JOURNAL OF MANAGED CARE PHARMACY · JMCP volume seven • number four july/august 2001...

JMCPvolume seven • number four

july/august 2001

Peer-reviewed Journal of the Academyof Managed Care Pharmacy

AMCP Guidance for Submission of Clinical and Economic Evaluation Data to Support Formulary Listing in U.S. Health Plans and Pharmacy Benefits Management Organizations

Driving Market Share in an Integrated Health System without Therapeutic Interchange

An Investigation of Allergic Rhinitis, Asthma, and Medication Use in a Privately Insured Population

Collaborating with Community Pharmacists to Improvethe Quality of Diabetes Care in an IPA-model HMO

Prior Authorization Programs: A Critical Review of the Literature

Evaluating Asthma Medication Use Before and After an Acute Asthma-related Event

JOURNAL OF MANAGED CARE PHARMACY®

Page

272

283

292

297

287303

C O N T E N T S

272 AMCP Guidance for Submission of Clinical and Economic Evaluation Datato Support Formulary Listing in U.S. Health Plans and Pharmacy BenefitsManagement OrganizationsSean D. Sullivan, Alan Lyles, Bryan Luce, and Joseph Grigar

283 Driving Market Share in an Integrated Health System without TherapeuticInterchangeJoseph F. Fischer, Robert M. Mowers, David J. Ormerod, and Ellen S.Burriss

287 An Investigation of Allergic Rhinitis, Asthma, and Medication Use in aPrivately Insured PopulationJodi Crystal-Peters, Cheryl Neslusan, and Amy White

292 Collaborating with Community Pharmacists to Improve the Quality ofDiabetes Care in an IPA-model HMODavid P. Nau, Joshua D. Blevins, and Stephen E. Neal

297 Prior Authorization Programs: A Critical Review of the LiteratureNeil J. MacKinnon and Ritu Kumar

303 Evaluating Asthma Medication Use Before and After an Acute Asthma-related EventJung H. Lee, Sandra D. Cassard, Peter E. Dans, Clare Wheelock, andJoseph D. Ober

D E P A R T M E N T S

249 Cover ImpressionsDesert Eloquence (1994)Bo Newell

253 PerspectivesZero Tolerance

255 SpotlightFocus on Drug Utilization

257 Caveats Getting a Quick Fix Online

259 CampusAn Inside Look at the Benefits of aStudent Pharmacy and Therapeutics

(P&T) Committee Competition fromthe University of Illinois at Chicago

309 Continuing EducationA Prescription for Change: Bridgesto Cross the “Quality Chasm”Linda L. Norton

315 CE Exam

321 Author’s Guidelines

323 AMCPROGRESS

AMCP/USP Receive Grant ForInitiative

325 Membership Application

EDITORIAL MISSION

JMCP is dedicated to providing managed care pharmacists, associates, and students with the toolsto excel in their daily practices by focusing on:Policy: Providing a forum for in-depth discussion of

issues of topical and long-term importance.Practice: Presenting information of interest and

educational value to the membership.Research: Publishing research that increases the

quality of research standards used in managedcare pharmacy practice and helps apply thatresearch to improve the practice of managed carepharmacy.

EDITORIAL STAFF

Editor-In-ChiefCraig S. Stern, R.Ph., M.B.A., Pharm.D., ProPharma

Pharmaceutical Consultants, Inc., Northridge, CAPublisherJudith A. Cahill, Executive Director • AMCP

HeadquartersContributing EditorsTracy S. Hunter, Ph.D.—AcademicsPerry Cohen, Pharm.D.—FeaturesShane Desselle, Ph.D.—CampusLinda Norton, Pharm.D.—Continuing EducationRenwyck Elder, R.Ph., M.B.A.—SpotlightJ. Warren Salmon, R.Ph., M.B.A.—InternationalDiane B. Ginsburg, M.S., R.Ph., F.A.S.H.P.—CaveatsCeleste d’Elliott—Cover ImpressionsJennifer F. Bloom, M.S.W. —AMCProgressEditors’ Review PanelCraig S. Stern, R.Ph., M.B.A., Pharm.D.Tracy S. Hunter, Ph.D.Diane B. Ginsburg, R.Ph., M.S., F.A.S.H.P.Editorial Advisory Board

Robert J. Anderson, Pharm.D., Mercer University,Marietta, GA

Jean Brown, Ph.D., AdvancePCS, Fountain Hills, AZJoan Deady, Pharm.D., Sutter Health, Sacramento, CAShane Desselle, Ph.D., Duquesne University,

Pittsburgh, PAColonel George J. Dydek, Pharm.D., U.S. Army,

Aberdeen Proving Ground, MDLeslie Fish, Pharm.D., Fallon Healthcare System,

Worcester, MARichard E. Geller, R.Ph., CIGNA HealthCare,

Tarzana, CADiane B. Ginsburg, R.Ph., M.S., F.A.S.H.P., University of

Texas at Austin, College of Pharmacy, Austin, TXAlan Heaton, Pharm.D., Prime Therapeutics, Eagan, MNTracy S. Hunter, Ph.D., Nova Southeastern University,

Fort Lauderdale, FLEric G. Klein, Pharm.D., Eli Lilly & Co.,

Indianapolis, INNeil MacKinnon, Ph.D., Dalhousie University, College

of Pharmacy, Halifax, Nova Scotia, Canada Daniel C. Malone, Ph.D., R.Ph., College of Pharmacy,

University of Arizona, Tucson, AZDarlene Mednick, R.Ph., M.B.A., Merck-Medco

Managed Care, LLC, Franklin Lakes, NJMichael J. Sax, Pharm.D., The Pharmacy Group,

LLC, East Glastonbury, CTFounding EditorLouise J. Sargent, M.S., R.Ph.

Journal of Managed Care Pharmacy(ISSN 1083–4087) is peer-reviewed and publishedbimonthly by the Academy of Managed CarePharmacy, 100 North Pitt Street, Suite 400, Alexandria,VA 22314; 703-683-8416; 800/TAP-AMCP; 703-683-8417 (fax). Postmaster: Send address changes to 100North Pitt Street, Suite 400, Alexandria VA 22314.

www.amcp.org

? FEATURES

263 Women’s Health: Issues and Opportunities for Managed CarePharmacyAnn M. McNamara and Mary Claire Wohletz

268 Employee Benefits Consulting: An Essential Role for PharmacyConnie Perry

Volume 7, No. 4

246 Journal of Managed Care Pharmacy JMCP July/August 2001 Vol. 7, No. 4

painting within a painting is thesignature of Bo Newell. HereNewell legibly discloses the aus-

tere, powerful elements of the desertwhile captivating his audience with atraveling herd of addax.

Because addax are one of the world’srarest mammals, they have been intro-duced into areas beyond their naturalSahara Desert habitat. Newell actuallyobserved this herd of addax on a gamereserve. The addax survive on very littlewater and have large, wide hooves foreasy travel through the desert; hunters arelargely responsible for the reduction intheir numbers. In Desert Eloquence,Newell casts a discerning eye on theexpedition of this herd; his mixture ofbeauty and stark reality introduces anaphoristic, sobering view. He draws theparallel between the harsh elements andthe attempted evisceration of this desert-dwelling antelope.

While Newell has been influenced bysuch wildlife artists as Bob Kuhn andDavid Shepherd, he also credits SalvadorDali, Claude Monet, and FrederickRemington for affecting his personal artis-tic development.

It is easy to see why Newell believesthat artists are born, not converted. At theage of five, he declared his love for animalswhile simultaneously making his debutinto the art world. He created a clay sculp-ture of a turkey that was exhibited at theBoston Museum of Fine Arts. During highschool, his dual fascination of art and ani-mals continued and he won two covetedawards for his wildlife creations.

In 1974, after receiving his B.F.A. infine arts from Texas Tech University,Newell traveled to Africa. His steadfastadmiration and respect for wildlife founda home there; the trip informed him witha sense of responsibility for animals thatare endangered species. Subsequent tripsto Africa confirmed his calling, and,today, Newell’s advocacy for wildlife con -

servation extends far beyond the Africancontinent.

In 1976 Newell had his first one-manshow in Houston. Its success ended hisamateur status. In 1980 his work waspart of an international art show inLondon held at the Grosvenor HouseHotel. Currently, his work is exhibited bygalleries throughout the United States;limited edition prints are handled byVirtual Gallery—Archetype Publishing inLos Gatos, California.

Bo Newell lives and works inHouston; his studio is a diversionarysymposium. While wildlife remains at thecore of Newell’s passion, he punctuateshis oeuvre with landscapes of Africa,Europe, Mexico, and the southwesternUnited States. Newell also paintsseascapes and spherical themes. Mindfulof his childhood success, he continues tosculpt. Yet although he is accomplishedin depicting many venues, Africa remainsa special location for him personally andprofessionally.

For the past 10 years, though Newell’sart has continued to deal with animalsfrom many areas of the globe, he hasintroduced a surreal approach. Yet he isritualistic about authenticity, often takinghundreds of photographs to capture themovement and traits of each species.

To visit the African continent is toobserve different inhabitants, territories,cultures, and wildlife. To visit the art ofBo Newell is to witness the eloquent andexotic reconciliation of art and life. ?

Celeste d’ElliottJMCP Contributing Editor

Author Correspondence: [email protected]

Cover Credit

Bo Newell. Desert Eloquence, oil on canvas.Houston, Texas. Copyright 1994.

Source

Interview with the artist.

C O V E RI M P R E S S I O N S

Desert Eloquence (1994) ? ?Bo Newell

A

AMCP HEADQUARTERS

100 North Pitt Street, Suite 400 Alexandria, VA 22314Tel: 703-683-8416Fax: 703-683-8417AMCP Staff Editor and Liaison:Jennifer F. Bloom, M.S.W., ext. 308

BOARD OF DIRECTORS

President: Cynthia J. Pigg, R.Ph., M.H.A.,CIGNA HealthCare, Richmond, VA

President-Elect: C.E. (Gene) Reeder, R.Ph.,Ph.D., University of South Carolina,Columbia, SC

Past President: John D. Jones, R.Ph., J.D.,Prescription Solutions, Costa Mesa, CA

Treasurer: Peter M. Penna, Pharm.D., P.M.Penna, LLC, Farmington, CT

Director: Michael E. Bailey, R.Ph., MedImpactHealthcare Systems, San Diego, CA

Director: James R. (Rusty) Hailey, M.B.A.,Coventry Health Care, Inc., Franklin, TN

Director: Dianne Kane Parker, Pharm.D.,Pharmacia Inc., Irvine, CA

Director: Debbie Stern, R.Ph., Rxperts, Irvine, CA

ADVERTISING

Advertising for Journal of Managed CarePharmacy is accepted in accordance with theadvertising policy of the Academy of ManagedCare Pharmacy.

For advertising information, contact:Professional Media Group, Inc., P.O. Box 18940 N. Woodbury Road, Pitman, NJ 08071Tel: 800-486-5454 or 856-589-5454 Fax: 856-582-7611

EDITORIAL

Correspondence related to editorial contentshould be mailed to:

Mitchell Petersen, Inc., 1775 Jamieson Avenue,Suite 210, Alexandria, VA 22314Tel: 703-518-4700; Fax: 703-518-8495 Managing Editor: Jeanne BurkeProduction Manager: Laura Mahoney

SUBSCRIPTIONS

Annual Subscription Rates: USA, individuals,institutions–$60; other countries–$80. Singlecopies cost $10. Missing issues replaced free ofcharge only up to six months after date ofissue. Send requests to AMCP headquarters.

REPRINTS

For article reprints, contact Diana Sholl,Reprint Management Services, 717-560-2001,x162. Microfilm and microfiche editions ofJournal of Managed Care Pharmacy are avail-able from University Microfilms, 300 N. ZeebRoad, Ann Arbor, MI 48106.

All articles published represent the opinions of theauthors and do not reflect the official policy of theAcademy of Managed Care Pharmacy or theauthors’ institutions unless so specified.

Copyright© 2001 Academy of Managed CarePharmacy, Inc. All rights reserved. No part of thispublication may be reproduced or transmitted inany form or by any means, electronic or mechan-ical, without written permission from theAcademy of Managed Care Pharmacy.

A b o u t o u r c o v e r a r t i s t

Vol. 7, No. 4 July/August 2001 JMCP Journal of Managed Care Pharmacy 249

P E R S P E C T I V E SR e f l e c t s t h e e d i t o r ’ s v i e w p o i n t o n i m p o r t a n t i s s u e s i n m a n a g e d c a r e p h a r m a c y

he health care community,including pharmacy, has beenforced to face the issues of

errors, medication errors, and fallibility.Unfortunately, the stimulus was not fromwithin, but from a government study: theInstitute of Medicine’s (IOM’s) report, ToErr is Human. Building a Safer HealthSystem.1 The world of health care wasalerted by payors, patients, and the gov-ernment that it was expected to performlike other industries that abhor even oneerror. Their expectation is that health careis too important to be practiced within asystem that is not dedicated to perfection,and that demonstrates its disdain bydelivering poor service. Health careproviders have answered the IOM studynot by rapidly embracing a culture ofchange, but by debating the accuracy ofthe numbers. Is health care really so dif-ferent from other industries?

We are surrounded by systems that aredesigned to deliver accurate services everytime. We expect telephones to deliver a dialtone, computers to boot up every time, andautomatic teller machines to deliver thecorrect amount of cash. We expect waitersto give courteous and excellent service,service stations to repair a car correctly thefirst time, and accountants to provide error-free tax forms. Unfortunately, in the healthcare arena errors in service delivery happenroutinely. The difference is that other indus-tries and organizations have a policy of zerotolerance for error. They design and imple-ment their processes to always deliver out-standing performance.

Health care is unique. Patients are com-plicated biological organisms with fre-quently unpredictable problems. The man-agement of medical problems requiresyears of study and practice treating numer-ous patients. Patients and payors accept theinevitable consequences of individualizedcare and unpredictable patient problems.Yet while emphasis is placed on the clinicalaspects of health care, little attention ispaid to the processes of care. Are thereorganizations that are as complex as

health care that have learned how tomanage processes to deliver uncompro-mising service every time?

Organizations do exist that managesystems every bit as complex as healthcare. They employ high-risk technologiesthat must be operated with maximumaccuracy. Errors and “bad luck” in theseorganizations can lead to disrupted oper-ation, destruction of major equipment,and even death. Two researchers studiedsuch organizations and identified the ele-ments of their success. Health care canlearn a great deal from their research.

Todd La Porte of the University ofCalifornia at Berkeley studied the opera-tions of nuclear-powered aircraft carriers,air-traffic control, large power-distribu-tion grids, and nuclear power plants.These systems are as complex as any youcan find. They cannot afford errors, badluck, or statements of “It’s not my job,” or“It’s not my responsibility.” As a result,they have developed management sys-tems that minimize errors (i.e., one erroris too many). These systems workbecause of management systems thatmorph between traditional hierarchicalsystems and loosely organized structuresbased on collaboration. For example, LaPorte studied how aircraft carriers canlaunch so many fighter planes, so quick-ly, with so few errors. He learned thatwhile the military is still highly regiment-ed and is the quintessential hierarchicalorganization, each and every member ofthe team has the power and the responsi-bility to shut down operations immedi-

ately if the circumstances warrant. Teammembers are continually talking to oneanother, sharing what they have doneand what they will be doing. Youths intheir early 20s are constantly trained intheir very specialized jobs. Training isbased on years of experience that haveproduced collections of best practices,defined by thick manuals of standardoperating procedures, whose purpose isto extend the organization’s control overas many eventualities as possible.2

How do these systems respond todynamic situations? They train, train, andtrain some more. They encourage “peopleto work together across the system toanticipate and avoid problems, so thatevents that cannot be controlled inadvance by following the rules are effec-tively dealt with on the fly.”2

Kathleen Eisenhardt of StanfordUniversity studied organizational character-istics that help companies make more effec-tive decisions in rapidly changing environ-ments. Eisenhardt’s organizations sharecommon elements with health care. Shefound that these organizations divided jobsinto highly specialized areas, emphasizedconstant communication and monitoring ofinformation, and were organized aroundcomposites of centralized and decentralizedstyles. These systems provide examples ofcomplex, dynamic systems that havelearned to function, thrive, and minimizeerror through constant communication,training, and evolving procedures.2

Additional examples of the commit-ment to zero tolerance are provided byPerot Systems, siteROCK, Applied Inform-ation Technologies, and NOCpulse. Eventhough these companies operate understrict military-style top-down proce-dures, they are committed to a zero toler-ance for errors. In their world, customersecurity is paramount. They emphasizetraining, process rules (strict processes orstandard operating procedures), wargames (detailed planning blueprints andtroubleshooting sessions), and mettletesting (interrogating managers to see

TZero Tolerance


Organizations develop

processes to minimize

risk once they have

developed a culture of

zero tolerance.

how they handle themselves under pres-sure).3

Organizations develop processes tominimize risk once they have developeda culture of zero tolerance. Six Sigma, or3.4 defects-per-million (i.e., the quest forzero defects), is the latest quality trend inbusiness. The emphasis is on continuousimprovement in reducing waste, ineffi-ciency, and variability. The results arehigher-quality products and services at alower cost. Six Sigma is being implement-ed with spectacular results by GeneralElectric, Lear Corporation, AmericanExpress, 3M, Toshiba, General Motors,and Ford to name just a few companies forwhich it is a major emphasis of competi-tive strategy. The hallmarks of Six Sigmaare training, DMAIC (define, measure,analyze, improve, control), and commit-ment. For example, 3M’s quality program,PPU (process and product understand-ing), includes four phases: understandingthe customer, understanding key qualitycharacteristics, establishing capabilities,and improving continuously. Commu-nicating concepts and information across

3M is the key to the program. A blueprint for health care is provided

in the elements necessary for any SixSigma program. One of the Six Sigmatraining companies, Sigma BreakthroughTechnologies, Inc., has summarized theseelements as:• clear need, strategy, and monetary goals

for improvement;• management that is involved;• procedures for selecting strategic

improvement projects;• good improvement methodology;• user-friendly software to implement the

improvement methodology;• dedicated and trained resources;• periodic reviews by various levels of

management;• communication of need and benefits;

and• recognition, rewards, and celebration.

The lessons to be learned from theseorganizations are to establish a culturethat is dedicated to driving error, ineffi-ciencies, and waste out of the system; andto ensure that outstanding customer serv -ice is an expectation from which any

variance is not tolerated. Health caremust adopt similar philosophies, and inthis cannot differ from these organiza-tional examples.

Managed care could do worse than tocommit to improving health care bychanging the culture from cost contain-ment to zero tolerance for anything shortof perfection in patient care. Costimprovements will then be a conse-quence, rather than a goal. ?

Craig Stern, R.Ph., Pharm.D., M.B.A.,Editor-in-Chief. Author Correspondence:[email protected].

References

1. Kohn LT, Corrigan JM, Donaldson MS, eds. Toerr is human. Building a safer health system.Committee on Quality of Health Care in America.Institute of Medicine. Washington: NationalAcademy Press, 1999.

2. Pool R. In the zero luck zone. Forbes ASAP.November 27, 2000, 85.

3. Information technology. Business Week, January29, 2001, 125–26.

Perspectives


ver the past decade the growthand evolution of interest inhealth outcomes research, espe-

cially pharmacoeconomic research, hasbeen tremendous. For those working inmanaged care, the interest is still explod-ing. The obvious reason is that we allwant to find the best value for our dollar.It was this desire and the need it reflectsthat fueled the creation and progress of theCenter for Pharmacoeconomic Studies atthe University of Texas.

The center was opened seven years agowhen faculty at the university decided thatthe best approach to studying the econom-ics and management of pharmaceuticals andthe related health outcomes was through amultidisciplinary team of researchers. Thus,the center combines the strengths of two dif-ferent research groups: clinical pharma-cotherapy practitioners and research spe-cialists, and researchers whose skills lie inthe methods, statistics, economics, manage-ment, and social behavioral issues related todrug use. The merging of these two researchgroups has been more than successful; it hasbeen synergistic.

The center brings together researchersfrom different disciplines and health caresectors and encourages their collaborationwith other health care providers and otheracademic institutions. The result is that thecenter’s researchers have investigated a widevariety of issues. Some examples are: • estimating the cost of Alzheimer’s disease

in a managed care setting; • measuring the impact of direct-to-con-

sumer prescription-drug advertising; • measuring the impact of a nonsteroidal

anti-inflammatory drug (NSAID) algo-rithm in managed care;

• comparing the cost-effectiveness of selec-tive serotonin reuptake inhibitors (SSRIs),antibiotics, and other agents; and

• examining the extent to which Mexicanprescription drugs enter the United States.

The center conducts clinical, economic,and policy research on how pharmaceuti-cals and pharmacy services affect patients’health care outcomes. For example, a recent

study by center researchers has shown thatincreasing the use of newer-generationantipsychotic medications within a popula-tion may lead to fewer costly inpatient hos-pitalizations. Another group found thatusing a lower dosage of a third-generationcephalosporin than usual to treat otitismedia minimized costs while achieving thesame outcomes.

One example of center research thathas had an impact on U.S. health care pol-icy was the study measuring the extent towhich Mexican prescription drugs wereentering the United States. This researchwas a major factor in the decision by U.S.authorities to ban importation of fluni-

trazepam (Rohypnol).As the director of the Center for

Pharmacoeconomic Studies, I have hadmany inquiries about the cost-effectivenessor benefit of pharmaceuticals and the costimpact of pharmacy services. Despite all thecurrent concern about the rise in pharma-ceutical expenditures in the public or pri-vate health care systems, I am convincedthat pharmaceuticals overwhelmingly pro-vide a great return on investment.

We all know that pharmaceuticalexpenditures have had double-digit annualgrowth in most health care plans. A recentstudy by the center investigating the causeof drug expenditure growth within theTexas Medicaid Drug Vendor Programshowed that the top three reasons for theincrease over the past three years havebeen: (1) the use of new pharmaceuticalproducts, (2) increased utilization of exist-ing drugs, and (3) price inflation. The over-whelming drivers were factors (1) and (2),which accounted for 80% of the increase:

patients taking more drugs, patients com-plying more with their therapies, andpatients using the newest—which are usu-ally the most expensive—products. (Theseresults do not differ greatly from those ofother managed care studies.)

Unfortunately, the component ofexpenditures receiving the most publicattention recently has been product cost(pricing). While this component is impor-tant, we should not curtail our efforts toensure that products, whatever their price,are used appropriately. Both over- andunder-utilization of drug products lead toinefficient delivery of health care, and thusultimately to higher costs.

Making sure that the right patient getsthe right dose of the right drug at theright time should be the responsibilitynot only of physicians but also of phar-macy benefit managers, pharmacists, andpatients. Monitoring and ensuring appro-priate utilization is perhaps the biggestchallenge that now confronts the drug-delivery process.

These results point to future direc-tions for center researchers. Questions tobe investigated by the center’s researchersinclude: • Is the right drug being given to the right

patient and is it being used correctly?• To what extent do drug therapies reduce

or increase other health care costs?• Are there less costly alternatives that do

not sacrifice efficacy and quality of care?A specific example of such current

center research deals with the question,“To what extent do patients have injuriesdue to falls as a result of urinary inconti-nence?” The urge to urinate may cause

S P O T L I G H T

O

I n n o v a t i v e M a n a g e d C a r e P h a r m a c y P r a c t i c e

Managed care plans must find ways not only

to control the rising costs of pharmaceuticals,

but also to ensure the correct use of drug

products provided.

Focus on Drug Utilization


patients to hurry, be careless, stumble,and fall, thus suffering injuries. Thoughthere is a plethora of anecdotal evidencethat this does occur, objective researchdata are scarce. Do drug therapies for uri-nary incontinence reduce the incidenceof falls and thus reduce the probability offractures or other costly injuries to theelderly? In other words, are such thera-pies cost-effective?

Another area where center researchersare devoting more time is measuring theimpact of pharmacy-based value-addedservices. Such services usually increase

patient drug compliance and thus drugutilization and costs, but they also affectthe use of other health care services. Forexample, consider diabetes care servicesprovided by community pharmacists.Center researchers have shown that com-munity pharmacists who provide high-quality diabetic counseling services reducehemoglobin A1c values in their patientsnearly 1% over a 12-month period. In thissame study, patients who were enrolled inthe program reduced their overall healthcare costs by an average of $2,382 becauseof fewer hospitalizations.

Managed care plans must find waysnot only to control the rising costs ofpharmaceuticals, but also to ensure thecorrect use of drug products provided.They must measure the outcomes of theproducts. Researchers at the Center ofPharmacoeconomic Studies are dedicatedto this goal. ?

Marv Shepherd, Ph.D., is Director, Centerfor Pharmacoeconomic Studies, College ofPharmacy, University of Texas, Austin.Author Correspondence:[email protected].

Spotlight


A n U p d a t e o n t h e L e g i s l a t i v e a n d R e g u l a t o r y I s s u e s F a c i n g P h a r m a c y P r a c t i c e

o you know that right now mostconsumers can purchase anyprescription medication online

without a prescription? That’s right, anyconsumer with a valid credit card can pur-chase lifestyle-enhancing medications silde-nafil (Viagra) and finasteride (Propecia).They can also purchase controlled sub-stances such as hydrocodone/acetamino-phen (Vicodin), acetaminophen/codeine(Tylenol #3), or even diazepam (Valium).Gaining access to injectible testosterone andother steroids is easy. Unscrupulous entre-preneurs have even written manuals (avail-able for purchase) that describe how toexploit the Internet for prescription med-ications.

How did this virtual deregulation of pre-scriptions occur? The problem stems fromtwo sources: Pharmacies or entities based inother countries whose laws and regulationsare less stringent than those in the UnitedStates operate freely on the Internet; andunscrupulous operators everywhere claimto diagnose conditions online without everhaving met the patient. These “online phar-macies” are not pharmacies at all, becausethey do not conform to U.S. state and feder-al regulations. Most importantly, they placethe consumer’s health in jeopardy.

I went online and found various for-eign pharmacies offering manuals forpurchase in the Caribbean, Mexico, Asia,the Philippines, and many other places. Ipaid $29.99 to order each of these man-uals. I quickly received five glossybrochures that detailed Web sites and e-mail addresses of international pharma-cies. These pharmacies had agreed toprovide consumers prescription medica-tions, no questions asked. I was able topurchase diazepam (Valium), hydroco-done/acetaminophen (Vicodin), testos-terone (Deca-Durabolin), and many othermedications. I ordered 30 diazepam 10mg tablets, 30 hydrocodone/acetamino-phen tablets, and 30 testosterone tablets.My orders totaled $200.

It took nearly six weeks to receive thetestosterone tablets, which arrived in a

manila envelope addressed by hand. Theenvelope was postmarked in Thailand.

Five months later I received theenvelopes that should have contained theValium and the Vicodin. The envelopeshad been opened, were emptied, and car-ried a bright green sticker that read“Examined by U.S. Customs.” A note(that had been photocopied so much thatit was barely legible) explained thatreceiving the medications violated vari-ous customs laws.

One month later I received a formalnotice from the Department of theTreasury, U.S. Customs Service, that themedications I had purchased were seized.The notice said that I could forfeit themedications, no questions asked, or sub-mit a valid reason (from my physician)stating my need for the medications.

I did neither. I called U.S. Customsand described “my experiment.” Since Iwanted to help, I forwarded the manualsI had purchased to the customs office.

This experience helped me realize thatthe reason the testosterone tablets were notseized was likely because they were in asmall package. The diazepam and hydroco-done/acetaminophen tablets had been sentin a bulk bottle that could be readily iden-tified. I wondered whether the customsoffice’s intervention was chance, spurred byrecent media attention, or part of a concert-ed effort to control rampant abuse.

The most worrisome experience I hadwas with the Mexican pharmacies that Icontacted via e-mail. I simply stated that Iwas interested in purchasing 10 fluni-trazepam (Rohypnol) tablets. Rohypnol is abenzodiazepine that commonly known asthe “date-rape” drug. The e-mail response Ireceived read, “Sorry all out, check backnext week.”

What about these so-called pharma-cies that offer Viagra, Propecia, andXenical without a prescription? Thesesites aren’t difficult to locate. They claimthat from an online questionnaire, theycan diagnose problems and assess theneed for the medication. No physical

examination or prescription required.Anyone can answer these questionnaires,falsely if necessary. The message here isthat the questionnaires are not reviewed;what is reviewed carefully is the cus-tomer’s credit card number. Some ques-tionnaires even have all of the correctanswers pre-selected, to simplify theordering process.

These “diagnosing” pharmacies haveundergone a great deal of scrutiny andmany lawsuits have been filed against them.Individual states have put in place regula-tions prohibiting these entities from sellingprescription medications to residents with-in that specific state. Unfortunately, not allstates have such restrictions.

Even with regulation, enforcement isdifficult. These sites can operate for a shorttime, then simply change URLs when theycome under scrutiny. Our state and federallaws do nothing to identify and track theseWeb sites. While the pharmacy professionis indeed over-regulated, federal regulationsand funding could help identify these oper-ators and impose heavy penalties. Our cur-rent state board investigators do not havethe training, expertise, or time for thisresponsibility.

Unregulated purchasing of prescrip-tion drugs online is a big problem. Whyaren’t we hearing more about it? That’sthe scary part: because consumers whofrequent these sites are not willing toreport them. ?

Candy Tsourounis, Pharm.D., is AssistantClinical Professor and Acting Director, DrugInformation Analysis Service, University ofCalifornia, San Francisco, School of Pharmacy.Author Correspondence: [email protected].

C A V E A T S

D

P o l i c y , l e g i s l a t i v e , a n d r e g u l a t o r y i s s u e s a n d t r e n d s

Getting a Quick Fix Online


C A M P U SR e s e a r c h a n d i n n o v a t i v e m a n a g e d c a r e c u r r i c u l a a t s c h o o l s o f p h a r m a c y

An Inside Look at the Benefits of a Student Pharmacy and Therapeutics (P&T)Committee Competition from the University of Illinois at Chicago

harmacy students across the coun-try are faced with a dilemma. Howcan students enrich their under-

standing of the complexity of pharmacypractice beyond what is taught in the class-room? Many students obtain part-time jobsand summer internships, but often thesejobs do not provide the students with thedeeper hands-on experience they need.More professional development is sought,usually outside the classroom.

Five years ago, the University of Illinoisat Chicago (UIC) College of PharmacyAcademy of Managed Care Pharmacy(AMCP) Student Chapter developed aPharmacy & Therapeutics (P&T) Commit-tee Competition to give students a “behindthe scenes” experience of the inner workingsof a P&T committee. Such a committee inmany different pharmacy environmentsenables pharmacists to demonstrate theirclinical expertise in a critical, expanded rolevis-á-vis their physician colleagues. The pri-mary mission of the P&T committee is topromote cost-effective and rational drugtherapies through a formulary-managementsystem. In doing so, the P&T committeerecommends policy to medical staff andadministrators about the desired therapeuticuse of drugs and other health-related issues.The P&T committee can be considered thebackbone of appropriate drug usage forhospital and long-term care pharmacies,pharmacy benefits managers (PBMs), andmanaged care organizations (MCOs).

The P&T Committee Competitionwas initiated, developed, and sponsoredby the AMCP student chapter and hasbeen held annually at the UIC College ofPharmacy for five years. The competitionis designed to simulate a “real world”experience outside the classroom. Pastcases have addressed evaluations of: • pharmaceutical manufacturer propos-

als on H2-receptor antagonists, antihy-perlipidemic products, and a disease-state management program that cen-tered on use of the manufacturer’sosteoporosis treatment drug (1997);

• bundling a group of drugs for preferredstatus on a formulary (1998);

• use of selective serotonin reuptakeinhibitors (SSRIs) in a specific patientpopulation (1999);

• restricting the use of nonsedating anti-histamines in a student-based healthmaintenance organization (HMO)(2000); and

• the use of prescription versus over-the-counter products (2001).

The goals of the P&T Competition are to:• enhance pharmacy students’ under-

standing of the P&T committee deci-sion-making process;

• further improve student professionalskills through research, presentation,and discussion of case-related issues;

• augment student experience and edu-cation in formulary design and man-aged care pharmacy;

• provide students with an opportunityand environment to network with sea-soned managed care pharmacy profes-sionals, including alumni; and

• recruit pharmacy students to participatein a professional-organization activity.In the competition, four students make

up a team which represents the pharmaciston P&T committee. Each team mustinclude a first-, second-, and third-year stu-dent; the fourth member can be from anypharmacy class. This structure creates anacademically diverse team and exposeseach student to knowledge levels andthought processes involved in varyingstages of educational development. Thefour competition judges are selected basedon various backgrounds, which mayinclude managed care professionals, indus-try professionals, faculty, and/or alumni.The primary purpose of the selection is toprovide a diversity of opinions from differ-ent health care settings and professionalperspectives.

Each team is presented with the samecase-specific scenario, which incorpo-rates many aspects that a “real” P&Tcommittee might encounter. The team is

asked to evaluate the case and develop aproposal to be submitted for committeeconsideration. The proposal must includea clinical research review, a pharmacoeco-nomic analysis, and an ethical-issueassessment. Within the clinical evaluation,the case information provided to the stu-dents includes the health plan’s demo-graphics, current formulary information,percent of members with related diseasestates, and drug utilization by drug classesunder review. In addition, the students aregiven the current annual expendituresper drug, the amount of drugs used byplan members, the current drug pricesincluding any rebates, and prior physi-cian switch rates to assist them with theirpharmacoeconomic analysis. The stu-dents also are informed of the use of dis-ease-state management programs, phar-macoeconomic software for their prod-ucts, and any academic detailing. All ofthis information may be used to assist thestudents in performing an ethical analy-sis. The ultimate goal of the studentteams in performing these analyses is todetermine if the overall proposal is thera-peutically valuable, economically viable,and ethically preferred.

Once the teams have researched andevaluated all of the information, they mustprepare a written executive summary and a35-minute oral presentation for the panel ofjudges. Each team’s write-up must includepertinent clinical data, and the economicand ethical analyses of the overall case, end-ing with a team decision and rationale tosupport the final recommendation to theP&T committee. During the first 15 min-utes of the oral presentation, each teammust explain the written summaries andpresent the rationale of its findings. Duringthe second 15 minutes, the panel of judgesasks a series of standard questions address-ing the clinical, economic, and policyaspects of each team’s recommendation. Thefinal five minutes are reserved for partici-pant questions and feedback from thejudges. Continued next page

P


Campus

At the conclusion of the oral presenta-tions, the judges select the first-placeteam. They use a standard judging criteriato evaluate the quality of the teams’ writ-ten and verbal presentations, problemsolving and critical thinking skills, appli-cation of relevant clinical research, andoverall coordinated team effort. The win-ning team is chosen based on its members’ability to think “out of the box” and extractpertinent data in order to synthesize theteam’s final decision. All participating stu-dents learn from each other in a team set-ting through research, thinking, and pres-entation experiences, and through con-structive feedback from the judges.

The benefits to the students of partic-ipating in a P&T Committee Competitionare numerous. First, the competitionincreases student interest in managedcare environments and provides aninsight into the roles and responsibilitiesof a managed care pharmacist. Indirectly,the competition promotes considerationof career choices toward managed carepharmacy. The students are given theopportunity to develop and refineresearch, analytical, and presentation

skills and gain a functional understand-ing of the P&T committee in a managedcare environment. Students learn to usepharmacoeconomic and pharmacothera-peutic evaluations to enhance futuredecision-making processes as they partic-ipate in making “real life” therapeutic rec-ommendations and decisions. In summa-ry, the P&T Competition serves to incor-porate student learning into a well-rounded managed care education for stu-dent participants.

In October 2000, at the AMCPEducational Conference in San Diego,California, several members of the UIC-AMCP student chapter presented a postertitled “Learning Managed Care ConceptsOutside the Classroom: A Pharmacy andTherapeutics Committee Competition.”Three chapter officers and two partici-pants from the winning team of thefourth-annual UIC-AMCP P&T Commit-tee Competition developed the poster.This poster-presentation opportunityincreased the exposure of the P&TCompetition and sparked interest fromstudents in other leading pharmacyschools. Many students expressed interest

in holding their own P&T competition attheir respective AMCP student chapters.

The success of the AMCP studentP&T Committee Competition led to thedevelopment of the first National P&TCommittee Competition sponsored byAMCP at the Annual Meeting in Tampa,Florida, this past April. The nationalcompetition provides all AMCP studentchapters with an excellent learningopportunity and increases interactionamong chapters to promote professionaland personal growth within the studentmembership.

Joy E. Zarlenga, Pharm.D. (cand.), VishalN. Goyal, Pharm.D. (cand.), J. WarrenSalmon, Ph.D., and Margaret H. Byun,Pharm.D., M.S. University of Illinois atChicago—AMCP Student Chapter. Author Correspondence: [email protected].

The authors thank JoAnn Stubbings, R.Ph.,M.H.C.A., and Mark J. Bachleda, Pharm.D.,for their invaluable comments on earlier drafts.

Readers are invited to submit

ideas and articles to Campus.

Contact Shane Desselle, Ph.D.,

Duquesne University, Mylan School of

Pharmacy, Pittsburgh, PA 15282;

Tel: 412-396-6363; Fax: 412-396-5130;

E-mail: [email protected].

The winning team is chosen based on its

members’ ability to think “out of the box”

and extract pertinent data in order to

synthesize the team’s final decision.


Women’s Health: Issues and Opportunities for Managed Care Pharmacy

POLICYF E A T U R E

raditionally, discussions of women’shealth were limited to conditionsthat are unique to women, such as

contraception, fertility, and menopause. Yetvery little was known about even theseissues, as illustrated by an article written in1967, which described menopause as achronic and incapacitating deficiency dis-ease.1 A menopausal woman was considereda castrate, with all the inherent physical andpsychic disorders observed in patients whohave undergone bilateral oophorectomy.The understanding of menopause, and ofwomen’s health in general, has growntremendously since that time.

Because gender differences in medicinehave become clear, we realize that we can-not treat males as the norm; thus, the def-inition of women’s health has expanded.Carolyn Clancy, director of the Center forOutcomes and Effectiveness Research,U.S. Agency for Healthcare Research andQuality (AHRQ), explained:

Women’s health may be defined as screen-ing for, preventing, diagnosing, and treatingconditions unique to women; conditionsmore prevalent or more serious in women;or conditions that have different risk factorsor require different interventions forwomen, and thus deserve special attention.2

The purpose of this article is toincrease recognition of the importance ofwomen’s health issues and the role thatmanaged care pharmacists can play inwomen’s health care.

? ? Women’s Health Issues in Managed Care

Financial IssuesThe cost of care is a general issue for managedcare organizations (MCOs), as well as forgovernment agencies and many employers.An integrated database of medical and phar-macy claims reported that nearly 60% of totalhealth plan expenditures are attributed towomen and 59% of all prescription drugs areprescribed to women.3 Reproductive issues—menopausal symptoms, menstrual disorders,and contraceptive management—accountedfor 16% of all health plan expenditures, morethan cardiovascular disease, diabetes, andasthma combined.3

One factor that contributes to theincreased cost of medical care for women isthat they visit their doctors more often thanmen do. Women are more likely than mento suffer health problems associated withtheir reproductive systems. Prenatal careand routine cervical and breast cancerscreening bring women into the health caresystem often. Another contributing factor isthat women are more susceptible than mento certain diseases, among them rheumatoidarthritis, osteoporosis, urinary incontinence,anxiety, depression, and Alzheimer’s disease.

In addition, some diseases have moresignificant consequences in women. Amongfemales attending family planning clinics,the prevalence of chlamydia ranges from4%–12%.4 Complications of chlamydia in

women are severe and frequent. Onceinfected, women are more susceptible toreproductive cancers and infertility.

The sheer number of women in govern-ment programs (Medicare and Medicaid)presents significant financial concerns. In1999, 57% of the 39 million people enrolledin Medicare were women, as were 57% ofMedicaid enrollees.5, 6Of the Medicare mem-bers older than 85 years, 75% were women.Medicaid is a major payor for nursing homecare; three quarters of nursing home resi-dents covered by Medicaid are women.6

Medicaid is also America’s largest single pur-chaser of maternity care (prenatal, delivery,and postpartum services). In 1995, Medi-caid paid for 39% of the more than 3.1 mil-lion live births in the United States.6

MarketingThe phrase “women’s health” is often usedin marketing, especially within the man-aged care and employer markets, becausewomen make 75%–90% of the health caredecisions for themselves and their fami-lies.7, 8 Some managed care plans havefound that promoting preventive care towomen increases the likelihood of their re-enrollment by as much as 28%. Employersmay also benefit from promoting women’shealth care because women now make up44% of the paid U.S. workforce.7

Quality of Care Some conditions that may affect a woman’s

Authors

ANN M. MCNAMARA, Pharm.D., is Clinical Specialist, Women’s Health, and MARY CLAIRE WOHLETZ, Pharm.D., is Clinical Specialist, Respiratory, Express Scripts,Inc., Bloomington, MN.

AUTHOR CORRESPONDENCE: Ann M. McNamara, Pharm.D., Express Scripts, 6625 W. 78th St., P.O. Box 390842, Bloomington, MN 55439-0842; Tel: 952-893-4640;Fax: 952-837-7189; E-mail: [email protected].

Copyright© 2001 Academy of Managed Care Pharmacy, Inc. All rights reserved.

B y A n n M . M c N a m a r a a n d M a r y C l a i r e W o h l e t z


T



quality of life have received little attention,in part because they are not central to thereproductive role. Osteoporosis, inconti-nence, and dysmenorrhea are often cited asareas of neglect. Uterine fibroids (uterineleiomyomata), benign tumors of the uterus,are exceptionally common in women aged25–45 years old (30%).10 Uterine fibroidsare the most common indication for hys-terectomy, accounting for 30% of hysterec-tomies in white women and over 50% inblack women.10 However, a recent reportfrom AHRQ stated that, “with the exceptionof trials of gonadotropin-releasing hormone-agonist therapy (e.g., goserelin, leuprolide,nafarelin) as an adjunct to surgery, there is aremarkable lack of randomized trials datademonstrating the effectiveness of therapieswith nonsteroidals, progestins, or oral con-traceptives in the management of womenwith symptomatic fibroids.”10 It is surprisingthat so little data are available on therapiesfor a very common condition.

Hormonal changes play a role in somedisease exacerbations. A significant per-centage of women with asthma, migraine,diabetes, depression, or epilepsy experi-ence worsening of their disease premen-

strually. Little is known about how oralcontraceptives and hormone replacementtherapy affect the course of many diseasesor the response to therapy. Additionally,women have often been excluded from orseriously underrepresented in clinical tri-als (e.g., major studies of coronary heartdisease). Lack of understanding surelyaffects quality of care.

In order to remedy this disparity inresearch and assure that the care of womenis adequate, it is necessary that women beincluded in the study populations of feder-ally funded clinical research and clinical tri-als on medications submitted to the Foodand Drug Administration (FDA). Onenotable trial, designed to address the gap inknowledge, is the Women’s Health Initiative,which will evaluate strategies for preventingheart disease, breast cancer, colorectal carci-noma, and osteoporosis in postmenopausalwomen. This will be the first trial to providedisease data on hormone therapy for ethnic-minority women. Through advocacy bywomen’s organizations, federal funding hasbecome available, particularly for breastcancer research.

The National Committee for Quality

Assurance (NCQA) evaluates and reports onthe quality of the nation’s MCOs. One of itsmeasures, the Health Plan Employer Dataand Information Set (HEDIS), will soonaccount for 25% of an MCO’s NCQA accred-itation. HEDIS measures the followingwomen’s health indicators: chlamydiascreening, management of menopause,breast cancer screening, cervical cancerscreening, prenatal care in the first trimester,and check-ups after delivery (see Table 1,left). Future HEDIS measures are likely toinclude osteoporosis and family planningcounseling.

? ? Opportunities for Managed Care Pharmacy

Women’s health care is also important tomanaged care pharmacy. In terms offinances only, women represent 62% of pre-scription drug use and 58% of total pre-scription costs.11 Approximately 70% of allprescriptions for antidepressants and anti-anxiety agents and 80% for migraine drugsare filled for women. 11 Several other classesof medications (e.g., gastrointestinal drugs,pain medications, antihistamines) are morecommonly filled for women.

In the past decade, a dramatic 75%increase in pharmaceutical industry researchon areas of women’s health has led to a fullpipeline of drugs to help meet the specialhealth needs of women.12 U.S. pharmaceuti-cal companies are developing 348 newmedicines directed at women’s health needs(see Table 2, page 265). Many are for dis-eases that disproportionately afflict women,but for some gender-based research is beingconducted. With all these medications inthe pipeline, managed care will face thechallenge of improving care while managingcosts.

The significant opportunities toimprove the care of women are illustratedby the wide variation of care in HEDISmeasures, especially chlamydia screening(see Table 1). Smoking rates are decliningmore slowly for women, and obesity andphysical inactivity are more prevalent inwomen than men. Recently, the Centers forDisease Control and Prevention revealedthat clinicians are missing opportunities for

TABLE 1 HEDIS Quality Measures Specific to Women’s Issues—1999 National Results

Results of Average Health Results of Plans in the Plan Results Plans in the

10th Percentile 90th Percentile

Breast cancer screening 64% 73% 82%

Cervical cancer screening 60% 72% 83%

Checkups after delivery 54% 71% 87%

Chlamydia screening, 5% 18.5% 32.6%ages 16–20 years

Chlamydia screening, 4.6% 16% 28%ages 21–26 years

Management of menopause 48.7% 56.6% 63.7%counseling (composite of subscores)

Prenatal care in the first 71% 85% 95%trimester of pregnancy

Note: HEDIS is the Health Plan Employer Data and Information Set.


heart-disease-prevention counseling. In astudy of 29,273 routine office visits, womenwere counseled less often than men aboutexercise, nutrition, and weight reduction.13

Managed care pharmacists have an opportu-nity to work toward improving the care ofwomen by educating members and provi-ders (see Table 3, page 266).

Managed care pharmacists must recog-nize differences in the safety of medicationsbetween men and women. A recent report

by the General Accounting Office stated that8 out of 10 prescription drugs pulled fromthe U.S. market since 1997 posed greaterhealth risks for women than for men.14 Fourof these (Rezulin, Redux, Pondimin, andLotronex) may have had more adverseevents in women than in men because theywere prescribed more often to women. Theother four (Posicor, Seldane, Hismanal, andPropulsid) had more adverse events inwomen even though they were widely pre-

scribed to men as well as women. Women inparticular have a higher incremental risk ofsuffering an arrhythmia after taking thesedrugs, probably because the intervalbetween heart muscle contractions is natu-rally longer for women than for men andbecause male sex hormones moderate theheart muscle’s sensitivity to these drugs.

Gender differences in efficacy may alsoexist. Because phenytoin is cleared morerapidly in women during the luteal phase of


Therapy Class Number Example(s)

Arthritis/musculoskeletal 54 • Recombinant parathyroid hormone for osteoporosis• Xyvion (tibolone) for osteoporosis prevention• Prempro in a lower dose for osteoporosis prevention• Premarin in combination with trimegestone for

osteoporosis prevention

Autoimmune diseases 37 • Amevive (recombinantly engineered protein) for psoriasis

• Rebif (interferon beta-1a) for multiple sclerosis

Cancer 85 • Evista (raloxifene) for the prevention of breast cancer • Cancer vaccines

Diabetes 19 • Inhaled insulin

Eye 4 • Medications for glaucoma or prevention of secondary cataracts

Kidney/urologic 16 • S-oxybutynin and several other medications for incontinence

Lung/respiratory 33 • Anti-IgE monoclonal antibody for asthma

Neurologic 33 • Almotriptan, frovatriptan, and Relpax for migraine headache

• Premarin for Alzheimer’s disease

Obstetric/gynecologic 52 • EVRA (transdermal contraceptive)• Female testosterone patch• Yasmin (drospirenone/ethinyl estradiol oral

contraceptive)• Medications for sexual dysfunction

Psychiatric 26 • Sarafem (fluoxetine) was recently approved for premenstrual dysphoric disorder

• There are trials ongoing for paroxetine in premenstrual dysphoric disorder

Other 18 • Zelmac (tegaserod) for irritable bowel disease in women

TABLE 2 Medicines and Vaccines in Development for Women



the menstrual cycle, serum concentrationsmay be reduced below the therapeutic win-dow. Decreased serum phenytoin concen-trations correlate with increased seizureactivity in women with catamenial epilepsy(an increase in seizure activity four daysprior to menses and six days after onset).Generally, women tend to respond less wellthan men to tricyclics but better than mento selective serotonin reuptake inhibitors(SSRIs) and monoamine oxidase inhibitors.

? ? Managed Care Implications

Benefit DesignWomen’s health issues cross all areas ofmanaged care pharmacy, including bene-fit design, formulary management, andclinical programs. A survey of more than14,000 women demonstrated that theinclusion of prescription drug benefitshas important ramifications for customer

satisfaction.15 The survey found that 51%of women made their health plan deci-sion based on whether there was a pre-scription drug benefit and 33% statedthat their decision was based on medicaland pharmacy copayment differencesbetween health plans.

Managed care pharmacists should beable to defend the economic and healthrationale for including contraceptives withinthe pharmacy benefit. For example, report-edly because of the increased use of contra-ceptives, from 1987 to 1994 the rate ofunplanned pregnancies dropped 16%,abortions declined 11%, and unplannedbirths declined 22%.16

Express Scripts’ oral contraceptivemodel, based on actual use of oral contra-ceptives in 1996 by 1.8 million employerlives, can estimate the cost of contraceptivecoverage based on the age demographics ofa group’s female population. The model isbased on average wholesale price (AWP),and thus does not include discounts orcopayments. Overall, the cost of oral con-traceptives is $0.77 per member per monthor $1.48 per female member per month.Payors should also be made aware that theymay already be covering some oral contra-ceptives to treat medically appropriate con-ditions such as endometriosis or dysfunc-tional uterine bleeding.

An analysis of contraceptive coverage isbeyond the scope of this article, but one ref-erence that does describe the cost to employ-ers of adding all contraceptive options isavailable on the Alan Guttmacher InstituteWeb site at www.agi-usa.org/pubs/kaiser_0698.html.

Coverage of fertility medications withinthe pharmacy benefit is another issue thatmay seem to specifically affect women.Thirteen states have now mandated someform of infertility coverage, yet there are nonational standardized protocols. Managedcare pharmacists may be involved indesigning infertility medication benefits ordeveloping protocols for their use. Forexample, the expense of one of the treat-ments, gonadotropins, may be compound-ed by waste (e.g., cycles are often can-celled, patients may have remaining

gonadotropins after a cycle). Therefore, adistribution system or benefit design strat-egy that reduces gonadotropin waste canreduce costs. A resource that describessome issues in developing infertility treat-ment algorithms is an article by Gleicher inHuman Reproduction Update.17

Clinical ProgramsCare of chronic disease is one of the fastest-growing and costliest segments of women’shealth care. Conditions that lend themselvesto clinical/educational programs or multi-disciplinary guidelines in women includechronic headaches, chronic rheumatologicconditions, obesity, coronary artery disease,osteoporosis, and depression. Effective ther-apy adherence programs are needed forosteoporosis and hormone replacementtherapies. Clinical programs that strive toimprove women’s knowledge of menopau-sal therapies, both pharmacologic and non-pharmacologic, would also fill a gap inmember education.

Formulary Management There are at least two perspectives of howthe care of women falls into formulary man-agement. One is the need for safety and effi-cacy information specifically directed atwomen. A second is the safety and efficacyof how a medication is to be used in combi-nation with oral contraceptives or hormonereplacement therapy products.

? ? Case Examples

Below we describe programs that con-tribute to understanding women’s health,improving the quality of care, or reducingcosts of pharmaceuticals.

Accurate Health Statistics A World Health Organization technicalpaper states that, first and most impor-tantly, it is essential that the situation ofwomen be more accurately reflected inroutinely collected health statistics.18 It hasbeen a frequent complaint that most sta-tistics are not separated by gender. ExpressScripts’ Drug Trend Report now has a chap-ter that describes use of medication inmen and women across the lifespan.11

• The National Osteoporosis Foundation rec-ommends that all adults receive at least1,200 mg/day of elemental calcium.However, the typical American diet pro-vides less than half that amount.

• 30%–70% of women who start estrogentherapy discontinue therapy within the firstyear. Of women who start alendronate,35% of patients discontinue it within eightmonths.

• Other than hormone replacement therapy,women receive very little information onways to manage menopause. Less than 2%of women received information on calcium,other medications, vitamins, exercise, nutri-tion, or vaginal creams for managingmenopause.

• Only 42% of women of reproductive age(18–44) take a mulivitamin or folic acidsupplement daily. Only 30% of women areaware that taking folic acid during earlypregnancy might prevent neural-tubedefects.

TABLE 3 Some EducationalOpportunities For Managed CarePharmacy

One important finding was that therewas a greater percentage change in costper prescription for females (+10.96%)than for males (+7.24%) as they aged.Mean AWP per prescription increased18.2% for women in their seventh decadeof life and 20.25% for women at age 80years of age and older. Increases in costper prescription hit elderly females dis-proportionately hard.

Patient CounselingA pharmacist-to-patient counseling pro-gram focuses on members who are mostat risk for negative outcomes related topolypharmacy. Two-thirds of the mem-bers contacted are women. These servic-es promote formulary or generic andlower-cost medicines as well as preventionof adverse drug events and interactions.Members are educated about their medica-tions, including dosing schedules and con-traindications, via letters and telephonecounseling. All members are encouraged todiscuss pharmacist recommendations andany additional questions with their physi-cian. The top five interventions discussedwith women are (1) calcium requirements;(2) bisphosphonates, including proper use,calcium requirement, and side effects; (3)bone-density tests for osteoporosis riskassessment; (4) incontinence/bladder con-trol; and (5) management of menopause.

One example highlights the value ofpharmacist counseling.During a phone con-sultation, a pharmacist noticed that thefemale patient that she was speaking withwas having difficulty breathing and talkingat the same time. The pharmacist noted twosimilar heart medications (metoprolol andatenolol) on the patient’s record and notifiedthe physician immediately about the dupli-cate therapy. One of the medications wasthen discontinued.

OsteoporosisA recent National Institutes of Health reportstated that fewer than 5% of patients withosteoporotic fractures are referred for med-ical evaluation and treatment.19 This is wor-risome because up to 20% of post-menopausal women with a vertebral frac-

ture will have a second fracture within ayear.20 A program being designed by ExpressScripts will use integrated pharmacy andmedical claims to target postmenopausalwomen diagnosed with osteoporosis or whohave had a hip, vertebral, or radial fractureand have not received a medication forosteoporosis. Letters will be sent to thepatients and their physicians, discussingprevention and treatment of osteoporosis,lifestyle modifications, and therapy options.Patients who start on medication will beenrolled in a compliance program. The goalis to promote evaluation and treatment ofpatients at risk for osteoporotic fractures.

? ? Conclusion

Women’s health is a priority for many healthcare systems for financial, marketing, andquality-of-life reasons. Significant advancesin women’s health care have been made inthe past 30 years. The inclusion of womenin clinical trials will continue this effort tounderstand gender-specific differences inmedicine.

Managed care pharmacists have oppor-tunities to improve the care of women byincorporating the information from clinicaltrials into their practice. They are well posi-tioned to improve quality of care througheducating providers and members.

Lastly, managed care faces the chal-lenge of managing the potential costs pre-sented by a vast pipeline of new drugsspecifically designed for women.

References

1. Rhoades FP. Minimizing the menopause. AmGeriatr Soc 1967; 15: 346–54.

2. Clancy C. National Summit on Women’s Healthand Managed Care. Washington: 1999 Mar 10–11.

3. Practice Pattern Science, Data on File, March 2000.

4. U.S. Preventive Task Force. Screening forChlamydial infection: recommendations andrationale. Am J Prev Med. 2001; 20(3S): 90–94.

5. Medicare and women. Washington: Henry J.Kaiser Family Foundation, 1999. Available atwww.kff.org/content/1999/1481/women.pdf.

6. Medicaid: A primer. An introduction andoverview. Washington: Henry J. Kaiser FamilyFoundation, August 1999. Available atwww.kff.org/content/1999/2161/pub2161.pdf.

7. Reaching women, health’s gatekeepers.Healthcare PR & Marketing News. 1998; 7: 10.

8. Gonen JS. Value purchasing: investing inwomen’s health: strategies for employers.Washington: Jacob’s Institute of Women’s Healthand the Washington Business Group on Health,April 2000.

9. Data on file, Wyeth-Ayerst Laboratories,CareData Reports, Inc. 1997.

10. Management of uterine fibroids. Summary, evi-dence report/technology assessment: Number 34.Rockville, MD: Agency for Healthcare Researchand Quality; 2001 Jan. AHRQ Pub. No. 01-E051.Available at www.ahrq.gov/clinic/utersumm.htm.

11. Teitelbaum F, Martinez R, Parker A et al.Express Scripts Drug Trend Report. St. Louis, MO:Express Scripts Inc., June 2000.

12. Survey finds dramatic increase in research onwomen’s health; 348 medicines in development.Washington: Pharmaceutical Research andManufacturers of America, November 1999.Available at www.phrma.org/searchcures/newmeds/women1999/women99.pdf.

13. Missed opportunities in preventive counselingfor cardiovascular disease—United States, 1995.MMWR Weekly. 1998; 47(5): 91–95.

14. Drug safety: most drugs withdrawn in recentyears had greater health risks for women.Washington: United States General AccountingOffice; January 2001. GAO-01-286R. Available atwww.gao.gov/new.items/d01286r.pdf.

15. Wilson J. A qualitative perspective of women’shealth: who makes the consumer-based health caredecisions? In Proceedings of the Women’s HealthForum at the National Managed Health CareCongress’ Annual Disease Management Congress.Pasadena, CA: 1999 Feb 2–5, 1999. Bronxville,NY: Medicom International, Inc., 1999.

16. Henshaw SK. Unintended pregnancy in theUnited States. Fam Plann Perspect 1998; 30:24–29, 46.

17. Gleicher N. Cost-effective infertility care.Human Reproduction Update 2000; 6(2): 190–99.

18. Garcia-Moreno C. for the Gender WorkingGroup. Gender and health: A technical paper.Geneva, Switzerland: World health Organization,1998. Available at www.who.int/frh-whd/GandH/GHreport/gendertech.htm#TOC.

19. NIH Consensus Development Panel onOsteoporosis Prevention, Diagnosis, and Therapy.Osteoporosis prevention, diagnosis and therapy.JAMA 2001; 285(6): 785–95.

20. Lindsay R, Silverman SL, Cooper C, et al. Riskof new vertebral fracture in the year following afracture. JAMA 2001; 285(3): 320–23.



Employee Benefits Consulting: An Essential Role for Pharmacy

POLICYF E A T U R E


mployers today face a complexassortment of health care deci-sions as employees clamor for

expanded benefits and the cost of healthcare continues to grow. The pharmacybenefit, although one of the most desir-able for employees, is financially a highrisk for payors. The role of the pharmacybenefits consultant has grown to matchemployers’ needs; it now includes notonly review of medication expendituresbut also analysis of the overall health careplan design and increased use of clinicalprograms to optimize medical and pre-scription-drug management.

The importance of good benefits con-sulting can be underscored by some recentcost trends. In the early 1980s, prescriptiondrugs accounted for about 2.3% of the U.S.health care dollar. That figure grew to 10%in the early 1990s, and to 15% by 2000.1

Clients outside of a managed care planexperienced higher annual increases (about18%) than did managed care clients(approximately 16.3%).2 Clearly, drug-benefit management has had an impact.

During the next five years, double-digitannual increases are expected.1 It is com-monly accepted that the cost of the prescrip-tion benefit will only be curbed if a clinicalconsulting approach is used when pharma-cy benefits are introduced or refined. Thisrepresents an opportunity for managed carepharmacists to have considerable impact onthe cost and quality of health care.

? ? History Is Important

Managed care has arrived at this point as thenatural consequence of its history. Managedcare is a form of insurance. Until World WarII (WWII), health insurance was a fringebenefit, meaning it was an optional incen-tive to attract and retain employees. DuringWWII, wages were frozen but benefits werenot. Consequently, the health care benefit asa form of compensation moved from thefringe into the limelight.3 The governmentprovided incentives to promote employeebenefit plans, including tax laws favoringbusinesses that provided health insurance.

Over the next two decades, fee-for-serv-ice health care flourished, and the cost ofhealth care was unfettered. Benefits that wereonce considered optional or attractive—pharmacy, dental, and vision—became high-ly desirable to employees, and pressure tomake them standard offerings increased. Bythe 1960s, health care financing had becomea national concern, and various methods ofcontrolling costs were introduced.

From the inception of managed care,performance benchmarks focused oncost savings, not clinical outcomes.4 Plandesigns assumed that providers wouldmanage insurable risk to maximize out-come and minimize complications at a rea-sonable cost.3 In the 1980s, pharmacy ben-efit management relied on controlling unitcost and unit utilization. How successfullymanaged care principles were applieddepended on a number of factors. The ini-

tial cost containment began to level out inthe 1990s.

Today, there are concerns about qual-ity of care, member satisfaction, andbreadth of coverage as well as cost. Thus,health care plans must use cost manage-ment as a foundation, use managed careprinciples to design benefits, use a deliv-ery system based on quality-managementprinciples, and comply with numerouslaws. For the pharmacy benefit today,three factors are essential:• controlling unit cost and utilization; • sharing financial risk (and some deci-

sion making) by having members pay anominal fee whenever they access thehealth care system; and

• obtaining discounts from all providersand vendors that influence drug costs.The complexities of drug therapy, the

entry of cosmetic and life-enhancing med-ications, new product pricing, and clinicalprogram opportunities have all affectedmedication use and opened new venues forbenefit management. Pharmacy-benefitconsulting must give history its due, usetoday’s tools efficiently, and look to thefuture to anticipate changes on both theimmediate and the distant horizons.

? ? Who Manages the Benefit?

Employers have options when evaluatingcost drivers in their populations. Theycan use a third-party administrator, theirmedical insurance carrier, or a pharmacy

Author

CONNIE PERRY, Pharm.D., is Assistant Vice President, Aon Consulting, Inc., Chicago, IL.

AUTHOR CORRESPONDENCE: Connie Perry, Pharm.D., Assistant Vice President, Aon Consulting, Inc., 123 North Wacker Dr., Suite 1000, Chicago, IL 60606-1770;Tel.: 312-701-3695; Fax: 312-701-4855; E-mail: [email protected].


E

B y C o n n i e P e r r y


benefits management company (PBM).Regardless of which they select, employ-ers expect recommendations based ondata and reports that respond directly tothe issues in ways that are easily under-stood and supportable.

These reports can serve as tools toeducate employers on what drives costsin today’s market (see Table 1, above).Tom Lerche, health and welfare practiceleader at Aon Consulting, describes thepharmacist’s role succinctly: “Our cus-tomers are self-funded employers whofind that the prescription drug benefit isone of their top three problems. This is adifficult area because it is high cost, butof high value to employees. Pharmacistsbring to bear clinical knowledge andskills.”5 He then demonstrates that dis-ease management is a cousin to the drugbenefit, and that when a consulting phar-macist is able to apply disease-manage-ment theories creatively, the customerbenefits. These pharmacists are particu-

larly helpful when employers requestadditional clarification on current pre-scription drug benefit issues.

Defining what a pharmacy benefits con-sultant does is difficult. Table 2, below, showsseveral definitions. While the general dutiescan be specified, how they are performedvaries. Often, information about benefitmanagement is not available because it isproprietary to the companies that own it.3

In general, consultants use two tech-niques to manage the benefit: economictechniques (plan design) and benefit man-agement (clinical tools). Three unique func-tions pharmacy consultants perform today

are (1) evaluation and analysis of the bene-fit, comparing it with national cost trends;(2) guiding the proposal and marketingprocesses; and (3) auditing vendors.

? ? Evaluating the Pharmacy Benefit

Evaluating the pharmacy benefit requiressimultaneous analysis of the financial andclinical techniques employed.

Financial areas include member cost-sharing, the exclusion medication list, retailand mail service pricing and fees, use ofgeneric medication, formulary analysis, andrebate/discount arrangements. Despite theirdesignation as financial, each of these areascan be addressed better if the pharmacist’sinherent clinical insight adds perspective. Forexample, though the decision to change froma two-tier to a three-tier cost-share affectsmembers, the impact can be mitigated ifsound clinical judgment is used to determinewhere drugs are placed in the tiers.

Research has confirmed that current costdrivers in pharmacy include increased drug

use by members, with a changing mix ofdrugs contributing approximately equallyand inflation contributing less.1, 6 Reasonsfor increased use include an aging popula-tion, new prescribing guidelines thatemphasize combination therapy, a plethoraof new therapeutic drug advances, increasedcompliance among members, and pharma-ceutical company advertising (both direct-to-consumer and to providers).

Clinical program opportunities abound:retrospective, concurrent, or prospectivedrug-utilization review; prior-authorizationprotocols; provider profiling; and wellnessor disease-management programs. Clearly,implementing these programs can reducecosts and minimize the need to increasemember cost share.

Pharmacy consultants have learnedseveral lessons over the decades:• Introducing several changes at once is

likely to be less successful than intro-ducing changes gradually with amplecommunication to and lead-time foraffected parties.6

• Analysis of plan design should startwith the member cost-share to ensurethat it is appropriate, encourages use ofgenerics and mail-in programs, andaddresses cosmetic, life-enhancing, ordiscretionary drugs in a way that is fairand humane.

• Pharmacy benefits consultants shouldand can successfully address utiliza-tion, affecting cost significantly.

• Excluding drugs after they have beenapproved and have gained users createsdiscontent among providers and mem-bers; each new drug choice should beanalyzed well before its introduction tothe market.

• Formularies can promote step care, limit-ing inclusion only to those doses anddelivery forms that are most cost-effective.1

Measuring the pharmacy benefit againstregional standards, reconciling employerhuman resource strategy and employeeneed, integrating clinical and financialprocesses, and anticipating new drugsbefore they enter the market placetogether create the best foundation forcost-effective, high-quality care.


Pharmacy benefit consultants:

• Control costs by managing unit cost (price) and number of units used (quantity or utilization)

• Introduce or recommend enhancements to tools and techniques to manage the prescription benefit

• Design and implement organizational activities to influence prescriber, pharmacist, and patient behavior in such a way that the cost and use of prescription coverage is decreased

TABLE 2 What Pharmacy Benefit Consultants Can Do

TABLE 1 Pharmacy CostDrivers in 2001

• Aging population

• Explosion of new drugs

• Medical guidelines specifying earlier drug interventions

• New, multidrug regimens

• New treatments for previously untreatablediseases

• Better patient compliance

• Direct-to-consumer advertising

• Price inflation


? ? The Proposal Process

Among the pharmacy benefit consultant’smost popular projects is the request forproposal (RFP) or vendor marketing andselection process. Pam Hodge, an assistantvice president with Aon Consulting, sug-gests bringing in a pharmacy consultantwhenever a client requests a prescriptiondrug study. Her clients are generally inter-ested in a long-term strategy for controlling

prescription drug costs. These strategiesmay include plan redesign, employee edu-cation, a common formulary approach, andnegotiating rebates and lower administrativecosts with vendors. She has found that theexpertise of a pharmacy benefits consultantcan be helpful to clients in designing andexecuting these strategies.

Selecting the best vendor to meet anemployer’s needs can be challenging today.

Legal, clinical, and business changes inhealth care have created an environment inwhich complex plan designs, superioraccount management, and performance-measurement agreements are the rule ratherthan the exception. This complicates theproposal process. In the past, fee or discountnegotiation often gave the complete answerto an employer’s cost concerns. Today, clini-cal issues abound, and performance meas-


Key Measure Description of Measure Best Practice Standard Reporting Frequency

ID card production Accuracy in delivery of ID 100% within 2 business days Quarterlycards after receipt of processable

eligibility information

Reporting standards Standard reports and Core Within 45 days of QuarterlyPerformance Measures are reported end of quarterly cycle

on a regular basis

Prior-authorization The response time for prior- 1 business day Quarterlyturnaround time authorization requests is reasonable

Claim-adjudication Prescription claims are 99% of claims are processed Quarterlyaccuracy rate processed with accuracy with 100% accuracy

Average speed of answer Average speed of answer for 100% of calls answered Quarterlyall calls received by PBM’s within 25 seconds

customer service unit

Mail-service prescription Number of days for mail- 95% of “clean” orders are Quarterlyturnaround time service prescriptions to mailed to members within

be delivered to members 2 business days

Access to claims data Ability of client’s benefits Access is granted with Notice of 15 daysstaff and their designee(s) to timely noticesaccess claims data, etc., for

the purpose of auditing, etc.

Member reimbursement Each eligible person is reimbursed Vendor agrees with the N/Afor inaccurate dispensing the cost of any inaccurately description of measure and will

dispensed prescription drug also reimburse for the cost of thereplacement drug

Call-abandonment rate Percentage of calls lost No more than 2.5% of Quarterlydue to hang-ups total calls are abandoned

Call-resolution rate Calls answered and resolved 95% of calls are resolved Quarterlyby customer service unit on first call

Telephone busy signals Minimal busy signals No more then 3% of total calls Quarterlyfor members calling the will result in a busy signalcustomer service unit

Issuance of rebate check Rebate check received in 60 days post end of Quarterlya timely manner reporting quarter

Denied/rejected claims No administrative fees Vendor agrees with N/Aare charged for denied description measure

or rejected claims

TABLE 3 Core Performance Measures


ures must be incorporated to address cus-tomer service concerns. In addition, mem-ber access to services must be considered.

One state-government manager com-mented after working with a pharmacy ben-efits consultant to develop an RFP for a100,000-member program: “Any organiza-tion that is going through the painful task oflooking at prescription drugs is bombardedwith information from lobbyists, vendors,and other interested parties. They all havean interest in the outcome. Having a phar-macy benefits consultant helps us direct ourenergies appropriately.”

Table 3, page 270, describes some cur-rent performance measures that define theemployer, provider, and consumer expecta-tions of acceptable service. Too often thecommunity pharmacist is swamped withproblems when new identification cardsshould have been mailed but were not, thephysician is frustrated with prior approvaldelays, or the member’s prescription arrivesafter several doses have been missed.Incorporating performance measures intocontracts improves the quality of service byclarifying tasks, time frames, appropriatemeasures, and reporting frequencies.Performance, however, must be measuredwith both the client’s or business environ-ment’s perspective and accepted or pendingmitigating factors in mind.

? ? Vendor Audits

Vendor audits ensure quality service. Theydetermine the effectiveness of vendor per-

formance and compliance with contracts. Audits are necessary because benefits

managers and pharmacy directors maynot always have the capability or docu-mentation to verify certain aspects ofservice. For example, rebate amounts arebased on volume of product sold; anaudit can determine if the rebate wasappropriate to the volume sold. Also, keyperformance measures can be audited toensure that they are being met. In short,almost anything—financial or clinical—can be audited.

Table 4, above, describes the basic ele-ments of an audit. Audits generally rely onreview of databases. Once the audit is con-ducted, the pharmacy consultant mustreport findings and make recommendationsfor corrective actions. These recommenda-tions may include changes to the PBMagreement or benefit plan. The report mustalso document any variances in terms ofbilling errors or overpayments and describetrends or specific areas where new process-es can prevent future problems.

? ? Implications for Managed Care Pharmacists

The average employee benefit packageconsumes 30%–50% of employer payrollcosts, yet it is a source of dissatisfactionand skepticism for many employees.3 Thedrug benefit has come to be consideredmore of an entitlement than an option formany. Many studies have shown thatdrug therapy is often the most-effective

and least-expensive form of treatment.With its increasing importance, the phar-macy benefit will continue to be scruti-nized, changed, and audited.

Managed care pharmacists must recog-nize that their information and clinicalmanagement functions will grow in signif-icance. The consultant pharmacist whoprovides good data and sound clinicaljudgment will gain more opportunity todemonstrate how employers can containcosts and optimize medication use in thepopulations they serve.

? ? Conclusion

Pharmacists working with health careplans should work with the plan’s mar-keting department to ensure that employ-ers understand the pharmacy-manage-ment programs used. Meeting face-to-face with employers is an excellent way tofacilitate accurate communication.

As pharmacy benefits grow in impor-tance, there will be more opportunity in thebenefit-consulting field. Pharmacists withmanaged care experience, gained fromemployment with either a health plan or aPBM, will be ideal candidates for these posi-tions. In this way, the effect of managed careon health care will continue to expand.

References

1. Disease management: Next step in managedcare. Aon Consulting Forum, December1999–January 2000.

2. Express Scripts. Drug trend report. St. Louis:Express Scripts, 1999; 1–51.

3. Stern CS. The history, philosophy, and princi-ples of pharmacy benefits. J Managed Care Pharm1999; 5(6): 525–31.

4. Wertheimer AI, Navarro RP. Pharmacy benefitmanagement principles and practices. In: Managedcare pharmacy practice: Principles and practice.New York: Pharmaceutical Products Press, 1999:29–46.

5. Regulatory developments. Aon ConsultingForum. December 1999.

6. Merck-Medco. Managing pharmacy benefitcosts, new insights for a new century, 2000 edi-tion. www.merckmedco.com, accessed January 13,2001.


• Analyze manufacturer rebates and their application to client

• Identify apparent duplicate claim payments

• Identify prescribing/utilization patterns within specific therapeutic categories

• Analyze formulary alternatives used and the formulary-conversion process

• Identify generic fill rate

• Validate eligibility updating and procedures for terminations and additions

• Validate claim-system edits for plan limitations, maximums, and prescription preauthorizations

• Analyze for possible fraudulent, drug abuse, or misuse-related claims

• Validate pharmacy reimbursement rates and accuracy of claim payments

• Evaluate impact of retail versus mail reimbursement formulas (if applicable)

TABLE 4 Elements of a Database Audit


This paper reports on the Academy ofManaged Care Pharmacy’s (AMCP’s)suggested process for pharmaceuticalmanufacturers to submit a uniform,evidence-based dossier for eachproduct submitted for formularyapproval.

The process requires that manufacturers provide clinical andeconomic evidence and an economicmodel that projects the potential economic consequences and impactof product coverage on health outcomes. This formulary-submissionprocess is designed to achieve twomain goals: (1) improve the timeli-ness, scope, quality, and relevance of

information provided to pharmacy andtherapeutics (P&T) committees; and(2) streamline the process of acquir-ing data and reviewing products forhealth plan staff pharmacists. In time,it is hoped, this process will lead toprogressive improvements in thequality of formulary submissions andprovide P&T committees with evi-dence that previously was oftenunavailable.

KEYWORDS: Economic model, formu-lary, health outcomes, P&T Committee

J Managed Care Pharm 2001: 272–82

AMCP Guidance for Submission of Clinical andEconomic Evaluation Data to Support Formulary Listingin U.S. Health Plans and Pharmacy BenefitsManagement Organizations

RESEARCH

SEAN D. SULLIVAN, Ph.D., is Professor of Pharmacy and Health Services,University of Washington, Seattle; ALAN LYLES, Sc.D., M.P.H., is AssociateProfessor, University of Baltimore, and Adjunct Associate Professor, JohnsHopkins School of Public Health, Baltimore, MD; BRYAN LUCE, Ph.D., is ChiefExecutive Officer, MEDTAP International, Bethesda, MD; and JOSEPH GRIGAR, M.S., is an independent consultant, New York, NY.

AUTHOR CORRESPONDENCE: Sean D. Sullivan, Ph.D., Professor ofPharmacy and Health Services, University of Washington, Box 357630, Seattle, WA 98195, Tel.: 206-685-8153, Fax: 206-221-5347, E-mail: [email protected].


Authors

he Academy of Managed Care Pharmacy (AMCP) hascreated a process for pharmaceutical manufacturers tosubmit a uniform, evidence-based dossier for each

product to help pharmacy and therapeutics (P&T) committeesmake informed decisions about which products to include onhealth plan formularies. The process requires that manufactur-ers provide clinical and economic evidence from published andunpublished studies and an economic model that projects thepotential economic consequences and impact of product cov-erage on health outcomes. This objective formulary-submis-sion process is designed to achieve two main goals: (1) improvethe timeliness, scope, quality, and relevance of information pro-vided to P&T committees; and (2) streamline the process ofacquiring data and reviewing products for health plan staffpharmacists. The intent is to encourage effective formularydeliberations through a partnership between the manufacturerand the health plan. In time, it is hoped, this process will leadto progressive improvements in the quality of formulary sub-missions and, at a minimum, provide P&T committees withevidence that previously was often unavailable.

Rational decisions about product adoptions based on clinical,economic, and other outcomes data should be the foundation of asound formulary system. These precepts were affirmed by the newguidance, “Principles of a Sound Drug Formulary System,” endorsedby the AMCP, American Society of Health-System Pharmacists,American Medical Association, Department of Veterans Affairs,National Business Coalition on Health, and United StatesPharmacopeia.1

AMCP has designed a companion Format for FormularySubmissions so that manufacturers will submit evidence of clin-ical and economic benefit in a standard format to health plans,pharmacy benefit management firms, hospitals, and otherorganizations so that they can make objective evaluations ofpharmaceuticals for coverage or reimbursement. These data willenable P&T committees to consider a wider array of informa-tion in their decisions about products to include on a healthplan’s formulary.

The process is intended to foster a meaningful partnershipbetween the manufacturer and the health plan and to elicit

T

by Sean D. Sullivan, Alan Lyles, Bryan Luce, and Joseph Grigar



TABLE 1 Formulary Submission Checklist

A COMPLETED FORMULARY SUBMISSION CHECKLIST SHOULD ACCOMPANY EACH SUBMISSION, WITH A BRIEF EXPLANATION FOR ALL MISSING DATA.

SUBMISSION PROCESS YES NO

Have you met with [PLAN NAME] staff to review the submission process?

Have you agreed upon a submission date with [PLAN NAME]?

Have you requested summary data to identify baseline characteristics of the plan population?

Have you included an explanation for any missing data? (Check yes if not applicable)

PRODUCT INFORMATION YES NO

Has a product description been provided?

Has a list of approved indications been provided?

Has the place of this product in therapy for each indication been included?

Have copies of treatment guidelines been provided?

Have the intermediate and final outcomes of therapy been listed?

Have you listed any coprescribed drugs by indication?

Have you identified the comparator drugs by indication?

SUPPORTING CLINICAL INFORMATION YES NO

Have you identified all relevant clinical and other experimental studies for the product?

Have you identified all relevant clinical and other experimental studies for comparator products?

Are copies of all studies identified included in the submission package?

Have you provided a spreadsheet summary of all studies identified?

Do the outcomes need to be translated into effectiveness terms?

Have these translations been included in the submission?

Have you included all relevant nonexperimental studies for the product?

Have you included all relevant nonexperimental studies for comparator products?

Have you provided a spreadsheet summary of all nonexperimental studies for the product?

Do the outcomes in nonexperimental studies need to be translated into effectiveness terms?

Have these translations been included in the submission?

SUPPORTING ECONOMIC INFORMATION YES NO

Have you identified all relevant pharmacoeconomic (PE) studies?

Have you justified the relevance of these PE studies for this population?

Have you provided a spreadsheet summary of these PE studies, detailing their relevance?

Have you developed a therapy intervention framework for each indication?

Have you confirmed the therapy intervention framework with the health plan?

Have you identified the characteristics of patients to be switched to this product?

Have you identified the characteristics that would exclude patients from using the product?

Continued next page



information on economic impact that can be customized to thepopulation of the specific health plan. This paper reviews theAMCP Format for Formulary Submissions and details the for-mulary submission dossier. The template is published on theAMCP Web site at www.amcp.org for public use.2

? ? Guidelines and Drug Coverage Decisions

For most health plans, the formulary has become the foundationfor the entire drug benefit program. Successful management of adrug formulary often hinges on the integrity of the criteria and evi-dence used to make approval or removal decisions. Unfortunately,adequate data often are not readily available; unpublished studiesand information on unapproved indications are difficult to obtain;manufacturers do not routinely supply data addressing quality-of-life and economic outcomes; and the time required for clinicalpharmacists to assemble, critically evaluate, and summarize data

for the P&T committee is excessive. To minimize these problems, formulary guidelines may be

used. Guidelines support the selection of a rational drug for-mulary by (1) standardizing the information required from themanufacturer; (2) formalizing the importance of a drug’s impacton both the health plan and its enrolled patient population; and(3) making the assumptions and evidence influencing formula-ry choices explicit, transparent, and relevant.

AMCP’s format is consistent with an international trend towardevidence-based health care decision making. Perhaps the mostprominent effort is that of the National Institute for ClinicalExcellence (NICE; www.nice.org.uk) in the United Kingdom, wherethe National Health Service (NHS) chartered NICE to provide guid-ance for England and Wales on the appropriate use of selected tech-nologies, including drugs, based on evidence of their social valueand impact on health budgets. In turn, NICE guidelines detail the

TABLE 1 Formulary Submission Checklist (continued)

COST IMPACT ASSESSMENT YES NO

Is a baseline prevalence analysis of resource utilization and costs included?

Have you structured these baseline estimates in terms of your therapy intervention framework?

Have you detailed the scenarios for cost impact assessments?

Have you highlighted the assumptions made for projecting patient switching behavior?

Have you justified the scenarios and assumptions for this plan’s patient population?

Have you provided aggregate cost impact assessments for the next three years?

Have you provided a breakdown of the costs by medical resource utilization and drug categories?

Have you included a proposal on how these cost-impact projections might be monitored?

Have you explained how differences between projections and actual costs might be resolved?

Is the cost of your proposed intervention program included in the cost assessment?

OUTCOMES IMPACT ASSESSMENT YES NO

Is a baseline prevalence analysis of patient outcomes included?

Have you structured these baseline estimates in terms of your therapy intervention framework?

Have you provided details for the scenarios for outcome impact assessment?

Have you provided details for the assumptions made for projecting patient switching behavior?

Have you justified the scenarios and assumptions for this plan’s population?

Have you provided aggregate patient-outcome impact assessments for the next three years?

Have you provided a breakdown of the costs by medical-resource utilization and drug categories?

Have you included a proposal on how patient outcomes might be monitored?

Have you explained the differences between the projected and actual patient outcomes?

Have you provided electronic copies of all spreadsheets or models used?

Will a disease- or care-management strategy be used with the introduction of your product?

Is documentation on this intervention program included in the submission?

YES NO



content of requests for clinical and economic evaluations ofselected technologies from manufacturers. These data are inde-pendently reviewed and used in the NHS appraisal process.Guidelines not unlike AMCP’s have now been issued in a grow-ing number of other European countries and by the CanadianCoordinating Office of Health Technology and the AustralianPharmacy Benefit Advisory Committee, which in the early 1990swas the first authority to issue “pharmacoeconomic” guidelines.3–5

In the United States the promulgation of health economicguidelines for formulary submissions has been led by RegenceBlueShield in Seattle in the private sector and in the public sec-tor by the Centers for Disease Control and Prevention and theUnited States Public Health Service, with some academicianscontributing.6–10 Evidence-based decision making about phar-maceuticals is increasingly embedded in the process of numer-ous other health-related entities, including the United StatesMedicare Coverage Advisory Committee (MCAC), the BlueCross Blue Shield Technology Evaluation Center (TEC), and thenew Blue Cross Blue Shield RxIntelligence group. They aresending a clear message to producers and sponsors of healthcare technologies that decisions will increasingly be made onevidence of value.

Unfortunately, early evidence from Australia suggests thatthe quality of vendor submissions can vary substantially; how-ever, the Australian authorities continue to endorse the processas helpful in formulary decision making. 11 The AMCP guide-lines promote practices that enhance the use of consistent sub-missions and reduce the perceived threats to credibility notedwith current practices.12 In the past, standard formulary kitssupplied limited clinical information and no economic data.These kits failed to communicate the value of pharmaceuticalsand were at best of limited use to health plans. By consideringtotal cost and health impact, the new AMCP guidelines shouldbe able to move managed care away from the pharmacy silo-budgeting approach typical of formulary decisions.

The AMCP guidelines offer both a process and a template.Although economic considerations receive substantial atten-tion, they are subordinate to clinical benefit, notably safety andefficacy. However, because the state of practice in economicevaluations is just evolving, more detailed guidance is needed inthis area. Submission of information in the recommended for-mat does not guarantee product approval, but it is a necessaryfirst step in rational drug selection within constrained budgets.

? ? Overview

The Food and Drug Administration (FDA) requires, within cer-tain limits, that information provided to health professionals(including health plans) be supported by evidence detailed onthe product label. Health economic and outcomes data (includ-ing computer simulation models) can be supplied in accor-dance with section 114(a) of the Food and Drug Modernization

Act of 1997.13 Manufacturers cannot routinely make some infor-mation, especially comprehensive economic evaluation data,available to health plans without an explicit request.

The AMCP guidelines will constitute a request for all avail-able information on products that allows the manufacturer toprovide broader information than would be available without it,such as projections of effectiveness from efficacy data, includinglong-term outcomes and other “off-label” unapproved usages.Health plans could also get retrospective database studies,health economics information, and other outcomes data.

In general, the AMCP guidelines do not restrict formularysubmission dossiers to a specific set of information, study

TABLE 2 Product Description

• Generic name, brand name, and therapeutic class of the product

• All dosage forms, including strengths and package sizes

• NDCs for all formulations

• Copy of the official product labeling and literature

• Cost per unit size (AWP and plan contract price, if available)

• DPS/AHFS drug classification

• FDA-approved and other studied indications

• Pharmacology

• Pharmacokinetics

• Contraindications

• Warnings/precautions

• Adverse effects

• Interactions (drug/drug, drug/food, drug/disease) and suggestionsfor avoiding them

• Availability, dosing, and administration

• Coprescribed or concomitant therapies, including dosages

• Comparison with the pharmacokinetic/pharmacologic profile ofother agents in the therapeutic area

Notes: AHFS is American Hospital Formulary Service; AWP is averagewholesale price; DPS is department of pharmacy services; FDA is Foodand Drug Administration; NDC is National Drug Code.

TABLE 3 Disease Description

• Epidemiology and risk factors• Pathophysiology• Clinical presentation• Approaches to treatment—principal options/practice patterns• Description of alternative treatment options (both drug and non-

drug)• Place of the proposed therapy in treatment (e.g., first line)• Expected outcomes of therapy• Other key assumptions and their rationale



research design, or format. Rather, the guidelines are intendedto delineate what information the health plan needs.Manufacturers are encouraged to supply all available data onclinical and economic benefit, rather than just the clinical trialsused to support licensing or those used in promotional materi-als. Consequently, it is recommended that the manufacturer col-laborate early in the process with the health plan’s representa-tives (see Section 5.4: Agenda for Presubmission Meeting).

Recommended Process

The recommended steps for the formulary submission process are:Step 1. At least six months before submission, a letter

(Notice of Intention to Submit) should be sent to notify thehealth plan’s pharmacy director or formulary manager of themanufacturer’s intent to submit a product for formulary consid-eration. This letter should include the anticipated timelines forthe submission, allowing the health plan to schedule a reviewand assign the submission to a subcommittee.

Step 2. The manufacturer should schedule a presubmissionmeeting with representatives of the health plan to review theformat’s requirements and to identify any data needed to estab-lish a baseline for assessing product impact. This meetingshould also address how to capture these data (see Section 5.4:

Agenda for Presubmission Meeting).Step 3. At least two months before the P&T committee

meets, copies of the submission should be provided to thehealth plan’s designee. It should be accompanied by an execu-tive summary, a completed checklist (see Table 1, page 273),and justification for any incomplete or missing data.

Step 4. Once the dossier has been received, the healthplan’s designee will review the submission and may ask themanufacturer to submit additional information to complete thedossier.

Step 5. At least two weeks before the P&T committeemeets, the health plan should inform the manufacturer in writ-ing as to whether the dossier is considered complete andwhether it will be abstracted for the committee’s consideration.If it is not considered complete or useful, the dossier will bereturned to the manufacturer with an explanation of why it wasnot submitted.

Step 6. At the P&T committee meeting, the health plan’sdesignee will summarize the manufacturer’s dossier, present theprincipal arguments for and against listing the product on theformulary, and describe any conditions that apply.

TABLE 4 Study Summaries

• Name of the clinical trial or study, location, and study date

• Trial design, randomization, and blinding procedures, researchquestions,* type of economic study,* study perspective*

• Washout, inclusion, and exclusion criteria

• Sample characteristics (demographics, size, disease severity, comor-bidities), treated population (actual or assumed)*

• Patient follow-up procedures (e.g., if an intention-to-treat designwas used, were drop-outs followed?), treatment period*

• Treatment and dosage regimens, treatment framework,* resourceutilization classification,* unit costs*

• Clinical outcome measures; outcomes evaluated*

• Other outcome measures (e.g., quality of life), principal findings*

• Statistical significance of outcomes and power calculations

• Validation of outcomes instrument (if applicable)

• Compliance behavior

• Generalizability of the population treated, relevance to health plan’senrolled population being treated*

• Publication citations/references used

• Limitations of study design

Note: Items with asterisks are necessary for economic studies.

TABLE 5 Information to be Included in theAnalytical Model

• Disease or condition, its natural history, clinical course, and out-comes

• Primary treatment options and the treatment process (clinical path-way) for each option. If the health plan employs a treatment guide-line for this condition, follow this framework. Alternative clinicalpathways presented by the manufacturer also may be considered.

• Proportion and characteristics of patients being treated by eachclinical pathway

• Product and other medical resources used to support each clinicalpathway

• Costs of product and other medical resources consumed withineach clinical pathway

• Outcomes of therapy for each clinical pathway, including expectedproportion of treatment failures and mean or median time to fail-ure, if known. These outcomes can be broadly and uniquelydefined by the manufacturer and can be modeled from other datasources. The manufacturer should address the relevance of theselected outcomes measure and generate both baseline and project-ed outcomes and impact assessments.

• Incremental cost and outcome analyses, presented in eithercost/consequences tables or as cost-effectiveness ratios, and totalcosts

• Results tables that provide outcomes in raw form (e.g., total eventsavoided) and prevalence of disease (events avoided/1,000 patients)

Step 7. The manufacturer is informed in writing of the com-mittee’s decisions about listing the product on the formulary andany recommendations for restricting access. If the product isdenied or restricted, the health plan should provide a detailedrationale, with guidance for reconsideration or appeal.

? ? Health Plan Role

The health plan must provide health plan–specific data to sup-port the manufacturer putting together the formulary submis-sion. The specific data that will be used should be agreed uponduring the presubmission meeting. If the health plan cannotprovide the necessary data on which to build the impact model,it should agree to accept a model that uses data from othersources (e.g., national, regional, another health system).

To assess the complexities of the outcomes data provided bythe manufacturer, a health plan may use the guidelines forauthors and peer reviewers published in the British MedicalJournal, which provides a checklist to health plans for evaluat-ing economic and outcomes models.14

? ? The Manufacturer’s Role

After soliciting data from the health plan, the manufacturer inte-grates into a clear, concise, and comprehensive document pub-

lished and unpublished data evaluating the efficacy, safety, eco-nomic impact, and other medical outcomes associated with theuse of the product. If data are unavailable or incomplete, themanufacturer is advised to explain why, and state when theywill be provided. The manufacturer should reveal the identitiesof the authors of the submission document and all primary eco-nomic evaluations. The exchange of information will beimproved if the manufacturer specifies a person who can answerquestions and give health plan reviewers additional informa-tion. The manufacturer forwards the completed dossier to theformulary manager at the health plan.

? ? The Formulary Submission

The formulary submission dossier includes the following sec-tions: • Section 1: Product Information • Section 2: Supporting Clinical and Economic Information • Section 3: Impact Model Report • Section 4: Clinical Value and Overall Cost • Section 5: Supporting Information: Bibliography, Checklist,

and Appendices.A detailed description of the content of each of these sec-

tions, with suggested maximum page lengths, follows.



Care Pathways: A general method of using predetermined, time-staged, evidence-based actions for managing the care of patients whohave clearly defined diagnoses or require certain procedures. Ideally,care pathways should apply to managing patients moving among amanaged health care system’s multiple levels of care and practice set-tings. Other terms for care pathways include clinical care plans, clini-cal pathways, critical pathways, care guides, and care maps.

Cost and Outcome Modeling: A quantitative modeling method usedto estimate the impact of formulary changes on (1) potential healthoutcomes and (2) total costs of drug and medical care in a popula-tion. One possible use of cost and outcome modeling, for example, isto extrapolate trial-based efficacy data into effectiveness and cost-effectiveness end points; cost and outcomes impact data from modelscan then be used to assess the health and overall fiscal consequencesof formulary changes. Estimating effects of a new product on the totalcosts and outcomes of care for health plan members is preferable.

Dossier: A detailed report for each product submitted by the manu-facturer for consideration that contains (1) clinical and economicdata from published and unpublished studies and (2) a disease-basedeconomic model to project effect introducing the product would haveon health and economically across the entire health plan system.

Effectiveness: The actual effects of treatment by the drug under reallife conditions (e.g., patients forgetting to take their doses, physiciansprescribing doses higher than those recommended, uncontrolled sideeffects). Head-to-head effectiveness studies with similar medicationsare preferable.

Efficacy: The potential effects of using the drug for treatment underoptimal circumstances (e.g., patients taking all doses at the righttimes, physicians prescribing correct doses, side effects appropriatelymanaged). Efficacy studies are typically the foundation of new drugsubmissions to the Food and Drug Administration. Active comparatortrials evaluating efficacy are preferred to placebo comparisons.

Formulary: A continually updated list of medications, related prod-ucts, and information representing the clinical judgment of physi-cians, pharmacists, and other experts in the diagnosis and treatmentof disease and the promotion of health.

Formulary System: An ongoing process whereby a health careorganization, through its physicians, pharmacists, and other healthcare professionals, establishes policies on the use of drugs, relatedproducts, and therapies, and identifies those that are the most med-ically appropriate and cost-effective to best serve the health interestsof a given patient population.

TABLE 6 Terms and Definitions

Section 1: Product InformationSection 1.1: Product Description (10 pages maximum). Inthis section, the new product should be compared with otheragents commonly used to treat the condition, whether or notthese products are currently on the health plan’s formulary. Theproduct description consists of information that traditionallyhas been incorporated in a product monograph (see Table 2,page 275). It should include a detailed discussion of the FDA-approved indications, the date approval was granted (or isexpected to be granted), and any data on off-label use. The sec-tion should also discuss comparative products or services thatthe proposed product is expected to replace (including non-drug as well as drug interventions). The information in this sec-tion should be presented in tabular form for clarity.

Section 1.2: Place of the Product in Therapy (3 pagesmaximum per disease). This section provides the disease con-text: to assess the impact of the new product effectively, the

clinical condition being treated and the role of the product inits treatment must be well understood. Although the manufac-turer is responsible for determining the relevant treatmentoptions for comparison, the determination should be madewith guidance from the health plan during the presubmissionmeeting. The manufacturer should include information aboutthe disease, characteristics of patients treated for the condition,a brief summary of the literature for each topic, and the infor-mation identified in Table 3, page 275, and when feasibleshould attempt to generalize these findings to the health plan’spopulation. This section should also address the implications ofany differences between the literature and the health plan’spractice patterns and patient population. If more than one dis-ease is addressed, each separate condition should be described,with results of studies presented in tabular form.

Section 2: Supporting Clinical and Economic Information This section summarizes published and unpublished clinicalsafety, efficacy, and economic evaluations. Studies should besummarized in a clear, concise format; presenting data frommultiple studies in tabular form is strongly encouraged. Table4, page 276, lists items that should be summarized for eachstudy.

Section 2.1: Presenting Clinical Study Results (1 page max-imum per study). The manufacturer should summarize each clin-ical trial and is encouraged to provide any head-to-head clinicalstudies between the proposed product and the principal com-parators, with no more than five trials from each of the follow-ing categories:• safety and efficacy trials; • prospective effectiveness (e.g., large sample) trials; • trials examining noneconomic end points, such as health sta-

tus and quality-of-life measures (previous validation and reli-ability studies of the instruments used should be referenced);and

• retrospective studies.Summaries of principal trial results of key comparator prod -

ucts, while desirable, are not required. Review articles andmeta-analyses, with particular emphasis on inclusion andexclusion criteria and main outcome measures, also may besummarized. For each study, the section should describeimportant study findings and comment on their implicationsfor the health plan’s patient population.

Information from all known studies on the product shouldbe summarized in a spreadsheet. The spreadsheet should incor-porate citation, if published; sample size; end points; studydates; treatments; results; design; inclusion/exclusion criteria;statistical significance; and study limitations.

Section 2.2: Clinical and Disease Management InterventionStrategies (3 pages maximum). This section should summarizeany studies or reports that evaluate the impact of the product being



1. Review of intended indications

2. Review of clinical studies• Explanation of comparators used and determination of their

appropriateness• Determination of the level of data needed for efficacy claims• If the data are insufficient, discussion about using nonexperi-

mental data to supplement them • Determination of whether the submission can proceed without

additional data

3. Review of cost-impact assessment• Evaluation of cost data and how they compare to the health

plan’s data• Discussion of the level of patient switching• Discussion of future studies and enhancements• Discussion of incorporation of health plan data• Determination of the level of data needed for submission• Determination of whether submission can proceed without addi-

tional data

4. Review of outcomes-impact assessment • Evaluation of the outcome markers used• Discussion of the level of anticipated patient switching• Discussion of the level of patient compliance• Discussion of the conversion of efficacy to effectiveness• Discussion of the appropriateness of modeling assumptions• Discussion of future outcomes monitoring• Determination of the level of data needed for submission• Determination of whether the submission can proceed without

additional data

TABLE 7 Proposed Agenda for Presubmission Meeting

proposed as part of a disease or care management interventionstrategy, with particular attention to variables that have provento be problematic for the health plan.

Section 2.3: Economic Evaluation Supporting Data (1page maximum per study). Economic evaluations mayinclude prospective cost-efficacy and cost-effectiveness studies,cross-sectional or retrospective economic evaluations, reviewarticles, and meta-analyses. This section may include studieswith a variety of analytic designs, such as prospective studiespiggy-backed onto pivotal clinical trials or naturalistic compar-ative, retrospective, or modeling studies. Since this portion ofthe document is to be a comprehensive assessment of availableevidence, the number of studies is not to be restricted bymethodological standards. However, the health plan will judgethe merit of individual studies based on published standards forconducting and reporting them.6, 8, 9, 15–22

Section 3: Impact Model Report (15 pages maximum) Section 3.1: Model Overview. Properly constructed pharma-coeconomic models can combine estimates of treatment effective-ness, resources consumed (i.e., costs) by each treatment process,and the uncertainty of these estimates for predicting systemwideconsequences of formulary changes. Such models can supportdecisions about adding a new product to the formulary, help defineits specific role in the environment of a given health plan, and helpcreate benchmarks against which the product’s future performancecan be measured. To minimize the potential for bias in economicevaluation, manufacturers should follow generally accepted rulesof scientific conduct.23, 24

The type of analytic model described in this section is a pre-condition for the health plan to evaluate how the new product, ifadopted, is likely to affect costs and clinical and humanistic out-comes for the plan’s enrolled population. Even though the specificformats may vary, each should incorporate a comprehensive, dis-ease-based analytical model tailored to the plan (or that can bemodified) and incorporating all the items listed in Table 5 (see page276).21 The manufacturer also should separate volume of resourcesused and unit costs for each resource, perform sensitivity analyseson pivotal estimates and assumptions, consult with the health planearly in model development to ensure the incorporation of appro-priate comparator products and end points, and present the fol-lowing information in tabular form: • total event rates (number of events avoided), • total resource utilization, • total costs, • total effectiveness, • incremental costs, and • incremental effectiveness.

When possible, measures of total and incremental effective-ness should incorporate natural units as well as quality-adjust-ed life years. Furthermore, to assist decision makers who may

be less familiar with pharmacoeconomic techniques, informationshould ideally be presented in terms familiar to the average healthplan administrator (events/1,000 patients, per member per month[PMPM] cost, etc.).

The model should be based on the clinical trial and economicdata, as modified by realistic expectations of the plan, practice pat-terns within the plan, and the patient population enrolled. For themodel to be realistic, the manufacturer may need either to obtaininformation from the health plan or, if these data are not available,to provide best estimates and a supporting rationale. The manu-facturer is encouraged to contact the formulary manager early inmodel development to find out what data are available. Otherinformation sources include randomized controlled trials, retro-spective analyses, case-control studies, cross-sectional surveys, casereports, and expert opinion.

The model’s time frame is a critical element. For chronic ill-nesses, both a one-year and a longer period should be adopted, asappropriate for the clinical problem and its resolution. For thelonger period, determining a final health outcome is recommend-ed. For acute illnesses, shorter periods may be appropriate.However, the overall effect on the health plan should be presentedon a yearly basis.

The model should consider recommendations published by thePanel on Cost-Effectiveness in Health and Medicine convened bythe U.S. Public Health Service.17 Although no standard approach isproposed, good modeling practices should always be followed.18



• List of intended indications• Summary of all studies to be included in the formulary submis-

sion, including: ? Clinical trials (experimental and nonexperimental)? Outcomes studies? Meta-analyses? Retrospective studies? Pharmacoeconomic models

• A general description of how cost and outcomes impact assess-ments will be developed, including:

? List of data sources (studies, databases, etc.)? Discussion of the conversion of efficacy to effectiveness for

both drug and comparators? Approach to modeling the environment of the health plan? Assumptions ? Suggested approach for determining patient characteristics

for switching• Summary of studies expected to be completed within one to three

years• A completed submission checklist

TABLE 8 Manufacturer Responsibilities for thePresubmission Meeting



Section 3.2: Clinical Trials: Claims for Safety and Efficacy(10 pages maximum). The primary considerations for adding aproduct to a formulary are how safe and effective the product willbe for the plan’s eligible population. Although clinical trials typi-cally focus on efficacy, the outcomes for modeling purposes mustbe translated to effectiveness (see Table 6, page 277). The bestquantitative estimates of effectiveness are required, with uncer-tainty handled analytically via sensitivity analysis. If these data arenot available, manufacturers should give their best estimate of theexpected effectiveness outcomes in usual practice. Translatingclaims from an efficacy to an effectiveness context also should beconsidered when (1) the model’s treatment period extendsbeyond the clinical trial; (2) outcomes supported by the trial areintermediate or surrogate; or (3) compliance, dosing, comorbidconditions, and the population of interest (e.g., children, elderly)are expected to differ from the efficacy trial data.

Poor compliance, especially for chronic conditions, canundermine claims that are based exclusively on clinical trials.All claims made for new products (whether in therapeutic oreconomic terms) should clearly state assumptions about patientcompliance. It is recommended that manufacturers documentanticipated compliance patterns from populations similar to theplan’s treatment population, if available. Additional clinical trialdata issues include establishing a trial’s external validity andcontrolling for provider and patient behavioral characteristics.

Section 3.3: Incidence and Prevalence Impact Assessments.An analytic model should reflect a prevalence rather than anincidence framework for chronic diseases. The prevalenceframework represents the patterns of treatment experienced bythe health plan over a specified time (e.g., a 12-month period),irrespective of the disease state reached by individual members.Incidence analysis, however, can be an acceptable modelingperspective for some acute diseases.

Outcomes should be differentiated by incidence and preva-lence. Typically, in incidence analysis, a cohort of patients istracked from start of therapy to an intermediate or final out-come. The manufacturer should translate such point-estimateimpact claims into prevalence-based claims, if possible, to clar-ify how the outcomes are achieved and how they are distributedwithin the treated population. If this is not possible, the manu-facturer should work with the health plan to estimate the neteffect of treatment across the entire patient population.

Section 3.4: Optimizing Patient Care. The impact assess-ment should start by assessing resource utilization and associat-ed medical costs at baseline for the designated therapy area,using data aggregated from service claims. This will allow themanufacturer to describe treatment options, determine patternsof resource utilization, and determine imputed costs for phar-macy and medical claims.

Treatment-pattern models should characterize the healthplan’s population and reflect best practices as promulgated by

task forces, learned societies, or government agencies. If theseutilization patterns differ from actual practice, the actual treat-ment patterns also should be modeled. It is desirable for themodel to depict both scenarios when actual and best practicesseem to differ.

Evidence for care-pathway effects on patient outcomes,resource utilization, costs, and therapy options should be pro-vided, as available, (1) under the evidence supporting clinicaland pharmacoeconomic claims, and (2) under the modelassumptions chosen for the impact assessment. Direct evidenceof health outcomes often may not be available, however; thehealth plan and the manufacturer must agree on whichapproach or assumptions will be acceptable. All assumptionsshould be justified as consistent with the prevalence frameworkof the analysis. These assumptions may be justified usingknown characteristics of patient population, epidemiologicalprofiles and clinical trials, meta-analyses, literature reviews, andexpert panels.

When a product is to be used to treat more than one disease,its effect should be modeled in each therapeutic area. Becauseof the complexity of constructing a model that simultaneouslyaddresses several therapeutic areas, a separate model for eachcondition is recommended.

Section 3.5: Presentation of Model Results. Results of themodel should be presented as follows:1. Estimates must include the cost of any additional resources

associated with implementing the therapy (e.g., disease man-agement).

2. Costs should be presented as total net costs of introducingthe new product.

3. Based on discussions between the plan and the manufactur-er, the submission should include recommendations aboutthe use of medical and pharmacy data to monitor costs andpatient outcomes and validate claims.

4. Effects should be estimated for the first three budget periodsafter product launch.Section 3.6: Exceptions. In some situations, a model devel-

oped for another health plan may eliminate the need for a newmodel. To be acceptable, the existing model should follow theframework described in this document and must demonstratethe systemwide impact of introducing the product to the healthplan’s formulary. The manufacturer must justify the adequacy ofpre-existing substitutes.

Section 4: Clinical Value and Overall Cost (2 pages max-imum) This section of the formulary submission dossier is the princi-pal opportunity for a manufacturer to communicate the value ofits product to the health plan. The manufacturer should brieflysummarize the information presented, state the expected perunit product cost, and estimate the health plan’s expenditures



for the product. Based on this information, the company shouldthen articulate the value argument to justify expenditures onthis product in terms of its anticipated effects on health-relatedquality of life and the economic consequences for the healthplan and its members. Through this process, product value isredefined as both parties move beyond simple cost containmentto optimize drug utilization given limited resources.

Section 5: Supporting InformationSection 5.1: References Contained in Dossiers. Submissionsshould list and provide copies of all clinical and pharmacoeco-nomic references used in previous sections and of informationsources from which estimates were drawn for use in the econom-ic evaluation. Referenced articles can be attached as appendices.

Section 5.2: Spreadsheet Models (Media). In addition tothe written report, the manufacturer should provide a transpar-ent, unlocked electronic copy of the model without the graphi-cal interface on a 3.5-inch disk as an Excel workbook, ASCIItab-delimited file, or an alternative format agreed upon by thehealth plan and the manufacturer. The model should be trans-parent to allow plan staff to investigate assumptions and calcu-lations, and to perform independent sensitivity analyses byvarying individual parameters. This model will be retained bythe health plan for internal analyses; it will not be released toany other party without the manufacturer’s permission.

Section 5.3: Data and Information from the Health Plan.Specific data elements will vary from plan to plan and model tomodel.

Section 5.4: Agenda For Presubmission Meeting. The pre-submission meeting should take place at least four to six monthsbefore the actual date of the formulary review to allow time forthe manufacturer to compile the necessary data. The proposedagenda can serve as a discussion guide to ensure that all relevanttopics are raised (see Table 7, page 278). At this meeting, themanufacturer should provide a copy and be prepared to discussall of the items listed in Table 8, page 279.

? ? Conclusion

All clinical decisions should be based on the best available evi-dence at the time of decision making; the formulary decisionprocess is a particularly opportune time for systematic review ofthe body of evidence for pharmaceuticals. The most importantevidence needed for informed decision making is clinical effica-cy; its source is the traditional clinical literature. However, asnumerous authors, scientific bodies, and authorities around theworld have come to believe, clinical efficacy evidence is neces-sary but not sufficient to guide optimal clinical care. Clinicalefficacy evidence alone cannot determine the value for moneyto be spent. The AMCP format is a modest effort to help P&Tcommittees improve the decision process. It envisions a collab-orative effort between the health plan that must decide whether

to make a drug available as a benefit to its patient populationand the manufacturer of the drug.

Performed appropriately, the process should reward manu-facturers of high-value drugs by providing a vehicle for the sys-tematic presentation of value for money. This is an opportunitythat manufacturers have wanted for a number of years.Similarly, doctors want their patients to have access to effectivedrug products. Furthermore, all parties who pay for care,including patients, employers, and taxpayers, want access tocost-effective products. Health plans, caught in the middle,need mechanisms to help assure that they receive optimal valuefor money spent. These guidelines are intended to accomplishthis objective; they are consistent with many other guidelinesbeing implemented worldwide.

The ultimate goal of this process is optimal patient care, tak-ing into account the reality of constrained budgets. The AMCPFormat for Formulary Submissions offers a clear, shared vision ofthe formulary process and information requirements that facili-tate partnership between the managed care plan and the productmanufacturer. The format describes the minimum informationrequired to support a comprehensive assessment of a product.With increased use of this format, the quality of submissions areexpected to improve over time; at a minimum, they will givepharmacists data that were often unavailable in the past.

References

1. US Pharmacopeia. Principles of a sound formulary system. USPharmacopeia, August 2000. Available at:www.amcp.org/public/legislative/DrugFormulary.pdf. Accessed on January 25,2001.

2. AMCP’s format for formulary submissions. Available at: www.amcp.org.Accessed on January 25, 2001.

3. Hoffmann C, Graf von der Schulenburg JM. The influence of economicevaluation studies on decision making. A European survey. The EUROMETgroup. Health Policy 2000; 52: 179–92.

4. Canadian Coordinating Office for Health Technology Assessment.Guidelines for economic evaluation of pharmaceuticals: Canada. 2nd ed.Ottawa, Ontario: Canadian Coordinating Office for Health TechnologyAssessment, 1997.

5. Commonwealth Department of Health and Aged Care. Guidelines for thepharmaceutical industry on preparation of submissions to the PharmaceuticalBenefits Advisory Committee, including major submissions involving econom-ic analysis. Available at: www.health.gov.au/haf/docs/pharmpac/gusubpac.htm.Accessed on April 1, 2001.

6. Mather DB et al. Incorporating clinical outcomes and economic conse-quences into drug formulary decisions: A practical approach. Am J Man Care1999; 5: 277–85.

7. Haddix AC et al., eds. Prevention effectiveness: A guide to decision analysisand economic evaluation. Oxford: Oxford University Press, 1996.

8. Gold MR et al. Cost-effectiveness in health and medicine. New York:Oxford University Press, 1996.

9. Drummond MF, Stoddart GL, Torrance GW. Methods for the economicevaluation of health care programmes. 2nd ed. Oxford: Oxford UniversityPress, 1997.



10. Warner KE, Luce BR. Cost-benefit and cost-effectiveness analysis in healthcare: Principles, practice, and potential. Ann Arbor, MI: Health AdministrationPress, 1982.

11. Hill SR, Mitchell AS, Henry DA. Problems with the interpretation of phar-macoeconomic analyses: A review of submissions to the AustralianPharmaceutical Benefits Scheme. JAMA 2000; 283: 2116–21.

12. Rennie D, Luft HS. Pharmacoeconomic analyses: Making them transpar-ent, making them credible. JAMA 2000; 283: 2158–60.

13. US Food and Drug Administration. FDA Modernization Act. November21, 1997. Available at: www.fda.gov/opacom/7modact.html. Accessed on April5, 2001.

14. Drummond MF, Jefferson TO. Guidelines for authors and peer reviewersof economic submissions to the BMJ. The BMJ Economic Evaluation WorkingParty. BMJ 1996; 313: 275–83.

15. Agro KE et al. Sensitivity analysis in health economic and pharmacoeco-nomic studies: An appraisal of the literature. PharmacoEconomics 1997; 11:75–88.

16. Detsky AS. Guidelines for economic analysis of pharmaceutical products:A draft document for Ontario and Canada. PharmacoEconomics 1993; 3:354–61.

17. Glick H, Kinosian B, Schulman K. Decision analytic modeling: Some usesin the evaluation of new pharmaceuticals. Drug Information J 1994, 28:691–707.

18. Henry D. Economic analysis as an aid to subsidisation decisions: Thedevelopment of Australian guidelines for pharmaceuticals.PharmacoEconomics 1992; 1: 54–67.

19. Johannesson M, Jönsson B, Göran K. Outcome measurement in economicevaluation. Health Econ 1996; 5: 279–96.

20. Kassirer JP, Angell M. The journal’s policy on cost-effectiveness analyses. NEngl J Med 1994; 331: 669–70.

21. Langley PC, Sullivan SD. Pharmacoeconomic evaluations: guidelines fordrug purchasers. J Managed Care Pharm 1996; 2: 671–77.

22. Sheldon TA. Problems of using modelling in the economic evaluation ofhealth care. Health Econ 1996; 5: 1–11.

23. Task Force on Principles for Economic Analysis of Health CareTechnology. Economic analysis of health care technology. Ann Intern Med1995; 123: 61–70.

24. Hillman AL, Eisenberg JM, Pauly MV et al. Avoiding bias in the conductand reporting of cost-effectiveness research sponsored by pharmaceutical com-panies. N Engl J Med 1991; 324: 1362–65.

OBJECTIVE: Shift selective serotoninreuptake inhibitor (SSRI) market sharein a large integrated health system toa more cost-effective alternative with-out the need for therapeutic inter-change.

SETTING: University of California DavisMedical Group (UCDMG)

PRACTICE INNOVATION: Integrate themanaged care team into the routineoperation of the medical group clinics.Influence prescribing patterns by aprogram of continuous physician edu-cation, giving the messages that (1) citalopram was the most cost-effective SSRI; (2) fluoxetine is theleast cost-effective, especially atdoses above 20 mg; and (3) thatcitalopram should be considered asinitial therapy for depression and forpatients not achieving optimal therapyon their current SSRI.

OUTCOMES MEASURES: Change inmarket share for SSRIs: prescriptionsfor citalopram and fluoxetine weremonitored for a year after the physi-cian education program began.

RESULTS: Market share for citalopramincreased dramatically. Starting belowthe reported national usage, the mar-ket share for citalopram at UCDMGincreased to above the national aver-age and the market share for fluoxe-tine decreased. Yearly estimated costsavings totaled $126,000—8% ofSSRI ingredient cost.

CONCLUSION: It is possible to shiftmarket share even in a large healthsystem without requiring therapeuticinterchange.

KEYWORDS: Market share, citalopram,fluoxetine, SSRI



RESEARCH

JOSEPH F. FISCHER, Pharm.D., Ph.D., is Managed Care Pharmacist, Universityof California (UC) Davis Health System, Sacramento, CA, Assistant ClinicalProfessor, Internal Medicine, UC Davis School of Medicine, and Assistant ClinicalProfessor, University of California, San Francisco (UCSF) School of Pharmacy;ROBERT M. MOWERS, Pharm.D., BCPS, is Managed Care Pharmacist, UCDavis Health System, Associate Clinical Professor, Internal Medicine, UC DavisSchool of Medicine, and Associate Clinical Professor, UCSF School of Pharmacy;DAVID J. ORMEROD, M.D., is Medical Director Managed Care, UC DavisHealth System and Associate Clinical Professor, Family Practice, UC Davis Schoolof Medicine; and ELLEN S. BURRISS, R.N., M.S., is Director, Woodland HealthFoundation, Woodland, CA.

AUTHOR CORRESPONDENCE: Joseph F. Fischer, Pharm.D., Ph.D., UC DavisMedical Center, 2315 Stockton Blvd., Sacramento, CA 95817; Tel: 916-734-3305; Fax: 916-734-5688; E-mail: [email protected]

ACKNOWLEDGMENT: Forest Pharmaceuticals provided trial script vouchersfor the project as well as the printing cost for the Primary Care Strategy forAdult Depression that was disseminated to physicians.


Authors

t is rare for a new drug, in a major therapeutic class, to bereleased at a significantly lower acquisition cost compared tothe existing drugs in that class. In March 1999 we targeted

the use of the selective serotonin reuptake inhibitors (SSRI) forreview; in July 1998 citalopram (Celexa) had become the fourthSSRI marketed in the United States.1 The average wholesale price(AWP) for citalopram was significantly less than the other SSRIdrugs available, especially fluoxetine (Prozac).

The managed care team at the UC Davis Medical Group(UCDMG) set out to treat depression more cost-effectively byincreasing the use of citalopram and decreasing the use of fluox-etine.2 The goal was to save at least $100,000 per year on SSRIprescriptions. In the first quarter of 1999 citalopram made up3% of the SSRIs prescribed at UCDMG but 9% of the SSRI mar-ket in the United States. (Market share is defined here as thenumber of prescriptions for each individual SSRI divided by thetotal number of SSRI prescriptions, expressed as a percent.)

A review of the literature established that citalopram was aneffective SSRI with a side-effect profile comparable to the otherdrugs in its class; it was comparable to the other SSRIs in over-dose and safer than tricyclic antidepressants.3, 4, 5 Since the AWPfor citalopram was lower than the older SSRI drugs, patientscould get a state-of-the-art SSRI as first-line therapy, physicianscould use an SSRI as first-line therapy, and managed care wouldbe providing, based on our analysis of the literature, the mostcost-effective drug in the class.

UCDMG is composed of 12 off-campus medical clinics andseveral outpatient clinics on the hospital grounds. The clinicsare staffed by 40 family-practice physicians and 34 internal-medicine physicians. UCDMG has 90,000 managed carepatients under contract with several large health maintenanceorganizations. Our three major insurers added citalopram totheir formularies.

The authors reviewed the data on the effectiveness, sideeffects, and costs associated with SSRI antidepressants with thechair of the department of psychiatry and the medical directorfor managed care. The managed care team was commissionedto implement an effort to move market share at UCDMG tocitalopram. The team consists of the medical director, the nursemanager, and two managed care pharmacists. Rather thanincreasing the presence of drug manufacturer representatives,internal detailing was used because it has more credibility withour providers. It would also allow the managed care pharma-


I

by Joseph F. Fischer, Robert M. Mowers, David J. Ormerod, and Ellen S. Burriss



cists to respond to drug information being provided to thephysicians from other sources, such as the pharmaceuticalindustry.

Because of the nature of depression, it would not have beenappropriate to ask physicians to do a therapeutic interchange,switching patients off their current SSRI if it was working. Wetherefore developed the following message to deliver to ourphysicians:• Citalopram is an antidepressant that is as effective as the other

SSRI agents.• Citalopram has an adverse-event profile similar to the other

SSRI agents.• Citalopram costs less than the other SSRI agents.• Fluoxetine is the least cost-effective antidepressant at UCDMG,

especially at doses greater than 20 mg.• Citalopram should be considered for all new patients who are

candidates for an SSRI and for patients who have not achievedoptimum therapeutic benefit from their current SSRI.

This message was delivered by as many methods as possible.The main pathway was at the monthly utilization management(UM) meetings at each clinic site. The managed care teamalready participated in these meetings. At each meeting, phar-macists comment on the treatment of depression. Initial discus-sions focused on the effectiveness and safety of citalopram incomparison to the other agents. The managed care medicaldirector was present at these meetings to lend full support tothe program. The medical director thanked the providers fortheir help in this area and shared cost-savings data with them.

To reinforce this message, the managed care pharmacistsprepared a clinical synopsis of citalopram and the treatment ofdepression. At least quarterly, the pharmacist e-mails all theclinic providers drug-information sheets related to managedcare and pharmacoeconomics. The citalopram sheet detailedthe basic pharmacology and toxicology of the drug and gave asynopsis of the clinical trials and cost comparisons.

Because of Joint Commission on Accreditation of HealthcareOrganizations (JCAHO) restrictions on storage and dispensingof samples, most of the UCDMG clinics have chosen not to sam-ple. The manufacturer gave trial script vouchers for citalopramto the managed care pharmacists, who distributed them at theUM meetings after initial discussions of the product. Severalphysicians requested an additional supply. In all, about 200vouchers were given out to UCDMG physicians; more than 100of them were redeemed at retail pharmacies.

Once the clinicians had been introduced to citalopram, thepharmacists discussed the pharmacoeconomics of the treatmentof depression. Each clinic is routinely given a list of its top-50drugs by cost. The pharmacist would highlight all the antide-pressants on this list, emphasizing the number of prescriptionsand the cost per prescription. This technique graphically showedthe physicians the significant cost reduction possible in the SSRIclass. The top-50 report shows the average ingredient cost for all

prescriptions; it reflects the strengths and quantities of eachdrug as actually prescribed by UCDMG physicians. The reportshowed that paroxetine was the second most cost-effective SSRIfor our medical group. This was incorporated into the pharma-cist message.

It also became important to counter misinformation that ourphysicians were getting from some pharmaceutical representa-tives. A common ploy was to imply that the cost savings usingcitalopram would be lost because the dose had to be increasedfrom 20 mg to 40 mg. This was deliberately misleading: theAWP as published in the Redbook (2000) for citalopram 20 mgwas $2.10 each versus the 40-mg tablet at only $2.19 each.There were no reports from physicians of increased office visitsto titrate dose. This misleading tactic of the pharmaceutical rep-resentatives has helped to solidify the position of the managedcare team as a source of unbiased drug information.

The managed care pharmacists then published a “PrimaryCare Strategy for Treatment of Adult Depression” in conjunctionwith the department of psychiatry. The front of the brochurelisted the treatment algorithm psychiatry suggests for the fami-ly practice and internal medicine physicians who are the pri-mary care providers (PCP) for patients. For patients withoutanxiety, two SSRIs were listed: citalopram and sertraline. Forthose with anxiety, the recommended SSRIs were citalopramand paroxetine. The brochure also contained information oncost for the most common antidepressants and a grid showingwhat drugs the local HMOs’ formularies covered.

Updates on the success of this project and cost savings arepresented to the medical group at the UM site meetings at leastevery other month. The initiative is monitored through pharma-cy claims data for the three largest health maintenance organiza-tions (HMOs) contracting with UCDMG. Plans H, P, and W allcover citalopram. Plans H and W cover all four of the availableSSRIs; plan P covers only citalopram and paroxetine. On plan P,fluoxetine and sertraline require prior approval (PA).

The market share percentage report represents the averagemarket share for the quarter.

? ? Results

By March 1999 citalopram had been available for nine months.The usage by UCDMG physicians was 2% for plan H, 3% forplan P, and 1% for plan W. Since the initiative began in March1999 there has been an almost linear increase in the marketshare for citalopram (see Figure 1, page 285). By the 2nd quar-ter of 2000, it had reached 21% (plan H), 26% (plan P), and14% (plan W).

All quarterly mean market-share percentages were com-pared for significance using chi-square analysis with appropri-ate p values reported. The increase in the percent market sharewas statistically significant (p<0.01) by the start of the thirdquarter of 1999 for all three plans. The national market share ofthis drug, its manufacturer reports, is currently 14% (August



2000). Plan P, with its more restrictive formulary, had the high-est increase. However, the difference between plan P and planH was not statistically significant.

Table 1, above, shows SSRI market shares for the second

quarter of 2000 for each SSRI and for new prescriptions only. Thepercent of new prescriptions for citalopram was 32% (plan P) and27% (plan H). The increase in mean market share of new pre-scriptions for citalopram compared to its total market share wasstatistically significant for the two plans for which data on new pre-scriptions were available (plan H; p<0.01, plan P; p=0.05).

Table 2, left, shows that market share for fluoxetine decreasedby 10% from the first quarter 1999 to the second quarter of 2000.The difference was significant for all plans (p< 0.01). The marketshares for paroxetine and sertraline were not significantly altered.

? ? Limitations

Pharmacy claims data were only available from 3 of the 10managed care companies doing business with UCDMC. Two ofthe companies are large HMOs while the third is a small com-pany (fewer than 50,000 lives). Not all of the managed carecompanies for which UCDMC has pharmacy risk cover citalo-pram. UCDMC has no control over formulary selection at thesecompanies.

? ? ?Conclusion

The managed care team achieved both its objectives. First, mar-ket share for citalopram increased significantly. In March 1999,citalopram use at UCDMG was 7%–8% below the nationalaverage; currently, it exceeds the national average by 6%.Second, the use of fluoxetine has decreased. It is possible tomove market share in both a positive and negative directionwithout therapeutic interchange.

Analysis of new prescriptions for SSRIs written in the sec-ond quarter of 2000 (Table 1) indicates the market share forcitalopram continues to increase. The percent of new prescrip-tions for citalopram was significantly higher than its marketshare for all (new and refill) prescriptions. Since this programdoes not involve therapeutic interchange, new prescriptions arean important indicator of market trends. While this is not a per-fect indicator, it does give an indication of initial therapy. (Somenew prescriptions will be the result of continuing therapywhere refills have expired.)

Plan W was the slowest to respond. This is the only plan tohave California Medi-Cal Managed Care (MCMC) members.They account for 61% of the membership of just over 13,000.Analysis of this subgroup of Plan W patients is shown in Table3, page 286.

The MCMC section of Plan W had a market share for citalo-pram of only 8% by the end of the first quarter of 2000, whileits commercial plan members had citalopram utilization of18%. Of the MCMC patients, 30%–50% have contracted physi-cian groups as their primary care providers. Because thesegroups are not part of the integrated health system and do nothave regularly scheduled meetings with the managed care team,the team only had access to about 50% of those who providecare to MCMC patients.

FIGURE 1 Citalopram Market Share as a Percent ofTotal Prescriptions for SSRIAntidepressants at UCDMG

30

25

20

15

10

5

0

Plan PPlan HPlan W

1stQuarter

1999

2ndQuarter

1999

3rdQuarter

1999

4thQuarter

1999

1stQuarter

2000

2ndQuarter

2000

1st Quarter 1999 2nd Quarter 2000

Plan H 38% 28%*

Plan P 25% 15%*

Plan W 40% 31%*

*Difference significant; p<0.01.

TABLE 2 Fluoxetine Percent of Market Share at UCDMG

Plan P Plan HNew Total New Total

Citalopram 32%** 26% 27%* 21%

Paroxetine 46% 49% 31% 30%

Fluoxetine 13% 15% 24% 28%

Sertraline 9% 10% 18% 21%

*p<0.01; **p=0.05

TABLE 1 Percent SSRI Market Share: New Prescriptions vs. Total Prescriptions, 2nd Quarter, 2000



MCMC patients function as an internal control group. Theimportance of the managed care team repeatedly reinforcing theinitiative is reflected in the increased use of citalopram in thecommercial versus the MCMC patients. Based on a comparison ofthe market share for the MCMC group versus the average marketshare for all the commercial plans the estimated yearly cost savingis $126,000, or 8% of total SSRI ingredient cost. This is a conser-vative estimate, as the managed care team did have contact with

the providers of care for about 50% of the MCMC patients, whichundoubtedly affected the use of citalopram.

As the market share for citalopram increases while fluoxetinedecreases, cost savings will continue. Savings from this program willcontinue until generic fluoxetine costs less than citalopram. Genericfluoxetine is scheduled to become available in September 2001.

This process should work for any large integrated health sys-tem, provided that a mechanism is in place to educate providersand monitor the movement of market share. The managed careteam identified the following factors as important for the successof this initiative: (1) the providers must be at risk for all or part ofthe pharmacy cost; (2) the medical director must give active sup-port to the project; (3) individual clinics and providers must beregularly informed about their role in the success of this project;and (4) the message must be repeated and reinforced often.

References

1. Anonymous. Citalopram approved for treatment of depression. Am J HealthSyst Pharm 1998 (Oct 1); 55(19): 1963.

2. Anonymous. Citalopram for depression. Med Lett Drugs Ther 1998 Dec 4;40 (1041): 113–14.

3. Tan JY, Levin GM. Citalopram in the treatment of depression and otherpotential uses in psychiatry. Pharmacotherapy 1999 (June); 19(6): 675–89.

4. Dewan MJ, Anand VS. Evaluating the tolerability of the newer antidepres-sants. J Nerv Ment Dis 1999 (Feb); 187(2): 96–101.

5. Barbey JT, Roose SP. SSRI safety in overdose. J Clin Psychiatry 1998; 59(suppl 15): 42–48.

1st Quarter 1st Quarter 1st Quarter 1st Quarter 1999 2000 1999 2000

Medi-Cal Group Commercial

Citalopram 0% 8% 7%* 18%*

Paroxetine 36% 37% 32% 30%*

Fluoxetine 42% 34% 41% 32%

Sertraline 22% 21% 20% 20%

*First quarter 2000 difference between Medi-Cal group and commercialgroup is significant; p<0.01.

TABLE 3 SSRI Use by Plan W Commercial Patientsvs. Medi-Cal Managed Care Patients

OBJECTIVE: To provide further evi-dence of the link between asthmaand allergic rhinitis.

DESIGN: A retrospective analysisusing medical and pharmacy claimsdata from 1994 and 1995. Prevalencerates of asthma and allergy and themean number of prescription claimsfor these patients were calculated.Patients with both asthma and aller-gic rhinitis, asthma only, and thosewith allergic rhinitis only were com-pared.

PATIENTS: Patients with allergicrhinitis or asthma in the 1994MarketScan file, with a diagnosis ofallergic rhinitis or two or more nonse-dating antihistamine prescriptions ortwo or more nasal inhaled steroidprescriptions during the year. Asthmapatients were identified by an asthmadiagnosis and the presence of atleast one beta-agonist prescription,or in the absence of an asthma diag-nosis, by two or more beta-agonistprescriptions during the year. Thefinal study population excluded thoseunder 12 or over 60 years of age,who did not have prescription drugdata in 1994 and 1995, and who werenot continuously enrolled in 1995.

RESULTS: Asthma was more preva-lent in the allergic rhinitis population

(10%) than in the general population.In addition, the rate of allergic rhinitisin the asthmatic population (44%)was much higher than the rate ofallergic rhinitis in the overall popula-tion (11%). On average, patients withboth conditions had approximately30% more asthma prescriptions(10.9) than did those with asthmaalone (8.4). Likewise, patients withboth conditions also had approxi-mately 31% more allergic rhinitis pre-scriptions (4.62) than did those withallergic rhinitis alone (3.52).

CONCLUSION: The increase in med-ication use by people with both asth-ma and allergic rhinitis lends supportto the idea that nasal inflammation isa marker for increasing dysfunctionof the entire respiratory tract. Giventhe increased prevalence of thesediseases, effective aggressive treat-ment would benefit a large segmentof the population. As the link betweenallergic rhinitis and asthma continuesto be established, it is probable thattreatments for one condition couldalleviate the coexisting condition.

KEYWORDS: Allergic rhinitis, asthma,drug therapy, disease prevalence, retrospective claims data


An Investigation of Allergic Rhinitis, Asthma, andMedication Use in a Privately Insured Population

RESEARCH

JODI CRYSTAL-PETERS, B.A., is Associate Director, The MEDSTAT Group,Washington, DC; CHERYL NESLUSAN, Ph.D., is Associate Director, ICOMHealth Economics, Johnson & Johnson; AMY WHITE, M.S., is Manager, TheMEDSTAT Group.

AUTHOR CORRESPONDENCE: Jodi Crystal-Peters, The MEDSTAT Group,4301 Connecticut Ave NW, Suite 330, Washington, DC 20008; Tel.: 202-719-7800; Fax: 202-719-7801; E-mail: [email protected].

ACKNOWLEDGMENT: Funding for this study was provided by IntegratedTherapeutics Group, Inc., a subsidiary of Schering-Plough.


Authors

llergic rhinitis is the most common allergic disorder.The incidence of this condition has increased inrecent years.1, 2 Asthma diagnoses have also increased

worldwide for several decades.2 Although allergic rhinitis andasthma commonly occur together, clinicians have yet to deter-mine the exact link between them. Evidence supporting a rela-tionship between asthma and allergic rhinitis has been widelyreported in the clinical literature.1 Both diseases share a num-ber of possible risk factors, including family history, allergicsensitization, adjuvant factors, and lifestyle factors.2

A study by Pedersen of patients with both allergic rhinitisand asthma examined the years elapsed from the onset of onedisease until the development of the other. 3 This study foundthat 75% of patients with both diseases experienced onset ofthe second disease within two years of the first. Other studieshave suggested that rhinitis frequently precedes asthma andthat upper airway dysfunction may be a predictive factor forsubsequent development of lower airway disease.4 Whetherrhinitis is the first manifestation of respiratory allergy for apatient who may eventually have asthma, or whether nasal dis-ease plays a direct role in causing asthma, has not been deter-mined. Wright et al. studied allergic rhinitis in children undersix years of age and found that having physician-diagnosedallergic rhinitis during the first year of life increased the chanceof having a later asthma diagnosis.5 Another opinion holds thatallergic rhinitis and asthma exist on a continuum of inflamma-tion involving one common airway, rather than consideringthem as separate diseases.6

Although the prevalence of asthma in the general populationis 3%–5%, asthma affects far greater proportion of patients diag-nosed with allergies. Recent evidence suggests that asthma inci-dence in people with allergic rhinitis ranges from 21%–58%.7, 8

The variation in rates is explained in part by differences in thepopulations studied and in how allergic rhinitis is defined.Estimates of allergic rhinitis incidence in asthmatics appear to bequite high, ranging from 71%–86%.7, 9 Allergic rhinitis seems tobe more common among asthmatics than is asthma among peo-ple with allergic rhinitis. In a review paper on allergic rhinitis andasthma, Yawn and colleagues note that the prevalence of allergicrhinitis in asthma patients differs among those diagnosed withseasonal as opposed to perennial allergic rhinitis.1 Only 6% ofasthmatics had perennial allergic rhinitis alone, whereas 24%had seasonal allergic rhinitis alone. Twenty-two percent of asth-


A

by Jodi Crystal-Peters, Cheryl Neslusan, and Amy White



matics had both forms of the condition. Some studies have suggested that appropriate treatment of

allergic rhinitis may alleviate symptoms of asthma.Pharmacotherapy for allergic rhinitis may increase airway cal-iber and decrease bronchial hypertension. For example, Grantet al. found that asthma symptoms improved in allergic rhinitispatients treated with an antihistamine in comparison to patientswho did not receive treatment.10 In their examination of theeffects of intranasal corticosteroids in patients with chronicperennial allergic rhinitis and mild asthma, Henriksen andWenzel found that four weeks of therapy significantly reducedboth rhinitis and asthma symptoms.11 Welsh et al. found thatallergic rhinitis patients receiving intranasal corticosteroids alsoexperienced improved lower airway symptoms attributed toasthma.12 Watson and colleagues found that intranasal corti-costeroids therapy reduced both rhinitis and asthma symptomseven though less than 2% of the drug was deposited in thelower airways.13

The purpose of this study is to provide further evidence thata link exists between asthma and allergic rhinitis. First, we com-pared the prevalence of allergic rhinitis and asthma in theMarketScan population to each disease population (all asthmat-ics and all persons with allergic rhinitis) to determine whethereither condition is more prevalent in those individuals alreadysuffering from the other condition. Second, we compared con-dition-specific medication use for people having allergic rhinitisonly or asthma only to those with both conditions. If asthmaand allergic rhinitis are related, it may be that persons with bothconditions have more complex upper and lower airwayinvolvement and thus need to use more medications to suc-cessfully manage their symptoms.

? ? Study Methods

Data The analytical file used in this study was constructed from the1994 and 1995 MarketScan Private Pay Fee-for-Service databas-es. These files contain data on more than four million coveredlives per year. Indemnity type insurance as well as noncapitatedmanaged care plans are represented. Included in these databasesare diagnosis, procedure, provider, benefit plan, and paymentinformation from medical claims for nearly 200 large, self-insuredemployers located throughout the United States. (Plan-specificbenefits information was available for only a subset of individualsin this study and so was not incorporated in the analyses.) Datafrom employees and their dependents are included. Outpatientprescription drug data—National Drug Code (NDC), copayment,deductible, total payment, and other elements—are available forroughly one million members per year.

Study Population and Analytical VariablesTo construct the study sample, we first identified patients with

allergic rhinitis or asthma in the 1994 MarketScan file. A patientwas considered to have allergic rhinitis if they had a diagnosisof allergic rhinitis (477.x) or two or more nonsedating antihis-tamine prescriptions or two or more nasal inhaled steroid pre-scriptions during the year. Asthma patients were identified byan asthma diagnosis (493.x) and the presence of at least onebeta-agonist prescription, or in the absence of an asthma diag-nosis, by two or more beta-agonist prescriptions during theyear. (International Classification of Diseases, 9th Revision,Clinical Modification [ICD-9-CM] codes used to identify asth-ma and allergic rhinitis are available upon request.) To arrive atour final study population, we excluded persons who wereunder 12 or over 60 years of age, who did not have prescriptiondrug data in 1994 and 1995, and who were not continuouslyenrolled in 1995. Note that continuous enrollment was identi-fied using the MarketScan enrollment file, and a claims-basedproxy. For those who did not have enrollment data, we checkedto see if they had any service claims during the first and lastquarters of 1995. If this criterion was met, we assumed that thepatient was continuously enrolled. Patients with a diagnosis ofchronic obstructive pulmonary disease also were excluded.

For each patient in the sample, we constructed a count ofthe number of prescription drug claims during 1995 for variousasthma and allergy medications. For allergy treatment, wegrouped prescriptions into two categories, nonsedating antihis-tamines (NSA) or nasal inhaled steroids (NIS). Medicationsexamined for the treatment of asthma were beta-agonists,inhaled steroids, oral steroids, theophylline, and cromolyn.Sedating antihistamines were excluded because many are soldover-the-counter and thus would not be represented by phar-macy claims.

Methods Using the definitions noted above, we calculated overall preva-lence rates for asthma and allergic rhinitis in the 1994MarketScan database. The initial prevalence rates were calculat-ed on the entire 1994 database and were not limited to thosepersons who were continuously enrolled. After applying theexclusion criteria to the analytic file, we compared the percent-age of patients in each of the three mutually exclusive subsam-ples by age grouping and gender type. To determine whetherdifferences exist along these dimensions between the threegroups, associated F-tests of significance were calculated. Meanmedication use was then examined. The overall mean numberof asthma drug claims for patients with asthma alone was com-pared to the mean numbers of such claims in those with bothconditions. We also compared the overall mean number ofallergic rhinitis drug claims for patients with allergic rhinitisalone to the mean numbers of such medications in those withboth conditions. These calculations examined whether therewere differences in the mean number of claims by drug type(beta-agonists, inhaled steroids, oral steroids, theophylline, or



cromolyn for asthma, and NSA or NIS for allergic rhinitis)across the groups of patients with one of the conditions versusthe group with both conditions. Pair-wise comparisons ofmeans were tested with t-tests.

? ? Results

Prevalence RatesApproximately 2.5% of the overall MarketScan population hadasthma in 1994. Asthma was more prevalent in the allergicrhinitis population (10%) than in the general population. Inaddition, the rate of allergic rhinitis in the asthmatic population(44%) was much higher than the rate of allergic rhinitis in theoverall population (11%). After limiting the sample to onlythose persons who were continuously enrolled in 1995, thefinal study population consisted of 54,171 individuals: 5,525with both allergic rhinitis and asthma, 42,686 with allergicrhinitis alone, and 5,960 with asthma alone.

Table 1, this page, presents the numbers and percentages ofpatients in each subgroup in the final study population by agecategory and gender. An ANOVA procedure revealed that theage composition of patients differed significantly among thethree study groups (p<.001). Those with both asthma and aller-gic rhinitis clustered in the middle-age groups, with 53.2%between 35 and 54 years in age; the age distribution was simi-lar among those with allergic rhinitis alone. Asthmatics, how-

ever, were divided relatively evenly across all age groups. Thedistribution of gender was also statistically different across thethree subsamples (p<.001), although in all cases women com-prised the majority of the samples.

Medication Use Tables 2, page 290, and 3, page 290, present the average num-ber of claims for asthma and allergic rhinitis medications bydrug type. Table 2 compares the mean number of asthma med-ications for patients with asthma alone and for those with bothasthma and rhinitis. The mean number of asthma-medicationclaims was significantly higher for those with both conditionsthan for those in the asthma-only category. This finding wasconsistent for men and women across all age groups. Of the5,525 patients with both asthma and rhinitis, 93% had at leasttwo asthma prescriptions; of the 5,960 people with asthmaalone, 91% filled at least two prescriptions for asthma drugs.Overall, those with both conditions had approximately 30%more prescriptions for asthma medications during 1995 thanpersons with asthma alone.

Although there were statistically significant differences inthe types of asthma medications between people with asthmaalone and those with both conditions, the mean number ofclaims appears to be similar. As explained above, the identifica-tion criteria for asthma included at least one prescription forbeta-agonists; thus, every participant had at least one asthma

Asthma Only Rhinitis Only Asthma and Rhinitis

N % N % N % p value *

Age Group

12-17 1,225 20.55 4,002 9.38 1,011 18.30 0.000

18–34 1,268 21.28 7,566 17.72 996 18.03 0.000

35–44 1,223 20.52 11,896 27.87 1,414 25.59 0.000

45–54 1,456 24.43 13,424 31.45 1,526 27.62 0.000

55–60 788 13.22 5,798 13.58 578 10.46 0.000

Gender

Male 2,475 41.53 15,578 36.49 1,967 35.60 0.000

Female 3,445 57.80 27,026 63.31 3,534 63.96 0.000

Not reported 40 0.67 82 0.19 24 0.43 0.070

Total 5,960 43.41 42,686 16.36 5,525 40.24

*P values were calculated using an F-test from the ANOVA procedure.

TABLE 1 Descriptive Statistics for Age Groups and Gender, by Illness Category



drug claim. Therefore, the fact that in 1995 the greatest numberof average claims was found among the beta-agonists, at 5.4claims per asthma/rhinitis patient versus 5.1 claims per asth-matic, is not unexpected. The second largest category of pre-scriptions was inhaled steroids, prescribed at an average of twoclaims per asthma/rhinitis patient compared to one claim perasthmatic. Inhaled steroids are intended through regular use todecrease symptom frequency for patients with chronic asthma.14

Both inhaled steroids and cromolyn are intended for long-termcontrol of mild, persistent asthma. Both groups of study partic-ipants filled less than one prescription claim for cromolyn peryear on average. They also had less than one claim on averagefor oral steroids, which are indicated for more severe and per-sistent asthma. There were more claims for theophylline, abronchodilator used for long-term asthma control. On average,patients with both asthma and rhinitis had 1.4 prescriptionclaims during the study period, while patients with asthmaalone had 1.3 theophylline claims.

Table 3 presents the mean number of allergic rhinitis med-ications by asthma status. Patients with asthma alone or withboth conditions had approximately 31% more allergic rhinitisprescriptions on average than the rhinitis-only patients. Of the5,525 asthmatics having rhinitis, 84% submitted at least oneclaim for an allergy drug and 75% submitted two or moreclaims. The proportion with one claim was slightly smalleramong people with rhinitis alone, about 80%. In total, asthmaand allergic rhinitis patients had an average of 4.6 prescriptionsfor allergic rhinitis in 1995 versus 3.5 prescriptions on averagefor the rhinitis-only patients. People having both asthma andallergic rhinitis also had significantly more NSA and NIS pre-scriptions than did patients with rhinitis alone.

? ? Discussion

This study provides further evidence of the substantial linkbetween allergic rhinitis and asthma. The high prevalence ofallergic rhinitis among asthmatics observed in these data is con-sistent with results from recent studies, and the rate of asthmaamong persons with allergic rhinitis was notably higher than inthe general population.3, 5, 7, 8 These findings lend support to theview that certain upper and lower airway conditions consideredto be separate diseases should be understood instead as differ-ent points on a continuum of airway inflammation.6

Because incidence and prevalence rates for both diseaseshave been increasing, effective treatment for both diseaseswould benefit a large segment of the population. As the linkbetween allergic rhinitis and asthma continues to be estab-lished, it is probable that treatments for one condition couldalleviate the coexisting condition.

The prevalence of allergic rhinitis (10.7%) in our insuredpopulation falls within the 9%–20% rates previously reportedfor the U.S. population as a whole.1 The prevalence rate of asth-ma (2.5%) is lower, however, than the 4%–6 % rate estimated

for the U.S. population.16 The disparity in asthma rates has sev-eral potential causes. Asthma prevalence varies by age, race andethnicity, family income, urbanization, and birth weight, amongother characteristics.17 Our data represent employees of largefirms and their dependents, a group that is wealthier than aver-age and more likely to be older, white, and urban than the gen-eral population. In addition, rising asthma prevalence duringthe 1990s suggests that estimates from 1994–1995 will belower than those from more recent years.

In this study, patients with both asthma and allergic rhinitisappear to have a significantly greater number of prescription

Asthma Only Asthma and Allergic Rhinitis

Mean Std Dev Mean Std Dev p value *

Beta-agonists 5.11 5.56 5.40 5.87 0.007

Inhaled steroids 1.04 2.41 1.86 3.19 0.000

Oral steroids 0.58 1.56 0.76 1.66 0.000

Theophylline 1.25 3.21 1.44 3.41 0.002

Cromolyn 0.37 1.52 0.75 2.18 0.000

Total 8.35 9.49 10.21 10.88 0.000

*P values are the result of a t-test comparison.

TABLE 2 Mean Number of Claims for VariousAsthma Medications for Patients withAsthma, by Presence of ComorbidAllergic Rhinitis

Rhinitis Only Asthma and Allergic Rhinitis

Mean Std Dev Mean Std Dev p value *

Nonsedating 2.26 2.88 2.69 3.43 0.000antihistamines

Nasal-inhaled 1.26 2.23 1.93 2.90 0.000steroids

Total 3.52 3.76 4.62 4.70 0.000

* p values are the result of a t-test comparison.

TABLE 3 Mean Number of Claims for VariousTypes of Allergic Rhinitis Medications forPatients with Allergic Rhinitis, byPresence of Comorbid Asthma.


drug claims for asthma and allergy medications on average thando those with only one of the conditions. Similarly, Halpernfound that patients with symptomatic rhinitis had more asthmamedications, especially more inhaled and supplemental corti-costeroids, than did the general population.15 As postulated byCorren, the increase in the use of asthma medications inpatients with co-occurring conditions may indicate that thesepatients have increased asthma severity. 4 This study reveals,however, that patients with both conditions also have higherutilization of allergic rhinitis medications, lending support toCorren’s alternative explanation that nasal inflammation is amarker for increasing dysfunction of the entire respiratory tract.

Because we did not adjust the claims for the number of dayssupply per prescription and cannot verify that patients adheredto their treatment regimens, however, it is important to inter-pret these data with caution.

Medications that can be purchased without a prescriptionare often used in the treatment of allergic rhinitis. Individualswith this condition who use only over-the-counter medicationscannot be identified in the MarketScan database if they did nothave a doctor’s visit for their allergies. As a result, there may beindividuals in the asthma-only subsample who have allergicrhinitis who were not identified for this study.

References

1. Yawn BP et al. Allergic rhinitis in Rochester, Minnesota residents with asth-ma: frequency and impact on health care charges. J Allergy and ClinImmunology 1999; 103: 54–59.

2. Lundbäck B. Epidemiology of rhinitis and asthma. Clinical andExperimental Allergy 1998; 28:3–10.

3. Pederson PA, Weeke ER. Asthma and allergic rhinitis in the same patients.Allergy 1983; 38: 25–29.

4. Corren J. The impact of allergic rhinitis on bronchial asthma. J Allergy andClinical Immunology 1998; 101: S352–56.

5. Wright AL et al. Epidemiology of physician-diagnosed allergic rhinitis inchildhood. Pediatrics 1994; 94: 895–901.

6. Grossman J. One airway, one disease. Chest 1997; 111: 11S–16S.

7. Greisner WA, Settipane RJ, Settipane GA. Co-existence of asthma and aller-gic rhinitis: a 23-year follow-up study of college students. Allergy and AsthmaProcedures 1998; 19: 185–88.

8. Rachelefsky GS. National guidelines needed to manage rhinitis and preventcomplications. Annals of Allergy, Asthma, and Immunology 1999; 82:296–305.

9. Janson C et al. Increased prevalence of sleep disturbances and daytimesleepiness in subjects with bronchial asthma: a population study of youngadults in three European countries. European Respiratory J 1996; 9: 2132–38.

10. Grant AJ et al. Clinical aspects of allergic disease: Cetirizine in patientswith seasonal rhinitis and concomitant asthma: prospective, randomized,placebo-controlled trial. J Allergy and Clin Immunology 1995; 95: 923–32.

11. Henriksen JW, Wenzel A. Effect of an intranasally administered corticos-teroid (budesonide) on nasal obstruction, mouth breathing and asthma. AmerRev Resp Disease 1984; 130: 1014–18.

12. Welsh PW et al. Efficacy of beclometasone nasal solution, flunisolide andcromolyn in relieving symptoms of ragweed allergy. Mayo Clinical Procedure1987; 62: 125–34.

13. Watson WT, Becker AB, Simons FE. Treatment of allergic rhinitis withintranasal corticosteroids in patients with mild asthma: Effect on lower airwayresponsiveness. Journal of Allergy and Clinical Immunology 1993; 91:97–101.

14. Holzer SS et al. Asthma treatment costs using inhaled corticosteroids.Amer J Manage Care 1997; 3: 891–97.

15. Halpern M et al. Allergic rhinitis may increase asthma costs [abstract].Amer J Resp Critical Care Medicine 1996; 153: A860.

16. Expert panel report II: Guidelines for the diagnosis and management ofasthma, clinical practice guidelines, National Institutes of Health: NationalHeart, Lung and Blood Institute, 1997.

17. Sly RM. Changing prevalence of allergic rhinitis and asthma. Annals ofAllergy, Asthma, and Immunology 1999; 82: 233–48.



OBJECTIVE: To assess the ability ofcommunity pharmacists to identifymanaged care patients with diabeteswho are not achieving therapeuticgoals.

SETTING: A network of independentcommunity pharmacists in WestVirginia and southeastern Ohio in col-laboration with The Health Plan of theUpper Ohio Valley.

METHODS: Pharmacists conducted assessments of patients’ glycemiccontrol (HbA 1c), blood pressure (BP),lipid levels (total cholesterol, low-den -sity lipoprotein [LDL], high-densitylipoprotein [HDL], triglycerides), andbody mass index (BMI). The therapeu-tic goals were: HbA1 c less than 7%, BPlower than 130/85 mmHg, total cho-lesterol under 200 mg/dl, LDL lessthan 100mg/dl, HDL lower than 45mg/dl, triglycerides under 200 mg/dl,and BMI lower than 30. These indiceswere measured during scheduledappointments in the pharmacy bypharmacists who had completed acertificate program in diabetes care.Reports on each patient’s status,along with recommendations, weresent to the patient’s physician.

RESULTS: Fifty-four persons wereenrolled in the pharmacist programand complete clinical data wereobtained for 47 patients. The following percentages of patientswere identified as not achieving thetherapeutic goal for a particularmeasure: HbA 1c: 63.9%, BP: 56.3%,total cholesterol: 38.3%, LDL: 69.8%,HDL: 76.5%, triglycerides: 57.4%, BMI: 61.9%. Patients who were notreaching the therapeutic target werereferred to their physicians for addi-tional evaluation.

CONCLUSION: Pharmacists can identi-fy a substantial number of personswith diabetes who are not achievingthe goals for HbA1 c, blood pressure,lipids, and weight. This approach canfacilitate the continuous assessmentand improvement of care for managedcare enrollees with diabetes.

KEYWORDS: Diabetes, qualityimprovement


Collaborating with Community Pharmacists to Improvethe Quality of Diabetes Care in an IPA-model HMO

RESEARCH

DAVID P. NAU, R.Ph., Ph.D., is Assistant Professor, College of Pharmacy,University of Michigan, Ann Arbor, MI; JOSHUA D. BLEVINS, M.S., is DistrictSales Manager, Celtic Insurance, Charlotte, NC; STEPHEN E. NEAL, M.B.A.,R.Ph., is Director of Pharmacy Services, The Health Plan of the Upper OhioValley, St. Clairsville, OH. At the time of this study, Dr. Nau and Mr. Blevinswere with the School of Pharmacy, West Virginia University, Morgantown, WV.

AUTHOR CORRESPONDENCE: David Nau, R.Ph., Ph.D., College ofPharmacy, University of Michigan, 428 Church St., Ann Arbor, MI 48109-1065; Tel: 734-764-4483; Fax: 734-763-2022; E-mail: [email protected].

ACKNOWLEDGMENT: This project was supported in part by The Health Planof the Upper Ohio Valley, National Community Pharmacists AssociationFoundation, Bristol-Myers Squibb, and Hoerchst-Marion Roussel.

The authors would like to acknowledge the support of Kathy Parsons for the cre-ation of this program, as well as the contributions of all of the pharmacists andtechnicians who work at the participating pharmacies.


Authors

he treatment of persons with diabetes is often expensiveand inadequate. In 1992, 14.2% of direct health careexpenditures was used to treat patients with diabetes who

make up less than 5% of the population.1 This represents an aver-age annual expenditure on medical care for persons with diabetesof $9,493, compared to $2,604 for patients in general. Many ofthese costs are associated with the management of the complica-tions of diabetes, such as myocardial infarction or end-stage renaldisease (ESRD).2 The cost of treating an acute myocardial infarctionin a person with Type-2 diabetes is estimated at $27,630, whileESRD costs have been estimated at $53,659 per year.3

The American Diabetes Association (ADA) estimates that thecomplications of diabetes could be reduced dramatically ifpatients maintained adequate control of their diabetes. Resultsfrom the Diabetes Control and Complications Trial (DCCT) sug-gest that intensive treatment and monitoring could reduce therisk of retinopathy (76%), nephropathy (50%), neuropathy(60%), dyslipidemia (34%), and cardiovascular disease (41%).4

To decrease the risk of diabetes complications, the ADA rec-ommends that patients receive annual assessments of lipids andmicroalbumin, and that glycosylated hemoglobin (HbA1c) bemeasured two to four times per year, depending upon thepatient’s glycemic control.5 Numerous studies have shown thatthese assessments frequently are not done. A recent study ofMedicare claims from three states found that only 16% ofMedicare recipients with diabetes received at least one HbA1c testover the course of one year, 46% saw an ophthalmologist, and55% were screened for high cholesterol.6 Managed care organiza-tions tend to fare better than fee-for-service providers, but are farfrom meeting the ADA guidelines. A large health maintenanceorganization (HMO) in California reported that HbA1c tests weredone for 44% of its patients with diabetes and United HealthCareCorporation recently reported rates of about 60% for HbA1c test-ing in its enrollees.7, 8

As part of the Health Plan Employer Data and Information Set(HEDIS) 2000, the National Committee for Quality Assurance(NCQA) requires managed care organizations (MCOs) to trackkey indicators of the quality of care for persons with diabetes. Thefrequency of glycosylated hemoglobin and lipid tests for this pop-ulation are such indicators. MCOs strive to implement mecha-nisms to continuously improve their performance in monitoringthe care provided to their members with diabetes. Pharmacistscould potentially increase the number of persons who receivethese tests by conducting the assessments in the pharmacy.

Tby David P. Nau, Joshua D. Blevins, and Stephen E. Neal


Collaborating with Community Pharmacists to Improve the Quality of Diabetes Care in an IPA-model HMO

Increasingly, health care organizations are using pharmacists ornurses to assist in the monitoring and management of patients withdiabetes. The Veterans Affairs (VA) Medical Center in Pittsburghfound that persons with Type-2 diabetes who were enrolled in itspharmacist-based program experienced significant improvementsin glycemic control within six months.9 After adjusting for the costsof the program, they estimated that net savings to the VA MedicalCenter for 15 of their most severely ill patients was more than$103,000 per year. In 1997, Fincham and Lofholm suggested thatcommunity pharmacists could reduce health expenditures for dia-betes by $4,295 per patient.10 A network of community pharma-cists saved the city of Asheville, North Carolina, more than $900per patient per year on diabetes care, while a pharmacy in Virginiadocumented significant improvements in their patients’ glycemiccontrol after enrollment in a diabetes care program.11, 12

? ? Practice Innovation

The Ohio Valley Pharmacist Care Network (OVPCN) is a group ofindependent community pharmacists located in the northern pan-handle of West Virginia and southeastern Ohio. This network wasformed in 1998 with the goal of establishing common pharmaceuti-cal care programs to meet the needs of patients, physicians, and pay-ors in the region. At the time of this publication, the membership ofOVPCN consisted of thirteen pharmacists at seven pharmacies.

Representatives of OVPCN met with the Director of Pharmacyand Director of Quality Improvement at The Health Plan of theUpper Ohio Valley to identify the needs of this independent prac-tice association (IPA)-model HMO. In addition, faculty from WestVirginia University (WVU) School of Pharmacy were asked to helpidentify opportunities for quality improvement and to developtraining programs for the pharmacists. A collaborative effort wasinitiated to assess and improve the quality of diabetes care for theHMO’s enrollees.

The goals of the collaborative effort were to:• increase the percentage of patients with HbA1c lower than 8%

(ideally lower than 7%);• increase the percentage of patients with low-density lipoprotein

(LDL) cholesterol less than 130 mg/dl (ideally less than 100mg/dl);

• increase the percentage of patients whose blood pressure is lowerthan 130/85 mmHg;

• decrease diabetes-related ER visits, hospitalizations, and unsched-uled physician visits;

• increase treatment guideline adherence (regular eye exams, footexams, immunizations, microalbumin, HbA1c, fasting blood glu-cose [FBG], lipids, blood pressure [BP], weight);

• enhance health-related quality of life; and• optimize the flow of information between patient, pharmacist,

and physician.Standardized diabetes care services were established at the net-

work pharmacies to ensure the consistency of The Health Planmember benefits. Equipment such as the Bayer Diagnostics

DCA2000+ and the Cholestech LDX Analyzer were purchased byeach pharmacy to facilitate the collection of data at each site. Theservices consist of an initial assessment and education program, aswell as follow-up visits with the patients. These services are pro-vided through scheduled appointments that last between 30 and60 minutes. The initial assessment of the patients includes the fol-lowing:• collection of baseline data: weight, blood pressure, blood glucose,

lipid panel, HbA1c;• assessment of guideline adherence: primary care physician (PCP)

visit, eye/foot exams, immunizations;• review of medication regimen and patient adherence to regimen;

and• patient education (three sessions of one hour each).The documentation and recommendations are forwarded to thepatient’s primary care physician.

Follow-up visits are scheduled every three months and general-ly last less than 30 minutes. The pharmacists collect monitoringdata (weight, BP, blood glucose diary); perform a lipid panel andmicroalbumin testing on an annual basis; perform HbA1c testingevery three to six months based on HbA1c levels; assess guidelinecompliance (PCP visit, eye/foot exam, immunizations); re-evaluatemedication regimen and patient adherence; re-educate patient ifnecessary; and forward documentation to the PCP.

Additionally, the pharmacists agreed to work with the WVUfaculty to develop a continuous-quality-improvement process forthe care they provided. This entails monthly meetings to discusscases, as well as periodic reviews of the pharmacists’ documenta-tion by the WVU faculty. Feedback is provided by the faculty toboth the pharmacists and The Health Plan.

The Health Plan agreed to compensate the pharmacists for dia-betes services provided that the pharmacists:• were registered pharmacists; • completed an approved certification program in diabetes-related

pharmaceutical care from a national organization (APhA,NIPCO, or AADE Comprehensive Review Program);

• completed the WVU certificate program in Pharmaceutical Carefor Persons with Diabetes;

• possessed and were trained in the proper use of DCA2000+(HbA1c), Cholestech LDX or equivalent (lipid profiles), bloodpressure monitoring equipment, and glucose meters;

• obtained Clinical Laboratory Improvement Act waivers for allequipment requiring this waiver;

• demonstrated competence in using the OVPCN diabetes educa-tion materials;

• had a private area in which to meet with patients; and• maintained appropriate patient care records.

? ? Methods

Study PopulationAdults over 18 years of age with Type-2 diabetes were eligible for



inclusion. It was anticipated that the majority of participantswould be members of The Health Plan of the Upper Ohio Valleybecause this HMO is paying for the educational component ofthe pharmacists’ services. However, the assessments were avail-able to all patients with Type-2 diabetes who attended theOVPCN pharmacies. The pharmacists solicited the involvementof their current patients and also accepted referrals from physi-cians. Nearly all health plan patients who were offered the oppor-tunity to participate chose to enroll in the pharmacy-based pro-gram (54 out of 60 enrolled).

Clinical Data CollectionThe clinical data are collected and recorded by the pharmacists aspart of their standard care program during 1999 and 2000. TheHbA1c levels are determined by analyzing blood samples in theBayer DCA2000+ machine; these tests can be performed in thepharmacy by trained personnel. The HbA1c reagent cartridgerequires only one microliter of blood from a fingerstick and takesonly five minutes to obtain results.

The lipid profile is conducted through the use of theCholestech LDX analyzer using a fingerstick sample of blood.This test provides estimates of the total cholesterol, HDL, LDL,and triglyceride and glucose levels. All samples were drawnfrom patients in a fasting state.

The pharmacists also monitor the patients’ blood pressure ateach visit. Two blood pressure readings were taken with thepatient in the sitting position over the course of the assessmentvisit. The mean of the two blood pressures was used for analyses.

The body mass index (BMI) for each patient was estimatedusing the height and weight in the following formula: (poundsx 703)/(inches x inches). Thus, a 200-pound, 72-inch personwould have a BMI of (200 x 703)/(72 x 72) = 27.1. All meas-urements were obtained in the pharmacy.

? ? Results

Patient CharacteristicsThe pharmacists enrolled 54 patients in the clinical program, ofwhich 32 (59%) were female. The average age of the partici-pants was approximately 60 years (range: 35–81 years).

Clinical AssessmentUsable clinical data were obtained for 47 patients. The baselineassessment revealed that a considerable number of personswere not “at goal” for the clinical indicators (see Table 1, thispage). For HbA1c, 63.9% of persons had not reached the desiredgoal of HbA 1c lower than 7%, and 36.2% were above 8%.

Blood pressure was also elevated for about half of thepatients, with only 44.7% reaching the ADA recommended tar-get of 130/85 mmHg. Approximately 15% of patients could becategorized as having Stage 2 or 3 hypertension.

Lipid levels also were not ideal for many of the patients.

Almost seven out of ten (69.8%) patients had LDL levels abovethe recommended target of 100 mg/dl, and 57.4% of patientshad LDL levels greater than 130 mg/dl, putting them in thehigh-risk level for coronary heart disease.13

HDL levels were also elevated in a substantial number ofpatients. Over three-fourths (76.5%) of patients had HDL levelsbelow the target of 45 mg/dl, and 40.4% of patients had HDLlevels below 35 mg/dl, which places them in the high-risk levelfor coronary heart disease.13 Total cholesterol was elevated in38.3% of patients, while triglyceride levels were above 200 mg/dlin 57.4% of patients. Fourteen percent of patients had triglyc-eride levels above 400 mg/dl.

Approximately 85.7% of persons were above the ideal BMIof 25, and nearly 62% were obese (BMI over 30).

? ? Discussion

For persons with diabetes, glycemic control is an important

TABLE 1 Percentage of Patients Not AchievingTherapeutic Goals (N=47)

Indicator Percentage of Patients Not Achieving Goal

HbA1ca

7–8% 27.7%> 8% 36.2%Total above goal 63.9%

Blood Pressureb, c

Stage 1 hypertension 31.9%Stage 2 hypertension 8.5%Stage 3 hypertension 6.4%Total 46.8%

LipidsTotal cholesterol >200 38.3%LDL >100 69.8%HDL <45 76.5%Triglycerides >200 57.4%

Body Mass Index (BMI)d

25–30 23.8%>30 61.9%

aThe ADA recommends a target HbA1c of 7% for persons with diabetes,and that additional clinical action be taken for patients with an HbA1c over8%.bThe JNC-VI guidelines define hypertension using the following parame-ters: Stage 1=systolic >140 or diastolic >90; Stage 2=systolic >160 ordiastolic >100; Stage 3=systolic >180 or diastolic >110.cThe ADA recommends a target blood pressure of 130/85 mmHg; 56.3% ofpatients failed to meet this goal.dThe recommended BMI is 20–25; a BMI > 30 is considered obesity.



predictor of micro- and macrovascular complications.4, 14

Additionally, blood pressure and lipid levels are importantmarkers for cardiovascular mortality in persons with diabetes.13, 15

The ADA recommends frequent monitoring of HbA1c, bloodpressure, and lipids to help ensure that patients are maintainingcontrol of their condition. 5 Regular eye, foot, and renal screen-ings are also recommended.

In this study, a novel approach was used to identify personswith diabetes who were not reaching the recommended thera-peutic goals for HbA1c, lipids, and blood pressure, as well as toidentify those who might benefit from weight-management pro-grams. Community pharmacists collected information on theirpatients with diabetes, and sent reports to each patient’s physi-cian. The pharmacists were able to identify a substantial num-ber of persons who were not reaching the desired endpoints oftherapy (see Table 1), and recommend adjustments to the drugregimen when appropriate. The patients and many of the physi-cians were quite appreciative of the pharmacists’ efforts.

Achieving control of diabetes is challenging. Behavioralchanges in diet, exercise, and medication use are difficult tomaintain, and patients often experience decline in theirglycemic control over time. Consequently, close monitoring isessential, and a supportive pharmacist can be a great help inmaintaining control of the diabetes and other health indicators.Pharmaceutical care espouses a closer relationship amongpatient, pharmacist, and physician. It is essential that all threeparticipants in this relationship understand the goals for diseasecontrol, that they agree on a plan for monitoring, and that infor-mation flows smoothly among the participants to continuallyenhance the patient’s health outcomes.

Pharmacists are in an excellent position not only to provideinformation about drug therapy, but also to assess the patient’sprogress toward therapeutic goals. In this study, pharmacistsidentified 36.2% of their patients as having an HbA1c above 8%.The ADA suggests that an HbA1c above 8% should prompt addi-tional action by providers to enhance the patient’s glycemic con-trol.5 The pharmacists notified the physicians that these patientswere not “at goal” and suggested specific therapeutic optionswhen requested. Although the long-term impact of these rec-ommendations could not be assessed at this time, identifyingthe patients in need of additional help prevented more thanone-third of the patients from “slipping through the cracks” andhaving their uncontrolled diabetes go undetected.

Many patients with Type-2 diabetes also have hypertensionand/or dyslipidemia and often suffer complications such asheart attack or stroke.3 Thus, it is important to monitor bloodpressure and lipid levels in these patients. The pharmacists inthis network identified over half (55.3%) of the patients as notmeeting the ADA recommend blood pressure goal of 130/85mmHg. Additionally, about 15% of the patients had blood pres-sure readings that were consistent with Stage 2 or Stage 3hypertension. Thus, at least 15% and perhaps as many as 55%

of the patients could benefit from additional intervention. The pharmacists also found that 57.4% of patients were within

the high-risk category for coronary heart disease as predicted byLDL over 130 mg/dl. The LDL level is an important indicator of riskfor cardiovascular mortality. 13 Many patients (76.5%) had HDL lev-els below the recommended target, and 57.4% had elevated triglyc-erides. This seems consistent with the finding that 85.7% of patientswere above their ideal body weight and about 62% of patients wereconsidered obese. Clearly, many of the patients enrolled in the phar-macy-based program were in need of help in reducing their risk ofmacrovascular complications.

Pharmacists’ involvement with diabetes care is not a newconcept. An increasing number of certified diabetes educatorsare pharmacists, and several studies have demonstrated theimpact of a pharmacist’s care on diabetes outcomes.9–12

However, very few community pharmacists have conductedassessments of their patients with diabetes in a manner as com-prehensive as the program described here. In addition to exam-ining blood glucose meter readings, weight, and blood pressure,the pharmacists in this network obtained equipment to collectHbA 1c and lipid levels within the pharmacy. Performing thesetests while the patient was in the pharmacy allowed the phar-macist to give the patient immediate feedback on their diseasecontrol, and facilitated more timely modifications of drug ther-apy. Rather than the pharmacist waiting for the physician toorder the test, and then hoping that the patient would go to thelaboratory and that the physician would share the data andmake appropriate changes in drug therapy, the pharmacist canmore proactively identify problem areas and make informed,specific recommendations for drug therapy enhancement.

Having the pharmacist collect and report this informationcan be in the best interest of physicians and health plans. Ifphysicians can rely on the pharmacist to coordinate the educa-tion and monitoring functions for diabetes care, then the physi-cians may be able to save time and be more efficient in provid-ing care to their patients with diabetes. The pharmacist can per-form the key monitoring tests recommended by the ADA andprovide reports directly to the physician, along with recom-mendations for drug therapy modification. Additionally, thepharmacist can ensure that the patients are seeing their physi-cian on a regular basis and can promote positive health behav-iors (e.g., regular eye and dental exams, flu shots, smoking ces-sation). By enhancing the frequency of regular exams and mon-itoring tests, a pharmacist can assist health plans in improvingtheir diabetes indicator scores for HEDIS. Although an individ-ual pharmacy may have little impact on the overall HEDIS scoreof a health plan, a network of pharmacies may be able to pro-duce a measurable difference.

As a result of this project, The Health Plan is working withOVPCN to develop a multi-disciplinary diabetes education pro-gram that will be integrated with the monitoring component of thepharmacists’ services. In this model, the pharmacists help not only



to monitor the patient’s clinical progress, but also to ensure thatthe patient is attending the education classes and adhering to diet,exercise, and medication recommendations.

The costs of implementing the diabetes care program withinthe OVPCN were substantial. Pharmacists spent $5,000–$7,000at each store to acquire the equipment, supplies, and training nec-essary to provide this service. Although this is not an insurmount-able barrier to implementation, the compensation levels to thepharmacist by the health plan should be adjusted to offset thesecosts within an acceptable time frame. The pharmacists were ableto offset some of these costs through small grants from pharma-ceutical manufacturers and foundations; however, contracts withthird-party payors are currently the primary source of revenue.

Paying pharmacists to identify patients with suboptimalglycemic control or undetected hypertension or dyslipidemia maybe quite cost-effective for managed care plans. A recent studydemonstrated that patients with poor glycemic control (HbA1c

over 10%) were hospitalized with complications of diabetes at afar greater rate than those with fair (HbA1c 8%–10%) or good(HbA1c lower than 8%) glycemic control.16 The average inpatientcharges over three years for a patient with good control was $970,versus $3,040 per person for patients with poor control. Anotherrecent study found that improving glycemic control in managedcare patients can reduce average total expenditures by $685–$950per year within the first four years of improvement.17 These costreductions are achieved not only through preventing the compli-cations of diabetes, but also through the immediate impact onpatients’ functional ability. If a pharmacist were paid only $250 perpatient, per year, to identify persons with diabetes who were fail-ing to meet their treatment goals, helping just a few patients toachieve better control of their diabetes would offset the costs of thepharmacist-care program. Consequently, it appears that commu-nity pharmacists can produce savings to third-party payors andpatients that clearly exceed the typical pharmacist compensation.

? ? Limitations

The percentage of persons who did not reach the therapeuticgoal was based upon the total number of those who agreed toparticipate in the pharmacist-based program. It is not clearwhether the persons enrolled in this monitoring program weredifferent from the general population of persons with diabetes.Because very few persons declined participation in the program,the study population is believed to be fairly consistent with thegeneral population of managed care enrollees with diabetes.

? ? Conclusions

Community pharmacists can play an important role in diabetescare by identifying patients who are not achieving their thera-peutic goals, and by working with physicians to make drugtherapy modifications. Through identifying patients not “atgoal,” the pharmacists have the opportunity to prevent thedevelopment of diabetes-related complications and reduce total

health care expenditures. Implementing a diabetes-monitoringprogram may require a significant investment by a pharmacist;however, this service should be of great value to patients, physi-cians, and third-party payors.

References

1. Rubin RJ, Altman WM, Mendelson DN. Health care expenditures for peo-ple with diabetes mellitus, 1992. J Clin Endocrin Metabol 1994; 78:809A–09F.

2. American Diabetes Association. Position Statement: Diabetic Nephropathy.Diabetes Care 1997; 20: S24–27.

3. O’Brien JA et al. Direct medical costs of complications resulting from Type2 Diabetes in the U.S. Diabetes Care 1998; 21: 1122–28.

4. The Diabetes Control and Complications Trial Research Group. The effectof intensive treatment of diabetes on the development and progression oflong-term complications in insulin-dependent diabetes mellitus. N Engl J Med1993; 329: 977–86.

5. American Diabetes Association standards of medical care for persons withdiabetes mellitus. Diabetes Care 2000; 23 (Supplement 1); S32–42.

6. Weiner JP et al. Variation in office-based quality. JAMA 1995; 273:1503–08.

7. Peters AL et al. Quality of outpatient care provided to diabetic patients.Diabetes Care 1996; 19: 601–06.

8. Burton TM. United HealthCare finds drugs, test are often underutilized.Wall Street Journal (Interactive Edition) July 8, 1998.

9. Coast-Senior EA et al. Management of patients with Type 2 diabetes bypharmacists in primary care clinics. Annals Pharmacother 1998; 32: 636–41.

10. Fincham JE, Lofholm PW. Saving money and lives–Pharmacist care fordiabetes patients. America’s Pharmacist 1998; (March): 49–52.

11. The Asheville Project. Pharmacy Times 1998 (October supplement).

12. Berringer R et al. Outcomes of a Community Pharmacy-Based DiabetesMonitoring Program. J Am Pharm Assoc 1999; 39: 791–97.

13. Summary of the second report of the National Cholesterol EducationProgram (NCEP) Expert Panel on Detection, Evaluation and Treatment ofHigh Blood Cholesterol in Adults. JAMA 1993; 269: 3015–23.

14. United Kingdom Prospective Diabetes Study Group. Intensive blood glu-cose control with sulphonylureas or insulin compared with conventional treat-ment and risk of complications in patients with type 2 diabetes. Lancet 1998;352: 837–53.

15. United Kingdom Prospective Diabetes Study Group. Tight blood pressurecontrol and risk of macrovascular and microvascular complications in type 2diabetes. BMJ 1998; 317: 703–13.

16. Menzin J et al. Potential short-term economic benefits of improvedglycemic control. Diabetes Care 2001; 24: 51–55.

17. Wagner EH et al. Effect of improved glycemic control on health care costsand utilization. JAMA 2001; 285: 182–89.

OBJECTIVE: Though prior authorization(PA) programs are widely used in themanaged care pharmacy environment,some stakeholders question whetherthese programs are effective. Theobjective of this article is to criticallyexamine the effect of PA programs onhealth-related outcomes.

DATA SOURCES: A computer-aidedsearch of the literature was conductedusing several online databases to findstudies that have evaluated PA pro-grams. Other sources of informationused were reference lists, authors ofprevious studies, and meetingabstracts.

STUDY SELECTION: In order for a studyto be included in our analysis, it had to(1) appear in the peer-reviewed litera-ture and (2) investigate the effects of aPA program on specified drugs. Weexcluded papers that studied the effec-tiveness of formulary systems, ofwhich PA may be a component.

DATA EXTRACTION: From each study

evaluated, we extracted data related tothe study design and to the effect ofthe PA program on economic, clinical,and humanistic outcomes.

DATA SYNTHESIS: Six studies met ourcriteria. Overall, PA programs appearto be effective at reducing drug-relatedcosts. There is some evidence thatthey reduce non–drug-related costsbut little evidence that they have apositive impact on clinical or humanis-tic outcomes. None of the studies hada randomized, controlled design; mostof the studies had severe methodologi-cal limitations.

CONCLUSION: Rigorously designedstudies are urgently needed in order toevaluate the effects of PA on health-related outcomes.

KEYWORDS: Prior authorization, specialauthorization, drug cost containment,pharmacy benefit management, pre-scribing restrictions



REVIEW

NEIL J. MacKINNON, Ph.D., R.Ph., is Assistant Professor and the MerckFrosst Chair of Patient Health Management, College of Pharmacy, andAssistant Professor, Department of Community Health and Epidemiology,Faculty of Medicine, and RITU KUMAR, M.H.S.A., R.N., is Research Fellow inHealth Outcomes Management, Dalhousie University College of Pharmacy,Halifax, Nova Scotia, Canada.

AUTHOR CORRESPONDENCE: Neil MacKinnon, Ph.D., R.Ph., DalhousieUniversity College of Pharmacy, 5968 College St., Halifax, Nova Scotia,Canada B3H 3J5; Tel: 902-494-6379; Fax: 902-494-1396; E-mail: [email protected].

ACKOWLEDGEMENTS: The authors would like to acknowledge the contribu-tions of Sheri Axworthy, Sharon Boriel, Derek Chaudhary, and Ingrid Sketris.Funding was provided by Merck Frosst Canada and Company.


Authors

n modern health care it is unethical not to be concernedwith evaluation, and no longer acceptable to be ‘evaluationilliterate.’”1

Managed care organizations (MCOs) use many pharmacybenefit management tools and techniques to help guide appro-priate medication usage. Popular methods are formularies, pre-scription drug caps, patient copayments, maximums, manda-tory use of generics, drug-utilization review, and therapeuticinterchange. The challenge faced by many MCOs is to deal withrising drug costs while not denying or limiting access to thosedrugs that improve therapeutic outcomes and health-relatedquality of life (HRQoL). The challenge is likely to become moredifficult; prescription drug expenditures in the United Statesare projected to increase 11.0% in 2001 and 10.7% in 2002.2

One pharmacy benefit management technique being usedwith increasing frequency is prior, or special, authorization(PA); a recent survey revealed that in 1996 43%, in 1997 54%,and in 1998 61% of employers reported using PA programs.3

PA is an administrative tool that requires the prescriber to getpre-approval for prescribing a drug in order to qualify for reim-bursement.4 The broad purpose of PA is to change prescribingbehavior.5 The goal of the PA process is to encourage appropri-ate use of medications, both to reduce the incidence of pre-ventable drug-related morbidity and to contain costs. The phi-losophy behind this mechanism, which intuitively seems tohelp promote the delivery of quality health care, is to targetnew, costly, or potentially toxic medications, and to encourageuse of less-expensive, safer alternatives. Some view this tech-nique as simply a means to contain costs rather than a quality-improvement or risk-management tool.

Not all agree with the use of PA to direct prescribing.Understandably, the pharmaceutical industry views PA pro-grams as a barrier to market access. Similarly, patients often feelthat PA programs impede their access to drugs that they per-ceive as necessary. Many physicians are exasperated by the timededicated to PA paperwork. Some physicians in NewBrunswick, Canada, reported filling out up to 10 PA forms aday; in September 2000, physicians in the province refused tofill out any PA forms because of the time burden.6, 7

Some physicians feel that PA programs actually preventtheir patients from getting the medications they need in a time-ly manner. A survey in Ontario, Canada, revealed that only34% (17 of 50) of the physicians who responded felt that a lim-ited-use (PA) listing for medications helped them to more


Iby Neil J. MacKinnon and Ritu Kumar

“



appropriately prescribe drugs.8 Sixty-six percent (29 of 44) saidthat in the past they have chosen not to prescribe a product pri-marily because of its limited-use status, even though the physi-cian felt the patient might benefit from the drug. 8

Community pharmacists also complain about the adminis-trative burden of PA programs; a recent study showed that, onaverage, a supermarket chain pharmacy spent 2.15 minutes andan independent pharmacy spent 2.97 minutes just on rejectionresolution for each prescription that required PA. 9 Last year, theNorth Carolina Board of Pharmacy passed special regulations tohelp reduce community pharmacists’ burdens of dealing withthird-party payor administrative policies such as PA.10

Perhaps the greatest controversy over the use of PA is theunintended effect of other prescribing restrictions such asrestrictive formularies and benefit caps. One of the first studiesto document unintended effects was a 1985 study of a closedformulary for drugs used in treating peptic ulcer disease for theWest Virginia Medicaid program. After a formulary policychange, outpatient drug expenditures were reduced by 78.9%,but monthly physician payments increased 3.1% and monthlyinpatient hospital costs increased 23.6%. 11 A 1991 study foundthat drug use decreased but nursing home admissions increasedafter a three-prescription limit per patient per month wasimplemented in the New Hampshire Medicaid program.12

A controversial 1996 study by Horn and colleagues furtheradded to the literature on the unintended effects of prescribingrestrictions by concluding that health maintenance organiza-tions (HMOs) with more-restrictive drug formularies had high-er overall utilization and costs of health care resources.13 Therehas been considerable debate over the methodology of the Hornstudy in particular and over prescribing restrictions in general,including several editorials in this journal.14

Since PA programs are common in the managed care phar-macy environment, and because of the questions about theseprograms stimulated by previous studies, we considered iturgent to examine the effectiveness of PA programs. The objec-tive of this article is to review the peer-reviewed literature on PAprograms and to assess their effects on economic, clinical, andhumanistic outcomes of health care.

? ? Methods

Data SourcesA computer-aided search of the medical and pharmacy litera-ture in English was conducted in spring 2001, using Medline,International Pharmaceutical Abstracts (IPA), Health Star, andEcolit. Keywords such as prior authorization, prescribingrestrictions, prior approval, special authorization, cost contain-ment, exception drug status, and restrictive formularies wereused in the search. Other studies on PA were found in managedcare textbooks, references, and reading materials previously col-lected by the lead author. We attempted to contact authors of

published studies on PA and researchers on PA in search of studiesthat were not identified by our computer-aided search. Wereviewed abstracts from recent educational conferences and annu-al meetings of the Academy of Managed Care Pharmacy (AMCP)and annual meetings of the Canadian Association of PopulationTherapeutics (CAPT), but a study had to be published in completeform in order to be included in our final analysis.

Study SelectionIn order for a study to be included in our analysis, it had to (1)appear in the peer-reviewed literature, and (2) investigate theeffects of a PA program on specified drugs. We excluded papersthat studied the effectiveness of formulary systems, of whichprior authorization may be a component, as it would be impos-sible to distinguish the effect of the PA program from the effectof the formulary itself.

? ? Data Extraction

From each study evaluated, we extracted data related to the studydesign and the effect of the PA programs on health-related out-comes. In the critique of each study, each author of this article inde-pendently used a standardized data collection form based on theECHO (economic, clinical, and humanistic outcomes) model pro-posed by Kozma, Reeder, and Schulz as our framework for evalua-tion.15 More specifically, for all studies we critically evaluated themethodology, study sample, outcomes measures, drugs studied,and economic (both drug costs and other health expenditures),clinical (both drug-related and non–drug-related), and humanis-tic outcomes (satisfaction and HRQoL).

? ? Results

Six studies met our criteria for review. 5, 16–20 The study design,study sample, outcomes measures, and drugs in the PA pro-grams are contained in Table 1, page 299.

Because no study had a randomized, controlled experimen-tal design, all studies had significant threats to validity. Thestudy by Smalley and colleagues had the most rigorous experi-mental design.17 Four of the six studies had no control group.5,

18–20 One study did not use a baseline measurement periodbefore the PA program was set up, and only one study had a fol-low-up period of more than one year to measure the long-termeffects of the PA program.18, 17

Four of the six studies used a state Medicaid program for thestudy sample.5, 16–18 The other studies used an urban teachinghospital and secondary data from a national survey. 19, 20 None ofthe studies was multi-center. The intended unit of analysis wasoften hard to determine; indeed, three of the studies did notspecify the exact number of patients considered.5, 18, 19

Outcome measures also varied considerably. One studymeasured simply the cost and utilization of the PA drugs.18 Onlyone study included clinical outcome measures.19 Four of thestudies looked at a single drug class (nonsteroidal anti-inflam-

Continued on page 301


TABLE 1 Summary of Reviewed Studies on Prior Authorization Programs

Kotzan, McMillan, Jankel,and Foster, 199316

Time-series analysis thatincluded data up to one yearbefore policy change and upto seven months after

80,064 continuously eligiblerecipients in the State ofGeorgia Medicaid program

Cost and utilization ofNSAIDs, other anti-arthriticagents, non-narcotic anal-gesics, physician claims, andother medical services (exactkinds not specified)

NSAIDs (except those avail-able in generic form)

Smalley, Griffin, Fought et al., 199517

Baseline year before policychange and the two-yearperiod after policy change;interrupted time-seriesanalysis with a controlanalysis

Two separate study groups:(1) enrollees at any timeduring the three-year periodin the Tennessee Medicaidprogram (495,821 in base-line year to 547,403 in year3), and (2) enrollees withuninterrupted enrollment forall three years who were reg-ular users of NSAIDs (3,174regular users of nongenericsand 1,849 regular users ofgenerics)

Cost and utilization ofNSAIDs, other analgesic orantiinflammatory drugs, psy-chotropic drugs, outpatientservices, and inpatientadmissions for managementof pain or inflammation

NSAIDs (except those avail-able in generic form)

Kotzan, Perri, and Martin,199618

Cross-sectional, equivalentcontrol group; comparisonof two groups for a three-month period

All prescription claimsprocessed for the State ofGeorgia Medicaid programfor Jan–Mar 1994(2,957,850 prescriptions,including 71,187 for PAdrugs), and cash prescrip-tions (6,347,617 total,including 357,546 for PAdrugs) from approximately1,100 Georgia pharmacies

Cost and utilization of 46drugs; market-share analysis

46 drugs that were part ofthe Georgia Medicaid pro-gram and also represented inprivate-payment markets

Phillips and Larson, 19975 Baseline (12–24 months,depending on the drug stud-ied) before policy changeand the 12-month periodafter the policy change; nocontrol group. Operationalperformance based on twoweeks of data

Iowa Medicaid enrollees forwhom a prescription requir-ing PA was filled during thestudy period (approximately250,000 enrollees; no num-ber for patients for whom aPA prescription was filled)

Cost and utilization of 16drugs; also administrativeoutcomes, such as approvalrates and program responsetimes

16 categories of individualmedications

White, Atmar, Wilsonet al., 199719

Baseline (July–Dec) six-month period before policychange and a six-monthperiod (July–Dec) sixmonths after the policychange; no control group

Patients in a 575-bed urbanteaching hospital during thestudy period who receivedan antimicrobial agent (exactnumber of patients or pre-scriptions filled is not pro-vided, though the total num-ber of patient-days permonth in the hospitaldecreased from 14,694 to13,738)

Cost of parenteral anti-microbials, antimicrobialsusceptibility patterns, gram-negative bacteremia survivalrates, time from initial bloodculture to receipt of anappropriate antibiotic, inpa-tient and ICU length-of-stay

Six intravenous antibioticsinitially, plus two otherantibiotics added over thenext six months

Feldman, Fleischer, andChen, 199920

Retrospective cross-sectionalstudy of data from theNational AmbulatoryMedical Care Survey; sensi-tivity analyses were per-formed to determinewhether a PA age of 25 iscost-effective

A cost and utilization modelwas created from previouslypublished data (the NationalAmbulatory Medical CareSurvey), normalized to100,000 covered lives

Cost and utilization of topi-cal tretinoin; costs of admin-istering a PA program to aninsurer were also considered

Topical tretinoin

Article Study Design Study Sample Outcome Measures Drugs Studied

Notes: NSAID is nonsteroidal anti-inflammatory drug. PA is prior authorization. ICU is intensive-care unit.




TABLE 2 Outcomes Measured in Prior Authorization Studies

Article Economic Outcomes:Drug Costs

Economic Outcomes:Non-drug Costs

ClinicalOutcomes:

Drug-Related

ClinicalOutcomes:Non-drug-

Related

HumanisticOutcomes:

Satisfaction

HumanisticOutcomes:

HRQoL

Kotzan, McMillan, Jankel,and Foster, 199316

NSAIDs: monthly reduction incosts of about 54% from base-line to policy implementationperiod ($3,018,308 estimatedsavings over seven months);monthly non-narcotic analgesic use increased about37% ($193,540 over firstseven months); no other significant changes

No significant changes inphysician claims or othercategories of medical services (exact categories notspecified); administrativecosts of the program werenot measured

Not measured

Not measured

Not measured

Not measured

Smalley, Griffin, Foughtet al., 1995 17

All enrollees: Annualexpenditures of NSAIDsdecreased by 53% duringthe two years after the startof the policy ($12,800,000estimated savings over twoyears); no other significantchanges; regular nongener-ic NSAID users: a relativedecrease in expenditures of64% compared to genericNSAID users; no other sig-nificant changes

All enrollees: no significantchangesRegular nongeneric NSAIDusers: no significantchanges where sample sizepermitted analysisAdministrative costs to theMedicaid program:$75,000 for one year

Not measured

Not measured

Not measured

Not measured

Kotzan, Perri, and Martin,1996 18

Total estimated drug costssavings attributable to theGeorgia Medicaid PA pro-gram for all 46 drugs:$8–$20 million annually

Effects on non-drug costswere not measuredAdministrative costs to theMedicaid program: About$1 million

Not measured

Not measured

Not measured

Not measured

Phillips and Larson, 1997 5 Net savings (gross drug sav-ings minus administrativecosts) for four drug cate-gories (anti-arthritics, benzo-diazepines, antihistamines,and antiulcer drugs) estimat-ed to be $2.51 million–$3.83million

Effects on non-drug costswere not measuredAdministrative costs to theMedicaid program for fourcategories of drugs totaled$162,000

Not measured

Not measured

Satisfactionnot directlymeasured,thoughresponsetimes andapproval rateswere

Not measured

White, Atmar, Wilson et al.,1997 19

Expenditures for all parenter-al antimicrobials decreasedby $431,548 (32%) for thesix-month period in 1994 ascompared to 1993 (thisincluded an increase inexpenditures for someantimicrobials not includedin the PA program)

No significant change in inpatient or ICU length ofstay Administrative costs tothe hospital: less than$150,000 per year (estimated)

Increased susceptibilityto isolates inICUs andinpatient unitsbut not outpa-tient sites;time to receiptof appropriateantibioticsunchanged

No significantchange in sur-vival rates inpatients withgram-negativebacteremia

Satisfactionnot directlymeasured,thoughresponse timesand approvalrates were

Not measured

Feldman, Fleischer,and Chen, 1999 20

Assuming a topical tretinoin unit cost per prescription of $28and a unit expense of $10 for performing a single PA, the totalcost per 100,000 covered lives is estimated to be $23,226 for aPA age of 25 and $22,685 for a PA age of 35. The tretinoin costwith no PA program is $23,800. Effects on non-drug costs werenot measured.

Not measured

Not measured

Not measured

Not measured

Notes: NSAID is nonsteroidal anti-inflammatory drug. PA is prior authorization. ICU is intensive-care unit. HRQoL is health-related quality of life.


matory drugs [NSAIDs] in two, intravenous antibiotics in one,topical tretinoin in one), while the two others evaluated severaldrug classes.

Table 2, page 300, contains the economic, clinical, andhumanistic outcomes that were measured in these studies. Wedistinguished between drug and non-drug outcomes in boththe economic and clinical outcome categories and betweenpatient satisfaction and HRQoL humanistic outcomes.

All six studies documented drug cost savings from the PAprograms. We had initially hoped to conduct a meta-analysis tobetter summarize the amount of drug cost savings from PA pro-grams, but inconsistencies in the descriptions of the study sam-ples and outcomes made it impossible to calculate an effect size.It is interesting, though, that the two studies that focused on aPA program for NSAIDs found similar drug cost savings(approximately 54% in Kotzan et al. and 53% in Smalley etal.).16, 17 Some studies failed to distinguish between cost savingsresulting from lower overall drug product cost (switch to gener-ic or less expensive drug) or from lower drug utilization.

Of the three studies that measured the effect of the PA pro-gram on non-drug costs, none found a significant increase incosts elsewhere in the health care system.16, 17, 19 Five of the sixstudies calculated the administrative costs of operating the PAprogram, although they did not provide thorough descriptionsof how these costs were measured, what they included, andcosts to stakeholders outside of the direct organization of inter-est (e.g., community pharmacists, physicians, etc.). 5, 17–19 Onestudy concluded that the administrative costs of the PA programoutweighed the reduction in drug costs for a majority of agegroups considered. 20

Only the study by White and colleagues measured how thePA program affected clinical outcomes.19 None of the studiesmeasured how PA programs affected satisfaction (patient, phar-macist, physician, nurse, or other) or HRQoL, two primaryhumanistic outcomes.

? ? Discussion

Our critical analysis of the literature indicates that although PAprograms are common, their outcomes have not been ade-quately evaluated. PA is not alone, however; evaluation ofadministrative policies and programs in health care and in phar-macy benefit management today is rarely adequate.12, 21 Still, thescarcity of quality evaluations of the outcomes of PA programsshould be of concern to patients, health care professionals,administrators, and others who work in managed care pharma-cy since these programs are widely used. Little has changedsince 1993, when Kotzan, McMillan, Jankel, and Foster lament-ed: “The long-term impact of PA programs has not been docu-mented. If the drug programs are devised solely on the basis ofeconomic consideration without regard for medical conse-quences, then it is likely that more expensive services willreplace those expensive drugs removed from the formulary.”16

Why is there a lack of rigorous evaluations of PA programs?Ray has reflected on the general problem of inadequate healthpolicy evaluations and concluded that a primary barrier is pol-itics; conducting a randomized, controlled trial is an admissionof uncertainty. 21 The persons or organizations involved in PAprograms may have a vested interest in the success of their pro-grams. Moreover, expensive randomized controlled trials maynot be practical for many organizations, although repeatedtime-series analyses, such as the one conducted by Smalley andcolleagues, may be possible.

One can understand the reluctance to measure humanisticoutcomes of PA programs, such as satisfaction, given that PA isan administrative policy. Still, measuring what happens topatients who are denied a PA request would be valuable.Fortunately, some MCOs are now trying to improve physicianand patient satisfaction with PA programs, some by automatingthe PA process to eliminate paperwork or pharmacist interven-tion.22, 23 Finally, another barrier to quality evaluations is thatsome organizations may have difficulty in separating the out-comes of a PA program from those of the total formulary-management system.

Why is there a need for more PA program evaluations thatmeasure all three types of health-related outcomes? The princi-pal reason is to determine how PA programs affect clinical andhumanistic outcomes. Proof that PA programs improve patientoutcomes would more strongly support their use. If they affectpatient outcomes negatively, all stakeholders should reassesstheir use. Failure to measure the clinical outcomes of PA pro-grams is of special concern: Our literature search found no pub-lished studies, and just one presentation abstract, that measuredclinical outcomes outside the hospital. 24

Secondly, evaluation of programs and policies is a key partof a continuous quality improvement (CQI) philosophy, wherebenchmarks are determined and an attempt made to improveperformance to exceed those benchmarks.25–27 As Phillips andLarson acknowledge, currently there are not even PA programbenchmarks for such basic outcomes as processing times,approval rates, and administrative costs.5 Standard principlesfor PA programs could be helpful, perhaps like those recentlydeveloped for drug-formulary systems by AMCP and otherorganizations.28 Setting, and reporting on, standards should leadto increased accountability and transparency for PA programs.The accountability that must clearly become a priority for eachstakeholder involved in putting such programs in place shouldinclude continual monitoring to determine if the program’smandate is being achieved.

The burden of proof whether PA programs improve patientoutcomes should be on those who have programs in place, evenif this is a difficult process. As Hepler says, “It may be painful tobe objective about our own sacred cows.”29 Program evaluationis especially urgent given that many policies that regulate accessto and utilization of pharmaceuticals can have unintended neg-



ative outcomes.12 PA programs that direct prescribers to followevidence-based clinical practice should, in theory, lead to posi-tive clinical and HRQoL outcomes. Yet, as at least two of thestudies we reviewed acknowledged, because these outcomeswere not measured, we cannot be certain whether PA programshave a positive or a negative effect on these outcomes.5, 18 Giventhese important but still unanswered questions, now wouldappear to be an opportune time for evaluation of all policies thatrestrict prescribing, including PA.30

? ? Limitations

In any analysis of critical literature, some may differ with theinclusion/exclusion criteria or identify studies that have beenomitted. We tried to minimize these problems by making ourinitial search as broad as possible through the use of multiple lit-erature-retrieval methods and by making our criteria fairly con-servative. As with any literature review, we are limited by inher-ent publication biases to publish only statistically significantresults. Finally, we did intend to conduct a meta-analysis, butthis proved impossible given the inconsistency in the descrip-tion of the study samples and outcomes.

? ? Conclusion

From a critical review of the literature, PA programs appear toreduce drug-related costs. There is some evidence that they mayalso reduce non-drug-related costs, but little evidence that theyimprove clinical or humanistic outcomes. Most existing studieshave severe methodological limitations. There has been not onerandomized controlled study to better establish the relationshipof PA programs to these health-related outcomes. Resources forthorough program evaluations may be scarce, but an unin-formed acceptance of PA programs without consideration oftheir effects on health outcomes may be suboptimal at best, anddangerous at worst.

References

1. Øvretveit J. Evaluating health interventions: An introduction to evaluationof health treatments, services, policies, and organizational interventions.Buckingham, UK: Open University Press, 1998.

2. Mehl B, Santell JP. Projecting future drug expenditures–2000. Am J HealthSyst Pharm 2000; 57: 129–38.

3. The Wyeth-Ayerst prescription drug benefit cost and plan design surveyreport, 1999 ed. Philadelphia, PA: Wyeth-Ayerst Laboratories, 1999, 11.

4. Soumerai SB et al. Effects of Medicaid drug-payment limits on admission tohospitals and nursing homes. New Engl J Med 1991; 325: 1072–77.

5. Phillips CR, Larson LN. Evaluating the operational performance and finan-cial effects of a drug prior authorization program. J Managed Care Pharm1997; 3(6): 699–706.

6. Robb N. Some suffer adjustment pains as Blue Cross changes drug-benefitprogram on East Coast. Can Med Assoc J 1995; 153(3): 339–41.

7. Health Edition Newsletter. Merck Frosst Canada Inc., September 22, 2000;4(37): 1.

8. PSL Group for Glaxo Wellcome Canada, Inc. (2000). Study A6548, impact

of limited use policy on prescribing of asthma therapies in Ontario. Montreal,PQ, Canada: PSL Group, 2000.

9. Herrier RN, Spencer JR, Davis CD. Case study using descriptive analysis toestimate hidden costs in processing third party prescriptions. J Am PharmAssoc 2000; 40(5): 658–65.

10. English T. Reactions mixed on North Carolina rule that seeks to alleviatethird party hassles. Pharmacy Today Sept 2000; 1, 15.

11. Bloom BS, Jacobs J. Cost effects of restricting cost-effective therapy. MedCare 1985; 23 (7): 872–80.

12. Soumerai SB et al. A critical analysis of studies of state drug reimburse-ment policies: Research in need of discipline. Milbank Quarterly 1993; 71(2):217–50.

13. Horn SD et al. Intended and unintended consequences of HMO cost-con-tainment strategies: Results from the Managed Care Outcomes Project. Am JManaged Care 1996; 2(3): 253–64.

14. Jones J, Cronin JM. The pros and cons of formularies. J Managed CarePharm 2000; 6(3): 203–07.

15. Kozma CM, Reeder CE, Schulz RS. Economic, clinical, and humanisticoutcomes: A planning model for pharmacoeconomic research. Clin Ther 1993;15: 1121–32.

16. Kotzan JA et al. Initial impact of a Medicaid prior authorization programfor NSAID prescriptions. J Res Pharm Econ 1993; 5(1): 25–41.

17. Smalley WE et al. Effect of a prior-authorization requirement on the use ofnonsteroidal antiinflammatory drugs by Medicaid patients. New Engl J Med1995; 332: 1612–17.

18. Kotzan JA, Perri M, Martin BC. Assessment of Medicaid prior-approvalpolicies on prescription expenditures: Market-share analysis of Medicaid andcash prescriptions. J Managed Care Pharm 1996; 2(6): 651–56.

19. White AC et al. Effects of requiring prior authorization for selected antimi-crobials: Expenditures, susceptibilities, and clinical outcomes. ClinicalInfectious Diseases 1997; 25: 230–39.

20. Feldman SR, Fleischer AB, Chen GJ. Is prior authorization of topicaltretinoin for acne cost effective? Am J Managed Care 1999; 5: 457–63.

21. Ray WA. Policy and program analysis using administrative databases. AnnIntern Med 1997; 127: 712–18.

22. Kielty M. Improving the prior-authorization process to the satisfaction ofcustomers. Am J Health Syst Pharm 1999; 56: 1499–1501.

23. Chiefari DM, Hopsicker JR. Using automated point-of-service pharmacyedits to improve criteria-based prior authorization processes. Presentationabstract. Academy of Managed Care Pharmacy, 12th Annual Meeting &Showcase, Phoenix, AZ, April 5–8, 2000.

24. Sketris IS et al. Proton pump inhibitor use under the Nova ScotiaPharmacare’s Exception Drug Program (Canadian Association of PopulationTherapeutics presentation abstract). Can J Clin Pharmacol 1999; 6(2): 50.

25. Berwick DM. Continuous improvement as an ideal in health care. NewEngl J Med 1989; 320(1): 53–56.

26. Christensen DB, Penna PM. Quality assessment and quality assurance ofpharmacy services. J Managed Care Pharm 1995; 1(1): 40–51.

27. Donabedian A. The quality of care: How can it be assessed? JAMA 1988;260(12): 1743–48.

28. Sullivan S, Lyles A, Luce B, Grigar J. AMCP Guidance for Submission ofClinical and Economic Evaluation Data to Support Formulary Listing in U.S.Health Plans and Pharmacy Benefits Management Organizations. J ManagedCare Pharm 7(4): 272–82.

29. Hepler CD. Where is the evidence for formulary effectiveness? Am JHealth Syst Pharm 1997; 54: 95.

30. MacKinnon NJ. When it comes to formularies, where is our focus? Can JHosp Pharm 1999 June: 52(3): 143–44.


OBJECTIVE: Pharmacy and medicalclaims data have often been used as asource of data on the asthma popula-tion in managed care settings.However, there are very few data onpatterns of drug utilization surroundingan acute event. This is a report of anobservational pilot study that evaluatedutilization patterns in asthma treat-ment before and after an acute event.

DESIGN: The study was performed byevaluating pharmacy and medicalclaims data submitted betweenJanuary 1, 1994, and September 30,1995, in a 275,000-member preferredprovider organization (PPO) in themid-Atlantic region; 83 patients metthe inclusion and exclusion criteria.

RESULTS: Evaluating the data usingthe McNemar test revealed a statisti-cally significant greater number ofpatients using short-acting beta2-ago-nists after an acute event who did notuse them before the event comparedto the number of patients whochanged in the reverse (i.e., patients

who used short-acting beta2-agonistsbefore an acute event but not after)(p<0.05). The number of patients whoused inhaled anti-inflammatory med-ications, however, did not change sig-nificantly. In addition, the proportionof patients using the short-actingbeta2-agonists or anti-inflammatoryagents was highest immediately afterthe acute event.

CONCLUSION: Our findings suggestthat an acute event only partiallyaffects adherence to recommendedtherapy. Efforts must be made toincrease and maintain the use ofimportant long-term anti-inflammatoryasthma medications. This analysis mayhelp lay the foundation for larger stud-ies focusing on utilization patterns andcompliance around an acute event fora variety of disease states.

KEYWORDS: Acute event, adherence,asthma, claims data, compliance, uti-lization



RESEARCH

JUNG H. LEE, Pharm.D., is a Medical Information Scientist, AstraZeneca,New York, NY (at the time this article was written Dr. Lee was a Resident,Managed Care Pharmacy, Advance Paradigm, Hunt Valley, MD); SANDRA D.CASSARD, SC.D. is an independent health services researcher, Lutherville,MD; PETER E. DANS, M.D., F.A.C.P., is a Medical Consultant, AdvancePCS,Hunt Valley, MD; CLARE WHEELOCK is Senior Director of DatabaseDevelopment, Disease Management, AdvancePCS, Hunt Valley, MD; andJOSEPH D. OBER, Pharm.D., BCPS, is Senior Regional Manager, CareManagement Division, GlaxoWellcome, Rockville, MD (at the time this articlewas written, Dr. Ober was Senior Director, Disease Management, AdvanceParadigm).

AUTHOR CORRESPONDENCE: Jung H. Lee, Pharm.D., AstraZeneca, 1775York Avenue, New York, NY 10128; Tel: 212-860-0998; Fax: 212-860-0998;E-mail: [email protected].

ACKNOWLEDGEMENTS: The authors wish to thank Babette Duncan,Pharm.D., and Vikki Oates for their input and assistance.

This study was presented in part at the Academy of Managed Care PharmacyNinth Annual Meeting, New Orleans, LA, May 9, 1997.


Authors

sthma is a chronic inflammatory disease of the air-ways that affects approximately 4%–5% of the gener-al population.1 This disorder may cause recurrent

wheezing, breathlessness, chest tightness, and coughing.Variable airflow and bronchial hyperresponsiveness to a varietyof stimuli are characteristic. The disease, which affects peopleof all ages, is responsible for approximately 470,000 hospitaladmissions annually in the United States, with hospitalizationrates highest among African-Americans and children.2

Over the past 20 years, the prevalence, morbidity, and mor-tality of asthma has increased in the United States and world-wide.3 In response to this alarming trend, in 1991 the NationalEducation and Prevention Program (NAEPP) instituted by theNational Heart, Lung, and Blood Institute (NHLBI) publishedguidelines for the diagnosis and management of asthma. 4 Twoadditional reports have since been prepared by asthma expertsin cooperation with the NHLBI: The International ConsensusReport on Diagnosis and Treatment of Asthma (NHLBI 1992) andthe Global Initiative for Asthma (NHLBI/WHO 1995). 5, 6

The 1991 Expert Panel Report classified asthma severity asintermittent (step 1), mild persistent (step 2), moderate per-sistent (step 3), and severe persistent (step 4). 4 Key elements ofthe recommendations included recognition of the role ofinflammation in the pathogenesis of asthma and the use of anti-inflammatory pharmacotherapy early in the treatment of mod-erate to severe disease; inhaled corticosteroids were consideredthe standard daily preventive treatment.4 Several clinical trialshave demonstrated their efficacy using parameters such asbronchial responsiveness, peak expiratory flow, and symptomscores.7 Inhaled corticosteroids have been shown to reduce thefrequency of acute episodes of asthma and the need for con-current medications, lessen the requirements for oral steroids,and lower airway reactivity. 8–16 They have also been observed toreduce the risk of asthma hospitalization.7

New guidelines were published and new therapies approvedafter the period of our study. In April 1997, NHLBI publishedExpert Panel Report 2: Guidelines for the Diagnosis and Managementof Asthma.17 This revised guideline continues to emphasize theimportance of daily long-term anti-inflammatory therapy in addi-tion to medications to manage exacerbations in patients withpersistent asthma. The latest guidelines incorporate new drugsthat add to the armamentarium of long-term control medica-tions. The long-acting inhaled beta2-agonist salmeterol is indi-


A

by Jung H. Lee, Sandra D. Cassard, Peter E. Dans, Clare Wheelock, and Joseph D. Ober



cated for long-term prevention of bronchospasm, includingnighttime symptoms. The leukotriene modifiers are the newestclass of long-term control medications. Zafirlukast and mon-telukast are leukotriene receptor antagonists, while zileuton is a5-lipoxygenase inhibitor.

The national guidelines emphasize the importance of patienteducation, especially on adherence to therapy. It is estimated, how-ever, that adherence to therapy in both the pediatric and the adultasthma populations is only about 50%.18 Nonadherence to a pre-scribed therapeutic program is one of the factors suspected of con-tributing to asthma morbidity and mortality in all populations.19

Pharmacy and medical claims data have often been used tostudy the asthma population in the managed care settingbecause these large databases can be used to identify specificpatient populations (e.g., high risk), monitor appropriateness ofcare, and evaluate utilization patterns. 20 Prescription usage hasbeen used to infer patient compliance with therapy as well as toobserve prescribing patterns of physicians.21–24 Although thistype of analysis has its limitations, it can provide a broad pic-ture of the utilization patterns of a population and suggest prob-lem areas that may require special attention.

An area that has not yet been extensively explored is clinicalmanagement and patient behavior immediately before and afteran acute event such as an emergency room visit or hospitaliza-tion. An acute event might influence patients to become moreaware of their condition as well as more compliant with medica-tion therapy. From the provider’s perspective, the acute eventcould provide a “teachable moment,” an opportunity to educatepatients on an appropriate therapy. This article describes utiliza-tion of relevant prescriptions before and after an acute asthma-related event (hospitalization or ER visit) in an effort to determinepatterns in patient compliance and physician prescribing.

? ? Study Methodology

This retrospective observational pilot study consists of an analy-sis of pharmacy claims for short-acting beta2-agonists andinhaled anti-inflammatory agents and of medical claims sub-mitted between January 1, 1994, and September 30, 1995, in a275,000-member preferred provider organization (PPO) in themid-Atlantic region. Asthmatics were identified from theInternational Classification of Diseases, Ninth Revision (ICD-9-CM) diagnostic codes on the medical claims.25 Membersbetween the ages of 1 and 60 years with at least one encounterfor a primary or secondary diagnosis code for asthma (ICD-9codes of 493–493.99) were identified as “asthmatic.” Memberswith only one acute asthma-related event (ER visit or hospital-ization) occurring between July 1, 1994, and March 31, 1995,were included for analysis to ensure that sufficient medical-claim-free periods and pharmacy data were available for analy-sis before and after the acute event.

Acute events were defined as (1) a hospitalization caused byasthma (primary ICD-9 code of 493–493.99) or (2) an asthma-

related ER visit (primary ICD-9 code of 493–493.99 andCurrent Procedural Terminology (CPT) codes of 99281, 99282,99283, 99284, 99285, or 99288).26 The study population wasalso limited by including only those who had a continuouspharmacy benefit. It was assumed that patients were continu-ously enrolled with a pharmacy benefit throughout the studyperiod if they had pharmacy claims submitted in the first(January–March 1994) and last (July–September 1995) quartersof the study. Patients were excluded if they had a diagnosis ofcystic fibrosis (ICD-9 codes of 277–277.99) or chronic obstruc-tive pulmonary disease (ICD-9 codes of 491–492.99).

Once the study cohort was identified, the medical and phar-macy claims were integrated to create a profile for each patient.The acute event was considered to be “time zero” and the phar-macy data surrounding the acute event was incorporated into atime line. The period before “time zero” was divided into 90-day pre-acute event periods and the time period after “timezero” was divided into 90-day post-acute event periods. Theprofiles were then evaluated for trends in utilization of short-acting beta2-agonists and inhaled anti-inflammatory medica-tions before and after “time zero.”

TABLE 1 Pre-event Short-Acting Beta-Agonist Useby Post-event Short-Acting Beta-AgonistUse

Frequency/Percent

No

Yes

Total

No

3/3.61%

9/10.84%

12/14.46%

Yes

21/25.30%

50/60.24%

71/85.54%

Total

24/28.92%

59/1.08%

83/100.00%

Post-event

Pre-event “yes” — Patient had prescription filled for short-acting beta-agonist in the pre-event period

Pre-event “no” — Patient did not have prescription filled for short-actingbeta-agonist in the pre-event period

Post-event “yes” — Patient had prescription filled for short-acting beta-agonist in the post-event period

Post-event “no” — Patient did not have prescription filled for short-actingbeta-agonist in the post-event period.

McNemar test of significant changes reveals a statistically significantchange (p<0.05) in that a greater number of patients changed from notusing short-acting beta-agonists in the pre-event period to using them in thepost-event period (n=21) than the number of patients who changed fromusing the short-acting beta-agonists in the pre-event period to not usingthem in the post-event period (n=9).



The asthma medications analyzed in this study were theshort-acting beta2-agonists (albuterol, pirbuterol, bitolterol,metoproterenol, and terbutaline), inhaled corticosteroids(beclomethasone, flunisolide, fluticasone, and triamcinolone),inhaled cromolyn, and inhaled nedocromil. Statistical differ-ences were determined using the McNemar test for significantchanges.27 The McNemar test can be used when there are twonominal scale responses that are recorded before and after someevent. The test was appropriate for this study because we hadpaired dichotomous data. The McNemar test compared the off-diagonal frequencies in a 2 x 2 table (see Tables 1, page 304,and 2, above). The a priori level of significance was defined asp<0.05.

? ? Results

Of the approximately 8,000 members with a primary or secondarydiagnosis code of asthma, 83 met the inclusion and exclusion cri-teria. Acute exacerbations of asthma led to ER visits for 56% ofthese patients and hospitalizations for 42%. Hospitalizationsinclude patients who were initially admitted to the ER.

A greater number of patients (86%) received a prescription forshort-acting beta2-agonists after an acute event than before (71%).

The McNemar test for significant changes found a statistically sig-nificant greater number of patients using short-acting beta2-ago-nists in the post-acute event period who had not used them in thepre-acute event period compared to patients who did the reverse(i.e., patients who used short-acting beta2-agonists pre-acute eventbut not post-acute event) (p<0.05). In the calculation, the numberof patients who have a “no” response pre-acute event and a “yes”response post-acute event are compared to the number of patientswho have a “yes” response pre-event and a “no” post-event. Table1 represents the patients who changed their usage of short-actingbeta2-agonists. A statistically significant result was obtained(p<0.05) comparing the 21 patients in the upper right cell to the 9patients in the lower left cell of the 2 x 2 table.

The number of patients using inhaled anti-inflammatoryagents, however, did not change significantly between the pre-acute event period and the post-acute event period (p=n.s.)(Table 2). A closer look at the cross-tab analysis of the inhaledanti-inflammatory medications (Table 2) reveals that 47 of the83 patients were not on any inhaled anti-inflammatory medica-tion before or after the acute event; 7 were on an inhaled anti-inflammatory medication before but not after; 14 were on aninhaled anti-inflammatory medication after but not before; andonly 15 were on an inhaled anti-inflammatory medicationbefore and after the acute event.

In total, only 29 out of 83 patients (35%) were on aninhaled anti-inflammatory medication after an acute asthma-related ER visit or hospitalization.

In a separate analysis, the proportion of patients usingshort-acting beta2-agonists or inhaled anti-inflammatory med-ications was determined for each time period (see Figure 1,page 306). The proportion of patients using a short-actingbeta2-agonist or an inhaled anti-inflammatory medication washighest during post-acute event period one (70% of patients[58] received short-acting beta2-agonists and 28% [23] receivedinhaled anti-inflammatory medications). Figure 1 depicts theoverall use of asthma medications in the periods surroundingan acute event. As expected, more patients used asthma-relatedmedications in the 90 days after the acute event, but that num-ber declined with time.

? ? Discussion

Nonadherence with recommended therapy has been document-ed as a contributing factor in asthma morbidity and mortality. 19 Astudy that evaluated five urban teaching hospital emergencydepartments evaluated the correlates of compliance with follow-up appointments and prescription filling after an emergencydepartment visit.28 Of the 1,386 patients interviewed, only 45%(408 patients) recalled being advised to take a medication and12% (50) reported that they did not obtain the medications. In a1997 study by Ordonez et al., children aged 3 to 15 years admit-ted to an Australian hospital with an acute asthma attack wereevaluated to identify factors that might prevent future hospital

TABLE 2 Pre-event Inhaled Anti-inflammatory Useby Post-event Inhaled Anti-inflammatoryUse

Frequency/Percent

No

Yes

Total

No

47/56.63%

7/8.43%

54/65.06%

Yes

14/16.87%

15/18.07%

29/34.94%

Post-event

Pre-event “yes” — Patient had prescription filled for inhaled anti-inflammatories in the pre-event period.

Pre-event “no” — Patient did not have prescription filled for inhaled anti-inflammatories in the pre-event period.

Post-event “yes” — Patient had prescription filled for inhaled anti-inflammatories in the post-event period.

Post-event “no” — Patient did not have prescription filled for inhaled anti-inflammatories in the post-event period.

McNemar test of significant changes reveals no statistically significantchange in the number of patients using inhaled anti-inflammatories com-pating the pre-event period to the post-event period (p=n.s.).

Total

61/73.49%

22/26.51%

83/100.00%



admissions.29 This study found evidence of inadequate preventivetreatment as well as poor compliance within the 12 months priorto an acute asthma attack.

There is also evidence that many physicians do not adhere toexisting guidelines for the emergency management of asthma.30 Forexample, in a 1997 Canadian study, a cross-sectional survey deter-mined that many Canadian emergency physicians did not followpublished recommendations for the care of patients with acute asth-ma.31 This was especially true with regard to the aggressive use ofbeta2-agonists and the use of corticosteroids.

Testing our assumption that an acute event would lead tomore optimal therapy and better patient compliance, our resultsshowed that this was only partially true, in that the only asthmamedications whose use significantly increased after the acuteevent were short-acting beta2-agonists. Although more patientsdid receive inhaled anti-inflammatory medications after theacute event, the number was not statistically significant. Thisresult may indicate either poor patient compliance or physicianfailure to prescribe these important long-term therapies.

Our results show that the number of patients receiving short-acting beta2-agonists steadily increases as time nears the acuteevent, with 42 of 83 patients receiving inhaled short-acting beta2-

agonists during the 90 days beforehand (Figure 1). This may be anindication that patients were relying on the short-acting beta2-ago-nists to obtain quick relief of their worsening symptoms.

The proportion of patients using both the short-acting beta2-agonists and the inhaled anti-inflammatory medications was high-est during the 90 days after the event. This suggests some supportfor the hypothesis that an acute event leads to more optimal treat-ment. However, as time passed, the use of anti-inflammatory agentsdecreased.

Finally, the low proportion of patients receiving inhaled anti-inflammatory medications throughout the periods before and afterthe acute event suggests either low patient compliance or physicianfailure to prescribe these long-term controller medications.

? ? Limitations

This study used medical and pharmacy claims data for the pur-pose of observing utilization trends in a broad population.Although informative, administrative claims data have severallimitations:• The accuracy can be disputed. Accuracy of both medical and

pharmacy claims relies on how well the data are entered (e.g.,the ICD-9-CM and CPT diagnostic coding).

FIGURE 1 Patients on Inhaled Short-Acting Beta-Agonists and/or Inhaled Anti-inflammatories by Period(n=83) (90-day periods)

70

60

50

40

30

20

10

0

Period 5 Period 4 Period 3 Period 2 Period 1 AcuteEvent

Period 1 Period 2 Period 3 Period 4 Period 5

Pre-event Post-event

• • • • •?

?

? ?

?

?

? ?

?

?

•• • •

•8

22

32 32

42

8

3

1311 12

13

23

58

35

20

35

9

2

1014

? ?Number of patients on inhaled beta-agonists

•Number of patients on inhaled anti-inflammatories


• Use of prescription claims only assures that the medicationwas dispensed, not that the patient is actually taking the med-ications or using them correctly.

• Likewise, we cannot distinguish between patient noncompli-ance and physician failure to prescribe when a medication isnot dispensed.

• Claims data does not take into account the use of samples. • Additionally, claims data does not provide the clinical infor-

mation required to assess the severity of an illness. • Also, because of the exclusion criteria, we studied only those

patients with one acute event during the study period. It ispossible that these patients were not moderate-severe asth-matics who required anti-inflammatory medications butmild-intermittent asthmatics who only required the “as need-ed” use of short-acting beta2-agonists.

This pilot study specifically evaluated the short-acting beta2-agonists, inhaled corticosteroids, inhaled cromolyn, and inhalednedocromil. The use of other asthma medications, including theo-phylline, ipratropium, oral corticosteroids, salmeterol, andleukotriene modifiers (which were not Food and DrugAdministration–approved or available during the study period),was likely to have affected utilization of the short-acting beta2-ago-nists and inhaled anti-inflammatory medications. Extending thisstudy’s methodology to all the available asthma therapies couldprovide a better picture of actual drug utilization in the asthmaticpopulation.

Finally, the generalizability of this study’s conclusions is limitedby the small sample size resulting from the strict inclusion criteria.A larger study population and evaluation of data for a longer periodof time might provide a better picture of the trends in utilization.

? ? Conclusion

Nevertheless, the study provides some valuable lessons. First,although evaluations of claims data may lack clinical detail to ren-der definitive judgments, they can raise important issues about thequality of care.20 Claims data can be used to improve appropriate uti-lization, target continuing medical education, help manage complexpatients, identify underserved patients, and detect misprescribing aswell as fraud and abuse.32–39 This study examined the pharmacyclaims of a population with a condition that is highly prevalent yetdifficult to control in many cases. The results suggest a lack of com-pliance with recommended therapies and missed opportunities ofeducation and intervention during an acute event. Finally, this eval-uation can provide a framework for future initiatives that focus onpatient compliance and utilization around a sentinel event, a fruitfularea of research that has not been sufficiently exploited.

References

1. McFadden Jr, ER, Gilbert IA. Asthma. N Engl J Med 1992; 327(27):1928–37.

2. Centers for Disease Control and Prevention. Asthma mortality and hospital-ization among children and young adults—United States, 1990–1993.

MMWR 1996; 45: 350–53.

3. Asthma—United States, 1982–1992. MMWR 1995; 43: 952–55.

4. National Heart, Lung, and Blood Institute. National Asthma EducationProgram, Expert Panel Report. Guidelines for the diagnosis and managementof asthma. Pub. No. 91–3042. Bethesda, MD: US Department of Health andHuman Services, 1991.

5. National Heart, Lung, and Blood Institute. International consensus reporton diagnosis and treatment of asthma. Pub. No. 92–3091. Bethesda, MD: USDepartment of Health and Human Services, 1992.

6. National Heart, Lung, and Blood Institute. Global initiative for asthma. Pub.No. 95–3659. Bethesda, MD: US Department of Health and Human Services,1995.

7. Donahue JG et al.. Inhaled steroids and the risk of hospitalization for asthma. JAMA 1997; 277: 887–91.

8. Rafferty P et al. Comparison of budesonide and beclomethasone dipropi-onate in patients with severe chronic asthma: assessment of relative pred-nisolone-sparing effects. Br J Dis Chest 1985; 79: 244–50.

9. Barnes PJ, Pedersen S. Efficacy and safety of inhaled corticosteroids in asth-ma. Am Rev Respir Dis 1993; 148: S1–S26.

10. Dahl R et al. A dose-ranging study of fluticasone propionate in adultpatients with moderate asthma. Chest 1993; 104; 1352–58.

11. Fabbri L et al. on behalf of an International Study Group. Comparison offluticasone propionate with beclomethasone dipropionate in moderate tosevere asthma treated for one year. Thorax 1993; 48: 817–23.

12. Gustafsson P et al. Comparison of the efficacy and safety of inhaled fluti-casone 200mcg/day with inhaled beclomethasone dipropionate 400mcg/day inmild and moderate asthma. Arch Dis Child 1993; 69: 206–11.

13. Haahtela T et al. Comparison of a beta2-agonist, terbutaline, with aninhaled corticosteroid, budesonide, in newly detected asthma. N Engl J Med1991; 325: 388–92.

14. Jeffery PK et al. Effects of treatment on airway inflammation and thicken-ing of basement membrane reticular collagen in asthma. Am Rev Respir Dis1992; 145: 890–99.

15. Kamada AK et al. and the Asthma Clinical Research Network. Issues in theuse of inhaled glucocorticoids. Am J Respir Crit Care Med 1996; 153:1739–48.

16. van Essen-Zandvliet EE et al. Effects of 22 months of treatment withinhaled corticosteroids and/or beta 2-agonists on lung function, airway respon-siveness, and symptoms in children with asthma. Am Rev Respir Disease1992; 146: 547–54.

17. National Heart, Lung, and Blood Institute. National Asthma EducationProgram, Expert Panel Report 2. Guidelines for the diagnosis and manage-ment of asthma. Pub. No. 97-4051. Bethesda, MD: US Department of Healthand Human Services, 1997.

18. Cochrane GM. Therapeutic compliance in asthma: Its magnitude andimplications. Eur Respir J 1992; 5: 122–24.

19. Horn CR, Clark TJ, Cochrane GM. Compliance with inhaled therapy andmorbidity from asthma. Respir Med 1990; 84: 67–70.

20. Dans PE. Caveat doctor: how to analyze claims-based report cards. JtComm J Qual Improv 1998 (Jan); 24(1): 21–30.

21. Grymonpre RE et al. Pill count, self-report, and pharmacy claims data tomeasure medication adherence in the elderly. Ann Pharmacother 1998 (Jul-Aug); 32(7–8): 749–54.

22. Laumann JM, Bjornson DC. Treatment of Medicaid patients with asthma:comparison with treatment guidelines using disease-based drug utilizationreview methodology. Ann Pharmacother 1998 (Dec); 32(12): 1290–94.

23. Blandford L et al. Analyzing variations in medication compliance relatedto individual drug, drug class, and prescribing physician. J Managed CarePharm 1999 (Jan-Feb); 5(1): 47–51.


24. Nestor A et al. Cross-sectional analysis of the relationship between nation-al guidelines recommended asthma drug therapy and emergency/hospital usewithin a managed care population. Ann Allergy Asthma Immunol 1998 (Oct);81(4): 327–30.

25. Jones MK et al (eds): ICD-9-CM code book, Vols 1–3. Reston, VA: StAnthony Publishing, 1994.

26. Kirschner CG, ed: CPT 96: Physician’s current procedural terminology.Chicago: American Medical Association, 1995.

27. Zar JH, 2nd ed. Biostatistical analysis. Englewood Cliffs, NJ: Prentice-Hall,Inc. 1984, 156–58.

28. Thomas EJ et al. Patient noncompliance with medical advice after theemergency department visit. Ann Emerg Med 1996; 27(1): 49–55.

29. Ordonez GA et al. Preventable factors in hospital admissions for asthma.Arch Dis Child 1998; 78: 143–47.

30. Beveridge RC et al. Guidelines for the emergency management of asthmain adults. Can Med Assoc J 1996; 155(1): 25–37.

31. Grunfeld A et al. Management of acute asthma in Canada: an assessmentof emergency physician behavior. J Emerg Med 1997; 15(4): 547–56.

32. Kassirer JP. The use and abuse of practice profiles. N Engl J Med 1994;330: 634–36.

33. Weiner JP et al. Applying insurance claims data to assess the quality ofcare: A compilation of potential indicators. Qual Rev Bull 1990; 16: 423–38.

34. Lasker RD, Shapiro DW, Tucker AM. Realizing the potential of practicepattern profiling. Inquiry 1992; 29: 287–97.

35. McNeil BJ, Pedersen SH, Gatsonis SC. Current issues in profiling qualityof care. Inquiry 1992; 29: 298–307.

36. Iezzoni LI. Data sources and implications: Administrative databases. In:Iezzoni LI, ed. Risk adjusting for measuring health care outcomes. Ann Arbor,MI: Health Administration Press, 1994.

37. Epstein A. Performance reports on quality: Prototypes, problems andprospects. N Engl J Med 1995; 333: 57–61.

38. Welch HG, Miller ME, Welch WP. Physician profiling: An analysis of inpa-tient practice patterns in Florida and Oregon. N Engl J Med 1994; 330:607–12.

39. Weinter JP et al. Variation in office-based quality: A claims-based profileof care provided to Medicare patients with diabetes. JAMA 1995; 273:1503–8.



The second report of the Committeeon the Quality of Health Care inAmerica, Crossing the QualityChasm: A New Health Systemfor the 21st Century is currentlyavailable as an advanced copy (uncor-rected proofs).1 It expands the com-mittee’s work beyond the focus onmedical and medication error in ToErr Is Human: Building a SaferHealth System to a larger issue: theneed to improve the quality of healthcare.2 The new report outlines aimsand principles for an improved designfor the delivery of care.

The text emphasizes that it is writtennot to present specifics but to pres-ent a new perspective on the pur-pose, intents, interactions, andprocesses of health care. The authorsacknowledge that it will be necessaryto redesign structures and processesof organizations and of professionalsand their interactions. They alsoacknowledge that the practices of thenation’s health care system, providers,and users must change to includeimprovements in dissemination andapplication of knowledge and potentialadvances in care. They state that the

nation’s health care system, providers,and users must have access to anduse information technology in prac-tice. Changes must be made to fiscal(payment) policies to facilitate newand better options and to reward bet-ter outcomes, and health careproviders must be better educated.

The committee emphasizes that thereport was written as a vision of whatis possible and that changes must bemade to the system to achieve thevision. The first report of theCommittee on the Quality of HealthCare in America, To Err Is Human:Building a Safer Health System,was summarized previously in thisjournal (J Managed Care Pharm 2001:62–68).

The goals of this continuing educationprogram are to inform readers aboutthis important report and to presentthe committee’s recommendations forbuilding organizational support forchange.

KEYWORDS: Quality, health system,health care delivery

A Prescription for Change: Bridges to Cross the “Quality Chasm”

LINDA L. NORTON, Pharm.D., is Associate Professor, Department ofPharmacy Practice, and Director, Postgraduate Professional Education,University of the Pacific School of Pharmacy and Health Sciences, Stockton, CA.

AUTHOR CORRESPONDENCE: Linda L. Norton, Pharm.D., Director,Postgraduate Professional Education, University of the Pacific School ofPharmacy and Health Sciences, 751 Brookside Road, Stockton, CA 95211;Tel: 209-946-2414; Fax: 209-946-3180; E-mail: [email protected].

The Academy of Managed Care Pharmacy is approved by the American Council on Pharmaceutical Education as a provider of continuing pharmaceutical education. Individuals may obtainup to 1 contact hour of credit or 0.10 Continuing Education Unit

(CEU). The ACPE number is 233-000-01-004-H04. Certificates will be mailedwithin eight weeks to participants who successfully complete the CE exam andachieve a score of 70% or more and submit the exam to AMCP prior to August31, 2002. Learning objectives and test questions follow on page 315.

Copyright© 2001 Academy of Managed Care Pharmacy, Inc. All rightsreserved.

Author

CONTINUING EDUCATION

he Committee on the Quality of Health Care inAmerica was formed three years ago and was given adirective to develop a plan to make enormous improve-

ments in health care over the next 10 years. The members of thecommittee have taken a multipronged approach to achievingtheir mission. They have:• conducted a literature review with a focus on quality of care;• organized workshops to improve understanding of the issues

surrounding health care quality;• identified factors that impact improvement efforts;• considered ideas for improving accountability for quality; and• identified research that may lead to improved quality in

health care.Throughout all these activities, the committee has been con-

cerned with the delivery of care to individuals rather than gen-eral public health.

The work chronicles the failure of the current health care sys-tem and the problems that often make it impossible for newhealth-related knowledge and information to improve the qualityof our health care. It details how the quality of health care canimprove when technology is applied to make such informationaccessible and useful to clinicians and administrators.

Most of us already know that inconsistencies in health carecause the quality of health care to vary from exemplary to ineffec-tive. We also know that more than 40 million Americans haveinadequate access to care. This is often because of a lack of healthinsurance. We know that, at times, some if not most areas of healthcare can be considered poor stewards of the resources we have.The committee’s report also details how fragmentation inhibitsuseful sharing of information and clinical communication. Overall,there has been little progress toward restructuring and improvingthe system even though we know that it is not operating at peakefficiency for patients, providers, or payors.

?? Prescription for Change

Given the need for improvements in our current system and thecontroversies surrounding the provision of health care, one largequestion looms: How can health care, and managed care pharma-cy, be positioned for the future when we are struggling simply tomeet the demands of today? The health care system, and pharma-cy in particular, is facing decreased payments, inadequate staffing,more demands from the uninsured or the underinsured and a shiftfrom more easily contained and defined episodic care for acuteneeds to a need for long-term continuity of care for chronic condi-tions. This last demand has resulted in increased reliance on med-


Tby Linda L. Norton

ications in the face of what are often difficult fiscal arrangements.The quality committee has responded to today’s health care

challenges with a surprising conclusion. It has announced thatit is increasingly apparent that pushing harder to make the cur-rent system work cannot be the answer. The system does notand will not work on a patient-centered, timely, effective, effi-cient, equitable, or even safe basis for all Americans. It has pro-claimed that change, a major change, in the health care systemis needed. This change should include, among other things, asignificant improvement in meeting the six aims set forth by thecommittee for a new health care system (see Table 1, page 311).

The committee has suggested a change toward care that issafer, more efficient, timely, effective, equitable, and centered onthe patients who get or need care. However, to provide this levelof care, the new health care system will need to overcome sev-eral challenges. It must:• change the way care is delivered; • use more information technology, and use it more effectively,

to support both administrative and clinical efforts;• manage patient care knowledge and skills to produce better

outcomes;• put together effective multidisciplinary care teams;• improve coordination of patient care; and• maximize the use of performance and outcome measures.

To help organizations meet these challenges, the committeepresents basic information on principles for change that havebeen successful in other industries, with explanations to helpguide the process of change. These five principles are summa-rized below. The committee also presented 10 basic rules for thehealth care system. These are summarized in Table 2, page 311.Finally, the committee presented 13 specific recommendationsto keep the health care system focused on a positive change.These recommendations are shown in Table 3, page 312.

? ? Five Principles for Change Design

It has been said that change is the only constant. Many health carepractitioners would agree. Health care has already undergone atremendous amount of change. However, health care is not theonly sector in the U.S. economy that is undergoing change.Change seems to be endemic in our economy.

Take banking. Less than 20 years ago, most of our bankinghad to be completed at our bank branch on Monday throughThursday from 10:00 a.m. to 3:00 p.m. or on Friday between9:00 a.m. and 6:00 p.m. Now, we can access and review accountsin the middle of night at automatic teller machines around theworld or while sitting at home at our personal computers. Thisradical change has occurred in less than a generation. Most of usremember the days when “banker’s hours” represented a schedulemany of us longed for. This is no longer the case.

In engineering, architecture, and even medicinal chemistryand pharmacology, computer models are taking the place of

painstakingly assembled manual models of proposed projects.In the computer models, projects can be reviewed for feasibili-ty, fit, efficacy, or cost before any solid models are developed.Design and turnaround times can be shortened while costlymistakes are avoided.

These examples represent what could be a positive change forhealth care. Imagine your patients or subscribers accessing healthcare in the middle of the night while sitting at their personalcomputers. Or imagine a team of surgeons working toward a newtechnique through virtual surgery before any real incisions aremade. Picture a patient viewing digital instructions for exercisesdesigned by a physical therapist. Or envision a patient with ques-tions about medications accessing a reliable site on the Internet toreview side effects and interactions, and even discussing thoseproblems online with a pharmacist. Furthermore, imagine that apharmacist has access to a patient’s entire medical record withoutcalling a hospital, primary care provider, specialist, insurancecompany, or pharmacy benefit manager.

The vision suggested by the Committee on the Quality ofHealth Care in America would allow each of these examples tobecome a reality. The committee does, however, recognize thatmost health care institutions, providers, payors, and patientswill need some help making this vision reality. To aid in thatprocess, the committee has suggested five principles for design-ing the suggested changes.

The first principle is the 80/20 principle: “Design for theusual, but recognize and plan for the unusual.” The bulk of anynew health care process should be dedicated to handling 80%of the work that needs to be done in that area—the predictablework. The design for this predictable work should allow sim-plification and standardization, with low variance from thestandards of care. The contingency plan for the other 20% ofthe work can be developed as needed. Other applications of the80/20 principle should be considered in designing the new sys-tem. For example, 20% of patients will likely account for 80%of the work and cost, and a high percentage of an effect may beattributed to a low number of causes.

The second principle is to design for safety. This principlehas three specific components:1. Systems should be designed to prevent errors by avoiding

dependence on an individual health care provider’s memoryand attentiveness. Instead, designs should simplify processesusing checklists, protocols, and guidelines.

2. Systems should be designed to make any errors that do occurvisible before harm comes to a patient.

3. Designs should include procedures to reduce any harm thatdoes occur. Technical support systems such as computerizedprotocols for antidote administration and harmless or nearharmless default modes for equipment will most likely playan important role in this area.The third principle is mass customization, which combines



the efficiency of mass production with the unique attributes thatare brought by a custom consumer-driven product. A commonexample is the way many of us order laptop and desktop com-puters. All the computer components are mass-produced, buteach customer can specify which components to include in agiven computer. In this way, the computer can meet the indi-vidual needs of each customer.

If components are appropriately designed with the 80/20principle in mind, the first principle can apply to mass cus-tomization also: 80% of customers or patients will select their“custom design” options from 20% of the selections. However,another idea for design is that under the right circumstances,each option should be a consideration for 80% of customers orpatients. An example for managed care pharmacy that combinesmass customization and the 80/20 principle could be the designof prescription copayment options. Of the menu of possibilities,


TABLE 1 The Six Major Aims Suggested by theCommittee on the Quality of Health Carein America

Aim Explanation

Safe Should not cause injury

Effective Provide evidence-based service to patients withneed and avoid overuse

Patient-centered Provide responsive care with courtesy

Timely Reduced waiting times and delays

Efficient Avoid waste

Equitable Limit individual and geographic variations in care

The Rule

1 Care must be based on continuous healing relationships rather than episodic encounters. Patients should be able to access care whenever and wherever they need care. Care should not be restricted to office, urgent care, and hospital visits. Patients should have access to responsive care 24 hours a day every day of the year. Care should not be restricted or be available only at a lower level on the weekends or at night. This care would be accessible by telephone or even over the Internet, but these methods of access should augment rather than replace face-to-face visits. Many patients and clinicians will still need and want the assurance of face-to-face encounters, which provide valuable information that may be missed during “virtual visits.”

2 Customization must be based on the patient’s needs and values, not the provider’s. Systems should be designed to meet the most common needs, but also to respond to individual needs, choices, and requests.

3 Patients should be able to control their own care. When provided with the necessary health information at an appropriate and under-standable level, patients should be able to employ the level of control they want to exercise over decisions that affect their health and health care. The system design must allow patients to select the level of control they desire. They should be encouraged and trained to share in the decision-making process. Some patients will opt to give the responsibility to their health care provider, and the system should accommodate that choice.

4 Knowledge must be shared and information must flow freely. Patients must be allowed access to their own medical information. They should also have access to the latest clinical information. Both clinicians and patients should be encouraged and trained to communicate effectively and share information.

5 Decisions must be based on the best evidence rather than habit or training. Clinicians should use the best scientific evidence available to reduce illogical variations in care.

6 Safety must be designed into the system. Patients should not be injured by the health care they receive. Systems designed to prevent error and reduce the impact of errors that do occur are key to reducing the risks to which patients are exposed.

7 Evidence on system performance should be readily available and the process should be transparent to patients. Information on the health care system must be available to patients. Access to such information will allow patients and their families to make decisions based on evidence rather than rumor. Information on alternative treatments and choices should include details describing system performance asit relates to safety, evidence-based medicine, and satisfaction.

8 The system should anticipate patients’ needs. The health care system must do more than just react to demands. It should anticipate needs.

9 Waste in time and resources must be decreased.

10 Clinicians should cooperate, collaborate, and communicate. All clinicians and health care facilities should work together to ensure appropriate care and exchange of information.

TABLE 2 Rules for Redesigning the Health Care System



The purpose of health care in the United States

Six major aims of health care

Monitoring and tracking the progress of change

Ten basic rules for redesigning health care

Identifying 15 priority conditions and ways to improve

those conditions

The Health Care Quality Innovation Fund

Workshops to address six identified challenges

Advances in science that are useful to health care’s constituents

Information and documentation system to support health care

Payment methods that reward, not impede, quality improvement

Research to improve the connection between payments and quality goals

Summit to strategize and assess restructuring of clinical education

Research on how the regulatory and legal systems’ affect the changes

needed, and how they can be modified to support the needed changes

The purpose of the U.S. health care system, including all of its constituents,should be to decrease, on an ongoing basis, the impact of illness, injury,and disability, and to increase the health and functioning of the people.

The health care system and its constituents should pursue six aims. That is,health care should be safe, effective, efficient, equitable, patient-centeredand timely. Overall, health care should meet the needs of each patient.

The federal government, including congress and the Department of Healthand Human Services (HHS), should establish processes to monitor andtrack the progress of the health system in reaching the six aims for healthcare. The Secretary of HHS should prepare an annual report on thatprogress.

The constituents of the health care system should work together toredesign the system’s processes in accordance with the 10 rules listed inTable 2, page 311.

The Agency for Healthcare Research and Quality (AHRQ) should identifyat least 15 priority conditions and, with constituents of the health care sys-tem, create processes for achieving significant improvements in those prior-ity conditions within five years.

Congress should establish a fund to support projects that are designed tomeet the six aims for health care and to produce improvements in priorityconditions. Funds should be invested in projects that will produce resultsfor the public that have widespread application.

The AHRQ and other funding sources should develop workshops for thehealth care community in order to identify and implement the best optionsto:

1. Redesign practice-based care processes.2. Improve the use of technology and allow access to patient data andother clinical information.3. Enhance the management of knowledge and skills.4. Build multidisciplinary teams.5. Coordinate care in all situations.6. Incorporate and use performance and outcome measurements.

HHS should make advances in scientific knowledge more useful for andavailable to clinicians and patients.

Leaders in government, industry, and health care should commit to assem-bling an information infrastructure that meets all of the needs of healthcare. One outcome should be the elimination of handwritten clinical databy the end of the decade.

Purchasers of care should remove barriers that impede quality improve-ment and should provide incentives to spur development.

Health care system constituents should formulate a research plan to identi-fy and test options for aligning payment methods with quality goals.

Health care leaders should develop plans to align all clinical education withthe ideals of the new health system, and should assess how the proposedchanges will affect credentialing, funding, and sponsorship of clinical edu-cation programs.

The AHRQ should fund research to discover how the regulatory and legalsystems will affect the changes that will be required in the health care sys-tem and how these systems can aid in the accomplishment of the six aimsproposed for the new health care system.

1

2

3

4

5

6

7

8

9

10

11

12

13

TABLE 3 The Recommendations of the Committee on the Quality of Health Care in America


payors/employers could choose the 20% of options that fit thespecific needs of 80% of their employees.

The fourth principle is to design for continuous flow or “abatch size of one.” The concept behind continuous flow is thatthe care process is matched to the demand and no queuing ofpatients or samples is necessary before processing.

There are important assumptions embedded in this principle,including the belief that the demand for service can be satisfied andthat the demand is relatively constant and predictable. If thedemand is believed to be beyond what may reasonably be satisfied,then a system must be designed to manage the demand.Unfortunately, one way to manage demand is to design in barriers,such as a waiting period or a queue. However, the committeebelieves that continuous flow and matching design and demandwill result in better care by reducing missed appointments and pro-cedures, not to mention duplication of effort on the part of bothpatient and the provider.

The fifth principle presented is production planning. Thisallows resources such as staff and equipment to be used mostappropriately to meet patients’ needs and to lower wastes andcosts. As in mass production, the repetitive process and the nat-ural flow of work in many aspects of health care can be defined;resources can be allocated to meet this repeating cycle. Thecycle may repeat daily, weekly, or even based on the phase of themoon. Although it is possible that only those individuals mostfamiliar with the work will recognize the cycle; a good under-standing of the work will allow many of these cycles to be iden-tified. Once the cycle is identified and confirmed, resources canbe allocated to increase or decrease staff and meet the needs ineach phase of the cycle.

? ? Aims, Rules, and Recommendations

The committee has based much of its work on the idea thateven after extensive redesign, our health care system will not besimple. Nor, even subject to rules, will it become rigid andmechanical. The committee fully expects the system to remaincomplex but adaptive.

Complex implies that the system will remain highly interactiveand interdependent. Adaptive means that the system can change,and it will be less predictable than a mechanical system. This is inlarge part because at least one component of the system reacts tostimuli in a variety of ways: In general, if humans are involved, thesystem will be adaptive. The human component of the systemintroduces a fundamental lack of predictability.

In light of the belief that the 21st-century version of ourhealth care system will remain complex and adaptive, the com-mittee has provided a limited number of aims and rules (seeTables 1 and 2). The aims are part of the committee’s vision forthe results of the redesign of care. The rules can provide theguidance needed to help that vision become reality. In addition,complex adaptive systems produce amazing results from appli-cation of just a few rules.

In addition to this effect, systems like the U.S. health caresystem exhibit an interesting array of outcomes that are beyondthe inherent adaptability of the system. These effects includenonlinear results (e.g., small changes can result in large effects),an overall lack of predictability, inherent abilities to evolve newbehaviors, an intrinsic order even in the absence of a centralcontrol (acceptance of a few basic rules or concepts mayaccount for some of this order), interdependence, and even co-evolution.

In spite of the impressive potential for positive outcomesfrom a complex adaptive new system—or perhaps because ofthe potential for the lack of predictability outcome—the com-mittee has adopted 13 recommendations, endorsing the phras-ing of the Commission on Consumer Protection and Quality inthe Health Care Industry (see Table 3, page 312). The commit-tee’s recommendations should provide direction for the con-stituents of the health care system. The recommendationsshould also provide initial guidance as participants movethrough the process of creating new designs and changing care.

? ? Application to Managed Care Pharmacy

The challenge for managed care pharmacy is how to accomplisha major change to meet the six aims of the Committee on theQuality of Health Care in America. There has clearly been someprogress through the use of disease management for providerquality improvement, the requirements for pharmacoeconomicstudies to support Food and Drug Administration and formula-ry submissions, and an emphasis on evidence-based medicineto decrease variance in patient care. However, there is a largechasm between the vision of the committee and the develop-ment of a pragmatic action plan. Formularies, preferred druglists, differential copayments, and other managed care toolshave come under significant criticism although they areattempting to apply some standards to the use of drugs. The“five principles” are directed to total system approaches thatrequire cooperative team-oriented problem solving. Someimprovements are being made in designs for safety (e.g.,progress in electronic prescribing and electronic medicalrecords), but there has been relatively little progress in masscustomization and continuous flow designs. The accomplish-ment of these principles is dependent on health care providerscoming together in teams with the goal of improving the systemfor patients.

What can managed care pharmacy do to achieve the goals ofthe committee? As the payor or payor advocate managed carepharmacy can stress the aims and principles of the committeein provider contracts. Managed care can define the “best prac-tices” of patient-centered care and contract with those providerswilling to provide evidence that they are meeting the practicesor incrementally moving toward those practices. “Centers ofexcellence” exist in all industries including health care anddemand higher prices for their efforts. These centers can be fos-




tered through preferred contracts and moving patients to thesecenters when appropriate. These concepts can also apply to theservices desired of pharmacists, the standards of care required,and the expected outcomes. Once these issues are definedappropriate fee structures can be developed to reimburse foroutcomes, and distribution systems can be designed using theprocesses (i.e., mass customization, batch processing) and tech-nology required to meet safety and financial goals.

The committee has emphasized a broad philosophy forchange. The aims and principles required are not different fromthe broad goals of managed care and managed care pharmacy.What is required of managed care pharmacy is to declare thevision, model the culture and principles, and settle for nothingless than the standards established by the committee.

? ? ConclusionAs discussed in To Err is Human, many patients are injured bythe health care that was intended to improve their lives.2

Although the United States has some of the best care available,the health care system too frequently fails to deliver on thepromise of excellent care. In specifying the need for change andsuggesting the most appropriate changes to be made, theCommittee has presented models for change and enunciated 6

aims, 10 rules, and 13 recommendations to improve health carein the United States. All this effort is intended to be both a guideand the impetus for change in health care.

The committee’s effort presents a vision of what may be possible.No matter what course is taken, in a system as complex and adap-tive as the U.S. heath care system, change is inevitable. One ques-tion that arises is: Will that change be for better or worse?

As with previous changes, there will be supporters anddetractors of the role that managed care medicine and managedcare pharmacy will have in the new heath care system.However, the committee has suggested that with the appropri-ate vision—focused on their six aims, adhering to their rules,and adopting their recommendations—the change as a wholewill be very much for the better. ?

References

1. Committee on Quality of Health Care in America. Crossing the QualityChasm: A New Health System for the 21st Century. Institute of Medicine.Washington: National Academy Press, Advance Copy (uncorrected proofs),2001; available at www.nationalacademies.org/Webextra/chasm.

2. Committee on Quality of Health Care in America, Institute of Medicine. Toerr is human. Building a safer health system. Kohn LT, Corrigan JM,Donaldson MS, eds. Washington: National Academy Press, 1999.


C E E X A M

Upon completion of this article, the suc-cessful participant should be able to:1. Describe the 13 recommendations

presented by the Committee toimprove the quality of health care.

2. Outline the six aims for improvementset forth by the Committee.

3. List at least 5 of the 10 rules forhealth care in the 21st century.

4. Summarize five design principles usedsuccessfully in health care that canserve as patterns for redesign andchange in the system.

5. Suggest an agenda for change in amanaged care pharmacy setting that isbased on the committee’s aims, rules,and recommendations.

SELF-ASSESSMENT QUESTIONS

1. The six aims stated by the Committee on the Quality of Health Care in America should serve as a vision of what may be possible with a redesign of health care.a. Trueb. False

2. The six aims include specific information on all of the following except:a. focus of care.b. the use of evidence-based medicine.c. equitability of care.d. cost reduction.

3. The 10 rules for redesigning the health care system include at least one rule on safety. When consideringsafety, health care in the 21st centuryshould be designed to:a. prevent errors.b. make any errors that do occur visible.c. reduce the risk of serious harm

when an error does occur.d. A and B only.e. A, B, and C.

4. Health care in the 21st century should be based on continuous healing relationships. To achieve this,patients should have access to care:a. during face-to-face visits.b. via e-mail or through “virtual visits.”c. 24 hours a day every day of the year.d. A and C only.e. A, B, and C.

5. If health care is to reach its full potential for providing excellent care,knowledge must be shared between clinicians and patients. Implicit in this statement are all of the followingexcept:a. Patients should have access to

their medical information.b. Clinicians should share only the

information they believe to be appropriate and patients should inform clinicians of their progress.

c. Both patients and clinicians shouldhave access to the latest clinical information.

d. Both patients and clinician should be trained to communicate effectively.

6. The purpose of health care in the United States should be to decrease the impact of injury, illness, and dis-ability and to increase the health andfunctioning of the people.a. Trueb. False

7. To provide the level of care suggestedby the committee, challenges must be overcome. These include all of thefollowing except:a. a change in the way health care is

delivered.b. improved access to and use of

information technology. c. increased reliance on measures

of process and decreased reliance on measures of patient outcomes.

d. improved coordination of patient care.

8. A number of barriers stand in the way of innovations in health care. One of the barriers specifically addressed in the committee’s recom-mendations is:a. a reduction in the current use of

multidisciplinary care teams.b. payment methods that fail to

reward innovation and may impede efforts to improve the quality of care.

c. the use of technology to manage clinical knowledge.

d. all of the above.

9. Based on the 80/20 principle:a. 80% of health care expenditures

pay for patient care; the other 20% cover administrative cost.

b. 80% of patients use their health care coverage regularly; the other 20% do not.

c. 80% of health care resources are used for the care of 20% of the patients.

d. none of the above.

10.Techniques that have been successfully used for the redesign of health care include all of the following except:a. design for safety to (1) prevent

errors, (2) make errors that do occur more visible, and (3) reducethe impact of errors that do occur.

b. design for the usual but recognize and plan for the unusual.

c. continuous flow or “a batch size ofone.”

d. The use of patient teams to develop a plan for redesign.

The Academy of Managed Care Pharmacy is approved by the American Council on Pharmaceutical Education (ACPE) as a provider of continuingpharmaceutical education. Individuals may obtain up to 1 contact hour of credit or 0.10 Continuing Education Unit (CEU). The Universal ProgramNumber is 233-000-01-004-H04. Certificates will be mailed within eight weeks to participants who successfully complete the CE exam and achievea score of 70% and submit the exam to AMCP prior to August 31, 2002. See text of article beginning on page 309.



INSTRUCTIONS

This test affords 1 hour (0.10 CEU) of continuing pharmaceutical education in all states that recognize the American Councilon Pharmaceutical Education. To receive credit, you must score at least 70% of your test answers correctly. To record ananswer, darken the appropriate block below. Mail your completed answer sheet to: Academy of Managed Care Pharmacy,

100 N. Pitt Street, Suite 400, Alexandria, VA 22314. If you score 70% or more, a certificate of achievement will be mailed to you with-in eight weeks. If you fail to achieve 70% on your first try, you will be allowed only one retake. The ACPE Provider Number for thislesson is 233-000-01-004-H04. This offer of continuing education credit expires August 31, 2002.

A B C D E

1. ? ?

2. ? ? ? ? ?

3. ? ? ? ? ?

4. ? ? ? ? ?

5. ? ? ? ?

A B C D

6. ? ?

7. ? ? ? ? ?

8. ? ? ? ?

9. ? ? ? ?

10. ? ? ? ?

11. ? A ? B ? C ? D ? E

12. ? Yes ? No

13. Minutes

14. ? Yes ? No

15. ? A ? B ? C ? D

Participant Identification: Please type or print. Date:

Social Security #: Work Phone #:For Identification Purposes Only

Name:LAST FIRST MIDDLE

Company:

Address:STREET (with Apt. No.) or P.O. Box CITY STATE ZIP

State & Lic. No.:STATE LICENSE NO.

Member Type: ? Active ? Supporting Associate ? Student ? Nonmember

Signature:

11. In what type of setting do you work?(Leave blank if none of the responsesbelow applies.)

a. HMO

b. PPO

c. Indemnity insurance

d. Pharmacy benefits management

e. Other

12. Did this program achieve its educational objectives?

a. Yes b. No

13. How many minutes did it take youto complete this program, includingthe quiz? (Fill in on answer sheet.)

14. Did this program provide insightsrelevant or practical for you or yourwork?

a. Yes b. No

15. Please rate the quality of this CE article.

a. Excellent c. Fairb. Good d. Poor


I verify by my signature above that I have completed this examination independently.

DEMOGRAPHIC INFORMATION (not for scoring)

CE Exam

Note: If you wish, you can now take this CE Exam online. Go to www.amcp.org.


The Journal of Managed Care Pharmacy(JMCP) is peer reviewed and only articlesmeeting high scholarly standards indica-tive of sound, reproducible, and validresearch are accepted.

Editorial content is determined by theEditor-in-Chief with suggestions from theEditorial Advisory Board. The views andopinions expressed in JMCP do not nec-essarily reflect or represent official policyof the Academy or the authors’ institu-tions unless specifically stated.

? ? Editorial Content

The Journal of Managed Care Pharmacycontains three basic types of editorialmaterial: peer-reviewed scholarly articles,feature articles, and departments. Articlesshould be organized, written, and for-matted for a specific part or section ofthe journal.

The Journal of Managed Care Pharmacyseeks contributions from authors in theareas of managed care pharmacy practice,pharmacotherapy, research, education,economics, and other areas of pharmacypractice.

Peer-Reviewed ArticlesThe heart of the journal is its peer-reviewed scholarly research, review, andreport articles. JMCP accepts:

• Comparative Research: articles usingthe scientific method to compare defini-tively two or more hypotheses.• Review Articles: papers that review recent clinical, economic, or manage-ment literature and offer synthesized summaries relevant to the managed care pharmacy field.• Descriptive Reports: articles describ-ing experiences or solutions to practicalproblems within managed care pharmacy settings.• Continuing Education: invitedreviews of timely topics with continuingedu- cation credit.

Feature ArticlesFeature articles highlight news andinformation of interest to managed carepharmacists, pharmacy personnel, andother health care providers. Topics areselected based on advice from JMCP’s

Editorial Advisory Board of managedcare pharmacists and administrators,educators, and industry representatives.

DepartmentsDepartments provide member-contributedand staff-written information of practicaland immediate value to journal readers:

• Letters: JMCP’s letters to the editor.• Perspectives: editorials by outside contributors as well as the editors.• Campus: updates from the pharmacy education world about inclusion of managed care topics in college curricula,activities of faculty and students inter-ested in managed care, and techniquesfor teaching managed care principles topharmacy students.• Caveats: updates from the legal and regulatory world.• International: articles describing man-aged care and/or pharmacy developmentsand emerging markets in foreign countries.• AMCProgress: news and information about the Academy and the activities of its leaders, members, and staff on behalf of managed care pharmacy.

? ? Advertising Policy

A copy of the full advertising policy forJMCP is available from AMCP headquar-

ters and the Advertising Representative.All aspects of the advertising sales andsolicitation process are completely inde-pendent of the editorial process. Adver-tising is interspersed throughout theeditorial material other than peer-reviewed articles. Advertising is notaccepted for placement opposite or nearrelated editorial copy.

Employees of advertisers may submitarticles for publication in the editorialsections of JMCP, subject to the usualpeer-review process. Financial disclo-sure and conflict of interest statementsare required when manuscripts are sub-mitted, and these may be published atthe discretion of the Editor-in-Chief ifthe article is printed.

Editorial Office:Contact: Mitchell Petersen, Inc.1775 Jamieson Avenue, Suite 210Alexandria, VA 22314-5715Tel: 703-518-4700Fax: 703-518-8495e-mail: [email protected]

Advertising Sales Office:Professional Media Group, Inc.40 N. Woodbury Road, Pitman, NJ 08071Tel: 800-486-5454 or 856-589-5455Fax: 856-582-7611

JMCP EDITORIAL POLICY

Mission Statement

JMCP, a publication of the Academy of Managed Care

Pharmacy, is dedicated to providing managed care phar-

macists, associates and students with the tools to excel in

their daily practices by focusing on:

? Policy: Providing a forum for in-depth discussion of issues

of topical and long-term importance.

? Practice: Presenting information of interest and educa-

tional value to the membership.

? Research: Publishing research that increases the quality

of research standards used in managed care pharmacy

practice and helps apply that research to improve the prac-

tice of managed care pharmacy.

JMCP accepts for consideration manu-scripts prepared in accordance with theUniform Requirements for theSubmission of Manuscripts toBiomedical Journals.1

?? Manuscript Preparation

Manuscript length should be 10–20 type-written pages (1500–3000 words), includ-ing tables, figures, and references. Due tospace limitations, authors are encouragedto limit the number of tables/figures to nomore than three. Manuscripts shouldinclude, in this order, a title page, a sepa-rate page identifying all authors (includingdegrees, employers, contact information,and financial disclosure and conflicts ofinterest), an abstract of no more than 250words, text, appendices, references, figurecaptions, tables, and figures. Each sectionshould begin on a new page with one-inch margins on all sides.

JMCP abstracts should be written nar-ratives and contain the informationdescribed for each type of article shownbelow, where applicable:Comparative Research

• Objective• Design• Setting• Patients/Participants• Interventions• Main Outcome Measures• Results• Conclusion

Review Articles• Objective• Data Sources• Study Selection• Data Extraction• Data Synthesis• Conclusion

Descriptive Reports• Objective• Setting• Practice Description• Practice Innovation• Interventions• Main Outcome Measures• Results• Conclusion

?? Reference Style

References should be prepared in the

style of Index Medicus. Shown below areexamples of common types of referencesprepared in JMCP style.1. Standard journal article(list all authors when four or less; when five ormore, list only the first three and add et al.)Lennard EL, Feinberg PE. Overview of the NewYork State program for prescription drug benefits.Am J Hosp Pharm 1994; 51: 2944-48.2. No author givenAnonymous. Top 25 U.S. hospitals, ranked by admis-sions, 1992. Man Healthcare 1994(Sep); 4(9): 64.3. Journal paginated by issueCorrigan PW, Luchins DJ, Malan RD, Harris J. User-friendly CQI for the mental health team. MedInterface 1994(Dec); 7(12): 89-92, 95.4. Book or monograph by authors Tootelian DH, Gaedeke RM. Essentials of pharmacymanagement. St. Louis, MO: C.V. Mosby, 1993.5. Book or monograph with editor, compiler, orchairman as authorChernow B, ed. Critical care pharmacotherapy.Baltimore, MD: Williams & Wilkins, 1995, 430 pp.6. Chapter in a bookKreter B, Michael KA, DiPiro JT. Antimicrobial pro-phylaxis in surgery. In: DiPiro JT, Talbert RL, HayesPE, Yee GC, Matzke GR, Posey LM, ed.Pharmacotherapy: a pathophysiologic approach.Norwalk, CT: Appleton & Lange, 1992: 1811-12.7. Government agency publication Akutsu T. Total heart replacement device. Bethesda,MD: National Institute of Health, National Heartand Lung Institute; 1974 Apr. Report no.: NIH-NHLI-69-2185-4.8. Dissertation or thesisYoussef NM. School adjustment of children withcongenital heart disease [dissertation]. Pittsburgh,PA: University of Pittsburgh, 1988.9. Paper presented at a meetingReagan ME. Workers’ compensation, managed care,and reform. Paper presented at the 1995 AMCRAMidyear Managed Care Summit. San Diego, CA:1995 Mar 13.

?? Submission of Manuscripts

Four complete copies of the manuscript,including originals of figures and tables,should be submitted to the JMCP

Managing Editor at Mitchell Petersen,Inc., 1775 Jamieson Ave., Suite 210,Alexandria, VA 22314-5715; 703-518-4700, 703-518-8495 (fax).

Please send us an electronic version ofthe manuscript, either on a disk or via e-mail to [email protected]. Alltext should be in a word processing pro-gram (preferably Microsoft Word) and allfigures and tables should be saved in adrawing program (preferably Photo-shopor Illustrator) or saved as a separate tifffile. Identify the format (PC or MAC), allprograms used, and all file names.

In a cover letter, the correspondingauthor should: • Briefly describe the importance and

scope of the manuscript;• Certify that the paper has not been

accepted for publication or published previously and that it is not under consideration by any other publication;

• Suggest names of possible reviewers when appropriate; and

• Identify the nature and extent of any financial interest or affiliation that any author has with any company, product, or service discussed in the manuscript.One of the following statements must

be signed by all authors and submittedwith the manuscript:

“In consideration of the Academy of ManagedCare Pharmacy (AMCP) taking action and review-ing and editing this submission, the author(s)undersigned hereby transfer(s), assign(s), or other-wise convey(s) all copyright ownership to AMCP inthe event that this work is published by AMCP.”

“I was an employee of the United StatesGovernment when this work was prepared forpublication; it is therefore not protected by theCopyright Act, and there is no copyright that canbe transferred.”

Once peer review is complete (general-ly 6–8 weeks), the corresponding authorwill be notified of the status of the article.

REFERENCES

1. International Committee of Medical Journal

Editors. Uniform requirements for manuscripts

submitted to biomedical journals. N Engl J Med

1991; 324: 424-48.

Author’s Guidelines


The Journal of Managed Care

Pharmacy is indexed by

International Pharmaceutical

Abstracts (IPA), Iowa Drug

Information Service,

Medscape (www.Med-

scape.com), and Pharmscope

(www.pharmscope.com).

he Academy of Managed CarePharmacy (AMCP) has received amajor grant from Janssen Pharma-

ceutica to help finance a coalition initiative toimprove the effectiveness of online prospec-tive drug-utilization review (OPDUR).

The AMCP/U.S. Pharmacopeia (USP)–led coalition has proposed the establish-ment of an OPDUR system that includes“dead stops” in instances in which adrug/drug or drug/class interaction couldcause serious harm to the patient. Thissystem would require a pharmacist todocument an intervention on behalf ofthe patient before the prescription couldbe dispensed. To make the system practi-cal in daily use, these “dead stops” wouldconcentrate only on the potentially mostserious interactions, rather than all possi-ble interactions.

Funds such as those received fromJanssen Pharmaceutica will be used todefray a variety of expenses, including thedevelopment of drug/drug interaction cri-teria by an expert committee of decisionmakers, consultants to advise on liabilityand other legal issues, and the testing ofthe methodology developed.

Also serving on the Steering Com-mittee for the initiative are the AmericanSociety for Automation in Pharmacy, theAmerican Pharmaceutical Association,the Food and Drug Administration, theNational Community Pharmacists Asso-ciation, the National Association of ChainDrug Stores, and the Pharmaceutical CareManagement Association.

? ? Sterler, Nudelman Honored

Lowell Sterler received the 2001 AMCPDistinguished Service Award and PhillipNudelman received the Foundation forManaged Care Pharmacy’s (FMCP’s)Award for Achievement at AMCP’s 13thAnnual Business Meeting in Tampa,Florida.

Sterler, an AMCP member since 1990,is vice president of professional relationsat AdvancePCS in Scottsdale, Arizona. Hehas served as AMCP president, chair of the

Nominations Committee, chair of theLegislative Committee, and as president ofthe Board of Trustees for FMCP, where heled efforts to promote the value of man-aged care pharmacy and pharmaceuticals.

The AMCP Distinguished ServiceAward recognizes an AMCP member forexceptional and sustained volunteer serv-ice and commitment to the Academy.Candidates for this award must have atleast five years of extraordinary volunteerservice to the Academy and must havebeen involved in activities critical to theachievement of AMCP’s mission. A selec-tion committee made up of AMCP boardmembers and committee chairs who areappointed by the AMCP president choos-es the award recipient.

In choosing Sterler for this award, thecommittee recognized not only his manyleadership roles within AMCP, but alsohis continuing efforts to represent theAcademy and the profession of pharmacyto external audiences, to educate themabout the value and importance of phar-macy. Sterler has represented AMCPbefore the Food and Drug Administrationin discussions regarding the agency’sauthority over health plans and phar-macy benefits management companies.He has also participated in the WhiteHouse Summit and other meetings relat-ing to Y2K planning and drug stockpil-ing; spearheaded talks with the Centersfor Disease Control and Preventionregarding the role of pharmacists inassisting public health initiatives; and hasbeen a frequent participant in AMCPLobby Days, representing the Academyand its members on Capitol Hill.

Sterler authored the chapter on phar-macy distribution systems and networkmanagement in the classic textbookManaged Care Pharmacy Practice; servesas AMCP’s diplomat to his alma mater,South Dakota State University; and wasamong the first of AMCP’s e-mail men-tors, working with students to help themunderstand the complexities of managedcare pharmacy practice.

The FMCP Award for Achievement rec-ognizes an individual for lifetime sustained,exemplary, and distinguished service tomanaged care pharmacy. Candidates mustbe professional role models who are mak-ing or have made significant and sustainedcontributions to the advancement of theprofession of managed care pharmacy in aclinical or an administrative practice. Therecipient is chosen by the Past Presidentsand Founders Advisory Council, and is notrequired to be an AMCP member.

Phillip Nudelman, Ph.D., is the inau-gural recipient of this prestigious award.He is chief executive officer of HopeHeart Institute of Seattle, Washington,and former chairman and president ofKaiser/Group Health. Nudelman beganhis career as a clinical pharmacist beforebecoming director of professional servic-es at Group Health Cooperative (GHC) ofPuget Sound. While at GHC, Nudelmanpioneered the use of technology in phar-macy, leading the effort that created andimplemented CoopRx, one of the firstpharmacy information systems. He alsoimplemented decentralized pharmacypilots at GHC, and put pharmacy in thelead role in the health care system’s for-mulary and pharmacy and therapeuticscommittee processes. He implementedthe use of the pharmacy database as aresearch tool at GHC. As chair of theWashington State Board of Pharmacy(1973–1977), Nudelman led efforts torequire pharmacist-patient counselingand mandatory continuing education forpharmacists.

Nudelman was chairman of the boardof the American Association of HealthPlans from 1998–2000 and has served onthe White House Task Force on HealthCare Reform (1993–1995); the PewHealth Professionals Commission (1997–1999), which dealt with pharmacy man-power, among other issues; and on thePresident’s Advisory Commission onConsumer Protection and Quality in theHealthcare Industry (1997–1998).

A M C P r o g r e s sN e w s a n d i n f o r m a t i o n a b o u t t h e A c a d e m y

AMCP/USP Receive Grant for Initiative

T


Continued on page 324


? ? AMCP Issues Practice Advisory On Tablet Splitting

Recognizing that tablet splitting has been anaccepted practice in health care for manyyears, AMCP has issued a ProfessionalPractice Advisory to recommend situationsin which tablet splitting may be beneficial,when it is inadvisable, and how to ensurepatient safety when the practice is used.

Tablet splitting has been used to obtainthe appropriate prescribed dose of a med-ication when the dosage is not availablefrom the manufacturer. It has been partic-ularly useful for pediatric patients and theelderly, who often require doses other thanthose supplied by manufacturers. Patientson flexible dosing schedules, or those whoneed to gradually increase or decreasedosages of a medication, may also need tosplit tablets.

Managed health care systems haveused tablet-splitting strategies to helpcombat the skyrocketing cost of prescrip-

tion drugs while providing quality, cost-effective care. In many cases, manufactur-ers price different strengths of the samemedication equally. By splitting a tabletthat is twice the strength of the dosedesired, the cost could be cut in half.Given the extraordinarily high cost ofprescription drugs, many life-savingmedications are beyond the reach ofsome patients. Tablet splitting can helppreserve access to comprehensive, high-quality drug benefits without impairingthe quality of care.

Because patient safety is the primaryconcern of managed health care systems,health plans should have guidelines inplace to determine which medicationsmay be appropriate for tablet splitting.The guidelines should detail precautionsthat health plans should take whenchoosing tablets to be split both to obtainaccurate dosing and to minimize thechance of error and adverse events.

The Professional Practice Advisorycontains suggestions on precautions thathealth care organizations should takewhen deciding which medications can besplit, advice on types of medications inap-propriate for splitting, and the role of thepharmacist in ensuring patient under-standing and safety when tablet splitting isrequired.

To review the Professional PracticeAdvisory on tablet splitting, see the“Policy Digest and Practice Advisory” cat-egory at www.amcp.org. ?

? ? Correction

In the list of AMCP Committee Chairs for2001–2002 in the May/June AMCProgress,the chair of the Past Presidents andFounders Advisory Council was inadver-tently omitted. He is Michael Dillon, CHDMeridian Healthcare.

AMCProgress

JOURNAL OF MANAGED CARE PHARMACY · JMCP volume seven • number four july/august 2001...

Documents

Transcript of JOURNAL OF MANAGED CARE PHARMACY · JMCP volume seven • number four july/august 2001...