washington, d.c. april 20-21, 2017 - world...
TRANSCRIPT
SecondAnnualUHCFinancingForum
GreaterEfficiencyforBetterHealthandFinancialProtection
Backgroundpaper(Forumedition)
Washington,D.C.April20-21,2017
1
Thisisabackgroundpapertothe“SecondAnnualUHCFinancingForum:GreaterEfficiencyforBetterHealthandFinancial
Protection”.ThispapersetsthestageforthepresentationsanddiscussionsattheForumandwaspreparedundertheguidanceoftheForumTechnicalWorkingGroup.Theinformationprovidedinthisdocumentdoesnotnecessarilyrepresenttheviewsorpositionofthe
organizationsrepresentedontheTechnicalWorkingGroup.
2
ContentsI. ExecutiveSummary........................................................................................................................3
II. Introduction...................................................................................................................................6
III. Whatisefficiency?.....................................................................................................................7
IV. Sourcesofinefficiency................................................................................................................9
1) Doingtherightthings...............................................................................................................11
i. Mixofhealthinterventionsandpackages...........................................................................11
ii. Preventionversustreatment...............................................................................................12
iii. Thebalancebetweengovernance,administration,publichealthfunctionsandpersonalservices.........................................................................................................................................13
iv. Inter-sectoralandmulti-sectoralaction...............................................................................13
v. Financialprotectionandservicecoverage...........................................................................14
2) Doingtherightthingsintherightplace...................................................................................15
3) Doingthingsright.....................................................................................................................17
i. HealthServices.....................................................................................................................18
ii. HealthSystemsandtheHealthFinancingComponent........................................................22
V. Identifyingthemostimportantsourcesofinefficiency...............................................................24
VI. Measuringandmonitoringinefficiency...................................................................................25
1) Macro-efficiency.......................................................................................................................25
2) Efficiencyincomponentsofthehealthsystem.......................................................................27
VII. Strategiestoimproveefficiency...............................................................................................32
1) Technicaloptionsforimprovingefficiency..............................................................................33
2) Whatwedonotknow..............................................................................................................43
i. Payingforresults..................................................................................................................43
ii. Theprivatesectorandefficiency.........................................................................................46
iii. Humanresourcestrategiesforefficiency............................................................................46
iv. Costsofimprovingefficiency...............................................................................................47
v. Politicaleconomyissues.......................................................................................................47
VIII. Efficiencyandequity................................................................................................................48
IX. Conclusions..............................................................................................................................49
X. Recommendations.......................................................................................................................53
XI. References................................................................................................................................55
3
I. ExecutiveSummaryThenatureandextentofinefficiencyinhealth
Theconceptofefficiencyinhealthincludesdoingtherightthings,intherightsettings,andintherightway.Efficienthealthsystemsproducethebestpossiblehealthandfinancialprotectionoutcomesfromtheavailableresources.
Inefficienciesexistinhealthsystemseverywhere:
§ Doingthewrongthings(e.g.fundinghighcost,lowimpactinterventionsbutnotfullyfundinglowcost,highimpactinterventions,particularlyintertiarycaresettings.)
§ Doingtherightthingsinthewrongsettings(e.g.relyingonhospitalsratherthanprimaryhealthcare.
§ Doingthingsbadly(e.g.leakagesandwasteintermsofpilferingofmedicinesthroughthesupplychainandmedicineslefttoexpireorstoredinpoorconditions.)
Themagnitudeofinefficienciescanbestaggering:takingallformsofinefficienciesintoaccount,countriescanwasteanestimated20%-40%oftheirhealthresources,missingtheopportunitiestouseresourcesmoreefficientlytoachievemuchmore.
Sourcesofinefficiency
Thevarioussourcesofinefficiencyinhealthhavebeenwellestablished.Manyarelinkedtothemaincost-driversinhealthsystems:medicines,humanresourcesforhealth,andhealthfacilitiesandinfrastructure,particularlyhospitals.
Thenatureofinefficiencies,however,differbysetting.Countriesmightberelativelyefficientinoneareaandlessefficientinanother.Foreachcountrytounderstandhowtoreducetheirinefficiencies,theymustassessitsmostimportantcauses(usingthetypeofchecklistprovidedinthisdocument),thendecidewhicharemostfeasibletoaddresstechnicallyandpolitically.
Countriesthenneedtodevelopstrategies,atime-tableforchange,andastrategytomonitorprogress.
Animportantimplicationisthatcountrieswillneedtochooseindicatorstotracktheirprogressthatrelatetotheareasofinefficiencytheyareabouttotackle.
Improvingefficiency
Manyoptionsforimprovingefficiencyexist,andallcountriescantakeactionstoachievebetterhealthandfinancialprotectionusingtheiravailableresources.
Someoftheoptionslieinthehealthfinancingsystem.Theyincluderaisingrevenuemoreeffectively,usingthetaxsystemtoreduceconsumptionofproductsthatharmhealth,reducingfragmentationinpooling,ensuringthatpooledfundspurchasetheinterventionsthatdeliverthegreatestimpactforthemoney,andmodifyingproviderpaymentmechanismstoencouragebothefficiencyandquality.
Othersolutionsrequireactionsinthewiderhealthsystem.Forexample,improvingmedicines-relatedefficiencycertainlyinvolvestheabilitytobuyatthelowestcost,butmayalsoincludethecapacitytotestandensurequalitythroughoutthedistributionchain,tomodifyregulationsorlegislationtoencouragetheuseofgenerics,anddemand-sidestrategiestoovercomeconcernsinprescribersandpatientsaboutthequalityofgenerics,andtoencouragerationaluseofallmedicines.
4
Thepoliticsofimplementingpoliciestoimproveefficiencycanbecomplexbecausechangesmayfaceoppositionbypowerfulstakeholders.
Someofthepossiblesolutionsaremorelikelytoproducerapidreturnsthanothers.Theseinclude,butarenotlimitedto:
§ taxationtoreduceharmfulproductconsumption;§ improvingbudgetflexibilitytoimprovetimelyreleaseandexpenditureofavailablefunds;§ introductionofgenericmedicinespolicies,andtransparent,competitivebidding(where
possible)forpurchases.§ allowhealthworkersatlowerlevelstotakeonmoreresponsibilityasappropriate(task
shifting
Othersmayneedshort-terminvestmentsbutrequirelonger-termcommitmentbeforetheresultsstarttobeseen.Forexample,movingtowardsactivepurchasing(purchasingbasedonanexplicitassessmentofneeds,pricesandvalue)requiresstaffskillsandcomputerizedinformationsystemsthatmightrequiretimetodevelop.
Toaccompanytheseactions,internationalcollaborationisneededtocontinuallysearchfortechnologiesthat“shiftthefrontier”,identifyingfurtheropportunitiestoimprovehealthandfinancialprotectionatlowcost.Thismayinvolvetranslatingexistingtechnologiestolow-costsettings,ordevelopingnewapproaches,suchasvaccinesforHepatitisCandHIV/AIDS.
Controversiesandmissinginformation
Despiteknowingalotaboutthenatureofinefficiencyandpossibletechnicalsolutions,thereisstillasurprisingamountthatisnotknown,orthatengenderssubstantialdisagreement.
Remarkablylittleisknownaboutwhatworksatthesystemiclevel.Howcanhospitalefficiencybeimproved?Whatistheappropriateroleoftheprivatesectorinimprovingefficiency?Whattypesofincentiveskeepstaffmotivatedandensurequality,butareaffordable?Theevidenceissimplynotyetconclusive,whichmeansthesequestionscannotyetbeanswereddefinitively.
Partoftheproblemisthatmanystudiesofspecificinterventions,forexample,results-basedfinancing,havefocusedonserviceutilizationandquality,withlittleexaminationofcosts.Thisresearchtrendmakesitverydifficulttodeterminewhethertheseinterventionsareagooduseofscarceresourceseveniftheywork,andiftheycanbefinancedinthelongterm.
Inothercases,themethodsforundertakingthenecessaryanalyticalworkarenotparticularlyuseful.Forexample,healthtechnologyassessmentbasedoncost-effectivenessanalysisisfrequentlyusedtohelpguidedecisionsabouttherightinterventionmix.Itisappropriateforhigh-incomecountrieswherethequestioniswhethersmallchangesinexpenditure,ontopofwell-establishedexistingpackagesofservices,arewarranted.ThistypeofHTAisnotparticularlyusefulinmanylowerincomesettingsbecausethechangesneededarenotmarginal.Techniquesforassessingwhatmixofinterventionsshouldbeavailableatprimarylevel,forexample,havethecapacitytotakeintoaccountthefactthatcostsandimpactvarywithfactorssuchasscaleandscopeofinterventions,andwhetherthestaffneededtodelivercost-effectiveinterventionsareavailable.HTAisalsooflimiteduseinassessingtheappropriatemixbetweenpersonalandpopulationbasedhealthservices,andbetweengovernanceandpublichealthfunctionsandpopulation-basedandpersonalhealthinterventions.Theseareasneedfurtherdevelopment.
5
Afinallimitationtocountrieswantingtotakeontheefficiencyagendaisthatthedataneededtoformallyassessthemajorcausesofinefficiencyandtomonitorprogressisextremelylimited.Thisispartlybecausefewcountriescurrentlyundertakeregularassessmentsoftheirefficiency,sotheindicatorsthatareavailablearemostlycollectedforotherpurposes.
Recommendations
Countries§ Undertakeanassessmentofthemajorcausesofinefficiencyandthosethatarefeasibleto
changeintheshort,mediumandlonger-term.§ Developandimplementastrategyforimprovingefficiencyintheshorttomediumterm–this
shouldbepartofahealthfinancingstrategyalthoughsomeoftheactionswillneedtoextendbeyondhealthfinancing.
§ Starttoputinplacethebackgroundinvestmentstoensurethelonger-termoptionscanbeundertaken–e.g.legislation,consultation,computerizedinformationsystems,staffskills.
§ Undertakebothpoliticalandtechnicalanalysisto identifywhichreformshavethegreatestchanceofsuccess,thenbuildsupportandnegateopposition.
§ Develop a set of efficiency indicators specific to the country’s main causes of healthinefficiencies,anddevelopanagendaforachievingmorefortheavailableresources.
§ Investinmethodstocollectindicatordataandtoevaluateprogressregularly.§ Identify areas of possible inter-sectoral or multi-sectoral actions that would achieve the
largesthealthimpacts,andthepoliticalfeasibilityofinfluencingothersectorstoimplementthem(perhapsincollaborationwiththeMinistryofHealth).ThiswouldhelptheMinistryofHealth target the key ministries and make the best use of their own limited time andresources.
Internationalcommunity(includingresearchersinallcountries)§ Routinelyassessthecostsaswellasimpactofeffortstoimproveefficiencysothatcountries
candetermine theefficiency and financial sustainabilityofdifferentoptions for improvingefficiency.
§ Developanagendatoidentifythecost-effectivenessofeffortstoredresshealthinequalitiesaspartoftheefficiencyandequitydiscussion.
§ Developmethodswhichcanbeusedtohelpcountriesdeterminewhichofthemyriadofinter-sectoralormulti-sectoralactionstoimprovehealthshouldbegivenprioritywiththelimitedtimeandfinancialresourcesavailabletoaMinistryofHealth.
§ Continue to invest in the technologies that might “shift the frontier” of possibilities”,identifyingfurtheropportunitiestoimprovehealthandfinancialprotectionatlowcost,suchasvaccinesforHepatitisCandHIV/AIDS.
6
II. Introduction
TheseriesofAnnualForumsonFinancingUniversalHealthCoverage(UHC)focusonhealthfinancingactionstosmooththepathtowardsUHC–definedformallyastheambitionthatallpeoplewillobtainthehealthservicestheyneed,ofgoodquality,andwithoutfinancialhardshiplinkedtopayingforthemout-of-pocket(WHO2010a).ThefirstForumin2016dealtwithrevenuegeneration,orhowtoraisefundstomeettheneedsanddemandsofthepopulationforgoodqualityhealthservicesandfinancialprotection,whicharekeycomponentsofUHC.Thisisimportanttocountriesatallincomelevels,althoughtheabsoluteneedforadditionalrevenueisparticularlyhighinpoorercountries.Insomecountries,theshortageoffundsissoseverethatmeaningfulprogresstowardsUHCwillnothappenwithoutsubstantialincreasesinrevenuegeneration.ThisiswhyrevenuegenerationwasthetopicofthefirstForum.
ThesecondAnnualForumonFinancingforUHCturnstothequestionofhowtousetheavailableresourcesinthemostefficientway.Theextentofinefficiencyandwasteinhealthcanbestaggering,andeveniftheestimatesarerelativelyimprecise,theygiveanideaoftheorderofmagnitudeoftheproblem.Acommonlyquotedfigureisthatbetween20%and40%ofallhealthresourcesmightbeeffectivelylosttovariousformsofinefficiency(WHO2010;Chisholm&Evans2010).Alongsimilarlines,theOECDrecentlyreportedtheresultsofanumberofstudiesonwaste(onecomponentofinefficiency),andshowedinselectedOECDcountries,somewherebetween20%and50%ofhealthexpendituresarelikelybeingwastedduetoinefficiencies(OECD2017).
Improvedefficiency(achievingmorewiththeavailableresources)enablescountriestoobtaingreatercoverage,andtodeliverqualityhealthservicesandfinancialprotectionforthesameexpenditures.Itcanalsoimprovehealthoutcomes.Forexample,arecentIMFworkingpapersuggeststhatAfricancountriescouldraiselifeexpectancyatbirthbyaboutfiveyearsonaverageiftheyusedtheirhealthresourcesmoreefficiently(Grigoli&Kapsoli2013).
Improvedefficiencycansometimesalsosavemoneyorreducetherateofincreaseofhealthexpenditures–thishasbeencalled“bendingthecurve”thatdepictstherelationshipbetweenhealthexpendituresandGDPovertime(Coady,FranceseandShang2014;OECD2017).This,however,doesnotalwayshappen.Manyefficiencyreformsrequireupfrontinvestmentsbeforetheystarttoshowtheimprovementsinhealthand/orfinancialprotectionthatareorganizationofwhatisdone,orthewaythingsaredoneaspirestoachieve.
Recognizingthatthehealthsectormustcompetewithothersectorsintheallocationofpublicfinances,MinistriesofFinancehavesometimesbeenreluctanttoincreaseallocationstoMinistriesofHealthwhichareperceivedaseithernotfullyspendingthefundstheyalreadyhave,ornotusingthemefficiently(Gillingham2014;Tandonetal.2014).Investmentsinothersectorsaresometimesconsideredtooffergreatervalueformoney.Therefore,improvingefficiencyinhealthcanalsohelptoconvinceMinistriesofFinancetoallocatemorepublicfundstohealth:byachievingmorewithexistingresources,additionalresourcesmaybecomeavailable.
ImprovingefficiencyandhealthfinancingstrategiesisjustapartofprogresstowardsUHC.Withinthehealthsector,furtherrequirementsinclude:sufficientmotivatedhealthworkersoftherighttypelocatedclosetopeople;goodqualityinfrastructure,appropriatelylocated;afocusonhealthservicequality;sufficientessentialmedicinesandotherhealthproducts;highqualityleadershipandgovernance;andtimely,accurateinformation.Progresscanalsobefacilitatedbypromotingkeyinter-sectoralactions:bothactionsinothersectorsthatimprovehealth,andactionsinthehealthsectorthatimproveincome,educationanddevelopmentmoregenerally–whichin
7
turnfeedbackintohealthimprovements.Whilerecognizingthis,thefocusonthisForumisonhealthfinancingandachievingmorewiththeavailableresources.
Themainobjectiveofthepaperistohighlightwhatisknownandwhatisnotknownaboutthemaincausesofinefficiencyinhealthandwhatcanbedonetoreducethem.TheagendaoftheForumwasdevelopedinresponsetothisanalysis,focusingonareaswhereeitherknowledgeisstillinadequateforinformeddecisionmaking,orwherethereiscontroversyordisagreementamongexperts.
Thenextsectionofthispaper(Section2)beginsbydefiningtheeconomicconceptofefficiencyandhowithasbeenusedinhealth.Section3thenidentifiesthecommonsourcesofinefficiencyinhealth,animportantstartingpointforcountriesseekingtoachievemorewiththeavailableresources.Section4considersindicatorsthatcanbeusedtoidentifywhichformsofinefficiencyarethemostimportantineachsettingbeforedevelopingpoliciestoredressthem,andthentotrackprogressinreducinginefficiency.
Section5turnstoidentifyingrecognisedtechnicalsolutionsplusareaswheretherearestillimportantgapsinknowledge.Italsoconsiderswhatisknownabout“how”toensurethedesiredstrategiesareimplementedtakingintoaccountthepoliticaleconomyofefficiencyreforms.
Makingahealthsystemmoreefficientdoesnotnecessarilymeanthatitwillbecomemoreequitable.Forexample,expandingcoveragebyfirsttargetingpeopleinisolatedareasratherthanthoseinmoredenselypopulatedsettingsmightnotbethemostefficientoptionintermsofimprovingpopulationhealthlevelsorprovidingfinancialprotection,yetacountrymayconsiderthisapproachtobedesirableonequitygrounds.Thebulkofthispaperfocusespurelyonefficiencybutturnstothequestionofequityandpossibletrade-offswithefficiencyinSection6.
Thefinalsection(Section7)summarizesthemainfindingsofthepaper,highlightingareaswherethereisinsufficientevidencetoguidepolicy,wherethereiscontroversy,andwheretherearepossiblequickwinsintermsofimprovingefficiency.
III. Whatisefficiency?
FormaldefinitionsofefficiencyfromeconomicsandhealtheconomicsaresummarizedinBox1.Inessence,ahealthsystemthatisefficientproducesthemixofhealthservicesthatmaximizestheoutcomessocietyexpectsfromitshealthsystem,usuallypopulationhealthimprovements,usesthemixofinputsthatcoststheleast,andcombinestheseinputstoproducethemaximumpossibleoutputs.Itwouldnotbepossibletogetmorehealthforthesamelevelofexpenditurebyeitherchangingthemixofinputs,gettingmoreoutofthechosenmixofinputs,orproducingadifferentsetofhealthgoodsandservices.Box1:FormalDefinitionsofEfficiency
Generaleconomicsdefinesthreetypesofefficiency.Technicalefficiencyisachievedwhenaparticularsetofinputsachievesthemaximumpossibleoutput(s).Technicalefficiencycouldbeachievedwithaveryexpensivesetofinputs,soproductiveorproductionefficiencyiswhentheinputsusedtoproducethisoutputhavetheleastcost,whileallocativeefficiencyrequirestheproductionofthesetofoutputsthatpeoplevaluethemostforthegivenresources(e.g.seeHollingsworth2008).
8
Healtheconomicshastypicallydefinedonlytwosortsofefficiency-allocativeandtechnical.Allocativeefficiencyhasbeendefinedasrequiringthatthehealthgoodsandservicesproducedarethosethatmaximizesociety’sobjectivesforthehealthsector,usuallytranslatedintosomemeasureofpopulationhealthstatusorimprovement.1Technicalefficiencyistakentomeanachievingthemostwiththeavailableinputs.Thequestionofwhatmixofinputsintheproductionprocessistheleastcostmix(productionefficiencyingeneraleconomics)isfrequentlysubsumed,sometimesimplicitly,intotechnicalefficiencyandsometimesintoallocativeefficiency(Yip&Hafez2015;seeCylus,Papanicolas&Smith2013forausefuldiscussionofthevarioustypesofinefficiencyandhowtheyhavebeenusedinhealth).
Ratherthanusetheformaldefinitionsofefficiencyasanorganizerforthediscussion,we
followaframeworksuggestedbyYip&Hafez(2015)whichfocusesonthekeypolicyquestionsfacingcountriesseekingtoimprovetheefficiencyoftheirhealthsystems:
1. Doingtherightthings(allocativeefficiency:whatmixofinterventionsmaximizehealthoutcomesfortheavailableresources?);
2. Doingthemright(acombinationoftechnicalandproductiveefficiency:arethemixofinputsthelowestcostmix,anddotheyachievethemaximumpossibleoutputs?).Weaddanadditionalcategory–doingtherightthings“intherightplaces”.Decisionson
whichcaresettingservicesshouldbedeliveredhaveasignificantimpactontheabilityofhealthsystemstoimproveormaintainhealthandfinancialprotection.Commontrade-offsarebetween:thedifferentlevelsofcare(e.g.community,primary,secondary,tertiary);dayversusinpatientcare;long-terminstitutionalcareversushomecare;andsocialcareversusmedicalcare.Adequatecoordinationandcontinuityofcare,bothwithinandacrosslevels,arecriticalelementstoensuringthatservicesaredeliveredintherightcaresettings.
Althoughhigh-incomecountrygovernmentsoftenurgelineministriestoimproveefficiency
inthefaceofbudgetcuts,ortorestrainexpendituregrowth,themainobjectiveofimprovingefficiencyinlowandlower-middleincomecountriesisnottoreduceoverallspendinginhealthorcutbudgets.ItistomakebetteruseofavailableresourcestoachievefasterprogresstowardsUHC,betterhealthandgreaterfinancialprotection.Thisiswhythepaperfrequentlyusestheterm“achievingmorewiththeavailableresources”todescribetheterm“efficiency”.
ThishastwoimplicationsforthewayefficiencyisconsideredinthecontextofUHC.First,thetraditionalwayofconsideringefficiencyfocusesonhealthoutcomesforpatientsorapopulation.Inthiscontext,increasingcoveragewithhealth-sectorinterventionsisonlyonepathwaytoimprovinghealthoutcomes.Multi-sectoralorinter-sectoralapproachesarealsoimportantandneedtobeconsideredinanyassessmentofthemostefficientwayofimprovingpopulationhealth.TheyarediscussedinSection3.
Second,theconceptofUHCacknowledgesthatpeoplevaluenotonlythehealthimprovementsthatresultfromappropriateuseofhealthservicesbutalsotheassurancethatusinghealthserviceswillnotresultinseverefinanciallosses.This“valueofinsurance”-whereinsuranceisinterpretedinthebroadsenseofareductionintheriskofpeopleneedingtofindthefundsforhigh,unexpectedhealthcostsinthefuturethroughsomeformofprepaymentandpooling-hasbeenwellacceptedineconomicsashavingtwocomponents.Thefirstisthewelfarebenefitassociatedwithpeopleknowingtheywillnotsufferunexpectedfinanciallossesintheeventofillness,andthe
1Sometimestheassumedmaximispopulationwelfare(seePalmer&Torgerson1999)orthevalueofthehealthimprovement(aninterpretationofQALYs).
9
secondisthewelfarebenefitofknowingthattheywillbeabletoaffordtousehealthservicesthatwouldhavebeenunaffordableintheabsenceofinsurance(seeNyman1999).2
Todate,littleattentionhasbeenpaidtounderstandingthisbroaderconceptofefficiencyinthecontextofasocietyseekingtoimprovebothaspectsofUHCatthesametime–coveragewithqualityhealthservicesofalltypes,andcoveragewithfinancialprotection.Arecentexceptionisthedevelopmentofanextendedcost-effectivenessanalysis,whichseekstounderstandtheimplicationsofvariouschoicesofinterventionmixes,notonlyonhealthoutcomesbutalsoonimpoverishment,oneconsequenceofalackoffinancialprotection(seeVerguetetal.2013;Verguetetal.2014;Verguetetal.2015;Shrimeetal.2015).ThisisalsodiscussedfurtherinSection3.
IV. Sourcesofinefficiency
Countriescannotmakeinformedchoicesaboutwhichinefficienciestheycantacklewithoutidentifyingthemostimportantcausesinagivensetting;thepragmaticwayofachievingthisistostartwithoneofthechecklists,whicharebasedonananalysisoftheliteratureoncountryexperiences.Table1providesachecklistbasedonanapproachtakenbyWHOin2010(WHO2010a).ThischecklisthasbeenmodifiedtoincorporatemorerecentexperienceandsomemanagerialandadministrativeinefficienciesthatwerenotconsideredintheWHOreport,drawingonrecentworkbytheOECD(OECD2017).
Thetableisorganizedaccordingtothethreekeypolicyquestionsdiscussedearlier:doingtherightthings,doingthemintherightplacesanddoingthemright.Thesectionondoingthingsrightbuildsfromhealthsysteminputs(medicinesandothermedicalproducts,healthworkforce,infrastructureandequipment)totheoutputsandoutcomestheyproduce(healthservices).Italsoincludesinefficiencieswithintheoverallhealthsystemmanagement,organizationandgovernanceandwithinthehealthfinancingsystem.Table1:CommonCausesofInefficiency
DoingtherightthingInefficiencycouldresultfromanimbalancebetween:population-basedpromotionandpreventionversuspersonalandcurativeservices;highcost,lowimpacthealthservicesversuslowcost,highimpactservices;governanceandpublichealthfunctionsversusotherhealthservices.NotdoingtherightthinginthecontextofUHCcanmanifestalsoasinadequateattentiontofinancialprotectioncomparedtotheavailabilityandqualityofhealthservices,orviceversa.
Doingthingsintherightplace
Inefficiencieswouldcommonlyincludeservicesbeingprovidedathigherlevelinstitutionsthatcouldbedonewiththesamequalitybutwithlowercostsatlowerlevelsofthesystem(e.g.avoidableoutpatientspecialistvisitsandemergencydepartmentvisits),avoidableinpatientadmissionsorlongerthannecessarylengthofstay,oracuteinpatientsurgerieswhichcouldbeperformedinday-care.
2Formoreinformationontheextentofhealthshocksonconsumptionandtheabilitytosmoothconsumptionwithinsurance,seeforexample,Limwattananonetal.2015,Wagstaff&Lindelow2010,Chetty&Looney2006.
10
Doingthingsright–inefficienciescanbelinkedtohigher-costinputschosen,orinputsnotachievingtheirmaximumpotentialA. Inputs
1. Medicines:a) Under-utilizationofgenericsorpayingtoomuchforanyspecificmedicine;b) Useofineffectivemedicines,thewrongmedicines,orusingthematthewrongtime;c) Overuseorunnecessaryuse.
2. Infrastructure(e.g.healthfacilities)andequipment:a) Inappropriatehealthfacilitysize,particularlyhospitals,foroptimalefficiency;b) Underorover-capacityinhealthfacilities;c) Equipmentthatispurchasedandcannotberepairedorisnotusedoptimally.
3. Personnel:inappropriatemixbetweendifferentcadres;locatedinthewrongplaces;demotivatedworkerswithlowproductivity(e.g.lowvisitsperhealthworkersperday,highratesofabsenteeism);poorqualityofcareprovided.
4. Inappropriatemixofinputs:e.g.healthworkersbutnomedicinesorothermedicalproducts,alowercostmixofinputsispossibletoachievethesameoutputs.3
B. Outputsandoutcomes
5. Healthservices:a) Unnecessary,tests,procedures,treatments/surgerycomparedtoneed;b) Medicalerrorsandlowqualitycare,includingdoingtheinterventionsatthewrong
time(e.g.late)meaningtheinputsandoutputsdonotachievethedesiredoutcomes;c) Underuseofneededhealthservices(prevention,treatment,rehabilitation,palliation;
includesmedicines)leavingpatientsvulnerabletounnecessarydisease,suffering,andpossibleincreasedmedicalcostssubsequently.
C. HealthSystemStructure,OrganizationandGovernanceincludingtheHealthFinancing
System6. Waste(includingexpiredmedicines),corruption,fraud.7. Sub-optimalpublicfinancialmanagementpracticesincludinglatedisbursementsfrom
theMinistryofFinance,alargenumberoflineitemsorinflexibilityacrosslines,lowbudgetexecutionrates.
8. Inefficiencyinraisingrevenues(forhealth),particularlywhenrevenueraisingforhealthisindependentfromgeneralgovernmentrevenuecollection.
9. Fragmentationinthesystem:inpooling,butinthebroaderhealthsystemaswell-e.g.procurement,supplychains,laboratories,servicedelivery.Thiscanbeassociatedwithdomesticdecisionssuchasestablishingseparateinsuranceschemesfordifferentpopulationgroups,ortodecisionsmadebyexternalpartnerstobypassexistingnationalsystemsandestablishparallelsystemsandstructurese.g.financialflows,audit,M&E,servicedelivery,laboratories.
10. Administrativeinefficiency:higher-than-necessarycostsfortheservicesoffered,includinginhealthinsuranceagencies.
3Therearesomeoverlapsinevitablybetweenthevariousformsofinefficiency.Forexample,healthworkerswithoutmedicinesordiagnostictestsmightbelinkedtoadministrativeefficiency,whileusinghighercostinputstoachieveresultsthatlowercostinputscouldachieveisoneaspectofhealthworkforcemanagementaswell.
11
1) DoingtherightthingsWithresourcescarcity,doingtherightthingsrequiresexplicitdecisionsaboutwhich
interventionsshouldbeavailable,andwithinthat,thebalancebetweenpreventionandtreatment,governanceandpublichealthfunctionsandpersonalhealthservices,andwhattypesofinter-sectoralactionstoimprovehealthshouldbepursued.TheseissuesarediscussedbrieflyinturnfollowingtheorderofTable1.
i. MixofhealthinterventionsandpackagesMostoftheliteratureondoingtherightthingshasfocusedontheappropriatemixofhealth
interventionsusingcost-effectivenessanalysis.Theliteratureonthecost-effectivenessofspecificinterventionsoragroupofinterventionsforaparticularhealthproblemisvast,toovasttoreferencefullyhere(e.g.Cambianoetal.2015;Edejeretal.2005;Mocketal.2015;Kimetal.2015;Ortegonetal.2012;Rozeetal.2015;Whiteetal.2011).Thereislessanalysiscomparingtheefficiencyoffundingasetofdifferenttypesofinterventionsacrosspriorityhealthproblems,thesortofanalysisrequiredifcountriesaregoingtodevelopanessentialpackageofhealthservicesfromscratchormodifytheirpackagebasedonthebestpossibleevidence.4Fromthefewanalysesthatareavailable,theevidenceshowsthatmanycountriesarenotfullyfundinghealthinterventionsthatarelowcost,highimpactintermsofpopulationhealthbenefits(suchaschildhoodimmunizations)whilehighcost,lowimpactinterventions(suchasformsoftertiarycareforchronicdiseases)doreceivefunding(Chisholmetal.2012;Evansetal.2005;Laxminarayanetal.2006;Jamisonetal.2006;WHO2017a).
Inthesecases,reallocationofresourcestowardsthelowcost,highimpactinterventionswouldimprovepopulationhealthforthesameexpenditure.Therearefewexplicitanalysesofthelikelygainsofdoingthis.OneexceptionistheWHO(2010)estimatethatswitchingbetweeninterventionsinthismannercouldproducethesamehealthbenefitsatbetween16%and99%ofthecurrentcosts(dependingonthehealthproblembeingstudiedandthecountry).5
Thereareanumberoftechnicalandpracticalproblemswithmuchofthecost-effectivenessliteraturethatlimititsvaluetocountrieswishingto“dotherightthing”bychangingtheirinterventionmixesordevelopingapackageofinterventionstowhichallpeoplewillbeguaranteedaccess.First,muchoftheanalysisasksifnewinterventionsshouldbefundedwithnoguidanceaboutwhatinterventionsshouldbereducedifresourcesarescarce.Thefundamentalquestionofwhichmixofinterventionswouldmostimprovepopulationhealthwiththeavailableresourcesisrarelyasked.Second,thecost-effectivenessofanygiveninterventiondependsonmanylocation-specificfactors,includingcoststructures,diseasepatterns,thepopulationagepyramid,whatothertypesofinterventionsarebeingundertakenthatmightinteractintermsofcostsoreffects,andcurrentlevelsofcoverage.Itisnotnecessarilyappropriatetoextrapolatetheresultsofastudyundertakeninonecountrytoothercountries,whosevariablesmaydiffersubstantially.Third,costsandeffectscanchangeovertimewithchangesindiseasepatternsandcosts,andtechnologicalinnovations.
Itwouldbedifficultforanycountrytoundertakestudiesofthecost-effectivenessofallpossiblehealthinterventions–promotion,prevention,treatment,rehabilitationandpalliation,aswellaspersonalandpopulation-based–intheirownsettings,andupdatethemfrequently,even
4Bymodify,wemeanaddandsubtract.Traditionalincrementalcost-effectivenessanalysislooksatwhatservicesshouldbeadded,butrarelyisguidancegivenaboutwhatshouldbereducedtomakewayforthenewinterventionwhenadditionalresourcescannotbefound.5Theseestimatesdidnottakeintoaccountthetransactioncostsinvolvedinmakingthechanges.
12
withgoodtechnicalcapacities.Internationalinitiativessuchasthevariousiterationsofthediseasecontrolprioritiesproject(DCP)andtheWHO-CHOICEprojecthavesoughttocollatetheinformationoncost-effectivenessfromawidevarietyofinterventionsinwaysthatcouldhelpcountries,particularlythosewithlimitedtechnicalcapacity,understandwhichinterventionscouldbegivenhighpriorityintheirsettings(Jamisonetal.2006;Mocketal.2015;Ortegonetal.2012;WHO2017a).WHO-CHOICEalsoreportsresultsatdifferentlevelsofcoverage,takingintoaccountpossibleinteractionsbetweeninterventionsintermsofcostsandeffects.6
Thereareseveralwell-acceptedlessonsfromthistypeofwork.Incountrieswithahighburdenofcommunicablediseaseandhighmaternalandchildmortality,packagesinvolvingpreventionandtreatmentformaternalandneonatalcare,childhealth,HIV/AIDS,tuberculosisandmalariaarehighlycosteffective(Laxminarayanetal.2006).Mostcountriescurrentlyseektomakethesepackagesuniversallyavailable,evenatlowlevelsofincomepercapita.Otherinterventionsthatseemtoberelativelycost-effective(e.g.preventionoftrafficaccidents,reductionoftobaccouse,surgicalwardsindistricthospitals,treatmentofacutemyocardialinfarctionandsomesecondarypreventionforcardiovascularevents)arenotyet,however,universallyavailableinlowandmiddle-incomecountries.ThereasonswhytheyarenotwidelyavailableareconsideredfurtherinSection6.
Thishighlightsanotherproblemwiththisliterature–itrarelyconsiderstheresourceenvelope.Financialconstraintspreventmostlowandlower-middleincomecountriesassuringuniversalcoveragewithalltheinterventionsthattheliteraturesuggestsarecost-effective:thisisafailureoftheliteraturemorethanafailurebycountries.
Moneyisnot,however,theonlyconstraint,atleastintheshort-to-mediumterm.Othersincludethenumber,skillmixandlocationofhealthworkers,andthetypeofhealthinfrastructureavailable,includinglocationandtypeofhospitals,primarycarefacilitiesandcommunityfacilities.Effortstodevelopasetofguaranteedhealthservicesmusttakeintoaccountalloftheseconstraints.Thisrequiresarelativelycomplicateddecision-makingprocess.Formally,itwouldrequiresomeformofprogrammingmodelratherthansimplycomparingcost-effectivenessratios,althoughasecond-bestalternativeistocomparethenon-financialresourceneedsofanyproposedsetofcost-effectiveinterventionswiththeavailablenon-financialresourcestoassessthefeasibilityofimplementation.
ii. PreventionversustreatmentPartofthecost-effectivenessliteraturehasfocusedonpreventioneitherbyitselfofin
comparisontotreatmentofvarioustypes(e.g.DiabetesPreventionProgramResearchGroup,2012;Chisholmetal.2012;Granichetal.2012;Jamisonetal.2006;Kuykenetal.2015;WHO2017a).Theresultsarenotparticularlysurprising:sometypesofpreventionareverycost-effectiveandsomearenot.Sometypesoftreatmentarecost-effectiveandsomearenot.Preventionisnotalways“better”thancureinthesenseofproducingmorehealthforthemoney,althoughoftenitis.Sometimes,thoughnotalways,expenditureonpreventionresultsinanetsavingoffuturetreatmentcosts.7Thedecisionaboutfundingpreventionortreatmentneedstobetakenonacase-by-casebasis,andeachcountrywillendupwithitsownmixdependingonfactorssuchasdiseasepatterns,coststructures,andhealthworkercapacities.Itisnotpossible,therefore,tooffergenericguidanceastowhatshare
6Thisallowsforeconomiesanddiseconomiesofscale,economiesofscope,andinteractionsbetweeninterventionsintermsofeffectiveness.7Thatisthepresentvalueofthecostsoftheinterventionarelowerthanthepresentvalueofthesavingsinthefuturecostsoftreatment,rehabilitationandpalliation.
13
ofhealthexpendituresshouldbeallocatedtopreventionandpromotionratherthansubsequenttreatment,rehabilitationandpalliation.
iii. Thebalancebetweengovernance,administration,publichealthfunctionsandpersonalservices
Improvingandmaintaininghealthrequiresanumberofactionsbeyondensuringprevention,treatment,rehabilitationandpalliationattheindividuallevel.Population-basedpromotionandprevention(suchaspublichealtheducationcampaigns),forexample,issomethingthatisunlikelytobeachievedwithouttheinvolvementofgovernment.Thesetofadditionalactivitiesthatrequiregovernmentinvolvement(generallyincludingfunding),areoftencalledessentialpublichealthfunctionsorservices.Thesecanbedefinedwithdifferingdegreesofaggregation(e.g.WHO2017band2017c;CDC2017)andgenerallyinclude:diseaseandoutbreaksurveillanceandcontrol;population-basedhealthpromotion;linkingpeopletopersonalservices;developingthehealthworkforce;settingandenforcingstandardsinservicedelivery;andhealthresearch.
Sometimesthegovernanceandadministrativeactivitiesrequiredtokeepahealthsystemfunctioningarealsoincluded,includingdevelopingandimplementingplansforthehealthsector,developingandenforcinglegislationandregulationsasappropriate,interactionswiththecommunityandpromotinginter-sectoralactiontoimprovehealth.Countriesstruggletofindtheappropriatebalancebetweenfinanceforroutineadministration,governanceandtheotherpublichealthfunctionsinthefaceofcontinualdemandforpersonalservicesforindividualpeople.
Scientificallyvalidguidanceontheappropriatebalancetoensureefficiencyisdifficulttofind.Thecost-effectivenessliteratureexaminessomeofthesefunctions,mostlyrelatingtopopulation-basedhealthpromotionorlegislationandregulationtocontrolthingsthatareharmfultohealthsuchastobaccoortraffic(e.g.Ortegonetal.2012;Goetzeletal.2014;Masonetal.2014;Gordon&Rowell2015).Beyondarguingthatmanyinterventionsarecost-effectiveorthatmorehealthpromotionofvarioustypesshouldbeimplemented(e.g.Lobsteinetal.2015),thereislittleguidancewhatpersonalhealthservicesshouldbecuttomakeroomfortheincreasedexpenditureintheseareas.
iv. Inter-sectoralandmulti-sectoralactionIthaslongbeenunderstoodthatmanyfactorsoutsidethehealthsectorinfluencehealth
(suchasincome,education,inequalities,environmentaldegradationandgenderandsocialnorms)andthathealthalsoinfluencesmanyexternalfactors(suchastheabilitytoearnandtogotoschool)(Grossman1976;Cumper1984;Wilkinson1997;CommissiononSocialDeterminantsofHealth2008;Marmotetal.2008).Asaresult,thereisincreasingliteraturearguingthata“wholeofgovernment”or“multi-sectoral”approachisneededtocomplementhealthservicedeliverytoimprovehealthandreducehealthinequalities(e.g.Marmotetal.2008;Carey,CrammondandKeast2014;WHO2014).Thisbuildsonthealreadylargeliteraturearguingfor“inter-sectoral”actions(thehealthministryworkingbilaterallywithothersectors)toimprovehealth(e.g.Dahlgren1994;WHO1997;Adamsetal.2014;Daviesetal.2014).
Thereisalsoagrowingliteraturedocumentingwhereinter-sectoralormulti-sectoralactionshavebeentaken,andhowmorecouldbefacilitated(e.g.Anafetal.2014;Larsenetal.2014;Dawson,Huikuri&Armada2015;deAndradeetal.2015;DeLeeuw&Peters2015).Asyet,however,thereisverylittleinformationthatcanbeusedtoguidegovernmentsindecidingwhetheritismore
14
efficienttoallocatemoreresourcestoothersectorsattheexpenseoffinancingthehealthsector.Partoftheproblemisthatactionsoutsidethehealthsectorcanimprovehealthalongsideotheraspectsofhumanwelfare,suchaseducationalattainment,theenvironmentandincomeearningcapacities.Cost-effectivenessanalysiscapturesonlytheimpactsonhealthsoisnotanappropriatetooltoguidesuchbroaddecisions.
Cost-benefitanalysishassometimesbeenusedinthesecases,convertingallbenefitsincludingreducedmortalityandmorbidityintomoneyterms.Livessavedaretranslatedintoamonetaryvalue,frequentlyusingamethodknownasthevalueofastatisticallifewhichisbasedonanassessmentofthevaluethatindividualsplaceonamarginalchangeintheirriskofdeath,mostcommonlyrevealedbythesalarypremiumtheywouldaccepttoworkinamoredangerousoccupation(e.g.Kniesner,ViscusiandZiliak2014;Laxminarayanetal.2014;Viscusi2015).Theapproachhasastrongfollowingforitsuseinvaluinghealthbenefitsassociatedwithinterventionssuchastransportimprovements,andintheassessmentofhealthinterventions.However,theappropriatenessofstatisticallifevaluemethodsarewidelydebatedbyhealtheconomistsanddislikedbyhealthprofessionalswhoconsiderthathumanlifeispriceless(e.g.Alberini&Ščasný2013;Angevine&Berven2014).
Anarrowerperspectiveformultisectoralorinter-sectoralanalysisistodeterminewhichofthepossibleoptions(whicharenotcurrentlybeingundertaken)theMinistryofHealthshouldfocusoninitsefforttoconvinceothersectorstotakeactiontoimprovehealth.Mostoftheexistingliteratureonthesocialdeterminantsofhealthimpliesthatthehealthsectorshouldseektohaveallofthemimplemented,butaminister’stimeisscare,asisthetimeoftheministrystaffandthefundstheywouldhaveavailabletosupporttheseactions.Fromthehealthperspective,itismoreefficienttotargettheactivitiesinothersectorsthatofferthegreatesthealthimprovementsfortheirinvestmentsinmoneyandtime.Theliteratureofferslittleguidanceonthis,withtherecentexceptionofapaperthatsoughttoidentifythehealthandnon-healthinterventionsthathadthebiggestimpactonlifeexpectancyin54lowerincomecountriessince1990basedonaformofregressionanalysis(Hauck,MartinandSmith2016).Interestingly,ofthenon-healthinputs,genderequalityhadthebiggestimpactonlifeexpectancywhileprimaryschoolenrolmentalsohadanimportanteffect.Thisisapromisingstart,butmoreworkneedstobedonetohelpindividualcountriesunderstand,lookingforward,whattypeofspecificinterventionsoutsidethehealthsectorarelikelytoimprovehealththemost.Forexample,mostcountriesreachedclosetouniversalprimaryenrolmentduringtheMDGera,soincreasingenrolmentmorewouldnothavelargeimpactsinthefuture.
v. FinancialprotectionandservicecoverageAsarguedearlier,discussionsaboutdoingtherightthingsinthefaceoffinancialconstraints
inhealthhavegenerallyassumedthattheoutcomeofinterestisanimprovementinhealth.Whenitisrecognizedthatpeoplealsovaluefinancialprotection,thereisanotherefficiencytrade-off:betweenusingscarcefundstoincreasecoverage(and/orquality)withexistinghealthinterventions,therebyimprovinghealth,orimprovingfinancialprotectionbyreducingout-of-pocketpayments.Littleattentionhasbeengiventohowtoaddressthistrade-offtodate.
Arecentexceptionhasbeencalled“extendedcost-effectivenessanalysis”.Itexaminestheimpactofdifferenttypesofinterventionsonhealthoutcomesasinstandardpractice,butalsoconsiderstheirimpactonafinancialprotectionindicatorsuchastheincidenceofimpoverishmentlinkedtoout-of-pockethealthpayments(Verguetetal.2013;Shrimeetal.2015;Verguetetal.2015a;Verguetetal.2015b).Theinformationonbothtypesofoutcomesarepresentedseparatelyratherthanseekingtoputrelativeweightstothetwocomponents.
15
Tousethistypeofanalysis,decision-makerswouldneedtoconsiderpairsofoutcomes(healthimprovementandfinancialprotection),andimplicitlyattachweightstotherelativevalueofeachwhendecidingwhattypesofhealthservicestosupport.Therearesomeproblemswiththisapproach.Forexample,itisevenmorecontextualtoaspecificcountrythannormalcost-effectivenessanalysisbecauseout-of-pocketpaymentlevelsandtheirdistributionacrosstypesofhealthservices,andincomesvarygreatly.Theanalysishasalsobeenappliedtohealthinterventionsandnotyettointerventionsaimedspecificallyatreducingoutofpocketpaymentsorstrengtheningpoolingarrangement.However,itisaninterestingdevelopmentthatcancontributetounderstandingatleastsomeofthetrade-offsinvolvedindecisionsaboutefficiencyanddoingtherightthings.8
2) DoingtherightthingsintherightplaceDoingtherightthingsalsorequiresanassessmentofwhichcaresettingsarethemost
efficientforservicestobedelivered.Iftherearenoeffectsonquality,servicesshouldbedeliveredintheleastcostlycaresetting.Commonexamplesofservicesbeingprovidedininappropriatecaresettingsinclude(i)acuteinpatientadmissionswhichcouldhavebeenavoidedorshortenedthroughtheavailabilityofadequateambulatorycare(includingdaycare),home-basednursingcareorsocialcareand(ii)outpatientspecialistoremergencycarevisitsforserviceswhichcouldhavebeenprovidedinprimarycare.Althoughsomeoftheseservicedeliveryandorganizationmodelsarenotyetwidelyavailableinlower-incomecountrieswherepatients(particularlythepoor)sometimesdonotobtainanytypeofformalcareatall,experiencesfrommoredevelopedcountriescanprovideimportantlessonstopreventcommonsetbacksanddesignchallengesinachievinggreaterefficiencyastheselower-incomecountriesdeveloptheirservicedeliverysystems.
Evidenceofavoidablehospitaladmissionsincludeadmissionsforconditionsthatarenotsevereenoughtowarrantanadmission(andthuscouldbetreatedatlesscostlycaresettings)aswellasforambulatorycare-sensitiveconditionswhichcouldhavebeenavertedthroughtheprovisionofadequatepreventativecareinlowerandlesscostlycaresettings.Forexample,thelargevariationinadmissionsforlow-mortalityconditions(e.g.asthmaexacerbation)fromtheemergencydepartmentindicatesthatsomeoftheseadmissionsmaybeunnecessary(Sabbatini,NallamothuandKocher2014;).Inaddition,studiesshowthatcostlyhospitalizationsforambulatorycaresensitiveconditions(e.g.diabetes)couldbeavoidedthroughaccesstoadequatepreventionandearlytreatmentinprimary/communitycare(James,BerchetandMuir2017;Rosanoetal.2013).Thereiswidevariationinageandsex-standardizedhospitaladmissionratesforAsthmaandCOPD(bothambulatorycaresensitiveconditions)acrossOECDcountries(Figure1).AsthmaadmissionratesarehighestforKorea,theUnitedStatesandSlovakrepublicwhileCOPDadmissionratesarethehighestforHungaryandIreland.InEstonia,avoidablechronicobstructivepulmonarydisease(COPD)andasthmaadmissionsconstituted76.9%ofadmissionsforlowerchronicrespiratorydisease,whileavoidablecongestiveheartfailure(CHF)andhypertensionadmissionscomprised84.3%ofadmissionsforhypertensionandotherformsofheartdisease(WorldBank2015).
8Anothermoretechnicalissueisthattheextentofimpoverishmentduetoout-of-pockethealthpaymentsdoesnotreallyreflecttheex-antevalueofthefinancialprotection,butmoretheresultofnothavingsufficientfinancialprotection.
16
Figure1:AsthmaandCOPDhospitaladmissioninadults,2013(ornearestyear)
Source:OECDHealthStatistics2015
Aprolongedlengthofstayinhospitalsisanotherindicationofcarethatisnotbeingdeliveredinappropriatecaresettings.Oftenpatientscanremaininthehospitalduetotheneedfornursingorpalliativecare,whichtheyareunabletoaccessinothercaresettings.Forexample,inEstonia,about32.6%ofhipfracturepatientsremainhospitalizedbeyondtheinternationalstandardof28days,whilethepercentageofstrokepatientsremaininginacuteinpatientcarebeyondtheinternationalstandardof56-daysis6.91%(WorldBank2015).Home-basednursingcarehasbeenshowntobemoreefficient(achievingatleastthesamequalityofcareatlowercost)thaninstitutionalcareforthefrailelderly(deJongeetal.2014).Whilethefrailelderlyoftenprefertoremainathome,theycannoteasilytraveltouseoffice-basedprimarycareservices.Thissometimesresultsinhospitalizationthatwouldnothaveoccurredinthepresenceofhome-basedprimarycareservices(e.g.Stall,NowaczynskiandSinha2013).Similarly,theavailabilityofsocialcareservices(institutionalorhome-based)whichprovideservicessuchasassistancewithactivitiesofdailylivinghasbeenshowntosignificantlyaffecttheextentofdelayeddischargesfromhospitals(NationalAuditOffice2016).Inadequatecoordinationwithoralackofaccesstosocialcareleadstoa“defaulttodoctor”phenomenon,whichinturncreatesadditionalpressuresonmedicalstafftime,whichtheyareill-equippedtohandle(Dorell2015).
Surgeryformanyconditionsmaybeperformedinlesscostlydaycaresettingsinsteadofacuteinpatientcaresettings,whileachievingthesameresults.Forexample,almostallcataractscouldberemovedwithambulatorysurgery.Despitetheseopportunities,daycaresurgeriesarenotfullyused,eveninhigh-incomecountries.AlthoughwelloverhalfoftheOECDcountriesnowconductover90%ofcataractsurgeriesindaycaresettings,somestillhaveratessubstantiallylower–31%reportedforPoland,50%forHungaryand72%inAustria(James,BerchetandMuir2017).
Thereisalsoconsiderableevidenceofunnecessaryuseofoutpatientspecialistandemergencydepartmentsinbothlowandhighincomesettings,whereunnecessaryisdefinedasvisitswhichcouldhavebeentreatedinlesscostlylevelsofthesystemwithnoreductioninoutcome.Forexample,inEstoniaarecentstudyshowedthatapproximately20%ofvisitswithoutpatientspecialistsfordiabetespatientsandnearly70%ofvisitsforhypertensionpatientscouldhavebeen
17
treatedinprimarycare9(WorldBank2015).Emergencydepartmentvisitsper100populationvariedfrom10to70inOECDcountriesinrecentyearssuggestinginefficiencyinatleastsomecountries.Recordreviewsfromanumberofthesecountriessuggestedthatbetween12%and56%ofthesevisitscouldhavebeentreatedinlesscostlysettingswithoutanydeclineinqualityofcare(James,BerchetandMuir2017).Bypassinglesscostlyprimarycareprovidersformoreexpensivehospital-basedcareisalsoacommonphenomenoninlowandmiddle-incomecountriesincludingChina,Kenya,NamibiaandTanzania(Wuetal.2016;Nshimirimanaetal.2016;Lowetal.2001;Kahabukaetal.2011).
Thedeliveryofcareininappropriatecaresettingsisdrivenbyanumberoffactors.Theliteraturesuggeststhatthesefactorsmayinclude:
• Systemdesignfactors(e.g.organizationalseparationofhealthandsocialcare,weak
gatekeepingfunctionsofprimarycareproviders,lackofaccesstoprimarycareprovidersafterhours,etc.);
• Contractingandfinancing(e.g.weakfinancialincentivesforsolvingmedicalproblemsinprimarycare,strongfinancialincentivesforincreasingthevolumesofoutpatientspecialistandinpatientcare,etc.);
• Clinicalandprofessionalmechanismsandprocess(e.g.lackofadherencetoclinicalguidelinesandpathways);and
• Healthsysteminputs(e.g.shortageofhomenursingproviders,lackoffullyfunctionale-platformforelectronicmedicalrecordsandpatientreferralcoordination)(WorldBank2015).
InEstonia,weakprimarycareincludingweakmanagementofpatientswithchronicdiseases,
lowadherencetoevidence-basedpractice,limitedservicescopeandknowledgegapswereshowntobeparticularlyimportantcontributorstoavoidablespecialistvisitsandhospitaladmissions.Asaresultofthesefactors,lowtrustofprimarycareprovidersamongpatientsmaycontributetoself-referralsandbypassing.InKenya,forexample,patientspreferredtobypassthelesscostlyprimarycaregatekeepersandgodirectlytomoreexpensivehospitalsbecauseofperceivedpoorcommunication,longwaitingtimesandbeingtreatedwithoutdignityandrespectinprimarycare(Nshimirimanaetal.2016).
Adequatecoordinationandcontinuityofcarewithinandbetweencaresettings(e.g.receiptoffollow-upcareafterahospitaladmission)isalsocriticaltopreventingfurtherdeteriorationinpatient’shealth,whichinturnmayrequirerepeatoutpatientspecialistoremergencydepartmentvisits,orreadmissionstoacuteinpatientcare.Thisincludescoordinationwithendoflifepalliativecarewheremanagedcarebypalliativeteamshasbeenshowntoreducehospitalizationrates(e.g.Reyniersetal.2014;Seowetal.2014).
3) Doingthingsright
Oncedecisionshavebeenmadeaboutwhichinterventionsshouldbeavailable,wheretheyshouldbeavailable,andthebalancebetweenexpandingservicesandexpandingfinancialprotection,thenextquestionishowtogetthemostoutofthedifferenttypesofinterventionsforthelowestcost.HerewediscussbrieflytheelementsincludedinTable1under“doingthingsright”.
9Thisindicatorlooksatspecialistvisitsbypatientswhoseconditionsareconsidereduncomplicatedbasedontheprimarydiagnosesmade.Ofthese,visitswereconsideredavoidableifpatientspresentedtoaspecialistnotspecifiedinnationalEstonianguidelines.Ifseveralvisitswerebilledunderthesameclaim(e.g.,pertainingtoonecarecycle),thedecisiononwhetherthesevisitswereavoidablewasmadebasedontheprimarydiagnosiscodeassignedtotheclaim.
18
i. HealthServices
ProbablythebestdocumentedsourceofinefficiencyfromTable1relatestomedicineswhichaccountforsomewherebetween20%and60%ofallhealthexpendituresinlowandmiddleincomecountries,andaround18%inhighincomesettings(Niessen&Khan2016).Widespreadover-prescriptionandoveruseofmedication,andincorrectprescriptionanduse,particularlyofantibiotics,iswelldocumented(e.g.Kalyangoetal.2015;Lietal.2012;Hollowayetal.2013;Maoetal.2015;OECD2017).Holloway(2011)andHollowayandDijk(2011)suggestthatglobally,lessthan50%ofpatientsreceiveappropriatemedication(comparedtotreatmentguidelines)fortheirconditions:evenfewerinlowandmiddleincomecountries-40%ofpatientsinthepublicand30%intheprivatesectorsrespectively.
Lowadherencetorecommendedtherapywastesresourcesbecauseofanincreasedneedforsubsequentmedicalcare(Pereiraetal.2014;Ryanetal.2014;Ali,Abou-TalebandMohamed2016;Choudhryetal.2016).Therearemanydeterminants,butlowadherenceismorelikelywithlong-termtherapythanforanacuteepisode,andalsowhereoutofpocketpaymentsformedicinesarerelativelyhigh.Anincreasingliteratureontheaffordabilityofmedicinessuggeststhat,ingeneral,affordabilitydeclineswithlevelsofnationalincomeperhead(e.g.Cameronetal.2009;Cameronetal.2012;Jiangetal.2015;Iyengaretal.2016;Khatibetal.2016).10Inlowandmiddleincomecountries,medicinesfrequentlyhavetobepaidforout-of-pocket,sothelackofaffordabilitytranslatesintoreducedadherencetoafullcourseoftreatmentand,forsomepeople,aninabilitytopurchaseandbenefitatallfromneededtherapy(Luetal.2011;Niesen&Khan2016).
Thelackofaccesstomedicinesthatpeopleneedisalsorelatedtomanycountriespayingtoomuchformedicines.Pricesforthesamemedicinesvarysubstantially,evenacrossEuropeanandOECDcountries,forbothgenericsandbrandnamemedicines(Cameronetal.2012;Simoens2012;Vogler&Kilpatrick2015;OECD2017).11Whilesomevariationsinpricescanbeexpectedgiventhedifferencesacrosscountriesinthesizeofthemarket(population,diseaseprevalence)andtransportcosts,Iyengaretal.(2016)illustratetheremarkablevariabilityinpricesacrossOECDcountries.Asanexample,thepriceforacourseofSofosbuvir(forHepatitisC)rangedfromUS$37,729toUS$64,680,withamedianofUS$42,017.Inastudyof46largelylowandmiddleincomecountries,publicsectorpricesforselectedgenericswerebeen5and17timesabovetheinternationalreferenceprice,withoriginatorbrandsalmost30%higher(Cameronetal.2012).Thegapbetweenthepricespaidandtheinternationalreferencepriceintheprivatesectorwasevengreater.SimilarresultswerefoundmorerecentlyinChina(Jiangetal.2015).Otherstudiesrevealcountrieswherebrandnamepricesarenohigherthantheequivalentgeneric,suggestingthosecountriesarepayingtoomuchfortheirgenericmedicines(Cameron&Laing2010;Vogler&Kilpatrick2015).
Reducingoveruseandinappropriateuse,andreducingthenegotiatedpriceformedicines,freesupresourcesthatcanbereinvestedinwaysthatimprovehealthorfinancialprotection,althoughsometimesthepharmaceuticalindustrylowersthepriceofsomemedicines,butincreasesthepricesofothers.Policiestoreplaceoriginatorbrandpurchaseswithgenericorbiosimilarmedicinesalsoresultinsavings.12Whileanumberofcountries
10Evenifthesecountriescannegotiatelowerprices(notallcan),thedifferenceinpricebetweenricherandpoorercountriesislessthanthedifferenceinaveragehouseholdincomesmeaningthataffordabilityfallswithdeclinesinnationalincomepercapita.11OECD(2017)pointsoutthatpricecomparisonscanbedifficultbecauseofdifferencesinpackagingacrosscountriesaswellasthesecretnatureofsomeofthediscountsthatcountriesnegotiate.12“Abiosimilar(alsoknownasfollow-onbiologicorsubsequententrybiologic)isabiologicalmedicalproductwhichismostanidenticalcopyofanoriginalproductthatismanufacturedbyadifferentcompany.Biosimilars
19
havegenericpoliciesofvarioustypes,marketpenetrationismuchlowerthanoptimalfromtheperspectivepurelyoffreeingscarcehealthresourcesbyswitchingfromhighercosttolowercost,equivalenteffect,options(Dylst,VultoandSimoens2014;Hassalietal.2014;OECD2017).Estimatesoftheproportionalcostsavingfromswitchingfromspecificbrandnametogenericmedicinesinclude51%inPakistanand53%inChina(Cameron&Laing2010);between4%and23%inAustria(Heinzeetal.2015)andbetween11%and73%in17lowandmiddleincomecountries(Cameronetal.2012).Indollarterms,Haasetal.(2005)suggestedthattheUScouldsavearound$9billionannuallybyintroducingagenericspolicy,whileMulcahyetal.(2014)suggestthesavingsfromswitchingtobiosimilarswouldbelower-around4%ofcurrentspendingonbiologicals-butstillsubstantialataround$44.2billionfrom2014to2024.
Inefficienciesrelatedtoinfrastructurehavelargelyfocusedonhospitals,andlessfrequentlyonlowerlevelhealthfacilities.Theconsiderablevariationinefficiency(measuredessentiallyastheratioofhealthfacilityoutputstotheirinputs)acrosshealthfacilitieshasbeenextensivelydocumentedincountriesatallincomelevelsthroughtheuseoffrontierproductionfunctionanalysis(e.g.Kirigiaetal.2011;Besstremyannaya2013;Kiadaliri,JafariandGerdtham2013;Duetal.2014;Jehu-Appiahetal.2014;Kittelsenetal.2015).13Inefficiencyinthiscontextgenerallymeansthatthesamethroughputofpatientscouldbehandledwithfewerinputs(comparedtothemostefficienthospitals).Anumberofthestudiesinlowerincomecountrieshaveidentified“over-staffing”ininefficienthospitals(KirigiaandAsbu2013;Kirigia,SamboandLambo2015),whichcanbelinkedtotheinappropriatemixofinputsdescribedinTable1asoneofthecausesofinefficiency.Overstaffingcouldalso,however,beinterpretedasshowingthatthesamestaffandinfrastructurecouldcopewithmorepatients,suggestingunder-utilizationoftheinfrastructureandstaff.
Someofthesestudiesalsoexaminedeconomiesofscaleoroptimalhospitalsize.Frequently,inefficienthospitalsaresmallerthantheirefficientcounterparts,althoughastudyfromSouthAfricashowedthatsomehospitalswere“toolarge”andsome“toosmall”comparedtotheefficientsetofhospitals(Preyra&Pink2006;Kristensenetal.2012;Leleu,MoisesandValdmanis2012;Kirigia,SamboandLambo2015).AstudyfromtwoprovincesinCanadashowedthattheoptimalhospitalsizevariedbyprovincesoitisreasonabletoassumevariationacrosscountries(Asmild,Hollingsworth&Birch2013).InsomeoftheseminalworkonhospitalcostfunctionsandeconomiesofscaleintheUSA,LaveandLave(1984)suggestthatitisdifficulttoidentifyoptimalsizeofhospitalsfromcross-sectionaldatabecausethenatureofsmallandlargerhospitalscanbequitedifferent.Theyarguethattheoptimalsizedependsonthescopeandcomplexityoftheservicesoffered,sothereislittlegenericguidancethatcanbegiventocountriesabouttheoptimalsize.Detailedstudiescontrollingforscopeandcomplexityofserviceswouldneedtobedoneineachsetting.Questionsofpossible“over-staffing”could,however,beexaminedinotherwaysdiscussedmoreinSection4.
Theavailabledatafromlowandmiddle-incomecountriesonhospitaloccupancyratesalsosuggestssubstantialinefficienciescanexistindistricthospitals.Astudyof18countriesin2007reportedanaveragebedoccupancyrateindistricthospitalsof55%,rangingfrom20%to98%(Chisholmetal.2010).InBotswana,districthospitalbedoccupancyratesin2009werebetween40%and61%,butatthesametimethetworeferralhospitalshadoccupancy
areofficiallyapprovedversionsoforiginal"innovator"products,andcanbemanufacturedwhentheoriginalproduct'spatentexpires”,Wikipedia,accessed8February2017.13ThesemethodsarediscussedmorethoroughlyinSection4inrelationtotrackingprogressinimprovingefficiency.
20
ratesof143%and222%respectively,suggestingefficiencyandprobablyqualitygainsfromencouragingadmissionsintodistricthospitalsratherthanreferralhospitals(Seitio-Kgokgweetal.2014).
Healthcaredeemedtobeinappropriate,ineffectiveorharmful,sometimescalled“unnecessary”care,hasbeenwidelydocumentedinhighincomecountries(OECD2017;Sainietal.2017;Brownleeetal.2017).Unnecessarytreatmentorinvestigationsarethosewhichhavelittlelikelihoodofimprovingthepatient’squalityordurationoflifeorwhichhavemorechanceofdoingharmthangood.Thelistofcommonformsofunnecessarytreatmentincludesimagingforlowerbackpainandheadaches,antibioticsforupperrespiratorytractinfections,preoperativetestingforlowriskpatients,cardiacimaginginlowriskpatients,inductionoflabour,caesareansectionandsometypesofcancerscreening(Hurley2014;OECD2017;Brownleeetal.2017).
Ineffectivecareincludes:theuseofvitaminandmineralsupplementsaimedatpreventingcardiovasculardisease;antipsychoticprescriptioninolderpatientswhichincreasetheirriskoffalls;orinterventionsundertakenatthewrongtime,ornotatall.Anextremeversionofineffectivecareiscarethatisharmfultopatients,includingmedicalerrors.Ineffectiveandharmfulcarecanleadtosubsequentoutpatientvisits,hospitaladmissionsandotherformsoftreatmentthatcouldhavebeenavoided,andattheextreme,toavoidabledeaths.Reducingallformsofunnecessary,inappropriateandharmfulpatientmanagementsavesmoney(sometimesalsoimprovinghealthoutcomes)thatcanbere-investedintointerventionsthatimprovehealthand/orincreasefinancialprotection.
Theevidenceofover-useisfrequentlycomplicatedbythefactthatpatientsdifferintermsofneed,butformanyoftheseexamples,theevidenceofwidevariationsintheiruseacrosscountries,adjustingforpopulationdifferences,istakentoimplyover-servicinginatleastsomeofthecountries.Forexample,thenumberofCTexaminationsper1000populationvariedfrom31.9inFinland,to254.7intheUSA,with15ofthe28countriesabovetheOECDaverageof131.6(2014data;OECD2017).
ForCaesareansectionandtheuseofantipsychoticsinelderlypatients,therearestandardsofneedthatenableanassessmentofoveruse–15-20%ofdeliveriesforCaesareansand0%foranti-psychotics(reportedinOECD2017).In2014,ratesofdeliveryusingCaesareansectioninOECDcountriesvariedfrom15.3per100livebirthsinIcelandto51.1inTurkey,withanOECDaverageof27.5.Theaveragerateincreasedbetween2007and2014,althoughitfellin12ofthe32countriesforwhichinformationwasavailable(OECD2017,Figure2.2).
Innon-OECDcountries,ratesofCaesareansectionarealsoincreasingandinmanyarealreadyabovetherecommendedrange(Yeatal.2015;Betranetal.2016).Theaverageacrossthecountriesclassifiedas“lessdeveloped”usingtheHumanDevelopmentIndexin2014wasalreadyalmost21%,rangingfrom1.7to56.4per100livebirths(Betranetal.2016).Oftheestimated6.2millionunnecessaryCaesareansectionsannuallyacrosstheworld,50%occurinBrazilandChinaalone(Berwick2017).
Scatteredexamplesofothertypesofunnecessary,inappropriateandharmfuluseofhealthservicesinlowandmiddle-incomecountriesexist.Presumptivetreatmentoffeverswithanti-malarialmedicationinsomecountrieswastesresourcesandisnolongernecessary,giventheavailabilityofinexpensiverapiddiagnostictests(Ochodo,GarnerandSinclair2016).Otherexamplesinclude:theoveruseofmedicineswhenprescriberssellthem(officiallyorunofficially)andthelongerlengthofhospitalstaylinkedtopaymentperday(Chenetal.2014;Gao,XuandLiu,2014;Rahmanetal.2014;Zhangetal.2015b);substantialoveruseofantibiotics,forexample,inchildrenwithacutediarrhoea,inIndiaandThailand(Brownleeetal.2017);andunnecessarycardiacproceduresinBrazilandIndia(Brownleeetal.2017).Thereisnoreasontobelievethatovertreatmentandinappropriate
21
treatmentisanylessinlow-incomesettingsthaninhigh-incomecountriesforpeoplewhocanpayorwhosecostsarecoveredbygovernmentorinsurance.
AdverseeventsthatcouldhavebeenpreventedarealsodocumentedinanumberofOECDcountries;thesecanleadtotheneedforadditionaltreatmentandsometimesresultinunnecessarydeaths(OECD2017).Ratesofpreventableadverseeventsinhospitalsacrossthe10studiesreportedbytheOECDrangedfrom1.0%to8.5%.Similarevidencefromlowandmiddleincomecountriesislessreadilyavailable,althoughthereisevidencethatanumberofcountries,includingChina,aredevelopingapproachestoimprovepatientsafetyandconsideringstaffperceptionsoftheproblemasthebasisfordevelopingpossiblesolutions(e.g.Wangetal.2014;Zhouetal.2015).
Under-useofnecessaryservicesexistsalongsideoveruseglobally,inthesamecountry,andeveninthesamepatient(Sainietal.2017).Underuseisprobablymorewidespreadinlowandmiddle-incomecountriesthaninhigh-incomesettings.Indirectevidenceofunderuseisfoundinthefiguresthat1.5millionchildrendieeachyearfromvaccinepreventabledeaths,andanestimated84%ofpre-termdeathsarepreventablewithappropriatecare(Glasziouetal.2017).
Moredirectevidencecomesfromanassessmentofparticulartypesofinterventions.Mostoftheannualestimatedunmetneedforsurgicalinterventionsof320millionprocedureswasinlowandmiddle-incomecountries(Glasziouetal.2017).Forexample,inthe“leastdeveloped”groupofcountriesbasedontheHumanDevelopmentIndex,theaveragerateofCaesareansectionin2014wasonly6%ofalldeliveriescomparedtotherecommended10-15%(rangeacrosscountriesfrom1.4to41.1%).IntheAfricancountriesforwhichdatawereavailable,theaverageratewasonly7%suggestingthatmanywomenwhoneedaCaesareansectionstilldonotobtainthisimportantlife-savingintervention.
Underuseisinefficientinthesensethatbyfailingtoaccessappropriate,low-cost,effectiveinterventions,manyoftheaffectedpeoplewillneedtousemorehealthservicesinthefuture,atalaterstageinthenaturalhistoryofthedisease,leadingtounnecessarydeathsandmorbidity.
Problemswiththehealthworkforcehavebeenwidelydocumented.Inhighincomecountries,insufficientdomesticproductionhasledtotheneedtoimporthealthworkersfromothercountries,whichcanhaveunintendedeffectsonefficiencyifthearrivalsdonothaveagoodcommandofthelocallanguageorculture(Aluttis,BishawandFrank,2014;CrispandChen2014).Inlowandmiddle-incomecountries,thereisasevereshortageofhealthworkers.Only5of49high-needcountriesareachievingtheminimumthresholdof23nurses,doctorsandmidwivesper10,000populationneededtodeliveranessentialsetofmaternalandchildhealthservices(CrispandChen2014;WHO2017e).Theshortageofhealthworkersismuchmoreacuteinremoteandruralareas,atrendnotlimitedtolowerincomecountries(Morrelletal.2014;Abimbolaetal.2015;WHO2017e).
Themixofskillsalsovariesconsiderablyacrosscountries,suggestingpossibleinefficiencyatleastinsomesettings.Forexample,Indiahasaratioofapproximatelyoneallopathicdoctortoeachnurseandmidwife(Rao2014).Althoughthereisnogoldstandard,mostcountrieshaveconsiderablyfewerdoctorspernurseandmidwife–e.g.Indonesiahad0.16;Thailand0.23atthetimeofRao’sIndiastudy(Rao2014;WHO2017f).
Communityhealthworkersbecamepopularinthe1970sinanefforttomoveservicesclosertothepeoplewhoneededthem.Theirnumberdeclinedinthe1980s,butarecentsurgeininteresthasoccurred,linkedtotheneedtogetservicesclosetopeopleandhelpmitigateshortagesinothertypesofhealthworkers(Perry,ZulligerandRogers2014).Anumberof
22
apparentsuccesseshavebeenreported,includingladyhealthworkersinPakistan(Yousafzaietal.2014);Ethiopiaisintheprocessnowofexpandingthenumberanddistributionoftheircommunityextensionworkers(Nejmudinetal.2011).Alargenumberofstudiesoftheseworkers’efficacyhaveproducedinconsistentresults.WHOisdevelopingasetofguidelinesforcommunityhealthworkerprogramsandlastyearcalledforexpressionsofinteresttoundertakesystematicreviewsofthevariousstudiestohelpitinthisprocess(WHO2017g).
Otherproblemsaffectingtheefficiencyofhealthworkersinlowandmiddleincomecountriesarelinkedtomotivation,absenteeism,retentionanddualpractice(Abimbolaetal.2015;Hotchkiss,BanteyergaandTharaney2015,Witteretal.2015).Absenteeismreducesthenumberofservicesthehealthworkercanprovide,whilehighturnoverofstaffrequiresretrainingandrelearningthatalsoreducesefficiency(Daouk-Öyryetal.2014).Limitedinformationisavailableontheextentofabsenteeism–twostudiesfromTanzaniasuggestedsubstantiallostproductivityfromabsenteeism.Kurowskietal.(2004)estimatedthattherewasa26%reductioninhealthworkerproductivitybecauseofunexplainedabsencesandbreaks,whileManzietal.(2012)reportedthat44%ofthestaffwerenotavailableatthetimeofthestudyvisit,andnursesworkedonlyfor57%oftheirallottedtimebecauseofbreakstakenwhileonduty.Comparabledataonthesefactorsarenotpubliclyavailableacrosscountries,soitisdifficulttomakegeneralizationsontheirimportance–althoughhealthministrieswillwanttotrackthesevariablesintheirsearchforgreaterefficiency.
In2010,salariedhealthworkersaccountedforjustover42%ofgovernmenthealthexpendituresglobally,lowerinAfricaandSouth-EastAsiacomparedtoEuropeandtheAmericas(Hernandez-Peñaetal.2013).Intermsoftotalhealthexpenditures,remunerationofhealthworkersinthegovernmentsectoraccountedfor34%whileindependenthealthworkerremunerationaccountedforanother10%globally.Whenfundsareshort,ministriesofhealthfrequentlyprioritisepayinghealthworkers,leadingtoreportsofhealthworkersbeinginpost,butdeprivedoftheinputssuchasmedicinesthattheyneedtodotheirwork(Moszynski2016).However,thereislittlesystematicdataonthistypeofinefficiencyinthemixofinputsacrosscountries.
ii. HealthSystemsandtheHealthFinancingComponentWasteintheformoffraudandcorruptionoccursinhealthsystemsatanumberoflevels(seeOECD2017).Inservicedelivery,patientscanmakewrongfulinsuranceclaimsandproviderscanbillforpatientsorservicestheydidnotprovide.Theopportunitiesareparticularlyhighinprocurement,includingthebiddingprocess,andinthesubsequentdistributionoftheinputsthatwereprocured.Inhumanresourcemanagement,thisextendstotakingbribesinreturnforanofferofemployment,oremployingfriendsandrelativeswhomightnotbethebestpersonforthejob.
ATransparencyInternationalReportquotedbyOECDsuggeststhatathirdofrespondentsacrossthe28participatingOECDcountriesbelievedtheirhealthsystemswerecorruptorextremelycorrupt,withNGOs,themilitaryandtheeducationsystembelievedtobelesscorruptthanthehealthsystem(TransparencyInternational2013;OECD2017).Forty-fivepercentofglobalrespondents(103countries)consideredtheirhealthsystemstobecorruptorextremelycorrupt,andinthiscasehealthfaredworsethanreligiousorganizationsandthemedia,inadditiontoNGOs,themilitaryandtheeducationalsystem.
Arecentstudyoffraudinthehealthsectorsuggeststhatsomewherebetween3%and10%ofhealthexpendituresarelostannually,withameanof5.6%.Extrapolatingtotheworld,corruptionandfraudisestimatedtocosttheworld$426billionannually,resourcesthat
23
couldbeusedtoimprovecoveragewithneededhealthservicesandwithfinancialprotection.14
Poorpublicfinancialmanagementpracticesalsolimittheefficiencyofgovernmenthealthsystems.Therearemanyproblems,includingunpredictablebudgetallocations,fragmentationinrevenuestreamsandfundingflows,unpredictableorlatedisbursementfromMinistriesofFinancetolineministriesandfromcentralministriestosub-nationalunits,lowbudgetexecutionrates,andinadequatefinancialaccountabilityandtransparency(Brixietal.2012;Foxetal.2013;Cashinetal.2017).Thesepracticesresultintheavailablemoneynotbeingspent,inpoorspendingbecausemoneyarriveslateorcannotbetransferredbetweenlineitems,orinleakagesbecauseoflimitedtransparency.Toillustrateonlyonepartofthis,recentpublicexpenditurereviewsfrom6Africancountriessuggestunderspendsoftheapprovedbudgetrangefrom$10to$120millionayear(WHO2016).Inpercapitaterms,thistranslatesintolossesofpotentialspendingofbetween$1and3.50annually.Datafromtwocountriesallowedacomparisonofexecutiononsalariesversusothertypesofexpenditure.Almosttheentirebudgetforsalarieswespent,whileexpenditureontheinputsneededtokeepthehealthsystemrunningranataround40%.Additionalsourcesofwasteinasystemarelinkedtoadministrativeformsofadministrativeinefficiencysuchashigherthannecessaryadministrativecosts,bureaucraticredtapeanddelays,complexsystemsthattaketimetonegotiate,andincreasinglyhealthcareprovidersspendingtheirtimeonadministrativeissuesratherthaninteractingwithpatientsorthepopulation(OECD2017).Informationontheextentofadministrativecostsinhealthsystemsisdifficulttofindforlowandmiddle-incomecountriesinparticular.Itexistslargelyforhealthinsuranceadministrationbuteventhen.Forexample,intheearly2000s,socialhealthinsuranceadministrationcostsaveraged3.8%oftotalinsuranceexpendituresin15high-incomeOECDcountries,datacomingfromcountryhealthaccountstudies.Thelowestproportionwassomewhereunder2%inEstonia(rangingfrom1.1to1.9%dependingontheyear)andthehighestaround7%inLuxembourg(yearlyrangefrom6.6to7.0%).Twoofthethreemiddle-incomecountriesforwhichdatawereavailablehadadministrationcostsoflessthan3%oftotalinsuranceexpenditures(GeorgiaandTurkey)butinMexicotheywerealmost17%.
Administrativecostsinprivateinsurance,againusingcountryhealthaccountsdata,wereingeneralsubstantiallyhigher,consistentlyover10%in14ofthe23high-incomecountriesforwhichdatawerecollated,reachingasmuchas30%.Ontheotherhand,inNewZealandtheywereonly5%.Thevariabilityintherangeofadministrativecostsacrosscountriesandtypesofinsurancesuggestssomearesubstantiallymoreefficientthanothersandtherearepossiblegainstobemadefromreducingtheshareofinsuranceexpendituresgoingtoadministration.
Thefinalsourceofinefficiencydiscussedhereisfragmentationwithinsystems.Infinancingsystems,thiscanmanifestitselfintermsofmultiplepayersandpurchasersfordifferenttypesofhealthservicesordifferentpartsofthepopulation.Thisisfrequentlycausedbyfragmentationinfundpools–healthinsuranceco-existingwithgovernmentfinancingandprovision,ormultipleinsurancepoolseachofwhichpurchasesservicesfortheirclients.Suchfragmentationleadstohigheradministrativecostsandhigherpricesthancouldbenegotiatedbyasinglelargepayer.Fragmentationcanalsobeanobstacletoequitable
14Globalhealthspendingin2014wasUS$7.6trillion(WHOGlobalHealthExpendituredatabase).
24
coverage-poorerpopulationgroupsarecoveredbyschemesthatarelesswellfundedthanthosefortheformalsectororgovernmentemployees,forexample,soofferasmallerrangeofhealthservices(Tucmeanu2014;Mengetal.2015;Kutzin,YipandCashin2016).
Inbroaderhealthsystems,examplesoffragmentationincludelaboratoriesfortuberculosisseparatetothoseforHIV/AIDS;multipleprocurement,purchasing,distribution,accounting,monitoringandservicedeliverysystems;orpatientsrequiringlongtermcarebeingpushedbetweennursinghomesandhospitalsbecausedifferentpartsofgovernmentpayforthedifferenttypesofcare,andeachseekstominimizetheirowncosts(Sidibé&Campbell2015,Raoetal.2014).Someoftheseissuesarefoundincountriesatallincomelevels(e.g.Hall2015;Lewis2015),butanadditionalfeatureofmanylowandlowermiddle-incomecountriesisthefragmentationassociatedwithinflowsofdevelopmentassistanceforhealthwhereexternalpartnershavechosentoestablishsystemsparalleltothosethatalreadyexistedratherthanstrengtheningandsupportingexistingsystems(e.g.Kienyetal.2014;Panter-Brick,EggermanandTomlinson2014;Gostin&Friedman2015).TheproblemsassociatedwithhowtobestpreservethehealthgainsassociatedwiththeseprogramsarenowbeingfacedincountriestransitioningfromfundingfromGavi,GlobalFundandsomebilateralaidagencies.
Thequestionofwhetherdecentralizationisaformoffragmentationassociatedwithinefficiencyisstillcontroversialdespiteyearsofexperience.Decentralizationpolicieshavebeenwidelyimplemented,notjustinhealth,withthegoalofimprovingoneormoreofthefollowing:efficiency,servicequality,management,responsivenesstolocalneedsandequity(Saltman,BankauskaiteandVrangbaek2004).Thereismixedevidence.Somestudiesfindthatithasbeenassociatedwithincreasedaccountabilityofgovernmenttocitizens,orwithimprovementsinmanagementthatthenresultedinhighercoveragewithhealthservicesandimprovedhealthoutcomes(Alves,PeraltaandPerelman2013;Loubiereetal.2009;Samadietal.2013)Otherstudiesshowlittleimpactofhighercostsandincreasedinequalitiesbetweenlocalgovernmentunitsduetodifferencesinfiscalcapacity(Azfar,KähkönenandMeagher2001;Atkinson&Haran2004;Saltmanetal.2007;Langenbrunner,XuandChu2016).Therehavealsobeensuggestionsthatrenewedcentral-levelfinancingwouldimproveequityandefficiencyindecentralizedsettings(Langenbrunneretal.2016).
Oneofthereasonsforcontradictoryresultsisthatdecentralizationtakesmanyforms,andcanhavemanycomponentseachrequiringlocalgovernmentcapacity–e.g.raisingrevenue,planning,purchasingorprovidingservices,monitoringandevaluation,audit.However,thetopicremainscontentiousandwillbediscussedatthisForum.
V. IdentifyingthemostimportantsourcesofinefficiencyNotallcountrieswillhaveeachofthesourcesofinefficiencydescribedinTable1,andeven
wheretheyexist,theirrelativeimportancewillvaryacrosssettings.However,someinefficiencyexistsineverycountry–everycountrycouldachievemorewiththeavailableresources(WHO2010a).Thereislittleglobalguidanceavailableonhowtoidentifythemostimportantsourcesinaspecificsetting,whichiswhythissectionhasspentsometimeexplainingthecommoncausesofinefficiencyandtheirsources.Thiscanbeastartingpointforcountriesthinkingthroughwhichonesarethemostimportantintheirsettings,andwhichtheywouldtacklefirst.Indicatorsofinefficiencycanhelpprovideevidenceinsupportofdiscussionsaboutthemostimportantsourcesofinefficiency
25
andsubsequentpolicydevelopment,sotheyarediscussedinthenextsection.
VI. Measuringandmonitoringinefficiency
Therearetwobroadapproachestomeasuringinefficiencyinhealthsystems.Thefirstseekstoobtainasinglescoresummarizingtheefficiencyofacountry’sentirehealthsysteminasimilarwaytothemeasurementofhospitalorhealthfacilityefficiencydescribedearlier.ThesecondapproachfocusesonmeasuringefficiencyrelatedtothedifferentcomponentsofahealthsystemandthepossiblesourcesofefficiencydescribedinSection3.Theformerwecallmacro-efficiencyandthelatter,micro-efficiencyanalysis.Theyarediscussedbrieflyinturn.
1) Macro-efficiencyFigure1depictstherelationshipbetweenlifeexpectancyatbirthandtotalhealthexpenditure
percapitain2014.Lifeexpectancyriseswithhealthexpenditurepercapita,thoughatadecreasingrate.Thereisalsoconsiderablevariabilityaroundtheaverageregressionlinewithcountrieswithsimilarexpendituresachievingverydifferentlevelsoflifeexpectancy.Thistypeofanalysis,sometimeswithlevelsofattainmentonotherhealthindicators(e.g.maternal,childorinfantmortality),orwithlevelsofcoverageonkeyinterventions(e.g.withchildhoodimmunizations)ontheverticalaxis,hasbeenusedasevidencethatsomecountriesaremoreefficientthanothersintranslatingexpenditureintohealthoutcomesorcoverage(WHO2010a;WHO2016).
Figure1:Lifeexpectancybyhealthexpenditurepercapita,2014
Manyfactorsotherthanhealthexpendituresobviouslycontributetotheobservedvariationsinlifeexpectancy,includingdifferencesininitialdiseaseburden,populationagestructuresanddistribution,coststructures,availabilityofinfrastructureandhistoricalpatternsofspendingon
40
45
50
55
60
65
70
75
80
85
90
0 1 2 3 4 5
Life
exp
ecta
ncy
Log health expenditure per capita
26
healthandsocialservices.Withoutcontrollingforthesefactors,itisnotpossibletostatecategoricallythattheobserveddifferencesarelinkedtodifferencesininefficiency.
Moreformalmethods(mostcommonlydataenvelopmentanalysis(DEA)andstochasticfrontierproductionfunctionanalysis(SFA))havebeendevelopedtomeasuretherelativeefficiencyofcountryorsubnationalhealthsystems(orhealthfacilities)andtocontrolfor,orexaminetheimpactof,possibledeterminantsofoutcomeinadditiontoexpenditure(e.g.Hollingsworth2003,2008,2016;Joumardetal.2010;Pang2005).Essentially,theyshifttheregressionlineuptothetopofthescatterplot,obtainingafrontierofthemostefficientperformers,giventheirlevelsofexpenditure(andotherdeterminants).Thoseonthefrontierarethemostefficient(generallygivenascoreof1),andthosebelowitaredeemedinefficientcomparedtothebestperformers,withascorebetween0and1dependingonhowclosetheyaretothefrontier.15
In2016,Hollingsworthidentifiedover400studiesapplyingthesetechniquestohealthissuesinthelast30years(DEAmorefrequentlythanSFA).Themethodswereinitiallyappliedtohealthfacilities,particularlyhospitals,butsincetheappearanceofthe2000WorldHealthReport,whichrankedhealthsystemefficiencyacrosscountriesusingSFA,theyhavebeenwidelyappliedtostudiesofthecomparativeefficiencyofhealthsystems(WHO2000,Sunetal.2017).
Overtime,thetechniqueshavebecomeincreasinglysophisticatedintheirtreatmentofrandomnoiseintermsofmeasurementerrors,samplenoise,specificationerrors,cross-countryheterogeneityandreturnstoscale(seeOlesen&Petersen2016;Greene2004&2008;Hamidi&Akinci2016).DEAcannowdealwithbothmultipleinputsandoutputs,andrecentstudieshaveevenmergedDEAwithSFAfordifferentstagesoftheanalysisofthedeterminantsofinefficiency(e.g.Berengueretal.2016).Inessence,however,theyalltreatdeviationsfromthefrontierthatarenotexplainedbythedifferencesininputsasinefficiency,thenseektounderstandthefactorsotherthanhealthsysteminputslinkedtothevariationsinefficiency.
Thereareanumberofproblemswiththisformofmacro-analysisfromapracticalpolicyperspective.First,theefficiencyscoresandrankingsaresensitive,sometimesverysensitive,tothemodelspecificationandthedatausedforinputs,outputsanddeterminants(Frogner,Frech&Parente2015;Gearhart2016a&2016b).Whencountriesaregroupedintodifferentcategoriesofperformers,thereismoreconsistency–e.g.somecountriesmostlyfallinthegroupwiththehighestefficiencywhileothersgenerallyfallintheleastefficientgroupregardlessofthemodelspecification(althoughthereisstillsomemovementbetweengroupsdependingonthemethodsused)(DeCos&Moral-Benito2014;Medeiros&Schwierz2015).Inthiscase,thepolicyfocuscanbeonthecountriesthatareconsistentlyfoundtobeinthelowestefficiencygroupalthoughlittlecanbesaidaboutcountriesthatmovebetweengroups.
Secondly,themodelsassumethatcountriesonthefrontierareefficient,whenmoremicro-levelworksuggestsformsofinefficiencyexistinallcountries.
Thirdly,noneofthemodelsasyetcapturethelagsthatmustbeimportantintranslatinginputsintooutputsandoutcomes.Thisisnotsimplyaquestionofusingpanelratherthancross-sectionaldata.Itrelatestotheassumptioninbothcasesthatexpendituresineachgivenyearproducehealthbenefitsinthatyearratherthanoveraperiodofyearsinthefuture.Tocapturethis,modelswherecurrentoutcomesareafunctionofhealthexpendituresinpreviousyearsaswellasthisyearwouldberequired.
Fourthly,themethodsandapproachesareverydifficultforthetypicalpolicy-makertounderstand,sotheresultsaresometimesdistrusted(Hollingsworth2016).
15Differentalgorithmsarepossibleinthisprocess–e.g.inDEA,efficiencycanbemeasuredintheoutputspacebymovingverticallyuptothefrontier,intheinputspacebymovinghorizontallytothefrontier,orasamixofthetwo.
27
Finally,andperhapsmostimportantly,theresultsrarelyhelppolicy-makersidentifyhowtheycanimproveefficiencyorpoliciesfordoingso.Thisisbecausethevariablesthatcanbeenteredintothemodelstoexplaindifferencesinefficiencyneedtobeavailableandcomparableacrosscountries.TypicalexamplesaretheGinicoefficientofincomeinequalityandaveragelevelsofeducationalattainment,variablesthatarenotunderthedirectcontrolofthehealthsector.
Thesemodelscan,however,beausefulstartingpointtothinkaboutwhichcountryhealthsystems(orwithinacountry,whichsubnationalsystems,hospitals,orprimarycarefacilities)seemtosufferfromthemostinefficiency.However,tounderstandthecausesofinefficiencyandthendevelopsolutions,moremicro-variablesneedtobeexplored.
2) EfficiencyincomponentsofthehealthsystemTable2reportssomeoftheindicatorsproposedintheliteraturethatdescribeaspectsofthe
causesofinefficiencydescribedinTable1,withthesourcesalsoindicated.
Table2:PossibleEfficiencyIndicatorsfortheSourcesofInefficiencyinTable1
Domainofinefficiency
Indicatorssuggested Source
Doingthewrongthings
Shareofpublicspendinginremoteareas,as%ofGGHE
WBFSD;Tandon&Cashin(2010)
Shareofpublicspendingthatgoestothepoorest40%ofthepopulation
WBFSD
Shareofnationalspendingonpharmaceuticals Smith&Nguyen(2013)
%offundingallocatedaccordingtoastrategicplanforthehealthsectororaccordingtodistributionofburdenofdisease
Tandon&Cashin(2010)
Doingthingsinthewrongplace
Shareofpublicspendinginprimarycare,as%ofGGHE
WBFSD;Yip&Hafez(2015)
Numberofoutpatientvisitsattertiaryhospitalsper100population
OECD(2017)
Shareofavoidablehospitaladmissionsforcertainconditions(asthma,COPD,diabetes,hypertension,CHF)
Marshall,LeathermanandMattke(2004)
Shareofavoidablespecialistvisitsforcertainconditions(diabetes,hypertension)
WorldBank(2015)
Delayeddischarges/returntousualplaceofresidence(hipfracture,stroke)
CompendiumofPopulationHealthIndicators–NHSDigital(2015);OECD(2017)
%ofsurgeriesconductedinambulatorysettings(cataract,tonsillectomy,inguinalherniarepair,cholecystectomy,laparoscopic)
OECD(2017)
28
SpendingBadly:Inputs
Medicines
Under-utilizationofgenericsorpayingtoomuchforanyspecificmedicine
PercentageofdrugspurchasedbytheMoHthataregenericorgenericsmarketsharesbyvolume(%)
WBFSD;OECD(2017);Heredia-Ortiz(2013)
Averagemedicinecostperencounteroraveragenumberofmedicinesprescribedperencounter
WHO/INRUD
Useofineffectivemedicines,thewrongmedicines,orusingthematthewrongtime
Cholesterol-loweringdrug&antidepressantconsumption
OECD(2017)
Percentageofprescriptionsinaccordancewithclinicalguidelines
WHO/INRUD
Overuseorunnecessaryuse
Percentageofencounterswithanantibioticorinjectionprescribed
Desalegn(2013);Hu,Liu&Peng(2003);Wangetal.(2014);Ferreiraetal.(2013)
Meannumberofdrugs/prescription WHO/INRUD;Bashrahil(2010);Ferreiraetal.(2013)
PercentageofmedicinesprescribedfromanEssentialMedicineListorformulary
WHO/INRUD;Desalegn(2013)
Personnel
Inappropriatemixbetweendifferentcadres;locatedinthewrongplaces;demotivatedworkerswithlowproductivity;poorqualityofcareprovided.
Averageratioofcommunityhealthworkertopopulation
McIntyre&Meheus(2014)
Absenteeismrateforhealthworkers WBFSD,Tandon&Cashin(2010);Heredia-Ortiz(2013);Okweroetal.(2010)
Healthworkerdensityinurbanvsruralareas Yip&Hafez(2015);Lannes(2015)
Ratioofdoctorstototalhealthpersonnelortonursesandmidwives
Heredia-Ortiz(2013)
Densityofphysicians/nurses(per1,000population) Heredia-Ortiz(2013);Lannes(2015)
Staffturnoverorretentionofhealthworkforce Dieleman&Harnmeijer(2006);Lannes(2015);Meessen,Soucat&Sekabaraga(2011)
Infrastructure(e.g.healthfacilities)andequipment
29
Inappropriatehealthfacilitysize,particularlyhospitals,foroptimalefficiency
Numberofhospitalbedsper100,000 OECD(2010);Smith&Nguyen(2013)
Bedoccupancyrate WBFSD;Heredia-Ortiz(2013);WHO(2010)
Turnoverrateforacutecarebeds OECD(2010)
Wrongscaleandscopeofhospitals Heredia-Ortiz(2013);WHO(2010)
Numberofadmissions,discharges
WHO(2010)
Underorover-capacityinhealthfacilities
NumberofconsultationsperdoctorOrnumberofoutpatientvisitsorinterventionsprovidedperfulltimeequivalentworkerorperfacility
OECD(2010);Dieleman&Harnmeijer(2006)
AveragePHCserviceutilizationrate Tandon&Cashin(2010)
Equipmentthatispurchasedandcannotberepairedorisnotusedoptimally
Inappropriatemixofinputs:e.g.healthworkersbutnomedicinesorothermedicalproducts
Availabilityofessentialmedicineslistorkeymedicinestopractitioners
WHO/INRUD
SpendingBadly:OutputsandOutcomes
Healthservices
Inappropriateandunnecessarycare
Averagelengthofstay(ALOS)forhospitalvisitsorforspecificadmissions(followingAMI,cancer)
WBFSD;OECD(2010);AustraliaNationalhealthperformancereporting;Heredia-Ortiz(2013);Cylus,Papanicolas&Smith(2013)
Relativestayindex(numberofdaysspentinhospitalforselecteddiagnostic-relatedgroups(DRGs)dividedbytheexpectednumberofdaysspentinhospital)
Davisetal.(2013);AustraliaNationalhealthperformancereporting
Caesareansectionrates OECD(2017)
MRI/CTscanexamsper1,000population OECD(2017)
30
Averageconsultationtime WHO/INRUD
Underuse Proportionofinfantsthatdidnotreceive3dosesofDTP3immunization
WHO/WorldBank(2015)
ProportionofHIVpositiveadultsandchildrenwhodonotreceiveantiretroviraltreatment
WHO/WorldBank(2015)
Proportionofpregnantwomenwhodonotreceive5antenatalcarevisits
WHO/WorldBank(2015)YipandHafez(2015)
Proportionofpeoplewithhypertensiontreatedandcontrolled
Glasziouetal.(2017)
Medicalerrorsandlowqualitycare
Unplannedreadmissions Davisetal.(2013);OECD(2017);AustraliaNationalhealthperformancereporting
30daymortalityrate(hospitalorforspecificconditionssuchasAMIorIschemicstroke)
Davisetal.(2013);Husseyetal.(2004)
Rateofhealthcareassociatedinfections AustraliaNationalhealthperformancereporting
Postoperativepulmonaryembolism(PE)ordeepveinthrombosis(DVT)inhipandkneesurgeries
OECD(2017)
Postoperativesepsisinabdominalsurgeries OECD(2017)
Obstetrictrauma,vaginaldeliverywith(orwithout)instrument
OECD(2017)
Foreignbodyleftinduringprocedure OECD(2017)
Incidencerateforpertussis,measles,andHepatitisB(vaccine-preventablediseases)
Husseyetal.(2004)
Maternalmortality,childmortality Sajedinejadetal.(2014)
SpendingBadly:HealthFinancingandHealthSystemOrganization
Waste,corruption,fraud.
Healthbudgetexecutionrates WBFSD
Percentageofmedicinesandothers(incl.IVfluids)destroyedduetoexpiration,breakageand/orinappropriatestorageconditions
WBFSD;Heredia-Ortiz(2013);Okweroetal.(2010)
Degreeofcorruption(internationalbenchmarknotspecifictohealthsector)
Tandon&Cashin(2010)
31
Arehealth-specificanti-corruptionpoliciesinplace? WBFSD
Informalpaymentsinhealth,as%oftheOOPexpenditure
WBFSD
%ofgovernmenthealthfundingthatreachesservicesdelivery
TandonandCashin(2010)
Inefficiencyinraisingrevenuesforhealth,particularlywhenrevenueraisingforhealthisindependentfromgeneralgovernmentrevenuecollection.
Governmentrevenueas%ofGDP TandonandCashin(2010)
ElasticityofhealthexpenditurewithrespecttoGDP TandonandCashin(2010)
Healthbudgetas%oftotalgovernmentbudget TandonandCashin(2010)
Internationalhealthassistanceas%oftotalandgovernmenthealthspending
TandonandCashin(2010)
Fragmentationinthesystem:inpooling,butinthebroaderhealthsystem
Arethereadequatedonorcoordinationmechanismstoalignexternalfinancingwithgovernmentpriorities,processes,andthehealthbudget?
WBFSD
Howareproviderspaidanddoesthepaymentmodalitycreateincentivesforcostcontainment,qualityofservicedeliveryorprovisionofservicestospecificgroupsofpeople?
WBFSD,McIntyreandMeheus(2014)
Administrativeinefficiency:higherthannecessarycostsfortheservicesoffered,includinginhealthinsuranceagencies.
Healthsectoradministrativecost,as%ofgovernmenthealthexpenditure
WBFSD
Whattypeofbudgetingprocessisusedinthecountry,e.g.,input-basedoroutput-based,andhowdoesthisaffectproviders’/purchasers’abilitytoallocateresourcestobeinlinewithpriorities?
WBFSD;TandonandCashin(2010)
The list is long, and certainly not exhaustive. For example, an indicator of the extent ofunderuseofneededservicescouldbedefinedforalldiseasesandtypesofinterventionsfocusingonthatdisease,suchasscreeningforcertaintypesofcancers,andsubsequenttreatment.
Someoftheindicatorssuggestedintheliteraturearealsodifficulttoobtain,particularlyinlower income countries, andwould require additional expenditures to establish andmaintain thesystemstocollectandanalysethemroutinely,forexample:theextentofinformalpayments(surveys);share of public expenditures going to the poorest 40%of the population (utilization surveyswithinformationonhouseholdexpendituresorincomes);percentageofmedicinesprescribedaccordingto national guidelines (surveys or observation of encounters);wrong scale and scope of hospitals(production function analysis based on intense data collection); and degree of corruption (publicexpenditurereviewsorspecialformsofaudit).
Othersaredifficulttointerpretbecauseamoveinonedirectiondoesnotunequivocallymeananincreaseordecreaseinefficiency.Theymaystillbeuseful,however,andinternationalyardsticksderivedfrommulti-countrycomparisons,particularlythoseofsimilarcountries,couldbeusedtomake
32
thesejudgements.16Examplesarehealthworkerdensity,distributionofhealthworkerstoruralareas,averagelengthof inpatientstay,andtheshareofnationalspendingonprimarycareormedicines.Eventhoughitispossibletodetermineifacountry’sscoresarehigherorlowerthanthatofsimilarcountries,itisdifficulttobesureifmoreorlessisgoodorbad.Theanalysisismostusefulwhenthereareoutliers–wherethevariableinquestioninonecountryissubstantiallyhigherorlowerthaninothercountries.
Anothersetisreadilyeasytointerpretandmostcountriesshouldbeabletomeasurethemwithoutgreatadditionalexpense.Fromthevariables inTable2,twelvefall intothiscategory.Thisdoesnotmeanthattheyarethemostimportanttomonitororthatthesetofvariablesisthebesttounderstandtheoverallefficiencyofthesystem,butonlythattheyshouldbeabletobemeasuredroutinelyincountriesatallincomelevels.Examplesincludeabsenteeismratesforhealthworkers,theproportionofpregnantwomenwhodonotreceive4antenatalconsultations,infantcoveragewith3doseDTP,post-operativesepsisinabdominalsurgery,Caesareansectionrates,budgetexecutionratesand bed occupancy rates. Higher-income countries with established measurement systems canobviouslymonitoramuchgreaternumberandothercountrieswillchoosetodevelopthesystemstomonitorotherindicatorsdependingontheareasofinefficiencytheychoosetotackle.
Theadvantageofthemicro-efficiencyoverthemacro-efficiencyapproachisthattheindicatorsareeasierforpolicy-makers,healthworkers,patientsandthepopulationtounderstand.Theyarealsodirectlyrelevanttotheanalysisofthemostimportantcausesofinefficiencyinacountry,whichinformsthepolicydebateaboutwhattodoaboutthem.
Thedownsideofusingadashboardofindicatorsisthatitcanbedifficulttounderstandifasystemisgettingmoreorlessefficientunlessallindicatorsimproveatthesametime.Eventhen,itwillnotbeclearifeffortstoimproveefficiencyinonearearesultinlowerefficiencysomewhereelse,inareasthatarenotbeingmeasured.
Weobservethatfewcountrieshavedevelopedaspecifiedsetofindicatorswithwhichtheytrackefficiencyandimprovementsovertime.Itisimportantthattheydoso,perhapsusingTable2asastartingpointandtakingintoaccounttheircapacitytomeasureandanalyseandthecostsofobtainingeachvariable.Theremaywellneedtobedifferentsetsfordifferentactors–e.g.ahospitalmanagermayrequireasetofrelativelyspecificindicatorswhiletheMinisterofHealthmightrequireabroadersetthatsummarizesefficiencyacrosstheentirehealthsystem,butincludeslessdetailontheindividualcomponents.Withoutanappropriatesetofindicators,itwillbedifficulttodetermineifthestrategiestoimproveefficiencyhavebeensuccessful.
VII. StrategiestoimproveefficiencyOncethemajorsourcesofinefficiencyhavebeenidentified,thenextstepistodevelopand
implementstrategiesforincreasingefficiency.Understandingthereasonswhyinefficienciesexistcanhelptoidentifyappropriatestrategiesforreducingthem,sointhissectionweturntoavailabletechnicalsolutions.Thelistofpossiblesolutionsdrawsonpoliciescommonlyputinplaceinresponsetothereasonswhyinefficienciespersist.Remarkably,thereislittleknownaboutwhat
16TheWorldBankGroupisintheprocessoftryingtoprovidetherangeofvaluesformanyoftheseindicatorssocountriescandeterminewheretheyareinrelationtoothercountries.
33
worksatthesystemlevelandthesectionconcludesbyidentifyingsomeofthekeyunresolvedandcontroversialissues.
1) TechnicaloptionsforimprovingefficiencyTable3buildsonTable1,suggestingreasonsforeachpossiblesourceofinefficiency,andsolutionsthathavebeenidentifiedintheliteraturebasedonconsiderablecountryexperience.Solutionsdirectlylinkedtohealthfinancingarehighlightedinbold.Thesesolutionshavebeendrawnfromanextensivereviewoftheliteraturereportingonwhattypesofinterventionshavebeenimplementedwiththeintentionofimprovingefficiency.Thelistis,however,unlikelytobecomprehensive–theliteratureisvastanditispossiblesomeoptionshavebeenmissed;somestrategiesmighthavebeenimplementedatcountrylevelwithoutaformalevaluation;andotherevaluationsmightnothavebeenpublishedormadepublic.Readersareencouragedtoindicatetotheauthorsotheroptionsthatshouldbeincludedinanylistthatcountriescanusetoidentifypossiblesolutionstotheirproblems,particularlythoseforwhichthereisgoodevidencethattheyworkordonotwork.
Table3:PolicyOptionsforImprovingEfficiency17
SourceofInefficiency CommonReasonsforInefficiency
PossibleSolutions
1. Doingthewrongthings:Highcost,lowimpactinterventionsfundedattheexpenseoflowcost,highimpact;inappropriatemixbetweenlevelsofcare;prevention,promotionversustreatment;publichealthandgovernancefunctionsversuspersonalservices;mixofintersectoralactionsorintersectoralversushealthservices
Difficulttoobtainthenecessarytechnicaldatatoguidedecisions;noclearyardsticksfordecidingwhatisenough(preventionvstreatment;publichealthandgovernancevspersonalservices);political,healthworkerorcommunitypreferences(typesofinterventions,levelsofcare);financialinterests(littlepreventionfundedbyinsurance).
Increasedcountrycapacitytogenerateandusekeyepidemiologicinformationonburdenofdiseasetoguidedecisions;furtherdevelopmentofmethodsforassessmentofintegratedpackagesandforincorporationoffinancialprotection;patientempanelment;gatekeeping;increasedservicedeliverycapacityandqualityatlowerlevelsofcare;healthtaxesandfinancialincentivesforpersonalactiononprevention;civilsocietyorganizationsandprofessionalassociationspromotingadequateproviderbehavior.
17ThisTablebuildsonanapproachtakeninWHO(2010).
34
2.Doingthingsinthewrongplaces:Inappropriateuseofhigherlevelversuslowerlevelsettingsforinpatientandoutpatientcare(includingdaysurgery),overuseofhealthversussocialcareandinstitutionalizedversushome-basedlong-termcare.
Insufficientinformationontheefficiencyofdeliveryofpackagesatdifferentlevels;organizationalseparationofhealthandsocialcare,weakcoordinationofcare,includingreferralsystems,poorqualityatlowerlevelsofcare,financialincentivesacrosscaresettingsthatpromotecareathigherlevels.
Methodstoassesstheefficiencyofpackagesdeliveredatdifferentlevelsofthesystem;coordination/integrationofhealthandsocialcare;patientempanelmentandgatekeeping;clinicalpathways;increasedservicedeliverycapacityandqualityatlowerlevelsofcare,includingITinnovationse.g.,tele-medicine,e-consultations);providermanagementnetworks(e.g.,primarycarenetworks);appropriateblendingofpaymentmethods.
3. Spendingbadly:highercostinputschosenorinputsnotachievingtheirmaximumpotential
HealthServiceDelivery1. Medicines:
a) under-utilizationofgenericsorpayingtoomuchforanyspecificmedicine;
b) useofineffectivemedicines,thewrongmedicines,orusingthematthewrongtime;
Nogenericspolicy;provider/patientperceptionsthatgenericsarepoorquality;financialincentivestoprescribebrandedmedicines;poorpurchasingpracticesorcorruption;lackofknowledgeofinternationalprices;highmark-upsortaxesonmedicines.
Inadequateregulation/administrationtocontrolsubstandardmedicines;poorknowledgebyproviders;demandorlowadherencefrompatients.
Genericspolicyandessentialmedicineslistforhealthfacilitiesaccompaniedbyqualitycontrols;informationongenericstoproviders/populationwithqualitycontrolsystem;financialincentivesforprescribinggenericsandnotbrandedmedicines;activepurchasingwithappropriatecompetitivebidding;centralizedprocurement;multi-yearprocurementframeworks;increasedtransparencyinpurchasesandtenders;zerotaxesonessentialmedicines;monitoringandpublicationofmedicineprices.Increasedgovernmentcapacitytoregulatemedicinestoensuresafetyandquality;informationexchangeforprovidersandthepopulation.
35
c) overuseorunnecessaryuse.
Financialincentivestoprescribeandsellmore;industrypromotion;consumerdemand.
Appropriatefinancialincentives–e.g.separateprescribingfromsales;increaseinformationtoprovidersandpatients;regulateandenforcestandardsforindustrypromotion.
2. Infrastructure(e.g.healthfacilities)andequipment:a) Inappropriatehealth
facilitysize,particularlyhospitals,foroptimalefficiency;
b) Underorover-capacityinhealthfacilities;
c) Equipmentthatis
purchasedandcannotberepairedorisnotusedoptimally.
Lackofinformationaboutappropriatesize;patientschoosetogotohigherlevelsofcare.
Toofewfacilitiesforthedemandormaldistribution;patientschoosehigherlevelfacilities–overandundercapacitycoexist;poormanagement;financialincentivesforhighadmissionandlonglengthofstay.
Donationsofequipmentthatcannotbeservicedlocallyorwheresuppliesandmaintenancearetooexpensive;poorprocurementpractices;corruption.
Monitoringefficiencyofhealthfacilities;gatekeeping;increasedservicedeliverycapacityandqualityatlowerlevelsofcare.
Masterplansforstreamlininginfrastructure;gatekeeping;increasedservicedeliverycapacityandqualityatlowerlevelsofcare;appropriateblendinganduseofpaymentmethods;improvedmanagementcapacitywithappropriateincentives.
Refusalofdonationsofnewtechnologywherebudgetswillnotbeabletopaysuppliesandmaintenance;improveddonorpractices;improveprocurementpracticesandcontrolsofcorruption.
3. Healthservices:a) Unnecessarytests,
procedures,orunderutilizationofthesecomparedtoneed;
b) medicalerrorsandlowqualitycare.
Poormanagement&control(perhapslinkedtoinsufficientmanagementresourcesorinadequateinformationonpatterns);financialincentivestoover-service;defensivemedicine.
Inadequateproviderknowledge;insufficientdatacollectionoruseofdatabymanagers;noincentivesforquality;poorinfrastructure;lowqualityincludinghygiene;poorcompliancewithinfection
Improvemanagementandavailabilityanduseofdata;clinicalguidelines;financialincentivestopreventoveruseandtopromotequality.
Continuoustrainingforproviders;improveddataavailabilityanduse(e.g.clinicalaudits);clinicalguidelines;incentivesforquality(contracting,providermonitoring,paymentsystems,
36
preventionandcontrolstandards.
complaintsystems);investmentininfrastructure,enforcementofminimumstandardsofservicequalityincludingmandatoryinspectionsandclosedownofunsafefacilities;accreditation.
4. Personnel:a) Inappropriatemix
betweendifferentcadres;
b) Locatedinthewrong
places;
c) Demotivatedworkerswithlowproductivityandpoorqualityofservices(e.g.lowvisitsperhealthworkersperday,absenteeism).
Poorplanning;inappropriatetrainingintakes(canbelinkedtostudentdemand);outmigrationorlackofretentionofsomecadres;resistancebyvariouscadrestolessskilledpeopletakingmoreresponsibility.
Incentivesinsufficientforsomelocations;
Poorwagesandincentivestructures;poormanagementandsupervision;poorworkingconditions;recruitmentandpromotionnotbasedonmerit.
Healthworkforceplanningbasedonlabourmarketassessmentandlinkstotrainingintakes;HRHtrainingandrecruitmentalignedwithbroaderhealthsystem’sobjectives;strategiesandincentivestorecruitandretainkeyhealthworkersinremoteandunderservedareas;skill-taskmatching,includingtaskshifting;
Revisesalarystructuresandincentivesforunderservedlocations.
Salaryandincentivestructuresinlinewithsystemobjectives;regulationofdualpractice;improvedmanagement,supervisionandworkingconditions;multi-disciplinaryteams,eliminate“cronyism”inhiringandpromotion(establishclearprocessforhiring,deploying,andpromotionofhealthpersonal).
5. Inappropriatemixofinputs:e.g.healthworkersbutnomedicinesorothermedicalproducts.
Poormanagementorbudgetingpractices;inflexiblecontractswithworkers.
Improvedmanagementofinputsforservicedelivery;budgetpracticesprovidinggreaterflexibilityofuseofinputs.
37
7. Waste(includingexpiredmedicines),corruption,fraud.Lowbudgetexecutionratesareaformofwastebecausetheavailablefundsarenotused.
Poorprocurement,inventorycontrolpractices;poorstorageandtransport;lackoftransparencyandaccountabilitywithappropriatemanagement,auditandlegislation;inadequatesupervision;poormanagementorinflexiblelineitembudgetsleadingtolowbudgetexecution(canalsorelatedtodelayeddisbursementfromministriesoffinance).
Improvedprocurement,supplychainmanagement,inventorycontrol,storageofsuppliesofdrugs;improvedregulationandgovernancewithsanctionsforcorruptionandfraud;codesofconduct;improvedpublicfinancialmanagementincludingbudgetpracticesandresourcetrackingandaccountability.
8. Inefficiencyinraisingrevenuesforhealth,particularlywhenrevenueraisingforhealthisindependentfromgeneralgovernmentrevenuecollection.
Efficiencyofraisinggeneralgovernmentrevenuesisbeyondthehealthsector.Inefficiencyincollectingchargesleviedbyministriesofhealthorhealthinsurancepremiums–poormanagement,poorinformationsystems,lackofmotivationofstafftocollectrevenues,inabilitytoenforcepaymentofcontributions,inabilitytoidentifytheindigentwhodonotpay,corruption.
Feeandpremiumsystemsenforceableatlowcost,improvedskills,managementandinformationsystemsfortrackingpaymentsandserviceuseparticularlyinnationalhealthinsurancesystems;organizational/staffincentivestocollectfees/premiums;user-friendlywaysforpeopletopaycontributions;methodsofidentifyingtheindigent;corruptioncontrol.
9. Fragmentationinthesystem:inpooling,butinthebroaderhealthsystemaswell-e.g.procurement,supplychains,laboratories,servicedelivery.
Donorpractices(developingsystemsfor:channellingandtrackingfunds;procurementanddistribution;employment;servicedelivery;monitoringandevaluation)outsidegovernmentstructures;powerstructuresintheministryofhealth;responsibilitiesinafederalsystem(e.g.centralgovernmentresponsibleforhospitals,lowerlevelgovernmentforotherservices);pressurefromthealreadyinsuredtomaintaintheirbenefitswhenhealthinsurancefortheinformalsectorisdeveloped.
Organizationalorvirtualintegrationasappropriate–e.g.standardizedinformation,budgetingandaccountingsystems,planning,M&Eacrossfragmentedunits.Riskadjustmentacrosspools.
38
10. Administrativeinefficiency:higherthannecessarycostsfortheservicesoffered,includinginhealthinsuranceagencies.
inflexiblestaffcontracts;lackofincentivesforefficiency;inadequatetrainingandknowledgeofmanagersandstaff;highstaffturnover.
contractsystems;incentivesforefficiency;improvedknowledgeandcapacitiesofstaff;policiesandincentivestoretainstaffwherehighstaffturnover
MostoftheoptionsconsideredinTable3involvemakingchangestoexistingstrategiesthat
eitherobtainmorefortheexistingresourcesorensurethesameoutcomesforfewerresources.TheyrepresentmovementtowardsthefrontieridentifiedinFigure1.Thereareothertypesofinterventionsthathavethepotentialtoshiftthefrontierbyexpandingthetechnicaloptionsforimprovinghealthandfinancialprotection.SomeexamplesareprovidedinBox2,althoughtheremainderofthissectionfocusesontheoptionslistedinTable3.
Box2:ShiftingtheFrontier:InnovationsinHealth
Alongsideeffortstodeliverhealthservices,productdevelopersandpractitionersareadvancinginnovationstoensurethatpeopleinlow-resourcesettingshaveaccesstonewapproachesforlife-savinginterventionsthatvastlyexpandcurrentoptions.Manyoftheseinnovativetechnologies,systems,andservicesseektoprovideaffordablesolutionsspecificallydesignedtoaddresstheneedsofvulnerablepopulationsaroundtheworld.Onenecessityfortheprovisionofqualityhealthcareisdatamanagement.Digitalhealthinformationsystemsbreakdownbarriersthatpreventtechnologiesandsystemsfromscaling,andenablesupportforplatformsthatcanbereused,adapted,andbuiltupon.Forexample,anationalelectronicimmunizationregistrybeingdevelopedinTanzaniaandZambiabytheBetterImmunizationDataInitiativeautomaticallysendsimmunizationinformationtohealthcareworkersinadvanceofvaccinationdayswithinformationregardingthenumberofchildrendueforvaccines,whichimmunizationstheyneed,andthevolumeofvaccinestockorsuppliestheclinicshouldhaveonhand.Thissystemminimizesmissedvaccines.Vaccinesareamongthemostpowerfullifesavingtoolsforchildrenunderfive.Withaccessibilitytobetterdata,healthworkersareequippedtomakebetterdecisionsaboutvaccinedeliveryandachievehigherimmunizationcoveragerates.Similarly,redesigningbiomedicaltechnologiesforuseinlow-resourcesettingswillacceleratelivessavedandaverthealthcarecosts.Forexample,oneinnovationunderdevelopmentisbetterrespiratoryratemonitorsandportablepulseoximeterstoimprovetimelydetectionandtreatmentofpneumoniaamongchildrenunderfive.Difficultiesindiagnosingpneumoniaamongyoungchildreninlow-resourcesettingsoftenleadtounnecessarytreatmentdelaysandincreasedriskofdeath.Innovationsindiagnostictechnologiescanhaveasignificantimpactwhenbettermonitoringleadstoexpandedcoverage.ArecentstudyconductedbyPATH,Harnessingthepowerofinnovationtosavemothersandchildren,modelledtheimpactofinnovativepneumonia-detectiontechnologieswhicharemorereliableandeasiertousethanexistingtools,estimatingthatjustoveronemillionlivescouldbesaved(PATH2016).18
18PATH.IC2030:Harnessingthepowerofinnovationtosavemothersandchildren.2016.http://www.path.org/publications/detail.php?i=2647;ModellingwasappliedtowardincreasedinterventioncoveragewithinthecontextofUSAID'sActingontheCallmodel.
39
Innovativeapproachesareevenmoreimportantwherenotoolcurrentlyexiststoexpandcoverageforexistingproducts.ExtendingcoveragefortheHepatitisBvaccinewouldsaveupto6millionlives,forthehaemophilusinfluenzatypebvaccinewouldsaveupto1.7millionlives,forthepneumococcusvaccinewouldsaveupto1.8millionlives,fortherotavirusvaccinewouldsave900,000lives,andforthehumanpapillomavirusvaccinewouldsave500,000livesoverthenext10years(WHO2013).19Additionalbenefitsarelongerandmoreproductivelives,higherearnings,andavertedhealthcarecosts,withapotentialofupto$44returnacrossthelifespanofanimmunizedchildforevery$1invested(Ozawaetal.2016).20PrioritizinginnovationwithinaUHCframeworkwillenablecountriestotakeadvantageofarichsetofemergingtoolsthatcanenabledramaticchange,includingreducedmortalityandmorbidity,inclusiveandsustainablegrowththroughcosteffectiveness,andimpressivegainsinhealthoutcomes(LancetCommissiononInvestinginHealth2013).21Rapidadoptionandscaleofhealthinnovationsiscriticaltofosterthesegains,asisafocusedstrategyandfurtherpoliticalcommitment(Atun2012).22Throughinnovation,UHCcanbedeliveredwiththegreatesthealthvalueformoney,providingpeopletheopportunitytoleadhealthier,moreproductivelives.Source:PATH
AnimportantobservationfromTable3isthatonlyaminorityofthesolutionsaretotallywithintheremitofhealthfinancingwithitsfunctionsofrevenuegeneration,poolingandpurchasing/provision.BecausethisForumisspecificallyonhealthfinancing,webeginwithwaysthatthehealthfinancingfunctioncancontributetoimprovingefficiency.Itisnotintendedtodescribeallthedetailsofhowtoimplementthevarioushealthfinancingstrategiesrelatedtoefficiency,buttooutlinethebroadareaswherethereisagreementthatthesolutionswork.Oncecountrieschoosethetypeofstrategytheyareinterestedinimplementing,technicalworkwouldthenneedtobeundertakentodesignthespecificdetails,buildingonaverylargeliterature.
Afterthehealthfinancingoptions,thepaperconsidersbrieflysomeoftheadditionalstrategies,onthegroundsthatifacountryisseriousaboutreducinginefficienciesitwillneedtouseamixofmeasuresthatinclude,butarenotexclusivetohealthfinancing.
Onepartoftheefficiencyofrevenuegenerationisthecostofenforcementandadministration,togovernment,householdsandfirms,sometimescalledadministrativeefficiency.Asecondpartistheyieldgeneratedfromatax,orproductionefficiency.Thethirdpartisthecostsoftheeconomyofchangesinbehaviourandeconomicoutputresultingfromaparticulartax–forexample,peopleworkinglessbecauseofhighmarginaltaxratesontheirincome,typicallycalleddeadweightlosses(e.g.Okafor2012;Thompson,BeattyandThompson2012;Scott2014).Financedepartments,sometimesworkingwithexternalagenciesliketheIMF,theWorldBankandRegionalDevelopmentBanks,constantlyseektoreducethecostsofenforcementandadministrationandincreaseyields.Generalizationsarethatcorruptionincreasesadministrativecosts,andincomebased19WorldHealthOrganization.GlobalVaccineActionPlan.SecretariatAnnualReport.Geneva,2013.20Ozawa,S.etal.ReturnOnInvestmentFromChildhoodImmunizationInLow-AndMiddle-IncomeCountries,2011–20.HealthAffairs.2016,35:2199-207.21LancetCommissiononInvestinginHealth.Globalhealth2035:aworldconvergingwithinageneration.Lancet.2013;382:1898–1955.22RAtun.Healthsystems,systemsthinkingandinnovation.HealthPolicyandPlanning.2012;27(suppl4):iv4-iv8.
40
taxesaredifficulttocollectwhenthereisalargeinformalsector(e.g.Imam&Jacobs2014).Indirecttaxesyieldmorethanincometaxesonindividualsinthesesettings.Thresholdsatlevelsthatexcludesmallenterprisesavoidhighcostsofenforcementrelativetoyieldsandimproveequity.Varioustypesofreformscanincreaseyieldoradministrativeefficiencyincluding,forexample,theestablishmentofsemi-autonomousrevenuecollectionagenciesatnationaland/orsub-nationallevels(VonHaldenwang,VonSchillerandGarcia2014).
Thisisrarelysomethingthatisconsideredbythehealthsectorexceptwhenthecollectionofhealthinsurancepremiumsoruser-chargesispartofitsmandate.Drawingfromthepublicfinanceliterature,itisdifficulttousetheincometaxsystemtocollecthealthinsurancepremiumswhenthereisalargeinformalsector,soothermethodsneedtobedevised.Variousinnovationsfrombroadertaxationcouldalsobeuseful,includinglocatingagenciesthatcollectpremiumsclosertothepeoplethatwillpaythem,orlicencingfacilitiesthatarewelldistributedinthecountrytocollectthem.However,theexperienceinThailandsuggeststhattheadministrativecostsofcollectionoftheoriginal30Bahtco-paymentfromtheinformalsectorwerehighandoutweighedtheyield,somethingthatcanapplyalsotocollectionofpremiumsfromtheinformalsector(Limwattanonetal.2011).Forthisaswellasequityreasons,thereisabroadconsensusthathealthinsurancecoverageofbasicservicesshouldbefinancedfromgeneralrevenue,whilecontributionsareaconditionforeligibilitytoawidersetofhealthservices.
Thegeneralconsensusonpoolingisthatsmallpoolsareinefficientintermsofhighadministrativecosts,andtheycanresultininequitiesifonepartofthepopulationiscoveredwithmoreandbetterservicesandhighfinancialprotectionthanothers(e.g.Raoetal.2014;Mengetal.2015;Kutzin,YipandCashin2016).Itis,however,sometimesdifficulttomergeexistingpoolsforpoliticalreasons,andcountriesmovingtowardshealthinsurancefromscratcharebestadvisedtoavoidcreatingdifferentpoolscoveringdifferentpopulationgroupswithdifferentbenefits(Hanvoravongchai&Hsiao2007;Knauletal.2012).
ThequestionofhowtopurchaseinputsandservicesisimportanttomanyoftheefficiencyproblemsinTable3.Therearefourcomponents.Thefirstinvolvesachoiceoftheappropriatemixofpersonalhealthinterventionsavailableintherightplaces,themixbetweengovernance,publichealthandpersonalhealthservices,andtheappropriatemixofinter-sectoralandmulti-sectoralactionsinsupportofhealthsectorstrategies.Possiblemethodstoassesstheefficiencyofthedifferentoptions,andtheirweaknesses,werediscussedearlierbutmanycountriesareintheprocessofassessingtheappropriatemixofservicestoguaranteetoeveryoneandwheretheyshouldbeprovided.Redirectingresourcesfromunnecessaryandharmfuloverusedservicestolow-costhigh-impactservicesthatareunderusedaddressestwosourcesofinefficiencyatthesametime–over-andunder-use–andnegativelists,orlowvaluelistsofinterventionsthatshouldnotbeusedorwillnotbecoveredbypayersareoneofthepossibleoptionsofdoingthis(Elshaugetal.2017).
Reducingtheproportionofpatientswhobypasslowerlevelservicestousemorecostlyservicesatsecondaryortertiarylevelsbecauseoftheirperceivedbetterqualityisanotherpartofensuringthattheneededservicesareprovidedintherightplace.Thisrequiresnotonlyhavingsomeformofgatekeeping,butalsostrengtheningthequalityofprimarycareservices.Box3outlinessomerecentdevelopmentsonsupporttocountriesseekingtodeveloptheirprimaryhealthcare.
41
Box3:SupportingPrimaryHealthCare
ThePrimaryHealthCarePerformanceInitiative(PHCPI)isaglobalpartnershiptohelpcountriesbuildhigh-performingprimaryhealthcaresystemsinlow-andmiddle-incomecountries.ThePHCPIpartners—theBill&MelindaGatesFoundation,theWorldBank,theWorldHealthOrganization,AriadneLabs,andResultsforDevelopment—worktosupportcountriestoachievethehealthSDGsthrough collecting better primary health care data, unearthing and sharing best practices, anddeployingdataandevidencetomakeimprovementsinthequality,effectivenessandefficiencyofprimarycareservices.High-performingprimaryhealthcaresystemsarecriticaltoensuringthattherightthingsaredoneintherightsettings:
× Primarycare isthefront lineofhealth–closetopeopleanddeliveringessentialhealthserviceslikevaccinations,maternalandnewborncare,andfamilyplanning;
× Good primary care helps patients manage chronic diseases, avoiding unnecessaryhospitalizationsandcareandtimeawayfromfamilyandwork;
× Primaryhealthworkersformtheearlywarningsystemfordetectingandstoppingdiseaseoutbreaksbeforetheybecomedeadlyepidemics–thisisthefirstlineofdefenseagainstepidemics;
× Whentherightthingsaredoneintherightway,primaryhealthcaresystemscancoverthelargemajorityofhealthneedsinwaysthatareresponsive,safe,ofqualityandtrustedbythepeopletheyserve.
High-performingprimaryhealthcaresystemsarealsothebackboneofefficienthealthsystems.Inbothhighandlowincomesettings, ithasbeenshownthatstrongprimaryhealthcarepreventsmanyillnessesandcatchesothersearlywitheffectivelow-costtreatment,therebykeepingpeopleoutofhospitalandreducingsubsequenttreatmentcosts(e.g.Kruketal.2010;Kringosetal.2013).Effectiveservicedeliverymeansthatpatientsreceivetherightpreventivecareortreatment,attherighttime,intherightplace,andwithrespect.DoingthisrequiresattentiontofivePHCsystemsfeatures(adaptedfromStarfield1992):
× First-contactaccess:Primaryhealthcaresystemsshouldserveastheentrypointintothehealthcaresystem,wherepeoplecanaccessaffordablecareformosthealthneeds.
× Comprehensiveness: Primary health care systems should deliver a broad spectrum ofpreventative, promotive, curative and palliative care – for example, throughmultidisciplinary teams that contain health professionals with varied, complementaryskills.
× Continuity: Primary health care systems should support long-term patient-providerrelationships–allowingproviderstocareforpatientsateverystageoflife.
× Coordination: Primary health care systems should coordinate an individual’s journeythroughcomplexhealthsystems.
× Person-centeredness:Primaryhealthcaresystemsshouldbeorientedaroundtheneedsofpeopleandcommunities.
ThePHCPIrepositoryofpromisingpracticesexplainshowleadinglowandmiddleincomecountriesputthesefivefeaturesintopracticetoimprovethequalityandefficiencyoftheirprimaryhealthcaresystems:http://phcperformanceinitiative.org/tools/promising-practices
42
Thesecondcomponentinvolvesdevelopwaysofpayingforthechosenservices(Honda2014;Maedaetal.2014;Tangcharoensathienetal.2015;Xu,ChengandColón-Ramos2015;Bastanietal.2016;Kutzin,YipandCashin2016).23Providerpaymenthasbeenshowntobeapowerfultooltoinfluencethebehaviourofproviders,andinturntheprovisionofcare.Ingeneral,theevidenceshowsthatnosingleproviderpaymentmethoddeliversvalueformoneyinallsettings.Forinstance,capitationencouragesefficiencybutmayleadtounder-provisionofservicesandincreasedreferrals.Ontheotherhand,fee-for-servicecanleadtooverprovisionofservicesandcanleadtocostescalation.Therehasbeenalsobeengrowinginterestinperformance-basedfinancinginhealthinrecentyears;althoughthemorerecentevidenceonqualityofcareandefficiencyisscattered–seethenextsectionofthispaperonwhatwedonotknow.
Countriesshoulddecidewhicharethemostappropriateblendsofpaymentmethodsusewithineachcaresetting,andhowthesepaymentmethodblendsalignacrosscaresettingstoachieveitsownpolicyobjectiveswithregardstoprovisionofcare,costandquality,(Cashinetal.,2009).“Blending”differentpaymentmethodswithincaresettingsisdonetoencouragecertaindesiredoutcomesaswellastomitigatethenegativeincentivesofindividualpaymentmethods(BelliandHammer1999;DranoveandSatterthwaite2000).Forexample,inprimarycare,capitationandFFSpaymentsareoftencombined.TheFFSpaymentsareusedprimarilytoencouragetheprovisionofcertainpriorityservices(e.g.vaccinations),servicesandprocedureswhichrequirecostlysupplies(e.g.injectablemedicines),aswellasserviceswhichlieontheborderbetweenprimaryandspecialistcare(e.g.woundcare,drainageofabscesses,removalofbenignlesions)andthustendtobereferred(Robinson2001).Incombinationwithcapitationpayments,primarycareprovidersarestillencouragedtolimitthevolumeofservicesprovidedtoachievecostsavings(LangenbrunnerandWiley2002;LangenbrunnerandSomanathan,2011).Similarly,theblendsofpaymentmethodsacrosscaresettingsshouldbetakenintoaccounttoensurethatincentivesarealignedwithhealthsystemobjectives(e.g.encouragingbettermanagementofpatientsattheprimarycarelevel,reducingincentivestoincreasevolumesinoutpatientspecialistandacuteinpatientcare,increasingincentivesforcoordinationwithprimarycareafterdischarge,etc.).
Thethirdcomponentinvolvesstrengtheningcontractingandprovidermonitoringcapacitiesofthepurchaserinordertoenforcetheintendedbehaviourchangesofnewpaymentmethods(Cashinetal.,2009).Providercontractscanhelpcontributetohealthsystemobjectivesbytakingadvantageofprovisionsrangingfromethicalcodesanddatadisclosurerequirementstocostandvolumecapsandrisksharingmechanismstoreducetransactioncostsandshapeproviderbehaviors.Inaddition,purchaserscanestablishandusetransparentcriteriafromwhomtocontract,forexample,accreditationand/orminimumvolumerequirementstofosterqualityofcare.Providermonitoringenablescontinuousqualityimprovementbyhighlightingareasofcaredeliverythatneedoptimization,andbysettingcommontargetsfortheseareas.Whenemployingpaymentmethodsthatencouragecostreduction(e.g.capitation,bundledpayments,etc.),useofqualitymonitoring,inadditiontoriskadjustment,isessentialtoensurethatprovidersarenotcuttingcostsinwaysthatjeopardizepatienthealth.Providermonitoringcanincludeactivitiessuchasclinicalaudits,regularcollectionofdataonspecificqualityindicators(e.g.adherencetoclinicalguidelines,adverseevents,patientsatisfaction).Basedonthemonitoringresults,purchaserscanholdprovidersaccountableaccordinglybyrewardinghigher-levelperformersand/orsanctioninglower-levelperformers.
Finally,thefourthcomponentinvolvesempoweringpatientstoholdpurchasersandprovidersaccountablethroughformalrepresentationofconsumersinpurchasingorganizations,
23Thisissometimescalledstrategicoractivepurchasing.
43
developmentofpackagesofcarewithformalcoverageguarantees,establishmentofpatient’srightslegislation,chartersorethicalcodesandthedevelopmentofaformalmechanismtoreceiveandrespondtopatientcomplaints(Busseetal.2007).
Improvedpublicfinancialmanagement(PFM)wouldcontributetoreducingmanyoftheformsofinefficiencyidentifiedinTable3.Theseincludereducingthefundsreturnedtothetreasurybecausetheyhavenotbeenusedinthefinancialyear,helpingtocontrolcorruption,fraudandothertypesofwaste,andlinkingpublicexpendituresmorecloselytoresults(Fritz,SweetandVerhoeven2014;Barroy,SparkesandDale2016;WHO2016;Cashinetal.2017).DespitetheconsiderableattentiongiventoassessingPFMsystemsoverthelastdecade,itisnotclearthatreformshavebeenuniversallysuccessful(Hepworth2015).Thiscanbelinkedtothepoliticaleconomyofreforms,somethingthatistakenupinthenextsection.
WeusemedicinestoillustratesomeoftheotherstrategiesfromTable3thatcancomplementhealthfinancingchangestoachievemorewiththeavailableresources.Manystrategieshavebeenshowntoincreaseefficiencyinavarietyofsettingsrangingfromestablishingandenforcinganessentialmedicineslistwhichusesgenericstothemaximumextent,increasingthetransparencyandinformationavailableformedicineprocurementsothatcountriespaytherightpriceandcorruptioniseliminated,developingthecapacitytomonitormedicinesafetyandqualitynotjustwhenmedicinesarelicencedorregisteredbutoncetheyenterthedistributionsystem,eliminatingincentivesforover-prescriptionsuchasseparatingprescribingfromsales,andvarioustypesofstrategiestoimproveproviderandpatientknowledgeaboutgenerics,appropriateandtimelyprescriptions,andtheimportanceofadherence(e.g.Holloway2011;Hollowayetal.2013;Atavetal.2014;Chenetal.2014;Hassalietal.2014;Hurley2014;Choudhryetal.2016).Itis,however,clearthatitissimplertoimplementthesepoliciesinthepublicthanintheprivatesectorwhereregulationssometimesdonotexistandwheretheydo,enforcementisdifficult(e.g.VanNguyenetal.2013;Sheikh&Uplekar2016).
ThereissimplytoomuchliteraturedescribingattemptstoimprovetheefficiencyintheothercomponentsofhealthsystemstoreportitallherebutarecentWHOpublicationsuggeststhat,inadditiontoconsideringhealthfinancing,itisusefultoworkthroughthehealthsystemfunctionsofgenerationofhumanandphysicalinputs,governance/stewardshipandservicedelivery(Sparkes,DuránandKutzin,J.2017).Theythensuggestlookingforinefficienciescommontoeachfunctionandinefficienciesthatpreventtheintegrationofactivitiesacrossfunctions.WeturnnowtodiscussingsomeoftheareasonwhichthereisnoconsensusandwherefurtherevidenceanddiscussionatthisForumwouldbevaluable.
2) WhatwedonotknowSomerecentdevelopmentsthatmayimproveefficiencyinhealtharestillintheirinfancy.Forexample,thequestionofwhetherelectronicmedicalrecordsimproveefficiency,andifsoinwhatcircumstances,isstillbeingexplored(e.g.Nguyen,BellucciandNguyen2014;Campanellaetal.2015).Thistypeofquestioninnotdiscussedherewherethefocusisonareasofdisagreementorwherethereissimplynotenoughinformationonlongstandingquestionstomakeinformeddecisions.
i. PayingforresultsManyexperimentshavebeenundertakenwithformsofresults-basedpaymentsthat
supplementincomesforstafforinstitutionsinreturnforasetofagreedoutputs(e.g.Rudasingwa,SoetersandBossuyt2015;Das,GopalanandChandramohan,2016;Spisaketal.2016).Similarly,
44
therehavebeenrecentexperimentswithformsofvalue-basedpayment24asareplacementforfee-for-serviceintheUSwiththebroadergoalofimprovingcontinuityofcareandprovidercoordinationaswellasreducingthegrowthofcosts(e.g.Conradetal.2014;Dambergetal.2014;Carey2015;Press,RajkumarandConway2016).Analternativeiscalledpopulation-basedpayment.Bundledandpopulation-basedpaymentsarerelativelynew,butthepreliminaryresultsarepromising,althoughtheyrequiretheabilitytodesignandimplementariskadjustmentsystemandtomonitorresults.AbriefsummaryisprovidedinBox4.
Box4:NewOptionsforStrengtheningProviderPaymentMechanismsforCareIntegration
Twonewpaymentmechanismsarecurrentlybeingpiloted,mostlyinhigh-incomecountries,toimproveintegrationofcare.Theyare(i)bundledpaymentsforacuteepisodesofcareandchronicconditionsand(ii)population-basedpayments(PBPs)coveringspecificservicesfordefinedpopulationgroups.Thesepaymentmethodsspanacrosscaresettingstoincreaseincentivesforprovidercoordination.
Bundledpaymentsforacuteepisodesandchronicconditions.Bundledpaymentsinvolveasingle,fixedratepaidtotwoormoreproviderstocoverallservicesdeliveredfor:(i)treatmentofanacuteepisodeofcare,or(ii)managementofthecareforpatientswithaspecificchronicconditionordisease(AmericanMedicalAssociation,2016).BundledpaymentsforacuteepisodesofcarehavebeenpilotedintheUSandEuropesincethemid-2000s.ExamplesincludeBestPracticeTariffs(BPTs)intheUK,coveringadmissionsforhipfracture,stroke,cholecystectomyandcataractsurgery,andOrthoChoicebundledpaymentsinSweden,coveringorthopedicproceduresincludinghip,kneeandspinesurgery(Srivastaetal.2016).Thesepaymentstypicallycoverthecostsofallinpatientandoutpatientservicesfromtheinitialvisitthroughtreatment,recoveryandrehabilitation,includingpost-dischargecareandanycomplicationsthatmayresultwithinacertaintimeperiodafterdischarge(PorterandKaplan2015).Bundledpaymentsforchronicconditionshavebeenimplementedtohelpimprovecoordinationofcareforthesepatientsandaimtoencourageaholistic,long-termperspectiveratherthanone-offencountersorinterventions(Srivastaetal.2016).PilotsofthesebundledpaymentsinEuropeancountrieshavefocusedonbothrare,yethigh-costdiseases,andonmorecommonchronicconditions.Thebundlestypicallycoverallservicesrelatedtothemanagementofthediseaseorcondition,inlinewithclinicalguidelinesandpathways,andcanspanmorethanonecaresetting.Forexample,undertheDutchbundledpaymentscheme,servicesincludedregularprimarycarecheck-ups,additionalconsultations,imaging,labtests,examinations(e.g.footexaminations),counselling,medications,psychosocialcare,andcoordinationofspecialistservices.Thecostsofcomplicationsfromthesediseasesaretypicallycoveredoutsideofthebundle(PorterandKaplan2015).Thepaymentistypicallytime-based(permonthoryear)since,unlikeacutecareepisodes,thecycleofcareforachronicconditioncoveredbyabundledpaymentmaycontinueindefinitely.Bundledpaymentsforacutecareepisodeshavebeenshowntoachievesignificantcostssavings,withnoorlittledeteriorationofquality.Theimpactsofbundledpaymentsonchronicconditionsarelessclear.InPortugalcostsfortreatingHIV/AIDSdecreasedwhilequalityofcarewasmaintained(e.g.,patientadherencetomedication,controlledinfectionlevels,complianceofproviderswithtreatmentguidelines,etc.).However,whilequalityimprovementswereobserved
24Value-basedpurchasing“referstoabroadsetofperformance-basedpaymentstrategiesthatlinkfinancialincentivestohealthcareproviders'performanceonasetofdefinedmeasuresinanefforttoachievebettervalue”(Dambergetal.2014).
45
forbothdiabetescareprocessandpatientoutcomeindicatorsintheNetherlands,costsincreased.Reasonsforthisincreaseareunclear,butmaybeduetodelayingtheuseofspecialistcare,whichcouldhaveresultedinmorecostlycare.
Population-based-payments.Population-basedpayment(PBP)modelsinvolvesettingaprospectivebenchmarkbudget,whichformsthebasisforpaymentstogroupsornetworksofprovidersfortheprovisionofallorthevastmajorityofservicesforadefinedpopulation.ThesemodelshavebeencloselyrelatedtotheemergenceofAccountableCareOrganizations(ACO)-networksofhealthcareprovidersthatarecollectivelyaccountablefortheorganization,costsandqualityofhealthcarefortheirmembers–intheUnitedStatesandelsewhere.Similartobundledpayments,providersarepermittedtokeepatleastportionofthesavingsgeneratedbelowthebenchmarkbudget(contingentonachievingspecifiedqualitytargets)andmayberesponsibleforanycostsexceedingthetotalPBPamount.BecausePBPscoverarangeofservicesacrossprovidersandarenotlinkedtospecificcareepisodesorconditions,theyarethoughttopromotegreaterintegrationofcare,amoreholisticviewofpopulationwell-beingaswellasincentivesforinnovationstokeepcostsdown(e.g.,riskstratifiedcasemanagement,dischargeplanning,preventiveactivities,etc.).
ThelargestACOpilotsareintheUS,partofbroaderreformsmandatedbytheAffordableCareActof2010.Forexample,Medicarehascontractswithover400ACOs.ProvidersformingtheACOtypicallyincludeprimarycareprovidersandhospitals,butcanalsoincludespecialists,long-termcarefacilitiesandhomecare(Srivastaetal.2016).InEurope,examplesofPBPmodelsforACOshavebeenimplementedinGermanyandSpain.Therangeofhealthservicesprovidersarefinanciallyresponsibleforvariesacrosstheavailablemodels.IntheMedicaremodel,ACOsarefinanciallyresponsiblecostswhichincludeinpatienthospitalcare,skillednursingcare,hospiceandhomehealthservicesaswellashospitaloutpatientcareanddoctors`services.InGermany,providersareresponsibleforallheathcarecostsfortheinsuredpopulationwiththeexceptionoflong-termcare(Srivastaetal.2016).ThesizeofthepopulationthatisassignedtoanACOalsovarieswidely,rangingfrom5000to245,000patients.PBPimplementationisrelativelyrecentandanyimpactsobservedonqualityandcostshouldbeconsideredpreliminary.AsubsetoftheMedicareACOs,knownasPioneerACOs,wereabletoachieveimprovementsin28ofthe33requiredqualityindicatorsincludingcontrollinghighbloodpressure,screeningforfuturefallriskandscreeningfortobaccouseandcessation,forexample.Onanaggregatelevel,MedicareACOscontributedtoslowingthegrowthinhealthspending,thoughnotallwereabletoachievecostsavings.In2012,totalnetsavingsforMedicareinamountedto$383millionwhiletheGKmodelachievedsavingsofEUR4.6million(Srivastaetal.2016).Source:WorldBank,Forthcoming
Intermsoftheexperiencewithresultsbasedpaymentsorpaymentforperformanceinlower-incomecountries,therehavebeenmixedresults.Sometimestheyhavethedesiredimpactandsometimestheydonot.Theycanworkforsomeofthedesiredoutcomesbutnotforothers(e.g.Foxetal.2013;Mills2014;Binyarukaetal.2015;Das,GopalanandChandramohan2016).Thereisundoubtedevidencethattheyhavestimulatedstaffmotivation,qualityandefficiencyattimes,andthemoreintensiveinformationsystemstheyrequirehavealsobenefitedthewiderhealthsystem
46
(Cashinetal.,2014).However,systematicquantitativeassessmentormeta-analysisacrosstheexperimentsiscomplicatedbythefactthateachexperimenthasbeendifferentinthetypeofhealthworkerstheytarget,intheoutputsforwhichpaymentistriggered,orinthewaypeoplearepaid,sodatastrictlyshouldnotbepooledwithoutsomewayofcontrollingforthisheterogeneity(MarkovitzandRyan2017).Itisalsonotcleartheextenttowhichtheseexperimentsaresustainable,asmanyofthemhavebeenfundedbydevelopmentpartners,attimesindependentofacountry’sfinancingsystemandonlywithemergingevidenceoftheprogramcosts(DeBruinetal.,2011;Eijkenaar,2013;Cashinetal.,2014).
Furtherworkisneededtobeforeitispossibletobesurewhatpaymentmechanism,ormixofpaymentmechanisms,bestencourageshighstaffmotivation,quality,andefficiency.Tocontributetothis,theWorldBankGroupiscurrentlyintheprocessofsummarizingtheirexperienceswithformsofresults-basedfinancing.
ii. TheprivatesectorandefficiencyTherehasbeenconsiderablerecentinterestintheappropriateroleoftheprivatesectorin
healthandmanyorganizationshavedevelopedstrategiesforengagingwiththeprivatesector(IFC2011;USAID2009;WorldBank2013).Systematicreviewsoftheextensiveliteraturesuggest,however,thatthereisnoevidencethateitherpublicorprivatehealthservicesareinherentlymoreefficientorofhigherqualitythantheother(Coarasaetal.forthcoming;Berendesetal.2011;Basuetal.2012).Thekeyissuesandchallengesinservicedelivery,suchassubstandardpatientsafetyandqualityofcare,excessiverelianceonhospitalsandinadequateinfrastructure,arecommonacrossthepublicandprivatesectors.Moreover,deficientsafetyandqualityhaslesstodowiththeownershipoftheproviderthanwiththeincentivesfacedbytheprovider.Thismeansthatinseekingtoexpandservicecoverage,thereisnoreasontoarguethatmorepublic,ormoreprivate,isthepreferredoption.
Opportunitiesdoexisttoengagethepublicsectoratbothhospitalandprimarycarelevelsandexamplescanbefoundofpublic-privatepartnershipsimprovingcoverageandqualitywithhospitals.Attheprimarycarelevel,engagementwiththeprivatesectorhasincludedcontracting,providernetworking,implementationofvoucherschemes,andinclusionofinformalprovidersintheformaldeliverysystem(includingprogrammesaimingatretrainingandformalizingtheirstatus)(Montagu&Goodman2016).However,whilethereseemstobeawideconsensusthatmoreeffortsshouldbeundertakentoestablishadialogueandrelationshipbetweengovernmentandprivateproviders,theredoesnotyetseemtobeagreementonhowbesttodoit.
Thereismoreagreementthatsomeinnovationsintheprivatesectorhavebenefitedthepublicsector.Theseincludeinnovationsin:businessprocessfunctionssuchasmarketing,financing,andoperating;promotionofhealthservicestothepoorthroughsocialmarketingandservicedesign;redesigningcoststructuresthatallowproductsandservicestobemoreaffordableforthepoor,byloweringoperatingcoststhroughsimplifyingmedicalservices,loweringunitcoststhroughhighervolumesandcross-subsidization;andnewoperatingstrategiesthatincreasetheavailabilityofservicesinremoteareas,mainlyachievedthroughoptimizinghumanresources,processandproductreengineering,andincreasingoutreachactivities(Bhattacharyya,etal.,2010).
iii. HumanresourcestrategiesforefficiencyConsiderableworkhasbeendoneonhumanresourcestrategiesinadditiontohowtopay
providers.Thisincludesdevelopingstrategiesandincentivesfor:continuoustrainingorretainingwherenecessary;retainingstaff;ensuringtherearesufficientstaffwiththerightskillsinisolatedanddisadvantagedareas;andtryingtocontrolharmfuleffectsofdualpractice(e.g.Asanteetal.
47
2014;Rawaletal.2015;Araujo,EvansandMaeda2016;GwynneandLincoln2016;Yazbeck,RabieandPande2017).Thesestrategiesareoftenbasedonresearchonthefactorsthatmotivateordemotivatehealthworkersindifferentpartsoftheworld(e.g.Bonenbergeretal.2014;Hotchkiss,BanteyergaandTharaney2015;VanYperen,WörtlerandDeJonge2016;Wurie,SamaiandWitter,2016).
Salariesandfinancialincentivesarenottheonlyissue,buttheyareimportantinmostsettings.Inlowincomecountries,thekeyissueishowtoaddressthewiderangeofissuesthatwouldreducethehealthworkershortage,improveskillmixesandensuremotivatedworkersareatworkandlocatedwheretheyareneededfortheavailableresources.WhereMinistriesofHealth(andEducation)arehavelimitedresources,whereshouldtheystart?Thisrequiresanassessmentofwherethebiggestimpactwouldbeobtainedforgivenlevelsofexpenditure,butthisinformationisnoteasilyavailable–usingdualpracticeasanexample,therehavebeenmanyattemptstocontroltheproblemsassociatedwithpublicsectorhealthprovidersworkingalsointheprivatesector,buttherearefewgeneralizablelessonsofwhatstrategieshavehadsustainablesuccess(e.g.SandierandPolton2004).
iv. CostsofimprovingefficiencyManystudieshaveevaluatedtheimpactofattemptstoimproveefficiencyinthehealth
system,somereportedhereandsomeintheAnnextothisdocument.Rarelyarethecostsreported.Wheretheyare,informationisrelativelysparse–forexample,Conradetal.(2014)reportedthattherearesubstantialtransactioncostsinvolvedinintroducingvalue-basedpaymentsintheUSAincludingthoselinkedtochangesinthecomputerpaymentsystems(Conradetal.2014).Thereisscatteredinformationthattransactioncostsmightbehighforresults-basedfinancinginlowerincomecountries,totheextentthatBorgietal.(2015)questionediftheeffectsjustifythecosts.Allstrategiestoimproveefficiencyinvolvetransactioncosts,andthisinformationiskeyfordecisionmakersseekingtoallocatescarceresourcestointer-sectoralactions,improvethequalityandrangeofservicesavailableandincreasefinancialprotection–andimproveefficiencyatthesametime.
v. PoliticaleconomyissuesThehealthsectoriscomplex,shapedbypowerfulinterestgroupsandmanyinterestsinboth
the public and private sectors interact and collide on a daily basis (Daemmrich 2013). Efficiencyreformschallenge the statusquo in theprovisionofhealth services, and in their financingand/ororganisation,sotheynaturallytriggerbroaderpolitical,economicandethicalconcerns(Robertsetal.2004).Efficiency reformscannot, therefore,beviewedonly froma technical sideand thepoliticaleconomyaroundtheirpossiblesuccessorfailurealsoneedstobeunderstood.PerhapsthisiswhyFoxandReich(2015)arguethatsuccessfulreformsaretheexceptionratherthantheruleinhealth.
Formalmodelsofpoliticaleconomyarenowbeingappliedtohealthreformsmorefrequentlytoshedlightontheseprocesses:forexample,modelsofcompetinginterestgroupsandvotermodels(seeHauck&Smith,2015);politicalsettlementanalysis(Kelsall,Hart&Laws,2016);andstakeholderanalysis(Bumpetal.2014).Reichetal.(2016)arguethatmappingoutthevetopointsandvetoplayersiskeytoanyreforms–vetopointsare‘juncturesinthelegislativeandpolicydesignwherereformscanbeblocked’,whereasvetoplayersare the ‘individualsorcollectiveactorswhoseagreement isrequiredtomakepolicydecisions’.
Severalstudieshaveshowntheimportanceof‘events’(e.g.politicaloreconomiccrises,wars,ornaturaldisasters)intriggeringareformprocess.Thoseeventsarecrucialastheytendtochallengethepowerofinterestgroupsandcollectiveactorswithinthehealthsystemandinsomecases,theyleadtoabreakingpointwherethesystemisnolongerdeemedappropriate.Forinstance,Reichetal.
48
(2016)findthatthecollapseofcommunismledtoapoliticaldiscussiononsocialprotectionsystemsand design of major health reforms promoting the development of primary health care and theimplementation of essential benefit packages. In France, the United Kingdom and Japan, thedevelopment of universal coverage financing system was integrated as part of the post-warreconstructionprocess(Stuckleretal.,2010;Reichetal.,2016).InThailand,the1977AsianfinancialcrisiswasthestimulustodeveloptheUHCSchemethatextendedcoveragewithhealthservicesandfinancialprotectiontotheentirepopulation(Patcharanarumoletal.2011),whilethe2008financialcrisiswasthestimulusforaseriesofreformsincludingthosethatincreasedefficiencyinthehealthsystem(Houetal.2013).
StudiesfromKyrgystanandMexicoshowhowhealthreformbecameapoliticalagendathatobtained the support of the population, enabling health financing changes to be pursued despiteopposition from some interest groups (Kelsall, Hart & Laws 2016; Frenk et al. 2006; Parry andHumphreys2009).Bumpetal.(2014)traceouttheopponentsandthesupportersofrecenthealthfinancingreformsinTurkeyandactivestepsweretakentobuildsupportandreducetheopponents’supportbasewhileHarris(2015)pointstotheimportanceofagroupofcivilservantsinThailandwhobuiltsupportandcombattedoppositiontotheintroductionoftheiruniversalcoveragescheme.
Beyondthis,thereislittleknowledgeofwhatfactorsallowefficiencyreformstosucceedandfail,andhowgovernmentsstrengthenedsupportandovercameopposition.Itisnotyetclearifthereare generalized lessons from experience that would help countries learn from the experience ofothers.
That being said, it is clear that there can be political obstacles to implementing even thereformsthattechnicallyseemthesimplest.Differenttypesofhospitalreformstoimproveefficiencycan be opposed by managers, clinical staff, non-clinical staff, or the community (e.g. Galetto,MarginsonandSpieser2014).Closinghospitalsorreducinghospitalsbedsinthefaceofover-capacityis likelytobeopposedbypoliticians inwhoseelectoratethehospitalsare located,bytheaffectedcommunity,andbyemployees(e.g.Bloometal.2015).Animportantrequirementforthesuccessfulimplementationofefficiencyreformsistoundertakeaformofpoliticalmappinginthedesignphase,tounderstandwhoislikelytosupportoropposethereform.Strategiesfordealingwithopposition,includingengagementwithkeystakeholders,thenneedtobedevelopedandimplementedbefore,oratthesametime,asthereformsthemselves.
VIII. EfficiencyandequityReducing inefficiency is a means of moving more rapidly towards UHC for the available
resources,therebyimprovingpopulationhealthandfinancialwellbeing.Efficiencyanalysisdoesnottypically account for the distribution of coverage and outcomes across population groups butconsiders aggregate outcomes at the population level. Improving population health and financialprotectionareanimportantgoalsofhealthsystemdevelopment,butreducinginequityisalsocriticalandthereisalargeliteratureonthenatureofinequalitiesandinequitiesinhealthandhowtoimprovethem(e.g.WHO2000;deAndradeetal.2015;Mackenbachetal.2015;Marmot2015).
Thequestionofwhetherthereisanefficiency-equitytrade-offinhealthpolicyhasalsobeenwidelydiscussedinthelasttwodecades:drawingonthisliteraturetheWorldHealthReportof2000stated that ‘equity and efficiency can be easily in conflict’ (WHO, 2000). For instance, someinterventionsdirectedspecificallytovulnerablegroupsmightbemorecostlytoimplementifthosegroups are located in a remote location or present demographic, cultural or socio-economicconstraints.Ebong&Levy(2011)comparedtheefficiencyoffacility-basedandoutreachprogramsinauniversalimmunisationcampaigninCameroon.Theformerwasmorecost-effectivethanthelatter.
49
However,althoughtheoutreachprogramto increasepopulationcoveragewasmorecostly, itwasmoreeffectiveatreachingvulnerablegroups.Inthesamevein,twostudiessuggestshowedthatthemostefficientprogramsforHIV/AIDSinSouthAfricawerenotthemostequitable(Cleary,Mooney&McIntyre2010;Verguet2013).Targetingthepoorandthevulnerablecameatacostintermsoftheoverallpopulationimpact.
On the other hand, some strategies that improve efficiency can also have positivedistributionalimpacts.Strengtheninginvestmentinprimarycareisthemostobviousexample.Primarycareinterventionsareoftenthemostcost-effectivewayofreducingtheburdenofdiseaseinlow-andmiddle- income countries compared to other levels of care, and primary care tends todisproportionallybenefit thepoor (Jamisonetal.,2006;Asanteetal.,2016).ThisdoesnotalwaysmeanthattherichdonotbenefitmorethanthepoorfromPHC,butthattheypoorbenefitrelativelymorefromPHCthanfromhigher-levelservices.
Someevaluationsof trainingcommunityhealthworkersandvarious formsof task shiftinghavealsobeensuggestedtoreducedeliverycosts,increasecoveragelevelsandbenefitthepoorandpeoplelivinginremoteareas.Zachariahetal.(2009),forexample,assessedthreeprogramsoftask-shiftingtoprovideARTandtreatmentmonitoring(inLesotho,SouthAfricaandMalawi),showingareductionofcosts,increasedaccessandimprovementsingeographicandsocio-economicequity.
Benefit-incidenceandfinancial-incidenceanalysishaveincreasinglybeenusedtoassesswhichgroupsbenefitthemostfromaparticularpolicyorfinancingflow–e.g.asubsidyorvoucherschemeoraparticularlevelofcare(e.g.Asanteetal.2014;Chenetal.2015;Asanteetal.2016).Whilethisinformationisinteresting,itisnotparticularlyinformativeaboutidentifyingthemostefficientwaysofreducinginequity.
Anillustrationistherelativelylargeliteratureontargetingparticularvulnerablegroupsversusamoreuniversalapproach.Mostofthestudiesoftargetinghavefocusedontheimpact,andwhetherthepoorbenefitasintended.Sometimesthepoordonotbenefitinwhichcasethereisnoneedoffurther analysis (e.g. Coady, Grosh & Hoddinott, 2004). But where the targeted group benefits,questionsofcostsandefficiencybecomeparamount.
Therearemanycostsinvolvedintargeting-administrationoftargetingschemes,continuousupdatingoftoolsfortheidentificationofthepoor,fraudcontrolandresourcetransfercosts(Coady,Grosh&Hoddinott 2004;Dutrey 2007). These are very rarely reported even in studies that showimpact.Theevidenceonthemeritsofalternativewaystoreduceinequalitiesisanimportantmissingingredienttoevidence-basedpolicydevelopmentinthisarea.
Finally,equityconsiderationshaveinfluencedthedesignofhealthfinancingstrategiesoverthelastdecades.Receivedwisdominhealthinsurancehasbeenthatitworksmostefficientlywheninsuranceisofferedforlowfrequency,highcostoccurrences.Peoplewithchronicillnessandpeoplelivingclosetothepovertylinecansufferfinancialcatastropheorbepushedintopovertyfromhighfrequency,lowcostevents.ProtectingthesepeoplefromseverefinancialhardshiplinkedtopayingoutofpocketforhealthservicesisanequityissueatthefoundationoftheconceptofUHC.
IX. ConclusionsInefficiencycanbe found in thehealthsystemsofall countries. In lowandmiddle income
countries,itimposessubstantialcostsintermsofslowingtherateatwhichtheycanmovetowardsUHCandimprovethehealthoftheirpopulations.Inhigh-incomesettingsitcanmeanthatservicesare cut or out-of-pocket payments are increased unnecessarily at a time of financial constraint.Achievingmorehealthandfinancialprotectionwiththeavailableresourcesis,therefore,animportant
50
complement to efforts to raise thenecessary resources forhealth. Proof that thehealth sector isgettingmoreefficientmayalsopersuadetheMinistryofFinancetoallocatemorefundstohealth.
Somecountryhealthsystemsaremoreefficientthanothers,butnotallsufferfromthesametypesof inefficiency. It isalsopossible tobe relativelyefficient in someareasand lessefficient inothersbuteverycountrycoulddosomethingtoensuretheyachievemorewiththeavailablehealthresources.Afirststepistoidentifythemainsourcesofinefficiency,thereasonswhytheypersist,andwhichonesareamenable to technical solutions thatarepolitically feasible to implement.Politicalanalysis to understand the likely opponents and supporters of any particular reform, and tosubsequently build support and counteract opposition, is a critical step towards maximizing theprobabilityofsuccesswithanyreformsthatareadeveloped.
Someof thetechnicalsolutionsthatcanreduce inefficiencyarewellknownandaccepted.Within the health financing system, for example, the need to reduce fragmentation in pooling bymergingpoolsorbystartingwithasinglelargepoolisone.Soaretheideasofchoosingthemixofhealth services that is the most efficient, delivered in the right places; introducing paymentmechanismsandsystemsthatencouragequality,efficiencyandresultsbasedonreliableuptodateinformation;andoptimizingpublicfinancialmanagementpractices.
Outsidethefinancingsystem,someof thepossiblesolutionsrelatedtomedicinesarewelldocumentedalthoughtherearealsoimportantstepsthathavebeenshowntobeeffectivewithhealthworkersandinfrastructure.Theseincludeshiftingtasksfromdoctorswheretheyareinshortsupplytoothertypesofhealthworkerssoastoincreasecoveragewithoutadditionalcost.
ThesolutionsofTable3wereorganizedaccordingtothreemainpolicyquestionstheyseektoaddress:doingtherightthings;doingthemintherightplaces;anddoingthemright.Governmentswillwanttoknowwhichsolutionsofferthegreatestchancesofanimmediatebenefitandwhichoneswillreaplonger-termbenefits.Table4summarizesthosethatarethemostlikelytodelivershort-termefficiencyreturns.
Table4:PossibleQuickWins
§ Organizationandmanagementofcare- Establishmanagementnetworksfor(primary)careproviders;- EstablishandenforcegatekeepingatPHC-level;- Improveservicedeliverycapacity/qualityatlowerlevelsofcare(e.g.,PHC),including
telemedicine.§ Taxpolicies:- Implementhealthtaxesandfinancialincentivesforpersonalactiononprevention;- Removetaxesonessentialmedicines.
§ PublicFinancialManagement- Modifybudgetpracticesasnecessary–e.g.movefromlineitembudgetstomoreflexible
budgeting;timelyreleaseoffunds;- Improvecontrolstopreventcorruption;- Improveregulationandgovernancewithsanctionsforcorruptionandfraud;- Promotecodesofconduct.
§ Paymentsystems- Developmoreuser-friendlywaysforpeopletopaycontributions.
51
§ HumanResourcesforHealth- Improvehealthworkforceplanningwithlinksittotrainingintakes;- Improvemanagement,supervisionandworkingconditions;- Allowhealthworkersatlowerlevelstotakeonmoreresponsibilityasappropriate(task
shifting).
§ Medicines- Developandenforcegenericspolicyandessentialmedicineslistforhealthfacilitieswith
qualitycontrolwhichmightinclude:- Limitingfinancialincentivesforprescribingbrandedmedicines- Informationongenericstoproviders/populationwithqualitycontrolsystems;- Removeinappropriatefinancialincentives–e.g.separateprescribingfromsales;- Increaseinformationtoprovidersandpatients;regulateandenforcestandardsfor
industrypromotion.- Activepurchasingofmedicineswithappropriatecompetitivebidding;- Increasetransparencyinpurchasesandtenders;- Monitorandpublishmedicineprices.
§ Informationsystems- Increasecountrycapacitytogenerateandusenecessaryinformation;- Improvemanagementandavailabilityanduseofdata.
§ Infrastructure- Improveprocurementprocessesforinfrastructure;- Refusedonationswherelocalservicecannotbeassuredorwherebudgetswillnotbeable
topayforspareparts;- Ensureappropriatemaintenanceandcleanliness.
§ Politicaleconomy- Informationsharingwithkeystakeholdersonreasonsforefficiencyactions.
Thepoliticsofefficiencyimprovementsmightmeanthatsomeofthesestrategiescouldtakeconsiderablylongerthanexpectedinsomecountries,whileothersidentifiedinTable3mightbefeasibletoimplementmorequickly.Focusingonquickwinsshouldnot,ofcourse,divertattentionfromsomeofthekeylonger-termoptionssuchaschangingproviderpaymentmechanismsandintroducingmorecomprehensiveformsofstrategicpurchasing.Thelonger-termoptionsalsosometimeswillrequireimmediateinvestmentstoensurethattheycanhaveanimpact.Forexample,strategicpurchasinggenerallyrequiresstrengtheningcapacitiesintermsofpersonalskillsandcomputerizedinformationsystems.Changingproviderpaymentmechanismswillfrequentlyrequirenewormodifiedlegislationandaperiodofconsultationwithbothpatientsandproviders.
Despitetheknowledgeabouttechnicalsolutionsinsomekeyareas,someimportantgapsinknowledgeandanumberofcontroversiesremain:
A. MixofservicesThereisaconsiderablebodyofknowledgeaboutthecostsandeffectsofdifferentsortsof
healthservicesthatcanbeusefulforcountriesseekingtochoosethebestmix.Despitethis,thechoice
52
ofabenefitspackagestillreliesasmuchonanassessmentofwhatisfeasiblegiventheneedsofthepopulationasonevidenceoftheefficiencyofdifferentinterventionmixes.Theareaswhereadditionalinformationiscriticalinclude:
a) Economiesofscaleandscope:Whichinterventionsshouldbedeliveredinthesamesettingwith
thesamestaff(scope)andwhatisthemostefficientlevelofcoveragefortheavailableresources(scale)?
b) The right services in the right place: What is the efficiency of different services delivered indifferentplacesandhowtodecidewhatistheappropriatemixatcommunity,primary,secondaryandtertiarylevels?Howtoensurecontinuityofcareacrossthedifferentlevelsandacrossthelife-cycle, including the balance between prevention, promotion, treatment, rehabilitation andpalliation?
c) Governance,publichealthfunctionsandpersonalservices.Whatistheefficientbalancebetweentheseactivitiesandhowmuchshouldbespentonthem?
d) Expandingcoveragewithhealthservices,improvingquality,andfinancialprotection.Howcantheefficiencyof thesealternativesbeassessedand compared inaway thatpolicy-makers canincorporateintotheirpolicychoices?
e) Inter-sectoral ormulti-sectoral actions.What guidance can be offered toministries of healthaboutwhere their limited time andmoney should be focused if theywant to influence othersectorstotakeactionsthatimprovehealth?
B. Incentivesforefficiencyandqualitya) Arethereformsofvalue-basedpaymentthatbetterencourageefficiency,qualityandresultsthan
existing payment systems. How much do they cost, do they have unintended negativeconsequences and could they be routinely incorporated into health systems without externalfinancing? This is not simply an unresolved question but a controversial issuewhere there arestrongproponentsandopponents,forexample,ofdifferentformsofresults-basedpayment.
b) Whatcapacitiesdocountriesneedtohavetobegintopurchasestrategicallyandwhatsystems(forexample,budgeting,accountingand legalsystems)needtobe inplacetosupport it?Whathasbeentheexperiencewhereitacountryhassoughttomovetowardsstrategicpurchasingwithoutthe requisite systems in place and is it feasible tomove forward in very resource-constrainedsettings?
C. MeasurementThere are major gaps in the availability of data to identify the extent and sources of
inefficiencyincountries,particularlylower-incomecountries,partlybecausefewcountriesroutinelymonitor theirownprogress in reducing inefficiencyandpartlybecause routinehealth informationsystemsproduceonlyafewofthenecessaryindicators.a) Howcancountriesdeterminewhichindicatorsarecriticaltotheirowneffortstoachievemore
withtheavailableresourcesandwhataretheinvestmentcostsofensuringtheycanbemonitoredregularly?
D. Politicaleconomya) Aretheregeneralizablelessonstobelearnedaboutwhysomeefficiencyreformsworkandothers
fail,orreformsworkinonecountryandnotinanother?b) Arethereapproachesthatwouldbeusefulforallcountriestotakewhendevelopingtheagenda
for improving efficiency tomaximize the chances of success, such as involving civil society orparliamentarians?
53
E. Theefficiencyandsustainabilityofoptionstoimproveefficiency
Littleinformationisavailableonthecostsofthedifferentoptionsforimprovingefficiencytoincludealongsidethebenefitstoallowanassessmentofthemostefficientwayofimprovingefficiencyto be determined. Studies of payment for performance provide one example, and most of theliteraturehassimplereportedresults.
a) Howcancountriesrapidlyassessthecostsofthevariousoptionsforimprovingefficiency?Isthereanywaytheglobalcommunitycanassist?
F. Equityandefficiency
Some types of efficiency improvements can improve equity and some can exacerbateinequities.Sometimesitdependsonhowtheinterventionisdeliveredratherthanonthediseaseorhealthconditiontargetedorthetoolbeingused.Whatisrarelyassessedarethecostsandeffectsofdifferent ways of improving equity – recognizing that improving equity is a legitimate goal ofgovernment,andtheyshoulddoitinthemostefficientway.
b) How can countries rapidly assess the costs and impacts of the various options for reducinginequalitiesinhealth?Isthereanywaytheglobalcommunitycanassist?
X. RecommendationsEvery country has inefficiencies in their health systems and every country has technical
optionsforreducingthem.Basedonanassessmentofwhatisknownandwhatismissing,anumberofrecommendationsforimmediateactioncanbeformulated.Theyaredividedintoactionsthatneedtobetakenbycountriesandthosewheretheinternationalcommunityincludingfinancialpartnerswithlow-incomecountriesandresearcherscanassist.
Countries
§ Undertakeanassessmentofthemajorcausesofinefficiencyandthosethatarefeasibletochangeintheshort,mediumandlonger-term.
§ Developandimplementastrategyforimprovingefficiencyintheshorttomediumterm–thisshouldbepartofahealth financingstrategyalthoughsomeof theactionswillneedtobebroaderthanhealthfinancing.
§ Starttoputinplacethebackgroundinvestmentsthatneedtobemadetoensurethelonger-termoptions canbeundertaken–e.g. legislation, consultation, computerized informationsystems,staffskills.
§ Undertakeapoliticalaswellasatechnicalanalysistoguidewhichreformshavethegreatestchanceofsuccess,thenbuildsupportandnegateopposition.
§ Developasetofefficiencyindicatorsrelatingtothemaincausesofinefficiencyinthecountryandtheagendaforachievingmorefortheavailableresources.
§ Investinmethodstocollectthemandtoevaluateprogressregularly.§ Identifytheareasofpossibleinter-sectoralormulti-sectoralactionsthatwouldachievethe
largesthealthimpacts,andthepoliticalfeasibilityofinfluencingothersectorstoimplementthem(perhapsincollaborationwiththeministryofhealth).ThiswouldhelptheMinistryofHealth target the key ministries and make the best use of their own limited time andresources.
54
Internationalcommunity(includingresearchersinallcountries)
§ Routinelyassessthecostsaswellasimpactofeffortstoimproveefficiencysothatcountriescandeterminetheefficiencyofdifferentoptionsforimprovingefficiency.Anassessmentofthefinancialsustainabilityofthedifferentoptionsisalsohelpful.
§ Developanagendatoidentifythecost-effectivenessofeffortstoredresshealthinequalitiesaspartoftheefficiencyandequitydiscussion.
§ Developmethodswhichcanbeusedtohelpcountriesdeterminewhichofthemyriadofinter-sectoralormulti-sectoralactionstoimprovehealthshouldbegivenpriorityforthelimitedtimeandfinancialresourcesavailabletoaMinistryofHealth.Thiswouldfeedbackintoitem5above.
§ Continue to invest in the technologies that might “shift the frontier” of possibilities”,identifyingfurtheropportunitiestoimprovehealthandfinancialprotectionatlowcost,suchasvaccinesforHepatitisCandHIV/AIDS.
55
XI. References
Abimbola,S.,Olanipekun,T.,Igbokwe,U.,Negin,J.,Jan,S.,Martiniuk,A.,Ihebuzor,N.andAina,M.,2015.HowdecentralisationinfluencestheretentionofprimaryhealthcareworkersinruralNigeria.Globalhealthaction,8.
Adams,A.M.,Ahmed,T.,ElArifeen,S.,Evans,T.G.,Huda,T.andReichenbach,L.,2014.InnovationforuniversalhealthcoverageinBangladesh:acalltoaction.TheLancet,382(9910),pp.2104-2111.
Alberini,A.andŠčasný,M.,2013.Exploringheterogeneityinthevalueofastatisticallife:Causeofdeathv.riskperceptions.EcologicalEconomics,94,pp.143-155.
Ali,M.A.S.,Abou-Taleb,D.A.andMohamed,R.R.,2016.TreatmentadherenceandbeliefsaboutmedicinesamongEgyptianvitiligopatients.DermatologicTherapy,29(6),pp.413-418.
Aluttis,C.,Bishaw,T.andFrank,M.W.,2014.Theworkforceforhealthinaglobalizedcontext:globalshortagesandinternationalmigration.GlobalHealthAction.http://www.who.int/workforcealliance/01.pdf?ua=1
Alves,J.,Peralta,S.andPerelman,J.,2013.Efficiencyandequityconsequencesofdecentralizationinhealth:aneconomicperspective.RevistaPortuguesadeSaúdePública,31(1),pp.74-83.
Anaf,J.,Baum,F.,Freeman,T.,Labonte,R.,Javanparast,S.,Jolley,G.,Lawless,A.andBentley,M.,2014.Factorsshapingintersectoralactioninprimaryhealthcareservices.AustralianandNewZealandjournalofpublichealth,38(6),pp.553-559.
deAndrade,L.O.M.,PellegriniFilho,A.,Solar,O.,Rígoli,F.,deSalazar,L.M.,Serrate,P.C.F.,Ribeiro,K.G.,Koller,T.S.,Cruz,F.N.B.andAtun,R.,2015.Socialdeterminantsofhealth,universalhealthcoverage,andsustainabledevelopment:casestudiesfromLatinAmericancountries.TheLancet,385(9975),pp.1343-1351.
Angevine,P.D.andBerven,S.,2014.Healtheconomicstudies:anintroductiontocost-benefit,cost-effectiveness,andcost-utilityanalyses.Spine,39(22S),pp.S9-S15.
Araujo,E.C.,Evans,T.G.andMaeda,A.,2016.Usingeconomicanalysisinhealthworkforcepolicy-making.OxfordReviewofEconomicPolicy,32(1),pp.41-63.
Asante,A.D.,Martins,N.,Otim,M.E.andDewdney,J.,2014.RetainingdoctorsinruralTimor-Leste:acriticalappraisaloftheopportunitiesandchallenges.BulletinoftheWorldHealthOrganization,92(4),pp.277-282.
Asante,A.D.,Price,J.,Hayen,A.,Irava,W.,Martins,J.,Guinness,L.,Ataguba,J.E.,Limwattananon,S.,Mills,A.,Jan,S.andWiseman,V.,2014.AssessmentofequityinhealthcarefinancinginFijiandTimor-Leste:astudyprotocol.BMJopen,4(12),p.e006806.
Asante,A.D.,Price,J.,Hayen,A.,Jan,S.andWiseman,V.,2016.Equityinhealthcarefinancinginlow-andmiddle-incomecountries:asystematicreviewofevidencefromstudiesusingbenefitandfinancingincidenceanalyses.PloSone,11(4),p.e0152866.
Asmild,M.,Hollingsworth,B.andBirch,S.,2013.Thescaleofhospitalproductionindifferentsettings:Onesizedoesnotfitall.JournalofProductivityAnalysis,40(2),pp.197-206.
Atav,A.S.,2014,November.IrrationalUseofMedicines:AnAssessmentofTurkishPatientsinPrimaryCareSettings.In142ndAPHAAnnualMeetingandExposition(November15-November19,2014).APHA.
56
Atkinson,S.andHaran,D.,2004.Backtobasics:doesdecentralizationimprovehealthsystemperformance?EvidencefromCearáinnorth-eastBrazil.BulletinoftheWorldHealthOrganization,82(11),pp.822-827.
Azfar,O.,Kähkönen,S.andMeagher,P.,2001.Conditionsforeffectivedecentralizedgovernance:Asynthesisofresearchfindings.UniversityofMaryland,CenterforInstitutionalReformandtheInformalSector.http://www1.worldbank.org/publicsector/learningprogram/Decentralization/ConditionsEffective.pdf
Balabanova,D.,Mills,A.,Conteh,L.,Akkazieva,B.,Banteyerga,H.,Dash,U.,Gilson,L.,Harmer,A.,Ibraimova,A.,Islam,Z.andKidanu,A.,2013.GoodHealthatLowCost25yearson:lessonsforthefutureofhealthsystemsstrengthening.TheLancet,381(9883),pp.2118-2133.
Barroy,H.,Sparkes,S.andDaleE.2016.Assessingfiscalspaceforhealthexpansioninlow-and-middleincomecountries:areviewoftheevidence.WorldHealthOrganization,Geneva.HealthFinancingPaper16.3.
Bashrahil,K.A.,2010.IndicatorsofrationaldruguseandhealthservicesinHadramout,Yemen/Indicateursd'usagerationneldesmedicamentsetdesservicesdesantedanslegouvernoratd'Hadramout(Yemen).EasternMediterraneanHealthJournal,16(2),151.
Bassi,J.andLau,F.,2013.Measuringvalueformoney:ascopingreviewoneconomicevaluationofhealthinformationsystems.JournaloftheAmericanMedicalInformaticsAssociation,20(4),pp.792-801.LimitedgoodinfoonthecostsandbenefitsofvarioussortsofHIS,includingelectronicmedicalrecords.
Bastani,P.,Dinarvand,R.,SamadBeik,M.andPourmohammadi,K.,2016.PharmaceuticalstrategicpurchasingrequirementsinIran:Priceinterventionsandtherelatedeffectivefactors.Journalofresearchinpharmacypractice,5(1),p.35.
Basu,S.,Andrews,J.,Kishore,S.,Panjabi,R.,Stuckler,D.,2012.Comparativeperformanceofprivateandpublichealthcaresystemsinlow-andmiddle-incomecountries:asystematicreview.PLoSMed9:e1001244.
NejmudinK,B.,HerbstC.H.,ZhaoF.,SoucatA.HealthExtensionWorkersinEthiopia,2011.ImprovedAccessandCoveragefortheRuralPoor.YesAfricaCan:SuccessStoriesfromADynamicContinent,pp.433-444.
Berendes,S.,Heywood,P.,Oliver,S.,Garner,P.,2011.Qualityofprivateandpublicambulatoryhealthcareinlowandmiddleincomecountries:systematicreviewofcomparativestudies.PLoSMed8:e1000433.
Berenguer,G.,Iyer,A.V.andYadav,P.,2016.Disentanglingtheefficiencydriversincountry-levelglobalhealthprograms:Anempiricalstudy.JournalofOperationsManagement,45,pp.30-43.
Berwick,D.M.,2017.Avoidingoveruse—thenextqualityfrontier.TheLancet.http://dx.doi.org/10.1016/S0140-6736(16)32570-3
Besstremyannaya,G.,2013.TheimpactofJapanesehospitalfinancingreformonhospitalefficiency:Adifference-in-differenceapproach.TheJapaneseEconomicReview,64(3),pp.337-362.
Betrán,A.P.,Ye,J.,Moller,A.B.,Zhang,J.,Gülmezoglu,A.M.andTorloni,M.R.,2016.Theincreasingtrendincaesareansectionrates:global,regionalandnationalestimates:1990-2014.PloSone,11(2),p.e0148343.
57
Bhattacharyya,O.,Khor,S.,McGahan,A.,Dunne,D.,Daar,A.S.andSinger,P.A.,2010.Innovativehealthservicedeliverymodelsinlowandmiddleincomecountries-whatcanwelearnfromtheprivatesector?.HealthResearchPolicyandSystems,8(1),p.24.
Binyaruka,P.,Patouillard,E.,Powell-Jackson,T.,Greco,G.,Maestad,O.andBorghi,J.,2015.Effectofpayingforperformanceonutilisation,quality,andusercostsofhealthservicesinTanzania:acontrolledbeforeandafterstudy.PloSone,10(8),p.e0135013.
Bloom,N.,Propper,C.,Seiler,S.andVanReenen,J.,2015.Theimpactofcompetitiononmanagementquality:evidencefrompublichospitals.TheReviewofEconomicStudies,p.rdu045.
Bonenberger,M.,Aikins,M.,Akweongo,P.andWyss,K.,2014.TheeffectsofhealthworkermotivationandjobsatisfactiononturnoverintentioninGhana:across-sectionalstudy.Humanresourcesforhealth,12(1),p.43.
Borghi,J.,Little,R.,Binyaruka,P.,Patouillard,E.andKuwawenaruwa,A.,2015.InTanzania,themanycostsofpay-for-performanceleaveopentodebatewhetherthestrategyiscost-effective.Healthaffairs,34(3),pp.406-414.
Brixi,H.,Mu,Y.,Targa,B.andHipgrave,D.,2012.Engagingsub-nationalgovernmentsinaddressinghealthequities:challengesandopportunitiesinChina’shealthsystemreform.Healthpolicyandplanning,p.czs120.
Brownlee,S.,Chalkidou,K.,Doust,J.,Elshaug,A.G.,Glasziou,P.,Heath,I.,Nagpal,S.,Saini,V.,Srivastava,D.,Chalmers,K.andKorenstein,D.,2017.Evidenceforoveruseofmedicalservicesaroundtheworld.TheLancet.http://dx.doi.org/10.1016/S0140-6736(16)32585-5
BumpJ.,SparksS.,TatarM.,CelikY.,2014.TurkeyonthewayofuniversalhealthcoveragethroughtheHealthTransformationProgram(2003−13).HNPDiscussionPaper,Washington,DC:WorldBank.http://documents.worldbank.org/curated/en/704941468338368780/pdf/931720WP0UHC0C0hnp0discussion0paper.pdf
Cambiano,V.,Ford,D.,Mabugu,T.,Mavedzenge,S.N.,Miners,A.,Mugurungi,O.,Nakagawa,F.,Revill,P.andPhillips,A.,2015.Assessmentofthepotentialimpactandcost-effectivenessofself-testingforHIVinlow-incomecountries.JournalofInfectiousDiseases,212(4),pp.570-577.
Cameron,A.,Ewen,M.,Ross-Degnan,D.,Ball,D.andLaing,R.,2009.Medicineprices,availability,andaffordabilityin36developingandmiddle-incomecountries:asecondaryanalysis.Thelancet,373(9659),pp.240-249.
Cameron,A.andLaing,R.,2010.Costsavingsofswitchingprivatesectorconsumptionfromoriginatorbrandmedicinestogenericequivalents.Worldhealthreport2010BackgroundPaper35,WHO,Geneva.
Cameron,A.,Bansal,A.,Dua,T.,Hill,S.R.,Moshe,S.L.,Mantel-Teeuwisse,A.K.andSaxena,S.,2012.Mappingtheavailability,price,andaffordabilityofantiepilepticdrugsin46countries.Epilepsia,53(6),pp.962-969.
Campanella,P.,Lovato,E.,Marone,C.,Fallacara,L.,Mancuso,A.,Ricciardi,W.andSpecchia,M.L.,2015.Theimpactofelectronichealthrecordsonhealthcarequality:asystematicreviewandmeta-analysis.TheEuropeanJournalofPublicHealth,p.ckv122.
Cashin,C.,Bloom,D.,Sparkes,S.,Barroy,H.,Kutzin,J.,O'Dougherty,S.andWorldHealthOrganization,2017.Aligningpublicfinancialmanagementandhealthfinancing:sustaining
58
progresstowarduniversalhealthcoverage.http://apps.who.int/iris/bitstream/10665/254680/1/9789241512039-eng.pdf
Carey,K.,2015.MeasuringtheHospitalLengthofStay/ReadmissionCostTrade-OffUnderaBundledPaymentMechanism.Healtheconomics,24(7),pp.790-802.
Carey,G.,Crammond,B.andKeast,R.,2014.Creatingchangeingovernmenttoaddressthesocialdeterminantsofhealth:howcaneffortsbeimproved?.BMCPublicHealth,14(1),p.1087.
CDC,2017.ThePublicHealthSystemandthe10EssentialPublicHealthServiceswww.cdc.gov/nphpsp/essentialservices.htmlaccessed7February2017.
Chen,M.,Wang,L.,Chen,W.,Zhang,L.,Jiang,H.andMao,W.,2014.Doeseconomicincentivematterforrationaluseofmedicine?China’sexperiencefromtheessentialmedicinesprogram.Pharmacoeconomics,32(3),pp.245-255.
Chetty,R.andLooney,A.,2006.Consumptionsmoothingandthewelfareconsequencesofsocialinsuranceindevelopingeconomies.Journalofpubliceconomics,90(12),pp.2351-2356.
Chen,M.,Fang,G.,Wang,L.,Wang,Z.,Zhao,Y.andSi,L.,2015.Whobenefitsfromgovernmenthealthcaresubsidies?AnassessmentoftheequityofhealthcarebenefitsdistributioninChina.PloSone,10(3),p.e0119840.
ChisholmD.,EvansD.B.,2010.“Improvinghealthsystemefficiencyasameansofmovingtowardsuniversalcoverage.”Backgroundpaper28,WHOWorldhealthreport2010.http://www.who.int/healthsystems/topics/financing/healthreport/whr_background/en/index1.html
Chisholm,D.,Baltussen,R.,Evans,D.B.,Ginsberg,G.,Lauer,J.A.,Lim,S.,Ortegon,M.,Salomon,J.,Stanciole,A.andEdejer,T.T.T.,2012.Whataretheprioritiesforpreventionandcontrolofnon-communicablediseasesandinjuriesinsub-SaharanAfricaandSouthEastAsia?.BMJ:BritishMedicalJournal(Online),344.
Choudhry,N.K.,Denberg,T.D.andQaseem,A.,2016.Improvingadherencetotherapyandclinicaloutcomeswhilecontainingcosts:opportunitiesfromthegreateruseofgenericmedications:bestpracticeadvicefromtheClinicalGuidelinesCommitteeoftheAmericanCollegeofPhysicians.Annalsofinternalmedicine,164(1),pp.41-49.
Cleary,S.,Mooney,G.andMcIntyre,D.,2010.EquityandefficiencyinHIV-treatmentinSouthAfrica:thecontributionofmathematicalprogrammingtoprioritysetting.Healtheconomics,19(10),pp.1166-1180.
Coady,D.,Grosh,M.andHoddinott,J.,2004.Targetingoutcomesredux.TheWorldBankResearchObserver,19(1),pp.61-85.
Coady,D.,Francese,M.andShang,B.,2014.TheEfficiencyImperative.FinanceandDevelopment,IMF,December,pp.30-32.
Coarasa,J.,Harding,A.,Kanneganti,S.,Kim,E.,Secci,F.,Seiter,A.,Sjoblom,M.,forthcoming.BackgroundPapersforWorldBankGroupPrivateSectorinHealthRoadmap.WorldBank.
CommissiononSocialDeterminantsofHealth,2008.FinalreportoftheCommissiononSocialDeterminantsofHealth.Geneva:WorldHealthOrganization,pp.1-247.
Conrad,D.A.,Grembowski,D.,Hernandez,S.E.,Lau,B.andMarcus-Smith,M.I.R.I.A.M.,2014.EmergingLessonsfromRegionalandStateInnovationinValue-BasedPaymentReform:BalancingCollaborationandDisruptiveInnovation.MilbankQuarterly,92(3),pp.568-623.
59
deCos,P.H.andMoral-Benito,E.,2014.Determinantsofhealth-systemefficiency:evidencefromOECDcountries.Internationaljournalofhealthcarefinanceandeconomics,14(1),pp.69-93.
Cozad,M.andWichmann,B.,2013.Efficiencyofhealthcaredeliverysystems:effectsofhealthinsurancecoverage.AppliedEconomics,45(29),pp.4082-4094.Healthinsurancecoveragereducesefficiency–checkiffeeforservice
Crisp,N.andChen,L.,2014.Globalsupplyofhealthprofessionals.NewEnglandJournalofMedicine,370(10),pp.950-957.
Cumper,G.E.,1984.Determinantsofhealthlevelsindevelopingcountries.Wiley,NewYork,1984.
Cylus,J.,Papanicolas,I.,&Smith,P.,2013.Healthsystemperformancecomparison:anagendaforpolicy,informationandresearch:anagendaforpolicy,informationandresearch.McGraw-HillEducation(UK).
Daemmrich,A.,2013.ThepoliticaleconomyofhealthcarereforminChina:negotiatingpublicandprivate.SpringerPlus,2(1),p.448.
Dahlgren,G.,1994,December.Theneedforinter-sectoralactionforhealth.InEuropeanHealthPolicyConference:opportunitiesforthefuture,Copenhagen(Vol.2,pp.1994-97).
Damberg,C.L.,Sorbero,M.E.,Lovejoy,S.L.,Martsolf,G.R.,Raaen,L.andMandel,D.,2014.Measuringsuccessinhealthcarevalue-basedpurchasingprograms.RandCorporation.http://www.rand.org/pubs/research_reports/RR306z1.html
Daouk-Öyry,L.,Anouze,A.L.,Otaki,F.,Dumit,N.Y.andOsman,I.,2014.TheJOINTmodelofnurseabsenteeismandturnover:asystematicreview.Internationaljournalofnursingstudies,51(1),pp.93-110.
Das,A.,Gopalan,S.S.andChandramohan,D.,2016.Effectofpayforperformancetoimprovequalityofmaternalandchildcareinlow-andmiddle-incomecountries:asystematicreview.BMCpublichealth,16(1),p.321.
Davies,S.C.,Winpenny,E.,Ball,S.,Fowler,T.,Rubin,J.andNolte,E.,2014.Fordebate:anewwaveinpublichealthimprovement.TheLancet,384(9957),pp.1889-1895.
Davis,P.,Milne,B.,Parker,K.,Hider,P.,Lay-Yee,R.,Cumming,J.,&Graham,P.,2013.Efficiency,effectiveness,equity(E3).Evaluatinghospitalperformanceinthreedimensions.HealthPolicy,112(1),19-27.
Dawson,J.,Huikuri,S.andArmada,F.,2015.LiverpoolActiveCity2005–2010:IncreasingPopulationPhysicalActivityLevelsThroughInter-sectoralAction.JournalofPhysicalActivityandHealth,12(6),pp.749-755.
Desalegn,A.A.(2013).AssessmentofdrugusepatternusingWHOprescribingindicatorsatHawassaUniversityteachingandreferralhospital,southEthiopia:across-sectionalstudy.BMChealthservicesresearch,13(1),170.
DiabetesPreventionProgramResearchGroup,2012.The10-yearcost-effectivenessoflifestyleinterventionormetforminfordiabetesprevention.Diabetescare,35(4),pp.723-730.
Dieleman,M.,&Harnmeijer,J.W.,2006.Improvinghealthworkerperformance:insearchofpromisingpractices.Geneva:WorldHealthOrganization,5-34.
Du,J.,Wang,J.,Chen,Y.,Chou,S.Y.andZhu,J.,2014.IncorporatinghealthoutcomesinPennsylvaniahospitalefficiency:anadditivesuper-efficiencyDEAapproach.AnnalsofOperationsResearch,221(1),pp.161-172.
60
Dutrey,A.P.,2007.Successfultargeting?:reportingefficiencyandcostsintargetedpovertyalleviationprogrammes.UnitedNationsResearchInstituteforSocialDevelopment.UNRISTSocialPolicyandDevelopmentProgrammePaperNumber35.http://www.unrisd.org/__80256b3c005bccf9.nsf/0/0b87c67449c938edc12573d10049830b?OpenDocument&cntxt=032EE&cookielang=es&Click=
Dylst,P.,Vulto,A.andSimoens,S.,2014.AnalysisofEuropeanpolicytowardsgenericmedicines.GaBiJ,3(1),pp.34-5.
Ebong,C.E.,&Levy,P.(2011).Impactoftheintroductionofnewvaccinesandvaccinewastagerateonthecost-effectivenessofroutineEPI:lessonsfromadescriptivestudyinaCameroonianhealthdistrict.CostEffectivenessandResourceAllocation,9(1),9.
Edejer,T.T.T.,Aikins,M.,Black,R.,Wolfson,L.,Hutubessy,R.andEvans,D.B.,2005.Costeffectivenessanalysisofstrategiesforchildhealthindevelopingcountries.Bmj,331(7526),p.1177
Eijkenaar,F.,Emmert,M.,Scheppach,M.andSchöffski,O.,2013.Effectsofpayforperformanceinhealthcare:asystematicreviewofsystematicreviews.Healthpolicy,110(2),pp.115-130.
Elshaug,A.G.,McWilliams,J.M.andLandon,B.E.,2013.Thevalueoflow-valuelists.Jama,309(8),pp.775-776.
Elshaug,A.G.,Rosenthal,M.B.,Lavis,J.N.,Brownlee,S.,Schmidt,H.,Nagpal,S.,Littlejohns,P.,Srivastava,D.,Tunis,S.andSaini,V.,2017.Leversforaddressingmedicalunderuseandoveruse:achievinghigh-valuehealthcare.TheLancet.http://dx.doi.org/10.1016/S0140-6736(16)32586-7
Evans,D.B.,Lim,S.S.,Adam,T.,Edejer,T.T.T.andWHOChoosingInterventionsthatareCostEffective(CHOICE)MillenniumDevelopmentGoalsTeam,2005.Evaluationofcurrentstrategiesandfutureprioritiesforimprovinghealthindevelopingcountries.Bmj,331(7530),pp.1457-61.
Feldman,R.,2015.TheEconomicsofProviderPaymentReform:AreAccountableCareOrganizationstheAnswer?.JournalofHealthPolitics,PolicyandLaw,40(4),pp.745-760.
Ferreira,M.B.C.,Heineck,I.,Flores,L.M.,Camargo,A.L.,DalPizzol,T.D.S.,Torres,I.L.D.S.,Koenig,A.,Trevisol,D.J.,Melo,A.M.M.F.D.,Cardoso,T.F.M.andMonreal,M.T.F.D.,2013.Rationaluseofmedicines:prescribingindicatorsatdifferentlevelsofhealthcare.BrazilianJournalofPharmaceuticalSciences,49(2),pp.329-340.
Fox,S.,Witter,S.,Wylde,E.,Mafuta,E.andLievens,T.,2013.Payinghealthworkersforperformanceinafragmented,fragilestate:reflectionsfromKatangaProvince,DemocraticRepublicofCongo.Healthpolicyandplanning,p.czs138.
Fox,A.M.andReich,M.R.,2015.Thepoliticsofuniversalhealthcoverageinlow-andmiddle-incomecountries:aframeworkforevaluationandaction.Journalofhealthpolitics,policyandlaw,40(5),pp.1023-1060.
Frenk,J.,González-Pier,E.,Gómez-Dantés,O.,Lezana,M.A.,&Knaul,F.M.(2006).ComprehensivereformtoimprovehealthsystemperformanceinMexico.TheLancet,368(9546),1524-1534.
Fritz,V.,Sweet,S.andVerhoeven,M.,2014.Strengtheningpublicfinancialmanagement:exploringdriversandeffects.WorldBankPolicyResearchWorkingPaperNo.7084
Frogner,B.K.,Frech,H.E.,&Parente,S.T.(2015).Comparingefficiencyofhealthsystemsacrossindustrializedcountries:apanelanalysis.BMChealthservicesresearch,15(1),415.
61
Galetto,M.,Marginson,P.andSpieser,C.,2014.Collectivebargainingandreformstohospitalhealthcareprovision:AcomparisonoftheUK,ItalyandFrance.EuropeanJournalofIndustrialRelations,20(2),pp.131-147.
Gao,C.,Xu,F.andLiu,G.G.,2014.PaymentreformandchangesinhealthcareinChina.SocialScience&Medicine,111,pp.10-16.
Gearhart,R.,2016a.Notheory:anexplanationofthelackofconsistencyincross-countryhealthcarecomparisonsusingnon-parametricestimators.Healtheconomicsreview,6(1),p.40.
Gearhart,R.,2016b.Therobustnessofcross-countryhealthcarerankingsamonghomogeneousOECDcountries.JournalofAppliedEconomics,19(1),pp.113-143.
Gee,J.andButton,M.,2015.Thefinancialcostoffraud2015:whatthelatestdatafromaroundtheworldshows.CentreforCounter-FraudPolicies,UniversityofPortsmouth.http://eprints.port.ac.uk/18410/1/PKF_The_financial_cost_of_fraud_2015.pdf
Gillingham,R.,2014.FiscalPolicyforHealthPolicyMakers.Washington,DC:WorldBank.http://documents.worldbank.org/curated/en/936691468324008445/pdf/879810WP0Fisca00Box385214B00PUBLIC0.pdf
Glasziou,P.,Straus,S.,Brownlee,S.,Trevena,L.,Dans,L.,Guyatt,G.,Elshaug,A.G.,Janett,R.andSaini,V.,2017.Evidenceforunderuseofeffectivemedicalservicesaroundtheworld.TheLancet.http://dx.doi.org/10.1016/S0140-6736(16)30946-1
Goetzel,R.Z.,Henke,R.M.,Tabrizi,M.,Pelletier,K.R.,Loeppke,R.,Ballard,D.W.,Grossmeier,J.,Anderson,D.R.,Yach,D.,Kelly,R.K.andSerxner,S.,2014.Doworkplacehealthpromotion(wellness)programswork?.JournalofOccupationalandEnvironmentalMedicine,56(9),pp.927-934.
Gordon,L.G.andRowell,D.,2015.Healthsystemcostsofskincancerandcost-effectivenessofskincancerpreventionandscreening:asystematicreview.EuropeanJournalofCancerPrevention,24(2),pp.141-149.
Gostin,L.O.andFriedman,E.A.,2015.AretrospectiveandprospectiveanalysisofthewestAfricanEbolavirusdiseaseepidemic:robustnationalhealthsystemsatthefoundationandanempoweredWHOattheapex.TheLancet,385(9980),pp.1902-1909.
Granich,R.,Kahn,J.G.,Bennett,R.,Holmes,C.B.,Garg,N.,Serenata,C.,Sabin,M.L.,Makhlouf-Obermeyer,C.,Mack,C.D.F.,Williams,P.andJones,L.,2012.ExpandingARTfortreatmentandpreventionofHIVinSouthAfrica:estimatedcostandcost-effectiveness2011-2050.PloSone,7(2),p.e30216.
Greene,W.,2004.Distinguishingbetweenheterogeneityandinefficiency:stochasticfrontieranalysisoftheworldhealthorganization'spaneldataonnationalhealthcaresystems.HealthEconomics,13(10),pp.959-980.
Greene,W.H.,2008.Theeconometricapproachtoefficiencyanalysis.Themeasurementofproductiveefficiencyandproductivitygrowth,1,pp.92-250.
Grigoli,F.andKapsoli,J.,2013.Wastenot,wantnot:theefficiencyofhealthexpenditureinemerginganddevelopingeconomies.IMFWorkingPaperNo.13/187.
Grossman,M.,1976.Thecorrelationbetweenhealthandschooling.InTerleckyj,N.E.ed.,Householdproductionandconsumption.Washington,DC:NationalBureauofEconomicResearch(pp.147-224).www.nber.org/chapters/c3962.pdf
62
Gwynne,K.andLincoln,M.,2016.DevelopingtheruralhealthworkforcetoimproveAustralianAboriginalandTorresStraitIslanderhealthoutcomes:asystematicreview.AustralianHealthReview.
Haas,J.S.,Phillips,K.A.,Gerstenberger,E.P.andSeger,A.C.,2005.Potentialsavingsfromsubstitutinggenericdrugsforbrand-namedrugs:medicalexpenditurepanelsurvey,1997–2000.Annalsofinternalmedicine,142(11),pp.891-897.
Hall,J.,2015.Australianhealthcare—thechallengeofreforminafragmentedsystem.NewEnglandJournalofMedicine,373(6),pp.493-497.
Hamidi,S.andAkinci,F.,2016.MeasuringEfficiencyofHealthSystemsoftheMiddleEastandNorthAfrica(MENA)RegionUsingStochasticFrontierAnalysis.Appliedhealtheconomicsandhealthpolicy,pp.1-11.
Hansen,A.H.,Halvorsen,P.A.,Aaraas,I.J.andFørde,O.H.,2013.ContinuityofGPcareisrelatedtoreducedspecialisthealthcareuse:across-sectionalsurvey.BrJGenPract,63(612),pp.e482-e489.
Hanvoravongchai,P.andHsiao,W.C.,2007.Thailand:achievinguniversalcoveragewithsocialhealthinsurance.Socialhealthinsurancefordevelopingnations,p.133.
Harris,J.,2015.“DevelopmentalCapture”oftheState:ExplainingThailand'sUniversalCoveragePolicy.Journalofhealthpolitics,policyandlaw,40(1),pp.165-193.
Hassali,M.A.,Alrasheedy,A.A.,McLachlan,A.,Nguyen,T.A.,AL-Tamimi,S.K.,Ibrahim,M.I.M.andAljadhey,H.,2014.Theexperiencesofimplementinggenericmedicinepolicyineightcountries:Areviewandrecommendationsforasuccessfulpromotionofgenericmedicineuse.SaudiPharmaceuticalJournal,22(6),pp.491-503.
Hauck,K.andSmith,P.C.,2015.Thepoliticsofprioritysettinginhealth:apoliticaleconomyperspective.CenterforGlobalDevelopmentWorkingPaperNo.414.https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2671571
Hauck,K.,Martin,S.andSmith,P.C.,2016.Prioritiesforactiononthesocialdeterminantsofhealth:Empiricalevidenceonthestrongestassociationswithlifeexpectancyin54low-incomecountries,1990–2012.SocialScience&Medicine,167,pp.88-98.
Heinze,G.,Hronsky,M.,Reichardt,B.,Baumgärtel,C.,Müllner,M.,Bucsics,A.andWinkelmayer,W.C.,2015.Potentialsavingsinprescriptiondrugcostsforhypertension,hyperlipidemia,anddiabetesmellitusbyequivalentdrugsubstitutioninAustria:anationwidecohortstudy.Appliedhealtheconomicsandhealthpolicy,13(2),pp.193-205.
Hepworth,N.,2015.Debate:Implementingadvancedpublicfinancialmanagementreformindevelopingcountries.PublicMoney&Management,35(4),pp.251-253.
Heredia-Ortiz,E.,2013.DataforEfficiency:AToolforAssessingHealthSystems’ResourceUseEfficiency.Bethesda,MD:HealthFinance&GovernanceProject,AbtAssociatesInc.https://www.hfgproject.org/wp-content/uploads/2014/10/04-Data-for-Efficiency-A-Tool-for-Assessing-Health-Systems-Resource-Use-Efficiency.pdf
Hernandez-Peña,P.,Poullier,J.P.,VanMosseveld,C.J.M.,VandeMaele,N.,Cherilova,V.,Indikadahena,C.,Lie,G.,Tan-Torres,T.andEvans,D.B.,2013.HealthworkerremunerationinWHOMemberStates.BulletinoftheWorldHealthOrganization,91(11),pp.808-815.
Hollingsworth,B.,2003.Non-parametricandparametricapplicationsmeasuringefficiencyinhealthcare.Healthcaremanagementscience,6(4),pp.203-218.
63
Hollingsworth,B.,2008.Themeasurementofefficiencyandproductivityofhealthcaredelivery.Healtheconomics,17(10),pp.1107-1128.
Hollingsworth,B.,Healthsystemefficiency:measurementandpolicy(2016).InCylus,J.,Papanicolas,I.andSmith,P.C.(eds).Healthsystemefficiency:Howtomakemeasurementmatterforpolicyandmanagement.WorldHealthOrganization,Copenhagen.
Holloway,K.A.,2011.Combatinginappropriateuseofmedicines.Expertreviewofclinicalpharmacology,4(3),pp.335-348.
Holloway,K.andDijk,L.V.,2011.Rationaluseofmedicines.http://www.narcis.nl/publication/RecordID/publicat%3A3693
Holloway,K.A.,Ivanovska,V.,Wagner,A.K.,Vialle-Valentin,C.andRoss-Degnan,D.,2013.Haveweimproveduseofmedicinesindevelopingandtransitionalcountriesanddoweknowhowto?Twodecadesofevidence.TropicalMedicine&InternationalHealth,18(6),pp.656-664.
Honda,A.,2014.Whatisstrategicpurchasingforhealth?LSHTMResearchonLine.http://resyst.lshtm.ac.uk/sites/resyst.lshtm.ac.uk
Hotchkiss,D.R.,Banteyerga,H.andTharaney,M.,2015.Jobsatisfactionandmotivationamongpublicsectorhealthworkers:evidencefromEthiopia.Humanresourcesforhealth,13(1),p.83.
Hou,X.,Velényi,E.V.,Yazbeck,A.S.,Iunes,R.F.andSmith,O.,2013.LearningfromEconomicDownturns:HowtoBetterAssess,Track,andMitigatetheImpactontheHealthSector.WashingtonDC,WorldBank.
Hu,S.,Liu,X.,&Peng,Y.(2003).AssessmentofantibioticprescriptioninhospitalisedpatientsataChineseuniversityhospital.Journalofinfection,46(3),161-163.
Hurley,R.,2014.Candoctorsreduceharmfulmedicaloveruseworldwide?.BMJ:BritishMedicalJournal(Online),349.
Hussey,P.S.,Anderson,G.F.,Osborn,R.,Feek,C.,McLaughlin,V.,Millar,J.,&Epstein,A.,2004.Howdoesthequalityofcarecompareinfivecountries?HealthAffairs,23(3),89-99.
Imam,P.A.andJacobs,D.,2014.EffectofcorruptionontaxrevenuesintheMiddleEast.ReviewofMiddleEastEconomicsandFinanceRev.MiddleEastEcon.Fin.,10(1),pp.1-24.
InternationalFinanceCorporation(IFC),2011.HealthyPatnerships:HowGovernmentsCanEngagethePrivateSectortoImproveHealthinAfrica,Washington,DC:IFC
Iyengar,S.,Tay-Teo,K.,Vogler,S.,Beyer,P.,Wiktor,S.,deJoncheere,K.andHill,S.,2016.Prices,costs,andaffordabilityofnewmedicinesforhepatitisCin30countries:aneconomicanalysis.PLoSmed,13(5),p.e1002032.
James,C.,Berchet,C.andMuir,T.2017.AddressingoperationalwastebybettertargetingtheuseofhospitalcareinOECD,TacklingWastefulSpendingonHealth,OECDPublishing,Paris.http://dx.doi.org/10.1787/9789264266414-en.
Jamison,D,T.,Breman,J.G.,Measham,A.R.,Alleyne,G.,Claeson,M.,Evans,D.B.,Jha,P.,Mills,A.,Musgrove,P.,2006.DiseaseControlPrioritiesinDevelopingCountries,SecondEdition.Washington,DC:WorldBankandOxfordUniversityPress.
Jehu-Appiah,C.,Sekidde,S.,Adjuik,M.,Akazili,J.,Almeida,S.D.,Nyonator,F.,Baltussen,R.,Asbu,E.Z.andKirigia,J.M.,2014.Ownershipandtechnicalefficiencyofhospitals:evidencefromGhanausingdataenvelopmentanalysis.CostEffectivenessandResourceAllocation,12(1),p.9.
64
Jiang,M.,Zhou,Z.,Wu,L.,Shen,Q.,Lv,B.,Wang,X.,Yang,S.andFang,Y.,2015.Medicineprices,availability,andaffordabilityintheShaanxiProvinceinChina:implicationsforthefuture.Internationaljournalofclinicalpharmacy,37(1),pp.12-17.
deJonge,E.,K.,Jamshed,N.,Gilden,D.,Kubisiak,J.,Bruce,S.R.andTaler,G.,2014.EffectsofHome-BasedPrimaryCareonMedicareCostsinHigh-RiskElders.JournaloftheAmericanGeriatricsSociety,62(10),pp.1825-1831.
Joumard,I.,Andre,C.,Nicq,C.,andChatal,O.,HealthStatusDeterminants:Lifestyle,Environment,HealthCareResourcesandEfficiency(2010).OECDEconomicsDepartmentWorkingPaperNo.627.AvailableatSSRN:https://ssrn.com/abstract=1616544orhttp://dx.doi.org/10.2139/ssrn.1616544
Kalyango,J.N.,Lindstrand,A.,Rutebemberwa,E.,Ssali,S.,Kadobera,D.,Karamagi,C.,Peterson,S.andAlfven,T.,2012.IncreaseduseofcommunitymedicinedistributorsandrationaluseofdrugsinchildrenlessthanfiveyearsofageinUgandacausedbyintegratedcommunitycasemanagementoffever.TheAmericanjournaloftropicalmedicineandhygiene,87(5Suppl),pp.36-45.
Kelsall,T.,Hart,T.,&Laws,E.,2016.Politicalsettlementsandpathwaystouniversalhealthcoverage(No.432).OverseasDevelopmentInstituteWorkingPaper432.https://www.odi.org/sites/odi.org.uk/files/resource-documents/10382.pdf
Khatib,R.,McKee,M.,Shannon,H.,Chow,C.,Rangarajan,S.,Teo,K.,Wei,L.,Mony,P.,Mohan,V.,Gupta,R.andKumar,R.,2016.Availabilityandaffordabilityofcardiovasculardiseasemedicinesandtheireffectonuseinhigh-income,middle-income,andlow-incomecountries:ananalysisofthePUREstudydata.TheLancet,387(10013),pp.61-69.
Kiadaliri,A.A.,Jafari,M.andGerdtham,U.G.,2013.Frontier-basedtechniquesinmeasuringhospitalefficiencyinIran:asystematicreviewandmeta-regressionanalysis.BMChealthservicesresearch,13(1),p.312.
Kieny,M.P.,Evans,D.B.,Schmets,G.andKadandale,S.,2014.Health-systemresilience:reflectionsontheEbolacrisisinwesternAfrica.BulletinoftheWorldHealthOrganization,92(12),pp.850-850.
Kim,C.W.,Baik,S.H.,Roh,Y.H.,Kang,J.,Hur,H.,Min,B.S.,Lee,K.Y.andKim,N.K.,2015.Cost-effectivenessofroboticsurgeryforrectalcancerfocusingonshort-termoutcomes:apropensityscore-matchinganalysis.Medicine,94(22),p.e823.
Kirigia,J.M.andAsbu,E.Z.,2013.TechnicalandscaleefficiencyofpubliccommunityhospitalsinEritrea:anexploratorystudy.Healtheconomicsreview,3(1),p.6.someefficient,butmosttoosmall.
Kirigia,J.M.,Sambo,L.G.,Renner,A.,Alemu,W.,Seasa,S.andBah,Y.,2011.TechnicalefficiencyofprimaryhealthunitsinKailahunandKenemadistrictsofSierraLeone.Internationalarchivesofmedicine,4(1),p.15.considerableinefficiency–56%veryinefficient.
Kirigia,J.M.,Sambo,L.G.andLambo,E.,2015.ArepublichospitalsinKwazulu-NatalProvinceofSouthAfricatechnicallyefficient.Afr.J.HealthSci.
Kittelsen,S.A.,Winsnes,B.A.,Anthun,K.S.,Goude,F.,Hope,Ø.,Häkkinen,U.,Kalseth,B.,Kilsmark,J.,Medin,E.,Rehnberg,C.andRättö,H.,2015.DecomposingtheproductivitydifferencesbetweenhospitalsintheNordiccountries.JournalofProductivityAnalysis,43(3),pp.281-293.
KnaulF.M.,González-PierE.,Gómez-DantésO.,García-JuncoD.,Arreola-OrnelasH.,Barraza-LlorénsM.,SandovalR.,CaballeroF.,Hernández-AvilaM.,JuanM.,KershenobichD.,
65
NigendaG.,RuelasE.,SepúlvedaJ.,TapiaR.,SoberónG.,ChertorivskiS.,FrenkJ.,2012.Thequestforuniversalhealthcoverage:AchievingsocialprotectionforallinMexico.TheLancet380(9849):1259–1279.
Kniesner,T.J.,Viscusi,W.K.andZiliak,J.P.,2014.Willingnesstoacceptequalswillingnesstopayforlabormarketestimatesofthevalueofastatisticallife.JournalofRiskandUncertainty,48(3),pp.187-205.
Kringos,D.S.,Boerma,W.,vanderZee,J.andGroenewegen,P.,2013.Europe’sstrongprimarycaresystemsarelinkedtobetterpopulationhealthbutalsotohigherhealthspending.Healthaffairs,32(4),pp.686-694.
Kristensen,T.,Olsen,K.R.,Kilsmark,J.,Lauridsen,J.T.andPedersen,K.M.,2012.EconomiesofscaleandscopeintheDanishhospitalsectorpriortoradicalrestructuringplans.Healthpolicy,106(2),pp.120-126.
Kruk,M.E.,Porignon,D.,Rockers,P.C.andVanLerberghe,W.,2010.Thecontributionofprimarycaretohealthandhealthsystemsinlow-andmiddle-incomecountries:acriticalreviewofmajorprimarycareinitiatives.SocialScience&Medicine,70(6),pp.904-911.
Kurowski,C.,Wyss,K.,Abdulla,S.,Yémadji,N.andMills,A.,2004.Humanresourcesforhealth:requirementsandavailabilityinthecontextofscaling-uppriorityinterventionsinlow-incomecountries.CasestudiesfromTanzaniaandChad.HEFPworkingpaper01/04.LondonSchoolofHygieneandTropicalMedicine;London,UK.http://www.who.int/workforcealliance/knowledge/resources/priority_interventions_tanzaniachad/en/
Kutzin,J.,Yip,W.andCashin,C.,2016.Alternativefinancingstrategiesforuniversalhealthcoverage.Handbookofglobalhealtheconomicsandpublicpolicy,1.
Langenbrunner,J.,Xu,K.,Chu,A.,2016.DecentralizationinEastAsiaandPacific:MoreQuestionsthanAnswers.WHO(WPRO)PolicyNote,prepublicationdraft,July2016,Manila.
Lannes,L.(2015).AnanalysisofhealthservicedeliveryperformanceinRwanda(Doctoraldissertation,TheLondonSchoolofEconomicsandPoliticalScience(LSE)).http://etheses.lse.ac.uk/3093/1/Lannes_An_analysis_of_health_service_delivery.pdf
Larsen,M.,Rantala,R.,Koudenburg,O.A.andGulis,G.,2014.Inter-sectoralactionforhealth:TheexperienceofaDanishmunicipality.Scandinavianjournalofpublichealth,42(7),pp.649-657.
Lave,J.R.andLave,L.B.,1984.Hospitalcostfunctions.AnnualReviewofPublicHealth,5(1),pp.193-213.
Law,M.R.,Kratzer,J.andDhalla,I.A.,2014.TheincreasinginefficiencyofprivatehealthinsuranceinCanada.CanadianMedicalAssociationJournal,186(12),pp.E470-E474.
Laxminarayan,R.,Mills,A.J.,Breman,J.G.,Measham,A.R.,Alleyne,G.,Claeson,M.,Jha,P.,Musgrove,P.,Chow,J.,Shahid-Salles,S.andJamison,D.T.,2006.Advancementofglobalhealth:keymessagesfromtheDiseaseControlPrioritiesProject.TheLancet,367(9517),pp.1193-1208.
Laxminarayan,R.,Jamison,D.T.,Krupnick,A.J.andNorheim,O.F.,2014.Valuingvaccinesusingvalueofstatisticallifemeasures.Vaccine,32(39),pp.5065-5070.
deLeeuw,E.andPeters,D.,2015.NinequestionstoguidedevelopmentandimplementationofHealthinAllPolicies.Healthpromotioninternational,30(4),pp.987-997.
66
Leleu,H.,Moises,J.andValdmanis,V.,2012.Optimalproductivesizeofhospital'sintensivecareunits.InternationalJournalofproductioneconomics,136(2),pp.297-305.
Lewis,S.,2015.Asysteminnameonly—access,variation,andreforminCanada'sprovinces.NewEnglandJournalofMedicine,372(6),pp.497-500.
Li,Y.,Xu,J.,Wang,F.,Wang,B.,Liu,L.,Hou,W.,Fan,H.,Tong,Y.,Zhang,J.andLu,Z.,2012.OverprescribinginChina,drivenbyfinancialincentives,resultsinveryhighuseofantibiotics,injections,andcorticosteroids.HealthAffairs,31(5),pp.1075-1082.
Limwattanon,S.,Vongmongkol,V.,Prakongsai,P.,Patcharanarumol,W.,Hanson,K.,Tangcharoensathien,V.andMills,A.,2011.TheequityimpactofUniversalCoverage:healthcarefinance,catastrophichealthexpenditure,utilizationandgovernmentsubsidiesinThailand.ConsortiumforResearchonEquitableHealthSystems,MinistryofPublicHealth,http://r4d.dfid.gov.uk/Output/188980.
Limwattananon,S.,Neelsen,S.,O'Donnell,O.,Prakongsai,P.,Tangcharoensathien,V.,VanDoorslaer,E.andVongmongkol,V.,2015.Universalcoveragewithsupply-sidereform:TheimpactonmedicalexpenditureriskandutilizationinThailand.JournalofPublicEconomics,121,pp.79-94.
Lobstein,T.,Jackson-Leach,R.,Moodie,M.L.,Hall,K.D.,Gortmaker,S.L.,Swinburn,B.A.,James,W.P.T.,Wang,Y.andMcPherson,K.,2015.Childandadolescentobesity:partofabiggerpicture.TheLancet,385(9986),pp.2510-2520.
Loubiere,S.,Boyer,S.,Protopopescu,C.,Bonono,C.R.,Abega,S.C.,Spire,B.andMoatti,J.P.,2009.DecentralizationofHIVcareinCameroon:increasedaccesstoantiretroviraltreatmentandassociatedpersistentbarriers.Healthpolicy,92(2),pp.165-173.
Lu,Y.,Hernandez,P.,Abegunde,D.andEdejer,T.,2011.Theworldmedicinessituation2011.Medicineexpenditures.WorldHealthOrganization,Geneva.http://www.who.int/health-accounts/documentation/world_medicine_situation.pdf?ua=1
Maeda,A.,Araujo,E.,Cashin,C.,Harris,J.,Ikegami,N.andReich,M.R.,2014.UHCLessonsinHealthFinancing.InUniversalHealthCoverageforInclusiveandSustainableDevelopment:ASynthesisof11CountryCaseStudies,WorldBankPublications,WashingtonDC.http://elibrary.worldbank.org/doi/10.1596/978-1-4648-0297-3_ch5
Mackenbach,J.P.,Kulhánová,I.,Menvielle,G.,Bopp,M.,Borrell,C.,Costa,G.,Deboosere,P.,Esnaola,S.,Kalediene,R.,Kovacs,K.andLeinsalu,M.,2015.Trendsininequalitiesinprematuremortality:astudyof3.2milliondeathsin13Europeancountries.Journalofepidemiologyandcommunityhealth,69(3),pp.207-217.
Manzi,F.,Schellenberg,J.A.,Hutton,G.,Wyss,K.,Mbuya,C.,Shirima,K.,Mshinda,H.,Tanner,M.andSchellenberg,D.,2012.HumanresourcesforhealthcaredeliveryinTanzania:amultifacetedproblem.Humanresourcesforhealth,10(1),p.3.
Marmot,M.,2015.Thehealthgap:thechallengeofanunequalworld.BloomsburyPublishing,London.
Marmot,M.,Friel,S.,Bell,R.,Houweling,T.A.,Taylor,S.andCommissiononSocialDeterminantsofHealth,2008.Closingthegapinageneration:healthequitythroughactiononthesocialdeterminantsofhealth.TheLancet,372(9650),pp.1661-1669.
Mao,W.,Vu,H.,Xie,Z.,Chen,W.andTang,S.,2015.SystematicreviewonirrationaluseofmedicinesinChinaandVietnam.PloSone,10(3),p.e0117710.
67
Markovitz,A.A.andRyan,A.M.,2017.Pay-for-Performance:DisappointingResultsorMaskedHeterogeneity?.MedicalCareResearchandReview,74(1),pp.3-78.
Mason,H.,Shoaibi,A.,Ghandour,R.,O'Flaherty,M.,Capewell,S.,Khatib,R.,Jabr,S.,Unal,B.,Sözmen,K.,Arfa,C.andAissi,W.,2014.AcosteffectivenessanalysisofsaltreductionpoliciestoreducecoronaryheartdiseaseinfourEasternMediterraneancountries.PLoSOne,9(1),p.e84445.
McIntyre,D.,&Meheus,F.(2014).Fiscalspacefordomesticfundingofhealthandothersocialservices.CentreonGlobalHealthSecurity,ChathamHouse,WorkingGroupPaper.https://www.chathamhouse.org/publications/Papers/View/198263
Medeiros,J.andSchwierz,C.,2015.Efficiencyestimatesofhealthcaresystems(No.549).DirectorateGeneralEconomicandFinancialAffairs(DGECFIN),EuropeanCommission,EconomicPapers549.http://ec.europa.eu/economy_finance/publications/economic_paper/2015/pdf/ecp549_en.pdf
Meessen,B.,Soucat,A.,andSekabaraga,C.,2011.Performance-basedfinancing:justadonorfadoracatalysttowardscomprehensivehealth-carereform?BulletinoftheWorldHealthOrganization,89(2),153-156.
Meng,Q.,Fang,H.,Liu,X.,Yuan,B.andXu,J.,2015.ConsolidatingthesocialhealthinsuranceschemesinChina:towardsanequitableandefficienthealthsystem.TheLancet,386(10002),pp.1484-1492.
Meskarpour-Amiri,M.,Dopeykar,N.,Mehdizadeh,P.,Ayoubian,A.andMotaghed,Z.,2015.AStudyontheFactorsAffectingthePrescriptionofInjectionMedicinesinIran:APolicyMakingApproach.Globaljournalofhealthscience,7(3),p.291.
Mills,A.,2014.Healthcaresystemsinlow-andmiddle-incomecountries.NewEnglandJournalofMedicine,370(6),pp.552-557.
Mock,C.N.,Donkor,P.,Gawande,A.,Jamison,D.T.,Kruk,M.E.andDebas,H.T.,2015.Essentialsurgery:keymessagesfromDiseaseControlPriorities.TheLancet,385(9983),pp.2209-2219.
Montagu,D.andGoodman,C.,2016.Prohibit,constrain,encourage,orpurchase:howshouldweengagewiththeprivatehealth-caresector?.TheLancet,388(10044),pp.613-621.
Morell,A.L.,Kiem,S.,Millsteed,M.A.andPollice,A.,2014.Attraction,recruitmentanddistributionofhealthprofessionalsinruralandremoteAustralia:earlyresultsoftheRuralHealthProfessionalsProgram.Humanresourcesforhealth,12(1),p.15.
Moszynski,P.,2016.SouthAfricasrisingmaternalmortalityisduetohealthsystemfailuressaysreport.BMJ.BritishMedicalJournal,343,p.d5089.
Mulcahy,A.W.,Predmore,Z.andMattke,S.,2014.ThecostsavingspotentialofbiosimilardrugsintheUnitedStates.RandCorporation.https://130.154.3.8/content/dam/rand/pubs/perspectives/PE100/PE127/RAND_PE127.pdf
Niessen,L.W.andKhan,J.A.,2016.Universalaccesstomedicines.TheLancet,387(10013),pp.9-11.
Nshimirimana,D.A.,Mwaura-Tenambergen,W.,Kokonya,D.andAdoyo,M.,2016.AttitudesofPrimaryHealthCare(PHC)GatekeepersTowardsPatientReferralPolicy,MachakosCounty,Kenya.Science,4(4),pp.284-288.
68
Nguyen,L.,Bellucci,E.andNguyen,L.T.,2014.Electronichealthrecordsimplementation:anevaluationofinformationsystemimpactandcontingencyfactors.Internationaljournalofmedicalinformatics,83(11),pp.779-796.
Nyman,J.A.,1999.Thevalueofhealthinsurance:theaccessmotive.Journalofhealtheconomics,18(2),pp.141-152.
Ochodo,E.,Garner,P.andSinclair,D.,2016.Achievinguniversaltestingformalaria.BMJ,352,p.i107.
OECD,2017,TacklingWastefulSpendingonHealth,OECDPublishing,Paris.http://dx.doi.org/10.1787/9789264266414-en
Okafor,R.G.,2012.TaxrevenuegenerationandNigerianeconomicdevelopment.Europeanjournalofbusinessandmanagement,4(19),pp.49-56.
Okwero,P.,Tandon,A.,Sparkes,S.,McLaughlin,J.,&Hoogeveen,J.G.,2010.FiscalspaceforhealthinUganda.WorldBankWorkingPaper146.
Olesen,O.B.andPetersen,N.C.,2016.StochasticDataEnvelopmentAnalysis—Areview.EuropeanJournalofOperationalResearch,251(1),pp.2-21.
Ortegón,M.,Lim,S.,Chisholm,D.andMendis,S.,2012.Costeffectivenessofstrategiestocombatcardiovasculardisease,diabetes,andtobaccouseinsub-SaharanAfricaandSouthEastAsia:mathematicalmodellingstudy.Bmj,344,p.e607.
Palmer,S.andTorgerson,D.J.,1999.Definitionsofefficiency.BMJ:BritishMedicalJournal,318(7191),p.1136.
Pang,G.,2005.Efficiencyofpublicspendingindevelopingcountries:anefficiencyfrontierapproach.WorldBankPolicyResearchWorkingPaperNo.3645.https://papers.ssrn.com/sol3/papers.cfm?abstract_id=922917
Panter-Brick,C.,Eggerman,M.andTomlinson,M.,2014.Howmightglobalhealthmasterdeadlysinsandstriveforgreatervirtues?Globalhealthaction,7.
Parry,J.,andHumphreys,G.,2014.Healthamidafinancialcrisis:acomplexdiagnosis:theglobalfinancialcrisiscouldhaveprofoundimplicationsforthehealthspendingplansofnationalgovernments.Unlesscountrieshavesafetynetsinplace,thepoorandvulnerablewillbethefirsttosuffer.BulletinoftheWorldHealthOrganization,Jan.2009,p.4+.AcademicOneFile.
Patcharanarumol,W.,Tangcharoensathien,V.,Limwattananon,S.,Panichkriangkrai,W.,Pachanee,K.,Poungkantha,W.,Gilson,L.andMills,A.,2011.WhyandhowdidThailandachievegoodhealthatlowcost.Goodhealthatlowcost,25,pp.193-223.
Pereira,M.G.,Pedras,S.andMachado,J.C.,2014.Familyvariablesasmoderatorsbetweenbeliefstowardsmedicinesandadherencetoself-carebehaviorsandmedicationintype2diabetes.Families,Systems,&Health,32(2),p.198.
Perry,H.B.,Zulliger,R.andRogers,M.M.,2014."Communityhealthworkersinlow-,middle-,andhigh-incomecountries:anoverviewoftheirhistory,recentevolution,andcurrenteffectiveness."Annualreviewofpublichealth35:399-421.
Press,M.J.,Rajkumar,R.andConway,P.H.,2016.Medicare’snewbundledpayments:design,strategy,andevolution.Jama,315(2),pp.131-132.
Preyra,C.andPink,G.,2006.Scaleandscopeefficienciesthroughhospitalconsolidations.Journalofhealtheconomics,25(6),pp.1049-1068.
69
Rahman,M.H.,Agarwal,S.,Tuddenham,S.,Peto,H.,Iqbal,M.,Bhuiya,A.andPeters,D.H.,2014.Whatdotheydo?InteractionsbetweenvillagedoctorsandmedicalrepresentativesinChakaria,Bangladesh.Internationalhealth,p.ihu077.
Rantala,R.,Bortz,M.andArmada,F.,2014.Inter-sectoralaction:localgovernmentspromotinghealth.Healthpromotioninternational,29(suppl1),pp.i92-i102.25Localgovtsdointer-sectoral.Limitedevidenceofimpactincludingequity.
Rao,K.D.,2014.SituationanalysisofthehealthworkforceinIndia.HumanResourcesBackgroundPaper,1.http://uhc-india.org/uploads/SituationAnalysisoftheHealthWorkforceinIndia.pdf
Rao,K.D.,Petrosyan,V.,Araujo,E.C.andMcIntyre,D.,2014.ProgresstowardsuniversalhealthcoverageinBRICS:translatingeconomicgrowthintobetterhealth.BulletinoftheWorldHealthOrganization,92(6),pp.429-435.
Rawal,L.B.,Joarder,T.,Islam,S.M.S.,Uddin,A.andAhmed,S.M.,2015.DevelopingeffectivepolicystrategiestoretainhealthworkersinruralBangladesh:apolicyanalysis.Humanresourcesforhealth,13(1),p.36.
Reich,M.R.,Harris,J.,Ikegami,N.,Maeda,A.,Cashin,C.,Araujo,E.C.,...&Evans,T.G.,2016.Movingtowardsuniversalhealthcoverage:lessonsfrom11countrystudies.TheLancet,387(10020),811-816.
Reidpath,D.D.,Olafsdottir,A.E.,Pokhrel,S.andAllotey,P.,2012.Thefallacyoftheequity-efficiencytradeoff:rethinkingtheefficienthealthsystem.BMCPublicHealth,12(1),p.S3.
Renmans,D.,Holvoet,N.,Orach,C.G.andCriel,B.,2016.Openingthe‘blackbox’ofperformance-basedfinancinginlow-andlowermiddle-incomecountries:areviewoftheliterature.Healthpolicyandplanning,31(9),pp.1297-1309.
Reyniers,T.,Houttekier,D.,Pasman,H.R.,VanderStichele,R.,Cohen,J.andDeliens,L.,2014.Thefamilyphysician’sperceivedroleinpreventingandguidinghospitaladmissionsattheendoflife:afocusgroupstudy.TheAnnalsofFamilyMedicine,12(5),pp.441-446.
Roberts,M.J.,Hsiao,W.,BermanP.,Reich,M.R.,2004.GettingHealthReformRight:AGuidetoImprovingPerformanceandEquity,NewYork:OxfordUniversityPress.
Robyn,P.J.,Bärnighausen,T.,Souares,A.,Traoré,A.,Bicaba,B.,Sié,A.andSauerborn,R.,2014.Providerpaymentmethodsandhealthworkermotivationincommunity-basedhealthinsurance:amixed-methodsstudy.SocialScience&Medicine,108,pp.223-236.
Rosano,A.,Loha,C.A.,Falvo,R.,VanDerZee,J.,Ricciardi,W.,Guasticchi,G.andDeBelvis,A.G.,2013.Therelationshipbetweenavoidablehospitalizationandaccessibilitytoprimarycare:asystematicreview.TheEuropeanJournalofPublicHealth,23(3),pp.356-360
Roze,S.,Smith-Palmer,J.,Valentine,W.,Portu,S.,Nørgaard,K.andPickup,J.C.,2015.Cost-effectivenessofcontinuoussubcutaneousinsulininfusionversusmultipledailyinjectionsofinsulininType1diabetes:asystematicreview.DiabeticMedicine,32(11),pp.1415-1424.
Rudasingwa,M.,Soeters,R.andBossuyt,M.,2015.Theeffectofperformance-basedfinancialincentivesonimprovinghealthcareprovisioninburundi:acontrolledcohortstudy.Globaljournalofhealthscience,7(3),p.15.
Ryan,R.,Santesso,N.,Lowe,D.,Hill,S.,Grimshaw,J.,Prictor,M.,Kaufman,C.,Cowie,G.andTaylor,M.,2014.Interventionstoimprovesafeandeffectivemedicinesusebyconsumers:anoverviewofsystematicreviews.TheCochraneLibrary.
70
Sabbatini,A.K.,Nallamothu,B.K.andKocher,K.E.,2014.Reducingvariationinhospitaladmissionsfromtheemergencydepartmentforlow-mortalityconditionsmayproducesavings.HealthAffairs,33(9),pp.1655-1663.
Saini,V.,Brownlee,S.,Elshaug,A.G.,Glasziou,P.andHeath,I.,2017.Addressingoveruseandunderusearoundtheworld.TheLancet.http://dx.doi.org/10.1016/S0140-6736(16)32573-9
Saltman,R.B.,BankauskaiteV.,Vrangbaek,K.,2004.Decentralizationinhealthcare:strategiesandoutcomes.EuropeanObservatoryofHealthSystemsandPolicies.
Samadi,A.H.,Keshtkaran,A.,Kavosi,Z.andVahedi,S.,2013.Theeffectoffiscaldecentralizationonunder-fivemortalityinIran:apaneldataanalysis.Internationaljournalofhealthpolicyandmanagement,1(4),p.301.
Scott,J.,2014.RedistributiveimpactandefficiencyofMexico’sfiscalsystem.PublicFinanceReview,42(3),pp.368-390.
Seitio-Kgokgwe,O.,Gauld,R.D.,Hill,P.C.andBarnett,P.,2014.AssessingperformanceofBotswana'spublichospitalsystem:theuseoftheWorldHealthOrganizationHealthSystemPerformanceAssessmentFramework.IntJHealthPolicyManag,3:179-189,2014,DOI:10.15171/ijhpm.2014.85
Seow,H.,Brazil,K.,Sussman,J.,Pereira,J.,Marshall,D.,Austin,P.C.,Husain,A.,Rangrej,J.andBarbera,L.,2014.Impactofcommunitybased,specialistpalliativecareteamsonhospitalisationsandemergencydepartmentvisitslateinlifeandhospitaldeaths:apooledanalysis.Bmj,348,p.g3496.
Sheikh,K.andUplekar,M.,2016.WhatcanwelearnabouttheprocessesofregulationoftuberculosismedicinesfromtheexperiencesofhealthpolicyandsystemactorsinIndia,Tanzania,andZambia?.InternationalJournalofHealthPolicyandManagement,5(7),p.403.
Shrime,M.G.,Verguet,S.,Johansson,K.A.,Desalegn,D.,Jamison,D.T.andKruk,M.E.,2015.Task-sharingorpublicfinanceforexpandingsurgicalaccessinruralEthiopia:anextendedcost-effectivenessanalysis.DiseaseControlPriorities,ThirdEdition(Volume1):EssentialSurgeryMarch2015,pp:339–352.http://dx.doi.org/10.1596/978-1-4648-0346-8_ch19.
Siapka,M.,Remme,M.,Obure,C.D.,Maier,C.B.,Dehne,K.L.andVassall,A.,2014.IstherescopeforcostsavingsandefficiencygainsinHIVservices?Asystematicreviewoftheevidencefromlow-andmiddle-incomecountries.BulletinoftheWorldHealthOrganization,92(7),pp.499-511AD.
Sidibé,M.andCampbell,J.,2015.Reversingaglobalhealthworkforcecrisis.BulletinoftheWorldHealthOrganization,93(1),pp.3-3.
Simoens,S.,2012.AreviewofgenericmedicinepricinginEurope.GaBIJournal,1(1),pp.8-12.
Smith,O.,&Nguyen,S.N.,2013.Gettingbetter:improvinghealthsystemoutcomesinEuropeandCentralAsia.WorldBankPublications.
Sparkes,S.,Durán,A.andKutzin,J.2017.ASystem-WideApproachtoAnalysingEfficiencyacrossHealthProgrammes,HealthFinancingDiagnostics&GuidanceNo2,WHO,Geneva.
Spisak,C.,Morgan,L.,Eichler,R.,Rosen,J.,Serumaga,B.andWang,A.,2016.Results-basedfinancinginMozambique’scentralmedicalstore:areviewafter1year.GlobalHealth:ScienceandPractice,4(1),pp.165-177.http://www.ghspjournal.org/content/4/1/165.abstract
71
StallN.,NowaczynskiM.,SinhaS.K.,2013.Backtothefuture:home-basedprimarycareforolderhomeboundCanadiansPart1:wherewearenow.CanadianFamilyPhysician.Mar1;59(3):237-40.
Starfield.,B.,1992.PrimaryCare:Concept,Evaluation,andPolicy.NewYork,NY:OxfordUniversityPress.
Stuckler,D.,Feigl,A.,Basu,S.,&McKee,M.(2010).Thepoliticaleconomyofuniversalhealthcoverage.Backgroundpaperfortheglobalsymposiumonhealthsystemsresearch.http://www.pacifichealthsummit.org/downloads/UHC/the%20political%20economy%20of%20uhc.PDF
SunD,AhnH,LievensT,ZengW.,2017.Evaluationoftheperformanceofnationalhealthsystemsin2004-2011:Ananalysisof173countries.PLoSONE12(3):e0173346.doi:10.1371/journal.pone.0173346
Sweeney,S.,Obure,C.D.,Maier,C.B.,Greener,R.,Dehne,K.andVassall,A.,2011.CostsandefficiencyofintegratingHIV/AIDSserviceswithotherhealthservices:asystematicreviewofevidenceandexperience.SexuallyTransmittedInfections,pp.sextrans-2011.verticalaidsinefficient.
Tandon,A.,Fleisher,L.,Li,R.andYap,W.A.,2014.Reprioritizinggovernmentspendingonhealth:pushinganelephantupthestairs.WHOSouth-EastAsiaJournalofPublicHealth,3(3–4):206-212.
Tandon,A.,andCashin,C.,2010.Assessingpublicexpenditureonhealthfromafiscalspaceperspective.Worldbank.HNPDiscussionPaper,WorldBank.http://siteresources.worldbank.org/HEALTHNUTRITIONANDPOPULATION/Resources/281627-1095698140167/AssesingPublicExpenditureFiscalSpace.pdf
Tangcharoensathien,V.,Limwattananon,S.,Patcharanarumol,W.,Thammatacharee,J.,Jongudomsuk,P.andSirilak,S.,2015.AchievinguniversalhealthcoveragegoalsinThailand:thevitalroleofstrategicpurchasing.Healthpolicyandplanning,30(9),pp.1152-1161.
Teklehaimanot,A.,Kitaw,Y.,Girma,S.,Seyoum,A.,Desta,H.andYe-Ebiyo,Y.,2016.StudyoftheworkingconditionsofhealthextensionworkersinEthiopia.TheEthiopianJournalofHealthDevelopment(EJHD),21(3).
Thompson,F.,Beatty,K.andThompson,J.,2012.RankingStateTaxSystems:Progressivity,Adequacy,Efficiency(October12,2012).StateTaxNotes,Vol.141,No.7,2013.AvailableatSSRN:https://ssrn.com/abstract=2159486orhttp://dx.doi.org/10.2139/ssrn.2159486
Tiemann,O.,Schreyögg,J.andBusse,R.,2012.Hospitalownershipandefficiency:areviewofstudieswithparticularfocusonGermany.HealthPolicy,104(2),pp.163-171.Confirmsthatprivatehospitalsarenotnecessarilymoreefficientthanpublicingmy.
Torchia,M.,Calabrò,A.andMorner,M.,2015.Public–privatepartnershipsinthehealthcaresector:Asystematicreviewoftheliterature.PublicManagementReview,17(2),pp.236-261.
TransparencyInternational,2013.“GlobalCorruptionBarometer”www.transparency.org/gcb2013/report
Tucmeanu,A.I.,2014.Theeconomicsofhealthcaresystemreform.JournalofSelf-GovernanceandManagementEconomics,2(2),pp.72-77.
USAID,2009.Thevitalroleoftheprivatesectorinreproductivehealth,Washington,DC:USAID.
VanNguyen,K.,Do,N.T.T.,Chandna,A.,Nguyen,T.V.,VanPham,C.,Doan,P.M.,Nguyen,A.Q.,Nguyen,C.K.T.,Larsson,M.,Escalante,S.andOlowokure,B.,2013.Antibioticuseand
72
resistanceinemergingeconomies:asituationanalysisforVietNam.BMCpublichealth,13(1),p.1158.
VanYperen,N.W.,Wörtler,B.andDeJonge,K.M.,2016.Workers'intrinsicworkmotivationwhenjobdemandsarehigh:Theroleofneedforautonomyandperceivedopportunityforblendedworking.ComputersinHumanBehavior,60,pp.179-184.
Verguet,S.,2013.EfficientandequitableHIVprevention:AcasestudyofmalecircumcisioninSouthAfrica.CostEffectivenessandResourceAllocation,11(1),p.1.
Verguet,S.,Murphy,S.,Anderson,B.,Johansson,K.A.,Glass,R.andRheingans,R.,2013.PublicfinanceofrotavirusvaccinationinIndiaandEthiopia:anextendedcost-effectivenessanalysis.Vaccine,31(42),pp.4902-4910.
Verguet,S.,Laxminarayan,R.andJamison,D.T.,2015a.UniversalpublicFinanceoftuberculosistreatmentinIndia:anextendedcost-effectivenessanalysis.Healtheconomics,24(3),pp.318-332.
Verguet,S.,Olson,Z.D.,Babigumira,J.B.,Desalegn,D.,Johansson,K.A.,Kruk,M.E.,Levin,C.E.,Nugent,R.A.,Pecenka,C.,Shrime,M.G.andMemirie,S.T.,2015b.HealthgainsandfinancialriskprotectionaffordedbypublicfinancingofselectedinterventionsinEthiopia:anextendedcost-effectivenessanalysis.TheLancetGlobalHealth,3(5),pp.e288-e296.
Viscusi,W.K.,2015.Theroleofpublicationselectionbiasinestimatesofthevalueofastatisticallife.AmericanJournalofHealthEconomics,1(1),pp27-52.
Vogler,S.andKilpatrick,K.,2015.AnalysisofmedicinepricesinNewZealandand16Europeancountries.Valueinhealth,18(4),pp.484-492.
VonHaldenwang,C.,VonSchiller,A.andGarcia,M.,2014.Taxcollectionindevelopingcountries–Newevidenceonsemi-autonomousrevenueagencies(SARAs).JournalofDevelopmentStudies,50(4),pp.541-555.
Wagstaff,A.andLindelow,M.,2010.Arehealthshocksdifferent?Evidencefromamulti-shocksurveyinLaos.WorldBankPolicyResearchWorkingPaperNo.5335,https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1623713
Wang,X.,Liu,K.,You,L.M.,Xiang,J.G.,Hu,H.G.,Zhang,L.F.,Zheng,J.andZhu,X.W.,2014.Therelationshipbetweenpatientsafetycultureandadverseevents:aquestionnairesurvey.Internationaljournalofnursingstudies,51(8),pp.1114-1122.
White,M.T.,Conteh,L.,Cibulskis,R.andGhani,A.C.,2011.Costsandcost-effectivenessofmalariacontrolinterventions-asystematicreview.MalariaJournal,10(1),p.337.
Wilkinson,R.G.,1997.Socioeconomicdeterminantsofhealth.Healthinequalities:relativeorabsolutematerialstandards?BMJ:BritishMedicalJournal,314(7080),p.591.
Witter,S.,Wurie,H.andBertone,M.P.,2015.Thefreehealthcareinitiative:howhasitaffectedhealthworkersinSierraLeone?.Healthpolicyandplanning,p.czv006.
WorldBank,2013.PrivateHealthPolicyToolkit,Washington,DC:WorldBank.
WorldHealthOrganization,1997.Inter-sectoralactionforhealth:Acornerstoneforhealth-for-allinthetwenty-firstcentury.InReportonaninternationalconferenceheldinHalifax(pp.20-23).
WorldHealthOrganization,2000.TheWorldHealthReport2000.HealthSystems:ImprovingPerformance.Geneva,WorldHealthOrganization.
73
WorldHealthOrganization,2010a.TheWorldHealthReport2010.HealthSystemsFinancing:thePathtoUniversalCoverage,Geneva,WorldHealthOrganization.
WorldHealthOrganization,2010b.Administrativecostsofhealthinsuranceschemes:Exploringthereasonsfortheirvariability.HSFDiscussionPaperNo.8-2010.http://www.who.int/health_financing/documents/dp_e_10_08-admin_cost_hi.pdf
WorldHealthOrganization,2014.Healthinallpolicies:Helsinkistatement.Frameworkforcountryaction.Geneva:WHO.
WorldHealthOrganization,2016.PublicFinancingforHealthinAfrica:fromAbujatotheSDGsWHO/HIS/HGF/Tech.Report/16.2.http://apps.who.int/iris/bitstream/10665/249527/1/WHO-HIS-HGF-Tech.Report-16.2-eng.pdf?ua=1
WorldHealthOrganization,2017a.Cost-effectivenessandstrategicplanning.WHO-CHOICEhttp://www.who.int/choice/cost-effectiveness/en/
WorldHealthOrganization,RegionalOfficeoftheAmericas,2017b.Essentialpublichealthfunctions.www.paho.org/hq/index.php?option=com_topics&view=article&id=166&Itemid=40844accessed7February2017.
WorldHealthOrganization,RegionalOfficefortheEasternMediterranean,2017c.AssessmentofessentialpublichealthfunctionsincountriesoftheEasternMediterraneanRegion.www.emro.who.int/about-who/public-health-functions/index.htmlaccessed7February2017.
WorldHealthOrganization,2017d.Healthworkforce–Migration.http://www.who.int/hrh/migrationaccessed9February2017
WorldHealthOrganization,2017e.Achievingthehealth-relatedMDGs.Ittakesaworkforce!http://www.who.int/hrh/workforce_mdgsaccessed9February2017
WorldHealthOrganization,2017f.GlobalHealthObservatoryDataRepository.http://apps.who.int/gho/data/node.main.A1444?lang=en&showonly=HWFaccessed9February2017.
WorldHealthOrganization,2017g.Community-basedhealthworkers(CHWs)http://www.who.int/hrh/communityaccessed9February2017
WorldHealthOrganizationandtheWorldBank,2015.TrackingUniversalHealthCoverage:FirstGlobalMonitoringReport,Geneva,WorldHealthOrganization.http://apps.who.int/iris/bitstream/10665/174536/1/9789241564977_eng.pdf?ua=1
WorldHealthOrganization/InternationalNetworkforRationalUseofDrugs(INRUD)–WHO/INRUDdruguseindicatorsforhealthfacilitiesavailableat:http://apps.who.int/medicinedocs/en/d/Js4882e/8.3.html
Wurie,H.R.,Samai,M.andWitter,S.,2016.RetentionofhealthworkersinruralSierraLeone:findingsfromlifehistories.Humanresourcesforhealth,14(1),p.3.
Xiao,N.,Sharman,R.,Singh,R.,Singh,G.,Danzo,A.andRao,H.R.,2012.“MeaningfulUse”ofambulatoryEMR:Doesitimprovethequalityandefficiencyofhealthcare?.HealthPolicyandTechnology,1(1),pp.28-34.
Xu,Y.,Huang,C.andColón-Ramos,U.,2015.MovingTowardUniversalHealthCoverage(UHC)toAchieveInclusiveandSustainableHealthDevelopment:ThreeEssentialStrategiesDrawn
74
fromAsianExperience:Commenton'ImprovingtheWorld'sHealthThroughthePost-2015DevelopmentAgenda:PerspectivesfromRwanda'.IntJHealthPolicyManag.2015;4(12):869-872.doi:10.15171/ijhpm.2015.156brunner,J.,XuK.,andChu,A.,pre-publicationdraft,July2016,Manila.
Yazbeck,A.S.,Rabie,T.S.andPande,A.,2017.HealthSectorReformintheMiddleEastandNorthAfrica:ProspectsandExperiences.HealthSystems&Reform,3(1),pp.1-6.
Ye,J.,Zhang,J.,Mikolajczyk,R.,Torloni,M.R.,Gülmezoglu,A.M.andBetran,A.P.,2015.Associationbetweenratesofcaesareansectionandmaternalandneonatalmortalityinthe21stcentury:aworldwidepopulation-basedecologicalstudywithlongitudinaldata.BJOG:AnInternationalJournalofObstetrics&Gynaecology.
Yip,W.andHsiao,W.,2014.HarnessingtheprivatisationofChina'sfragmentedhealth-caredelivery.TheLancet,384(9945),pp.805-818.
Yip,W.andHafez,R.,2015.ImprovingHealthSystemsEfficiency.Geneva,WorldHealthOrganization.http://apps.who.int/iris/bitstream/10665/185989/1/WHO_HIS_HGF_SR_15.1_eng.pdf
Yousafzai,A.K.,Rasheed,M.A.,Rizvi,A.,Armstrong,R.andBhutta,Z.A.,2014.EffectofintegratedresponsivestimulationandnutritioninterventionsintheLadyHealthWorkerprogrammeinPakistanonchilddevelopment,growth,andhealthoutcomes:acluster-randomisedfactorialeffectivenesstrial.TheLancet,384(9950),pp.1282-1293.
Zachariah,R.,Ford,N.,Philips,M.,Lynch,S.,Massaquoi,M.,Janssens,V.,&Harries,A.D.(2009).TaskshiftinginHIV/AIDS:opportunities,challengesandproposedactionsforsub-SaharanAfrica.TransactionsoftheRoyalSocietyofTropicalMedicineandHygiene,103(6),549-558.
Zhang,X.,Zhao,L.,Cui,Z.andWang,Y.,2015a.StudyonequityandefficiencyofhealthresourcesandservicesbasedonKeyindicatorsinchina.PloSone,10(12),p.e0144809.
Zhang,S.,Zhang,W.,Zhou,H.,Xu,H.,Qu,Z.,Guo,M.,Wang,F.,Zhong,Y.,Gu,L.,Liang,X.andSa,Z.,2015b.HowChina’snewhealthreforminfluencesvillagedoctors’incomestructure:evidencefromaqualitativestudyinsixcountiesinChina.Humanresourcesforhealth,13(1),p.26.
Zhou,P.,Bundorf,M.K.,Gu,J.,He,X.andXue,D.,2015.SurveyonpatientsafetyclimateinpublichospitalsinChina.BMChealthservicesresearch,15(1),p.53.