetheeffectivenessofsuicidescreeningand treatment,thecandidateacknowledgesthatalltoooften,insightsfromneurosciencefailtoimproveclinical carebecausefewneuroscientistshavetheclinicalperspectiveneededtotranslatemechanisticknowledgeinto clinicalinnovations.TheprotectedtimefundedbythisCDAawardwillallowthecandidatetoparticipatein activitiesimpartingauniquecombinationofskillsandperspectivesthatwillallowthecandidatetobridgebasic andclinicalscienceintheserviceofreducingVeteransuicide.TheCDAtrainingplanbuildscompetenciesin threeareas:1)methodologiesandpracticalskillsforsuicideresearch,2)psychiatricassessmentandclinical perspective,and3)traininginbioinformaticsandmachinelearningmethodologies.Thecandidateiswell- establishedwithintheVAsystemholdingaHealthScienceSpecialistpositionsince2015.Thecandidate?s mentorshipteamiscomprisedprimarilyofVAclinician-scientistsactivelyworkingwithsuicidalorhigh-risk populationsandarewell-qualifiedtomentorthecandidatetowardthecareergoalofbecomingaleaderin translationalVeteransuicideresearch.
RecentVeteransarenowmorelikelytodiebysuicidethancombat.TheproposedCDA-2researchwill generatenovelbiologicalandbehavioralmakersofsuiciderisk,andwillconductapreliminarydemonstration oftheiruseinpredictivemodelsofsuicidalthoughtsandbehaviors.Methodsusedtodefineriskfactorsare implementedinatransdiagnosticVeteransample.Importantly,definingriskfactorsindependentofpsychiatric populationwillincreasegeneralizabilityofstudyfindingstoawiderVeteranpopulation.Thesenovelsuicide riskfactorswillidentifyhigh-riskVeteransthatmaynothavebeenidentifiedbytraditionalsuicideassessment methods.ThismeansthatmoreVeteransatriskforsuicidewillreceivelife-savinginterventions,thus,this researchisconsistentwiththeVA?smissiontocombatVeteransuicide.
