! 1! Science!Institute!!

Transcription

1 Electronic*Health*Record*(EHR)2enhanced*clinical*communication*and*quality2of2life* improvement*in*managing*menopausal*vasomotor*menopausal*symptoms*(vms)* Submittedby: TimothyDeVerDye,PhD ProfessorofObstetricsandGynecology,Pediatrics,PublicHealthSciences, andmedicalinformatics AssociateChairforResearch,ObstetricsandGynecology Director,BiomedicalInformatics ClinicalandTranslationalScienceInstitute UniversityofRochesterSchoolofMedicineandDentistry VoiceF(585) Abstract TheoverallgoalofthisprojectistoevaluatetheeffectivenessofanEHRFfacilitatedclinical interventiontohelpimprovecontrolofvasomotorsymptomsinmenopausalwomen.research objectivesinclude1)todevelop,build,test,andimplementanehrfbasedsetofactivitiesto facilitateclinicalcommunicationaroundmanagingvms,2)toassesstheimpactofehrfenabled VMStoolsonpractitionersandtheclinicalencounter,3)ToassesstheimpactofEHRFenabled patientengagementonclinicalinteractionsaroundmanagingtheirvms,and4)toimprove patienthealthfrelatedqualityoflife(qol)insymptomaticmenopausalwomenthroughbetter managementofvms.toaddressthesequestions,theprojectteamwilldesignandimplement arobust,clusterfrandomizedclinicaltrialtotesttheeffectofcombiningpatientfgeneratedvms symptominformationandqualityoflifeintousefulelectronicdashboardsforpatientsand practitioners,withpractitionerdecisionfaids(basedonbestpracticerecommendations)and electronichealthrecordtoolsthatfacilitatetreatmentactions.thestudydesignallowsfortwo experimentalarms(patientfonlyintervention,andpatientfpractitionerintervention),both comparedwithstandardcareatbaseline.theprojectishubbedwithinalargedepartmentof ObstetricsandGynecologyandwiththeBiomedicalInformaticsgroupoftheUniversity s acclaimedclinicalandtranslationalscienceinstitute.theprojecthastheintegralinvolvement ofthemedicalcenter senterprisefwideepicsystem(termedefrecord),guidedbyaproject teamofaccomplishedpractitioners,statisticians,epidemiologists,informaticists,and qualitativeresearchers.projectresultswillbedisseminatedlocallythroughbothpracticeand researchnetworks,andnationallyatscientificconferencesandinpeerfreviewedpublications.

2 Electronic*Health*Record*(EHR)2enhanced*clinical*communication*and*quality2of2life* improvement*in*managing*menopausal*vasomotor*menopausal*symptoms*(vms)* I.*Overall*Goal*and*Objectives TheoverallgoalofthisprojectistoevaluatetheeffectivenessofanEHR7facilitatedclinical interventiontohelpimprovecontrolofvasomotorsymptoms(vms)inmenopausalwomen. ThisgoalalignswithPfizer sinterestsinaddressingacaregapreflectedbythehighproportion ofwomenwhosemenopausalsymptomsarenegativelyimpactingtheirqualityoflifethough whoarenotpresentlyreceivingtreatmentforthesesymptoms.theproject sgoalalsoaligns wellwiththeuniversityofrochester snewstrategicplanthatemphasizesinnovation, populationhealth,widespreadintegrationoftechnologytoimprovecare,andimprovingvalue. TheDepartmentofObstetricsandGynecology,theacademichomeforthisproject,constitutes alargepracticegroupthatservesadiversepopulationofwomenexperiencingmenopausal symptoms,hasnationally7andinternationally7recognizedmenopausalresearchers,andhas clinicalchampionswithstrongvisibilityaroundmenopause7relatedissuesamongpractitioners andthecommunity.further,theuniversityofrochesterhashadalargeinstallationofepic (electronichealthrecordsystem)sincemarch2011withmechanismsforconductingthetypeof researchproposed.thatthebiomedicalinformaticsgroupatrochester sclinicaland TranslationalResearchInstitute 1 areinvolvedinallaspectsofthisprojectleveragessubstantial technicalresourcesinelectronichealthrecord7basedresearch,studydesign,and dissemination. The*objectives*of*this*research*include:* * 1)Todevelop,build,test,andimplementanEHR7basedsetofactivitiestofacilitateclinical communicationaroundmanagingvms, 2)ToassesstheimpactofEHR7enabledVMStoolsonpractitionersandtheclinicalencounter, 3)ToassesstheimpactofEHR7enabledpatientengagementonclinicalinteractionsaround managingtheirvms,and 4)ToimprovepatientHealth7relatedQualityofLife(QoL)insymptomaticmenopausalwomen throughbettermanagementofvms. I.a**Theoretical*Orientation Thisprojectisguidedbytwointersectinghealthbehaviortheoreticalmodels(seeFigure1). First,theclassicHealthBeliefModel(HBM)attemptstoexplainandpredicthealthbehaviors, byfocusingontheattitudesandbeliefsofindividuals(seenutbeametal.2012).inthisproject boththepatientandpractitionerareguidedbythehealthbeliefmodel:theyidentifyand 1NIHAward#5UL1TR (K.Kieburtz,PI)TheUniversityofRochester'sClinicalAndTranslational ScienceInstitute 1

3 assessrisks,benefits,andoptions,andthenactuponthem.awomanmustidentifyand recognizeanexperienceasa symptom andshe(andherpractitioner),inturn,makean assessmentofwhat,ifanything,shouldbedoneaboutit.thisprojectaimstohelpwomen betteridentifytheirsymptomswiththisinformation,whicharethenprovidedalongwith treatmentrecommendationstopractitionerssotheyarebetterabletotreatthem. Thisinteractionbetweenpatientandpractitionerishighlightedinasecondtheoretical orientation,socialcognitivetheory.socialcognitivetheorypositsthatbehaviorchangeoccurs duetoadynamicandreciprocalinteractionoftheperson,environment,andbehavior (Nutbeametal2012).Appliedtothisproject,byimprovingawoman s information environmentwithstructuredandeffectiveinformationprovidedthroughpublicandclinical sources,thatinformationandtheinvolvementoftheclinicianingeneratingtreatmentoptions shapesherbehavior(e.g.adoptingrecommendedtreatments). Theexperimentaldesignadoptedinthisprojectallowsformanipulationoftheinformation environmentinthreeways:first,bypackagingdetailedpatient7generatedsymptomandquality oflifedataintoactionabledashboardsforbothpatientsandforpractitioners;second,through enhancedinformationtargetingpatientneed(directedonlinehealtheducationmaterial)and practitionerneed(ehrdecision7toolsbaseduponbestpracticerecommendationsfromacog); andfinally,thebundledtreatmentoptionswillbebasedonelectronicalgorithmsincorporating thepatientsymptomprofileandpractitionerrecommendation. 2

4 II.*Technical*Approach* II.a.*Assessment*of*Need*for*the*Project* TheUniversityofRochesterMedicalCenter(URMC)isthelargesthealthcareproviderof inpatientandoutpatientwomen shealthcareinthefingerlakesregion,comprisinganetwork ofhospitals,primarycarepractitioners,andspecialists.becauseurmcusesepic(which accountsforthelargestmeaningfulusemarketshareamongehrprovidersintheusa), 2 the primarypopulationofinterestforthisparticularprojectisthepopulationservedbythe extendedurmcsystem.intheproposedpopulation,womenareseenprimarilyby obstetricians/gynecologists(68%)withremainderbeingservedbyfamilymedicineandinternal Medicinepractitioners. Table*1.*Estrogen*prescription*in*women*with*symptomatic*menopausal*climacteric*states,* by*age,*urmc*outpatient*population*march*11* *June*2014* * Age* (years)* ICD9*627.2**Symptomatic*menopausal* climacteric*states** Estrogen/* Combination* Prescribed* Percent* Prescribed* % % *Source:UniversityofRochesterCTSI,i2b2QueryandAnalysisToolappliedtoURMC se7recordpopulation (51,904outpatientwomenage45754and52,268outpatientwomenage55764) ThisprojectadoptstheAmericanCollegeofObstetriciansandGynecologists(ACOG)2014 PracticeBulletinonManagementofMenopausalSymptomsasoutliningbestpractice,which includesbothhormonalandnon7hormonaltreatments. ShowninTable1,overllonly15.4%ofURMCoutpatientswithdiagnosedsymptomatic menopausalorfemaleclimactericstateswereprescribedestrogen.asmentionedintherfafor thisinitiative,thispatternmaywellillustrateanimportantgapincareforwomenwhomay benefitfromhormonaltherapy.thisdatasuggests,perhaps,overallunderuseofmedical treatmentforvmssymptoms,thoughinsufficientdataexiststoconfirmthatinterpretation. Therefore,thefirstphaseofthisprojectwillbecollectbaselinedataregardingVMSsymptoms, qualityoflife,andclinicalpractice,providingabaseforcomparisonoftheproposed intervention.notethattheclinicaldiagnosisofapproximately2,000womenwithvms symptomsarefromanoverallpopulationofmorethan100,000outpatientwomenintheseage groups.thediagnosiscodemaywellindicateanundercountofwomenpresentingwiththese symptoms.thebaselineperiodwillhelptheteamassessthispotentialundercountincoding. 2 EPICisthetopElectronicHealthRecordsoftwareforallcategoriesofmeaningfuluse(see EHRMeaningfulUse MarketShareIndustryView March2014 availableat 3

5 TargetPopulation.Thetargetpopulationforthisstudyincludeswomenage45764without surgically7inducedmenopauseandwithoutknownmedicalhistorythatiscontraindicatedfor carestrategiesoutlinedinacog s2014practicebulletinonmanagementofmenopause Symptoms.Informedwrittenconsentwillbeobtainedintwostages,firstbyenrollingURMC7 associatedob/gyn,internalmedicine,andfamilymedicinepracticesintothestudy,and secondlybyelectronicallynotifyingallwomenwithscheduledprimarycareappointmentsin thosepracticesofthestudyandobtaininginformedconsent.becauseconsentwillbeobtained electronicallythroughmychart,andsincetheinterventionisdeployedinpartthrough MyChart,thestudywillmaynotberepresentativeofnon7MyChartusers(whocouldhave lowerhealthanddigitalliteracy,orperhapsinternetaccess,thanmychartusers).the UniversityofRochester sresearchsubjectsreviewboard(rsrb)willreviewandapprovethe projectpriortorecruitment. II.b.*Project*Design*and*Methods* II.b.i.Overallstudydesign:Thestudyisdesignedasaclusterrandomizedtrialtoinvestigatethe effectsoftwointerventions,apatient7centeredinterventionandprovider7centered intervention,onqualityoflifeinsymptomaticmenopausalwomenandtocomparethese interventionswithstandardcareidentifiedinthebaselinephase.thisdesignwillbesupported byqualitativeformativeresearchandaqualitativeprocessevaluationtohelpfacilitateinsight andengagement.thegeneralstudydesignispresentedinfigure2andtheoverallflowof interventionispresentedinfigure3. 4

6 II.b.iiRationale ArandomizedtrialisnecessarytoestablishwhetherornotanEHR7enhancedpractitioner7 targetedinterventionoffersadvantagesover1)standardcaremeasuredatbaseline,or2)an interventiontargetingonlypatienteducation.becausepatientsclusterinpractices,theonly optiontoeliminateanerroneouspractice7effectistoconductaclusterrandomizedtrial.also, becauseofthelogisticsoftheinterventions,practitionerscannotconceivablydelivercarein oneway(informedbyehr7enhancements)toonepatient,andinadifferentway(without enhancements)toadifferentpatient.thusthemodeloftheclusterrandomizedtrialwith practiceastheunitofrandomizationistheoptimaldesign.further,theprojectteamdecided thatsymptom7appropriatehealtheducationinformationiswarrantedforallparticipantsand shouldbepartofstandardcare.so,allparticipantswillformtheirowncontrolduringthe baselineperiod(nointerventions)andthenallwillreceivethepatientintervention.the experimentalarmwillthen,inadditiontothepatientintervention,receivethepractitioner (EHR)intervention. II.b.iiiInterventionDescription Thetwointerventionarms(Figure2)include:1) PatientIntervention only(includespatient symptomandqualityoflifeassessments,feedbackthroughadashboardtopatients,andaccess totargetedpractitioner7vettededucationalmaterials),and2)thepatientinterventionplusthe PractitionerIntervention (practitionerhasaccesstoehr7enhancedcaresupportswithpatient assessmentdashboardstailoredforclinicaluse,toolstofacilitateuseofacogpracticebulletin, andpre7plannedordersanddocumentation).allpatientsseekingcarefromanenrolled practitionerwillreceivetheinterventionappropriatetothegroupforwhichthatpractitioner wasassigned.theinterventionflowisshowninfigure3. PatientIntervention.Patientswillbeaskedtocompleteatstandardintervalsandbeforeclinical visitstwoinstruments:thegreeneclimactericscale(gcs)andtheutianqualityoflife Questionnaire(UQOL). TheGreeneClimactericScaleisoneofthemostwidelyusedscalesformeasuringmenopausal symptoms,andhasdemonstratedexcellentreliabilityandvalidity(greene1998).thisvms symptommeasureconsistsof21itemsscoredona47pointlikertscale(0,notatall;3 extremely),thatmeasurefourcategoriesofsymptomsa)psychologicalsymptoms;b)somatic symptoms;c)vasomotorsymptomsandd)sexualdysfunction. TheUtianQualityofLifeQuestionnaireisacommonlyusedtoolforclinicalresearchand practiceinperimenopausalwomenandhasdemonstratedhighreliabilityandvalidity(utianet al2002).thismeasureconsistsof23itemsscoredona57pointlikertscale(1,nottrueofme;5, verytrueofme)thatmeasurefourcomponentsofqualityoflife(qol):occupationalqol, healthqol,emotionalqol,andsexualqol. 5

7 Figure 3: Intervention Flow Patient Researcher Practitioner Sends out consent to qualified woman Send out GCS & UQOL Questionnaire Series Receives questionnaire series in baseline & every 3 months Questionnaire appears immediately in In Basket for review prior to visit Complete the questionnaire Review symptoms in EHR at beginning of visit Patient counseling is given by Practitioner Key: GCS Greene Climacteric Scale UQOL Utian Quality of Life Scale BPA Best Practice Advisory (an active decision support alert) ACOG The American College of Obstetricians and Gynecologists No Views questionnaire results in patient dahsboard Dashboard of patient s GCS & UQOL presented to Practitioner Are symptom specific recommendations needed? Yes Reviews results with patient VMS Risk driven targeted education BPA w/ ACOG Management of Menopause symptom recommendations/ counsel the patient Places orders and documentation through prepopulated list 6

8 Theinformationfromthesetwoscaleswillbesummarizedbacktowomenonadashboard summarizingandhighlightingtheirresponses,inadditiontopresentingtheindividual componentsofthescales.onlinehealtheducationmaterialsthathavebeenvettedbythe project sclinicalteamtoassurequalityandconsistencywiththeacogpracticebulletinon ManagingMenopausalSymptomswillbelinkedwithwomen sscoresonthetwoscales.as such,therecommendationstheyreceivewillbetargetedtotheirvmssymptomandqol profiles.theactualcontentandpresentationofthesedashboardsandfeedbackmechanisms willbedevelopedandtestedinthequalitativeformativestageoftheprojecttoassureusability andparticipantacceptability. PractitionerIntervention.Thepractitionerinterventionwillincludeseveralcomponents(Figure 3).First,practitionerswillhaveaccesstothewoman svmssymptomandqualityoflife informationdetail(question7specific)withine7record(epic).second,informationwillbe summarizedinacaredashboardforthepractitionertohelpthemidentifypotentialconcerns, andtointerpretandactupontheinformation.third,practitionerswillhaveeasyaccessto VMS7specificEHR7actions(e.g.,automaticdocumentation,pre7plannedordersets,prescription sets)thatenhanceadherencewiththeacogtechnicalbulletin.aswiththepatient Intervention,thedesignandusabilitydetailsofthePractitionerInterventionwillemergefrom thequalitativeformativephasethatinvolvespractitioners.note,giventheexperimental design,thatallenrolleesreceivethepatientintervention,butonlyenrolleeswhoarepatients ofpractitionersinthepractitionerinterventionarmwillreceivecarethatresultsfromclinical EHR7enhancedactivities. II.b.ivAnalyticPlan PowerCalculation.Theproposedstudywillenrollpatientsfrom12715practices.These practiceswillberandomlyassignedtothetwointerventionarmsasdescribedpreviously, stratifiedbysizeandpracticetype.forbotharms,allpatientswillreceivestandardcareduring thefirstphase.duringthesecondphase,arm1willreceivethepatientinterventiononlyand Arm2willreceivethepatientinterventionalongwiththepractitionerinvention.Patientswill berecruitedinthestudyduringbothphases.thoseenrolledinthefirstperiodwillbefollowed upeitheruntilcompletionofthestudyoruntildropout,whicheveroccursfirst.with200 subjectsenrolledineachphaseacrosspracticesassignedtothearminwhichpatientsreceive thefirstinterventionduringthesecondphase,thestudywillhavemorethan80%powerto detectastandardizeddifference(cohen seffectsize)of0.3intheuqolchangerelativeto baseline(see Statisticalmethods sectionbelowforadefinitionofthisoutcomevariable) usingatwo7sidedt7testwitha5%significancelevel.weexpectthestudytohavehigherpower orbeabletodetectsmallereffectsizebecausechangeinuqolwillbemeasuredatseveral timepoints.randomizationofpracticeswillbestratifiedbypracticesizetoassurecomparable groups. Statisticalmethods.Theprimaryobjectiveofthestatisticalanalysisistodeterminewhether eitherofthetwointerventionsthatwillbetestedimprovesqualityoflifeinsymptomatic menopausalwomenthroughbettermanagementofvmscomparedtostandardofcare.the primaryoutcomevariablethatwewillusetocompareinterventionswillbetheuqolscore.in 7

9 allstatisticalanalyses,wewilltreatthisscoreasacontinuousvariable.foreachpatient,the UQOLscorewillbedeterminedatbaselineandduringsubsequentassessments.Weexpectthat theseassessmentswilloccureverythreemonthsonaverage.atbaseline,theuqolscorewill bedeterminedbeforeintervention.thus,becauseofrandomization,thereshouldbeno differenceinuqolatbaseline,regardlessofphaseandinterventionreceived.ourprimary statisticalanalysiswillfocusonchangeinuqolrelativetobaseline,measuredrepeatedlyover timeineachpatientanddescribedusingmulti7levellinearmixedeffectsmodelstoaccountfor within7patientandwithin7practitionerdependencies.thesemodelswillincludethefollowing covariates(oranysubsetofthese,ifdeemedmoreappropriate)asfixedeffects:baseline UQOL,thetypeofinterventionreceivedwhenUQOLwasdetermined(athree7levelcategorical variable),timefrombaseline(definedasthedifferencebetweenthetimeatwhichuqolwas evaluatedandthetimeatwhichthepatiententeredthestudy).totestwhethertherateof improvementdiffersbypracticeorarm,wewillalsoconsiderincludinginteractionterms(the mostimportantonesbeingtime practiceandtime treatment).otherinteractionswillbe considered,forexample:time timeandtime treatment practice.twolevelsofrandom effectswillbeintroducedinthemodel.onelevelwilldescribepotentialdependenciesbetween observationsperformedinthesamepatient.thiswillbeaccomplishedusingpatient7specific randominterceptsandrandomcoefficientsforthevariabletime.thesignificanceofthese randomeffectswillbeassessedusingalikelihoodratiotestontheassociatedvariance components,accountingforthefactthatvariancesareconstrainedtobepositivewhen determiningtheasymptoticdistributionofthetest.asecondlevelofrandomeffectswill describepotentialdependenciesinducedbypractitioners.ourplanistoincludetheserandom effectsaspractitioner7specificrandomintercepts.modelingassumptionswillbethoroughly checked.forexample,theassumptionsofnormalityoftheerrortermsandoftherandom effectswillbeexaminedbyinspectingthedistributionoftheconditionalandunconditional residualsusingnormalprobabilityplots.scatterplotofresidualswillbeexaminedtoassessthe linearityassumption.ifanyassumptionappearsviolated,wewilleithermodifythemodelor transformthedata.weexpecttheprobabilityofanobservationtobemissingtoincreaseover timeduetoattrition.wewillinvestigatewhetherothervariablesisassociatedwiththe probabilitythatanyobservationmisses.onceanappropriatemodelwillhavebeendeveloped, wewillcompareinterventionsusinglikelihoodratiotests.pointestimatesofmodel parameterswillbereportedalongwiththeirstandarderrorsand95%confidenceintervals.all significancetestswillbetwo7sided(withtheexceptionofthelikelihoodratiotestsforvariance components)andconductedatthe5%significancelevel.allstatisticalanalyseswillbe conductedusingsas,rormatlab. III.*Evaluation*Design* * Asdescribedpreviously,thisstudyisprimarilypoweredtodetecta30percentchangeinUQoL andsecondarilypoweredtodetectchangeinvmsinmenopausalwomen.whilethemain studyisaclusterrandomizedtrial,itissupplementedbyasubstantialqualitativeandformative componenttoobtainpractitionerandpatientinputintothedesignoftheintervention.further, patientsandpractitionersarequeriedregularlythroughoutthestudytoascertaintheimpactof theinterventionontheircareparameters(lengthofvisits,documentationrequirements,etc). 8

10 Thestudyteamincludesexpertsinstatisticaldesign,epidemiology,clinicalcare,andbiomedical informatics.additionaldetailregardingmeasurementandevaluationoftheproject sobjectives follows. III.a.***Overview*of*Outcomes*Evaluation* Objective*1:Todevelop,program,test,andimplementanEHR>basedsetofactivitiesto facilitateclinicalcommunicationaroundmanagingvms. EHRBuild:AdaptingDarer sapproachtoehrinnovationinclinicalcare(darer2014),thecore interventionofthisprojectisaclusterofactiontoolsthatsupporteliminationofcaregaps(see Silow7Carroll2012).Todothis,wewillcreateaninterventionthatstartswithasetofstructured questionnairesbuiltintomychartthat:1)obtainsconsentfromeligiblewomen,2)implements thegreeneclimactericscale(gcs)toassesssymptomatology(seegreene1998),and3) implementstheutianqualityoflifescale(abrahametal2014).automatedalgorithmswill convertgcsanduqoldataintovisualdashboardsaccessibletopatientsandpractitioners alongwithpracticerecommendationsappropriatetothegcsanduqolinterpretationsbased upontheacogmanagementofmenopausalsymptomspracticebulletin(2014).notationsof thegcsanduqolresultswillbeplacedinanappropriatedocumentationlocation,triggering practitionerstoaddresstheseissuesatthenextvisit.womenwillhaveaccesstotheirown resultsofthegcsanduqolthroughmychart,anduponcompletionandenrollmentwill receivelinkagestoacogandmenopausesocietypatient7informationsources,universityof RochesterPeriFactsBulletinsaddressingmenopause,andguestcolumnsonmenopausefrom RochesterWomenauthoredbyco7investigatorDr.JamesWoods.Attheclinicalencounter, practitionersintheexperimentalgroupwilldiscussthewoman sresultstothegcsanduqol andwillpresentappropriatetreatmentrecommendationsfromtheacogpracticebulletin. Practitionerswillinitiateappropriateautomatedordersfortreatmentbasedontheirclinical encounterwithwomen.basicfieldswillbeincludedintheepicrecordtodocumentfidelityin implementationoftheinterventionasappropriatetothepractitioner sassignedgroup. ProviderTraining:Allpractitionersparticipatinginthisprojectwillparticipateintwoone7hour trainingsessionsorganizedbydr.dyeanddr.woods.thecontentofthisprovidereducation willincludediscussionofriskassessmentandmeasurementusingmrsanduqol,acog PracticeBulletintreatmentrecommendations,andPatientCommunicationaboutMenopause. TheEHRbuildwillbetestedusingusabilitytestingtechniquesandwillbeiterativelydeveloped untiltheprojectteamreachesconsensusonthetools,presentationdashboards,andpractice recommendations.afterthebuildconsensusisreached,theinterventionwillbetestedwith non7participatingpractitionerstoidentifyfinalusabilityissuesandissuethefinalbuild. Objective1,qualitativecomponent:Thesuccessoftheinterventiondependsinlarge partonpractitionerwillingnesstoengageinthepatient sgcsanduqolgenerated recommendations,documentation,andautomatedordersintheehr.tomaximize practitionerengagementandimprovetheehrbuild,semi7structuredfocusgroupswill beconductedwithpractitionersduringthefirstphaseoftheproject.wewillusea 9

11 purposefulsampleofpractitionersacrossavarietyofpracticesandagesuntil redundancyisreached.practitionerswillfirstbeaskedopen7endedquestionsabout theirattitudesregardingintegratingintotheehrthepatientsurveysandagog recommendations.theywillthenbeaskedspecificallyabouttheirreactionsto prototypesoftheehrbuild.asimilarapproachwillbetakenwithfocusgroupsoftarget patients,exceptwithafocusonrepackagingthemeasurementtoolresults(green ClimactericScaleandtheUtianQualityofLife),focusinguponwomen sinterpretation andunderstandingoftheresults,andcommentsoneducationsourcestowhichthey willbereferred.atrainedqualitativeresearcherwillconductthefocusgroupwitha researchassistanttakingnotes.allfocusgroupswillberecordedwiththepermissionof participants.theaudiotapeofthediscussionsandtheparticipantobserver snoteswill beanalyzedusingtheconstantcomparativemethodbytwoinvestigators(krueger 1994). Objective*2:ToassesstheimpactofEHR>enabledVMStoolsonpractitionersandtheclinical encounter. Practitionersinpracticesparticipatinginthisstudywillreceiveabriefgeneralizedautomated surveyintheirepicin7basketatbaselineandeverythreemonthsassessingtheirexperience withprovidingclinicalcaretowomenaroundvmsthatparticularlyassesseslengthofvisits, timespentansweringquestionsforpatientsoutsideofclinicalencounter,timespentwritingup notes,andtransparencyofdocumentation(seedarer2012,delbancoetal2010). Objective2,qualitativecomponent:Basedonpractitionersurveyresponses,a purposefulsub7sampleofpractitioners(varyinginengagementandsatisfactionwith intervention)fromtheehrenhancedarmwillberecruitedtoparticipateinsemi7 structuredfocusgroups.theobjectiveofthisprocessevaluationistobetterunderstand howpractitionersinteractwiththeinterventionandiftheyfinditahelpfuladditionto theircurrentstandardofcare.thefindingsfromthisanalysiswillbeusedtobetter understandtheoverallfindingsoftheinterventionandhowandiftheinterventioncan bedisseminatedorimproved.atrainedqualitativeresearcherwillconductthefocus groupswitharesearchassistanttakingnotes.allfocusgroupswillberecordedwiththe permissionofparticipants.theaudiotapeofthediscussionswillbetranscribedand coded.twoinvestigatorswillindependentlyandsystematicallyreviewedtranscriptsto identifythemesandsub7themesuniqueandsimilaracrossallgroups.systematiccoding scalesincludedfrequency(thenumberoftimesthatthetopicappearsintheanalysis), convergence(whetherthetopicextendsacrosssubjectclassificationgroups),and intensity(theemotionandimportanceoftopictothespeaker)ofthedataelements. Usinganiterativeprocess,thesethemeswillbereviseduntilaconsensusisachieved amongthetwoinvestigators.wewillusenvivo,aqualitativedataanalysissoftware,to manuallyrecordandcomparecodingoftranscripts. * * 10

12 Objective*3:ToassesstheimpactofEHR>enabledpatientengagementonclinicalinteractions aroundmanagingtheirvms Patientsinbothgroupswillbetriggeredeverythreemonthsfrombaselinetoassesswhetheror notthey:understoodtheirhealthandmedicalconditionsbetter,tookbettercareof themselves,rememberedtheplanfortheircarebetter,werebetterpreparedfortheirvisits, feltmoreincontroloftheirhealthcare,weremorelikelytotakemedicationsprescribed(e.g. seedarer2014,delbancoetal2010). Objective3,qualitativephase.Apurposefulsampleofpatientsfrombothstudyarms willberecruitedtoparticipateintelephoneinterviewsabouttheirexperience.the objectiveofthisprocessevaluationistobetterunderstandpatients experiencesfilling outthequestionnaires,theirclinicalinteractionwiththeirphysician,and(forthosein thepractitionerinterventionarm)iftheybelievetheinterventionimprovedtheirquality ofcare.thefindingsofthisqualitativeanalysiswillprovideimportantinsightintothe perceivedimpactoftheinterventionfromthepatient sperspective. Objective*4:ToimprovepatientHealth>relatedQualityofLifeinsymptomaticmenopausal womenthroughbettermanagementofvms.patientsenrolledwillbetriggeredtorepeatthe GCSandUQoLeverythreemonthsfrombaseline,todocumentbothabsolutechangefrom baselineandrelativechangefromeachquarter.wewillimplementfinalintent7to7treatanalysis ascertainingifuqolchangedovertimeandifthatchangecanbeattributedtothe interventionsinthisproject. Summary*of*Outcomes*Evaluation* AssessmentofproviderandpatientmetricswilloccurattheintervalsdescribedinSectionE, andaremodeledontheopennotesevaluation(darer2014,delbancoetal2010).alldatais gatheredthroughepicandmychart,designedbytheprojectteamandanalyzedbythe projectteamincoordinationwithabiostatistician.tocontrolforavarietyofexternalfactors, wehaveincludedaphaseofstandardcare(atbaseline)andtwointerventionarms,andwill measurefidelityoftheinterventionasimplemented.thestudyispoweredtodetectan estimated30%improvementintheuqolindexfrombaselinetooneyearpost7enrollment. Participantengagementwillbemanageddirectlythroughexaminingaccess/portalusemetrics, responsestoquarterlyexperienceassessments,andthroughthequalitativeevaluation. III.b***Dissemination* Scientifically,resultsfromthisprojectwillbepresentedatnationalconferencesandwillbe submittedforpeer7reviewedpublication.wewilltarget,specifically,methodologicaljournals interestedintheimpactofelectronichealthrecord7basedinterventions,andalsoinclinical journalsinterestedinwhetherornotvmssymptommanagementcanbeimpactedbyclinician7 orientedintervention.internaltotheuniversityofrochester stechnicalenvironment,theepic teamwillsharethemethods,operationalization,andimpactofthisprojectwithepicleadersin otherdisciplineswhomaywishtotakeasimilarapproachintheirownclinicalresearchusinge7 Record.Additionally,DrWoodswillsharetheproject sfindingswithotherlocal,regional,and 11

13 statepractitionerswhoworkwithwomenofmenopausalage.theproject sfindingswillfurther beintegratedwiththeperifactscontinuingeducationsystem(whichdrwoodsdirects)which willprovidethemostcomprehensivelocaloutletforresults. III.c***Innovation** Thisprojectisinnovativeinitsdesignandapproach.Afterconductingadetailedliterature search,wehavenotidentifiedanotherclusterrandomizedtrialthatrandomizedpractitioners tointerventiongroupsandassessedvmssymptomatologyoverthecourseofoneyear.as describedearlier,thereareinherentbiasesinnotrandomizingpractitionersinresearchwhere patientoutcomesareassessedthoughthemaintargetforinterventionisaclinical,practice7 basedone.secondly,thisprojectisuniquelysituatedinanidealcontextofalargee7record installationinacompletemedicalinstitution,withtheintegralinvolvementofpractitionersand informaticists,placingthestudyatthecenterofwheretheuniversity sresearchcapacityand clinicalcareintersect. IV.**Detailed*Workplan*and*Deliverables* Thisproject stimeline(table2)isanticipatedtobe24months,withfourmonthsforbuild planning,implementation,andtesting,onemonthfortraining,staggeredrandomizationand enrollment(estimatedforsixmonths),andfollow7upfortwelvemonths.threemain deliverablesguidethisprojectthatreflecttheproject slogicmodel:theinterventionsare developedandbuilt,theclusterrandomizedcontrolledtrial(theprimarysourceofinformation forthisproject)isdevelopedandimplemented,andtheinformationfromtherctisanalyzed, interpreted,anddisseminated. Table&2:&Timeline Overview&of&Deliverables&Timeline &EHR&Build/&Intervention&Development 2.&RCT&Developed&and&Implemented 3.&RCT&Evaluated&and&Disseminated Deliverables&Timeline&Detail Finalizing2team2members All:hands2team2meeting2(monthly) IRB2Approval2and2Registration Practitioner2focus2groups2to2inform2intervention Patient2focus2groups2to2inform2intervention Intervention2build2prototyped Deliverable&1:&Intervention&build&finalized Practices2recruited Practices2randomized Consent2created2in2MyChart Patients2recruited Baseline2measurement2of2GCS/2UQoL Repeated2measures2GCS/2UQoL2(Every232months) Deliverable&2:&Implementation&of&Full&RCT&Protocol Qualitative2Process2Evaluation Data2management2and2quality2control Statistical2analysis Conference2Presentation Deliverable&3:&Develop&and&Submit&Manuscripts Month&of&Project 12

14 Deliverable*1:*EHR*Build*and*Intervention*Development.Whiletheprojectteammembers haveideasabouttheehrintervention(e.g.,dashboards,clinicalpracticereminders,decision7 aids),thedetailsandusabilityoftheinterventionneedstobetestedwithpatientsand practitioners.thus,thequalitativephasewillbeginimmediatelywithconceptandprototype testingandfocusgroupingtorefineandfinalizetheelectronicinterventionsinthefirstquarter (bymonthfour)oftheproject.becausethetrialbeginswithabaselinephase(nointervention) thatconsistssolelyofassessmentdeployment(withoutfeedbacktopatients),thebuilderswill haveuptosixadditionalmonthstofinalizeandintegratethefinalinterventionsintothee7 Record(EPIC)system. * Deliverable*2:*RCT*Developed*and*Implemented.Themainfocusoftheprojectinthecluster randomizedcontrolledtrialoftheelectronicinterventions,withtwoarms(patientintervention only,andpatient7practitionerintervention).explainedearlier,thestudyquestionsnecessitate thisdesigngiventhelackofsystemsindependenceofpractitionersbelongingtothesame practice(thatis,allpractitionersinaparticularpracticeusethesamesystemforallpatients, makingsimultaneousimplementationofavarietyofelectronicinterventionsconfounded).as such,practiceswillberecruitedforthisstudy(estimated12715practices)andsubsequently randomized.allpatientswithineachpracticemeetingeligibilitycriteriawithinthesixmonth windowofrecruitmentwillbeapproachedelectronicallytoobtainconsenttoparticipate. MyChart(thepatientinterfacewithEPICandtheirpractitioner)willbedeployedwiththeGCS anduqolinstrumentstoconsentedparticipantsrepeatedatthreemonthintervalsthroughout theproject.theactualtrialisexpectedtolastfor18monthsoftheprojectperiod,withtwo monthsinthefirstquarterforadditionalplanningandadministrativeprocedures. * Deliverable*3:*RCT*Evaluated*and*Disseminated.Thefinalanalysisfortheprojectwilloccurin thelastquarter,togetherwithmanuscriptgenerationandconferenceabstractsubmission (dissemination).however,theprojectteamwillbeconductingevaluationactivitiesfromthe startoftheprojectperiodthrough:1)theongoingqualitativeformativeandprocessevaluation, 2)ongoingdataandenrollmentmanagementanddataqualityreviews,and3)anticipatory statisticalanalysisandanalyticplandevelopment. Administration*and*Management.TheprojectwillbemanageddailybytheProject CoordinatorandthePrincipalInvestigator.Anall7handsteammeetingwillmeetmonthly throughouttheprojectsoallprojectteammembersareapprisedofprogressandinvolved. Regularmeetingswillbeheldwithsubgroupsassignedtoparticulartasks.Theprojectteamwill useasana( willalsoformpartofthestandingagendaforthectsibiomedicalinformaticsteam,inthe eventthatotherinformaticsresourcesneedtobeaccessedforunforeseentechnical circumstancesandtoprovideabackupmanagementstructureintheeventofanemergency. ThePIandProjectCoordinatorwillcooperativelycompleteallRSRB(IRB)requirements,trial registrations,andotheradministrativeandregulatoryaspectsoftheproject.thepiandproject Coordinatorwillmeetatleastweeklytoreviewtheproject sprogress,andtheproject CoordinatorwillalsoattendtheweeklyCTSIBiomedicalInformaticsteammeetings. 13

15 References Abraham,L.,Pinkerton,J.V.,Messig,M.,Ryan,K.A.,Komm,B.S.,&Mirkin,S.(2014). Menopause7specificqualityoflifeacrossvaryingmenopausalpopulationswithconjugated estrogens/bazedoxifene.maturitas. AmericanCollegeofObstetriciansandGynecologists.(2014).ACOGPracticeBulletinNo.141: managementofmenopausalsymptoms.obstetgynecol,123, DarerJ.DisruptiveInnovationandData7DrivenCare.UniversityofRochesterDepartmentof MedicineGrandRounds,June2014.* Delbanco,T.,Walker,J.,Darer,J.D.,Elmore,J.G.,Feldman,H.J.,Leveille,S.G.,&Weber,V.D. (2010).Opennotes:doctorsandpatientssigningon.Annalsofinternalmedicine,153(2), GreenJ.Constructingastandardclimactericscale.Maturitas1998;29: KruegerRA.FocusGroups.2 nd ed.thousandoaks,ca:sagepublications,1994. NutbeamD,HarrisE,WiseM.TheoryinaNutshell:APracticalGuidetoHealthPromotion Theories(ThirdEdition).Australia:McGraw7Hill,2012. Silow7Carroll,S.,Edwards,J.N.,&Rodin,D.(2012).Usingelectronichealthrecordstoimprove qualityandefficiency:theexperiencesofleadinghospitals.thecommonwealthfund,17,1738. UtianWH,JanataJW,KingsbergSA,SchluchterM,HamiltonJC.TheUtianQualityofLife (UQOL)Scale:developmentandvalidationofaninstrumenttoquantifyqualityoflifethrough andbeyondmenopause.menopause2002;9(6):

16 V.*Organizational*Detail* OrganizationalCapacity ThisprojectwillbeledbytheUniversityofRochesterMedicalCenter sdepartmentof ObstetricsandGynecology.Dr.TimothyDye,ProfessorofOB/GYN,Pediatrics,Medical Informatics,andPublicHealth,andservesasAssociateChairforResearchintheDepartmentof OB/GYN,willserveasPI,andDr.JamesWoods,ProfessorandChairEmerituswillserveasCo7 Investigator.Dr.DyealsoleadsBiomedicalInformaticsintheClinicalandTranslationalScience Institute(CTSI),whichalongwiththeDivisionofMedicalInformaticswillformthetechnical teamfortheproject.dr.woodsisaworld7recognizedleaderinob/gynpatient7provider communication,menopauseresearch,andobstetricalandgynecologicalpractice.drs.dyeand Woodswillbesupplementedbyotherprofessorialandtechnicalstaffinimplementingthis work. University*of*Rochester*Medical*Center* Oneofthenation'stopacademicmedicalcenters,theUniversityofRochesterMedicalCenter formsthecenterpieceoftheuniversity'shealthresearch,teaching,patientcareand communityoutreachmissions.overthelastfiveyears,theurschoolofmedicineanddentistry hasreceivedalmost$1.3billionintotalresearchfunding.theschoolranksinthetopquartileof U.S.academicmedicalcentersinresearchfundingfromtheNationalInstitutesofHealth, attractingnearlythreetimesthefederalfundingreceivedbythemedicalschoolsinalbany, SyracuseandBuffalocombined.TheUniversity'shealthcaredeliverynetwork URMedicine isanchoredbystrongmemorialhospital7an8007bed,university7ownedteachinghospital whichsustainsspecialtyprogramsthatconsistentlyrankamongthebestinthenation accordingtousnews&worldreport. WhilethisprojectisnotlimitedtopractitionersintheURMCOB/GYNDepartment,the Departmentdoesprovideabout70percentofcaretowomenwithVMSseenattheUniversity, whichisthelargerproviderintheregion.theinvestigatorswillengagethepractice7based ResearchNetworkandtheOBPracticeNetworktoaccessadditionalpractitionerswhoprovide primarycaretowomenwithvms.nestedwithinthedepartmentofob/gynwillenhancethe project sresearchenvironment,accesstoclinicalresearchresources,andwillfacilitate communicationthroughexistingin7person,on7line,andwrittenmechanisms.further,the projectmayinvolvesomeoftheresidentsandmedicalstudentsbeingtaughtinthe Department,providingaresearchlearningexperienceforfuturepractitionersandwomen s carepractitioners. TheEPICsystemismaintainedbytheDivisionofMedicalInformatics,headedbyDr.David Krusch,alsoaCo7Investigatorofthiseffort.Dr.Dye steaminthebiomedicalinformaticsgroup ofthectsiworkscloselywithdr.krusch steaminmedicalinformaticstoimplementresearch interventionsusingepicanditsassociatedprojects.withthetechnicalteamnestedwithinthe investigatorteamandfullyresources,thereislittleriskthatthepriorityofthisprojectwilldelay implementationorresearchattention.dr.dye steamalsoworkscloselywiththedepartment ofbiostatisticstoprovideanalyticservices. 15

17 StaffCapacity Timothy*De*Ver*Dye,*PhD*willserveasPrincipalInvestigatorforthisproject.DrDyeisProfessor andassociatechairforresearchinthedepartmentofobstetricsandgynecologyatthe UniversityofRochester,andalsoservesasDirectorofBiomedicalInformaticsforUR sclinical andtranslationalresearchinstitute.drdyehasextensiveexperiencewithclinicaland translationalresearchwithanemphasisonwomen sandchildren shealth,andhasoverseen numerouslarge7scaleinformaticsprojectstheintegratewithhospitalsystems.specificallyfor thisproject,drdyewillliaisewiththepfizerteam,willoverseetheprojectteam sdevelopment andimplementationoftheproposedresearch,andwillleadinthedisseminationoffindings.as anepidemiologistandanthropologistwithexpertiseinbiomedicalinformatics,drdyeiswell7 suitedtorelatetoallaspectsoftheproposedresearch. Project*Coordinator.Apart7timeprojectcoordinatorwillbehired/assignedtocoordinateall aspectsofthestudy,workingcloselywithdrdyeandprojectteammemberstoassurethat projectplansaredevelopedandimplementedefficiently.theprojectmanagerwillalsoserveas themaininterfaceforprojectreportingandcompliance,bothwithpfizerandwiththe UniversityofRochester sinternalsystems.theuniversityofrochester sctsimaintainsa robustprofessionaldevelopmentorganizationofresearchstudycoordinators(study CoordinatorsOrganizationforResearchandEducation SCORE )thatprovidesnetworkingand trainingopportunities.thisprojectwillfirstaccessthescoregroupwiththisopportunity,only recruitingexternallyintheunlikelyeventthatanexistingscorestudycoordinatorwithstrong informaticsbackgroundcannotbeidentified.thecoordinatorwillbemaster sprepared,with clinical/clinicalresearchexperience,anddemonstratedinformaticsandehrexperience, preferablyinthecontextofprimarycareandwomen shealth. James*Woods,*MDwillserveasCo7Investigatorforthisproject.DrWoodsisProfessorand formerchairofthedepartmentofobstetricsandgynecologyattheuniversityofrochester, andhasextensiveexpertiseinwomen shealth,clinicalresearch,andmenopausalcare.dr WoodsisdeeplyinvolvedwithcontinuingprofessionaleducationasEditor7in7ChiefofPerifacts, subscribedbyover11,000ob/gynnursesandphysiciansacrosstheu.s.drwoodsalsowritesa monthlyarticleonmenopausemedicineandproducesamonthlyfeatureforrochesterwoman magazineonmenopause.heisasignificantopinionleaderinrochesterandwillworkwith Kalin*Warshoff*N.P.,*whoisanursepractitionerclinicianspecializinginwomen shealth, providingcontentexpertise,testingehrprototypes,reachingouttootherparticipating practices,overseeingtrainingofpractitionersimplementingtheinterventions,andvetting healtheducationmaterialsforthepatientcomponentoftheintervention. Ollivier*Hyrien,*PhDwillserveastheproject smethodologistandbiostatistician.drhyrienis AssociateProfessorofBiostatisticsandComputationalBiologyattheUniversityofRochester, andhasoverseenthedesignandstatisticalanalysisofmanyrandomizedtrialsofclinical interventions.drhyrienwilloverseetheoperationalizationofthestudydesignincluding randomizationofpractices,statisticalmanagementofstudyenrollment,andwillconductall analysesofstudydata. 16

18 Miriam*Weber*PhDwillserveasthepsychometricianfortheproject,inparticularoverseeing theoperationalizationanddeploymentofthegreeneclimactericscaleandtheutianqualityof LifeScale.DrWeber,anAssistantProfessorofNeurologyattheUniversityofRochester,will workcloselywithdrhyrienaroundanalysisofthescalesandsubscalesgeneratedfromthese instruments,andwillassisttheepicbuildteaminpresentingtheresultsoftheseinstrumentsin avalidandconcisemannertopractitionersinthepractitionerintervention.drweberisan experiencedclinicalresearcherwhohaspublishedextensivelyonmenopauseandqualityoflife, andwillbringthatexperiencetothisprojectastheexpertonmenopausalsymptomandquality oflifemeasurementandanalysis. Margaret*Demment*PhDwilloverseethequalitativeassessmentandpatient/practitioner engagementcomponentsofthisproject.drdemmentisanepidemiologistandseniorresearch AssociatewiththeClinicalandTranslationalResearchInstitute(CTSI)attheUniversityof Rochester.DrDemmenthasexperienceorganizingandconductingformativeandevaluative qualitativeresearchandwithintegratingqualitativeresearchmethodsintoquantitativedesigns (mixedmethods).drdemmentwillworkcloselywithdrdyeanddrwoodstoassesspatient andpractitionerexperiencewiththeinterventionsofthisproject,andwithmrtatroanddr Darertoassessoptimalinterventiondevelopment. Adam*Tatro,*MSNistheleadinformaticistfortheprojectandwilloverseedevelopmentand implementationoftheehrbuilds,includingmychartquestionnairedevelopmentandepic dashboardsforclinicians.mrtatrohasextensiveexperienceinbothclinicalandresearch informaticsandistheleadinformaticistforthectsiattheuniversityofrochester.inthis capacity,heroutinelyinterfaceswiththeepictechnicalstaff(locally,nationally,andatother institutions)andisfamiliarwithepicfunctionalityandoperations.mrtatrowillcoordinatethe EPICProgrammerandthetechnicaldevelopmentteamtoassureappropriatedevelopmentand implementationofdatacapture,feedback,anddisplayofallprojectinformation.finally,mr TatrowillworkwithDrDemmenttoconductusabilitytestingwithpilotgroupstoassurethat theinterventionsareacceptableandsensitivetopatientandpractitionerneeds. Jonathan*Darer*MD*MPH*isChiefInnovationOfficeratGeisingerHealthSystemandan experiencedclinicalinformaticist.drdarer,aprimarycarephysicianatgeisinger,hasextensive experienceadaptingepicelectronichealthrecordsystemsforoptimalclinicaluseandfor clinicalcareimprovement.drdarerwilladvisetheteamoninterventiondesignand implementation,andwillassistwithinterpretationoffindingsandpatient/practitioner evaluation.hisextensiveexperiencewithpatientsandpractitionersaroundelectronichealth recordinnovationwillassisttheprojectinmaximizingthepotentialoftheepicsystem. EPIC*Programmer.Apart7timeEPICprogrammerwillbehired/assignedtoprogramallbuilds, reports,andinterfacesasrequiredfortheproject.theprogrammerwillserveasthemain liaisonwiththedivisionofmedicalinformaticsstaffresponsibleforepicimplementationatthe UniversityofRochesterandwillassurethatallprojectdevelopmentandprogrammingisin compliancewithuniversityandepicstandardsandrequirements. 17

19

20 Timothy De Ver Dye, PhD Professor of Obstetrics & Gynecology, Pediatrics, Public Health Sciences, & Medical Informatics Vice Chair (Research), Obstetrics & Gynecology Director, Biomedical Informatics Clinical and Translational Science Institute University of Rochester School of Medicine and Dentistry Dear Tim: September 2, 2014 The University of Rochester (UR) School of Medicine and Dentistry (SMD) is pleased to support your application to Pfizer to conduct electronic health record research to improve patient and clinical experience managing vasomotor symptoms for women in menopause. The research as proposed aligns well with the UR SMD s new Strategic Plan that emphasizes innovation, population health, widespread integration of technology to improve care, and improving value. The Department of Obstetrics and Gynecology, the academic home for this project, constitutes a large practice group that serves women experiencing menopausal symptoms, has nationally- and internationally-recognized menopausal researchers, and has clinical champions with strong visibility around menopause-related issues among practitioners and the community. Further, the University of Rochester has had a large installation of EPIC (our electronic health record system) since March 2011 with mechanisms for conducting the type of research proposed. The grant application appropriately is centered with the Clinical and Translational Science Institute s Biomedical Informatics team working closely with our Medical Informatics Division (which runs the University s EPIC system), Biostatistics and Computational Biology on study design and statistical analysis, and of course our strong Department of Obstetrics and Gynecology. As Vice Dean for Research at the School of Medicine and Dentistry, I will do whatever I can to encourage others to conduct practical, EHR-based research using rigorous methods working with you and your team. I look forward to learning of this project s progress and extending those learnings to the intersection of our research and clinical mission, an intersection directly addressed by EHR-related research activities. Sincerely Stephen Dewhurst, Ph.D. Vice Dean for Research, School of Medicine and Dentistry Associate Vice President for Health Sciences Research University of Rochester

21 SCHOOL OF MEDICINE AND DENTISTRY Department of Obstetrics and Gynecology Eva K. Pressman, MD Henry A. Thiede Professor and Chair September 1, 2014 Timothy De Ver Dye, PhD Professor of Obstetrics and Gynecology Associate Chair for Research, Obstetrics and Gynecology University of Rochester School of Medicine and Dentistry 265 Crittenden Boulevard Rochester, NY Dear Tim, As the Chair of the Department of Obstetrics and Gynecology at the University of Rochester, I am very pleased to endorse your application to help improve care to women through creative uses of the electronic health record. We have been using EPIC for several years and research projects such as the one you and Dr.Woods (our former chair) propose will help our physicians connect more strongly with the potential of e- record to strengthen the care experience. Vasomotor symptoms in menopausal women is a common condition we see in providing care; indeed this area can be easily overlooked in the context of other screening and clinical activities and your study will help us explore the best way to package patientgenerated information for clinical use. Further, I think that since you re using educational materials to expand women s knowledge of menopause that have been vetted (and in some cases, created by) our clinicians will help steer patients toward useful and accurate resources rather than leaving them to navigate the internet without guidance. In many ways, this project can be seen as an example for other studies based on using the electronic health record as a mechanism to improve care. This project will be housed in our Department in collaboration with your informatics team at the Clinical and Translational Research Institute (CTSI). This project is also a great example of the intersection of informatics and clinical care for women, and the team you ve assembled (which includes Dr Woods and other clinical providers) will help assure that your work is relevant and visible to clinicians. Our Research Division will provide whatever support necessary to facilitate this project and I think, since the project is embedded within a clinical department, there is a great potential that the project will be successful. I think your project could provide a wonderful opportunity to help us see whether or not the electronic health record can provide unique opportunities to facilitate evidence-based care. That you are using the new ACOG Practice Bulletin from January 2014 on Managing Menopause Symptoms 601 Elmwood Avenue Box 668 Rochester, NY fax

22 as a base for clinical recommendations and information will certainly help us achieve that. On behalf of our department, I endorse this project, support your efforts and look forward to working with you on this important initiative. Sincerely, Eva K. Pressman, M.D. Henry A. Thiede Professor and Chair of Obstetrics and Gynecology

23

24 September 2, 2014 To: Timothy De Ver Dye, PhD Professor of Obstetrics and Gynecology, Pediatrics, Public Health Sciences, and Medical Informatics Director, Biomedical Informatics Clinical and Translational Science Institute University of Rochester School of Medicine and Dentistry 265 Crittenden Boulevard Rochester, NY Dear Tim, I am pleased to support your application to Pfizer entitled Electronic Health Record (EHR)-enhanced clinical communication and quality-of-life improvement in managing menopausal vasomotor menopausal symptoms (VMS). As you know, I head the Division of Medical Informatics at the University of Rochester and am responsible for overseeing our institution s installation of EPIC enterprise-wide. We transitioned to EPIC several years ago and our use is now widespread with well over one million patients managed in the system. This research that you propose will serve as a model for other researchers at the University who wish to learn from electronic record enhancements that could well improve care. The widespread use of EPIC is an incredible opportunity for us to improve care and provide effective clinical experiences for patients. That you ve included adequate funds and personnel in your application to oversee the EPIC builds that are required to implement the intervention will help us not have to reassign our existing staff and will allow sufficient attention for your project s needs. I can assure you that our team will provide whatever we can to help facilitate this work. erecord (as our local installation of EPIC is termed) was heralded at its inception as an opportunity to enhance communication between practitioners and patients, to help practitioners distill and interpret large amounts of clinical information, and to facilitate best practice. Your project will help us demonstrate all of these features applied to a very important clinical area that crosses disciplines. I look forward to participating in your project and helping others learn from this experience and move e-record-based research forward. Sincerely, David A. Krusch, MD, FACS Chief Medical Information Officer Director, Division of Medical Informatics Professor of Surgery and Informatics University of Rochester School of Medicine and Dentistry University of Rochester Medical Center 601 Elmwood Avenue, Box SURG Rochester, New York Phone: (585) David_Krusch@urmc.rochester.edu