Articulo Para Revisar

8/3/2019 Articulo Para Revisar

1/13

www.amcp.org Vol. 14, No. 9 November/December 2008 JMCP Journal of Managed Care Pharmacy 831

ABSTRACT

BACKGROUND: There is evidence that pharmacist interventions improveclinical outcomes. The few studies that address economic outcomes(a) often repor t estimated instead of actual medical costs, (b) report onlymedication costs, or (c) have been conducted in settings that are nottypical of community-based primary care.

OBJECTIVES: To (a) determine whether a clinical pharmacists recommen-dations to physicians regarding optimizing medication therapy are relatedto medical costs in capitated patients in an internal medicine practice, and(b) compare what primary care physicians (PCPs) in a comparison groupactually did proactively to optimize medication therapy versus what a

clinical pharmacist would have recommended to them.METHODS: This was a prospective, controlled study comparing 2 internalmedicine practices. Study enrollment was performed using a screeningprocess carried out every 1-2 weeks on a rolling basis for 1 year fromJuly 2001 through June 2002. Eligibility criteria for prospective enrollmentwere (a) 1 or more r isk factors: at least 1 chronic disease or an event(e.g., emergency room visit, adverse drug reaction, medication nonadher-ence) or aged 50 years or older, (b) a scheduled visit to see a PCP within2 weeks from the screening date or a diagnosis of diabetes without a PCPvisit during the first 6 months of the study, (c) need for optimization ofmedication therapy as determined by a clinical pharmacist on the screeningdate, and (d) 12 months of continuous insurance eligibility before enroll-ment in the study. For inclusion in the final study analyses, patients werealso required to have continuous insurance eligibility through 12 monthsfrom study enrollment. One clinical pharmacist made recommendations

to optimize medication therapy in the intervention group. For the compari-son group, the same pharmacist proposed recommendations that remainedconcealed from the physicians. The primary outcome measure wasper patient per year (PPPY) medical cost, based on plan liability (grossallowable costs minus patient costs), excluding prescription drug cost.

Additional outcome measures included numbers of outpatient visits,hospital admissions, emergency room (ER ) visits per 1,000 patients, andhospital days; and percent of recommendations that were accepted by thePCPs. Changes in outcome measures from the pre-intervention to post-intervention period were compared across study groups in a difference-in-difference analysis, using the Students t-test for normally distributed dataand the Mann-Whitney U-test (nonparametric) for skewed data.

RESULTS: There were 127 and 216 adult patients in the intervention andcomparison groups, respectively. The primary outcome, change in meanPPPY medical (excluding pharmacy) cost, did not dif fer significantly

between the groups (P= 0.711). The between-group difference in thechange in ER visits per 1,000 patients approached statistical significance(P= 0.054). Intervention group patients were more likely than comparisongroup patients to have the following issues addressed: medication non-adherence (85.7% vs. 40.0%, respectively; P= 0.032), untreated indication(72.6% vs. 11.5%, P< 0.001), suboptimal medication choice (60.0% vs.5.9%, P< 0.001) and cost-ineffective drug therapies (72.1% vs. 6.5% ,P< 0.001). Of the estimated number of actionable opportunities identifiedfor the comparison group (but concealed from the physicians), 23.5%were adopted by comparison group physicians without any assistancefrom a clinical pharmacist.

CONCLUSION: Compared with patients of PCPs who received no input froma clinical pharmacist, patients of PCPs who received clinical pharmacist

recommendations were more likely to have several medication-relatedissues addressed, including medication nonadherence, untreated indica-tions, suboptimal medication choices, and cost-ineffective drug therapies.However, total medical (excluding pharmacy) costs for the intervention andcomparison groups were not significantly different.

J Manag Care Pharm. 2008;14(9):831-43

Copyright 2008, Academy of Managed Care Pharmacy. All rights reserved.

A Prospective Trial of a Clinical Pharmacy Intervention

in a Primary Care Practice in a Capitated Payment System

Jeanette L. Altavela, PharmD, BCPS; Matt K. Jones, BA; and Merrilee Ritter, MS

RESEARCH

Pharmacists can optimize medication therapy, resulting in

improved patient outcomes, suchas decreased exacerbations

inpatients withcongestiveheart failureand improvedblood

sugarmanagementinpatientswithdiabetes.

Pharmacistsare effective at recognizingpotential andactual

drug-related problems,suchas drug-induced conditions and

clinicallyrelevantdruginteractions.

Pharmacistinterventionscanlimithealthcarecostsinspecific

groupsofpatientssuchasMedicaidandsomehealthmainte-

nanceorganizations.

What is already known about this subject

What this study adds

Patientsintheinterventiongroupweremorethantwiceaslikely

tohavemedicationnonadherenceissuesaddressed(85.7%vs.

40.0%,P=0.032),6 times as likelyto havea medication pre-

scribedthatwasindicatedbutnotprescribedpreviously(72.6%

vs.11.5%,P


2/13

832 Journal of Managed Care Pharmacy JMCP November/December 2008 Vol. 14, No. 9 www.amcp.org

Studieshaveshownthatpharmacistconsultationprogramscan improveclinical outcomesby optimizingmedicationuse inambulatory patients.1-5 Amongpatients in a heart

failure clinic, a program of pharmacist evaluation (medication

evaluationandrecommendations,patienteducationandfollow-uptelemonitoring)resultedinasignificantdecreaseinheartfail-ureeventsandall-causemortality.Studyauthorsattributedthisresulttocloserfollow-upandoptimizingdosesofangiotensin-convertingenzyme(ACE)inhibitors.1In2randomizedtrialsofpatients with hypertension, those who weretreated collabora-tivelybyphysiciansandpharmacistsachievedbettercontrolofbloodpressurethandidthosewhoweremanagedbythephysi-cianalone.2,3Thephysician-pharmacistteamin1studyincreasedmedication optimization by titrating doses more effectively,switchingtolessexpensiveormoreappropriateformulationsofmedications,andincreasingappropriatelaboratorymonitoring. 2

Evenwhenpatientsmedicationswerenotchanged,bloodpres-sureswerestillimproved.Theauthorsspeculatedthatimprovedmedicationadherenceandbeneficialeducationabouthyperten-sioncontributedtotheseoutcomes.2Collaborationbetweenphy-siciansandpharmacistshasresultedinahigherrateofpatientsmeetingtheirlipid-levelgoalsthan previouslyachievedwithoutcollaborationinthesamepractice.4TheAshevilleProjectdemon-stratedthatclosecollaborationbetweencommunitypharmacistsandpatientswithdiabetesmellituswasassociatedwithimprovedbloodsugarmanagement.5

TheNationalCommitteeforQualityAssuranceskeyprogramfor quality measurement is the Healthcare Effectiveness DataandInformationSet(HEDIS).Since2007,HEDIShasincludedmeasuresofhealthcareefficiencyinthecostofcare,referredtoas relative resource use for chronic conditions. Forexample,asthma and cardiovascular conditions are measured both forquality,suchasappropriatemedicationuseandmedicationadher-ence,andforthecostofcare.6,7Somestudiesofclinicalpharma-cistactivitieshaveconcentratedonloweringmedicationcosts, 8,9butfewhaveattemptedtolookattheimpactonmedicalhealthcarecostsandutilization.Loweringmedicationcostshasbeenaccomplishedbysimplifyingmedicationregimens,recommend-inglessexpensivealternatives,andprovidingpharmacotherapyconsultationdirectlytopatients.8-10Inanefforttodecreasemedi-calhealthcarecostsandutilization,somestudieshavedemon-

stratedthatpharmacistseffectivelyidentifypotentialandactualdrug-relatedproblems,potentiallyresultingincostavoidance.8,10

Previous studies that assessed clinical or medical cost out-comes were either conducted in U.S. Department of Veterans

Affairs (VA) systems, in a setting where the patient was seenataseparatepharmacistvisit,inapharmacist-runclinic,orinpopulationsthatweredissimilartogeneralprimarycareinternalmedicinepractices.10-14Althoughthesestudiesdescribeeffectivemodels,theydonotextrapolatewelltothetypicalprimarycare,internalmedicinepracticewheremedicalpatientsaremostoftenseenbyphysiciansandinwhichpharmaciststypicallyhaveno

access to pertinent medical information (e.g., medical history,progress notes, laboratory and other test results, and consultnotes)necessarytomakeclinicalrecommendationstoprescrib-ers. Embedding a clinicalpharmacistwithin the primary care

practicecanremovethosebarriers.Inthe2yearsbeforethepresentstudy,2clinicalpharmacists

working for the Greater Rochester Independent Practice Asso-ciation (GRIPA) hadgainedexperiencewith a number ofpri-marycarephysicians(PCPs)onhowtoimprovemedicationuseand prevent the known hazards associated with medicationmisuseintheirpatients.GRIPAisauniquepartnershipofmorethan600 physiciansand 2 hospitals in 2 counties in westernNewYork. Thepharmacists were locatedwithin thephysicianpracticewithlittledisruptiontothenormalofficeworkflow.Atthat time,the pharmacistsdid notmeetwith the patients,butprovidedwrittenrecommendationsto eachpatients physician.

The clinical pharmacist had opportunities to affect a patientsmedicationadherence,toensurethatthemostappropriatemedi-cationswerebothprescribedandmonitoredappropriately,andtohelp preventtherapeutic duplicationandadverse drugreac-tions.Inaddition,pharmacistsservedasadynamicdruginfor-mationresourceforthephysician.Forpatientswhosecarewasaffectedbytheclinical pharmacistsrecommendations,a trendtoward lowered medical health care costs and utilization wasobserved.15,16However,nocomparisongroupofpatientswithouttheservicesofaclinicalpharmacistwasavailableatthattime.

Theprimarypurposeofthepresentstudywastodeterminewhethertherecommendationsofaclinicalpharmacistembeddedin a primary care practice, whichhadnot previously receivedservicesfromGRIPAsclinicalpharmacists,woulddecreasethemedical costs of capitated patients. The secondary purposeofthestudywastocompareactionstakenbyphysiciansinacom-parisongroup,whichreceivednopharmacistinput,withactionstakenbyphysicianswhowereprovidedrecommendationsbyaclinicalpharmacist.

Methods

Study SettingThiswasaprospective,controlledstudyconductedin2primarycare practices located in the suburbs of Rochester, New York.One practice served as the intervention group, and the other

served as the comparison group. Physicians at both practicesweremembersofGRIPAandhadneverreceivedservicesfromGRIPAs clinical pharmacists. ViaHealth, GRIPAs parent com-pany, owns2 hospitals and one-half ofGRIPA; the physiciansowntheotherhalf.GRIPAoperatesunderfinancialriskcontractswith insurance companies. A portion of the patients in theseprimarycarepracticesweremembersofaninsurancecompanywithwhichGRIPAhadariskcontract.Thepatientswereenrolledin either the insurance companys commercial insurance planor its Medicare insurance product. The risk contractprovidedGRIPAwithanincentivetoproactivelyoptimizemedicalcareto

A Prospective Trial of a Clinical Pharmacy Intervention in a Primary Care Practice in a Capitated Payment System
http://www.ncqa.org/Portals/0/HEDISQM/Archives/2007/Vol2/Cost_of_Care_Update.pdfhttp://www.ncqa.org/Portals/0/HEDISQM/HEDIS2008/%202008_Measures.pdfhttp://www.ncqa.org/Portals/0/HEDISQM/HEDIS2008/%202008_Measures.pdfhttp://www.ajmc.com/files/articlefiles/AJMC1998JulGerber991_1000.pdfhttp://www.ajmc.com/files/articlefiles/AJMC1998JulGerber991_1000.pdfhttp://www.ncqa.org/Portals/0/HEDISQM/HEDIS2008/%202008_Measures.pdfhttp://www.ncqa.org/Portals/0/HEDISQM/Archives/2007/Vol2/Cost_of_Care_Update.pdf


3/13



decreaseitsfinancialrisk.Themedicalcostdata,termedplanliabilityinthisstudy,wereactual(notestimated)plansponsorcosts (gross allowable costs minus patient costs). The patientsforwhomGRIPAandtheirphysiciansareatriskarecapitated

patients.Although the plan was capitated, the physicians hadincentive to submit all claims to receive payment for servicesprovided.Theriskcontractprovidedopportunityforphysicianstogetpaidmorethanthestandardfeesreimbursedthroughtheclaimssubmissionandpaymentprocess.

Theintervention grouppracticehad 957 capitated patients,andthecomparisongrouppracticehad1,272capitatedpatients,with12.3%and31.6%enrolledintheMedicareinsuranceprod-uct,respectively.Theremainingcapitatedpatientsineachgroupwereenrolledinthecommercialinsuranceproduct.Bothprac-ticesconsistedofinternalmedicinephysicians,with2physiciansin the interventiongroup and 4physiciansin the comparison

group.Theinterventiongroupwasprivatelyowned,whereasthecomparisongroupwasownedbyViaHealth.Bothpracticesusedpaper-basedmedical records and appointment scheduling sys-tems.The2physiciansintheinterventiongrouphadpracticedfor18and6years,respectively,whereasthe4physiciansinthecomparisongrouphadbeeninpracticefor20yearsonaverage(range17-25years).

One clinical pharmacist worked within both practices andbrought a laptop computer to record her activity in a securedatabase.Attheinterventiongrouppractice,thepharmacistdidnot have Internet access. The comparison group practice wasequippedwithcomputerswithlimitedInternetaccess,whichthepharmacistcoulduseifneeded.Theclinicalpharmacistrecordedmedicationrecommendationsthatwereeitherprovidedtophysi-cians(interventiongroup)orconcealed(comparisongroup).

Writteninformedconsentwasobtainedfromthephysiciansatbothpractices.TheViaHealthClinicalInvestigationsCommittee(institutionalreviewboard)approvedthisstudy.

Patient Selection

Patientsenrolledinthisstudywerecontinuouslyenrolledin1ofthe2contractedinsuranceproducts(commercialorMedicare)fortheentire12monthsbeforetheirstudyenrollmentdatetoensurethattherewerecompletebaselineclaimsdata.Thepatientselec-tionperiod,duringwhichpatientswereenteredintothestudy

inarollingscreeningandenrollmentprocessconductedbytheclinicalpharmacistevery1-2weeks,beganonJuly1,2001,andendedonJune30,2002.Patientmembershipstatuswasprovidedtothepharmacistatstudyinitiation,andinsuranceclaimswereusedtodetermine eachpatients risk factors,which were usedaspartoftheentrycriteriainthestudy(Table1).Tobeeligibleforenrollmentintothestudy,patientshadtobescheduledforanappointmentwithaPCPwithin1-2weeksofthescreeningdateorhaveadiagnosisofdiabetesmellitusdocumentedintheirclaimsbutnoPCPvisitduringthefirst6monthsofthestudy.The second criterion served to identify patients with diabetes

mellitusthatdidnothaveoptimalfollow-upcare.Patientsidenti-fiedontheappointmentschedulehadtomeetatleast1ofthefollowing2 criteria: (a) 1 ormoreof the risk factors listed inTable1,identifiedthrough International Classification of Diseases,Ninth Revision, Clinical Modification (ICD-9-CM), DiagnosisRelatedGroup(DRG),orCurrentProceduralTerminology(CPT)codesfoundineachpatientsinsuranceclaimsforthe12monthsbefore April2001, or (b) absenceof any oftheaboverisk fac-tors,butaged50yearsorolder.Finally,aneedformedicationoptimizationwasrequiredforstudyentry;patientsmeetingtheotherstudycriteriawereenrolledonlyiftheclinicalpharmacistrecorded recommendations to optimize medication therapy,whetherreportedtothePCP(interventiongroup)orconcealed

(comparisongroup).Each patients study enrollment date was the first date on

whichthepharmacistmadearecommendationforthatpatient.Post-enrollment follow-up lasted 12 months for each patient.Thus,tobeincludedinthefinalstudyanalyses,thepatienthadtomaintaincontinuousinsuranceeligibilityandremaininthecareofthesamePCPforthe12monthsafterthestudyenroll-ment date.Insurance eligibility wasdeterminedbya monthlymembershiprostersenttoGRIPAfromtheinsurancecompany.ThemembershipstatusandriskfactorevaluationofthepatientsinthephysicianpracticeswereupdatedinJanuary2002.

TABLE 1 Risk Criteria for Study Entry:Hospital and Medical Claim Codes

ICD-9-CM DRG CPT

Diabetes 250.XX 294,295Congestiveheartfailure 428.XX 115,124,125,

127

Coronaryarterydisease 410, 410.9411.XX,412-414.XX,(Except414.1414.10, 414.11or 414.19)

106,107,109,112, 116, 121,122, 123, 132,140

33510-33545

Asthma 493.XX 096,097

Chronicobstructivepulmonarydisease

491-492.XX493.2496.XX

088

Hypertension 401-405.X X 134

Hypercholesterolemia 272.XX

Migraine 346.XX

Atrialfibrillation 427.31,427.32138, 139

Adversedrugreaction 995,995.1,995.2

Noncompliancewithmedicaltreatment

V15.81

Anyemergencyroomvisit 9

Tobaccoabusedisorder 305.1,989.84

CPT = Current Procedural Terminology; DRG = Diagnosis Related Group;ICD-9-CM = International Classification of Disease, Ninth Revision, ClinicalModification.


4/13


Description of the Intervention

In both practices, the same clinicalpharmacist reviewed eachpatients medical record and assessed whether the patientsmedication therapy could be optimized. For the interventiongroup, the clinicalpharmacist provided the PCP with writtenrecommendations(consultnote)regardingdrug-relatedproblemssimilar to those described by Strand et al.17 All consult noteswerecompletedbeforethepatientsappointmentwiththePCP.Theconsultnoteswerenotmeanttobeapermanentpartofthemedicalrecordandwerelabeledaccordingly,whichlikelylim-itedphysicianspotentialconcernsaboutmedicalmalpracticeorliabilityrelatedtothesenotes.Theconsultnoteswerewrittenoncolorful paper and placed conspicuously in the papermedicalrecord.

Forthecomparisongroup,theclinicalpharmacistdocumentedin the database the recommendations for each patient, which

remainedconcealedfromthePCP;physiciansinthecomparisongrouppracticewereaskedtoactasthoughtheclinicalpharma-cistpresentintheofficeisinvisible.Theestimatednumberofactionableopportunitiesforthecomparisongroupwas(a)calcu-latedbymultiplyingtherecommendationacceptanceratefortheinterventiongroup(thepercentageofclinicalpharmacistrecom-mendations that were actually adopted by intervention physi-cians)timesthenumberofconcealedrecommendationsforthecomparisongroup,andthen(b)comparedwiththeactualnum-berofchangesmade bycomparisongroup physicians withoutclinicalpharmacistassistance.Thepharmacistalsodocumentedintheprivacy-secureddatabaseallknownchronicdiseasesand

otherdemographicinformationforbothstudygroups,includingheight and weight, if these data wereavailablein the medicalrecord.

Medicalrecordreviewswereconductedforallpatientswhowereenrolledinthestudy.Themedicalrecordincludedmedicalhistory,physicalexam,consultnotes,laboratorydata,andothertestresults.For72.9%and39.3%ofthecapitatedpatientsintheinterventionandcomparisongroup,respectively,thepharmacisthad access to claims data reflecting the patients prescriptionrefill(pharmacy)claimsfromthepatientsinsurancecompany.Pharmacy claims data were available only for the capitatedpatients that hada prescriptionbenefit through the insurancecompanywithwhichGRIPAhadariskcontract.Forinstance,

therewerenopharmacydataonpatientswhohadmedicalinsur-ancebutfilledalltheirprescriptionsthroughtheVA.Noneofthephysicianshaddirect accesstothepharmacyclaimsdata. Thepharmacistinterpretedthepharmacyclaimsdataanddistilledthat information into her consult notes as needed tooptimizemedicationtherapy.However,becausepharmacy datawerenotavailableforallstudypatients,costsforprescriptiondrugscouldnotbeassessedexceptintheaggregate.

In addition toproviding proactive recommendations to theintervention group physicians, the clinical pharmacist wasavailabletohelpwithanymedication-relatedproblemsordrug

informationissuesatthephysiciansorstaffsrequest.Theclini-calpharmacistalsoofferedphysicianeducation,patientcounsel-ing,adherencemonitoringandeducationasdeemedappropriate.Patient counseling was done only onan as-needed basis, was

notdirectedatanyparticularcondition,andgenerallydealtwithmedication nonadherence. Otherwise, most of the medicationadherenceissuesweresimplybroughttotheattentionofthePCPsforthemtoaddressduringthepatientsvisit.

The clinical pharmacist was not available to the compari-songroupphysiciansforconsultationduringthestudyperiod.However, an a priori decision was made that, if a significantfinding were discovered during a medical record review inthe comparison group that required immediate attention toprevent patient harm, the clinical pharmacist would consultthe physician and the patient would be discontinued fromthestudy.

Theclinicalpharmacistrecordedphysicianresponsestoeachrecommendationat6monthsand12monthsaftertherecom-mendationwasmade,inboththeinterventionandcomparisongroups.Recommendationsmadebythepharmacistthatwerenolongerapplicablebythetimeofthepatientsappointmentwereexcluded from the analysis. The clinical pharmacist recordedaphysicianresponseasacceptediftherewasevidencedocu-mentedwithinthemedicalrecordindicatingthattherecommen-dationwasfollowed(e.g.,achangeinaprescription,alaboratorytestordered).

Once patients met all criteria for inclusion, the study wasconducted withan intent-to-treatanalysis.Whetherornotthe

physicianadoptedthepharmacistsrecommendation,thatpatientwasincludedinthefinalanalysis.

Outcome Measures

MedicalcostsandutilizationwereobtainedfrommedicalclaimsdatacontainedintheGRIPAdatawarehouse.Thesedataorigi-natedfromeachenrolledpatientsinsurancecompany.Cost(planliability)wascalculatedasaperpatientperyear(PPPY)amountfortheprimaryoutcomeandtabulatedforallclaimsforhospi-talizations, emergency room (ER) visits, radiology and labora-torytests,PCPvisits,andspecialtyvisits.Althoughincludedinmedical costs, inpatient costs also were tallied separately. Theutilization data includednumber of hospitalizations, ER visits,

PCP visits and specialty visits, and hospital length of stay indays.Hospitalizationswereidentifiedbyanyclaimwithavaliddiagnosisrelatedgrouporarevenuecodebetween100and219(room and board)as longas the facilitytype wasnot askillednursingfacility ornursing home. Medical costsandutilizationwere determinedfor 12monthsbeforeand aftereach patientsenrollmentdate.

Prescription cost data wereavailableonly in aggregate as aone-timereportprovidedbytheinsurancecompany.Investigatorsdidnothaveaccesstocompleteprescriptionmedicationclaimsdata because GRIPA was not at financial risk for medication



5/13


expenses. Thus, no patient-level analyses of prescription datawereperformed.

Episode Treatment Groups (ETGs) for eachgroup were notavailableatthestartofthestudybutwerecalculatedbasedon

historicinformationbeforestudyanalysiswascompleted.ETGsidentify and quantify an episode of care that spans inpatient,outpatient,andallancillaryservices,includingpharmaceuticals,and takes into consideration patient age and comorbidities.18ETGs were believed to be important to include in the studyanalysesto determinethe degreeof similarity ofthe 2 groupsthroughoutthe studybecauseETGsare aclinically useful tooltomeasurehealthcaredemand.18

Statistics

Beforethestudy,interesthadbeenexpressedinlookingattheresponse variables by differentage groups as well as over the

entirepopulation,becausepublishedstudiesaboutclinicalphar-macistinterventionshavetypicallybeeninpatientswithchronicdiseaseandofteninolderagegroups. 1,10-12,15Twosubgroupsage 65 oryounger versusolder thanage 65were compared.Othersubgroupswerecreatedfor3agecategories20-50years,51-65 years, and older than age 65and the data for these3subgrouppopulationswereanalyzedseparately.

Categoricaldata(e.g.,rates,percentages)wereanalyzedusingthelikelihoodratiochi-squaretestfordifferencesinproportions,comparing the interventiongroup andcomparisongroup. Thevariables analyzedincludedsex, agecategory, weight category,andpresenceorabsenceofcomorbiditiesandriskfactorsinclud-ing congestiveheart failure (CHF), diabetes mellitus, coronaryartery disease (CAD), asthma, chronic obstructive pulmonarydisease,andcurrentcigarettesmoking.

Continuousdatawereexamined,usinghistogramsandscatterplots,todeterminedistributioncharacteristicsandrelationshipswith other variables.Normally distributed data were analyzedusing Students t-tests for 2-group differences. Non-normallydistributeddatawereanalyzedusingtheMann-WhitneyU-test,whichisanonparametrictestfor2-groupcomparisons.Baselinevariablesanalyzedwiththesemethodsincludedage,ETGs,andbodymassindex(BMI).Studyoutcomemeasureswereassessedusingadifference-in-differenceanalysisbysubtractingpre-inter-ventionvaluesfrompost-interventionvaluesandcomparingthe

changeamountsbystudygroup.Statisticalsignificancewasdetermined using an alpha level

of 0.05. Statistical analysis was performed using the Minitabversion13.32(MinitabInc.,StateCollege,PA)andSPSSversions13.0and14.0(SPSSInc.,Chicago,IL)statisticalpackages.

Results

Study EnrollmentCountsofeligiblepatientswhowereenrolledbetweenJuly2001and June 2002, patients excluded, and patients included inthefinaldatasetareshowninFigure1.Morethan80%ofthe

enrolledpatientsmetmorethan1riskfactordeterminedfrominsuranceclaims(datanotshown).Twopercentoftheenrolledpatientswereidentifiedbecausetheyhadadiagnosisofdiabeteswith no scheduled appointment during the first 6 months ofthestudy.Theonlypatientinthecomparisongroupwithasig-nificantfindingthatrequiredtheclinicalpharmacisttomakean

urgentrecommendationtothecomparisongroupphysicianwasexcludedfornothaving12monthsofcontinuousinsuranceeligi-bilityafterstudyenrollment.Thus,nopatientsinthecomparisongroupwerediscontinuedfromthestudysolelybecauseofclinicalpharmacistinteractionwiththecomparisongroupphysicians.Ofpatientswhometallthecriteriaforenrollmentintheprospectivephaseofthestudy(i.e.,ofthosewhowereassignedtoeithertheinterventiongroup[n=159]orthecomparisongroup[n=290]),exclusionsforfailuretomaintaincontinuousinsuranceeligibilityweremadefor30(18.9%)ofinterventiongroupand71(24.5%)ofcomparisongroupsubjects.

http://www.thedeltagroup.com/Corporate/Pubs/ETGs.pdfhttp://www.thedeltagroup.com/Corporate/Pubs/ETGs.pdfhttp://www.thedeltagroup.com/Corporate/Pubs/ETGs.pdfhttp://www.thedeltagroup.com/Corporate/Pubs/ETGs.pdf


6/13



Demographics

PatientdemographicsatstudyenrollmentareshowninTable2.Themean [SD] ageof patients in the intervention group (59.6[11.6])wasyoungerthaninthecomparisongroup(68.2[12.7];P65years 2.81 2.89 0.596 t-test

Weight category,b number (%) (n = 118) (n = 210)

Normalweight 16 (13.6%) 59 (28.1%) 0.003 chi-square

Overweight 24 (20.3%) 49 (23.3%) 0.532 chi-square

Obese 63 (53.4%) 92 (43.8%) 0.095 chi-square

Morbidlyobese 15 (12.7%) 10 (4.8%) 0.009 chi-square

BMI(kg/m2),mean[SD]c 32.06 [7.51] 28.45 [5.67] < 0.001 t-test

a P values were determined from independent 2-sample t-tests for continuous variables and likelihood ratio chi-square tests for categorical variables; the Mann-WhitneyU-test for 2 independent sample groups was used when the continuous variables were not normally distributed.b Normal weight = BMI 25 kg/m2, overweight = BMI 25.1-27.99 kg /m2, obese = BMI 28-39.9 kg/m2, morbidly obese = BMI > 40 kg/m2.c 4.7% and 2.3% of patients in the intervention and comparison groups, respectively, did not have calculated BMI measures because their height data were unavailable.

BMI = body mass index; CAD = coronary artery disease; CHF = congestive heart failure; chi-square = likelihood ratio chi-square test; COPD = chronic obstructive pulmonarydisease; ETG = Episode Treatment Group; kg/m2 = ratio of weight in kilog rams to he ight in meters squared; M-W = Mann-Whitney U-test; t-test = Students t-test forindependent groups.


7/13



interventionandcomparisongroupsdidnotdifferwithrespectto the studys primary outcome, change in PPPY medical cost(P=0.711).

SecondaryoutcomesaredisplayedinTable4.Bothbeforeandafterthe intervention, interventiongroup patients had a loweraveragenumberofPCPvisitsthandidcomparisongrouppatients.However, the between-group difference in the mean changein PCP visits from pre-intervention to post-intervention wasnotstatisticallysignificant(P=0.914).Fromthepre-interventionto the post-intervention periods, hospital admissions per1,000patientsincreasedfrom206.0to221.0(7.3%)intheinter-ventiongroupandfrom121.0to204.0(68.6%)inthecomparisongroup,althoughthebetween-groupdifferenceintheamountofchangefrompre-interventiontopost-interventiondidnotreachstatistical significance (P=0.329). ER visits per 1,000 patientsdeclinedby44.1%intheinterventiongroup(from127.0to71.0)and increased by 57.6% in thecomparison group(from 144.0to227.0);thebetween-groupdifferenceinthechangeamountsapproachedstatisticalsignificance(P=0.054).

Prescriptioncostwascomparedatanaggregatelevel,withno

statisticalanalysesavailable.Theinterventiongroupsprescriptionclaimscost(insuranceplan liability) increased by17.4%(from$105,000to$123,227),whereasthecomparisongroupsprescrip-tionclaimscostdecreasedby10.1%(from$90,135to$81,042).

Patients Older Than 65 Years o Age: Forpatientsolderthan65 years ofage, studygroupsdid not significantlydifferwithrespecttothestudysprimaryoutcome,changefrompre-inter-ventiontopost-interventioninmedicalcosts(datanotshown).However, the interventiongroupsaveragePPPY costincreased29.7%, whereas the comparison groups cost increased 65.8%frombeforetoaftertheintervention.ERvisitsdecreasedby1.6%

intheinterventiongroupandincreasedby60.4%inthecom-parisongroup.

Clinical Pharmacist Interventions: Theclinicalpharmacistmade271recommendationstotheinterventiongroupwithanaverage of 2.1 recommendations per patient versus 286 con-cealed recommendations for patients in the comparison groupwithan average of1.3 per patient. In the intervention group,189 (69.7%) of the recommendations were accepted, whereas47(16.4%)oftheconcealedrecommendationswereactedonbycomparison group physicians. Thus, assuming thatabout 70%of the concealed (comparison group) recommendations wereactionable(i.e.,wouldhavebeenacteduponbythecomparisongroupphysiciansiftherecommendationshadbeenmadeandnotconcealed),comparisongroupphysiciansidentified47of200,orabout23.5%,ofactionableopportunitiesontheirownwithouttheservicesofaclinicalpharmacist.

Figure2showsbroadcategoriesofrecommendationsacceptedintheinterventiongroupresultinginmoreoptimalcareforthosepatients.Table5providesspecificexamplesofrecommendationswithinthesebroadcategories.Interventiongrouppatientswere

morethantwiceaslikelyascomparisongrouppatientstohavemedication nonadherence issues addressed (85.7% vs. 40.0%,P=0.032),and6timesaslikelytohaveamedicationprescribedthatwasindicatedbutnotprescribedpreviously(72.6%vs.11.5%,P


8/13


were10timesaslikelytobeprescribedanoptimalmedicationfor their condition(60.0% vs. 5.9%, P


9/13


Wewereunabletoassessdifferencesindrugcostbetweentheinterventiongroupandcomparisongroupinthepresentstudybecausestudy enrollmentcriteriadid notrequire thatpatientshadprescriptiondrugcoverageduringanypartofthestudy.Theinsurerdidnotgrantaccesstoindividualprescriptionmedica-tionfinancialdatabecauseGRIPAwasnotatriskformedicationcosts.However,aggregatedpharmacyclaimscostdatasuggested

anincreasedcostintheinterventiongroup.Thiscostfindingissimilartothoseofothersimilarstudiesinwhichthepharmacistshad access to the patients medical records and did not limitpharmacistservicestoonediseasestate.Thesestudiesshowedatrendofslightlyhigherannualcostofprescriptionmedication(5.7%-8.6%)intheinterventiongroups.24-26Inthepresentstudy,despitetheclinicalpharmacistsabilitytolowerthecostofsomemedications, one of the most common recommendations wastostartanewmedicationwhenitwasindicatedbutpreviouslyoverlookedbythephysician.Thispatternpotentiallyincreasedmedicationcost.Theinterventiongroupwas6timesaslikelyas

thecomparisongrouptohaveanewmedicationstarted.SomemedicationsinitiatedduringthestudywerecalciumandvitaminDsupplementsforthepreventionortreatmentofosteoporosis,ordailyaspirinforpatientswithdiabetesmellitus,whichwouldnot beexpected tochange the overall prescriptionmedicationcosts.However,othermedicationswereinitiatedtotreathyper-lipidemia, provide ACE inhibitors for patients diagnosed with

diabetesmellitusorCHF,orassurethatCADpatientshadfreshsublingualnitroglycerin.

Uniquetothisstudywasthatoutcomesforusualcarewithregardtomedicationmanagementweredocumentedandcom-paredwithoutcomesfortheinterventioninaprimarycareprac-tice.Thisdesignprovidedgreaterunderstandingofwhatmighthave potentially been accomplished for the patients receivingusualcare,hadtheyreceivedtheservicesofaclinicalpharmacist.Figure2 showsthat many potentialopportunities appearedtoexistforphysicianstooptimizemedicationtherapy.Hanlonetal.also recorded concealed recommendations for a randomized

http://www.pubmedcentral.nih.gov/picrender.fcgi?artid=1884656&blobtype=pdfhttp://www.pubmedcentral.nih.gov/picrender.fcgi?artid=1884656&blobtype=pdf


10/13


control group and found that, similar to the present studysresults, 55.1% of intervention group and 19.8% of controlgroupphysiciansenactedtheclinicalpharmacistsrecommendedchanges.27

Inthepresentstudy,between-groupdifferencesintheratesofoptimized medication therapy mayhave contributed to thetrend in lower hospital admissions and ER visits for patientsprovidedwithclinicalpharmacistservices.Forexample,medi-cation nonadherence, leading topoordisease control,also canleadtoincreasedhospitalizationsandcanbeanimportantdriver

ofoverallmedicalcosts.28Althoughfindingsofsomestudiescallinto question the relationship between improved medicationadherenceandclinicaloutcomesorhealthcarecosts, 26,29otherstudies have found a beneficial effect ofadherence onclinicaloutcomes.20,30 Recognizing drug interactionsand adverse drugreactionsarepartof the expertise ofa clinicalpharmacistandmayhave contributedto minimizing ERvisitsin the interven-tiongroupasevidencedinothersettings. 20-22,25Forexample,thecomparisongroupinthepresentstudyincludedawomanolderthan 80 years of age who was prescribed a low-dose tertiary


TABLE 5 Specific Examples of Optimal Care Opportunities

Description o Optimum CareIntervention Type

Examples oRecommendations:

Untreated indication:Recommendationtostartamedicationforamedicalconditionthatiscurrentlyuntreatedbutconsideredastandardofcare

Statinsforpatientswithcoronaryarterydiseaseandlow-densitylipoproteincholesterolabove goal

Angiotensin-convertingenzymeinhibitorforpatientwithdiabetesandmicroalbuminuria

Cost:Recommendationforanequallyeffectivebutlessexpensivemedication

Useone-halftabletofahigherstrengthtabletofthesamemedicationtoachievethedose(e.g.,80mgofatorvastatin,one-halftabletdaily,insteadof40mgofatorvastatindaily).

Changeprescriptionto1tabletofahigherstrengthinsteadofmultipletabletsoflowerstrengthtoachievethedose(e.g.,40mgofatorvastatintwicedailyto80mgofatorvastatinoncedaily).

Optimal drug:Recommendationtoreplaceacurrentmedicationwithamoreappropriatemedicationbasedonpatientcharacteristics,comorbidities,andpharmacokineticorothercharacteristicsofthemedication

Glipizideispreferredoverglyburideinpatientaged71yearswithchronickidneydisease.

Switchfromalong-actingbenzodiazepine(flurazepam)toashorter-actingbenzodiazepinesuchasoxazepaminelderlypatientwithinsomnia.

Adverse drug reactions:Identificationofapotentialoractualadversedrugreaction

Forpatientwithprostatecanceronleuprolideacetate,considercalciumandvitaminDadministrationandbonedensitytestbecausethereisbonelossassociatedwithadministration

ofleuprolide.

Avoidpioglitazoneorrosiglitazoneinpatientwithstage3congestiveheartfailure.

Nonadherence:Evidencethatthepatientisnottak-ingthemedicationasprescribed

Addressnonadherencewithpatientswithosteoporosiswhohavestoppedfillingtheirprescriptionforalendronate.

Addressnonadherencewithapatientprescribedastatinwhosecholesterolhasincreaseddramaticallyyetnotbeenaddressedatpreviousappointments.

Drug monitoring:Identificationofinappropriatemedicationmonitoringandrecommendingappropriatemedicationmonitoring

Orderaserumpotassiumdeterminationforpatientstartedonhydrochlorothiazidemorethan1yearago.

Orderthyroid-stimulatinghormonedeterminationforpatientwithchangeinlevothyroxinedosemorethan3monthsagowhodoesnothavecurrentbloodworkdone.

Drug interactions:Identificationofclinicallyrelevantdruginteractionsorwarningofpotentialdruginteractions

Assurethatpatienttreatedforhypothyroidismandstartingoncalciumsupplementdoesnottakecalciumandlevothyroxinetogether.

Limitacetaminophendosingtolessthan2gmperdayinpatientonchroniccarbamazepine,which

caninduceacetaminophenconversiontotoxicmetabolite.Subtherapeutic dose:Recommendationforalterna-tivedosingforsomeoneonasubtherapeuticdose

Increaseangiotensin-convertingenzymeinhibitordosetogoaldosepercongestiveheartfailurestandards.

IncreasecalciumandvitaminDsupplementtoachieverecommendedtotaldailyintake.

Supradose:Recommendationforalternativedosingforidentificationofapatientprescribedadosethatisinappropriatelyhighorshouldideallybetitrateddownward

Startingdoseofniacinextended-releasetabletsat1,000mgisunlikelytobetoleratedbypatient;suggest500mgatbedtime.

Patienttakingconjugatedestrogens,0.9mgdailyattempttitratingestrogendosetominimumeffectivedoseforpostmenopausalsymptoms.

No indication:Recommendationtodiscontinueamedicationthatappearstolackanindication

Discontinueprotonpumpinhibitorinapatientrecentlydischargedfromhospitalwithnewprescriptionforaprotonpumpinhibitorwithoutagastrointestinalcondition.

Discontinue1mgfolicaciddailysupplementinapatientwhodiscontinuedoralmethotrexatemorethan1yearago.
http://www.amcp.org/data/jmcp/Contemporary%20Subjects-383-393.pdfhttp://www.amcp.org/data/jmcp/Contemporary%20Subjects-383-393.pdfhttp://www.amcp.org/data/jmcp/Contemporary%20Subjects-383-393.pdfhttp://www.amcp.org/data/jmcp/Contemporary%20Subjects-383-393.pdf


11/13



aminetricyclic antidepressant forsuspectedurge incontinence.Withinweeksofstartingthiscentralnervoussystemactivemedi-cation with anticholinergicactivity, shesuffered falls,resultinginhospitalizationforfracture.

Unlike much of the published literature about health caresystems such as the VA, this study took place in a typicalprimary care practice that did not have a common electronicmedical record platform. This study also involved a privatelyowned medical practice that was not associated with eitherapharmacyormedicalschool,unlikemanyofthestudiescon-ducted in ambulatory care pharmacist practice environmentswithin the United States.1,2,9,21,31-33 The clinical pharmacistsapproachusedinthepresentstudycouldpotentiallytakeplacein any community, in any doctors office, with little disrup-tion to workflow. Space is a precious commodity in primarycarepractices;usingthisparticularmodelwouldallowclinical

pharmaciststoworkinanytypeofspaceandflextheirscheduleaccordingtotheneedsofthemedicalpractice.

Incontrasttootherstudies,patientswhomay haveneededthemosthelpwithmedicationtherapywerenotexcluded. 1,3,30,33The IMPROVEstudy excluded patients who had a psychiatricillnessrequiringmentalhealthservices,poorunderstandingofwritten and spoken English, visual impairment and residencefar from the physician office, or no working telephone.19 Theonlyabilityrequiredforpatientsinthecurrentstudywasabilitytophysically make it to aphysicianoffice visit; therewere nootherlimits.

Limitations

First, the medical practices were selected, not randomized.Recruitingphysicianstoparticipateinthecomparisongroupwasachallengingtask,asthecomparisongroupphysiciansdidnotbenefitfromparticipating.Thephysiciansinthepresentstudyscomparisongroupwerelikelywillingtoparticipatebecausetheyhadanunderstandingofthevaluableroleofaclinicalpharma-cist;theyhadpastexperienceworkingwithclinicalpharmacistswhomanagedanticoagulationandprovidedmonthlyeducationsessionsonmedicationswithinahealthmaintenanceorganiza-tion.Neitherphysiciansinthesamepracticenorpatientswererandomized,whichmayhavebiasedtheresults.However,itdidpreventthecontaminationthatcouldhaveoccurredifasingle

physician had workedwith both interventionand comparisonpatients.Thiscontamination, althoughnotideal fora researchstudy,istypicallysomethingthatclinicalpharmacistsstriveforwithin a medical practice. Ideally, after a clinical pharmacistmakesarecommendation2 or 3times, the physiciantends toapplythisknowledgeappropriatelytotheremainderofsimilarpatientsinhisorherpractice.

Second,therearemajorconcernsaboutwhetherthepatientcohortswerecomparable,particularlybecauseofthedifferenceinage.Theinterventiongroupandcomparisongroupsdifferedatbaseline;ofpatientswith12monthsofpre-interventioneligibil-

ity,25.9%ofinterventionand45.3%ofcomparisonpatientswereaged66yearsorolder.Thepercentagesofstudypatientsexcludedfromthefinalanalysisfornothaving12monthsofcontinuousinsurance eligibility following the date of study enrollment

were similar in the intervention group (30 of 159 patients or18.9%)andthecomparisongroup(71of290patientsor24.5%).However,just15.6%oftheexcludedpatientsintheinterventiongroupwereaged65yearsoryounger,comparedwith66.2%inthe comparison group. This pattern appearedto bea resultofaninsurancechangetoaself-insuredproductmadeby1largeemployer in Rochester during this study, thus removing itsparticipantsfromthecapitatedpopulation.Theemployerchangeexcluded so many younger patients in the comparison groupthat the difference inmean age between the 2groupsbecameevenlarger.

Third,we madeana prioridecisionto excludeall patients

that did not have 12 monthsof continuous insurance eligibil-ityafterstudyenrollment;thusitisunknownhowtheclinicalpharmacist interventions affected those patients that subse-quently either died or disenrolled from the insurance plan.Fourth,themedicalcostdatacontainedsomeoutliercasesthatwerenotremovedfromourstudysamplebecauseofouraprioridecisiontoretainalleligiblecasesforfinalanalysis.Therewerenopatientswithtraumaormotorvehicleaccidents,butaverysmall number of patients in both the intervention and com-parison groups had extreme changes in 12-month medicalcosts; these changes ranged from a decrease of $1.8 millionfor1patientintheinterventiongrouptoanincreaseof$1.7mil-lionforanotherpatientinthecomparisongroup.Thesechargeswere attributable to hospitalizations for cancer treatments,congestive heart failure, and major surgeries including 1 livertransplant.

Fifth,thegeneralapplicationofthestudyfindingscouldbeaffectedbyseveralfactors.The69.7%acceptancerateofrecom-mendationsby physicianswas higherthan inmany publishedoutpatientstudies. 23,24,26,34Thisoutcomemayhavebeenattrib-utabletotheuseofonly1person,theclinicalpharmacistwhoperformed the intervention, to determine the acceptance rateineachofthe2studygroups.However,therelationshipsbuiltbetweentheclinicalpharmacistandphysiciansintheinterven-tion group over the 12 months probably played a role in the

successoftheinterventionasdemonstratedinotherstudiesinwhichauthorssurmisedthatinterpersonalrelationshipsbetweenthe pharmacist and physicians contributed to improved out-comes.2,36Althoughthepresentstudydidnotmeasurewhetheracceptance of recommendations resulted in resolution of theidentifiedproblems,theacceptancesdidreflectpositivecaredeci-sionsmovinginthedirectionofresolution.TheIMPROVEstudyauthorsstatedthat69%oftheirrecommendationswereresolved,but when they removed the interventions performed directlyby the pharmacist (withoutneeding physician approval), theirresolutionratedeclinedto57%.19
http://www.pubmedcentral.nih.gov/picrender.fcgi?artid=1884656&blobtype=pdfhttp://www.pubmedcentral.nih.gov/picrender.fcgi?artid=1884656&blobtype=pdf


12/13


Sixth, the study may have underestimated the benefits ofthe clinical pharmacist because one of the comparison groupphysiciansalsowasamemberofapharmacyandtherapeuticscommitteeforanotherlargeinsurerinRochester,NewYork,and

wasacutely aware ofmedicationrelatedproblemsandmoney-savingopportunities.Theaveragenumberofrecommendationsper patient in the intervention group versus the comparisongroup (2.1 vs. 1.3, respectively) might also have contributedto study findings. Lastly, the inclusion criteria for this studywereratherbroad.Asaresultofourfindings,wehavenarrowedthecriteriaforconsultation,limitingourtargetpopulationtothemosthigh-riskpatientswithmultiplecomorbidities.

Conclusion

Aclinicalpharmacistcanpromoteoptimalmedicationtherapyinoutpatientsbyworkingwithprimarycarephysicianswithin

theirofficepractices.Althoughthemedical(excludingpharmacy)costsoftheinterventionandcomparisongroupsdidnotdiffersignificantly, a nonsignificant trend suggests that the interven-tionmayhavehadapositiveeffectonmedicalcostsandwarrantsfurtherinvestigationwithalargersamplesize.

3.BorensteinJE,GraberG,SaltielE,etal.Physician-pharmacistcomanagementofhypertension:arandomized,comparativetrial.Pharmacotherapy.2003;23(2):209-16.

4.CordingMA,Engelbrecht-ZadvornyEB,PettitBJ,EasthamJH,SandovalR.Developmentofapharmacist-managedlipidclinic.Ann Pharmacother.2002;36(5):892-904.

5.CranorCW,BuntingBA,ChristensenDB.TheAshevilleproject:Long-termclinicalandeconomicoutcomesofacommunitypharmacydiabetescare program.J Am Pharm Assoc (Wash).2003;43(2):173-84.

6.NationalCommitteeforQualityAssurance.Health Plan Employer Dataand Information Set for 2007; Volume 2: Technical Specifications.Washington,DC:NationalCommitteeforQualityAssurance;2007.Availableat:www.ncqa.org/Portals/0/HEDISQM/Archives/2007/Vol2/Cost_of_Care_Update.pdf.AccessedOctober1,2008.

7.NationalCommitteeforQualityAssurance.HealthPlanEmployerDataandInformation2008SummaryTableofMeasures,ProductLinesandChanges.Availableat:www.ncqa.org/Portals/0/HEDISQM/HEDIS2008/2008_Measures.pdf.AccessedOctober1,2008.

8.MasonJD,ColleyCA.Effectivenessofanambulatorycareclinical

pharmacist:Acontrolledtrial.Ann Pharmacother.1993;27(5):555-59.9.JamesonJ,VanNoordG,VanderwoudK.Theimpactofapharmaco-therapyconsultationonthecostandoutcomeofmedicaltherapy.J FamPract.1995;41(5):469-72.

10.BlakeySA,Hixson-WallaceJA.Clinicalandeconomiceffectsofpharmacyservicesingeriatricambulatoryclinic.Pharmacotherapy.2000;20(10):1198-203.

11.MaloneDC,CarterBL,BillupsSJ,etal.Aneconomicanalysisofarandomized,controlled,multicenterstudyofclinicalpharmacistinter-ventionsforhigh-riskveterans:theIMPROVEstudy.ImpactofManagedPharmaceuticalCareResourceUtilizationandOutcomesinVeteransAffairsMedicalCenters.Pharmacotherapy.2000;20(10):1149-58.

12.CowperPA,WeinbergerM,HanlonJT,etal.Thecost-effectivenessofaclinicalpharmacistinterventionamongelderlyoutpatients.Pharmacotherapy.1998;18(2):327-32.

13.LaiLL,SorkinAL.CostbenefitanalysisofpharmaceuticalcareinaMedicaidpopulationfromabudgetaryperspective.J Manag Care Pharm.1998;4(3):303-08.Availableat: www.amcp.org/data/jmcp/research_v3_303-308.pdf.AccessedOctober1,2008.

14.GerberRA,LiuG,McCombsJS.Impactofpharmacistconsultationsprovidedtopatientswithdiabetesonhealthcarecostsinahealthmain-tenanceorganization.Am J Manag Care.1998;4(7):991-1000.Availableat:www.ajmc.com/files/articlefiles/AJMC1998JulGerber991_1000.pdf.

AccessedOctober1,2008.

15.AltavelaJL,GeraciL.Decreasedhealthcareutilizationcostsafteradditionofclinicalpharmacyservicesinaninternalmedicinepractice.Posterpresentedat:2001AmericanCollegeofPharmacyAnnualMeeting;October21-24,2001;Tampa,FL.

16.AltavelaJL,BarbeauL,SorrentoT,SmeenkK,HaasC.Physicianresponsetoclinicalpharmacyservicesinaprivatepracticesetting.Poster

presentedat:1999AmericanCollegeofClinicalPharmacyInternationalCongressonClinicalPharmacy;April11-14,1999;Orlando,FL.

17.StrandLM,MorleyPC,CipolleRJ,RamseyR,Lamsa mGD.Drug-relatedproblems:theirstructureandfunction.DICP.1990;24:1093-97.

18.ForthmanMT,DoveHG,WoosterLD.EpisodeTreatmentGroups(ETGs):Apatientclassificationsystemformeasuringoutcomesperformancebyepisodeofillness.Top Health Inf Manage.2000;21(2):51-61.Availableat:www.thedeltagroup.com/Corporate/Pubs/ETGs.pdf.AccessedSeptember20,2008.

19.EllisSL,BillupsSJ,MaloneDC,etal.TypesofinterventionsmadebyclinicalpharmacistsintheIMPROVEstudy:impactofmanagedpharma-ceuticalcareonresourceutilizationandoutcomesinVeteransAffairsmedicalcenters.Pharmacotherapy.2000;20(4):429-35.


JEANETTE L. ALTAVELA, PharmD, BCPS, is Manager, Pharmacy

Services; MATT K. JONES, BA, is Senior Managed Care Analyst; andMERRILEE RITTER, MS, is Statistical Consultant, Greater Rochester

Independent Practice Association, Rochester, New York.

AUTHOR CORRESPONDENCE: Jeanette L. Altavela, PharmD,

BCPS, Greater Rochester Independent Practice Association,60 Carlson Road, Rochester, NY 14610. Tel.: 585.922.1548;

E-mail:[email protected]

Authors

DISCLOSURESThis research was not funded. The authorsare employees of the GreaterRochesterIndependentPracticeAssociation.StudyconceptanddesignwereprimarilytheworkofAltavela,andAltavelaperformedallofthedatacol-lection.AltavelaandRitterinterpretedthedatawithassistancefromJones.

AltavelawroteandrevisedthemanuscriptwithsomeassistancefromRitter. TheauthorsacknowledgeJamesR. Tobin,whocontributedtothe studyconceptanddesign;PeterB.Zajkowski,whohelpedwithdatainterpretation;

andCurtisE.HaasandJuneF.Johnson,whocontributedtothemanuscriptrevision.

REfERENCES

1.GattisWA,HasselbladV,WhellanDJ,OConnorCM.Reductioninheartfailureeventsbytheadditionofaclinicalpharmacisttotheheartfailuremanagementteam:resultsofthePharmacistinHeartFailureAssessmentRecommendationandMonitoring(PHARM)study.Arch Intern Med.1999;159(16):1939-45.

2.BogdenPE,AbbottRD,WilliamsonP,OnopaJK,KoontzLM.Comparingstandardcarewithaphysicianandpharmacistteamapproachforuncontrolledhypertension.J Gen Intern Med.1998;13(11):740-45.
http://www.ncqa.org/Portals/0/HEDISQM/Archives/2007/Vol2/Cost_of_Care_Update.pdfhttp://www.ncqa.org/Portals/0/HEDISQM/Archives/2007/Vol2/Cost_of_Care_Update.pdfhttp://www.ncqa.org/Portals/0/HEDISQM/Archives/2007/Vol2/Cost_of_Care_Update.pdfhttp://www.ncqa.org/Portals/0/HEDISQM/HEDIS2008/%202008_Measures.pdfhttp://www.ncqa.org/Portals/0/HEDISQM/HEDIS2008/%202008_Measures.pdfhttp://www.amcp.org/data/jmcp/research_v3_303-308.pdfhttp://www.amcp.org/data/jmcp/research_v3_303-308.pdfhttp://www.ajmc.com/files/articlefiles/AJMC1998JulGerber991_1000.pdfhttp://www.thedeltagroup.com/Corporate/Pubs/ETGs.pdfmailto:[email protected]:[email protected]://www.thedeltagroup.com/Corporate/Pubs/ETGs.pdfhttp://www.ajmc.com/files/articlefiles/AJMC1998JulGerber991_1000.pdfhttp://www.amcp.org/data/jmcp/research_v3_303-308.pdfhttp://www.amcp.org/data/jmcp/research_v3_303-308.pdfhttp://www.ncqa.org/Portals/0/HEDISQM/HEDIS2008/%202008_Measures.pdfhttp://www.ncqa.org/Portals/0/HEDISQM/HEDIS2008/%202008_Measures.pdfhttp://www.ncqa.org/Portals/0/HEDISQM/Archives/2007/Vol2/Cost_of_Care_Update.pdfhttp://www.ncqa.org/Portals/0/HEDISQM/Archives/2007/Vol2/Cost_of_Care_Update.pdfhttp://www.ncqa.org/Portals/0/HEDISQM/Archives/2007/Vol2/Cost_of_Care_Update.pdf


13/13



20.FarrisKB,KumberaP,HaltermanT,FangG.Outcomes-basedpharmacistreimbursement:reimbursingpharmacistsforcognitiveservices,part1.J Manag Care Pharm.2002;(8)5:383-93.Availableat:www.amcp.org/data/jmcp/Contemporary%20Subjects-383-393.pdf.AccessedOctober1,2008.

21.ChiquetteE,AmatoMG,BusseyHI.Comparisonofananticoagulationclinicwithusualmedicalcare:anticoagulationcontrol,patientoutcomes,andhealthcarecosts.Arch Int Med.1998;158(15):1641-47.

22.PauleyTR,MageeMJ,CuryJD.Pharmacist-managed,physician-directedasthmamanagementprogramreducesemergencydepartmentvisits.AnnPharmacother.1995;29(1):5-9.

23.MurrayMD,HarrisLE,OverhageJM,etal.Failureofcomputerizedtreatmentsuggestionstoimprovehealthoutcomesofoutpatientswithuncomplicatedhypertension:resultsofarandomizedcontrolledtrial.Pharmacotherapy.2004:24(3):324-37.

24.SorensenL,StokesJA,PurdieDM,WoodwardM,ElliottR,RobertsMS.Medicationreviewsinthecommunity:Resultsofarandomized,controlledeffectivenesstrial.Br J Clin Pharmacol.2004;58(6):648-64.Availableat:www.pubmedcentral.nih.gov/picrender.fcgi?artid=1884656&blobtype=pdf .

AccessedOctober2,2008.

25.McLeanW,GillisJ,WallerR.TheBCCommunityPharmacyAsthmaStudy:astudyofclinical,economic,andholisticoutcomesinfluencedbyanasthmacareprotocolprovidedbyspeciallytrainedcommunitypharmacistsinBritishColumbia.Can Respir J.2003;10(4):195-202.

26.CarterBL,MaloneDC,BillupsSJ,etal.InterpretingthefindingsoftheIMPROVEstudy.Am J Health Syst Pharm.2001;58(14):1330-37.

27.HanlonJT,WeinbergerM,SamsaGP,etal.Arandomized,controlledtrialofaclinicalpharmacistinterventiontoimproveinappropriateprescrib-inginelderlyoutpatientswithpolypharmacy.Am J Med.1996;100(4):428-37.

28.SokolMC,McGuiganKA,VerbruggeRR,EpsteinRS.Impactofmedi-cationadherenceonhospitalizationriskandhealthcarecost.Med Care.2005;43(6):521-30.

29.HanlonJT,LindbladCI,GraySL.Canclinicalpharmacyserviceshaveapositiveimpactondrug-relatedproblemsandhealthoutcomesincommunity-basedolderadults?Am J Geriatr Pharmacother.2004;2(1):3-13.

30.SolomonDK,PortnerTS,BassGE,etal.ClinicalandeconomicoutcomesinthehypertensionandCOPDarmsofamulticenteroutcomesstudy.J Am Pharm Assoc. (Wash).1998;38(5):574-85.

31.IsettsBJ,BrownLM,SchondelmeyerSW,LenarzLA.Qualityasses smentofacollaborativeapproachfordecreasingdrug-relatedmorbidityandachievingtherapeuticgoals.Arch Intern Med.2003;163(15):1813-20.

32.WhitleyHP,FermoJD,ChumneyEC.5-yearevaluationofelectronicmedicalrecordflagalertsforpatientswarrantingsecondarypreventionofcoronaryheartdisease.Pharmacotherapy.2006;26(5):682-88.

33.OdegardPS,GooA,HummelJ,WilliamsKL,GraySL.Caringforpoorlycontrolleddiabetesmellitus:arandomizedpharmacistintervention.AnnPharmacother.2005;39(3):433-40.

34.DoucetteWR,McDonoughRP,KlepserD,McCarthyR.Comprehensivemedicationtherapymanagement:identifyingandresolvingdrug-relatedissuesinacommunitypharmacy.Clin Ther.2005;27(7):1104-11.

35.HaggertySA,CerulliJ,ZeollaMM,CottrellJS,WeckMB,FaragonJJ.Communitypharmacytargetinterventionprogramtoimproveaspirinuseinpersonswithdiabetes.J Am Pharm Assoc.2003;45(1):17-22.

36.LeapeLL,CullenDJ,ClappMD,etal.Pharmacistparticipationonphysicianroundsandadversedrugeventsintheintensivecareunit.JAMA.1999;282(3):267-70.
http://www.amcp.org/data/jmcp/Contemporary%20Subjects-383-393.pdfhttp://www.amcp.org/data/jmcp/Contemporary%20Subjects-383-393.pdfhttp://www.pubmedcentral.nih.gov/picrender.fcgi?artid=1884656&blobtype=pdfhttp://www.pubmedcentral.nih.gov/picrender.fcgi?artid=1884656&blobtype=pdfhttp://www.amcp.org/data/jmcp/Contemporary%20Subjects-383-393.pdfhttp://www.amcp.org/data/jmcp/Contemporary%20Subjects-383-393.pdf

Documents

Articulo Para Revisar