Reforming private drug coverage in Canada: Inefficient drug benefit design and the barriers to change in unionized settings

Citation for this paper:

O’Brady, S., Gagnon, M. & Cassels, A. (2015). Reforming private drug coverage in

Canada: Inefficient drug benefit design and the barriers to change in unionized

settings. Health Policy, 199, 224-231.

http://dx.doi.org/10.1016/j.healthpol.2014.11.013

UVicSPACE: Research & Learning Repository

_____________________________________________________________

Faculty of Human and Social Development

Faculty Publications

_____________________________________________________________

Reforming private drug coverage in Canada: Inefficient drug benefit design and the

barriers to change in unionized settings

Sean O’Brady, Marc-André Gagnon, Alan Cassels

2015

© 2014 The Authors. Published by Elsevier Ireland Ltd. This is an open access

article underthe CC BY-NC-ND license (

http://creativecommons.org/licenses/by-nc-nd/3.0/

).

This article was originally published at:

http://dx.doi.org/10.1016/j.healthpol.2014.11.013

ContentslistsavailableatScienceDirect

Health

Policy

jo u r n al h om ep age :w w w . e l s e v i e r . c o m / l o c a t e / h e a l t h p o l

Reforming

private

drug

coverage

in

Canada:

Inefficient

drug

benefit

design

and

the

barriers

to

change

in

unionized

settings

Sean

O’Brady

_,

_Marc-André

_Gagnon

_,

_Alan

_Cassels

a_{Écolederelationsindustrielles,UniversitédeMontréal,Montreal,Quebec,Canada}

b_{InteruniversityResearchCentreonGlobalizationandWork(CRIMT),Montreal,Quebec,Canada}

c_{SchoolofPublicPolicyandAdministration,CarletonUniversity,Ottawa,Ontario,Canada}

d_{FacultyofHumanandSocialDevelopment,UniversityofVictoria,Victoria,BritishColumbia,Canada}

a

r

t

i

c

l

e

i

n

f

o

Articlehistory:

Received3June2014

Receivedinrevisedform

12November2014

Accepted14November2014

Keywords:

Privatedrugplans

Employeebenefits Drugcoverage Collectivebargaining Pharmaceuticals Healthinsurance

a

b

s

t

r

a

c

t

Prescriptiondrugsarethehighestsinglecostcomponentforemployees’benefitspackages inCanada.Whileindustryliteratureconsiderscost-containmentforprescriptiondrugcosts tobeapriorityforinsurersandemployers,theimplementationofcost-containment meas-uresforprivatedrugplansinCanadaremainsmoreofamyththanareality.Through18 semi-structuredphoneinterviewsconductedwithexpertsfromprivatesectorcompanies, unions,insurersandplanadvisors,thisstudyexploresthereasonsbehindthis incapac-itytoimplementcost-containmentmeasuresbyexamininghowprivatesectoremployers negotiatedrugbenefitdesigninunionizedsettings.Respondentswereaskedquestionson howemployeebenefitsarenegotiated;therelationshipsbetweentheplayerswho influ-encedrugbenefitdesign;theroleoftheseplayers’strategiesininfluencingplandesign;the broadsystemthatunderpinsdrugbenefitdesign;andthepotentialforauniversal pharma-careprograminCanada.Thestudyshowsthatthereisconsensusabouttheneedtoeducate employeesandemployers,morecollaborationanddata-sharingbetweenthesetwosets ofplayers,andforexternalinterventionfromgovernmenttohelptransformestablished normsintermsofprivatedrugplandesign.

©2014TheAuthors.PublishedbyElsevierIrelandLtd.Thisisanopenaccessarticleunder theCCBY-NC-NDlicense(http://creativecommons.org/licenses/by-nc-nd/3.0/).

1. Introduction

Canada’suniversalhealthcare systemdoesnot cover prescriptiondrugs.Publicdrugcoverageismostlyprovided onaprovincialbasistoseniorsandpeopleonsocial assis-tance.Manyprovincesalsoofferpubliccatastrophicdrug coveragefortherestofthepopulation(e.g.forpatients

∗ Correspondingauthorat:SchoolofPublicPolicyandAdministration,

CarletonUniversity(RB5224),1125ColonelByDrive,Ottawa,Ontario,

CanadaK1S5B6.Tel.:+16135202600.

E-mailaddress:ma.gagnon@carleton.ca(M.-A.Gagnon).

receivingpublicsubsidiesoncetheycontributemorethan 3–4%of theirannualincometowardprescriptiondrugs)

[1]. Most Canadians are covered through private drug plansofferedmostlybyemployersthroughsupplemental health benefits: 51% of Canadian workers have supple-mentalmedicalbenefits[2],andsincework-relatedhealth insurancealsocoversdependentsofemployeeswith cov-erage,asmanyastwo-thirdsofCanadiansarecoveredby healthinsuranceplans.

PrescriptiondrugspendinginCanada’sprivatesector hasincreasednearlyfivefoldin20years,from$3.6billion in1993to$15.9billionin2013[3].Privatedrugplansin Canadaareoftenconsideredwastefulbecausetheyaccept http://dx.doi.org/10.1016/j.healthpol.2014.11.013

0168-8510/©2014TheAuthors.PublishedbyElsevierIrelandLtd.ThisisanopenaccessarticleundertheCCBY-NC-NDlicense

payingforhigherpriceddrugsthatdonotimprovehealth outcomesforusersandusecostlysub-optimaldispensing intervalsformaintenancemedications.Asaconsequence,it isestimatedthatprivatedrugplansinCanadawasted$5.1 billionin2012,whichismoneyspentwithoutreceiving therapeuticbenefitsinreturn[4].Thisamountrepresented 52% of thetotal expenditures of $9.8billion byprivate insurersonprescriptiondrugsforthatyear[5].

Canadianemployershavedemonstratedgrowing con-cernforcost-containmentinthedesignoftheiremployees’ drug benefits. However, the implementation of cost-containmentmeasuresforprivatedrugplansremainsmore ofamyththanareality[6–10]sincefewplansrequirecaps fordispensingfees,premiumsfromclaimants,mandatory genericsubstitutionorrestrictionsonmoreexpensivebut nottherapeuticallysuperiornewdrugs[9].TheCanadian Life and Health Insurance Association, concernedabout thesustainabilityofprivatedrugcoverageinCanada,has askedforgovernmenthelptoreducecosts[11].Growing administrativecostsofprivatehealthplanscontinuesto putadditionalfinancialpressuresonthecapacitytooffer privatehealthbenefits[12].

Alackofpublishedliteratureonhowdrugbenefitsare negotiated andimplementedrequired ustoexplorethe subjectininterviewswithemployers,union representa-tives,insurersandconsultantsworkingforemployersor unions. We focused on unionized workplaces. In 2013, approximately 13.3% of all workers in Canada were unionizedprivatesectoremployees,18%wereunionized publicsectoremployees,andtherestbeingnon-unionized employees[13].Byfocusingondrugbenefitsinunionized settings,wewereabletobenefitfromtheinsightsofunion representativeswhohavesignificantexpertisein supple-mentalhealthbenefits.Drugbenefitsinunionizedsettings are often considered similar to those of non-unionized organizations[14].

2. Methodology

Weidentifiedkeyinformantsworkingwithinthemost prominent Canadianorganizationsin thefour organiza-tionalcategoriesexaminedinthisstudy,whoprovidedus withleadstocreateacohort ofpotentialinterview par-ticipants.Afterinitialcontactwiththesekeyinformants, anon-probabilitysamplingtechniqueknownassnowball sampling[15]wasemployedtoreachfurtherrespondents thathadkeyexposuretothedrugbenefitdesignprocess and couldprovideinsightsthatcouldbegeneralized,to some extent,across theirorganizational categories. We extendedaninvitationtoover60representativesfrom14 unions,9privatesectoremployers,19insurance compa-nies, and17 benefitsconsultanciestoparticipate inthe researchproject.Amongthoseinvited,18expertsagreed toparticipateintheinterviewprocess,fourofwhomwere from privatefirms,five from unions,five frombenefits consultancies,andfourrepresentedinsurancecompanies. We carried out one-to-one semi-structured interviews betweenSeptember2012andJanuaryof2013.

Thestudyfocusedonlargeunionizedworkplacesthat hadAdministrativeServicesOnly(ASO)plans,wherethe employerisresponsibleforthecostsofbenefitplansand

bearstherisksassociatedwithit,whileinsurersarejust hiredtomanageclaims.ThisstudyfocusedonASO arrange-mentsbecausetheyarethemostcommoninsuranceoption chosenby largeprivate-sectorfirms[16].Those organi-zationswhoseactivitiesresidedsolelyintheprovinceof Québec,wheretheregulationofprivatedrugplansdiffers

[17],wereexcluded.

Participants were asked to participate in semi-structured phone interviews lasting 20–30min. With a specificemphasisondrugbenefits,questionsfocusedon fourmainthemes:howemployeebenefitsarenegotiated; therelationshipsbetweentheplayerswhoinfluencedrug benefitdesign;theroleoftheseplayers’strategiesin influ-encingplandesign;andthebroadsystemthatunderpins drugbenefitdesign.Withrespecttothis lasttheme,the respondentswereaskedtodescribetheinequities inher-entinthesystemandtheirrecommendationsforreform, includingtheiropinionaboutanationalpublicdrugplan inCanada.Oneinsurerwasunabletorespondtoquestions pertainingtothelastthemebecausethetimeallottedin thisrespondent’sschedulepreventedtheinterviewfrom reachingthesequestions.

The researchdesign wasreviewed and approved by theCarletonUniversityResearchEthicsBoard. Sincethe natureof thetopicdiscussed wassensitiveforsomeof theorganizationsinvolved,theagreedprotocolguaranteed allparticipantsanonymitybynotdisclosingthenamesof theparticipantsandtheiraffiliatedinstitutions.Anydetails whichwould enablereaderstoidentifytheparticipants ortheorganizationsweredeliberatelyexcludedfromthis paper.

Wecarriedoutastandard thematicanalysisby tran-scribingandanalyzingthecontentsoftheaudiofiles.Based ontheresultsoftheinterviews,wedevelopedanarrative encompassingfournewthemeswhichdifferfromthe ini-tialthemesunderwhichthequestionswereorganized.The contentswerethenascribedinitialcodesand organized intothemesandsub-themesbasedonthetranscriptions’ contents.The authorsreliedontheirjudgmentto iden-tify themesfrom theinterview data,as noquantitative standardmeasuringtheprevalenceofsubjectcontentcan adequatelycapturethedepthofsuchqualitativedata[18]. Thus,ourstrategytoanalyzethesedatainvolvedcodingthe dataintoaconceptualframeworkfromwhichtheresearch resultsaredrawn[19].

3. Researchfindings

Thefollowingfoursectionsdescribethecorefindings through fourthemes: objectives; tactics and strategies; barrierstochange;andrecommendationsforreform. 3.1. Objectivesvis-à-visdrugbenefitdesign

Drugbenefit design decisions arearrived atthrough professionalnetworksofemployers,unions,insurers,and benefitsconsultants.Theinterviewsshowedthatthesesets ofplayershavedifferentinterpretationsofwhatisatstake indrugbenefitoutcomes,theirintentionsininfluencing theseoutcomes,aswellastheirperceptionsoftheother’s intentions.Table1categorizestheseplayer’sintentionsin

Table1

Monetaryandnon-monetaryobjectivesidentifiedbyparticipants.

Objectives Employers Unions Insurers Consultants

Monetary -Cost-savings. -Generousbenefits. -Contracts. -Employerspurchasingservices.

Non-monetary -Consistencyofbenefit provisionwithotherplayersin themarket.

-Talentattractionand retention.

-Control.

-Membershipapproval. -Demonstratinggainsand therebytheirutilityto employees.

-Amoredirectrelationship withtheirclients(cuttingout theconsultants).

-Maintaintheirusefulness.

accordancewiththeirstatedmonetaryandnon-monetary objectives.

Bothemployers and otheractors in this study cited cost-savingsas beingimportant. Typically, the employ-ersindicatedthatmonetaryitemsincollectivebargaining are discussed ona cost-neutralbasis; meaning that an increaseinabenefitlineitemmustbeoffsetbycost-savings elsewhere.Thus,changesinbenefitsarediscussedinthe contextofintroducingchanges tocompensation,as any costincreaseorsavinginoneareaaffectstheentire bas-ketofgoodsofferedtoemployeesintheircompensation packages.

Respondents from all categories indicated that con-sistency of benefits with other market players is of significance to employers. Three employers explicitly expressedadesiretoofferbenefitpackagesthatareatleast onpar withtheircompetitors. Firms constantly realign theirbenefitpackagestomeettheindustrystandardsince employeebenefitsareusedasatoolfortheattractionand retentionoftalent.Oneemployerdescribedthisneedinthe followingwords:“Thewholepackageofbothmandatedand discretionarybenefitshavetobewithinarangethatallows ustocompeteinthemarketplace.Soif,forexample,wewere tolookatourbenefitcostsandfindoutthatnooneinthe marketplaceprovidedprescriptiondrugcoverage,wewould havetoassessthatandtrytomakechangesinthecollective agreementssoastoeithereliminatethatormodifythebenefit tomakeitmorecosteffective.”

Respondentsfromallcategoriesmentionedthat,in con-trasttoemployers,theover-ridingobjectiveofunionsis tomaximizetheirbenefitswithminimalco-paymentsfor theiremployees.Inthewordsofonerespondentonthe unionside,“weasaunionwantasmuchaspossible.Wetry tokeepwhatever wehaveorimproveuponit.”Thispoint wasmadebyaunionmemberwhoclaimedthat“theunion sideisfarmoredemocraticandmore intensebecauseyou havetodealwiththemembership,aswellasjustthe leader-ship.”Asunionsfunctionaccordingtodemocraticmodels, theirleadershipishighlyresponsiveto(andinfluencedby) attitudesof membersrelating tobenefitsand collective agreements.Thisgovernancemodelpresentssome diffi-cultiesforachievingoptimalplan designsbecauseeven well-informedunionleaderswhowanttoimplement cost-containmentmeasuresmaybehamstrungbythedemands oftheirmembers.Thus,effectiveplandesignchangesneed significantemployeeengagementsothatthememberswill ultimatelybuy-intotheproposedchanges.

Ontheotherhand,consultanciesandinsurersindicated thattheyareprimarilyinterestedinkeeping their busi-nessesafloatandcontractingouttheirservicesthoughthey havedifferentmotivations.Theconsultanciesmostlywant

topreservetheirrelationshipswithplansponsors,while theinsurerswanttoplaya moreconsultativerolewith theirclients,andreducetheinfluenceofbenefits consul-tancies.

3.2. Tacticsandstrategies

Therespondentsvariedintheirinterpretationsofthe useofstrategiesinthenegotiatingprocess.Alloftheactors reportedusingsomesortofstrategytoachieveobjectives; however,theuseofstrategiesandtacticsbyemployersand unionswasmostprevalent.

Theemployersindicatedthattheirover-ridingstrategy istomaintaincost-neutralityinprovidingdrugbenefits– inthecontextofoverallcompensation–toemployees:any increasesinthecostsofaparticularbenefitsareamustbe off-setbycost-savingselsewhere.Controllingknowledge wasalso frequentlyreportedbythe union-side respon-dents(andbyoneconsultantthatservicesemployers)as astrategytoachievegreatercontrolovernegotiationsand plandesignbyfirms.Accordingtooneunion representa-tive,“theemployeralways hastheadvantage inthisstuff becausetheyhavealloftheinformationwithrespecttothe reportsandthecostsfromtheinsurerortheadvisor”andin onecircumstancetheunion“almosthadtothreatenthem (theemployer)withlegalactioninordertogetsomebasic informationoutofthemaroundcosting.”Thus,employers areperceivedbyunionsasemployingthistacticto pre-vent union negotiators from challenging theirpositions and proposals. Under this scenario,unionscannot opti-mallycontributeconstructiveanalysesandproposalssince theydonothavesufficientaccesstodata.Thispracticewas identifiedasbeingparticularlyproblematic,foritprevents unionsandemployeesfrombeingsufficientlyeducatedto contributetoplandesigndiscussionsinameaningful man-ner.

Oneemployerarguedthateven wheninformation is shared, there is a lack of trust which preventsthe two partiesfrommovingforwardintermsofplandesign.While thisemployerarguedthattheyfrequentlyconsultunions onspecific planchanges and “feel that they(the union) wouldbenefitbyhavingbettersupportaroundtheseissues,” theyalsoadmittedtonotsharinginformationor consul-tingtheunionwhenitisnotrequiredunderthecollective agreement.Theemployer“hadnoobligationtoaskforthe consent of the union and the company”for specific plan design changes and were “able to make changes” with-outanysortofdialog.Thiscausedabreakdownoftrust betweenthetwo partiesand hinderedtheircapacityto educatetheuniononplanchanges,sincetheunionisonly consultedafterthefact.

In theviewsof oneemployer,animportantstrategy used by unions negotiating with numerous employers within a sector has been to play employers against one another. Therefore, when one employeraccedes to demands,itscompetitorfeelspressuredtomakesimilar concessionstoensurethatitretainsatalentedand moti-vatedworkforce.Oneemployerindicatedthatthisstrategy canputlessfinanciallyviablefirmsatrisk,since wealth-iercompaniesmaywillinglymakeconcessionsthattheir competitorscannotafford.

However,anotheremployerindicatedthatcompanies may collaboratewhen their employeesare represented by the same union and cooperation is in their shared interest.Coordinationamongmultipleemployerscanlead to the adoption of consistent proposals between firms. Ifachieved,this harmonizationmakesitchallenging for unionsto playemployers againsteach otherand it can help employers to negotiate declines in benefits at an industrylevelforspecificoccupationgroups.Thistypeof coordinationeasesfirmspecificfearsoverlosingtalentto competitors.

Insurersplayasmallerroleinthenegotiatingprocess. Onestrategymentionedbyallinsurerswasaddingvalue toservices.Orasoneinsurerputsit,“wereallydon’thave astrategy,asidefromprovidingvalueandgettingclientsto seethatvalue.”Oneexamplewasintroducingcutting-edge informationtechnologytostreamlineclaimsadjudication. Oneinsurerhighlightedthetransitionfrompaperto elec-tronicclaimssubmissionsprocessestocutcostsandreduce theburdenofsubmittingphysicalclaimsasanexample ofvalue-addedservicestodifferentiatethemselvesfrom theircompetitors.Mostoftherespondentsindicatedthat insurancecompaniesarenotparticularlyconcernedwith strategiesandbenefitoutcomes,focusinginsteadonthe administration of claims and execution of plan designs demandedbytheirclients(i.e.employers).Thiswas evi-dentbythefactthatmostrespondentswhospokeonthis question(includingtheinsurersthemselves)indicatedthat insurancecompaniesareindifferentastowhether cost-containmentmeasuresareincludedinbenefitplans.Thus, insurerswerereportedashavingnoincentivesforreducing thecostsofplans.

On theotherhand,employersusestrategies intheir dealingswithinsurers.Withassistancefromtheirbenefits consultants,employershavetheoptionofgoingto mar-ket,andlookforcompetinginsurerswhomightofferlower administrativecosts,orotherwisepressuretheircurrent carriertolowertheirprice.Planadvisorshaveexperience in dealing withmultipleinsurerssimultaneously. Many employersseevalueinplanadvisors.Withadeeper knowl-edge ofclaimsmanagement and a rangeof methodsto cutcosts,planadvisorscanhelppoolemployerstogether, provideexpertadviceoncosting,assistemployersingoing tomarket,andcarryoutexpertnegotiationsonbehalfof clients.

3.3. Barrierstochange

Someclearbarrierstochangewereidentified.A major-ityofrespondentsindicatedthatpoorinformation-sharing wasa barriertoachievingcost-effectiveplanoutcomes.

Onebenefitsconsultantindicatedthatthisissymptomatic of the “old model” where communication over benefit design is limited to a few key players. Typically, lim-itedcommunicationbetweenunionsandemployers’plan advisorsand insurersresultsin a lossfor unions.Some workers’bargaining unitsmighthave accesstointernal expertisefromtheirunion;butthisismoreofan excep-tion.

Accordingtooneconsultant,“nooneknowsthecostof drugbenefitplans.”Thisrespondentwasarguingthatfew involvedinbenefitdesign,eitherinprivatefirms,unions, orinsurers,aresufficientlycompetenttoundertakeproper analysesofclaimsdatasotheydonotreallyknowhow pro-posedplanchangescouldaffectthem.Thislackofexpertise hasramificationsfortheeducationofstakeholdersonthe outcomesofbenefitdesign.

Wealsofoundthatalackoftrustwasperceivedasa fac-torinhibitingprogressinbenefitdesign,thebreakdownof whichisblamedonhistoricalprecedentsbetweenunions andtheiremployers.

Alackofemployeeengagementbytheprivatesector wasalsoconsideredtobeaproblem.Respondents from allsidesfrequentlyindicatedthatemployeesoftenwanted morecomprehensivebenefits,despitethefactthatthey achievesmallerpayraisesiftheyarepayingforinefficient benefitspackages,rather thanusingmoreefficientdrug benefitdesigntoredistributethesavingstowagesorother benefits.Employeesmaybecomeresistantanddistrustful whenchangesareintroducedwithoutsufficiently inform-ingthemonhowthosebenefitswillaffectthemandtheir families.

The democratic structure of union governance was identified as a barrier to change by respondents on the union and consultancy sides. Without information-sharingand employeeengagement,especially education ofemployeesonaspectsof theirbenefits,unionleaders cannotadoptaprogressiveapproachtodrugbenefits nego-tiations. One insurer referred to a conversation he had witha unionleaderonthebenefitsofintroducing cost-containmentmeasuresintoplandesign.Inthispurported conversation, the union leader expressedthat his sup-port for an initiative is meaningless unless the union’s membership has an appetite for a more cost effective plandesign.Furthermore,someunion,employer,and con-sultantrespondents claimed that when theunion lacks expertiseorcapacity,theyareworkingblindfolded, with-out the knowledge and expertise needed to negotiate betterbenefitpackages.

Variousunionrespondentsalsorecognizedthatmany Canadian unions have this capacity for generating knowledgeandexpertisethroughtheuseofin-house com-pensation specialistsand/or unionoriented consultants. Forexample,oneoftheinterviewedunionshasaresearch teamthatwould“parachutein”toassistthelocalforpart ofthenegotiatingprocessand“inacoupleofsetsof negoti-ations,theresearchdepartmentwillbeintherefromstartto finish.”Havingtheneededknowledgeandexpertisewhen andwhere youneedthemcantranslateintoinnovative plandesigns,solongastheinsightsgeneratedfromthis areeffectivelypassedontounionlocalsatthebargaining table.Oneunionrepresentativesaidthatlackoftrustand

information asymmetriesare less of anissue when the unionistheplansponsor.

Howtheinsuranceindustryisorganizedandoperates wasalsoreportedasbeingdetrimentaltobenefitdesign outcomes.Someintervieweessaidthatthebiggest prob-lemisthatinsurersdonothaveanincentivetointroduce cost-effectivenessclausesinplandesign.Forinstance,one benefitsconsultantstatedthatmostcompanies“inCanada willbuywhateverformularytheirinsurancecompanyoffers” andthat“wellover90%ofgroupinsuranceplansinCanada consistofeverysingledrugavailableforsalewithanoticeof complianceissuedfromHealthCanada.”

While thereare instances inwhich these companies consulttheirclientsonplanoptions,mostofthe insur-erssawthemselvesasreactive(andreceptive)toclient demands,ratherthanastheinitiatorinintroducing cost-effectivechangestoplandesign.Private-for-profitinsurers weredescribedasadministratorsofplandesignwho dif-ferinsomerespectsfromtheirnot-for-profitcounterparts. Asoneoftheinsurerputsit,anot-for-profitinsurer“will alwayslooktoreducecosts”and“additionalclaims”fortheir groups,while“afor-profitinsurermaywanttheadditional claimsbecausetheygetpaidperclaim”.

However,whenspeakingoffor-profitinsurers, partici-pantsfromallgroupsarguedthatinsurershavenofinancial incentivestocutcostsforemployers,asindicatedbyone employersaying:“frommyexperienceonthecommittees,I don’tgettheimpressionthattheinsurersaretheretosave costsforthe employers. Ihaven’t seenit.It’s always been theotherdirection.”Thisclaim wasalsocorroboratedby abenefitsconsultant, whoarguedthat“therehasbeena fairbitofinertia,youknow,amongsttheprovidersoutthere inactuallydoingsomethingtooradical, tooleading edge” because“there’snodirectfinancialincentiveforinsurance companiesorpharmacy benefit managersto actuallyhelp employerssavemoney”.Expandingonthis,another consult-antarguedthataninsurer’scommissionstructure,which isbasedonvolumesofclaimsexpressedinadollarvalue, mayinfactdiscourageinsurancecompaniesfrom propos-ingplandesignsthatreducethevolumesofclaims,asdoing sowouldadverselyaffectcompanyprofits.Furthermore, another benefits consultant indicated that insurers are expertswhocalculateriskandtherebyhavenoaptitudefor thecreationofformularies.Accordingtothisrespondent, theimpactisthatinsurancecompaniesexcelatmanaging risk,yetfarepoorlyindesigningcost-effectiveplansthat relyonthedesignandimplementationofformularies. 3.4. Recommendationsforreform

Greateropennessandtrust, intermsof information-sharingwassuggestedby anoverwhelming majorityof respondents from each category. Similarly, one insurer expressedimmensesupportforhavingan“intelligent con-versation”withalltherelevantinsurers, companiesand unionsatthenationallevelonthepotentialformore sus-tainable plan outcomes. Thiscomment was made with regard to a lack of information-sharing, as well as an overalllackofeducationintermsoftherelevantplayers’ understandingofdrugbenefitdesign. Eachofthe insur-ersthatprovidedrecommendationsforreform(3outof4)

envisionedaformofstakeholderengagementinwhichthe governmentwouldplayarole.Inthewordsofoneinsurer, “whenIsaystakeholders,weneedfederalandprovincial gov-ernments there.Weneeddoctors there.Weneedhospitals there.Pharmaciststhere.Weneedthedrugcompaniesthere. Theemployers,theunions,andtheinsuranceindustry.”Next, thisinsurerarguedthat“allofthosepartiesneedtocome togethercooperatively,puttheirself-interestaside,puttheir egosaside,andfigureoutawaytomosteffectivelymanagethe explodingcostofbenefits,whethertheyarepubliclyfunded benefitsorprivatelyfundedbenefits.”Thisassertionwas cor-roboratedbyoneconsultantandoneunionrepresentative. However,oneemployerarguedthattherehavebeen prece-dentswherethegovernmenthasregulated pharmaceuti-cals,withspecificmentionofgenerics,withoutadequately consultingindustry.Furthermore,anotheremployer rec-ommendedthatunionsandmanagementacknowledgethe meritsofmutual gainsand bargainforfosteringgreater cooperation and information-sharing between manage-ment andlabor. Mutualgains bargainingpresumesthat positive-sumscenariosareplausible,andcanbeachieved throughprinciplednegotiationsthatseparatethe negotia-torsfromtheproblemandfocusonprovidinggainstoboth sidesvestedinthenegotiation’soutcome[20].

Therewasalsosignificantsupportfor providing uni-versalcatastrophicdrugcoverage,universaldrugcoverage forseniors,oruniversaldrugcoverageforallCanadians. Aninterestingfindingfromtheinterviewdatawasthat respondentsfromallinterviewedgroupsdeclaredbeing in favor ofintroducing some sort of arrangement for a nationaldrugplan.Somefavoredhavingauniversal phar-macare program which wouldapply toalldrugs, while othersfavoredprogramstailoredforcatastrophicdrug cov-erage.Twooftheinsurersthatrespondedtothisquestion explicitlyfavoredsomeformofuniversalcatastrophicdrug coveragewhiletheotherfavoreduniversalpharmacare.

The benefits of such universal pharmacare, argued oneinsurer,isthatit provides“theemployerswithsome breathing room”;makes drug coverage and usage more transparent, which would result in bettereducationon whichdrugsarebeingconsumedandhow;andprovides opportunitiesforintegratingpublicandprivatesystemsto facilitatetheefficientdeliveryofdrugstoCanadians.Each oftheunionrepresentativesandoneemployerinterviewed forthisstudyexpressedtheirsupportforuniversal phar-macare.Threeoutoffiveconsultantsarguedinfavor of anationalpharmacareplanwhiletheothertwofavored someotherformofnationalriskpoolingorformulary man-agementtoaddresscosts.

While a majorityofinterviewees favoredsomeform ofuniversalcoverage,afewrespondentsfromtheinsurer and employer sides expressed concerns that universal pharmacareisnot feasible.These respondentsindicated thatexorbitantcostswouldbeassociatedwiththissortof program,andthatthesecostswouldexceedtheprojections of itsproponents.Furthermore, oneconsultantand one insurersuggestedthatanationalformularybeintroduced as a baseline for benefits providersacross the country. Ratherthanopenformularies,whicharethemost com-montoolusedtomanageprivatedrugbenefitsinCanada, the suggested formulary would vet drugs according to

measuresof effectiveness, safety and cost-effectiveness. One benefits consultant suggested that some sort of nationalpoolingand/orpurchasingarrangementbe intro-ducedtocontroldrugcostsforemployees.Interestingly, national pooling of high drug costs (individual claims over$25,000)hasbeenimplementedamong23insurance companiessinceJanuary1,2013.Suchpooling,however, covers fullyinsured plansonly,and excludeASOplans, thetypicalplanstructureforlargerenterprisesanalyzed inthispaper[21].

Finally,employersweremostconcernedwiththe gov-ernment’s role in distributing thecostsassociated with drug coverageamong public and privateplayersin the system. In fact, each employer expressed concern over this.Threeofthefouremployersexpressedconcernover thegovernment’sroleasaplansponsorandhow govern-mentsshift coststotheprivatesector. Asdescribed by oneemployer,“thegovernmentisaverybigconsumer of drugs”andifthedrugcompanies“startlosingmoneyonthe governmentside,theypassitontoprivateinsurance”.Thus, governmentregulationsthathelpemployerscontaincosts aredesired.

4. Discussion

Ouranalysisidentifiedkeyissuesatstakewhenitcomes tocreatingandnegotiatingprivatedrugcoverage.The com-munityofexpertswhoparticipatedinthisprojectappeared tohaveacommonunderstandingofwhatconstitutesthe salientissuesfacingthedrugbenefitsintheCanadian sys-tem.Thefivefollowingpointsweremoststrikinginour analysis.

4.1. Thenecessityofinformation-sharing

Thisstudy’smajorfindingisthatalackofinformation andpooreducationaboutdrugbenefitplandesignleadto poorer outcomes.Alack ofcooperationbetweenunions andemployershasresultedinthecreationofsiloswhich constrainthelinesofcommunicationandthecooperative useofexpertiseandinformation-sharingtoimprovingplan outcomesforemployees.

4.2. Democraticgovernanceofunionsrequires engagementofinformedemployees

A common theme is that a fundamental lack of employeeengagementhasmadeitdifficultforfirmsand unionstoachievetheemployeebuy-inneededtomove toward more rational, cost-effective drugbenefit plans. Employeestendtotreasuretheirbenefitpackages–even whenitisagainsttheirownbestinterest–andwillcall on their unions to resist changes that diminish cover-age.Thisoccursevenwhenpracticalplanchangeswhich mightrestrictaccesstohighercostdrugsthatareno bet-terthanlowercostalternativesareintroduced.Theidea that drugplansshouldpay forall drugs onthemarket (anddrugsprescribedbydoctors)isverystrong.Suchan expensivemindsetposesadisservicetoemployeeswho areoftenforcedtoendurecutstootherbenefits,wages,

andevenjobs.Thesearetheconsequencesofhaving pay-for-anythingdrugbenefits.

Ourfindings leadus tobelievethatCanada’s private sectorunionshavenotdoneenoughtoengagetheir mem-bers.Ratherthancreatingcampaignstoeducateemployees aboutprogressivechanges intheirbenefits,union lead-ers, by theirneed toappease those who elect them to leadershippositions,continuetoplacatetheirmembers’ desirefor generousbenefitsbyresistingchange.Forthe mostpart,theunionrepresentativesinterviewedforthis study were highly aware of the problems facing plan designandsupportiveofintroducingchanges.However, ananalysisoftheaggregateresponsessuggeststhatthis awarenessbyexpertsinprominentCanadianunionsstill needstobetransformedintoaproactiveeducation cam-paign on drug benefits in order to attain cost-efficient outcomes.

4.3. Theneedforincentivesforinsurancecompaniesto reducecosts

Canada’s insurance companies, particularly private insurers,werenotseenasparticularlyproactiveintermsof introducingplanchanges.Whilesomeoftherespondents indicatedthatsomeinsurerswerestartingtotakeonarole ineducatingplansponsorsaboutdifferentbenefitdesigns, insurerswerelargelyperceivedasbeingdemand-driven. Thatis,theirrolehasbeenrestrictedtoadministering ben-efitplanswithafocus onreducing administrativecosts andintroducinginnovativetechnologyforclaims manage-ment.Someinsurancecompaniesmighttrytoimplement efficientcost-containmentmeasures[22],butforthemost partarenotfinanciallyincentedtoworkinthatdirection. Asshownabove,for-profitinsurersactuallyfacethe oppo-siteincentives,drivingupthecostsofclaimsisconduciveto greaterprofitsforthiscriticalplayerinbenefitdesign.The majorpointhereisthattherecouldbearolefor policy-makerstocreateamorerationalincentivestructurethat motivatesinsurerstocontaincostsfordrugcoverage,for examplethroughtheprinciplesofmanagedcompetition

[23].

4.4. Theneedforgovernmentintervention

Itappearsthattheproblemsfacingbenefitprovisionin Canadawillnotresolvethemselveswithoutsomesortof governmentintervention.Allplayersappearedtobeaware ofmajorproblemsandconstraintsfacingprivatebenefitsin Canada.Themomentumisnotthereyet,incontrasttothe UnitedStates,wherefocusingoncost-effectiveformulary managementisanessentialpartofprivateinsurers’ offer-ings.Toresolvethissituation,allhavesuggestedsomeform ofgovernmentintervention,eitherthroughanational for-mulary,arisk-poolingscheme,anarrangementdesigned toprovidesomesortofuniversalpharmacaretoCanadians, orevenabroaddiscussionamongstalltheplayerswitha vestedinterestinplanoutcomes.Noneoftherespondents indicatedthatthegovernmentoughttoplaynoroleinthe futureofbenefitdesign.Thissuggeststhatthereisastrong needforstandardizationorregulationwhichwouldbe tol-eratedbythosewithavestedinterestinplanoutcomes,

butwelcomedasbeneficialbypractitionersinthefield,so longastheyplayaroleintheformulationofgovernment policy.

4.5. Disseminatingthesharedunderstandingofproblems andsolutionsrelatingtodrugbenefitdesign

Interestingly,thereappearstobeconsiderable consen-susonthenatureoftheproblemandpotentialsolutions. Noneof theparticipantsin thisstudy disputedthefact thatmostprivatesectorfirmsinCanadaarenot manag-ingformulariesbasedoncost-effectivenesssotheplans in placeprovide littleemphasis on value-for-money.In linewiththis finding,noneoftherespondentsdisputed thatemployeesareultimatelytheloserwhenfirmspayfor over-generousdrugbenefitdesignsthateatupfundsthat couldbeshiftedtootherformsofcompensation(i.e.wages andotherbenefits).Theircommonviewpointwasthatthe mainplayersinvolved incrafting drugbenefitplansfor themajorityofCanada’sprivatesectoremployeesarevery awareofthebarrierstoachievingsuperiorplanoutcomes, buttheyfeelthatoverallcontextinwhichthesebenefitsare designediswroughtwithconstraintsthatleavethem feel-ingpowerlesstoinstillprogressivechange.Theproblems arewidelyrecognized,butthisconsensusamongplayers mustberecognizedbypolicymakersinordertoarriveat apoliticalsolution,withoutwhichthecoreproblems fac-ingtheseorganizationsarenotlikelytodisappearinthe short-term.

4.6. Limitations

Thisexploratorystudyhasitslimitations.First,inorder toencourageparticipationinthisstudy,theresearchdesign providedtheintervieweeswithfullcontroloverwhat feed-backtheycouldprovide,andensuredthattheywereaware thatdoingsoisvoluntary.Thisconstrainedourabilityto probetheparticipantsforanswersonthemost controver-sialaspectsofthestudy,suchasthosepertainingtothe useofstrategiesinthenegotiatingprocess.Furthermore, thefactthatmanyaspectsofthistopicarecontroversialor dealinproprietarybusinessinformationlimitsthe partic-ipants’abilitytodisclosesensitiveinformationthatcould havebeenquiteinformativeforthisstudy.Participantsmay haveshiftedtheinterview discussionstowardthemore superficialelementstopreventthemselvesfrom disclos-ingindustrysecretspertainingtocollectivebargainingand plandesign. Infact,someconsultantsrefusedto partici-pateinthisstudybecausetheydidnotwanttodisclose theirfirm’sstrategies.

Inaddition,thisqualitativestudyissubjecttoselection bias,a common problemin qualitative research,where wearereflectingonlythepointsofviewofaselectgroup ofinterviewees,thuslimitingthegeneralizabilitytothe widerpopulation[24].Thus,theclaimsmadeinthispaper, suchas that concerning actors’ consensuson the issue ofcost-containment,mightbeoflimitedgeneralizability in how they reflect the views of all insurers, unions, employers, and consultancies, even if cross-referenced responsesremained consistent among the actors inter-viewed.Finally,theguaranteeoftheresearchparticipants’

anonymitypreventsusfromexploringwhattypesofunion structures,relationshipswithmanagement,anddelivery options(e.g.throughunionsvs.managementorprofitvs. non-profitinsuranceproviders)facilitatetheadoptionof morecost-effectiveplanoptions.Havinggreaterfreedom toexploretheinfluenceofthesevariablesonplandesign outcomeswouldprovideforaricheranalysisonthetopic.

5. Conclusion

This paperexplored how privatesector negotiations between unions, employers, consultants and insurance companiesfailtoachievedrugplandesignsthatareboth sustainableandcost-effective,andexplorespotential solu-tionstotheproblem.Sincepharmaceuticalsarethehighest singlecostcomponentofprivatehealthbenefits,thereis anurgentneedtoimplementmeasuresformanagingthe costsofdrugplanswithoutdegradinghealthoutcomes. Ourstudyshowsthatthereisconsensusabouttheneed toeducateemployeesandemployers,morecollaboration anddata-sharingbetweenthesetwosetsofplayers,and forexternalinterventionfromgovernmenttohelp trans-form established norms in terms of private drug plan design.

Up tonow,in spiteofrepeatedconcerns, employers and insurers have demonstrated little effort to imple-ment cost-containment measuresin privatedrugplans. Thisis instarkcontrasttoaggressively managedpublic plans which limit formularies in order to containcosts andensurecoveragetothosedrugproductsdeemed effec-tive,safeandcost-effective[11,25].Therefore,itbehooves private plansto emulate publicapproaches and strate-gies.

Currently,privatedrugplansforunionmembersshield employees from many additional costs, but also shield themfrommakingrationalchoicesondrugcoveragewhich are basedon considerationsof effectiveness,safety and valueformoney.Thisstudyhaspointedoutwherethose gapsare,especiallyinknowledge,capacityanddata,and itisnowincumbentuponunions,employers,insurers,and consultantstobeproactiveinorganizingdrugbenefitsina sustainableway.

Conflictofinterests

Sean O’Brady has no conflict of interest to declare. Marc-Andre Gagnon has received research funding by theCanadianFederationofNurses’Unionsforadifferent researchprojectrelatedtodrugcoverageinCanada.Alan Casselsisco-directorofDECA(DrugEvaluationConsulting andAnalysis).Theauthorswouldliketoacknowledgethe financialcontributionoftheCanadianHealthCoalitionin ordertopayforthetranscriptionofinterviews.

Acknowledgements

TheauthorswouldliketothanktheCanadianHealth Coalitionfortheirhelpwiththetranscriptionofinterviews. Theorganizationhadnoinvolvementinthedesignor con-tentofthearticle.

References

[1]Daw J, Morgan S. Stitching the gaps in the Canadian public drug coverage patchwork? A review of provincial pharmacare policy changesfrom 2000 to 2010. HealthPolicy 2012;104(1): 19–26.

[2]MorganS,DawJ,LawM.RethinkingpharmacareinCanada.CDHowe Institute;2013.Commentary384.

[3]ExpressScriptCanada.2013Drugtrendreport.Mississauga:ESI;

2014. http://www.express-scripts.ca/sites/default/files/uploads/

FINAL executive%20summary FINAL.pdf[accessed01.06.14].

[4]ExpressScriptsCanada,JunePoorpatientdecisionswasteupto$5.1

billionannually,accordingtoexpressscriptCanada.Pressrelease;

2013.

http://www.express-scripts.ca/about/canadian-press/poor- patient-decisions-waste-51-billion-annually-according-express-scripts[accessed01.06.14].

[5]Canadian Institute forHealthInformation. DrugExpenditure in Canada1985to2012.Ottawa:CIHI;2013.

[6]KratzerJ,KimberlynM,ErinS,LawMR.Cost-controlmechanisms inCanadianprivatedrugplans.HealthcarePolicy2013;9(August (1)):35–43.

[7]McDougallM.Actnow,savelater.BenefitsCanada2012;36(May (4)):33–8.

[8]French M. Do employers have what it takes? Mercer, Health

& Benefits Insights 2012;1(March):5–9. http://www.mercer.ca/

flipbook/H&BInsights/mercerinsights.pdf[accessed01.06.14].

[9]Martinez B. Drug insurance in Quebec, model or foil. In:

Presentation for the Canadian Association for Health Services

and Policy Research. 2012. A video of the presentation is

availableonline:https://www.youtube.com/watch?v=5Id5yhUg1YY

[accessed01.06.14].

[10]Lindenberg B. Benefits costs management: myth or reality? EmployeeBenefitsInquirer2011;4(March(4)).

[11]CanadianLifeandHealthInsuranceAssociation,Inc.CLHIAreport onprescriptiondrugpolicy;ensuringtheaccessibility,affordability andsustainabilityofprescriptiondrugsinCanada.CanadianLifeand HealthInsuranceAssociationInc.;2013.

[12]LawM,KratzerJ,DhallaIA.Theincreasinginefficiencyofprivate healthinsuranceinCanada.CanadianMedicalAssociationJournal 2014;186(4).

[13]Statistics,Canada,Table282-0078–Labourforcesurveyestimates

(LFS), employees by union coverage, North American Industry

ClassificationSystem(NAICS),sexandagegroup,annual(persons),

CANSIM. http://www5.statcan.gc.ca/cansim/a05?lang=eng&id=

2820078&pattern=2820078&searchTypeByValue=1&p2=35

[accessed20.09.14].

[14]ConferenceBoardofCanada.Benefitsbenchmarking2012,Ottawa; 2012.

[15]Salganik MJ, Heckathorn DD.Sampling and estimation in hid-den populations usingrespondent-drivensampling. Sociological Methodology2004;34(1):193–240.

[16]Sanofi.SanofiCanadahealthcaresurvey.Laval:RogersPublishing; 2012.

[17]Commissairedelasantéetdubienêtredu,Québec.,Lesmédicaments d’ordonnance:ÉtatdelasituationauQuébec.Québec:Gouvernement duQuébec;2014.

[18]BraunV,ClarkeV.Usingthematicanalysisinpsychology.Qualitative ResearchinPsychology2006;3(2):77–101.

[19]MilesMB,HubermanAM.Qualitativedataanalysis:anexpanded sourcebook,2nded.ThousandOaks:SagePublications;1994.

[20]FisherR,BillU.Gettingtoyes:negotiatingagreementwithoutgiving in.Boston:HoughtonMifflin;1981.

[21]BurnsT.Insurerstosharecostsofhigh-priceddrugs,3,April.Benefits Canada;2012.

[22]BlackwellT.Healthinsurer’sU.S.styleplanseeksto‘manage’ mem-bers’drugcosts,.In:NationalPost.July2012.

[23]EnthovenAC.Thehistoryandprinciplesofmanagedcompetition. HealthAffairs1993;12(Suppl.1):24–48.

[24]PayneG,WilliamsM.Generalizationinqualitativeresearch. Sociol-ogy2005;39(2):295–314.

[25]StevensonH.Anendto blankcheques;gettingmore valueout

ofemployersdrugplans.ReformularyGroup2011. http://www.

reformulary.com/filesdocs/content/pdf/en/AnEndtoBlank Cheques-May2011ENr.pdf[accessed01.06.14].