Problemática de la diferencia étnico – cultural en el Ecuador

initiative

evaluation

in

developing

countries

BarclayT. Stewarta,b,c,d,_{*, AdamGyedu}b,c_{,Robert Quansah}b,c_{,WilfredLarbi Addo}e_,

Akis Afokof,Pius Agbenorkub,c,Forster Amponsah-Manug,JamesAnkomahc, Ebenezer Appiah-Denkyirah_{,Peter Baffoe}i_{,Sam Debrah}j_{,Peter Donkor}b,c_,

TheodorDorvlog, KennedyJapiongk,Adam L. Kushnerl,m,n,MartinMornaj, AnthonyOfosuh,VictorOppong-Nketiao,StephenTabirie,p, CharlesMocka,q,r

a_{DepartmentofSurgery,UniversityofWashington,Seattle,WA,USA}

b_{SchoolofMedicalSciences,KwameNkrumahUniversityofScienceandTechnology,Kumasi,Ghana} c_{DepartmentofSurgery,KomfoAnokyeTeachingHospital,Kumasi,Ghana}

d_{DepartmentofInterdisciplinaryHealthSciences,StellenboschUniversity,CapeTown,SouthAfrica} e_{EasternRegionalHealthDirectorate,GhanaHealthService,Koforidua,Ghana}

f_{DepartmentofSurgery,TamaleTeachingHospital,Tamale,Ghana} g_{DepartmentofSurgery,EasternRegionalHospital,Koforidua,Ghana} h_{GhanaHealthService,Accra,Ghana}

i_{DepartmentofObstetricsandGynecology,UpperEastRegionalHospital,Bolgatanga,Ghana} j_{DepartmentofSurgery,UniversityofCapeCoast,CapeCoast,Ghana}

k_{DepartmentofEmergencyMedicine,PoliceHospital,Accra,Ghana} l_{SurgeonsOverSeas(SOS),NewYork,NY,USA}

m_{DepartmentofInternationalHealth,JohnsHopkinsBloombergSchoolofPublicHealth,Baltimore,MD,USA} n_{DepartmentofSurgery,ColumbiaUniversity,NewYork,NY,USA}

o_{DepartmentofAnesthesia,KomfoAnokyeTeachingHospital,Kumasi,Ghana} p_{DepartmentofSurgery,UniversityofDevelopmentStudies,Tamale,Ghana} q_{HarborviewInjuryPrevention&ResearchCenter,Seattle,WA,USA} r_{DepartmentofGlobalHealth,UniversityofWashington,Seattle,WA,USA}

ARTICLE INFO Articlehistory: Accepted12September2015 Keywords: Trauma Qualityimprovement Globalsurgery Developingcountry Ghana ABSTRACT

Introduction: Prospectiveclinicalauditoftraumacareimprovesoutcomesfortheinjuredinhigh-income countries(HICs).However,equivalent,context-appropriateauditfiltersforuseinlow-andmiddle- incomecountry(LMIC)district-levelhospitalshavenotbeenwellestablished.Weaimedtodevelop context-appropriatetraumacareauditfiltersfordistrict-levelhospitalsinGhana,waswellasother LMICsmorebroadly.

Methods:ConsensusontraumacareauditfilterswasbuiltbetweentwentypanellistsusingaDelphi techniquewithfouranonymous,iterativesurveysdesignedtoelicit:(i)traumacareprocessestobe measured; (ii) important features of audit filters for the district-level hospital setting; and (iii) potentiallyusefulfilters.Filterswererankedonascalefrom0to10(10 beingveryuseful).Consensus wasmeasuredwithaveragepercentmajorityopinion(APMO)cut-offrate.Targetconsensuswasdefined apriorias:amedianrankof_!9foreachfilterandanAPMOcut-offrateof_!0.8.

Results:Panellistsagreedontraumacareprocessestotarget(e.g.triage,phasesoftraumaassessment, earlyreferralifneeded)andspecificfeaturesoffiltersfordistrict-levelhospitaluse(e.g.simplicity, unassumingofresourcecapacity).APMOcut-offrateincreasedsuccessively:Round1–0.58;Round2– 0.66;Round3–0.76;andRound4–0.82.AfterRound4,targetconsensuson22 traumacareand referral-specificfilterswasreached.Examplefiltersinclude:triage–vitalsignsarerecordedwithin 15minofarrival(mustincludebreathingassessment,heartrate,bloodpressure,oxygensaturationif

* Correspondingauthorat:UniversityofWashington,DepartmentofSurgery,1959NEPacificSt.,SuiteBB-487,POBox356410,Seattle,WA98195-6410,USA. Tel.:+12065433680.

E-mailaddress:[email protected](B.T.Stewart).

ContentslistsavailableatScienceDirect

Injury

j ou rna l h ome p a ge : w ww . e l se v i e r. co m/ l oc a te / i n j ury

http://dx.doi.org/10.1016/j.injury.2015.09.007

Introduction

Injuriesareresponsiblefor5milliondeathsandincur52million disability-adjusted life years annually, comprising 15% of the globaldiseaseburden[1].Theurgencyofandresourcedeficiencies for trauma care improvement vary immensely between high- incomecountries (HICs),where10% ofthesedeathsoccur each year,andlow-andmiddle-incomecountries(LMICs),where90% occur[2,3].Nonetheless,certaintoolsfromtraumacarequality improvementinHICs,suchasprospectiveclinicalaudit,maybe usefultoLMICsworkingtoreduceavertabledeathanddisability duetoinjury[4,5].

Givencriticalresourcedeficiencies,traumacareimprovements shouldrelyonincreasingtheefficiencyandqualityofcareusing low-costinterventions,insteadofrelyingonabolusofresources [6,7].PreventabledeathpanelreviewsfromseveralLMICssuggest thatthere is an opportunitytoimprove traumacare outcomes through low-cost improvements in quality [8]. As example, a tertiaryhospital inBrazil foundthat 61% of in-hospitaltrauma deaths might have been preventable by meeting trauma care standards [9]. Similarly, a multidisciplinary panel review of traumaticdeathsata tertiaryhospitalinGhanafoundthat60% ofdeathswerepotentiallypreventable[10].Similarconclusions fromIranandPakistansuggestthatthesefindingsarenotisolated [11,12]. These studies imply that improving the processes of traumacarecanpreventdeathanddisability,evenincentreswith insufficientresources.

Inadditiontotrackingcrudeorrisk-adjustedoutcomes(e.g.in- hospitaldeath) or preventabledeath rates, traumacare canbe evaluatedbyauditfilters[13].Auditfiltersareroutinelytracked actions, processes or expectations of care that can be used to identifywhenstandardsarenotbeingmet[5,14].InHICs,routinely collectedauditfiltersareusedbytraumacentresandsystemsto evaluatecareefficiencyandquality[15].However,commonlyused filtersassumeahigh-levelofresources(e.g.sufficientandhighly trained personnel, timely access to advanced diagnostics) [16].Giventhattheassumedlevelofresourcesisnotappropriate formanyLMIChospitals,thesefiltersarenotapplicable[3].While context-appropriate filters have been used successfully for maternalhealth qualityimprovementinitiativesinLMICs,audit filtersusefulformonitoringandevaluatingtraumacareprocesses arenotwellestablished[4,17].

Toaddressthisgap,weaimedtousetheDelphitechniqueto develop consensus on trauma care audit filters that would be accurateproxiesof qualitytraumacare,simple tomeasureand feasibletocollectatdistrict-levelhospitalsinGhana.Bydoingso, theproposed filters couldbe used tobenchmark, monitor and evaluatetraumacareprocessesandqualityimprovementinitia- tivesatfirst-levelhospitalsinLMICsmorebroadly.

Methods Setting

Ghanaisaheavilyindebted,lower-middleincomecountryin

percapitaincomeofUS$1760[18].LikeotherLMICs,theburdenof injuryislarge[2].Theage-standardizedinjurydeathrateis58per 100,000persons,whichismorethan20%higherthanthatinHICs [2]. Most injured are brought to the hospital by commercial vehicles,oftenonlyafterpayment[19].Given longpre-hospital timesand lackof care,80%of traumadeathsoccur in thepre- hospital setting [20]. The injured that reach hospital care, particularly those injured in rural areas, almost uniformly encounter facilities without trained trauma teams or essential resources[3].District-levelhospitalsareusuallythefirstpointof healthcarecontactforinjuredpatients;themostseniorclinicianis usually a medical officer or a non-physician provider [21– 23]. Rarely are surgeons or physician anaesthetists available [24]. Nevertheless, many district-level hospitals offer some surgical services, namely caesarean section, and usually have between 50 and 100 beds [23]. As examples of trauma care resource deficiencies, the majority of district-level hospitalsin Ghanaareunabletoprovidebasicairwaysupport,transfuseblood, takeanX-rayorperformatraumalaparotomyformostpatientsin need,particularlyduringanemergency[7].Injuriesthatrequire morecomplexcarehavetobeidentifiedandreferredtoregionalor tertiaryhospitals,whichfacesimilarresourcechallenges[7].Re- ferral often requires prohibitively expensive pre-payment that prevents transfer [25]. Therefore, strengthening district-level trauma care may have greater than expected impact on the avertable death and disability from injury in Ghana given its importanceintheadhoctraumacaresystem[26].

Delphitechnique

WeusedtheDelphitechniquetodevelopconsensusonasetof useful district-level hospital trauma care audit filters from panellists with relevant expertise. The Delphi technique is an iterative,anonymous,data-drivensurveymethodthatfacilitates expert-group consensus building [27]. The four defining char- acteristicsoftheDelphitechniqueare:(i)anonymity;(ii)iteration; (iii) controlled feedback; and (iv) statistical group response [28].Thesetenetswereupheldthroughoutthestudy.

Panellistselection

Toensurethat thefinal traumacareauditfilters wereboth appropriatefor theGhanaiandistrict-level hospitalcontextand representedqualitytraumacare,weapproachedexpertsthatmet thefollowingcriteria:(i)Ghanaianprofessionalsorprofessionals that have significant experience with the Ghanaian healthcare system;and(ii)expertiseinanarearelevanttothedevelopmentof district-levelhospitaltraumacareauditfilters.Significantexperi- ence was defined as at least 1 year of workin a district-level hospitaland/orcurrentlyoverseeingtraumacareatadistrict-level hospital(s).Theseareasincludedtraumacare,clinicaldistrict-level hospitalexperience,anaesthesia,nursing,hospitaladministration, andhealthcarepolicyandevaluation(e.g.GhanaHealthService officials). Obstetricians have been collecting audit filters for maternalcareinLMICsformanyyears[17];thus,anobstetrician

available);circulation–alargeboreIVwasplacedwithin15minofpatientarrival;referral–ifreferralis activated,thereferringclinicianandreceivingfacilitycommunicatebyphoneorradiopriortotransfer. Conclusion: ThisstudyproposestraumacareauditfiltersappropriateforLMICdistrict-levelhospitals. GiventhesuccessesofsimilarfiltersinHICsandobstetriccarefiltersinLMICs,thecollectionand reportingofprospectivetraumacareauditfiltersmaybeanimportantsteptowardsimprovingcarefor theinjuredatdistrict-levelhospitalsinLMICs.

who met the aforementioned criteria were approached. All approached panellists participated in the Delphi process (i.e. participation rate 100%). The breakdown of the participants’ primaryareaofexpertisewasasfollows:

1.Traumacare–sixpanellists.

2.District-levelhospitalcare–sixpanellists. 3.Anaesthesia–onepanellist.

4.Nursing–twopanellists.

5.Healthcareadministration–twopanellists. 6.Healthcarepolicyandevaluation–twopanellists. 7.Obstetriccare–onepanellist.

Notethattheserepresentonlyprimaryareasofexpertise.Most panellistswereabletoprovideexpertiseinmorethanonearea (e.g. a surgeon who was also a healthcare administrator and teachestraumacarecoursestodistricthospitalstaff).

Surveymethods

Potential panellists were approached with an email that described the aims of the study, the Delphi technique and expectedoutputs.Thiswasfollowedbyatelephonecommunica- tiontoensuretheemailwasreceived.Potentialpanellistswere askedtorespondwithaconfirmatoryemailiftheywantedtotake part;allpotentialpanellistsresponded. Allcommunication was blinded;noneofthepanellistsknewwhomtheotherpanellists wereuntilconsensushadbeenmettoavoidsocialresponsebias. For each round, responses to open-ended questions were examinedusingacontentanalysisframework[29].First,qualita- tiveresponsesweregroupedintocategoriesbasedoncodesthat represented clustered responses. Then, categories were further refined into useful themes and described. Responses were triangulatedbetweenpanelliststoevaluatetheextentoftheme convergence.Particularly uniqueresponseswerealsodescribed andevaluatedbypanellistsinthesubsequentsurveyround.

Inadditiontoopen-endedquestions,panellistswereaskedto rankproposedauditfilterseachround.Filterswithamedianrank <7 out of 10 were not included in subsequent rounds. Target consensus(i.e.terminatingpoint fortheDelphitechnique)was definedapriorias:amedianrankof!9foreachproposedaudit filteron a scale from0 to10 and anaverage percent majority opinion(APMO)cut-offrateof!0.8.TheAPMOcut-offrateisa consensusmeasurethatiscalculatedbysubtractingagreements fromdisagreementsanddividingthedifferencebyallresponses; agreement wasdefinedas anauditfilterrank of!7 [28]. Asa sensitivity assessment,consensus wasalsomeasured using the coefficientofvariance perround andperfilter.By doingso,we couldevaluateconsensusbetweenroundswithouthavingtorely onourdefinitionofagreement(i.e.rank!7).SurveyMonkeywas usedfordatacollectionandStatav12(CollegeStation,TX,USA) wasusedfordataanalysis.

Surveyrounds

Round1consistedofthreeparts.Panellistswerefirstaskedto propose specific target processes tobe measured by the audit filters,aswellasimportantfeaturesof filterstobeusedin the district-hospitalsetting.Next,panellistswereaskedtorankalistof potentialfilters on a scale from 0to 10 (0wasuseless, 5 was somewhatusefuland10wasveryuseful).Lastly,panellistswere askedtoproposeauditfilterstoberankedduringthenextround. ThefindingsfromRound1werepresentedinRound2sothat each panellist could considerand evaluate process targetsand important features of audit filters offered by other panellists.

eachoftheproposedprocesstargetsandimportantfeatures.Next, filterswithamedianrankof!7fromthefirstround,aswellas those proposed by the panellists in Round 1, were ranked/re- ranked. Then, open-ended questions regarding highly ranked filters(i.e.medianrank!9)andlowrankedfilters(i.e.medianrank <7)wereaskedtofurtherunderstandcomponentsofbothvery usefulandlessusefulfilterfeatures.Panellistswereagaingivenan opportunitytoproposenewfiltersforthenextround.

Round 3 was designed to challenge successful filters. First, findings from the previous round were presented and an opportunitywasgiventosupport,modifyorrejectthefindings. Second, highly ranked filters from previous rounds and newly proposed filters from Round 2 were grouped into each of the respective district-level hospital-based trauma care categories identified asessentialtargetsby thepanellistsinRound 1(e.g. triage, airway, breathing, circulation, disability, exposure/burn, identificationofshock,earlyreferralofpatientsinneedofahigher- levelofcare,resuscitation,reassessment,outcome).Panelliststhen ranked the grouped filters side-by-side. By doing so, the most useful filter(s) in each trauma care category could be elicited. Additionally, open-ended questions followed each category of filters,whichaimedtoidentifymodificationsthatmightimprove thefilters.Lastly, panellistswereagaingivenanopportunityto proposenewfiltersforthenextround.

InRound4,allfilterswithmedianrank!7inpreviousrounds andnewlyproposedfiltersfromRound3wereranked/re-ranked within trauma care categories to force panellists to judge one against othersthat representedsame process.Targetconsensus was reached after Round 4 (i.e. median rank of !9 for each proposedauditfilterandanAPMOcut-offrateof!0.8).

Results

Importantprocessestomeasure

Processesthatemergedasimportanttargetsformonitoringand evaluation included: triage;componentsof theprimary assess- ment;earlyidentificationof shock;earlyreferral of patients in need of a higher-level of care or patients at high-risk for deterioration; resuscitation; reassessment; basic fracture man- agementandacompositemeasureofcare(Table1).

For hospitals with more resources (i.e. advanced district hospitals,regionalortertiaryhospitals)oranestablishedtrauma caresystem,anexpandedsetofauditfiltertargetprocesseswas agreedupon.Theseincludedpre-hospitalcare,advancedresusci- tation, injury diagnostics, timely surgical intervention, life- threateninginjuryspecificfilters,andrehabilitation.

Importantfeaturesofauditfilters

Panellistsagreedonanumberofimportantfeaturesofaudit filtersfordistrict-levelhospitals.Theseincludedfeaturesrelatedto their accuracy of process measurement, feasibility of data collection,applicability tothedistrict-levelhospitalsetting (i.e. acknowledgingdifferentialresourceconstrainsbetweenhospitals) and inclusiveness of facilities regardless of resources to aid comparison. Several particularly informative features are de- scribed(Table2).

Amongfeaturesmoststressedbypanellistsweresimplicityand feasibility, both withregards totheactions expected and data collectionmechanismitself.Forexample,panellistsdiscouraged theuseofcompoundfilters(i.e.if‘x’,then‘y’wasdone)orfilters that assumed a higher-than-average level of resources (e.g. functioning X-ray, focused assessment with sonography for trauma [FAST] scan training, ability to intubate or operate).

hospitalswithoutsurgicalcarecapacityistoquicklyidentifyand referpatientswhoareinneedofmoreadvancedtreatmentorat high-riskofdeterioration.Therefore,filtersthatrequiredsignifi- cantlongitudinaldatacollection(e.g.monitoringofurineoutput, detailsofpost-resuscitationorpost-operativecare),whileimpor- tant,werenotprioritized.Itwasalsoagreedthatfiltersshouldbe usefulforcomparing alldistrict-level hospitals;however,those with greater resources or an established trauma care system shouldbeableandencouragedtocollectagreaternumberofand/ ormoreadvancedfiltersattheirdiscretion.

Other considerations for auditfilters were agreedupon. For example,traumacareauditfiltersSHOULD:

"alignwithnationaltraumacareguidelines,aswellasinterna- tionallyacceptedstandardsfortraumacare;

"beusefulforallpotentialdistrict-hospitalproviders(e.g.nurses, non-physicianproviders,medicalofficers);

"rely on physical exam and serial assessment rather than diagnosticstudiesthatmayormaynotbeavailable(e.g.X-ray,

"beproxiesofqualitytraumacare,notcomprehensivechecklists. Additionally, panellists agreedthat traumacare auditfilters SHOULDNOT:

"requiredatatobecollectedonlow-riskorlow-acuitypatientsto streamline the care process (e.g. ambulatory patients, those triagedgreenusingtheSouthAfricanTriageScale)[30]; "inadvertentlycauseinexperiencedstafftoperformprocedures

beyond their scope of practice in an effort to increase the hospital’sauditscore,whichmightbemoredangerousthanearly referral(e.g.difficultintubation,traumalaparotomy);

"belinkedtoapatient’sabilityorinabilitytopay;

"neglect the importance of timing and effective triage, but appreciatethelargecase-loadsandinsufficientnumberofstaff tomanageallofthepatientsrapidly;and

"underestimatedistrict-levelhospitalcapabilitiesbysettingthe bartoolow;instead,thebarshouldbesetslightlyhigherthanthe currentsituationbutattainablewithlow-costqualityimprove-

Traumacareprocessesthatpanellistsagreedshouldbemonitoredbybasicandexpandedauditfilterswithexamplesfordistrict-levelhospitals.

Basic Expanded Examplefilter

Pre-hospitalcare PatientarrivedbyNationalAmbulanceService Timesinceinjury Patientarrivedwithin1hofinjury

Triage Vitalsignsarerecordedwithin15minofpatientarrival

(mustincludebreathingassessment,heartrate,blood pressure,oxygensaturationifavailable)

Componentsofprimary assessment

Examinationforpneumo-orhaemo-thoraxwasdone within15minofpatientarrivalbylisteningtobothsides ofthechestwithastethoscopeANDbilateralpercussion

Identificationofshock Ifdifficultybreathing,ORshockpresentattriage(HR