搜索
EVALUATINGTHEHUMANENGINEERINGOFMICROPROCESSOR-CONTROLLEDOPERATINGROOMDEVICESRichardI.Cook,MD,~'pScottS.Potter,MA,~-DavidD.Woods,PhD,~-pandJohnS.McDonald,MD¢~CookRI,PotterSS,WoodsDD,McDonaldJS.Evaluatingthehumanengineeringofmicroprocessor-controlledoperatingroomdevices.JClinMonit1991;7:217-226ABSTRACT.Althoughhumanengineeringfeaturesarewidelyappreciatedasapotentialcauseofoperatingroomincidents,evaluatingthehumanengineeringfeaturesofdevicesisnotwidelyunderstood.Standards,guidelines,laboratoryandfieldtesting,andengineeringdisciplineareallproposedmethodsforimprovingthehumanengineeringofdevices.Newmicro-processortechnologyoffersdesignersgreatflexibilityinthedesignofdevices,butthisflexibilityisoftencoupledwithcomplexityandmoreelaborateuserinteraction.Guidelinesandstandardsusuallydonotcapturethesefeaturesofnewequipment,inpartbecausetechnologyimprovementsoccurfasterthanmeaningfulguidelinescanbedeveloped.Profes-sionalhumanengineeringofnewdevicesreliesonabroad,user-centeredapproachtodesignandevaluation.Usedintheframeworkofcurrentknowledgeabouthumanoperatorper-formance,thesetechniquesofferguidancetonewequipmentdesignersandtopurchasersandusersofthesedevices.KEYWORDS,Equipment:standards;computers.Measurementtechniques:computer;cognitiveengineering;humanfactors.Fromthe*DepartmentofAnesthesiologyandtheJ'CognitiveSys-temsEngineeringLaboratory,DepartmentofIndustrialandSystemsEngineering,TheOhioStateUniversity,Columbus,OH.ReceivedDec27,1989,andinrevisedformJul10,1990.AcceptedforpublicationJul30,1990.AddresscorrespondencetoDrCook,DeptofAnesthesiology,TheOhioStateUniversity,410W10thAve,Columbus,OH43210-1228.Qualityhumanengineeringofoperatingroomdevicesisaclearlydesirablegoal.Butwhatdoesitmeantosaythatadeviceiswelldesignedforahumanoperator?Howdoesthedesignerknowwhenthedesignisgoodortheuserknowthatthedeviceiswelldesigned?Theevaluationofhumanengineeringfeaturesisacomplexactivityconductedbyprofessionals.Theycallonexpe-riencewithclassicalhumanfactors,human-computerinteraction,andcognitiveengineeringandemploymanydifferenttechniquestoevaluatethenatureofde-vicesandsystems.Thesetechniquesspanawiderange.Atoneendistheapplicationofwrittencriteriaasastaticreviewofdevicedesign.Thecriteriamaybeformalstandardsormoreinformalchecklistsorguidelines,forexample,theAssociationfortheAdvancementofMed-icalInstrumentation(AAMI)HumanFactorsEngineeringGuidelinesandPreferredPracticesfortheDesignofMedicalDevices[1].Guidelinescontainlistsoffeaturesorele-mentsthatshouldorshouldnotbeincludedinorderforthedesigntobehumanfactored.Attheotherendofthescalearemethodsthatrelyheavilyontestinginlaboratoryorcontrolledfieldconditions,analysisofuserinteractionswiththedeviceunderrealisticcondi-tions,andinteractionsofthedevicewithotherdevicesintheenvironment.Alltechniqueshavelimitationsthatdependcriticallyonthenatureofthedeviceandthestageinthedesignprocessatwhichtheyareapplied.Overthepast10yearstechnologyhasshiftedthecharacteristicsofdevicesinmanyfieldsincludingtheoperatingroom.Manymicroprocessor-baseddevicesCopyright©1991byLittle,BrownandCompany217218JournalofClinicaIMonitoringVol7No3July199thaveappeared.Theseincludedrip-ratecontrollers,fluidpumps,awidevarietyofmonitors,diagnosticequip-ment(e.g.,transesophagealDopplerechocardio-graphs),ventilators,andpatientstimulators(e.g.,con-tinuouspassivemotiondevices).Somedevicesarenewandintrinsicallydependentonthemicroprocessor(e.g.,pulseoximeters)whileothersaremodificationsofnon-microprocessor-basedpredecessordevices.Thistechnologicallyinducedchangeshowsnosignsofabat-ing;indeed,thefirstgenerationofhighintegrationpa-tientmonitorsisnowbeinginstalledinhospitalsthroughoutNorthAmerica.Ourpurposeistodemonstratethelimitationsofchecklist-basedguidelinesandtopointtoothertech-niquesusedbyhumanfactorsprofessionalstoensurequalityhumanfactorsindevicedesign.Weperformedbothsortsofassessmentsonasmalloperatingroomdevice.Theseevaluationsproducelistsofhumanengi-neeringdeficiencies(HEDs),thatis,identifiablefeaturesthataremoreorlesslikelytoleadtoproblemsintheiruse.WeshowrepresentativeexamplesofthesortsofHEDsrevealedandsomeoftheirlikelyconsequencesforhumanusers.Wealsoprovidesomejustificationforconcludingthatthedifferencesbetweenthestylesofassessmentareimportantandtrytomapoutthebroaderspaceofhumanfactorsevaluations.Suchademonstrationrequiresadevice,preferablyasimpleonewithaneasilyunderstoodpurpose.ThetestdevicechosenforthisevaluationwastheMarquestSCT2000ServoControlledTrackingHeatedHumidificationSystem.InourhospitalsitreplacedtheBirdheatedhumidifiersthathadbeeninuseformanyyears.TheBirddevicesrequiredsubstantialsetupeffort,weredif-ficulttomaintain,andproducedannoyingcondensationintheproximalcircuitlimb(rainout).Thetimingofintroductionofthenewdeviceallowedustoobservethemanufacturer'sin-servicetrainingsessionandearlyaswellaslateruse.Inprinciple,thedeviceshouldbeeasytounderstandandstraightforwardtooperate.Thedeviceisalsousedindependentlyofotherdevicesand,exceptthatitisintheanesthesiacircuit,doesnothavemuchpotentialforinteractionwithotherdevices.Thedevicewasde