approachrrrr333

ResearchTrendsTheImpactofaSuicidePreventionStrategyonReducingtheEconomicCostofSuicideintheNewSouthWalesConstructionIndustryChristopherM.
Doran1,RodLing2,JorgenGullestrup3,SarahSwannell4,andAllisonMilner51SchoolofHuman,HealthandSocialSciences,CentralQueenslandUniversity,Brisbane,QLD,Australia2HunterMedicalResearchInstitute,UniversityofNewcastle,NewLambton,NSW,Australia3MatesinConstruction,SpringHill,QLD,Australia4MaterHillPsychologyServices,Woolloongabba,QLD,Australia5McCaugheyVicHealthCentreforCommunityWellbeing,MelbourneSchoolofPopulationandGlobalHealth,TheUniversityofMelbourne,VIC,AustraliaAbstract.
Background:Littleresearchhasbeenconductedintothecostandpreventionofself-harmintheworkplace.
Aims:Toquantifytheeco-nomiccostofself-harmandsuicideamongNewSouthWales(NSW)constructionindustry(CI)workersandtoexaminethepotentialeconomicimpactofimplementingMatesinConstruction(MIC).
Method:Directandindirectcostswereestimated.
Effectivenesswasmeasuredusingtherelativeriskratio(RRR).
InQueensland(QLD),relativesuicideriskswereestimatedfor5-yearperiodsbeforeandafterthecommencementofMIC.
ForNSW,thedifferencebetweentheexpected(i.
e.
,usingNSWpre-MIC[2008–2012]suiciderisk)andcounterfactualsuicidecases(i.
e.
,applyingQLDRRR)providedanestimateofpotentialsuicidecasesavertedinthepost-MICperiod(2013–2017).
Resultswereadjustedusingtheaverageuptake(i.
e.
,9.
4%)ofMICactivitiesinQLD.
EconomicsavingsfromavertedcaseswerecomparedwiththecostofimplementingMIC.
Results:Thecostofself-harmandsuicideintheNSWCIwasAU$527millionin2010.
MICcouldpotentiallyavert0.
4suicides,1.
01fullincapacitycases,and4.
92shortabsences,generatingannualsavingsofAU$3.
66million.
ForeveryAU$1invested,theeconomicreturnisapproximatelyAU$4.
6.
Conclusion:MICrepresentsapositiveeconomicinvestmentinworkplacesafety.
Keywords:self-harm,suicide,prevention,workplace,MatesinConstruction,economic,costSuicidehasbeendefinedasadeliberateactofself-harmperformedwiththeexpectationthatitwillbefatal,whileasuicideattempthasbeendefinedasanonfatalactofself-harm(WorldHealthOrganization,2006).
Foreverydeathbysuicide,therearebetween10and20attemptedsuicides(WorldHealthOrganization,2006).
Suicidalbehaviorhasgainedrecognitionworldwideasasignificantpublichealthproblem.
InAustralia,suicideisaleadingcauseofdeathwith2,273deaths(agedover15years)in2011(1,747maledeathsand546femaledeaths),representing1.
5%ofalldeathsovertheageof15(Austral-ianBureauofStatistics,2013b).
Mostdeathsbysuicideareamongpeopleofworkingage,andsuicideisthelead-ingcauseofdeathformenaged25–44yearsandwomenaged25–34years(AustralianBureauofStatistics,2013b).
Anemergingareaofinterestinsuicideresearchistheimpactofemploymentstatusandindustryonratesofsu-icide(Milner,Spittal,Pirkis,&LaMontagne,2013;Sui-cidePreventionAustralia,2014;WorldHealthOrganiza-tion,2006).
Althoughbeingemployedisassociatedwithareducedriskofsuicideoverall,currentevidencesuggestssuicideratesaredifferentiallydistributedacrossindustryandoccupationalgroups.
ArecentreviewbyMilneretal.
(2013)onsuicidebyoccupationfoundastepwisegradientinrisk,withthelowest-skilledoccupationsbeingatgreaterriskofsuicidethanthehighest-skill-levelgroup(Milneretal.
,2013).
InaseparateanalysisusingdatafromtheNationalCoronialInformationSystem(NCIS),Milneretal.
confirmedthatthisgradientalsoapplieswithintheconstructionindustry(CI;Milner,Niven,&LaMontagne,2014).
Further,consistentwithpreviousresearch,ratesofsuicideamonglower-skilledworkerswerehigherthanratesinthegeneralmalepopulation(Heller,Hawgood,&DeLeo,2007).
122Unfortunately,thepreventionofsuicidehasnotbeenadequatelyaddressedinsocietyortheworkforce,perhapsowingtoalackofawarenessofsuicideasamajorprob-lemandthetabooinmanysocietiestodiscussitopen-ly(Germain,2014;WorldHealthOrganization,2006).
Mann,Apter,andBertolote's(2005)systematicreviewofsuicidepreventionstrategiesfoundthatarangeofsuicidepreventionstrategieshavebeenproposeddespiteknowl-edgedeficitsabouttheeffectivenessofsomecommonkeycomponents.
Theauthorssuggestthatthemostpromisinginterventionsaimtoimprovephysicianeducationaboutsuicide,meansrestriction(i.
e.
,reducingaccesstolethalmethods),andgatekeepereducation(i.
e.
,wheretherolesofgatekeepersareformalizedandpathwaystotreatmentarereadilyavailable;Mannetal.
,2005).
MatesinConstruction(MIC)isanexampleofamul-tifacetedworkplacesuicidepreventionstrategydevelopedinAustralia.
MICwasestablishedin2008bytheBuildingEmployeesRedundancyTrusttopreventsuicideintheCI(Gullestrup,Lequertier,&Martin,2011).
MICisamulti-modalpreventionandearlyinterventionprogram,consist-entwiththenational"livingisforeveryone"suicidepre-ventionstrategy(DepartmentofHealthandAgeing,2007)andwithMrazekandHaggerty'sspectrumofpreventionandintervention(1994).
MIChasthreemaincomponents:generalawarenesstraining(GAT);connectortraining;andappliedsuicideinterventionskillstraining(ASIST)(Gullestrupetal.
,2011).
GATinvolvesa1-hrtrainingsessionprovidedbyaccreditedtrainerstoconstructionworkersonsiteswiththeaimsofincreasingawarenessofsuicideasaworkplacehealthandsafetyissue,improvingknowledgeofwarningsigns,andencouragingworkerstoseeksupport.
Connec-tortraininginvolvesa4-hrtrainingsessionprovidedbyMIC.
Theroleofaconnectoristokeepcoworkerssafewhileconnectingthemtohelp,thatis,toanASIST-trainedworker,MICfieldofficer,orcasemanager.
ASISTworkersundergoanintensive2-daytrainingcoursetoenablethemtoidentifycuesandrespondappropriatelytocallsforhelpwiththegoalofreachingacontractorsafeplaninvolvingextrahelpandsafety.
MICaccreditedsitesoremployersalsoreceivepromotionalmaterialsandaccesstootherMICprogramsincludinga24/7helpline.
AlthoughMICoriginatedinQueensland(QLD),itisequallyrelevanttootherAustralianstatesandterritories.
Theaimofthecurrentstudywastoquantifytheeconom-iccostofself-harmandsuicideamongNewSouthWales(NSW)CIworkersandtoexaminethepotentialeconomicimpactofimplementingMICintheNSWCI.
MethodRatesofSuicideandSelf-HarmintheCISuicidedatawereobtainedfromtheNationalCoronialIn-formationSystem(NCIS)fortheperiod2001–2012.
OnlymalesubjectswereincludedinthisstudybecauseofthesmallnumbersofwomenintheCIwhosuicidedandcon-sequentconfidentialityissueswithreportingsmallsamplesizes.
OccupationalinformationwascodedaccordingtotheAustralianandNewZealandStandardClassificationofOccupations(ANZSCO)andtheAustralianandNewZealandStandardIndustrialClassifications(AustralianBureauofStatistics,2006;AustralianBureauofStatis-tics,2013a).
OccupationscodedasbeingintheCIfellintothreemajorgroups:techniciansandtradesworker(AN-ZSCOmajorlevel3);machineoperators(ANZSCOmajorlevel7);and,laborers(ANZSCOmajorlevel8).
SuicideintheCIwasfurtherclassifiedashigherskill(ANZSCOma-jorlevel3)andlowerskill(ANZSCOmajorlevels7and8).
Thisapproachtothecodingandclassificationofskilllevelfollowstheproceduredescribedinapreviousjournalarticle(Milner,Niven,etal.
,2014).
EstimatingCostofSelf-HarmandSuicideintheNSWCITheanalysisusedacostingmethodologyendorsedbytheNationalOccupationalHealthandSafetyCommissionandusedina2012reportbySafeWorkAustralia.
Bothdirectandindirectcostswereconsideredforarangeofeconom-icagents(includingemployers,workers,andthegovern-ment)andbyseverityofinjury.
ForCIworkers,totalcostsofself-harmandsuicidewereestimatedbymultiplyingaverageindirectanddirectcostsbycasesofself-harmandsuicides.
Asummaryofthekeyparameters,assumptions,anddatasourcesforcostitemsisprovidedinTable1.
Theclas-sificationstructureforeconomiccostsisbasedonsixcon-ceptualcostgroups:productiondisturbancecosts;humancapitalcosts;medicalcosts;administrativecosts;transfercosts;andothercosts.
Productiondisturbancecostsreflectshort-termimpactsuntilproductionisreturnedtopre-in-cidentlevelsandincludesthevalueoflostproductionandstaffturnovercosts.
Humancapitalcostsaremeasuredbyconsideringthevalueofpotentialfutureearningsfromtimeofinjurytoretirementage(i.
e.
,<65years)assumingadiscountprofileandproductivityloss.
Thediscountpro-fileconsidersthelikelychangesinthevalueofmoneyovertimebyincludingtheopportunitycostofsaving(4.
1%;ReserveBankofAustralia,2014)andtherateofinflation(2.
8%;AustralianBureauofStatistics,2013c).
Medicalcostsareexpensesincurredbyworkersandthecommuni-tythoughmedicaltreatment.
Administrativecostsincludeinvestigationcosts,travelcosts,andfuneralcosts.
Trans-fercostsrefertoadeadweightloss,measuredastheval-ueoftaxationreceiptsforegone,equivalentto28.
75centsineveryforegonetaxdollar(AccessEconomics,2009).
Othercostsincludedinthisanalysisarecostofcarersandaids/modificationsforfullincapacitycasesandthecostofpostventionservicesforfatalities.
Threelevelsofseverityareusedinthisanalysis:shortabsenceinvolvinglessthan5daysoffwork;fullincapac-itythatresultsintheindividualbeingpermanentlyunabletoreturntowork;andafatality.
123Table1.
SummaryofkeyparametersusedincostinganalysisCostcategoryDescriptionSourceProductiondisturbancecostsValueoflostproductionAverageweeklyearnings(AWE)*averagedurationofabsence(byseveritycategory);AWE*averagedurationofabsence*0.
4AustralianBureauofStatistics,2014;SafeWorkAustralia,2012StaffturnovercostsThecostofreplacingexistingstaffaffectedbywork-relatedincidents(26weeksofAWE)andtrainingofnewstaff(2.
5weeksofAWE)AustralianBureauofStatistics,2014;SafeWorkAustralia,2012HumancapitalcostsLossofincomeForfullincapacityorfatality:lossofearningsfromtimeofinjurytoretirementage(i.
e.
,<65years),discountrate=4.
1%,inflationrate=2.
8%,productivityrate=1.
6%.
Forfullincapacity,futureearningsincludesaveragesocialwelfarepaymentsreceived(sincethesecontributetopost-injuryincome)AustralianBureauofStatistics,2013c;AustralianBureauofStatistics,2014;CommonwealthofAustralia,2010;ReserveBankofAustralia,2014LossofgovernmentrevenueForfullincapacityorfatality,taxationandotherrevenueforegonewhenworkersareunabletoworkduetowork-relatedincidentsAustralianBureauofStatistics,2014;ReserveBankofAustralia,2014SocialwelfarepaymentsSicknessandsocialwelfarepaymentsbornebythegovernmentforpeoplewithdisabilities(disabilitysupportpensionpaymentsofAU$700perfortnight[in2010dollars]discountedtopresentvalueovertheperiodbetweentheincidentandreducedlifeexpectancyDepartmentofHumanServices,2014MedicalcostsMedicalcostsAveragemedicalcostsfromnationaldatasetforcompensation-basedstatisticsSafeWorkAustralia,2012AdministrativecostsInvestigationcostsInvestigationcosts:Asaproxyforthecoststofirms,investigationandinspectioncostsreportedinjurisdictionalannualreportsareassumedtomatchthecosttoem-ployersforthesefunctionsSafeWorkAustralia,2012TravelexpensesPaymentsmadefortravelexpensestoworkers'compensationjurisdictionsbyclaim-ants(asaproxy,assumingthatcompensationisadequatetocovertheseexpenses)SafeWorkAustralia,2012FuneralexpensesAveragefuneralcostsareestimatedatAU$4,000SafeWorkAustralia,2012TransfercostsTransfercostsTheredistributionofpublicsectorresourcestocareforincapacitatedpersonincursdeadweightcostsonsociety–foreverydollaroftaxraised,about28.
75centsisabsorbedinthedistortionsinducedandtheadministrationofthetaxsystemAccessEconomics,2009OtherCarercostsForfullincapacity,theadditionalcostofcare(estimatedapplicabledisabilitysupportpensionpaymentsofAU$2,056perannum,discountedtopresentvalueovertheperi-odbetweentheincidentandreducedlifeexpectancy)SafeWorkAustralia,2012Costofaids,equipment,andmodificationsForfullincapacitycasesonly,thepresentvalueoffuturecostsforaidsandmodifi-cations(estimatedapplicabledisabilitysupportpensionpaymentsofAU$646perannum,discountedtopresentvalueovertheperiodbetweentheincidentandreducedlifeexpectancy).
SafeWorkAustralia,2012PostventioncostsCostassociatedwithbereavementforsixfamily/friends–estimatedatAu$14,058perperson;employercostassociatedwithprovidingcounselingandtimeoffworkforthreecolleagueswhomayhavewitnessedfatality–estimatedatAU$10,000fromtimeofincidenttoreturntofulldutiesComansetal.
,2013,Corso,Mercy,Simon,Finkelstein,&Miller,2007TheWorldHealthOrganizationestimatesforeveryemployeewhodiesbysuicide,another10–20willmakeasuicideattempt(17%resultinginapermanentdisabil-ityand83%innodisability;WorldHealthOrganization,2006).
Inthisanalysis,forevery15suicideattemptsthereisonefatality,2.
55(17%)fullincapacitycases,and12.
45(83%)shortabsencecases.
Correspondingdurationsofabsence(foruseincalculationofproductiondisturbancecosts)are0.
2weeksforshortabsenceand2.
6weeksforfullincapacityandfatality.
Costswerederivedusinganin-cidence-basedapproachwithcoststhataninjuryimposesinfutureyearsdiscountedto2010dollars.
EstimatingtheEffectivenessofMICInanevaluationoftheUSAirForcesuicidepreventionprogram,Knox,Litts,Talcott,Feig,andCaine(2003)cal-culatedtherelativeriskratio(RRR,ortheratioofrisk)ofsuicideforairforceemployees,beforeandafterthein-124Table2.
DataunderpinningmeasurementofeffectivenessofMatesinConstructionQLDpre-MICperiodNumberofsuicidedeathsConstructionindustryworkforceQLDpost-MICperiodNumberofsuicidedeathsConstructionindustryworkforce200336115,000200845174,050200451126,375200943172,475200541145,250201045164,950200630154,275201153168,625200749168,050201236161,325Total207708,950Total222841,425QLDpre-MICsuiciderate29.
20QLDpost-MICsuiciderate26.
38QLDpre-MICsuicideratelower95%CI25.
22QLDpost-MICsuicideratelower95%CI22.
91QLDpre-MICsuiciderateupper95%CI33.
18QLDpost-MICsuiciderateupper95%CI29.
85QLDRRRQLDRRR(lower95%CI)QLDRRR(upper95%CI)QLDpost-MICrate/QLDpre-MICrate0.
9040.
9090.
900Equivalentpercentagechangeinsuiciderisk–9.
6%–9.
1%–10.
0%NSWpre-MICperiodNumberofsuicidedeathsConstructionindustryworkforce200843224,925200937206,050201057212,700201142209,350201242215,475Total2211,068,500NSWpre-MICsuiciderate20.
68NSWpre-MICsuicideratelower95%CI17.
96NSWpre-MICsuiciderateupper95%CI23.
41NSWconstructionindustryworkforce2013–20171,050,518MeanLower95%CIUpper95%CIEst.
no.
ofsuicidesinNSW2013–2017noMIC(i.
e.
,NSWpre-MICrate)217189246Est.
no.
ofsuicidesinNSW2013–2017withMIC(i.
e.
,QLDRRR)196171221Est.
differenceinsuicidenumber(over5-yearperiod)21Est.
reductioninsuicides(peryear)4.
19Est.
reductioninsuicidesattributabletoMIC(peryear)0.
40Est.
reductioninfullincapacitycasesattributabletoMIC(peryear)1.
01Est.
reductioninshortabsencecaseattributabletoMIC(peryear)4.
92Note.
QLD=Queensland.
NSW=NewSouthWales.
MIC=MatesinConstruction.
RRR=relativeriskratio.
Est.
=estimated.
tervention.
Asimilarapproachwasusedinthisanalysis(Table2).
Inparticular,weusedQLDtimeseriesdataonsuicidecasesandestimatesoftheconstructionindustryworkforcetogenerateasuicideriskbeforeandaftercom-mencementofMICactivities.
TheQLDpre-MICperiodcovered2003–2007withthepost-MICperiodspanned2008–2012.
TheQLDRRRwasderivedwithPoissonre-gressionusinginformationonpre-andpostsuiciderisks.
ForNSW,thepre-MICperiodwasdefinedastheperi-od2008–2012withthepost-MICperiodspanning2013–2017.
Apre-MICperiodsuicideriskwascalculatedusingtimeseriesdataonNSWsuicidecasesandtheNSWCIworkforce.
Thepre-MICperiodsuicideratewasmul-tipliedbyanestimateoftheCIworkforce(andadjustedby100,000),toderiveanestimateofsuicidecasesfortheperiod2013–2017.
AkeyassumptioninthiscalculationistheriskofsuicideremainsstableintheabsenceofMIC.
Toestimatecounterfactualsuicidenumbersfortheperiod2013–2017,thecalculatedQLDRRRwasusedtoestimatethereducedriskandchangeinfatalitybysuicideamongNSWCIworkersovertheperiod2013–2017.
Thediffer-encebetweentheexpectedandcounterfactualsuicidecas-esprovidesanestimateofsuicidecasesavertedinNSWinthepost-MICperiod(2013–2017).
Table3providesanoverviewofMICconstructionactivitiesinQLDovertheperiod2008–2013.
Allactivi-125tieshaveincreasedsteadilysincethecommencementofMICwiththeaverageuptake(orpenetration)ofgeneralawarenesstraining(GAT)being9.
4%(i.
e.
,theaverageproportionoftheCIworkforceexposedtoGATactivitiesovertheperiodofinterest).
Thisrateof9.
4%isappliedtothedifferenceinexpectedandcounterfactualNSWsui-cidecases(derivedpreviously)toderiveanestimateofthechangeinsuicidecasesattributabletoMIC.
PotentialEconomicImpactofImplementingMICintheNSWCIThepotentialeconomicimpactofimplementingMICintheNSWCIisderivedbycomparingtheeconomicsav-ingsfromfewersuicideandsuicideattemptswiththecostofimplementingtheprogram.
Resultsareexpressedasaratioofbenefitstocostswithapositiveratiorepresentingapositiveeconomicinvestment.
SensitivityAnalysisThreedifferentsensitivityanalyseswereundertakentotesttherobustnessofresultstochangesinkeyparameters.
First,theproportionofsuicideattemptsresultinginfullincapacity(i.
e.
,17%ofsuicideattempts)wasvariedby±5percentagepoints.
Second,theattributionofMICtoavert-edsuicideandsuicideattempts(i.
e.
,9.
4%)wasincreasedby5and10percentagepointsof9.
4%.
Third,thediscountrate(i.
e.
,4.
11%)wasadjustedto0%,3%,and5%.
ResultsSuicideAmongQLDandNSWCIWorkersThenumberofsuicidedeathsamongQLDCIworkersovertheperiod2003–2012andNSWCIworkersovertheperiod2208–2013isprovidedinTable2.
TheaverageageofeachsuicidefatalityamongCIworkerswas36.
8yearsand37.
7yearsinQLDandNSW,respectively.
TotalCostofSuicideandSuicideBehaviorintheNSWCITheaveragecostassociatedwithanincidentinvolvingashort-termabsenceisestimatedtocostAU$925;eachself-harmincidentresultinginfullincapacityisestimatedtocostAU$2.
78million;andeachsuicideincidentresult-inginafatalityisestimatedtocostAU$2.
14million.
Thekeycostdriverinbothfullincapacitycasesandafatalityislostincome(andtaxes),and,forfullincapacityonly,theadditionalcostofwelfarepayments.
Giventheaverageageofeachsuicideis37.
7yearsinNSW,thisequatestoalossof27.
3years(65years37.
7years)inpotentialproduc-tiveemployment.
In2010,amongmaleCIworkersinNSWtherewere:145self-harmincidentsresultinginfullincapacity;710self-harmincidentsresultinginashortabsencefromwork;and57fatalitiesbysuicide.
Multiplyingthesenumbersbyaveragecostperincidentsuggeststhatthecostofself-harmandsuicideintheNSWCIwasAU$527millionin2010(Table4).
EffectivenessofMICintheNSWCITable2providestimeseriesdataonsuicidecasesandes-timatesoftheCIworkforceforbothQLDandNSW.
ForQLD,therelativesuicideriskratewas0.
9036,suggestingadecreasedriskinthepost-MICperiodof9.
64%(96%CI=9.
1–10.
0%).
Thepre-MICsuicideratewasestimatedat29.
20per100,000CIworkers(95%CI=25.
22–33.
18)andthepost-MICsuicideratewasestimatedat26.
38per100,000CIworkers(95%CI22.
91–29.
85).
ForNSW,thepre-MICsuicideriskwasestimatedat20.
68per100,000CIworkers(95%CI=17.
96–23.
41).
CombiningthisriskwiththeestimateoftheCIworkforcefortheperiod2013–2017(i.
e.
,1,050,518)suggestsanestimated217suicidecases(95%CI=189–246).
Coun-terfactualsuicidenumberswerederivedbyapplyingtheQLDRRR(i.
e.
,0.
9036)totheestimatednumberofcases(i.
e.
,217)resultingin196suicidecases,adifferenceof21casesoverthe5-yearperiodof2013–2017or4.
19fewersuicidesperyear.
Theanalysisattributedonly9.
4%ofthisdifferencetoMICtoreflecttheaverageuptakeofMICTable3.
MatesinConstructionactivitiesinQueensland,2008–2013YearQueenslandconstructionindus-tryworkforceGeneralawarenesstrainingGATcumulativeProportionofworkforceexposedtoGATConnectortrainingASISTtraining2008174,0507307300.
4%5502009172,4753,1743,9042.
3%227302010164,9505,4659,3695.
7%333542011168,6257,27416,6439.
9%478472012161,3259,58826,23116.
3%507522013161,3259,53035,76122.
2%79355Note.
NumbersbasedonMatesinConstructiondatabase(MatesinConstruction,personalcommunication)ASIST=appliedsuicideinterventionskillstraining.
GAT=generalawarenesstraining.
126generalawarenesstrainingintheQLDCIpost-MIC(i.
e.
,9.
4%).
TheresultssuggestthatifimplementedintheNSWCI,MICcouldpotentiallyavert0.
4suicides,1.
01suicideattemptsresultinginfullincapacity,and4.
92suicideat-temptsresultinginashortabsencefromwork.
PotentialEconomicImpactofImplementingMICintheNSWCIThepotentialeconomicimpactofimplementingMICintheNSWCIisanestimatedsavingofAU$3.
66millioneachyear(Table5).
ThemajorityofbenefitsareestimatedtoflowtothegovernmentwithasavingofAU$3.
56mil-lioneachyear.
IfthebudgetforrollingouttheMICpro-graminNSWisAU$800,000eachyear,thebenefit–costratioisequivalentto4.
6:1,thatis,foreveryAU$1invest-edthereisareturnofAU$4.
60,representingapositiveeconomicinvestmentofpublicfunds.
SensitivityAnalysisTable6providestheresultsofthesensitivityanalysis.
Allvariationsinkeyparametershavelittleimpactonthepos-itiveeconomicbenefitofMIC.
ReducingtheproportionTable4.
CostofsuicideandsuicidebehavioramongNewSouthWalesconstructionindustryworkers,2010(inAustraliandollars)EmployerWorkerGovernmentTotalShortabsenceProductiondisturbancecosts$258,834$0$0$258,834Humancapitalcosts$0$0$0$0Medicalcosts$354,825$0$0$354,825Administrativecosts$19,870$2,839$19,870$42,579Other$0$0$0$0Transfercosts$0$0$0$0Subtotal$633,529$2,839$19,870$656,238FullincapacityProductiondisturbancecosts$6,085,272$0$0$6,085,272Humancapitalcosts$0$0$364,416,944$364,416,944Medicalcosts$72,675$280,707$1,590,674$1,944,056Administrativecosts$345,061$53,053$398,114$796,227Other$0$0$13,080,842$13,080,842Transfercosts$0$0$18,127,194$18,127,194Subtotal$6,503,008$333,760$397,613,767$404,450,535FatalityProductiondisturbancecosts$2,386,381$0$0$2,386,381Humancapitalcosts$0$0$105,281,868$105,281,868Medicalcosts$28,500$20,777$117,734$167,010Administrativecosts$163,590$228,000$163,590$555,180Other$1,710,000$0$4,807,836$6,517,836Transfercosts$0$0$7,108,704$7,108,704Sub-total$4,288,471$248,777$117,479,732$122,016,979Total$11,425,008$585,375$515,113,369$527,123,752Table5.
PotentialeconomicsavingsfromimplementingMatesinConstructionintheNSWconstructionindustry(inAustraliandollars)TypeofincidentNumberofavertedincidentsAveragecostperincidentTotalcostsavings%SavingstogovernmentShortabsence4.
92$925$4,5543%Fullincapacity1.
01$2,782,597$2,806,58598%Fatality0.
40$2,140,649$846,70796%Total6.
33$3,657,84697%Note.
NSW=NewSouthWales.
127ofsuicideattemptsresultinginfullincapacityfrom17%to12%changesthebenefit–costratiofrom4.
57to3.
54;attributingahigherproportionofincidentsreducedtoMICfrom9.
4%to19.
4%increasesthenumberofavertedin-cidents,theeconomicsavings,andthebenefit–costratio.
Allvariationsofthediscountrateresultedinapositivebenefit–costratio.
DiscussionMainFindingsTheCIisthefourthmajorcontributortoAustralia'seco-nomicoutputatoverAU$100billioneachyear,closeto8%ofgrossdomesticproduct(AustralianBureauofSta-tistics,2012).
OuranalysishasquantifiedtheaveragecostofaCImaleworkerdyingfromsuicideatAU$2.
14mil-lion,witheachworkerlosinganaverageof27.
3yearsofpotentialproductiveemploymentand42yearsofpotentiallifelost.
ThetotaleconomiccostofsuicideandsuicidebehaviortotheNSWCIalonewasestimatedatAU$527millionin2010.
Giventhefactthatsuicideislargelypreventable(Sui-cidePreventionAustralia,2014),itissurprisingthatwork-placesuicidepreventionstrategiesarenotcommonprac-tice.
Ouranalysishasdemonstratedthatimplementingamultifacetedworkplacesuicideprevention,thatis,MIC,intheNSWCIcanpreventfatalitiesandinjuriesfromself-harmatthesametimeassavingscarceeconomicre-sources.
ComparedwiththecostofimplementingMIC,theresultssuggestthatforeveryAU$1investedthereisareturnofAU$4.
60tosociety.
Onlyalimitednumberofsimilarstudieshavebeencon-ducted.
BeyondBlueestimatedthereturnoninvestmentforemployersinvestinginamentallyhealthyworkplace(BeyondBlue,2014).
Theactionsselectedfortheanalysisincludedprevention,earlyintervention,andrehabilitation/returntoworkstrategies.
Theauthorsdemonstratedthataninvestmentinmentalhealthintheworkplacegivesrisetoanumberofbenefitstoorganizations,peoplewithmen-talhealthconditions,coworkers,andclients.
Theauthorsfoundthatforeverydollarspentonsuccessfullyimple-mentinganappropriateaction,thereisonaverage$2.
30inbenefitstobegainedbytheorganization(BeyondBlue,2014).
Comans,Visser,andScuffham(2013)conductedaneconomicevaluationoftheStandByResponseService,acommunity-basedcrisisinterventionprogramforpeo-plebereavedbysuicide.
TheauthorsfoundStandBytobeadominantstrategy,thatis,moreeffectiveandlessex-pensivethanthealternative.
VanSpijker,Majo,Smit,VanStraten,andKerkhof(2012)conductedacost-effective-nessanalysisofanonline,unguided,self-helpinterven-tionforreducingsuicidalideation.
Theauthorsalsofoundtheinterventiontobedominant(VanSpijkeretal.
,2012).
Lifeline(2013)publishedanevaluationreportofitsOnlineCrisisSupportChatService,acrisisinterventionservicetosupportsuicidepreventioninAustralia(Lifeline,2013).
Theauthorsreportedthatforeverydollarinvestedintheonlineservice,therewas$8.
40inbenefits(Lifeline,2013).
LimitationsandStrengthsTheanalysisreliedonNCISdatatoidentifyfatalitiesbysuicideintheCI.
Thisdatasourcemay,however,haveun-derestimatedthenumberofsuicidesoccurringinAustral-iaowingtocoronialandsystem-relatedissues(DeLeoetal.
,2010).
Onlymalesuicideswereexaminedinthisstudybecauseofthesmallnumberoffemalesuicidesandsubse-quentconfidentialityissues.
ThepotentialeffectivenessofMICintheNSWCIreliedonacombinationofQLDandNSWdata.
AnyunderreportingofsuicidesintheNCISdataarelikelytoinfluencethecalculatedsuiciderisk,theQLDRRR,andthepotentialeconomicimpactofMIC.
Table6.
Sensitivityanalysisofkeyparameters(inAustraliandollars)ParametervariedNumberofavertedincidentsEconomicsavingsperyearCostofMatesinConstructionBenefit–costratioProportionofsuicideattemptsresultinginfullincapacitySensitivity1=12%6.
33$2,832,654$800,0003.
54Baseline=17%6.
33$3,657,846$800,0004.
57Sensitivity2=22%6.
33$4,483,038$800,0005.
60AttributionofMatesinConstructiontoavertedsuicideandsuicideattemptsBaseline=9.
4%6.
33$3,657,846$800,0004.
57Sensitivity3=14.
4%9.
65$5,577,850$800,0006.
97Sensitivity4=19.
4%13.
00$7,514,603$800,0009.
39DiscountrateusedtoconvertfuturecoststopresentvalueBaseline=4.
11%6.
33$3,657,846$800,0004.
57Sensitivity5=0%6.
33$7,083,972$800,0008.
85Sensitivity6=3%6.
33$4,289,637$800,0005.
36Sensitivity7=5%6.
33$3,249,461$800,0004.
06128Theresultsofthesensitivityanalysesusinglowerandup-perconfidenceintervalsaroundtheeffectivenessratepro-videdsomeconfidencethattheresultsarerelativelyrobusttosuchvariations.
Althoughthisanalysiscloselyfollowedthemethodol-ogyadoptedbySafeWorkAustralia,certainmethodologi-calvariationswererequired.
Forexample,unliketheSafeWorkAustraliaapproach,thisanalysisincludedpostven-tioncostsassociatedwithsuicidebereavementandcoun-seling.
Evidencesuggeststhatpostventioncostsaresignif-icanttoboththecommunityandtheindustryandfailuretoincludethesecostswouldunderestimateanycostestimate(CarsonJ.
SpencerFoundationetal.
,2013;Comansetal.
,2013;SuicidePreventionAustralia,2014).
TheanalysisdidnotattempttoestimatethecostssavedbythetransferofknowledgegainedthroughMICtrainingattheworkplacetofamilyandfriendsoutsideofwork.
Therippleeffectsofothersuicidegatekeeperprogramshaveshownthatforeachpersontrainedanotherfivepeo-plehaveconversationswiththattraineeandlearnaboutbestpracticesinsuicideintervention(CarsonJ.
SpencerFoundationetal.
,2013;SuicidePreventionAustralia,2014).
Thistransferofknowledgethenincreasesthepo-tentialofsavinglivesoutsideofwork.
Itisalsoimportanttonotethattheanalysisperiodof2008–2012includesyearsoftheglobalfinancialcrisis(GFC),aperiodwhensuicidesaregenerallyexpectedtoincrease.
EvidencesuggeststhattheGFChadminimalim-pactinAustralia(Wettenhall,2011)andthereforewasnotconsideredasapotentialconfounderinthisanalysis.
Fur-ther,arecentinvestigationofsuicideduringtheGFCfoundasmallandrelativelytransientincreaseinsuicideamongtheemployedpopulation,andamuchlargerimpactonthoseoutofwork(Milner,Morrell,&LaMontagne,2014).
WorldHealthOrganizationevidenceontherelation-shipbetweensuicideandsuicidebehaviorhasbeenusedtoestimatetherelationshipbetweensuicide,suicidebe-havior,andinjuryresultingfromsuchbehavior(WorldHealthOrganization,2006).
TheserelationshipshavebeensupportedbyAustraliandata(Sladeetal.
,2009;SuicidePreventionAustralia,2014).
InanexaminationofsuicideintheAustralianCI,Mil-ner,Niven,etal.
(2014)foundthatthemajormethodofsuicidewasconsistentwiththatofthegeneralpopulation(e.
g.
,hangingfollowedbyexposuretocarbonmonoxide,andfirearms).
Ouranalysistestedvariationsinthepropor-tionofsuicideattemptsresultinginfullcapacitywithonlysmallchangesinthebenefit–costratio.
Ouranalysisadoptedaconservativeapproachtoout-comemeasurementbyattributingonly9.
4%ofavertedsuicideincidenttoMIC.
Thisratereflectstheaverageup-take(orpenetration)ofMICactivitiesbytheCIovertheperiodofinterest.
ItisimportanttonotethatonceMIChadbeenestablishedinQLD,thepenetrationratewasashighas22.
2%(Table3).
Itmaybeexpectedthatpenetrationrateswouldbehigherthan9.
4%inNSWgiventhelessonslearnedfromtheQLDexperience.
Nevertheless,thiscon-servativeapproachalsoallowsfortheacknowledgmentofotherimplementedpopulation-levelsuicideprograms(Su-icidePreventionAustralia,2014).
ConclusionThisstudyprovidesnewevidenceonthecostsassociat-edwithself-harmandsuicideintheNSWCIfortheyear2010andthepotentialreturnoninvestinginaworkplacesuicidepreventionstrategy.
TheseresultssuggestthatMICcansavelivesatthesametimeassavingscarceresources.
Itrepresentsapositiveeconomicinvestmentintowork-placesafetyintheNSWCI.
ReferencesAccessEconomics.
(2009).
Theeconomicimpactofyouthmentalillnessandthecosteffectivenessofearlyintervention.
Mel-bourne,Australia:Author.
AustralianBureauofStatistics.
(2006).
AustralianandNewZea-landstandardindustrialclassification(ANZSIC)(CatalogueNo.
1292.
0).
Canberra,Australia:Author.
AustralianBureauofStatistics.
(2012).
1301.
0–YearbookAus-tralia.
Canberra,Australia:Author.
AustralianBureauofStatistics.
(2013a).
AustralianandNewZealandstandardclassificationofoccupations(CatalogueNo.
1220.
0).
Canberra,Australia:Author.
AustralianBureauofStatistics.
(2013b).
Causesofdeath,Aus-tralia(CatalogueNo.
3303.
0).
Canberra,Australia:Author.
AustralianBureauofStatistics.
(2013c).
Consumerpriceindex,Australia(CatalogueNo.
6401.
0).
Canberra,Australia:Au-thor.
AustralianBureauofStatistics.
(2014).
Employeeearnings,benefitsandtradeunionmembership(CatalogueNo.
63100TS0002).
Canberra,Australia:Author.
BeyondBlue.
(2014).
Creatingamentallyhealthyworkplace:Returnoninvestmentanalysis.
Canberra,Australia:Australi-anGovernmentNationalHealthCommission.
CarsonJ.
SpencerFoundation,CrisisCareNetwork,NationalActionAllianceforSuicidePrevention,&AmericanAsso-ciationofSuiciology.
(2013).
Amanager'sguidetosuicidepostventionintheworkplace:10actionstepsfordealingwiththeaftermathofsuicide.
Denver,CO:Author.
Comans,T.
,Visser,V.
,&Scuffham,P.
(2013).
Costeffectivenessofacommunity-basedcrisisinterventionprogramforpeoplebereavedbysuicide.
Crisis,34,390–397.
CommonwealthofAustralia.
(2010).
Australiato2050:Futurechallenges.
Intergenerationalreport.
Canberra,Australia:Author.
Corso,P.
S.
,Mercy,J.
A.
,Simon,T.
R.
,Finkelstein,E.
A.
,&Miller,T.
R.
(2007).
Medicalcostsandproductivitylossesduetointerpersonalandself-directedviolenceintheUnit-edStates.
AmericanJournalofPreventiveMedicine,32,474–482.
DeLeo,D.
,Dudley,M.
J.
,Aebersold,C.
J.
,Mendoza,J.
A.
,Barnes,M.
A.
,Harrison,J.
E.
,…Ranson,D.
(2010).
Achiev-ingstandardisedreportingofsuicideinAustralia:Rationaleandprogramforchange.
MedicalJournalofAustralia,192,452–456.
DepartmentofHealthandAgeing.
(2007).
Livingisforeveryone(LIFE)framework.
Canberra,Australia:CommonwealthofAustralia.
DepartmentofHumanServices.
(2014).
Disabilitysupportpen-sion.
Canberra,Australia:Author.
Retrievedfromhttp://www.
humanservices.
gov.
au/customer/services/centrelink/disability-support-pensionGermain,M.
l.
(2014).
Work-relatedsuicide.
AnanalysisofUSgovernmentreportsandrecommendationsforhumanre-sources.
EmployeeRelations,36,148–164.
129Gullestrup,J.
,Lequertier,B.
,&Martin,G.
(2011).
MATESinconstruction:Impactofamultimodal,community-basedprogramforsuicidepreventionintheconstructionindustry.
InternationalJournalofEnvironmentalResearchandPublicHealth,8,4180–4196.
Heller,T.
S.
,Hawgood,J.
l.
,&DeLeo,D.
(2007).
Correlatesofsuicideinbuildingindustryworkers.
ArchivesofSuicideResearch,11,105–117.
Knox,K.
l.
,Litts,D.
A.
,Talcott,G.
W.
,Feig,J.
C.
,&Caine,E.
D.
(2003).
RiskofsuicideandrelatedadverseoutcomesafterexposuretoasuicidepreventionprogrammeintheUSAirForce:Cohortstudy.
BritishMedicalJournal,327,1–5.
Lifeline.
(2013).
SocialreturnoninvestmentforecastoftheLifelineOnlineCrisisSupportChatService.
Retrievedfromhttps://www.
lifeline.
org.
au/ArticleDocuments/121/Lifeline_Online_Crisis_Support_Chat_SROI_report_final.
pdf.
aspxMann,J.
J.
,Apter,A.
,&Bertolote,J.
(2005).
Suicidepreven-tionstrategies:Asystematicreview.
JournaloftheAmericanMedicalAssociation,294,2064–2074.
Milner,A.
,Morrell,S.
,&LaMontagne,A.
D.
(2014).
Econom-icallyinactive,unemployedandemployedsuicidesinAus-traliabyageandsexovera10-yearperiod:Whatwastheim-pactofthe2007economicrecessionInternationalJournalofEpidemiology,43,1500–1507.
Milner,A.
,Niven,H.
,&LaMontagne,A.
(2014).
SuicidebyoccupationalskilllevelintheAustralianconstructionindus-try:Datafrom2001to2010.
AustralianandNewZealandJournalofPublicHealth,38,281–285.
Milner,A.
,Spittal,M.
J.
,Pirkis,J.
,&LaMontagne,A.
D.
(2013).
Suicidebyoccupation:Systematicreviewandmeta-analysis.
TheBritishJournalofPsychiatry,203,409–416.
Mrazek,P.
J.
,&Haggerty,R.
J.
(1994).
Reducingtherisksformentalhealthdisorders:Frontiersforpreventiveinterven-tionresearch.
Washington,DC:NationalAcademyPressIn-stituteofMedicine.
ReserveBankofAustralia.
(2014).
Cashratetarget(2008-2014).
Retrievedfromhttp://www.
rba.
gov.
au/statistics/cash-rate/cash-rate-2008-2014.
htmlSafeWorkAustralia.
(2012).
Thecostofwork-relatedinjuryandillnessforAustralianemployers,workersandthecommuni-ty:2008–09.
Canberra,Australia:Author.
Slade,T.
,Johnston,A.
,Teesson,M.
,Whiteford,H.
,Burgess,P.
,Pirkis,J.
,&Saw,S.
(2009).
ThementalhealthofAustralians2.
Reportonthe2007NationalSurveyofMentalHealthandWellbeing.
Canberra,Australia:DepartmentofHealthandAgeing.
SuicidePreventionAustralia.
(2014).
Workandsuicidepreven-tion:Positionstatement.
Sydney,Australia:Author.
VanSpijker,B.
A.
J.
,Majo,C.
,Smit,F.
,VanStraten,A.
,&Kerk-hof,A.
J.
(2012).
Reducingsuicidalideation:Cost-effective-nessanalysisofarandomizedcontrolledtrialofunguidedweb-basedself-help.
JournalofMedicalInternetResearch,14,e141.
Wettenhall,R.
(2011).
Globalfinancialcrisis:TheAustralianexperienceininternationalperspective.
PublicOrganizationReview,11,77-91.
WorldHealthOrganization.
(2006).
Preventingsuicide:Are-sourceatwork.
Geneva,Switzerland:Author.
ReceivedDecember14,2014RevisionreceivedAugust2,2015AcceptedAugust8,2015PublishedonlineDecember23,2015AbouttheauthorsProfessorDoranholdsaBachelorofEconomics(Honors)degreeandaPhDinhealtheconomics.
HeisaninternationalexpertintheareaofprioritysettingandisrecognizedasoneofAustralia'sleadinghealtheconomists.
Hisresearchhasmadeanimpactatthelocal,national,andinternationallevel.
Dr.
RodLingholdsaBachelorofCommerce(Economics)degreeandaPhDinsociology.
HeisahealtheconomistemployedattheHunterMedicalResearchInstitute,NSW,Australia.
Hismainresearchinterestsarementalhealthandeconomicevaluation.
JorgenGullestrupistheinauguralmanagerofMATESinCon-structionfollowing13yearsasaunionofficialintheQueenslandconstructionindustry.
Heholdsagraduatecertificateinsuicideprevention,adiplomainsocialwelfare,andcommunitydevelop-mentandanassociatediplomainlaborstudies.
Dr.
SarahSwannellisaregisteredpsychologistinprivatepracticeandwiththeQueenslandGovernment.
SheholdsaBachelorofSocialScience(Psychology)(Honors)degreeandaPhDinpsy-chiatry.
Dr.
Swannell'smainclinicalinterestsareself-destructivebehaviors,trauma,anddissociation.
Shehaspublishedinbothnationalandinternationalpeer-reviewedjournals.
Dr.
AllisonMilnerisaseniorresearchfellowattheMcCaugheyVicHealthCentreforCommunityWellbeing,MelbourneSchoolofPopulationandGlobalHealth,Australia.
Herresearchinterestsincludetheinfluenceofmacro-levelemploymentcharacteristics,qualityofwork,jobstress,andunemploymentasdeterminantsofmentalhealthandsuicidality.
Shehasanhonorsdegreeinpsy-chology,aMasterofEpidemiology,andaPhDinSuicidology.
ChristopherM.
DoranSchoolofHuman,HealthandSocialSciencesCentralQueenslandUniversityLevel4,160AnnStreetBrisbane4000AustraliaTel.
+6141293-5084E-mailc.
doran@cqu.
edu.
au