November 2014 - Academy of Managed Care Pharmacy
Transcription
November 2014 - Academy of Managed Care Pharmacy
J MCP ■ Journal of Mana g ed Care & Specialty Pharmac y ■ November 2014 Journal of Managed Care & specialty Pharmacy Volume 20 ■ Number 11 ■ November 2014 ■■ SPECIALty PHARMACY ■ Vol. 20, No. 11 Do Value Thresholds for Oncology Drugs Differ from Nononcology Drugs? Yuna Hyo Jung Bae, PharmD, and C. Daniel Mullins, PhD Predictors of Treatment Initiation with Tumor Necrosis Factor-α Inhibitors in Patients with Rheumatoid Arthritis Rishi J. Desai, PhD; Jaya K. Rao, MD; Richard A. Hansen, PhD; Gang Fang, PhD; Matthew L. Maciejewski, PhD; and Joel F. Farley, PhD ■ Pa g es 1069-1146 ■■ benefit management Price Elasticity and Medication Use: Cost Sharing Across Multiple Clinical Conditions Justin Gatwood, PhD, MPH; Teresa B. Gibson, PhD; Michael E. Chernew, PhD; Amanda M. Farr, MPH; Emily Vogtmann, PhD, MPH; and A. Mark Fendrick, MD North Carolina Medicaid Recipient Management Lock-In Program: The Pharmacist’s Perspective S. Rose Werth, BA; Nidhi Sachdeva, MPH, CHES; Andrew W. Roberts, PharmD; Mariana Garrettson, MPH; Chris Ringwalt, PhD; Leslie A. Moss, MHA, CHES; Theodore Pikoulas, PharmD, BCPP; and Asheley Cockrell Skinner, PhD The Impact of Information Technology on Managed Care Pharmacy: Today and Tomorrow Douglas Hillblom, PharmD; Anthony Schueth, MS; Scott M. Robertson, RPh, PharmD; Laura Topor, BA; and Greg Low, RPh, BSPharm, PhD Is There an Association Between the High-Risk Medication Star Ratings and Member Experience CMS Star Ratings Measures? Sara C. Erickson, PharmD; R. Scott Leslie, MPH; and Bimal V. Patel, PharmD, MS ■■ online exclusive A Case Study in Generic Drug Use: Should There Be Risk Adjustment in Incentive Payments for the Use of Generic Medications? Surrey M. Walton, PhD; Christine Rash, PharmD; Bruce L. Lambert, PhD; and William L. Galanter, MD, PhD Perceptions and Attitudes of Community Pharmacists Towards Generic Medicines Suzanne S. Dunne, BSc (Hons), MSc; Bill Shannon, MD, FRCGP, MICGP; Walter Cullen, MD, MICGP, MRCGP; and Colum P. Dunne, BSc (Hons), MBA, PhD Journal of Managed Care & Specialty Pharmacy A Peer-Reviewed Journal of the Academy of Managed Care Pharmacy ■ www.jmcp.org ■ www.amcp.org Help AMCP Recognize Managed Care Pharmacy’s Leaders Nominate Someone for an AMCP Award! The AMCP Distinguished Service Award recognizes an AMCP member who has made exceptional and sustained contributions to AMCP over at least a five year period. Last year’s winner was Marty Mattei. The AMCP Grassroots Advocacy Award recognizes an individual or group responsible for significant activity around a grassroots cause. Last year’s winners were Caroline Atwood and Marshal Abdullah. The Individual Contribution Award recognizes an AMCP member making a significant contribution to the Academy by any means other than service as an AMCP Committee Member. Last year’s winner was John Strezewski. The Spirit of Volunteerism recognizes a current AMCP Committee Member who has demonstrated exemplary and outstanding service to AMCP over the past year; and, has provided volunteer activities that resulted in successful and/or high quality AMCP program, projects or services for its members. Last year’s winner was Patrick Gleason. JMCP Award for Excellence recognizes an article published in JMCP that represents the best scholarly work in managed care pharmacy. The award presented in 2014 went to The Effect of Hepatitis C Treatment Response on Medical Costs: A Longitudinal Analysis in an Integrated Care Setting. And consider applying to become an AMCP Fellow! The Fellows program was established to acknowledge sustained excellence in the pharmacy profession, grant recognition for exceptional contributions, long term commitment and active participation in the Academy. e Submit your nomination or download a Fellow application at www.amcp.org/awards All nominations and completed Fellow applications are due December 16, 2014. a1c a1c simplify diabetes data management 1c a1c Connecting thousands of members with their healthcare teams and vital diabetes data is easy with ARK Care®. Generate reports for use in patient charts, pay-for-performance and HEDIS reporting. a1c For a personal demonstration, call 855.646.3235 or visit www.glucocardusa.com/arkcaredemo COMPLETEDIABETESSOLUTIONS™ USM 2014-00021 Rev 03/14 AMCP Headquarters 100 North Pitt St., Suite 400, Alexandria, VA 22314 Tel.: 703.683.8416 • Fax: 703.683.8417 Editorial staff Volume 20, No. 11 Editor-in-Chief John Mackowiak, PhD 919.942.9903, [email protected] C ONTENTS Publisher Edith A. Rosato, RPh, IOM, Chief Executive Officer Academy of Managed Care Pharmacy Assistant Editor Laura E. Happe, PharmD, MPH 864.938.3837, [email protected] ■ commentary 1073 The Impact of Information Technology on Managed Care Pharmacy: Today and Tomorrow Douglas Hillblom, PharmD; Anthony Schueth, MS; Scott M. Robertson, RPh, PharmD; Laura Topor, BA; and Greg Low, RPh, BSPharm, PhD ■ BRIEF REPORT 1086 Do Value Thresholds for Oncology Drugs Differ from Nononcology Drugs? Yuna Hyo Jung Bae, PharmD, and C. Daniel Mullins, PhD Assistant Editor Eleanor M. Perfetto, MS, PhD 410.706.6989, [email protected] Assistant Editor Karen L. Rascati, PhD 512.471.1637, [email protected] Managing Editor Jennifer A. Booker 703.317.0725, [email protected] Copy Editor Carol Blumentritt 602.616.7249, [email protected] 1093 A Case Study in Generic Drug Use: Should There Be Risk Adjustment in Incentive Payments for the Use of Generic Medications? Surrey M. Walton, PhD; Christine Rash, PharmD; Bruce L. Lambert, PhD; and William L. Galanter, MD, PhD Advertising ■ RESEARCH Advertising for the Journal of Managed Care & Specialty Pharmacy is accepted in accordance with the advertising policy of the Academy of Managed Care Pharmacy. Graphic Designer Margie C. Hunter 703.297.9319, [email protected] For advertising information, contact: Derek Lundsten, VP, Business Development American Medical Communications, Inc. 630 Madison Avenue, Manalapan, NJ 07726 Tel.: 973.713.2650 E-mail: [email protected] 1102 Price Elasticity and Medication Use: Cost Sharing Across Multiple Clinical Conditions Justin Gatwood, PhD, MPH; Teresa B. Gibson, PhD; Michael E. Chernew, PhD; Amanda M. Farr, MPH; Emily Vogtmann, PhD, MPH; and A. Mark Fendrick, MD 1110 Predictors of Treatment Initiation with Tumor Necrosis Factor-α Inhibitors in Patients with Rheumatoid Arthritis Rishi J. Desai, PhD; Jaya K. Rao, MD; Richard A. Hansen, PhD; Gang Fang, PhD; Matthew L. Maciejewski, PhD; and Joel F. Farley, PhD editorial Questions related to editorial content and submission should be directed to JMCP Managing Editor Jennifer Booker: [email protected]; 703.317.0725. Manuscripts should be submitted electronically at jmcp.msubmit.net. subscriptions Annual subscription rates: USA, individuals, institutions–$90; other countries–$120. Single copies cost $15. Missing issues are replaced free of charge up to 6 months after date of issue. Send requests to AMCP headquarters. reprints (continued on page 1073) Journal of Managed Care & Specialty Pharmacy (ISSN 1944-706X) is published 12 times per year and is the official publication of the Academy of Managed Care Pharmacy (AMCP), 100 North Pitt St., Suite 400, Alexandria, VA 22314; 703.683.8416; 800.TAP.AMCP; 703.683.8417 (fax). The paper used by the Journal of Managed Care & Specialty Pharmacy meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper) effective with Volume 7, Issue 5, 2001; prior to that issue, all paper was acid-free. Annual membership dues for AMCP include $90 allocated for the Journal of Managed Care & Specialty Pharmacy. Send address changes to JMCP, 100 North Pitt St., Suite 400, Alexandria, VA 22314. 1070 Journal of Managed Care & Specialty Pharmacy JMCP November 2014 Vol. 20, No. 11 Information about commercial reprints and permission to reuse material from JMCP may be found at amcp.org/ JMCP_Reprints_and_Permissions. Authors may order reprints from the Sheridan Press; contact contact Tamara Smith, [email protected], 800.352.2210 All articles published represent the opinions of the authors and do not reflect the official policy or views of the Academy of Managed Care Pharmacy or the authors’ institutions unless so specified. Copyright © 2014, Academy of Managed Care Pharmacy. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, without written permission from the Academy of Managed Care Pharmacy. www.amcp.org Facts and Figures: Part D Drug Utilization Provided by MedImpact Healthcare Systems, Inc. Top 10 Drugs by Rx Claim Cost 79% Top 10 Drugs by Rx Claim Volume 5.0% Sofosbuvir 3.3% Levothyroxine Sodium 3.6% Insulin Glargine 3.0% Lisinopril 1.8% Fluticasone/Salmeterol 2.9% Omeprazole 1.8% Rosuvastatin Calcium 2.7% Simvastatin 1.7% Adalimumab 21% 75% 1.6% Lenalidomide 25% 2.6% Atorvastatin Calcium 2.6% Amlodipine Besylate 1.5% Sitagliptin Phosphate 2.2% Metformin HCL 1.5% Aripiprazole 2.2% Hydrocodone BIT/Acetaminophen 1.5% Etanercept 1.8% Furosemide 1.4% Tiotropium Bromide 1.6% Losartan Potassium Top 10 Drugs by Claim Cost: Prior Authorization and Step Therapy Trends by Brand National Snapshot % of Total Medicare Rx lives (n=28,855,056) 100% 80% 60% 40% 20% 0% Sovaldi Lantus Advair Diskus Crestor Humira % of Lives with PA Revlimid Januvla Abilify Enbrel Spiriva % of Lives with ST Highest . . . . . . . . . . . . . . . . . . . . . . . . . . . . Share Top-10 Rx Claim Cost . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Lowest Regional Variation Top 3 Drugs by Claim Cost Prior Authorization West 1. Enbrel 2. Humira 3. Sovaldi Step Therapy Northeast 1. Enbrel 2. Humira 3. Revlimid Midwest 1. Enbrel 2. Humira 3. Revlimid West 1. Januvia 2. Crestor 3. Abilify Midwest 1. Januvia 2. Crestor 3. Advair Diskus Northeast 1. Januvia 2. Crestor 3. Advair Diskus South 1. Januvia 2. Crestor 3. Advair Diskus South 1. Enbrel 2. Humira 3. Revlimid Source: MedImpact August 2014 claims data and BusinessOne Technologies’ Maestro. Accessed 9/3/2014. Geographical regions as defined by U.S. Census Bureau. Visit www.medimpact.com today 10181 Scripps Gateway Ct. San Diego, CA 92131 Phone: 800.788.2949 www.medimpact.com www.amcp.org Vol. 20, No. 11 November 2014 JMCP Journal of Managed Care & Specialty Pharmacy 1071 Board of Directors President: Dana Davis McCormick, RPh President-Elect: Raulo S. Frear, PharmD Past President: Kim A. Caldwell, RPh Treasurer: H. Eric Cannon, PharmD, FAMCP Chief Executive Officer: Edith A. Rosato, RPh, IOM Director: Stanley E. Ferrell, RPh, MBA Director: James T. Kenney, Jr., RPh, MBA Director: Janeen McBride, RPh Director: Lynn Nishida, RPh Director: Gary M. Owens, MD Volume 20, No. 11 C ONTENTS (continued) Editorial advisory board ■ RESEARCH 1122 North Carolina Medicaid Recipient Management Lock-In Program: The Pharmacist’s Perspective S. Rose Werth, BA; Nidhi Sachdeva, MPH, CHES; Andrew W. Roberts, PharmD; Mariana Garrettson, MPH; Chris Ringwalt, PhD; Leslie A. Moss, MHA, CHES; Theodore Pikoulas, PharmD, BCPP; and Asheley Cockrell Skinner, PhD 1129 Is There an Association Between the High-Risk Medication Star Ratings and Member Experience CMS Star Ratings Measures? Sara C. Erickson, PharmD; R. Scott Leslie, MPH; and Bimal V. Patel, PharmD, MS 1138 Perceptions and Attitudes of Community Pharmacists Towards Generic Medicines Suzanne S. Dunne, BSc (Hons), MSc; Bill Shannon, MD, FRCGP, MICGP; Walter Cullen, MD, MICGP, MRCGP; and Colum P. Dunne, BSc (Hons), MBA, PhD ■ departments 1071 Facts and Figures: Part D Drug Utilization, MedImpact Healthcare Systems, Inc. Editorial Mission JMCP publishes peer-reviewed original research manuscripts, subject reviews, and other content intended to advance the use of the scientific method, including the interpretation of research findings in managed care pharmacy. JMCP is dedicated to improving the quality of care delivered to patients served by managed care and specialty pharmacy by providing its readers with the results of scientific investigation and evaluation of clinical, health, service, and economic outcomes of pharmacy services and pharmaceutical interventions, including formulary management. JMCP strives to engage and serve professionals in pharmacy, medicine, nursing, and related fields to optimize the value of pharmaceutical products and pharmacy services delivered to patients. JMCP employs extensive biasmanagement procedures intended to ensure the integrity and reliability of published work. 1072 Journal of Managed Care & Specialty Pharmacy JMCP November 2014 Vol. 20, No. 11 Committee Purpose: To advise and assist the editors and staff in the solicitation and development of JMCP content. For more information see: www.amcp.org/ eab. To volunteer to be a committee member, watch for the call for AMCP volunteers every September. Trent McLaughlin, BSPharm, Xcenda, Scottsdale, AZ (chair) Karen Worley, PhD, Humana, Inc., Cincinnati, OH (chair) Christopher Bell, MS, GlaxoSmithKline, Research Triangle Park, NC Gary Besinque, PharmD, FCSHP, Kaiser Permanente, Downey, CA Norman V. Carroll, BSPharm, PhD, Virginia Commonwealth University, Richmond, VA Mark Conklin, MS, PharmD, Pharmacy Quality Solutions, Sewickley, PA Bridget Flavin, PharmD, Regence, Portland, OR Renee Rizzo Fleming, BSPharm, MBA, PRN Managed Care Consulting Services, LLC, East Amherst, NY Patrick Gleason, PharmD, BCPS, FCCP, Prime Therapeutics LLC, Minneapolis, MN Todd A. Hood, MHA, PharmD, Celgene Corporation, Cumming, GA Mark Jackson, BSPharm, PIVINA Consulting, Inc., Windsor, ON Donald Klepser, MBA, PhD, University of Nebraska Medical Center, Omaha, NE Stephen J. Kogut, BSPharm, MBA, PhD, University of Rhode Island, Kingston, RI Bradley C. Martin, PharmD, PhD, University of Arkansas for Medical Sciences, Little Rock, AR Uche Anadu Ndefo, BCPS, PharmD, Texas Southern University, Houston, TX Robert L. Ohsfeldt, PhD, Texas A&M Health Science Center, College Station, TX Gary M. Owens, MD, Gary Owens Associates, Ocean View, DE Cathlene Richmond, BCPS, BS, PharmD, Kaiser Permanente, Oakland, CA Mark Christopher Roebuck, MBA, PhD, RxEconomics, Hunt Valley, MD Jordana Schmier, MA, Exponent, Alexandria, VA Marv Shepherd, BSPharm, MS, PhD, University of Texas at Austin, Austin, TX Jennifer Booker, Academy of Managed Care Pharmacy, Alexandria, VA John Mackowiak, PhD, Academy of Managed Care Pharmacy, Alexandria, VA Author Guidelines: http://amcp.org/JMCP_AuthorGuidelines Mission Statement: http://amcp.org/JMCP_MissionStatement www.amcp.org c o mm e n tary The Impact of Information Technology on Managed Care Pharmacy: Today and Tomorrow Douglas Hillblom, PharmD; Anthony Schueth, MS; Scott M. Robertson, RPh, PharmD; Laura Topor, BA; and Greg Low, RPh, BSPharm, PhD SUMMARY Understanding the use of health information technology (HIT) and its implications is crucial for the future of managed care pharmacy. Information is the cornerstone of providing and managing care, and the ability to exchange data is easier and more complicated than ever before. In this commentary, a subset of the Academy of Managed Care Pharmacy Healthcare Information Technology Advisory Council addresses how HIT supports managed care today and its anticipated evolution, with a focus on quality, patient safety, communication, and efficiency. Among the tools and functions reviewed are electronic health records, electronic prescribing, health information exchange, electronic prior authorization, pharmacists as care team members, formularies, prescription drug abuse, and policy levers to address these issues. J Manag Care Pharm. 2014;20(11):1073-79 Copyright © 2014, Academy of Managed Care Pharmacy. All rights reserved. I n health care, there is a silent and often unrecognized dependency on health information technology (HIT). It underlies every element of managed care pharmacy, from payers managing a pharmacy benefit through the provision of medication therapy management (MTM) and innovative programs that deliver quality pharmaceutical care to providers involved in direct patient care. It is also redefining how key stakeholders in medication management communicate with each other. This commentary describes how managed care pharmacy uses HIT today and how that will evolve in the near future. It examines how information is shared among managed care programs, providers, sponsors, government, and members, facilitating more informed decision making and enhancing productivity and efficiency. It also presents challenges and opportunities for moving forward. These insights were identified by a subset of the Academy of Managed Care Pharmacy (AMCP) Healthcare Information Technology Advisory Council, whose role is to advise AMCP membership on the role of HIT in managed care pharmacy and who believes that for HIT to continue to serve managed care pharmacy, AMCP must be involved in its development. ■■ HIT in Managed Care Pharmacy Today Managed care pharmacy currently employs many essential HIT tools including pharmacy claims processing, electronic prescribing (e-prescribing), computerized physician order entry (CPOE), and electronic health records (EHRs).1 Pharmacy Claims Processing Pharmacies began to make the technology leap from paper claims to real-time claims adjudication with the National www.amcp.org Vol. 20, No. 11 Council for Prescription Drug Programs’ (NCPDP) creation of the pharmacy universal claim form in 1977 and rollout, 11 years later, of the NCPDP Telecommunication Standard Version 1.0. Prospective drug utilization review (PDUR)—regulated by Medicaid in January 1993—then enabled pharmacists to receive real-time adverse drug event alerts when a prescription claim violates a pre-established criterion for appropriate drug use. The current NCPDP Telecommunications Standard (vD.0) supports over 20 transaction functions and facilitates the exchange of over 4 billion real-time pharmacy claims with such speed and simplicity that most participants do not realize they are using sophisticated HIT.2 Unlike medical claims that are still predominantly submitted in “batches” over the course of days or even weeks, the vast majority of pharmacy claims today are processed and adjudicated in a matter of few seconds through data centers that maintain high availability, such as those of Emdeon and RelayHealth—2 providers of health information exchange services. E-prescribing and CPOE The expanded influence of HIT is evident in many areas of managed care today, with the expanded adoption of e-prescribing and medication orders that are increasingly integrated in EHRs.3 Although the functionality to prescribe medications electronically has been available for decades, it was primarily in the acute care setting until 2006. While the ability to order medications has continued to expand in hospitals, the ability to electronically write and transmit prescriptions in the ambulatory setting is now mainstream and becoming the standard of care. At its core, this function eliminates misinterpretation of handwriting, reduces the need to rekey information and the corresponding possibility of error, and improves the transmission process. Most e-prescribing and CPOE solutions have incorporated formularies, providing complementary clinical and economic decision support, which provide additional quality and financial value. According to Surescripts, a leading network providing prescription transmission services, there were over 1 billion prescriptions sent electronically in 2013. Nearly three-quarters (73%) of office-based physicians are e-prescribing, and nearly all retail pharmacies are wired to receive e-prescriptions, providing meaningful value to managed care pharmacy.4 Over a 7-year period (2005-2012), a Michigan coalition of General Motors, Ford Motor Company, Chrysler, Blue Cross Blue Shield of Michigan, the Health Alliance Plan, Express Scripts, CVS Caremark, and Catamaran found that e-prescribing saved stakeholders $119 million in drug costs and $11 million from avoided hospitalizations.5 November 2014 JMCP Journal of Managed Care & Specialty Pharmacy 1073 The Impact of Information Technology on Managed Care Pharmacy: Today and Tomorrow Electronic Health Records (EHRs) An EHR is a systematic collection of electronic health information about an individual patient or population6 that is increasingly capable of being shared across different health care settings. EHRs may include a range of data, including demographics, medical history, medications and allergies, immunization status, laboratory test results, radiology images, vital signs, personal statistics such as age and weight, and billing information. Designed to represent data that accurately captures the state of the patient at all times, it allows an entire patient history to be viewed without the need to track down the patient’s previous medical records and assists in ensuring that data are accessible, accurate, appropriate, and legible. In 2013, about three-quarters of physicians had adopted EHRs due in part to federal incentive programs.7 The effect of this level of adoption is promising for managed care. In a study conducted from 2004 to 2009, Kaiser found a statistically significant decrease in emergency department (ED) visits (28.8 per 1,000) and hospitalizations among diabetes patients whose physicians used its EHR.8 The next step in developing EHRs is to connect disparate EHRs into health information exchanges (HIEs). ■■ Developing Areas for HIT in Managed Care Pharmacy Advances in HIT are creating new opportunities for managed care pharmacy to improve the efficiency and quality of health care. HIT and HIEs are creating new roles and responsibilities for managed care pharmacists. HIT will facilitate critical functions, such as obtaining prior authorization to ensure that patients receive medications that are safe, appropriate, and cost-effective. Further progress will enable use of HIT to identify and reverse the rise in prescription drug abuse. New forms of HIT, such as telehealth and mobile health, will be adopted by managed care pharmacists and patients and will reduce costs, improve outcomes, and engage patients in their care. Health Information Exchanges (HIEs) HIEs will knit together unrelated information sources to provide health care professionals with a more comprehensive view of a patient’s medical information. This is needed to connect hospitals, providers, payers, and pharmacies to effectively and efficiently collect and share clinical, pharmacy, and administrative data. HIE also provides the means to communicate interventions performed by pharmacists, such as medication counseling, to other providers and patients. There are nearly 300 HIEs in the United States that enable the electronic sharing of health-related information. One-half of the nation’s hospitals are now participating in a regional, state, or private HIE, and 71% plan to invest in technology to ensure their connectivity to HIEs in the next 2 years.9 Furthermore, nearly one-half of the nation’s physicians plan to join an HIE. The benefits of managed care involvement in HIE are plentiful. They include preventing avoidable admissions; simplifying and streamlining drug and medical authorizations; improving quality of information and streamlined distribution of reporting for quality improvement programs; reducing avoidable service 1074 Journal of Managed Care & Specialty Pharmacy JMCP November 2014 utilization and costs associated with ED visits; and simplifying and streamlining the flow and presentation of information in an EHR, thereby reducing the time spent interpreting data from a variety of sources.10 There are several examples of payer participation in HIEs. A noteworthy example is in Nebraska, where Coventry and Blue Cross Blue Shield of Nebraska (BCBSNE) are participating in the Nebraska Health Information Initiative (NEHII). NEHII’s area of focus is to lower costs, improve timeliness and reliability, and mitigate security risks associated with the exchange of clinical data via facsimile and paper-based methods for medical and drug authorization and care management activities.11 One of their programs provides external medication history to clinicians during the medication reconciliation process at transitions of care. BCBSNE provided the seed money and actively engaged in governance, business planning, and construction of infrastructure. Health plans currently pay a permember-per-year fee to the HIE. HIT-Enabled Pharmacist Involvement in Team Care The rise of new integrated approaches to the delivery of care— such as accountable care organizations (ACOs) and patientcentered medical homes—will create new roles and responsibilities for managed care pharmacists. The result: increased care coordination facilitated by HIT and a robust HIE infrastructure. For example, pharmacists will play key roles in the care team, which will be facilitated by HIT. According to an AMCP report, studies have demonstrated that pharmacists participating in team-based care models have made positive contributions to patient care and safe medication use with the help of HIT.12 Recently, California passed legislation that would expand this access to participation in care teams for pharmacists in an effort to help address shortages among other health care professions.13 As a clinical expert working as part of an interdisciplinary team, pharmacists can assess whether medication use is contributing to unwanted effects and can help achieve desired outcomes from medication use. Ideally, pharmacists will be considered “eligible providers” and thus entitled to additional compensation if meaningful use metrics are achieved.14 The pharmacist’s role in patient-centered team-based care can be enhanced by integration of prescription information with other sources of clinical information.15 These sources include EHRs, disease registries, and patients themselves, who can access and provide data through Web portals, personal health records, and mobile health applications. Electronic exchange of these data through a robust HIE infrastructure will be essential. The primary care team may be in the best position to coordinate a patient’s care, but often it will need information from other providers, including ambulatory providers. This need for coordination and data sharing will lead to increased reliance on EHRs and HIEs in managed care organizations— for e-prescribing, patient care, and to prevent readmissions. These capabilities will put managed care pharmacists in a better position to work with their patients and care team to make informed decisions about medication therapy options.16 Vol. 20, No. 11 www.amcp.org The Impact of Information Technology on Managed Care Pharmacy: Today and Tomorrow Pharmacists in managed care arrangements also will play a vital role in their organizations by leveraging HIT to provide patient care services and MTM. This includes comprehensive medication reviews, medication reconciliation, drug utilization review, ordering and reviewing lab tests, immunizations, drug-dosage adjustments, and identifying gaps in care.17 These programs are essential to care coordination by improving medication adherence, managing where and when care is delivered, and improving patient outcomes. Pharmacists have been providing MTM services for Medicare patients, and the model is being expanded to include similar services for non-Medicare patients. Some clinics are expanding the services to a commercial population, offering medication reconciliation postdischarge. Electronic Prior Authorization Prior authorization (PA) involves getting permission from the patient’s health insurer before a medication therapy can begin. It is a complex, multistep process, which many view as a necessary part of the health care system to help ensure that patients receive medication therapies that are safe, appropriate, and cost-effective. Going forward, PA will become an even bigger part of managed care pharmacists’ workflow, and newer tools, such as electronic prior authorization (ePA) submission, will be an essential workplace tool. Factors influencing the adoption of ePA include labor costs, workflow productivity and efficiency, and enhanced performance metrics and quality measurements. Plans are expected to require more PAs in general and for the specialty medications needed by increasing numbers of the elderly and chronically ill patients that will be in their panels. Growth in spending on specialty drugs is far outpacing spending on traditional drugs, and many new specialty pharmaceuticals are in the pipeline. The 2013 Express Scripts Drug Trend Report projects that specialty drug spending will jump by 67% by 2015, and nearly half of all prescription drug sales will be for specialty medications by 2016.18 In the future, managed care pharmacists increasingly will be moving away from yesterday’s phone-fax-paper processes toward ePA. At the point of prescribing, ePA will create value for pharmacies by eliminating the responsibility of the pharmacist to facilitate the PA process. This will allow pharmacists to focus on patients and revenue-generating activities. In addition, ePA will help PAs get approved more quickly, thus helping to reduce abandoned prescriptions due to patient frustration with the process, the number of rejected claims, and increase patient satisfaction. HIT can also enhance the ability to comply with the U.S. Food and Drug Administration’s (FDA) Risk Evaluation and Mitigation Strategies (REMS), which are needed for many specialty medications.19 REMS are structured plans to manage specific risks of drugs that are effective but associated with known or potential risks, such as death or injury. REMS help the FDA, drug manufacturers, and prescribers make sure that the benefits of such drugs outweigh their risks. The changes in PA will be facilitated by the availability of ePA standard transactions. NCPDP has developed a new ePA standard transaction for products covered by patients’ phar- www.amcp.org Vol. 20, No. 11 macy benefits. Based on SCRIPT—named by the Medicare Drug Improvement and Modernization Act as the standard for e-prescribing—ePA is envisioned to occur during the e-prescribing process. The new transaction allows the provider to request a PA question set from the payer, return the answers, and receive a response, all electronically (potentially in near real time). Questions may be customized, depending on the patient and the medication involved, and clinical attachments, such as subsets of the medical record, also are supported. With the widespread adoption of EHR technology, much of the information to be exchanged can be system-driven, reducing the burden of manually entering and reviewing the data. In addition, ePA will be required increasingly by states. They have recognized the value of an ePA process and are moving forward by statute and regulation. For example, Minnesota Laws, Chapter 336, Sec. 5, requires that “no later than January 1, 2016, drug prior authorization requests must be accessible and submitted by health care providers, and accepted by group purchasers, electronically through secure electronic transmissions. Facsimile shall not be considered electronic transmission.”20 In the short term, many states are considering use of a standardized PA form, which can be digitized later as standards and interoperability adoption increase. Get Ready for e-Health Telehealth and mobile health (m-health) are rapidly picking up traction in today’s health care environment.21,22 Growth in these areas will be partially driven by the expansion of the number of insured individuals and the limited number of health care providers available to meet the need. The impact will increasingly be felt by managed care and managed care pharmacy as ways to enable patient visits, consultations, medication adherence, and remote monitoring. This is especially important for the elderly and chronically ill—key populations for many managed care organizations, such as Medicaid managed care and Medicare ACOs. Telehealth patients in the United States are expected to rise nearly 6-fold by 2017, and telehealth revenue is set to rise to $707.9 million in 2017.23 The impact is already being felt through lower costs and fewer hospitalizations and readmissions.24 A study conducted for California and its Medicaid program, Medi-Cal, concluded that telemedicine used for “home monitoring for chronic diseases [such as] heart failure and diabetes . . . has the potential to produce savings to the Medi-Cal program of as much as several hundred million dollars annually.”25 It reported a 42% reduction in costs related to heart failure care and a 9% reduction in costs related to diabetes care. At the same time, m-health applications (i.e., “apps”) are skyrocketing—mostly because apps and mobile devices empower consumers to take charge of their own treatment and create more effective communications with providers and pharmacists, which will help bring down costs.26 There are already 100,000 health applications available in app stores, and the top 10 m-health applications are expected to generate up to 4 million free and 300,000 paid daily downloads.27 November 2014 JMCP Journal of Managed Care & Specialty Pharmacy 1075 The Impact of Information Technology on Managed Care Pharmacy: Today and Tomorrow What does this mean for managed care pharmacy? Both telehealth and m-health hold opportunities to provide highquality pharmacy services and follow-up care as well as engage patients. For example, telehealth consultations may be carried out by managed care pharmacists through clinical software in the cloud.28 This kind of interaction may be especially useful in rural areas and for homebound, chronically ill patients. These technologies can help lower costs by facilitating the delivery of care and connecting people to their health care providers.29 As plans increase their membership among a younger demographic, they will also demand the same access to services via technology as they are accustomed to receiving from other industries. M-health apps can also help patients order medication refills and inform the care process by electronically providing information on their regimens and clinical status. Patients can track their own medication administration activity, similar to the inpatient medication administration record, as well as tracking physical activity and nutrition information. M-health can provide tools to help patients manage their care and improve medication adherence.28 Similarly, m-health can be a tool for managed care pharmacists to establish relationships with patients and educate them on the importance of medication adherence. Such technology-enabled interactions can make a positive impact on patient outcomes.30 For example, alerts can be pushed out to patients to schedule visits and lab tests and to remind them to take particular medications or alert them to potential interactions. Pharmacists will be able to use m-health in their day-to-day operations. Apps have been developed for drug references, clinical references, medical calculators, laboratory references, news and continuing medical education, and productivity. Having information at their fingertips can improve their relationships with patients and other members of the care team. Addressing Prescription Drug Abuse Despite progress during the last several years, prescription drug abuse and fraud remain significant problems for all health plans and providers. The Centers for Disease Control and Prevention has declared prescription drug abuse a national epidemic that costs 20,000 lives and $72 billion dollars a year.31 More than 2.4 million people were considered opioid abusers in 2010. The number is growing, particularly among the elderly. Statistics show that seniors account for an increasing proportion of unintentional substance abusers, and they are a major population served by managed care pharmacists.32 Increasingly, HIT is being used innovatively to address substance abuse and drug diversion. For example, e-prescribing can help clinicians recognize substance abuse through medication history checks, which show controlled and noncontrolled medications that were filled by the patient. E-prescribing systems, as well as pharmacy systems, can also flag potentially deadly prescription errors and drug interactions related to opioid use, thus, preventing accidental deaths and overdoses. Renewal request monitoring can help flag abuse and diversion.33 In addition, as of June 2013, 47 states have implemented 1076 Journal of Managed Care & Specialty Pharmacy JMCP November 2014 a prescription drug monitoring program (PDMP).34 The typical PDMP program is voluntary for prescribers and consists of electronic databases that collect, monitor, and analyze prescribing and dispensing data sent electronically from pharmacies and dispensing practitioners. Massachusetts, New York, and Vermont, however, now require that prescribers check their PDMP state databases prior to prescribing controlled medications. New York was the first state to mandate that prescribers consult a PMDP prior to prescribing Schedule II, III, and IV controlled substances. Data Analytics One of today’s biggest trends is “big data,” that is, enhanced data analytics using new tools, more sophisticated analysis techniques, and the sharing of expanded datasets from multiple sources beyond traditional claims data. To be sure, pharmacy systems have been retrospectively mining their own claims data for years. While that will be important, pharmacies in the future will be looking to enhanced data analytics and data sharing in 3 areas: improved patient care; better benefit management; and fighting fraud, waste, and abuse. One way to improve care and patient outcomes is to put the results of data analytics into the hands of pharmacists and other providers at the point of care. Data from EHRs—including those connected to an HIE with access to other EHRs and claims data—will provide the basis for new work flows and actionable information at the point of care. The result: better care coordination by “pulling” the patient into the system before conditions worsen, become even more costly, or potentially adverse events occur, such as an emergency room visit or deadly drug interactions. As an example, a pharmacy team associated with the University of Massachusetts’ Medical School analyzed data elements from various sources—including pharmacy records and medical claims—to identify patterns that identify prescriber outliers, evaluate member outcomes, and quantify hospitalization rates.35 Results of the analyses also are used to develop recommendations that improve patient outcomes and reduce short- and long-term health care costs, such as listing medication classes where generic usage can be increased, predicting the effect of formulary restrictions on cost savings, and assessing diagnosis data specific to the member population.36 Analysis of data from multiple sources and platforms can provide more immediately actionable information beyond what can be determined from pharmacy claims analysis alone. In 1 mail order pharmacy, for example, hundreds of thousands of customer service logs were analyzed. Results detected a spike in calls between days 75 and 105 of some patients’ medication regimens. Looking closer, analysts found that the calls correlated with refill dates, and they discovered that some customers were calling for refills because their medications were taken with variable dosages. To reduce the number of lengthy customer service calls and expensive “emergency” refills and rush orders, the pharmacy began asking patients how many pills they had remaining at day 30 and day 60, so that they could better predict when the medication would run Vol. 20, No. 11 www.amcp.org The Impact of Information Technology on Managed Care Pharmacy: Today and Tomorrow out.37 Express Scripts is taking a big data approach to identify consistent patterns and tailor the most effective interventions for those patients most at risk for medication nonadherence.38 At the same time, electronic sharing of pharmacy claims data, coupled with real-time submission by pharmacies, creates an opportunity for analytics to become an effective tool for pharmacy benefit management. For example, utilization trends can be quickly identified, which can lead to the development of clinical strategies and follow-up services that create value for clients.39 These might include formulary management, medication therapy management, and clinical decision support, all of which are enabled by HIT. Pharmacies also are using large-scale data analytics to identify individual cases of blatant abuse. Sophisticated algorithms and other techniques can be employed to examine data from multiple sources and highlight suspect activity or patterns of abuse. Express Scripts’ analytical models helped identify a husband-and-wife team that had obtained approximately 7,000 pills for controlled substances—a total worth of about $150,000—by using 17 doctors and pharmacies in different cities. They had signed multiple exclusivity contracts with doctors, stating they would only get narcotics prescriptions from those doctors.40 ■■ Challenges Presented by HIT in Managed Care Despite the growing business imperative for the routine sharing of health information between multiple stakeholders in the managed care environment, gaps and challenges remain. Interoperability Interoperability describes the extent to which systems and devices can exchange data and interpret that shared data.41 In terms of connectivity, nursing homes, home health providers, and other postacute care and community-based providers lag behind other sites of care. Increased interoperability also will be challenging given the expected rise of Medicaid managed care organizations, since 85% of Medicaid enrollees will be in managed care organizations by 2020.42 Medicaid health IT systems tend to lag behind those in the private sector, which could potentially hinder the accessibility and sharing of necessary pharmacy and related clinical data as patients move in and out of Medicaid managed care plans. Standards and certified vendor solutions are needed. A particular need is the development of standards and enhancements to manage specialty prescriptions electronically, which NCPDP has begun to address. While nearly all retail and institutional pharmacies can receive and process e-prescriptions, specialty pharmacies are behind because of the additional information needed to support dispensing the medication (and related supplies and services). The lack of e-prescription connectivity of specialty pharmacies has implications for costs, quality, and patient safety. In fact, specialty medications are a significant and growing part of care regimens and drug spend. These costs in turn impact the financial success and risk sharing for ACOs and other managed care systems. www.amcp.org Vol. 20, No. 11 In addition, recent privacy violations have caused justifiable concerns about widespread sharing of protected health care information (PHI). While safeguards for PHI are covered under HIPAA, it will become increasingly important for covered entities to implement ongoing measures to help safeguard patient information as clinical information exchange becomes pervasive. Formulary Data Quality and System Support Although e-prescribing is becoming widespread, there appears to be limited use of its advanced features by physicians. These include identifying potential drug interactions and obtaining patient formulary information.43 Today, the availability and usefulness of medication history and formulary data is limited, which causes prescribers to question its reliability. Physicians in 1 study reviewed the formulary data only occasionally. Respondents observed that formulary information was inconsistently available, out-of-date, or inaccurate. Some respondents expressed a desire to see additional information, such as a patient’s actual copayment amounts, formulary alternatives for off-formulary medications, or whether a medication requires PA. More than two-thirds of the practices responding believed that the volume of formulary-related pharmacy callbacks was still burdensome because their e-prescribing solutions did not alert them to the need for PAs after the e-prescriptions were sent, resulting in inefficiencies for both parties.44 Alert Fatigue Part of the challenge of increasing use of available information, including clinical decision support (CDS), is alert fatigue. Alert fatigue is a well-known problem where a continuing large volume of notifications lead the user to become insensitive to the information presented.45 Some users or implementations will forgo full activation of CDS, or set alert thresholds high, to reduce the potential for alert fatigue. Balancing CDS information versus alert fatigue and patient safety is one of the conundrums of utilizing technology in health care. ■■ Policy Levers Are Key The federal government is aware of these issues and, according to the Office of the National Coordinator for HIT (ONC), will use policy levers to help address them.46 One possibility is the integration of HIT into federal payment policies. For example, the use of interoperable EHR systems to share information could eventually become part of reimbursement criteria in the Medicare and Medicaid programs.47 HIT adoption has been furthered by the Meaningful Use incentive program. While the incentives may not be available in the future because of budgetary constraints, requirements for HIT adoption and use are likely to continue. Meaningful Use Stage 2 is heavy on HIE requirements, while Stage 3 is expected to include requirements for advanced HIE and increased clinical detail.48 At some point in the near future, the federal government will be asked to issue guidance concerning the use of NCPDP’s new ePA standard transaction. To promote increased interoperability, the ONC is convening the HIT community to prioritize HIT challenges and November 2014 JMCP Journal of Managed Care & Specialty Pharmacy 1077 The Impact of Information Technology on Managed Care Pharmacy: Today and Tomorrow subsequently enable development and harmonization of related standards, specifications, and implementation guidance.49 Moreover, the ONC has developed a national HIE strategy that will help frame policy discussions and drive initiatives with federal backing. Efforts will continue by the private sector and public/private initiatives to address connectivity and interoperability issues. An example is the CommonWell Health Alliance, which is a group of health care stakeholders that have banded together to define and promote a national infrastructure for interoperability with common standards and policies for EHRs.50 Quest Diagnostics and Surescripts recently announced an agreement to pioneer the formation of an integrated service to make laboratory and prescription information broadly and easily accessible to prescribers. Standards development organizations will continue to create new technical standards and specifications for data exchange and interoperability. On the state level, legislatures and boards of pharmacy are expected to be addressing HIT-related issues. There are gaps in e-prescribing of controlled substance (EPCS) laws and regulations at the state level that are likely to be addressed. As of the publication of this article, EPCS is legal in 48 states and the District of Columbia.4 Clarity and synergy in state laws would eliminate a barrier to EPCS adoption with the added benefits of improving prescribers’ workflows, protecting against drug diversion, and increasing patient safety. Change can be difficult. The successful transition to electronic systems and workflows must be understood, championed, and managed. Managed care organizations must support development and use of HIT and HIE, including in their pharmacies. Pharmacists must be included as leaders championing and motivating staff to participate in the change process and to understand benefits in implementations and workflows. ■■ Conclusions Managed care pharmacy professionals must understand the value and effective use of HIT solutions, if they are to have an impact on how the future develops. Those who do not recognize and prepare for these changes will not only miss an opportunity, but will be at a competitive disadvantage in the marketplace. Managed care pharmacy professionals will look to AMCP to innovate and champion the development of HIT applications and disseminate their benefits—all of which are a value-add to the membership. As such, we urge AMCP to maintain the Healthcare Information Technology Advisory Council’s charter, which expired early in 2014. Without the assistance of the Advisory Council, AMCP leaders and members will have to ferret out such information on their own and try to interpret it in the context of their own business models and solutions. This is a difficult and daunting task, which can lead to costly mistakes in strategic positioning, HIT investments, and patient care. 1078 Journal of Managed Care & Specialty Pharmacy JMCP November 2014 Authors DOUGLAS HILLBLOM, PharmD, is Vice President, Professional Practice and Pharmacy Policy, OptumRx, Sacramento, California; ANTHONY SCHUETH, MS, is CEO and Managing Partner, Point-of-Care Partners, LLC, Coral Springs, Florida; SCOTT M. ROBERTSON, RPh, PharmD, is Principal Technology Consultant, Kaiser Permanente, Pasadena, California; LAURA TOPOR, BA, is President, Granada Health, Minnetonka, Minnesota; and GREG LOW, RPh, BSPharm, PhD, is Program Director, MGPO Pharmacy, Massachusetts General Hospital, Boston, Massachusetts. AUTHOR CORRESPONDENCE: Douglas Hillblom, PharmD, Vice President, Professional Practice and Pharmacy Policy, OptumRx, 8880 Cal Center Dr., Ste. 300, Sacramento, CA 95826. Tel.: 916.403.0703; Fax: 954.212.0200; E-mail: [email protected]. DISCLOSURES The authors are members of a subgroup of the AMCP Healthcare Information Technology Advisory Council. They declare no other potential conflicts of interest. REFERENCES 1. Academy of Managed Care Pharmacy. HIT primer. 2011. Available at: http://www.amcp.org/HITPrimer. 2. National Council for Prescription Drug Programs. Pharmacy: a prescription for improving the healthcare system. White paper. September 2009. Available at: http://www.ncpdp.org/Education/Whitepaper?page=4. Accessed August 27, 2014. 3. Agency for Healthcare Quality and Research. Health information technology: overview. Program brief. AHRQ Pub. No. 07-P006. February 2007. Available at: http://healthit.ahrq.gov/sites/default/files/docs/page/hitover.pdf. Accessed August 27, 2014. 4. Surescripts. 2013 national progress report and Safe-Rx rankings. May 2014. Available at: http://surescripts.com/news-center/national-progress-report2013. Accessed August 27, 2014. 5. Point-of-Care Partners. Building the foundation for an effective and efficient health care system: lessons learned in southeastern Michigan. White paper. April 2013. Available at: http://www.pocp.com/publications/semi-hitlessons-learned-wp.pdf. Accessed August 27, 2014. 6. Gunter TD, Terry NP. The emergence of national electronic health record architectures in the United States and Australia: models, costs, and questions. J Med Internet Res. 2005;7(1):e3. 7. Office of the National Coordinator for Health Information Technology. HealthIT dashboard. Fast facts about health IT adoption in health care. Available at: http://dashboard.healthit.gov/quickstats/. Accessed August 27, 2014. 8. Reed M, Huang J, Brand R, et al. Implementation of an outpatient electronic health record and emergency department visits, hospitalizations and office visits among patients with diabetes. JAMA. 2013;310(10):1060-65. 9. Agency for Healthcare Quality and Research. Trends in health information exchanges. Updated March 26, 2014. Available at: http://www.innovations.ahrq.gov/content.aspx?id=3944. Accessed August 27, 2014. 10. Office of the National Coordinator for Health Information Technology. The value of health information exchange from a payer’s perspective–a toolkit for HIE leaders. October 2013. Available at: http://www.pocp.com/Value%20of%20 HIE%20Payor%20Toolkit%20v10%20130926.pdf. Accessed August 27, 2014. 11. Nebraska Information Technology Commission. Nebraska operational eHealth plan. Version 7. May 2013. Available at: http://nitc.ne.gov/ehealth_ council/documents/Nebraska%20eHealth%20Operational%20Plan%20 May%206%202013.pdf. Accessed August 27, 2014. 12. Academy of Managed Care Pharmacy. Pharmacists as vital members of accountable care organizations. Updated March 2012. Available at: http:// www.amcp.org/ACO.pdf/. 13. California Legislative Information. SB-493: Pharmacy practice. 2013. Available at: http://leginfo.legislature.ca.gov/faces/billTextClient.xhtml?bill_ id=201320140SB493. Accessed August 27, 2014. Vol. 20, No. 11 www.amcp.org The Impact of Information Technology on Managed Care Pharmacy: Today and Tomorrow 14. Centers for Medicare and Medicaid Services. 2014 definition stage 1 of meaningful use. Updated July 18, 2014. Available at: http://www.cms.gov/ Regulations-and-Guidance/Legislation/EHRIncentivePrograms/Meaningful_ Use.html. Accessed August 27, 2014. 15. Pharmacy e-Health Information Technology Collaborative. The roadmap for pharmacy health information technology integration in U.S. health care. 2011. Available at: http://www.accp.com/docs/positions/misc/ HITRoadMap2011.pdf. Accessed August 27, 2014. 16. National Coordinator for Health Information Technology. Why adopt EHRs? 2012. Available at: http://www.healthit.gov/providers-professionals/ why-adopt-ehrs. Accessed August 27, 2014. 17. Edlin M. Pharmacists offer MTM services to support ACOs. Managed Healthcare Executive. April 1, 2013. Available at: http://managedhealthcareexecutive.modernmedicine.com/managed-healthcare-executive/ news/user-defined-tags/pharmacist/pharmacists-offer-mtm-servicessuppor?contextCategoryId=39. Accessed August 27, 2014. 18. Stettin G. Specialty drug spending to jump 67% by 2015. Prescription Drug Trends. May 21, 2013. Available at: http://lab.express-scripts.com/ prescription-drug-trends/specialty-drug-spending-to-jump-67-by-2015/. Accessed August 27, 2014. 19. U.S. Food and Drug Administration. FDA basics webinar: a brief overview of risk evaluation and mitigation strategies. June 2014. Available at: http://www.fda.gov/aboutfda/transparency/basics/ucm325201.htm. Accessed August 27, 2014. 20. Minnesota Office of the Revisor of Statutes. 62J.497 Electronic prescription drug program. 2013. Available at: https://www.revisor.mn.gov/ statutes/?id=62j.497. Accessed August 27, 2014. 21. Health Resources and Services Administration, Rural Health. Telehealth. 2012. Available at: http://www.hrsa.gov/ruralhealth/about/telehealth/. Accessed August 27, 2014. 22. Health Information Management Systems Society. Definitions of mHealth. January 5, 2012. Available at: http://www.himss.org/ ResourceLibrary/GenResourceDetail.aspx?ItemNumber=20221. Accessed August 27, 2014. 23. Walker S. Telehealth−an analysis of demand dynamics−world−2012. Catalogue number: 2396. InMedica. 2012. Available at: http://www.imsresearch.com/report/The_World_Market_for_Telehealth_An_Analysis_of_ Demand_Dynamics__2012_Edition. Accessed August 27, 2014. 24. Broderick A, Lindeman D. Scaling telehealth programs: lessons from early adopters. The Commonwealth Fund. January 2013. Available at: http://www.mtelehealth.com/pdf/studies/Case_Studies_in_Telehealth_ Adoption_-_Scaling_Telehealth_Programs_-_Lessons_from_Early_ Adopters_-_2013-01-30.pdf. Accessed August 27, 2014. 25. Newman M, McMahon T. Fiscal impact of AB 415: potential cost savings from expansion of telehealth. Blue Sky Consulting Group. September 30, 2011. Available at: http://cchpca.org/sites/default/files/Fiscal%20 Impact%20of%20AB%20415%20Potential%20Cost%20Savings%20from%20 Expansion%20of%20Telehealth_0_0.pdf. Accessed August 27, 2014. 26. Valencia R. Empowering consumers to take charge of their health: leveraging the power of mobile to deliver personalized health care. HIMSS News. March 11, 2013. Available at http://www.himss.org/News/NewsDetail. aspx?ItemNumber=17255. Accessed August 27, 2014. 27. Workman B. The obsession with medical costs will turn mobile health apps and devices into a major growth industry. Business Insider. November 6, 2013. Available at: http://www.businessinsider.com/mobile-will-be-thefuture-of-health-care-2013-11#ixzz2jya7qa9k. Accessed August 27, 2014. 28. Klauson KA, Elrod S, Fox BI, Hajar Z, Dzenowagis JH. Opportunities for pharmacists in mobile health. Am J Health Syst Pharm. 2013;70(15):1348-52. 29. Javitt JC. mHealth tools: a bond between patient and pharmacists. Drug Topics. December 10, 2013. Available at: http://drugtopics.modernmedicine.com/drug-topics/news/mhealth-tools-bond-between-patients-andpharmacists?page=full. Accessed August 27, 2014. 30. U.S. Department of Health and Human Services. Health Resources and Services Administration. Using health text messages to improve consumer health knowledge, behaviors, and outcomes: an environmental scan. Rockville, MD: U.S. Department of Health and Human Services, 2014. Available at: http://www.hrsa.gov/healthit/txt4tots/environmentalscan.pdf. Accessed September 3, 2014. www.amcp.org Vol. 20, No. 11 31. Centers for Disease Control and Prevention. CDC grand rounds: prescription drug overdoses−a U.S. epidemic. MMWR Morb Mortal Wkly Rep. 2012;61(1):10-13. 32. Friedman RA. The rising tide of substance abuse. The New York Times. April 29, 2013. Available at: http://newoldage.blogs.nytimes.com/2013/04/ 29/a-rising-tide-of-mental-distress/?_php=true&_type=blogs&_r=0. Accessed August 27, 2014. 33. Point-of-Care Partners. Moving e-prescribing forward. A thought paper prepared for the National Coordinator for Health Information Technology, U.S. Department of Health and Human Services. April 2012. 34. U.S. Department of Health and Human Services. Prescription drug monitoring program interoperability standards: a report to Congress. September 2013. Available at: http://www.healthit.gov/sites/default/files/fdasia1141report_final.pdf. Accessed August 27, 2014. 35. University of Massachusetts Medical School. Commonwealth medicine. Pharmacy. Available at: http://commed.umassmed.edu/services/pharmacy. Accessed August 27, 2014. 36. University of Massachusetts Medical School. Commonwealth medicine. Data analytics. Available at: http://commed.umassmed.edu/services/pharmacy/data-analytics. Accessed August 27, 2014. 37. Wegener R, Sinha V. The value of big data: how analytics differentiates winners. Bain & Company. September 17, 2013. Available at: http://www.bain.com/ publications/articles/the-value-of-big-data.aspx. Accessed August 27, 2014. 38. Express Scripts. Insights. Making big data actionable. October 10, 2013. Available at: http://lab.express-scripts.com/industry-updates/making-bigdata-actionable/#sthash.BHk8R9Kg.dpuf. Accessed August 27, 2014. 39. Groves P, Kayyali B, Knott D, Van Kuiken S. The ‘big data’ revolution in healthcare: accelerating value and innovation. McKinsey and Company. January 2013. Available at: http://www.mckinsey.com/~/media/mckinsey/ dotcom/client_service/healthcare%20systems%20and%20services/pdfs/ the_big_data_revolution_in_healthcare.ashx. Accessed August 27, 2014. 40. Express Scripts. Insights. Rx addiction: one family’s 7,000 pills. February 1, 2013. Available at: http://lab.express-scripts.com/insights/drug-safety-andabuse/rx-addiction-one-familys-7000-pills. Accessed August 27, 2014. 41. Health Information Management and Systems Society. What is interoperability? 2013. Available at: http://www.himss.org/library/interoperabilitystandards/what-is. Accessed August 27, 2014. 42. Sullivan K. Medicaid has a managed care future. AJMC.com. October 22, 2013. Available at: http://www.ajmc.com/focus-of-the-week/1013/Medicaidhas-a-Managed-Care-Future. Accessed August 27, 2014. 43. Grossman J. Even when physicians adopt e-prescribing, use of advanced features lag. Issue Brief Cent Stud Health Syst Change. 2010;(133):1-5. 44. Grossman JM, Boukus ER, Cross DA, Cohen GR. Physician practices, e-prescribing and accessing information to improve prescribing decisions. Res Brief. 2011;(20):1-10. Available at: http://www.hschange.com/ CONTENT/1202/#ib4. Accessed August 27, 2014. 45. Curtiss F, Fairman K. Quality improvement opportunities in prescriber alert programs. J Manage Care Pharm. 2010;16(4):292-96. Available at: http:// www.amcp.org/WorkArea/DownloadAsset.aspx?id=8330. 46. Raths D. Live from state HIT Summit: beyond meaningful use—policy levers to promote interoperability. HealthInformatics. April 1, 2014. Available at: http://www.healthcare-informatics.com/article/live-state-hitsummit-beyond-meaningful-use-policy-levers-promote-interoperability. Accessed August 27, 2014. 47. Office of the National Coordinator for Health Information Technology. Principles and strategies for accelerating HIE. 2012. Available at: http:// www.cms.gov/ehealth/downloads/Accelerating_HIE_Principles.pdf. Accessed August 27, 2014. 48. Office of the National Coordinator for Health Information Technology. How to attain meaningful use. 2013. Available at: http://www.healthit.gov/providers-professionals/how-attain-meaningful-use. Accessed August 27, 2014. 49. Office of the National Coordinator for Health Information Technology. Health information exchange. Standards and interoperability. 2014. Available at: http://www.healthit.gov/providers-professionals/standardsinteroperability. Accessed August 27, 2014. 50. Commonwell Health Alliance. Interoperability for the common good. Available at: http://www.commonwellalliance.org/. Accessed August 27, 2014. November 2014 JMCP Journal of Managed Care & Specialty Pharmacy 1079 Introducing Jardiance ® (empagliflozin) tablets 10 mg/25 mg JARDIANCE is an SGLT2 inhibitor for the treatment of adults with type 2 diabetes, in addition to diet and exercise • Significant A1C reduction • Once-daily oral dosing • Additional benefit of weight loss* * JARDIANCE is not indicated for weight loss. Weight change was a secondary endpoint in clinical trials.1 INDICATION AND LIMITATION OF USE JARDIANCE is indicated as an adjunct to diet and exercise to improve glycemic control in adults with type 2 diabetes mellitus. JARDIANCE is not recommended for patients with type 1 diabetes or for the treatment of diabetic ketoacidosis. IMPORTANT SAFETY INFORMATION CONTRAINDICATIONS JARDIANCE should not be used in patients with a history of serious hypersensitivity to JARDIANCE or in patients with severe renal impairment, end-stage renal disease, or dialysis. JARDIANCE is proven to significantly reduce A1C In addition to lowering A1C, JARDIANCE significantly reduced weight† JARDIANCE monotherapy vs placebo (24 weeks) JARDIANCE is not indicated for weight loss. Weight change was a secondary endpoint.1 † A1C REDUCTION 0.4 (24 weeks) Mean baseline=7.9% 0.2 WEIGHT CHANGE (N=228) (24 weeks) Mean baseline=172 lb (N=224) -0.2 Weight (%) change from baseline (adjusted mean) A1C (%) change from baseline (adjusted mean) 0.0 (N=224) 0.1% -0.4 -0.6 -0.8 -0.7 -1.0 ■ JARDIANCE 10 mg % -0.8 % p <0.0001 vs placebo‡ ■ JARDIANCE 25 mg (N=224) 0.0 ‡ A1C reduction: Difference from placebo (adjusted mean) was -0.7% and -0.9% for JARDIANCE 10 mg and 25 mg, respectively. Weight change: Difference from placebo (adjusted mean) was -2.5% and -2.8% for JARDIANCE 10 mg and 25 mg, respectively. § Study design: In a 24-week, double-blind, placebocontrolled study of 676 patients with type 2 diabetes mellitus, the efficacy and safety of JARDIANCE 10 mg (N=224) and 25 mg (N=224) were evaluated vs placebo (N=228). The primary endpoint was A1C change from baseline.1 JARDIANCE 10 mg and 25 mg significantly reduced systolic blood pressure (SBP)II by -2.6 mm Hg (placebo-adjusted, p=0.0231) and -3.4 mm Hg (placebo-corrected, p=0.0028), respectively, at 24 weeks1¶ (N=224) (N=228) -0.4% -1.0 -2.0 -3.0 -4.0 -2.8%-3.2% -5.0 ■ Placebo IMPORTANT SAFETY INFORMATION (continued) p <0.0001 vs placebo§ JARDIANCE 10 mg JARDIANCE 25 mg Placebo WARNINGS AND PRECAUTIONS Hypotension JARDIANCE causes intravascular volume contraction. Symptomatic hypotension may occur after initiating JARDIANCE particularly in patients with renal impairment, the elderly, in patients with low systolic blood pressure, and in patients on diuretics. Before initiating JARDIANCE, assess for volume contraction and correct volume status if indicated. Monitor for signs and symptoms of hypotension after initiating therapy. Impairment in Renal Function JARDIANCE increases serum creatinine and decreases eGFR. Renal function should be evaluated prior to initiating JARDIANCE and periodically thereafter. More frequent monitoring is recommended with eGFR below 60 mL/min/1.73 m2. The risk of impaired renal function with JARDIANCE is increased in elderly patients and patients with moderate renal impairment. JARDIANCE should be discontinued in patients with a persistent eGFR less than 45 mL/min/1.73 m2. JARDIANCE is not indicated as antihypertensive therapy. Blood pressure (BP) change was a secondary endpoint.1 II ¶ SBP mean baseline: 133.0 mm Hg, 129.9 mm Hg, and 130.0 mm Hg for JARDIANCE 10 mg, 25 mg, and placebo, respectively.1 Please see additional Important Safety Information and Brief Summary of full Prescribing Information on the adjacent pages. Learn more at www.Jardiance.com IMPORTANT SAFETY INFORMATION (continued) 75C, 63M, 63Y 5 4 3 75K 95 5 95 5 95 5 25 25 75 M+Y 25 50K 50C, 39M, 39Y 25K 25C, 16M, 16Y 25 25 25 75 25 25 25 25 M+Y 50 25 95 5 95 5 95 5 95 5 80K, 80C, 70M, 70Y 300% 4 96 4 96 4 4 96 96 1 2 C+Y 50 50 50 50 3 97 3 97 3 97 3 97 25 75 50 50 50 50 C+Y 50 95 96 96 4 96 4 96 4 97 3 97 3 97 3 75 50 25 97 98 99.5 99 1 0.5 2 98 99.5 99 1 0.5 2 98 99.5 99 1 0.5 2 98 99 99.5 25 50 75 C+M 75 75 75 75 GATF/SWOP Digital Proofing Bar 2 0.5 1 99 99.5 98 2 0.5 1 99 99.5 98 2 1 0.5 99.5 98 2 99 1 0.5 98 99.5 99 25 50 75 C+M 75 75 75 JARPROFISI 8.2.14 JARPROFISI 8.2.14 75 Genital Mycotic Infections WARNINGS AND PRECAUTIONS (continued) WARNINGS AND PRECAUTIONS (continued) JARDIANCE increases the risk for genital mycotic infections. Patients with a history of chronic or Genital Mycotic Infections Genital Mycotic Infections recurrent genital mycotic infections were more likely to develop these infections. Monitor and treat JARDIANCE increases the risk forfor genital mycotic infections. Patients with a history ofof chronic oror JARDIANCE increases the risk genital mycotic infections. Patients with a history chronic as appropriate. recurrent genital mycotic infections were more likely toto develop these infections. Monitor and treat recurrent genital mycotic infections were more likely develop these infections. Monitor and treat asUrinary appropriate. as appropriate. Tract Infections JARDIANCE increases the risk for urinary tract infections. Monitor and treat as appropriate. Urinary Tract Infections Urinary Tract Infections JARDIANCE increases the risk forfor urinary tract infections. Monitor and treat asas appropriate. JARDIANCE increases the risk urinary tract infections. Monitor and treat appropriate. Increased Low-Density Lipoprotein Cholesterol (LDL-C) Increases in LDL-C can occur with JARDIANCE. Monitor and treat as appropriate. Increased Low-Density Lipoprotein Cholesterol (LDL-C) Increased Low-Density Lipoprotein Cholesterol (LDL-C) Increases in LDL-C can occur with JARDIANCE. Monitor and treat asas appropriate. Increases in LDL-C can occur with JARDIANCE. Monitor and treat appropriate. Macrovascular Outcomes There have been no clinical studies establishing conclusive evidence of macrovascular risk reduction Macrovascular Outcomes Macrovascular Outcomes with JARDIANCE or any other antidiabetic drug. There have been nono clinical studies establishing conclusive evidence ofof macrovascular risk reduction There have been clinical studies establishing conclusive evidence macrovascular risk reduction with JARDIANCE oror any other antidiabetic drug. with JARDIANCE any other antidiabetic drug. ADVERSE REACTIONS The most common adverse reactions (>5%) associated with placebo and JARDIANCE 10 mg and 25 ADVERSE REACTIONS ADVERSE REACTIONS mg were urinary tract infections (7.6%, 9.3%, 7.6%, respectively) and female genital mycotic infections The most common adverse reactions (>5%) associated with placebo and JARDIANCE 1010 mgmg and 2525 The most common adverse reactions (>5%) associated with placebo and JARDIANCE and (1.5%, 5.4%, 6.4%, respectively). mgmg were urinary tract infections (7.6%, 9.3%, 7.6%, respectively) and female genital mycotic infections were urinary tract infections (7.6%, 9.3%, 7.6%, respectively) and female genital mycotic infections (1.5%, 6.4%, respectively). (1.5%, 5.4%, 6.4%, respectively). When5.4%, JARDIANCE was administered with insulin or sulfonylurea, the incidence of hypoglycemic events was increased. When JARDIANCE was administered with insulin oror sulfonylurea, the incidence ofof hypoglycemic events When JARDIANCE was administered with insulin sulfonylurea, the incidence hypoglycemic events was increased. was increased. DRUG INTERACTIONS Coadministration of JARDIANCE with diuretics resulted in increased urine volume and frequency of DRUG INTERACTIONS DRUG INTERACTIONS voids, which might enhance the potential for volume depletion. Coadministration ofof JARDIANCE with diuretics resulted in in increased urine volume and frequency ofof Coadministration JARDIANCE with diuretics resulted increased urine volume and frequency voids, which might enhance the potential forfor volume depletion. voids, which might enhance the potential volume depletion. USE IN SPECIAL POPULATIONS Pregnancy USE ININ SPECIAL POPULATIONS USE SPECIAL POPULATIONS There are no adequate and well-controlled studies of JARDIANCE in pregnant women. JARDIANCE Pregnancy Pregnancy should be used during pregnancy only if the potential benefit justifies the potential risk to the fetus. There are nono adequate and well-controlled studies ofof JARDIANCE in in pregnant women. JARDIANCE There are adequate and well-controlled studies JARDIANCE pregnant women. JARDIANCE should bebe used during pregnancy only if the potential benefit justifies the potential risk toto the fetus. should used during pregnancy only if the potential benefit justifies the potential risk the fetus. Nursing Mothers It is not known if JARDIANCE is excreted in human milk. Because of the potential for serious adverse Nursing Mothers Nursing Mothers reactions in nursing infants from JARDIANCE, discontinue nursing or discontinue JARDIANCE. It is not known if JARDIANCE is is excreted in in human milk. Because ofof the potential forfor serious adverse It is not known if JARDIANCE excreted human milk. Because the potential serious adverse reactions in nursing infants from JARDIANCE, discontinue nursing or discontinue JARDIANCE. reactions in nursing infants from JARDIANCE, discontinue nursing or discontinue JARDIANCE. Geriatric Use JARDIANCE is expected to have diminished efficacy in elderly patients with renal impairment. The Geriatric Use Geriatric Use incidence of volume depletion-related adverse reactions and urinary tract infections increased in JARDIANCE is is expected toto have diminished efficacy in in elderly patients with renal impairment. The JARDIANCE expected have diminished efficacy elderly patients with renal impairment. The patients ≥ 75 years treated with JARDIANCE. incidence of volume depletion-related adverse reactions and urinary tract infections increased in incidence of volume depletion-related adverse reactions and urinary tract infections increased in JARPROFISI 8.2.14 patients ≥≥ 7575 years treated with JARDIANCE. patients years treated with JARDIANCE. 0.5 WARNINGS AND PRECAUTIONS (continued) IMPORTANT SAFETY INFORMATION (continued) IMPORTANT SAFETY INFORMATION (continued) In adults with type 2 diabetes, In adults with type 2demonstrated diabetes, JARDIANCE similar A1C reduction vs glimepiride with the additional benefit of significant weight loss JARDIANCE demonstrated similar A1C reduction vs* glimepiride with the additional benefit of significant weight loss* A1C REDUCTION IN A 52-WEEK INTERIM ANALYSIS† Mean baseline=7.9%; 7.9% A1C REDUCTION IN A 52-WEEK INTERIM ANALYSIS† loss. Weight change was a secondary endpoint.1 *JARDIANCE is not indicated for weight loss. 1 design: In a 104-week, doubleStudy WEIGHT CHANGE IN A 52-WEEK ANALYSIS endpoint. Weight change was aINTERIM secondary § Mean baseline=7.9%; 7.9% Mean baseline=182 lb; 183 lb 0.00 WEIGHT CHANGE IN A 52-WEEK INTERIM ANALYSIS§ 3.0 -0.25 -0.25 -0.50 -0.50 -0.75 -0.75 4 Baseline 12 4 Baseline 12 28 40 52 Wk 52 40 52 p<0.0001 Wk 52 Weeks 28 Weeks JARDIANCE 25 mg + metformin (N=693) Glimepiride + metformin (N=700) JARDIANCE 25 mg + metformin (N=693) (mITT) (mITT) p<0.0001 WeightWeight (% change (% change from baseline) from baseline) (adjusted (adjusted mean) mean) ‡ ‡ A1C (%)A1C mean (%)change mean change from baseline from baseline (adjusted (adjusted mean) mean) 0.00 *JARDIANCE is not indicated for weight Mean baseline=182 lb; % 183 lb 2.0 3.0 1.0 2.0 0.0 1.0 -1.0 0.0 -2.0 -1.0 -3.0 -2.0 -4.0 -3.0 -5.0 -4.0 -5.0 2.0 2.0% Difference of 5.9% or 10.8 lbII Difference of 5.9% or 10.8 lbII -3.9% -3.9% p<0.0001 II JARDIANCE 25 mg + metformin (N=765) II p<0.0001 Glimepiride + metformin (N=780) JARDIANCE 25 mg + metformin (N=765) blind study of 1,545 patients with type 2 diabetes mellitus, the efficacy of Study design: In a 104-week, doubleJARDIANCE mg as add-onwith therapy blind study of25 1,545 patients type to metformin (N=765) evaluated 2 diabetes mellitus, thewas efficacy of vs glimepiride dose 2.7 mg) JARDIANCE(mean 25 mgdaily as add-on therapy added to metformin (N=780), to metformin (N=765) was evaluated vs administered oncedaily daily.dose 2.7 mg) glimepiride (mean †added to metformin (N=780), Completers only. ‡administered once daily. Mean change from baseline adjusted for baseline A1C, geographical Completers only. region, and eGFR at baseline. § ‡Modified intent-to-treat population (mITT). Last Mean change from baseline adjusted for baseline observation on study (LOCF) used to impute A1C, geographical region, andwas eGFR at baseline. §data missing at Week 52. Modified intent-to-treat population (mITT). Last # SBP mean baseline: 133.4 mmwas Hg and observation on study (LOCF) used to impute 133.5 mm Hg at forWeek JARDIANCE 25 mg and data missing 52. 1 #glimepiride, respectively. SBP mean baseline: 133.4 mm Hg and † 133.5 mm Hg for JARDIANCE 25 mg and glimepiride, respectively.1 Glimepiride + metformin (N=700) + metformin (N=780) ¶ JARDIANCE 25 mg significantly reduced SBPGlimepiride (-3.6 mm Hg) vs an increase with glimepiride (2.2 mm Hg) # at 52 weeks; adjusted mean, p<0.0001 JARDIANCE 25 mg significantly reduced SBP¶ (-3.6 mm Hg) vs an increase with glimepiride (2.2 mm Hg) ¶ JARDIANCE is not indicated as antihypertensive at 52 weeks; adjusted mean, p<0.0001# therapy. BP change was a secondary endpoint.1 ¶ • The recommended dose ofas JARDIANCE is 10 mg once daily. In patients JARDIANCE JARDIANCE is not indicated antihypertensive therapy. BP change was atolerating secondary endpoint.1 10 mg, the dose may be increased to 25 mg • The recommended dose of JARDIANCE is 10 mg once daily. In patients tolerating JARDIANCE 10 mg, the dose may be 1 • increased Primary endpoint to 25 mgwas A1C change from baseline after 52 weeks and 104 weeks. At 52 weeks, change from baseline (adjusted mean) was -0.7% with both JARDIANCE and glimepiride. Data at 104 weeks are not yet available • Primary endpoint was A1C change from baseline after 52 weeks and 104 weeks.1 At 52 weeks, change from baseline (adjusted mean) was -0.7% with both JARDIANCE and glimepiride. Data at 104 weeks are not yet available IMPORTANT SAFETY INFORMATION (continued) WARNINGS AND PRECAUTIONS (continued) (continued) IMPORTANT SAFETY INFORMATION Hypoglycemia with Concomitant Use with Insulin and Insulin WARNINGS AND PRECAUTIONS (continued) Secretagogues Hypoglycemia with Concomitant Use with Insulin and Insulin Insulin and insulin secretagogues are known to cause hypoglycemia. Secretagogues The use of JARDIANCE with these agents can increase the risk of Insulin and insulin secretagogues are known to cause hypoglycemia. hypoglycemia. A lower dose of insulin or the insulin secretagogue The use of JARDIANCE with these agents can increase the risk of may be required to reduce the risk of hypoglycemia when used in hypoglycemia. A lower dose of insulin or the insulin secretagogue combination with JARDIANCE. may be required to reduce the risk of hypoglycemia when used in combination with JARDIANCE. Please see additional Important Safety Information and Brief Summary of full Prescribing Information on the adjacent pages. Please see additional Important Safety Information and Brief Summary Reference: 1. Data on file. Boehringer Ingelheim Pharmaceuticals, Inc. Ridgefield, CT. 2014. of full Prescribing Information on the adjacent pages. Reference: 1. Data on file. Boehringer Ingelheim Pharmaceuticals, Inc. Ridgefield, CT. 2014. Copyright © 2014 Boehringer Ingelheim Pharmaceuticals, Inc. All rights reserved. (8/14) JAR622805MHC Copyright © 2014 Boehringer Ingelheim Pharmaceuticals, Inc. All rights reserved. (8/14) JAR622805MHC ® ® JARDIANCE (empagliflozin) tablets, for for oraloral useuse JARDIANCE (empagliflozin) tablets, b Female genital mycotic infections include the following adverse reactions: vulvovaginal mycotic infec-infecFemale genital mycotic infections include the following adverse reactions: vulvovaginal mycotic tion, tion, vaginal infection, vulvitis, vulvovaginal candidiasis, genital infection, genital candidiasis, genital vaginal infection, vulvitis, vulvovaginal candidiasis, genital infection, genital candidiasis, genital infection fungal, genitourinary tracttract infection, vulvovaginitis, cervicitis, urogenital infection fungal, vaginiBRIEF SUMMARY OF PRESCRIBING INFORMATION infection fungal, genitourinary infection, vulvovaginitis, cervicitis, urogenital infection fungal, vaginiBRIEF SUMMARY OF PRESCRIBING INFORMATION tis bacterial. Percentages calculated with the number of female subjects in each group as denominator: tis bacterial. Percentages calculated with the number of female subjects in each group as denominator: Please seesee package insert for for full full Prescribing Information. Please package insert Prescribing Information. placebo (N=481), JARDIANCE 10 mg JARDIANCE 25 mg placebo (N=481), JARDIANCE 10 (N=443), mg (N=443), JARDIANCE 25 (N=420). mg (N=420). c c adverse eventevent grouping, including, but not to, polyuria, pollakiuria, and nocturia Predefined adverse grouping, including, butlimited not limited to, polyuria, pollakiuria, and nocturia INDICATIONS ANDAND USAGE: JARDIANCE is indicated as an to diet and and exercise INDICATIONS USAGE: JARDIANCE is indicated as adjunct an adjunct to diet exercisedPredefined d MaleMale genital mycotic infections include the following adverse reactions: balanoposthitis, balanitis, genital mycotic infections include the following adverse reactions: balanoposthitis, balanitis, to improve glycemic control in adults withwith typetype 2 diabetes mellitus. Limitation of Use: to improve glycemic control in adults 2 diabetes mellitus. Limitation of Use:genital infections fungal, genitourinary tract infection, balanitis candida, scrotal abscess, penile infec-infecinfections fungal, genitourinary tract infection, balanitis candida, scrotal abscess, penile JARDIANCE is not recommended for patients withwith typetype 1 diabetes or for treatment of oftion.genital JARDIANCE is not recommended for patients 1 diabetes or the for the treatment Percentages calculated with with the number of male subjects in each groupgroup as denominator: placebo tion. Percentages calculated the number of male subjects in each as denominator: placebo diabetic ketoacidosis. diabetic ketoacidosis. (N=514), JARDIANCE 10 mg JARDIANCE 25 mg (N=514), JARDIANCE 10 (N=556), mg (N=556), JARDIANCE 25 (N=557). mg (N=557). b CONTRAINDICATIONS: CONTRAINDICATIONS: Thirst (including polydipsia) waswas reported in 0%, 1.7%, andand 1.5% for placebo, JARDIANCE Thirst (including polydipsia) reported in 0%, 1.7%, 1.5% for placebo, JARDIANCE • History of serious hypersensitivity reaction to JARDIANCE. • History of serious hypersensitivity reaction to JARDIANCE. 10 10 mg,mg, andand JARDIANCE 25 25 mg,mg, respectively. Volume Depletion: JARDIANCE causes JARDIANCE respectively. Volume Depletion: JARDIANCE causes • Severe renalrenal impairment, end-stage renalrenal disease, or dialysis [see[see UseUse in Specific • Severe impairment, end-stage disease, or dialysis in Specifican osmotic diuresis, which maymay leadlead to intravascular volume contraction andand adverse an osmotic diuresis, which to intravascular volume contraction adverse Populations]. Populations]. reactions related to volume depletion. In the poolpool of five placebo-controlled clinical tri- trireactions related to volume depletion. In the of five placebo-controlled clinical adverse reactions related to volume depletion (e.g.,(e.g., blood pressure (ambulatory) adverse reactions related to volume depletion blood pressure (ambulatory) WARNINGS ANDAND PRECAUTIONS: Hypotension: JARDIANCE causes intravascular WARNINGS PRECAUTIONS: Hypotension: JARDIANCE causes intravascularals, als, blood pressure systolic decreased, dehydration, hypotension, hypovolemia, decreased, blood pressure systolic decreased, dehydration, hypotension, hypovolemia, volume contraction. Symptomatic hypotension maymay occur afterafter initiating JARDIANCE volume contraction. Symptomatic hypotension occur initiating JARDIANCEdecreased, hypotension, andand syncope) werewere reported by 0.3%, 0.5%, andand 0.3% of of orthostatic hypotension, syncope) reported by 0.3%, 0.5%, 0.3% [see[see Adverse Reactions] particularly in patients withwith renalrenal impairment, the the elderly, in inorthostatic Adverse Reactions] particularly in patients impairment, elderly, treated withwith placebo, JARDIANCE 10 mg, andand JARDIANCE 25 mg respectively. patients treated placebo, JARDIANCE 10 mg, JARDIANCE 25 mg respectively. patients withwith low low systolic blood pressure, andand in patients on diuretics. Before initiatpatients systolic blood pressure, in patients on diuretics. Before initiat-patients maymay increase the the risk risk of hypotension in patients at risk for volume contracJARDIANCE increase of hypotension in patients at risk for volume contracing ing JARDIANCE, assess for volume contraction andand correct volume status if indicated. JARDIANCE, assess for volume contraction correct volume status if indicated.JARDIANCE [see[see Warnings andand Precautions andand UseUse in Specific Populations]. Increased Urination: Warnings Precautions in Specific Populations]. Increased Urination: Monitor for signs andand symptoms of hypotension afterafter initiating therapy andand increase Monitor for signs symptoms of hypotension initiating therapy increasetion tion poolpool five five placebo-controlled clinical trials, adverse reactions of increased urination In the placebo-controlled clinical trials, adverse reactions of increased urination monitoring in clinical situations where volume contraction is expected [see[see UseUse in inIn the monitoring in clinical situations where volume contraction is expected polyuria, pollakiuria, andand nocturia) occurred moremore frequently on JARDIANCE thanthan polyuria, pollakiuria, nocturia) occurred frequently on JARDIANCE Specific Populations]. Impairment in Renal Function: JARDIANCE increases serum Specific Populations]. Impairment in Renal Function: JARDIANCE increases serum(e.g.,(e.g., (see(see Table 1). Specifically, nocturia waswas reported by 0.4%, 0.3%, andand 0.8% on placebo Table 1). Specifically, nocturia reported by 0.4%, 0.3%, 0.8% creatinine andand decreases eGFR [see[see Adverse Reactions]. TheThe risk risk of impaired renalrenalon placebo creatinine decreases eGFR Adverse Reactions]. of impaired treated withwith placebo, JARDIANCE 10 mg, andand JARDIANCE 25 mg, respecof patients treated placebo, JARDIANCE 10 mg, JARDIANCE 25 mg, respecfunction withwith JARDIANCE is increased in elderly patients andand patients withwith moderate function JARDIANCE is increased in elderly patients patients moderateof patients Impairment in Renal Function: UseUse of JARDIANCE waswas associated withwith increases tively. Impairment in Renal Function: of JARDIANCE associated increases renalrenal impairment. MoreMore frequent monitoring of renal function is recommended in these impairment. frequent monitoring of renal function is recommended in thesetively. creatinine andand decreases in eGFR (see(see Table 2). Patients withwith moderate renalrenal in serum creatinine decreases in eGFR Table 2). Patients moderate patients [see[see UseUse in Specific Populations]. Renal function should be evaluated priorprior to toin serum patients in Specific Populations]. Renal function should be evaluated at baseline hadhad larger mean changes. [see[see Warnings andand Precautions andand impairment at baseline larger mean changes. Warnings Precautions initiating JARDIANCE andand periodically thereafter. Hypoglycemia withwith Concomitant initiating JARDIANCE periodically thereafter. Hypoglycemia Concomitantimpairment in Specific Populations]. in Specific Populations]. UseUse withwith Insulin andand Insulin Secretagogues: Insulin andand insulin secretagogues are areUseUse Insulin Insulin Secretagogues: Insulin insulin secretagogues known to cause hypoglycemia. TheThe risk risk of hypoglycemia is increased when JARDIANCE known to cause hypoglycemia. of hypoglycemia is increased when JARDIANCETable 2: Changes from Baseline in Serum Creatinine andand eGFR in the PoolPool of of Table 2: Changes from Baseline in Serum Creatinine eGFR in the is used in combination withwith insulin secretagogues (e.g.,(e.g., sulfonylurea) or insulin [see[see is used in combination insulin secretagogues sulfonylurea) or insulin FourFour 24-week Placebo-Controlled Studies andand Renal Impairment Study 24-week Placebo-Controlled Studies Renal Impairment Study Adverse Reactions]. Therefore, a lower dosedose of the insulin secretagogue or insulin Adverse Reactions]. Therefore, a lower of the insulin secretagogue or insulin maymay be required to reduce the the risk risk of hypoglycemia when usedused in combination withwith be required to reduce of hypoglycemia when in combination PoolPool of 24-Week Placebo-Controlled of 24-Week Placebo-Controlled JARDIANCE. Genital Mycotic Infections: JARDIANCE increases the the risk risk for genital JARDIANCE. Genital Mycotic Infections: JARDIANCE increases for genital Studies Studies mycotic infections [see[see Adverse Reactions]. Patients withwith a history of chronic or recurmycotic infections Adverse Reactions]. Patients a history of chronic or recurPlacebo Placebo JARDIANCE JARDIANCE JARDIANCE JARDIANCE rentrent genital mycotic infections werewere moremore likelylikely to develop mycotic genital infections. genital mycotic infections to develop mycotic genital infections. 10 mg 25 mg 10 mg 25 mg Monitor andand treattreat as appropriate. Urinary Tract Infections: JARDIANCE increases the the Monitor as appropriate. Urinary Tract Infections: JARDIANCE increases N N 825825 830830 822822 risk risk for urinary tracttract infections [see[see Adverse Reactions]. Monitor andand treattreat as approprifor urinary infections Adverse Reactions]. Monitor as appropri- Baseline Baseline Creatinine (mg/dL) 0.840.84 0.850.85 0.850.85 ate.ate. Increased Low-Density Lipoprotein Cholesterol (LDL-C): Increases in LDL-C Increased Low-Density Lipoprotein Cholesterol (LDL-C): Increases in LDL-C Mean Mean Creatinine (mg/dL)2 2 can can occur withwith JARDIANCE [see[see Adverse Reactions]. Monitor andand treattreat as appropriate. occur JARDIANCE Adverse Reactions]. Monitor as appropriate. eGFR (mL/min/1.73 m ) m ) 87.387.3 87.187.1 87.887.8 eGFR (mL/min/1.73 Macrovascular Outcomes: There havehave beenbeen no clinical studies establishing conclusive Macrovascular Outcomes: There no clinical studies establishing conclusive N N 771771 797797 783783 evidence of macrovascular risk risk reduction withwith JARDIANCE or any otherother antidiabetic drug. evidence of macrovascular reduction JARDIANCE or any antidiabetic drug. Week 12 12 Week Creatinine (mg/dL) 0.00 0.02 0.01 Creatinine (mg/dL) 0.00 0.02 0.01 Change ADVERSE REACTIONS: TheThe following important adverse reactions are are described below ADVERSE REACTIONS: following important adverse reactions described below Change -1.3-1.3 -1.4-1.4 eGFR (mL/min/1.73 m2) m2) -0.3-0.3 eGFR (mL/min/1.73 andand elsewhere in the labeling: Hypotension [see[see Warnings andand Precautions]; Impairment elsewhere in the labeling: Hypotension Warnings Precautions]; Impairment in Renal Function [see[see Warnings andand Precautions]; Hypoglycemia withwith Concomitant UseUse N N 708708 769769 754754 in Renal Function Warnings Precautions]; Hypoglycemia Concomitant 24 24 Week withwith Insulin andand Insulin Secretagogues [see[see Warnings andand Precautions]; Genital Mycotic Insulin Insulin Secretagogues Warnings Precautions]; Genital Mycotic Week Creatinine (mg/dL) 0.00 0.01 0.01 Creatinine (mg/dL) 0.00 0.01 0.01 Change Infections [see[see Warnings andand Precautions]; Urinary TractTract Infections [see[see Warnings andand Change Infections Warnings Precautions]; Urinary Infections Warnings -0.6-0.6 -1.4-1.4 eGFR (mL/min/1.73 m2) m2) -0.3-0.3 eGFR (mL/min/1.73 Precautions]; Increased Low-Density Lipoprotein Cholesterol (LDL-C) [see[see Warnings Precautions]; Increased Low-Density Lipoprotein Cholesterol (LDL-C) Warnings a a Moderate Renal Impairment Moderate Renal Impairment andand Precautions]. Clinical Trials Experience: Because clinical trialstrials are are conducted Precautions]. Clinical Trials Experience: Because clinical conducted under widely varying conditions, adverse reaction ratesrates observed in the trialstrials of aof a under widely varying conditions, adverse reaction observed in clinical the clinical Placebo JARDIANCE Placebo JARDIANCE drugdrug cannot be directly compared to rates in the clinical trialstrials of another drugdrug andand maymay cannot be directly compared to rates in the clinical of another 25 mg 25 mg not not reflect the the ratesrates observed in practice. PoolPool of Placebo-Controlled TrialsTrials evaluating reflect observed in practice. of Placebo-Controlled evaluating N N 187187 – – 187187 JARDIANCE 10 and 25 mg: TheThe datadata in Table 1 are derived fromfrom a pool of four 24-week JARDIANCE 10 and 25 mg: in Table 1 are derived a pool of four 24-week Creatinine (mg/dL) 1.491.49 – – 1.461.46 Baseline Creatinine (mg/dL) placebo-controlled trialstrials andand 18-week datadata fromfrom a placebo-controlled trialtrial withwith insulin. placebo-controlled 18-week a placebo-controlled insulin. Baseline JARDIANCE waswas usedused as monotherapy in one trialtrial andand as add-on therapy in four trials. eGFR (mL/min/1.73 m2) m2) 44.344.3 – – 45.445.4 JARDIANCE as monotherapy in one as add-on therapy in four trials. eGFR (mL/min/1.73 These datadata reflect exposure of 1976 patients to JARDIANCE withwith a mean exposure duraThese reflect exposure of 1976 patients to JARDIANCE a mean exposure duraN N 176176 – – 179179 tion tion of approximately 23 weeks. Patients received placebo (N=995), JARDIANCE 10 mg of approximately 23 weeks. Patients received placebo (N=995), JARDIANCE 10 mg Week 12 12 Week Creatinine (mg/dL) 0.01 – 0.12 Creatinine (mg/dL) 0.01 – 0.12 (N=999), or JARDIANCE 25 mg (N=977) onceonce daily.daily. TheThe mean age age of the waswas Change (N=999), or JARDIANCE 25 mg (N=977) mean of population the population Change – – -3.8-3.8 eGFR (mL/min/1.73 m2) m2) 0.1 0.1 eGFR (mL/min/1.73 56 years andand 3% 3% werewere olderolder thanthan 75 years of age. MoreMore thanthan half half (55%) of the 56 years 75 years of age. (55%) of population the population waswas male; 46%46% werewere White, 50%50% werewere Asian, andand 3% 3% werewere Black or African American. male; White, Asian, Black or African American. N N 170170 – – 171171 24 24 At baseline, 57%57% of the population hadhad diabetes moremore thanthan 5 years andand hadhad a mean Week At baseline, of the population diabetes 5 years a mean Week Creatinine (mg/dL) 0.01 – 0.10 Creatinine (mg/dL) 0.01 – 0.10 Change hemoglobin A1cA1c (HbA1c) of 8%. Established microvascular complications of diabetes hemoglobin (HbA1c) of 8%. Established microvascular complications of diabetes Change eGFR (mL/min/1.73 m2) m2) 0.2 0.2 – – -3.2-3.2 eGFR (mL/min/1.73 at baseline included diabetic nephropathy (7%), retinopathy (8%), or neuropathy (16%). at baseline included diabetic nephropathy (7%), retinopathy (8%), or neuropathy (16%). Baseline renalrenal function waswas normal or mildly impaired in 91% of patients andand moderately N N 164164 – – 162162 Baseline function normal or mildly impaired in 91% of patients moderately 2 52 52 Week impaired in 9% of patients (mean eGFR 86.886.8 mL/min/1.73 m2).mTable 1 shows common impaired in 9% of patients (mean eGFR mL/min/1.73 ). Table 1 shows common Week Creatinine (mg/dL) 0.02 – 0.11 Creatinine (mg/dL) 0.02 – 0.11 Change adverse reactions (excluding hypoglycemia) associated withwith the the use use of JARDIANCE. TheThe Change adverse reactions (excluding hypoglycemia) associated of JARDIANCE. 2 2 ) -0.3 – -2.8 eGFR (mL/min/1.73 m ) -0.3 – -2.8 eGFR (mL/min/1.73 m adverse reactions werewere not present at baseline, occurred moremore commonly on JARDIANCE adverse reactions not present at baseline, occurred commonly on JARDIANCE a of patients fromfrom renalrenal impairment studystudy with with eGFReGFR 30 to30less thanthan 60 mL/min/1.73 m2 m2 thanthan on placebo andand occurred in greater thanthan or equal to 2% of patients treated withwithaSubset Subset of patients impairment to less 60 mL/min/1.73 on placebo occurred in greater or equal to 2% of patients treated JARDIANCE 10 mg or JARDIANCE 25 mg. JARDIANCE 10 mg or JARDIANCE 25 mg. Hypoglycemia: TheThe incidence of hypoglycemia by study is shown in Table 3. The inci-inciHypoglycemia: incidence of hypoglycemia by study is shown in Table 3. The Table 1: Adverse Reactions Reported in ≥2% of Patients Treated withwith JARDIANCE Table 1: Adverse Reactions Reported in ≥2% of Patients Treated JARDIANCEdence of hypoglycemia increased when JARDIANCE waswas administered withwith insulin or or dence of hypoglycemia increased when JARDIANCE administered insulin andand Greater than Placebo in Pooled Placebo-Controlled Clinical Studies of Greater than Placebo in Pooled Placebo-Controlled Clinical Studies ofsulfonylurea [see[see Warnings andand Precautions]. sulfonylurea Warnings Precautions]. JARDIANCE Monotherapy or Combination Therapy JARDIANCE Monotherapy or Combination Therapy a a Severeb Hypoglycemic Table 3: Incidence of Overall and Events in Controlled Table 3: Incidence of Overall and Severeb Hypoglycemic Events in Controlled Clinical Studies Clinical Studies Number (%)(%) of Patients Number of Patients Placebo JARDIANCE 10 mg JARDIANCE 25 mg Placebo JARDIANCE 10 mg JARDIANCE 25 mg Monotherapy Placebo 10 mg JARDIANCE 25 mg Monotherapy PlaceboJARDIANCE JARDIANCE 10 mg JARDIANCE 25 mg N=995 N=999 N=977 N=995 N=999 N=977 (24 (24 weeks) (n=229) (n=224) (n=223) weeks) (n=229) (n=224) (n=223) a a 7.6% 9.3% 7.6% Urinary tracttract infection Urinary infection 7.6% 9.3% 7.6% b b 1.5% Overall (%) (%) 0.4% 0.4% 0.4% Overall 0.4% 0.4% 0.4% Female genital mycotic infections 5.4% 6.4% Female genital mycotic infections 1.5% 5.4% 6.4% Upper respiratory tracttract infection 3.1% 4.0% Upper respiratory infection 3.8% 3.8% 3.1% 4.0% Severe (%) (%) 0% 0% 0% 0% 0% 0% Severe c c Increased urination 1.0% 3.4% 3.2% Increased urination 1.0% 3.4% 3.2% JARDIANCE 25 mg JARDIANCE 10 mg In Combination with Placebo + JARDIANCE 25 mg JARDIANCE 10 mg In Combination with Placebo + Dyslipidemia 3.4% 3.9% 2.9% Dyslipidemia 3.4% 3.9% 2.9% Metformin (24 (24 weeks) + Metformin + Metformin + Metformin Metformin weeks) Metformin Metformin + Metformin Arthralgia 2.2% 2.4% 2.3% Arthralgia 2.2% 2.4% 2.3% (n=214) (n=217) (n=206) (n=214) (n=217) (n=206) d d MaleMale genital mycotic infections 0.4% 3.1% 1.6% genital mycotic infections 0.4% 3.1% 1.6% Nausea 1.4% 2.3% 1.1% Overall (%) 0.5% 1.8% 1.4% Nausea 1.4% 2.3% 1.1% Overall (%) 0.5% 1.8% 1.4% a a Predefined adverse eventevent grouping, including, but not to, urinary tracttract infection, asymptomatic Predefined adverse grouping, including, but limited not limited to, urinary infection, asymptomatic Severe (%) (%) Severe bacteriuria, cystitis bacteriuria, cystitis 0% 0% 0% 0% 0% 0% Secretagogues: Coadministration of empagliflozin withwith insulin or insulin secretagogues Secretagogues: Coadministration of empagliflozin insulin or insulin secretagogues increases the the risk risk for hypoglycemia [see[see Warnings andand Precautions]. Positive Urine increases for hypoglycemia Warnings Precautions]. Positive Urine Test: Monitoring glycemic control withwith urineurine glucose teststests is not Glucose Test: Monitoring glycemic control glucose is recommended not recommended 25 mg Placebo 10 mg In Combination withwith JARDIANCE 25 mg Glucose PlaceboJARDIANCE JARDIANCE 10 mgJARDIANCE In Combination taking SGLT2 inhibitors as SGLT2 inhibitors increase urinary glucose excretion in patients taking SGLT2 inhibitors as SGLT2 inhibitors increase urinary glucose excretion + + in patients + + + Metformin Metformin + Sulfonylurea + Metformin (n=225) + Metformin + Metformin Metformin + Sulfonylurea (n=225) will will leadlead to positive urineurine glucose tests. UseUse alternative methods to monitor gly-glyto positive glucose tests. alternative methods to monitor Sulfonylurea (24 (24 weeks) Sulfonylurea andand Sulfonylurea Sulfonylurea weeks) cemic control. Interference withwith 1,5-anhydroglucitol (1,5-AG) Assay: Monitoring cemic control. Interference 1,5-anhydroglucitol (1,5-AG) Assay: Monitoring (n=217) (n=224) (n=217) (n=224) glycemic control withwith 1,5-AG assay is not recommended as measurements of 1,5-AG glycemic control 1,5-AG assay is not recommended as measurements of 1,5-AG Overall (%) (%) 8.4% 16.1% 11.5% Overall 8.4% 16.1% 11.5% are are unreliable in assessing glycemic control in patients taking SGLT2 inhibitors. UseUse unreliable in assessing glycemic control in patients taking SGLT2 inhibitors. alternative methods to monitor glycemic control. alternative methods to monitor glycemic control. Severe (%) (%) 0% 0% 0% 0% 0% 0% Severe USEUSE IN SPECIFIC POPULATIONS: Pregnancy: Pregnancy Category C: There are are no no IN SPECIFIC POPULATIONS: Pregnancy: Pregnancy Category C: There 25 mg Placebo 10 mg In Combination withwith JARDIANCE 25 mg adequate PlaceboJARDIANCE JARDIANCE 10 mgJARDIANCE In Combination andand well-controlled studies of JARDIANCE in pregnant women. JARDIANCE adequate well-controlled studies of JARDIANCE in pregnant women. JARDIANCE (n=165) Pioglitazone +/- +/+ Pioglitazone should (n=165) + Pioglitazone + Pioglitazone + Pioglitazone Pioglitazone be used during pregnancy onlyonly if theif the potential benefit justifies the the potential risk risk should be used during pregnancy potential benefit justifies potential Metformin +/- +/Metformin Metformin (24 (24 weeks) Metformin to the Metformin +/- +/Metformin weeks) fetus. Based on results fromfrom animal studies, empagliflozin maymay affect renalrenal develto the fetus. Based on results animal studies, empagliflozin affect devel(n=168) (n=165) (n=168) (n=165) opment andand maturation. In studies conducted in rats, empagliflozin crosses the the placenta opment maturation. In studies conducted in rats, empagliflozin crosses placenta andand reaches fetalfetal tissues. During pregnancy, consider appropriate alternative therapies, reaches tissues. During pregnancy, consider appropriate alternative therapies, Overall (%) (%) 1.8% 1.2% 2.4% Overall 1.8% 1.2% 2.4% especially during the the second andand thirdthird trimesters. Nursing Mothers: It is Itnot known if if especially during second trimesters. Nursing Mothers: is not known Severe (%) (%) 0% 0% 0% 0% 0% 0% Severe JARDIANCE is excreted in human milk.milk. Empagliflozin is secreted in the milkmilk of lactating JARDIANCE is excreted in human Empagliflozin is secreted in the of lactating ratsrats reaching levels up to higher thanthan thatthat in maternal plasma. Since human reaching levels up5totimes 5 times higher in maternal plasma. Since human In Combination withwith Placebo 10 mg JARDIANCE 25 mg In Combination PlaceboJARDIANCE JARDIANCE 10 mg JARDIANCE 25 mgkidney maturation occurs in utero andand during the the firstfirst 2 years of life when lactational kidney maturation occurs in utero during 2 years of life when lactational c Insulin (18 (18 weeks ) c) (n=170) (n=169) (n=155) Insulin weeks (n=170) (n=169) (n=155) exposure maymay occur, therethere maymay be risk to the developing human kidney. Because many exposure occur, be risk to the developing human kidney. Because many drugs are are excreted in human milkmilk andand because of the potential for serious adverse drugs excreted in human because of the potential for serious adverse Overall (%) (%) 20.6% 19.5% 28.4% Overall 20.6% 19.5% 28.4% reactions in nursing infants fromfrom JARDIANCE, a decision should be made whether to to reactions in nursing infants JARDIANCE, a decision should be made whether Severe (%) (%) 0% 0% 0% 0% 1.3% Severe 1.3% discontinue nursing or toordiscontinue JARDIANCE, taking into into account the the importance discontinue nursing to discontinue JARDIANCE, taking account importance of the drugdrug to the mother. Pediatric Use:Use: TheThe safety andand effectiveness of JARDIANCE a of the to the mother. Pediatric safety effectiveness of JARDIANCE a Overall hypoglycemic events: plasma or capillary glucose of less thanthan or equal to 70tomg/dL Overall hypoglycemic events: plasma or capillary glucose of less or equal 70 mg/dL b b in pediatric patients under 18 years of age havehave not not beenbeen established. Geriatric Use:Use: in pediatric patients under 18 years of age established. Geriatric Severe hypoglycemic events: requiring assistance regardless of blood glucose Severe hypoglycemic events: requiring assistance regardless of blood glucose c c No JARDIANCE dosage change is recommended based on age. A total of 2721 (32%) Insulin dosedose couldcould not be during the initial 18 week treatment period No JARDIANCE dosage change is recommended based on age. A total of 2721 (32%) Insulin notadjusted be adjusted during the initial 18 week treatment period patients treated withwith empagliflozin werewere 65 years of age andand older, andand 491491 (6%)(6%) werewere patients treated empagliflozin 65 years of age older, Genital Mycotic Infections: In the five five placebo-controlled clinical trials, the incidence Genital Mycotic Infections: In pool the pool placebo-controlled clinical trials, the incidence75 years of age andand older. JARDIANCE is expected to have diminished efficacy in elderly 75 years of age older. JARDIANCE is expected to have diminished efficacy in elderly of genital mycotic infections (e.g.,(e.g., vaginal mycotic infection, vaginal infection, genital of genital mycotic infections vaginal mycotic infection, vaginal infection, genitalpatients withwith renalrenal impairment [see[see UseUse in Specific Populations]. TheThe risk risk of volume patients impairment in Specific Populations]. of volume infection fungal, vulvovaginal candidiasis, andand vulvitis) waswas increased in patients treated infection fungal, vulvovaginal candidiasis, vulvitis) increased in patients treateddepletion-related adverse reactions increased in patients whowho werewere 75 years of age depletion-related adverse reactions increased in patients 75 years of age withwith JARDIANCE compared to placebo, occurring in 0.9%, 4.1%, andand 3.7% of patients JARDIANCE compared to placebo, occurring in 0.9%, 4.1%, 3.7% of patientsandand olderolder to 2.1%, 2.3%, andand 4.4% for placebo, JARDIANCE 10 mg, andand JARDIANCE to 2.1%, 2.3%, 4.4% for placebo, JARDIANCE 10 mg, JARDIANCE randomized to placebo, JARDIANCE 10 10 mg,mg, andand JARDIANCE 25 25 mg,mg, respectively. randomized to placebo, JARDIANCE JARDIANCE respectively.25 mg. TheThe risk risk of urinary tracttract infections increased in patients whowho werewere 75 years of age 25 mg. of urinary infections increased in patients 75 years of age Discontinuation fromfrom study duedue to genital infection occurred in 0% of placebo-treated Discontinuation study to genital infection occurred in 0% of placebo-treatedandand olderolder to 10.5%, 15.7%, andand 15.1% in patients randomized to placebo, JARDIANCE to 10.5%, 15.7%, 15.1% in patients randomized to placebo, JARDIANCE patients andand 0.2% of patients treated withwith either JARDIANCE 10 or1025ormg. Genital mycotic patients 0.2% of patients treated either JARDIANCE 25 mg. Genital mycotic10 mg, andand JARDIANCE 25 mg, respectively [see[see Warning andand Precautions andand Adverse 10 mg, JARDIANCE 25 mg, respectively Warning Precautions Adverse infections occurred moremore frequently in female thanthan malemale patients (see(see Table 1). Phimosis infections occurred frequently in female patients Table 1). PhimosisReactions]. Renal Impairment: TheThe efficacy andand safety of JARDIANCE werewere eval-evalReactions]. Renal Impairment: efficacy safety of JARDIANCE occurred moremore frequently in male patients treated withwith JARDIANCE 10 mg (less(less thanthanuated in a study of patients with mild and moderate renal impairment. In this study, occurred frequently in male patients treated JARDIANCE 10 mg uated in a study of patients with mild and moderate renal impairment. In this study, 0.1%) andand JARDIANCE 25 mg (0.1%) thanthan placebo (0%). Urinary TractTract Infections: In the 0.1%) JARDIANCE 25 mg (0.1%) placebo (0%). Urinary Infections: In the195195 patients exposed to JARDIANCE hadhad an eGFR between 60 and 90 mL/min/1.73 m2, m2, patients exposed to JARDIANCE an eGFR between 60 and 90 mL/min/1.73 poolpool five five placebo-controlled clinical trials, the incidence of urinary tract infections (e.g., placebo-controlled clinical trials, the incidence of urinary tract infections (e.g.,91 patients exposed to JARDIANCE hadhad an eGFR between 45 and 60 mL/min/1.73 m2 m2 91 patients exposed to JARDIANCE an eGFR between 45 and 60 mL/min/1.73 urinary tract infection, asymptomatic bacteriuria, and cystitis) was increased in patients urinary tract infection, asymptomatic bacteriuria, and cystitis) was increased in patientsand and 97 patients exposed to JARDIANCE had had an eGFR between 30 and 45 mL/min/1.73 m2. m2. 97 patients exposed to JARDIANCE an eGFR between 30 and 45 mL/min/1.73 treated withwith JARDIANCE compared to placebo (see(see Table 1). Patients withwith a history of ofThe glucose lowering benefit of JARDIANCE 25 mg decreased in patients with worstreated JARDIANCE compared to placebo Table 1). Patients a history The glucose lowering benefit of JARDIANCE 25 mg decreased in patients with worschronic or recurrent urinary tracttract infections werewere moremore likelylikely to experience a urinary tracttractening renal function. The risks of renal impairment [see Warnings and Precautions], chronic or recurrent urinary infections to experience a urinary ening renal function. The risks of renal impairment [see Warnings and Precautions], infection. TheThe raterate of treatment discontinuation duedue to urinary tracttract infections waswas 0.1%, infection. of treatment discontinuation to urinary infections 0.1%,volume depletion adverse reactions andand urinary tracttract infection-related adverse reactions volume depletion adverse reactions urinary infection-related adverse reactions 0.2%, andand 0.1% for placebo, JARDIANCE 10 mg, andand JARDIANCE 25 mg, respectively. 0.2%, 0.1% for placebo, JARDIANCE 10 mg, JARDIANCE 25 mg, respectively.increased withwith worsening renalrenal function. TheThe efficacy andand safety of JARDIANCE havehave increased worsening function. efficacy safety of JARDIANCE Urinary tracttract infections occurred moremore frequently in female patients. TheThe incidence of ofnot been established in patients with severe renal impairment, with ESRD, or receiving Urinary infections occurred frequently in female patients. incidence not been established in patients with severe renal impairment, with ESRD, or receiving urinary tracttract infections in female patients randomized to placebo, JARDIANCE 10 mg, urinary infections in female patients randomized to placebo, JARDIANCE 10 mg,dialysis. JARDIANCE is not expected to be in these patient populations [see[see dialysis. JARDIANCE is not expected to effective be effective in these patient populations andand JARDIANCE 25 mg waswas 16.6%, 18.4%, andand 17.0%, respectively. TheThe incidence JARDIANCE 25 mg 16.6%, 18.4%, 17.0%, respectively. incidenceContraindications andand Warnings andand Precautions]. Hepatic Impairment: JARDIANCE Contraindications Warnings Precautions]. Hepatic Impairment: JARDIANCE of urinary tracttract infections in male patients randomized to placebo, JARDIANCE 10 mg, of urinary infections in male patients randomized to placebo, JARDIANCE 10 mg,maymay be used in patients withwith hepatic impairment. be used in patients hepatic impairment. andand JARDIANCE 25 mg waswas 3.2%, 3.6%, andand 4.1%, respectively [see[see Warnings andand JARDIANCE 25 mg 3.2%, 3.6%, 4.1%, respectively Warnings In the of anofoverdose withwith JARDIANCE, contact the Poison Control OVERDOSAGE: In event the event an overdose JARDIANCE, contact the Poison Control Precautions andand UseUse in Specific Populations]. Laboratory Tests: Increase in Low-Density Precautions in Specific Populations]. Laboratory Tests: Increase in Low-DensityOVERDOSAGE: Employ the the usual supportive measures (e.g.,(e.g., remove unabsorbed material fromfrom Center. Employ usual supportive measures remove unabsorbed material Lipoprotein Cholesterol (LDL-C): Dose-related increases in low-density lipoprotein cho-cho-Center. Lipoprotein Cholesterol (LDL-C): Dose-related increases in low-density lipoprotein gastrointestinal tract,tract, employ clinical monitoring, andand institute supportive treatment) gastrointestinal employ clinical monitoring, institute supportive treatment) lesterol (LDL-C) werewere observed in patients treated withwith JARDIANCE. LDL-C increased lesterol (LDL-C) observed in patients treated JARDIANCE. LDL-C increasedthe the by the patient’s clinical status. Removal of empagliflozin by hemodialysis has has as dictated by the patient’s clinical status. Removal of empagliflozin by hemodialysis by 2.3%, 4.6%, andand 6.5% in patients treated withwith placebo, JARDIANCE 10 mg, andandas dictated by 2.3%, 4.6%, 6.5% in patients treated placebo, JARDIANCE 10 mg, beenbeen studied. studied. JARDIANCE 25 mg, respectively [see[see Warnings andand Precautions]. TheThe range of mean JARDIANCE 25 mg, respectively Warnings Precautions]. range of meannot not baseline LDL-C levels waswas 90.390.3 to 90.6 mg/dL across treatment groups. Increase in inAdditional baseline LDL-C levels to 90.6 mg/dL across treatment groups. Increase information can can be found at www.hcp.jardiance.com Additional information be found at www.hcp.jardiance.com Hematocrit: In a In pool of four placebo-controlled studies, median hematocrit decreased by by Hematocrit: a pool of four placebo-controlled studies, median hematocrit decreased 1.3% in placebo andand increased by 2.8% in JARDIANCE 10 mg andand 2.8% in JARDIANCE © 2014 Boehringer Ingelheim International GmbH 1.3% in placebo increased by 2.8% in JARDIANCE 10 mg 2.8% in JARDIANCECopyright Copyright © 2014 Boehringer Ingelheim International GmbH 25 mg treated patients. At the endend of treatment, 0.6%, 2.7%, andand 3.5% of patients withwithALLALL RIGHTS RESERVED 25 mg treated patients. At the of treatment, 0.6%, 2.7%, 3.5% of patients RIGHTS RESERVED hematocrits initially within the the reference range hadhad values above the the upper limitlimit of the hematocrits initially within reference range values above upper of theJAR-BS-8/14 JAR582220PROF JAR-BS-8/14 JAR582220PROF reference range withwith placebo, JARDIANCE 10 mg, andand JARDIANCE 25 mg, respectively. reference range placebo, JARDIANCE 10 mg, JARDIANCE 25 mg, respectively. DRUG INTERACTIONS: Diuretics: Coadministration of empagliflozin with diuretics DRUG INTERACTIONS: Diuretics: Coadministration of empagliflozin with diuretics resulted in increased urineurine volume andand frequency of voids, which might enhance the the resulted in increased volume frequency of voids, which might enhance potential for for volume depletion [see[see Warnings andand Precautions]. Insulin or Insulin potential volume depletion Warnings Precautions]. Insulin or Insulin Table 3 (cont’d) Table 3 (cont’d) br e i f r e p o rt Do Value Thresholds for Oncology Drugs Differ from Nononcology Drugs? Yuna Hyo Jung Bae, PharmD, and C. Daniel Mullins, PhD ABSTRACT BACKGROUND: In the past decade, many oncologic drugs have been approved that extend life and/or improve patients’ quality of life. However, new cancer drugs are often associated with high price and increased medical spending. For example, in 2010, the average annual cost of care for breast cancer in the final stage of disease was reported to be $94,284, and the total estimated cost in the United States was $16.50 billion. OBJECTIVE: To determine whether value threshold, as defined by the incremental cost-effectiveness ratio (ICER), differed between oncology and other therapeutic areas. METHODS: The PubMed database was searched for articles published between January 2003 and December 2013 with calculated ICER for therapeutic drug entities in a specific therapeutic area. The search term used was “ICER” and “United States.” From 275 results, only those articles that reported ICERs using quality-adjusted life-years (QALY) were included. In addition, only those articles that used a U.S. payer perspective were retained. Among those, nondrug therapy articles and review articles were excluded. The mean ICER and value threshold for oncologic drugs and nononcologic drugs were evaluated for the analysis. RESULTS: From 54 articles selected for analysis, 13 pertained to drugs in oncology therapeutics, and the remaining 41 articles addressed ICER for drugs in other therapeutic areas. The mean and median of ICERs calculated for cancer-specific drug intervention was $138,582/QALY and $55,500/ QALY, respectively, compared with $49,913/QALY and $31,000/QALY, respectively, for noncancer drugs. Among the cancer drugs, 45.0% had ICERs below $50,000/QALY and 70.0% below $100,000/QALY. In comparison, 72.0% of noncancer drugs showed ICERs below $50,000/QALY, and 90.0% had ICERs below $100,000/QALY. When a specific threshold was mentioned, it was in the range of $100,000-$150,000 in cancer drugs, whereas drugs in other therapeutic areas used traditional threshold value within the range of $50,000-$100,000. CONCLUSIONS: The average ICER reported for cancer drugs was more than 2-fold greater than the average ICER for noncancer drugs. In general, articles that addressed the relative value of oncologic pharmaceuticals used higher value thresholds and reported higher ICERs than articles evaluating noncancer drugs. J Manag Care Pharm. 2014;20(11):1086-92 Copyright © 2014, Academy of Managed Care Pharmacy. All rights reserved. What is already known about this subject •Although the incremental cost-effectiveness ratio (ICER), expressed as the cost per quality-adjusted life-year (QALY), has long been used as a standard metric in cost-effectiveness analyses, its use has been met with challenges both in the United States and abroad. 1086 Journal of Managed Care & Specialty Pharmacy JMCP November 2014 •In the United States, the 2010 Patient Protection and Affordable Care Act prohibits the Patient-Centered Outcomes Research Institute (PCORI) from the use of cost/QALY ICER as a threshold to make recommendations on what type of health care or intervention should be utilized. •Other factors besides a drug’s ICER influence the drug formulary decisions of insurers and third-party payers. Often, they need to consider factors such as available resources, existence of alternatives, and/or anticipated impact of the new drug being considered for the formulary. What this study adds •This review of ICERs in oncology and other therapeutic areas documents wide variation in ICERs across disease states. •Our systematic approach to make a side-by-side comparison of ICERs of cancer drugs and noncancer drugs from the literature within the past decade suggests that higher ICER thresholds for anticancer agents may exist. T he U.S. Food and Drug Administration (FDA) approves new anticancer drugs based on evidence for safety and efficacy, which often is demonstrated by extending progression-free survival or overall survival by weeks to months. Although research and development of new drugs are imperative for continued improvement of cancer therapy, many have questioned how sustainable it is for government and third-party payers to continue paying for the increasingly high price of contemporary cancer drugs for the incremental benefit they bring to the patients. The cost of a 1-year supply of these drugs typically reaches $100,000, and pricing for the new incoming agents have been on an upward trend. For example, in 2010, the average annual cost of care for breast cancer in the final stage of disease was reported to be $94,284, and the total estimated cost in the United States was $16.50 billion.1 In order to assess the question of whether a new cancer drug holds adequate value for its price, cost-effectiveness analysis is often performed, which aids the health care decision makers with formulary listings or reimbursement policies. The incremental cost-effectiveness ratio (ICER) is used by many institutions to evaluate the value of a new drug in comparison with the established therapy from clinical efficacy and cost perspectives. The ICER is often expressed in the unit of Vol. 20, No. 11 www.amcp.org Do Value Thresholds for Oncology Drugs Differ from Nononcology Drugs? cost per quality-adjusted life-years (QALY), which incorporates components of quality of life as well as the duration to establish standardization in measuring health utility. It can be used in evaluating how much additional value a new drug can add compared with the current standard therapy at a measured cost. This is often performed with a predetermined threshold value that serves as the maximum ICER limit in deciding whether a drug is cost-effective. There are many factors in addition to the ICER that influence drug formulary decisions by institutions or third-party payers. The threshold value could be set based on an institution’s financial budget, or it could also be taken from previous decisions, which are often variable across institutions and globally.2 Currently, there is no uniformity in determination of a threshold across health care institutions, and a lack of standard exists.3 In the United States, the use of ICERs in assigning value in health outcomes has faced challenges, since the Patient Protection and Affordable Care Act in 2010 prohibited use of cost per QALY as a threshold within the research sponsored by the Patient-Centered Outcomes Research Institute (PCORI). This prohibition was in response to long-standing public concerns that the use of ICER as a threshold would discriminate on the basis of age and disability.4 Despite these concerns, ICER is often used in health care institutions and by third-party payers in private sectors and other countries as a valuable tool in the health care decisionmaking process. The objective of this study was to determine whether value threshold, as defined by the ICER, differed between oncology and other therapeutic areas. ■■ Methods Data Collection The lead author conducted searches and pulled data, which were reviewed by the coauthor. The PubMed electronic database was searched, using the search term “ICER” AND “United States.” Results were restricted to articles in English. Time frame was limited to the 11 years from January 1, 2003, to December 31, 2013, and focus was on the treatments developed during the recent advancement in cancer research. With these criteria, we were able to obtain 275 articles from the search results. Articles that addressed cost-effectiveness of nondrug therapy were excluded. Articles that reported ICER of nonprescription drugs were also excluded from the list. Included articles were those that assumed a U.S. payer perspective and reported ICER in unit of dollar per QALY. If a study reported ICERs from review of multiple independent studies, it was excluded (Figure 1). Analysis All articles were sorted into either an oncology-related drug group or a nononcology-related drug group. Drugs used for treatment of cancer or reducing the risk of cancer were categorized into the oncology-related drug group. Individual www.amcp.org Vol. 20, No. 11 FIGURE 1 Article Selection Process for Analyses of Average ICERs and Value Thresholds Total number of articles from PubMed database search (N = 275) Excluded: Articles not containing cost-effectiveness analysis (n = 54) Articles with nondrug intervention (n = 120) Articles with nonprescription drug intervention (n = 6) Non-U.S. perspective (n = 14) Not third-party payer perspective (n = 7) Articles containing review of other analyses (n = 9) ICER is not shown in dollar per QALY (n = 11) Number of articles after exclusions (n = 54) Articles pertaining to cancer drugs with ICERs (n = 13) Articles that mentioned a specific threshold (n = 5) Articles pertaining to noncancer drugs with ICERs (n = 41) Articles that mentioned a specific threshold (n = 11) ICER = incremental cost-effectiveness ratio; QALY = quality-adjusted life-year. values of ICER were obtained from the articles and analyzed for the mean and median in each group. If an article presented multiple ICERs of 1 drug to several comparators, each ICER was entered separately into the analyses. A similar method of analysis was employed for threshold values stated for the evaluation of ICER in some of the articles. The mean value thresholds obtained from each group of articles were compared for the analysis. If an article mentioned 1 value threshold and November 2014 JMCP Journal of Managed Care & Specialty Pharmacy 1087 Do Value Thresholds for Oncology Drugs Differ from Nononcology Drugs? TABLE 1 Reported ICERs and Value Threshholds for Oncologic Agents Disease State Prostate cancer 9 Colon cancer10 Stage IV lung cancer11 Cervical cancer12 Pancreatic cancer13 Breast cancer14 Breast cancer15 Ovarian cancer16 Drug Comparison Abiraterone Placebo Mitoxantrone Placebo Abiraterone Mitoxantrone FOLFOX 5-FU/LV 5-FU/LV Observation group Erlotinib Platinum-containing chemotherapy Gemcitabine/cisplatin Cisplatin Everolimus Sunitinib Denosumab Zoledronic acid Bevacizumab Standard care Carboplatin/paclitaxel and Carboplatin/paclitaxel additional paclitaxel cycle Carboplatin/paclitaxel/bevacizumab Carboplatin/paclitaxel Peg-filgrastim Filgrastim Peg-filgrastim Filgrastim Tamoxifen Standard care ICER ($/QALY) 94,000 101,000 91,000 54,000 14,000 110,644 33,000 41,000 697,000 745,000 13,000 Value Threshold ($) N/A N/A 100,000 100,000 N/A N/A 150,000 100,000 326,000 N/A 31,000 N/A 6,000 N/A 190,000 (low risk with uterus) 100,000 72,000 (low risk without uterus) 57,000 (high risk with uterus) 37,000 (high risk without uterus) Adjuvant trastuzumab Standard care 39,000 N/A Breast cancer 20 Anastrozole Tamoxifen 20,000 N/A Breast cancer 21 Total: 13 articles Total: 20 interventions Mean: 138,582 Mean: 110,000 FOLFOX = leucovorin, fluorouracil, oxaliplatin; 5-FU = fluorouracil; ICER = incremental cost-effectiveness ratio; LV = leucovorin; N/A = not available; QALY = qualityadjusted life-year. Breast cancer17 Non-Hodgkin lymphoma18 Breast cancer risk reduction19 presented more than 1 ICER for a particular drug, it was counted for each of the ICERs. We also assessed whether the article compared the reported ICER with the value threshold ICER. Because the prices of therapeutic agents, as well as ICERs, are likely to be higher in more recent years, we examined ICER thresholds of the historical benchmark of $50,000/ QALY and the more contemporary benchmark of $100,000/ QALY. Sixteen articles mentioned a specific value threshold for the evaluation of ICER. Of those 16 value thresholds, 5 were from the oncologic drug group, and 11 were from the nononcologic drug group. In the oncologic drug group, the range of value thresholds was $100,000-$150,000 with the mean of $110,000. In comparison, the range of thresholds used in the nononcologic drug group was $50,000-$100,000, and the mean was $68,181. ■■ Results For the analysis, we included 13 articles that addressed cancer treatment and 41 articles that related to treatment of other diseases or conditions. From these articles, we obtained 20 ICERs that were related to cancer treatment and 50 ICERs that were related to treatment of noncancer conditions. The range of ICERs reported for oncologic agents was $6,000-$745,000. The mean in this group was $138,582/QALY and the median was $55,500/QALY (Table 1). Among these values, 45.0% (9 of 20) were below $50,000/QALY, and 70.0% (14 of 20) were less than $100,000/QALY. As for noncancerrelated drugs, the range of ICERs reported in the articles was $-54,000-$332,309, and the mean and median were $49,913/ QALY and $31,000/QALY, respectively. In this group, 72.0% (36 of 50) fell below $50,000/QALY, and 90.0% (45 of 50) were below $100,000/QALY (Table 2). ■■ Discussion The data showed that the mean ICER of oncologic drugs was higher than the mean ICER of nononcologic drugs by more than 2-fold. Among the articles that mentioned specific value threshold in the analysis, all oncologic drugs were evaluated in context of thresholds between $100,000-$150,000, whereas the thresholds for nononcologic drugs were in the range of $50,000-$100,000. The results confirmed that oncologic drugs are often evaluated with value thresholds higher than the traditional range to adjust to high ICERs reported for these agents. A high range of ICERs is also observed in some specialty drugs, such as biologics—drugs made of biological rather than chemical properties—along with many of the cancer therapy drugs. These specialty biologic drugs are often approved for life-threatening illnesses, such as multiple sclerosis, hemophilia, cancer, human immunodeficiency virus, and diabetes. 1088 Journal of Managed Care & Specialty Pharmacy JMCP November 2014 Vol. 20, No. 11 www.amcp.org Do Value Thresholds for Oncology Drugs Differ from Nononcology Drugs? TABLE 2 Reported ICERs and Value Thresholds for Nononcologic Agents Value Threshold ICER ($) Disease State Drug Comparison ($/QALY) Chemoprevention in Barrett’s Esophagus22 Aspirin + statin Aspirin 158,000 100,000 Duloxetine Naproxen 47,678 N/A Osteoarthritis23 Boceprevir Standard dual-therapy: peginterferon 29,184 50,000 Chronic hepatitis C 24 alpha and ribavirin Telaprevir 44,247 Nitroglycerin Standard care to all patients 34,000 N/A Glaucoma 25 Neurogenic detrusor overactivity 26 OnabotulinumtoxinA Best supportive care 24,000 N/A Disease-modifying osteoarthritis drugs Standard care 57,000 N/A Knee osteoarthritis27 Duloxetine Naproxen 59,473 N/A Chronic low back pain 28 Generic-based antiretroviral therapy No antiretroviral therapy 21,000 100,000 Human immunodeficiency virus29 Branded antiretroviral therapy Generic-based antiretroviral therapy 114,000 Exenatide Insulin glargine 15,000 N/A Type 2 diabetes30 Guanfacine XR + stimulant Stimulant monotherapy 31,000 50,000 ADHD31 Low molecular-weight heparin No prophylaxis 90,893 N/A Anticoagulation in cancer patients32 Rivaroxaban Warfarin 27,000 100,000 Stroke prevention in atrial fibrillation 33 Fingolimod IFN beta-1a 73,000 100,000 Multiple sclerosis34 Vaccine (1% colonization rate) No vaccine 25,217 N/A S. aureus vaccine in hemodialysis patients35 Schizophrenia 36 Olanzapine ODT SOT 19,000 N/A Olanzapine ODT Risperidone SOT 39,000 Erythropoietin stimulating agents Routine blood transfusions 873 N/A End-stage renal disease37 Atorvastatin Simvastatin 45,000 N/A Hyperlipidemia 38 Ticagrelor Genotype-driven treatment 10,000 50,000 Acute coronary syndrome39 Atazanavir – ritonavir Lopinavir – ritonavir 26,000 50,000 Human immunodeficiency virus 40 Liraglutide Exenatide 40,000 N/A Type 2 diabetes 41 Darunavir – ritonavir Lopinavir – ritonavir 23,000 N/A Human immunodeficiency virus 42 Bevacizumab Ranibizumab -54,000 N/A Macular degeneration43 Rosuvastatin (20-year horizon) Placebo 10,000 N/A Cardiovascular disease44 Rosuvastatin (10-year horizon) 44,000 Adalimumab Etanercept 5,000 50,000 Psoriasis 45 Infliximab Etanercept 293,000 Omalizumab Usual care 172,000 N/A Asthma46 2-dose influenza vaccine No vaccine 6,787 50,000 Influenza during pregnancy47 Bisphosphonates No bisphosphonates 87,853 N/A Osteoporosis 48 (ages 75-79) Clopidogrel/aspirin Aspirin 36,000 N/A Cardiovascular disease49 Continuous subcutaneous insulin injection Multiple daily injection 16,000 N/A Type 1 diabetes50 Pioglitazone Rosiglitazone 20,000 N/A Type 2 diabetes51 Urokinase Alteplase 332,309 N/A Peripheral artery disease52 Tipranavir – ritonavir Protease inhibitor – ritonavir 56,000 N/A Human immunodeficiency virus53 Olanzapine No treatment 50,000 N/A Alzheimer’s disease54 Exenatide Standard care 35,000 N/A Type 2 diabetes55 Celecoxib NSAID 31,000 N/A Osteoarthritis56 Aprepitant Standard care 96,000 N/A Chemotherapy-induced nausea and vomiting57 Exenatide Insulin 13,000 N/A Type 2 diabetes58 Duloxetine Standard care -342 N/A Diabetic peripheral neuropathy59 -429 Pramipexole Levodopa 42,000 N/A Parkinson’s disease60 Boceprevir (response-guided therapy) Peginterferon-ribavirin 30,200 50,000 Hepatitis C61 Boceprevir (fixed-duration 48 weeks) Boceprevir (response-guided therapy) 91,500 HEPLISAV – diabetes patients Engerix – B 12,613 N/A Hepatitis B prophylaxis62 HEPLISAV – health care workers Engerix – B 11,062 HEPLISAV – travelers Engerix – B 5,564 Total: 41 articles Total: 50 interventions Mean: Mean: 49,913 68,181 ADHD = attention deficit hyperactivity disorder; ICER = incremental cost-effectiveness ratio; IFN = interferon; NSAID = nonsteroidal anti-inflammatory drug; ODT = orally disintegrating tablet; QALY = quality-adjusted life-year; SOT = standard oral tablet; XR = extended-release. www.amcp.org Vol. 20, No. 11 November 2014 JMCP Journal of Managed Care & Specialty Pharmacy 1089 Do Value Thresholds for Oncology Drugs Differ from Nononcology Drugs? This is one of the elements that keeps the prices of these drugs just as high as cancer drugs. Their high prices are also derived from the extended exclusivity protection of the patent for biologic drugs that is separated from regular pharmaceuticals that allowed the monopolistic pricing for these drugs. It is only recently that the patent for some of the earlier biologics expired allowing development of generic versions of these biologics, often referred to as biosimilars.5 However, the regulation of these products is expected to meet with challenges, since the different manufacturing process of biopharmaceuticals may affect the activity of the product. It is reasonable to expect the prices of these drugs to become more affordable as the knowledge of the production technology for biosimilars becomes more standardized in the future. More cancer patients benefit from continued advancement in cancer treatment research that allows patients to live longer. However, there has yet to be a cure for cancer. Current therapy includes drugs that delay cancer progression and extend overall survival as much as possible. Patients are treated with each approved agent sequentially or in combination over the course of the disease, since the effectiveness of the drugs is overcome by resistance, which requires a change in therapy. This need to continuously change treatment plans is the reason that prices of cancer drugs remained high in the past. The use of 1 drug did not invalidate the need for the other drug, creating a virtually monopolistic pricing scheme.6 When a new and improved version of a drug becomes available, the older drug is often viewed as substandard treatment and over time becomes an obsolete option rather than used in establishing a competitive pricing scheme. Many argue that it is unsustainable for the current health system to continue to pay for expensive cancer drugs that provide modest incremental benefits in therapy. The current task remains for society and payers to draw the line and decide when a life-saving drug is “too expensive.” The FDA approved Zaltrap (ziv-aflibercept) in 2012 for second-line treatment of advanced colon cancer based on a phase III clinical trial that showed ziv-aflibercept extended median overall survival by 42 days. However, ziv-aflibercept received disapproval for cost-effectiveness by Memorial Sloan-Kettering Cancer Center in New York and the National Institute for Health and Care Excellence (NICE) in the United Kingdom, based on the conclusion that ziv-aflibercept was no more effective than Avastin (bevacizumab), a similar drug already on the market, but was twice as expensive—priced at $10,000 for a month supply, compared with $5,000 a month for bevacizumab. The reported ICER for ziv-aflibercept by NICE was between $97,000-$102,656/QALY. This may also suggest that with persistent entry of similar cancer drugs into the market, the prices will decline over time, and the ICER of cancer drugs will also decrease in the future.7 1090 Journal of Managed Care & Specialty Pharmacy JMCP November 2014 Cancer drugs are not the only drugs that historically have high prices. Drugs that treat serious illnesses also tend to enter the market with prices in the higher ranges. Correspondingly, we have observed drugs being compared with different thresholds based on the seriousness of the disease they treat. For example, lifestyle drugs are often compared with a lower threshold, while life-saving drugs, such as orphan drugs, are compared with a much higher threshold. This practice raises the question of whether it is valid to have fixed $50,000/QALY or $100,000/QALY thresholds across all payers, types of care, and populations.8 Because of the variations among the types of third-party payers in the United States, it is reasonable that the threshold acceptance should also be based on those factors unique to the insurer and the care given. Limitations First, the reported ICERs included in our analyses are not in direct reference to what is being accepted in the real-world formulary decisions. The value thresholds used in the reports analysed have been chosen by the authors performing the cost-effectiveness analyses, and while it may indicate a general trend of higher value thresholds in oncology drugs, it is not directly attributed by the actual value thresholds utilized by insurers and third-party payers. Second, some of the articles included in our analysis addressed the cost-effectiveness of an old drug, for example, as an added therapy to standard care. Finally, the comparator drug in the cost-effectiveness analyses was not required to be the most appropriate standard therapy for the disease state by the practice guidelines or the most costeffective choice in current practice. The reported ICERs can vary significantly based on the value of the comparator drug. ■■ Conclusions The results of our analyses indicate that cancer drugs are associated with higher ICERs in comparison with ICERs reported for noncancer drugs. On average, the ICER for cancer drugs was more than 2-fold higher than other therapeutic areas, with the majority of cancer and noncancer ICERs falling in the $100,000-150,000 and $50,000-100,000 ranges, respectively. Authors YUNA HYO JUNG BAE, PharmD, is Postdoctoral Fellow in Pharmacoeconomics and Health Outcomes Research, Western University of Health and Sciences College of Pharmacy, Pomona, California, and C. DANIEL MULLINS, PhD, is Professor, Pharmaceutical Health Services Department, University of Maryland School of Pharmacy, Baltimore, Maryland. AUTHOR CORRESPONDENCE: Yuna Hyo Jung Bae, PharmD, 309 E. Second St., Pomona, CA 91766. E-mail: [email protected]. Vol. 20, No. 11 www.amcp.org Do Value Thresholds for Oncology Drugs Differ from Nononcology Drugs? DISCLOSURES There was no external funding for this research, and the authors have no financial or other potential conflicts of interest regarding the subject of this manuscript. Mullins receives grant support from Bayer and Amgen; payment from Genentech for lecturing and speaking engagements; is a paid consultant to Amgen, Bayer, BMS, Genentech, and Pfizer; currently has grants pending with Bayer and Pfizer; and holds an advisory board membership with Bayer, Pfizer, and Mundipharma. Bae and Mullins contributed equally to concept and design of this study. Bae collected the data, which were interpreted by both authors. The manuscript was written primarily by Bae, assisted by Mullins, and revised by both authors. 13. Casciano R, Chulikavit M, Perrin A, Liu Z, Wang X, Garrison LP. Costeffectiveness of everolimus vs sunitinib in treating patients with advanced, progressive pancreatic neuroendocrine tumors in the United States. J Med Econ. 2012;15(Suppl 1):55-64. 14. Snedecor SJ, Carter JA, Kaura S, Botteman MF. Cost-effectiveness of denosumab versus zoledronic acid in the management of skeletal metastases secondary to breast cancer. Clin Ther. 2012;34(6):1334-49. 15. Montero AJ, Avancha K, Gluck S, Lopes G. A cost-benefit analysis of bevacizumab in combination with paclitaxel in the first-line treatment of patients with metastatic breast cancer. Breast Cancer Res Treat. 2012;132(2):747-51. 16. Lesnock JL, Farris C, Krivak TC, Smith KJ, Markman M. Consolidation paclitaxel is more cost-effective than bevacizumab following upfront treatment of advanced epithelial ovarian cancer. Gynecol Oncol. 2011;122(3):473-78. ACKNOWLEDGMENTS We would like to thank Lisa Blatt, MA, for her guidance and assistance with the preparation of the manuscript. REFERENCES 1. Mariotto AB, Yabroff KR, Shao Y, Feuer EJ, Brown ML. Projections of the cost of cancer care in the United States: 2010-2020. J Natl Cancer Inst. 2011;103(2):117-28. 2. McCabe C, Claxton K, Culyer AJ. The NICE cost-effectiveness threshold: what it is and what that means. Pharmacoeconomics. 2008;26(9):733-44. 3. Eichler HG, Kong SX, Gerth WC, Mavros P, Jönsson B. Use of costeffectiveness analysis in health-care resource allocation decision-making: how are cost-effectiveness thresholds expected to emerge? Value Health. 2004;7(5):518-28. 4. Neumann PJ, Weinstein MC. Legislating against use of cost-effectiveness information. N Engl J Med. 2010;363(16):1495-97. Available at: http://www. nejm.org/doi/full/10.1056/NEJMp1007168. Accessed September 16, 2014. 5. Tsiftsoglou AS, Ruiz S, Schneider CK. Development and regulation of biosimilars: current status and future challenges. BioDrugs. 2013;27(3):203-11. 6. Siddiqui M, Rajkumar SV. The high cost of cancer drugs and what we can do about it. Mayo Clinic Proc. 2012;87(10):935-43. Available at: http:// www.mayoclinicproceedings.org/article/S0025-6196(12)00738-0/fulltext. Accessed September 16, 2014. 7. Bach PB, Saltz LB, Wittes RE. In cancer care, cost matters. The New York Times. October 14, 2012. Available at: http://www.nytimes.com/2012/10/15/ opinion/a-hospital-says-no-to-an-11000-a-month-cancer-drug.html?_r=0. Accessed September 16, 2014. 8. Bridges JF, Onukwugha E, Mullins CD. Healthcare rationing by proxy: cost-effectiveness analysis and the misuse of the $50,000 threshold in the U.S. Pharmacoeconomics. 2010;28(3):175-84. Available at: http://link.springer. com/article/10.2165/11530650-000000000-00000/fulltext.html. Accessed September 16, 2014. 9. Zhong L, Pon V, Srinivas S, et al. Therapeutic options in docetaxelrefractory metastatic castration-resistant prostate cancer: a cost-effectiveness analysis. PLoS One. 2013;8(5):e64275. 10. Ayvaci MU, Shi J, Alagoz O, Lubner SJ. Cost-effectiveness of adjuvant FOLFOX and 5FU/LV chemotherapy for patients with stage II colon cancer. Med Decis Making. 2013;33(4):521-32. 11. Handorf EA, McElligott S, Vachani A, et al. Cost effectiveness of personalized therapy for first-line treatment of stage IV and recurrent incurable adenocarcinoma of the lung. J Oncol Pract. 2012;8(5):267-74. 12. Phippen NT, Leath CA 3rd, Chino JP, Jewell EL, Havrilesky LJ, Barnett JC. Cost effectiveness of concurrent gemcitabine and cisplatin with radiation followed by adjuvant gemcitabine and cisplatin in patients with stages IIB to IVA carcinoma of the cervix. Gynecol Oncol. 2012;127(2):267-72. www.amcp.org Vol. 20, No. 11 17. Lyman GH, Lalla A, Barron RL, Dubois RW. Cost-effectiveness of pegfilgrastim versus filgrastim primary prophylaxis in women with earlystage breast cancer receiving chemotherapy in the United States. Clin Ther. 2009;31(5):1092-104. 18. Lyman G, Lalla A, Barron R, Dubois RW. Cost-effectiveness of pegfilgrastim versus 6-day filgrastim primary prophylaxis in patients with nonHodgkin’s lymphoma receiving CHOP-21 in United States. Curr Med Res Opin. 2009;25(2):401-11. 19. Melnikow J, Birch S, Slee C, McCarthy TJ, Helms LJ, Kuppermann M. Tamoxifen for breast cancer risk reduction: impact of alternative approaches to quality-of-life adjustment on cost-effectiveness analysis. Med Care. 2008;46(9):946-53. 20. Kurian AW, Thompson RN, Gaw AF, Arai S, Ortiz R, Garber AM. A costeffectiveness analysis of adjuvant trastuzumab regimens in early HER2/neupositive breast cancer. J Clin Oncol. 2007;25(6):634-41. 21. Locker GY, Mansel R, Cella D, et al. Cost-effectiveness analysis of anastrozole versus tamoxifen as primary adjuvant therapy for postmenopausal women with early breast cancer: a US healthcare system perspective. The 5-year completed treatment analysis of the ATAC (‘Arimidex’, Tamoxifen Alone or in Combination) trial. Breast Cancer Res Treat. 2007;106(2):229-38. 22. Choi SE, Perzan KE, Tramontano AC, Kong CY, Hur C. Statins and aspirin for chemoprevention in Barrett’s esophagus: results of a cost-effectiveness analysis. Cancer Prev Res (Phila). 2014;7(3):341-50. 23. Wielage RC, Bansal M, Andrews JS, Klein RW, Happich M. Cost-utility analysis of duloxetine in osteoarthritis: a US private payer perspective. Appl Health Econ Health Policy. 2013;11(3):219-36. 24. Chan K, Lai MN, Groessl EJ, et al. Cost effectiveness of direct-acting antiviral therapy for treatment-naive patients with chronic HCV genotype 1 infection in the Veterans Health Administration. Clin Gastroenterol Hepatol. 2013;11(11):1503-10. 25. Li EY, Tham CC, Chi SC, Lam DS. Cost-effectiveness of treating normal tension glaucoma. Invest Ophthalmol Vis Sci. 2013;54(5):3394-99. 26. Carlson JJ, Hansen RN, Dmochowski RR, Globe DR, Colayco DC, Sullivan SD. Estimating the cost-effectiveness of onabotulinumtoxinA for neurogenic detrusor overactivity in the United States. Clin Ther. 2013;35(4):414-24. 27. Losina E, Daigle ME, Suter LG, et al. Disease-modifying drugs for knee osteoarthritis: can they be cost-effective? Osteoarthritis Cartilage. 2013;21(5):655-67. 28. Wielage RC, Bansal M, Andrews JS, Wohlreich MM, Klein RW, Happich M. The cost-effectiveness of duloxetine in chronic low back pain: a U.S. private payer perspective. Value Health. 2013;16(2):334-44. 29. Walensky RP, Sax PE, Nakamura YM, et al. Economic savings versus health losses: the cost-effectiveness of generic antiretroviral therapy in the United States. Ann Intern Med. 2013;158(2):84-92. November 2014 JMCP Journal of Managed Care & Specialty Pharmacy 1091 Do Value Thresholds for Oncology Drugs Differ from Nononcology Drugs? 30. Samyshkin Y, Guillermin AL, Best JH, Brunell SC, Lloyd A. Long-term costutility analysis of exenatide once weekly versus insulin glargine for the treatment of type 2 diabetes patients in the U.S. J Med Econ. 2012;15(Suppl 2):6-13. 46. Campbell JD, Spackman DE, Sullivan SD. The costs and consequences of omalizumab in uncontrolled asthma from a USA payer perspective. Allergy. 2010;65(9):1141-48. 31. Sikirica V, Haim Erder M, Xie J, et al. Cost effectiveness of guanfacine extended release as an adjunctive therapy to a stimulant compared with stimulant monotherapy for the treatment of attention-deficit hyperactivity disorder in children and adolescents. Pharmacoeconomics. 2012;30(8):e1-15. 47. Beigi RH, Wiringa AE, Bailey RR, Assi TM, Lee BY. Economic value of seasonal and pandemic influenza vaccination during pregnancy. Clin Infect Dis. 2009;49(12):1784-92. 32. Pishko AM, Smith KJ, Ragni MV. Anticoagulation in ambulatory cancer patients with no indication for prophylactic or therapeutic anticoagulation: a cost-effectiveness analysis from a U.S. perspective. Thromb Haemost. 2012;108(2):303-10. 33. Lee S, Anglade MW, Pham D, Pisacane R, Kluger J, Coleman CI. Costeffectiveness of rivaroxaban compared to warfarin for stroke prevention in atrial fibrillation. Am J Cardiol. 2012;110(6):845-51. 48. Kelton CM, Pasquale MK. Evaluating the claim of enhanced persistence: the case of osteoporosis and implications for payers. Med Decis Making. 2009;29(6):690-706. 49. Chen J, Bhatt DL, Dunn ES, et al. Cost-effectiveness of clopidogrel plus aspirin versus aspirin alone for secondary prevention of cardiovascular events: results from the CHARISMA trial. Value Health. 2009;12(6):872-79. 34. Lee S, Baxter DC, Limone B, Roberts MS, Coleman CI. Cost-effectiveness of fingolimod versus interferon beta-1a for relapsing remitting multiple sclerosis in the United States. J Med Econ. 2012;15(6):1088-96. 50. St Charles M, Lynch P, Graham C, Minshall ME. A cost-effectiveness analysis of continuous subcutaneous insulin injection versus multiple daily injections in type 1 diabetes patients: a third-party U.S. payer perspective. Value Health. 2009;12(5):674-86. 35. Song Y, Tai JH, Bartsch SM, Zimmerman RK, Muder RR, Lee BY. The potential economic value of a Staphylococcus aureus vaccine among hemodialysis patients. Vaccine. 2012;30(24):3675-82. 51. Tunis SL, Minshall ME, St Charles M, Pandya BJ, Baran RW. Pioglitazone versus rosiglitazone treatment in patients with type 2 diabetes and dyslipidemia: cost-effectiveness in the U.S. Curr Med Res Opin. 2008;24(11):3085-96. 36. Ascher-Svanum H, Furiak NM, Lawson AH, et al. Cost-effectiveness of several atypical antipsychotics in orally disintegrating tablets compared with standard oral tablets in the treatment of schizophrenia in the United States. J Med Econ. 2012;15(3):531-47. 52. Olvey EL, Skrepnek GH, Nolan PE Jr. Cost-effectiveness of urokinase and alteplase for treatment of acute peripheral artery disease: comparison in a decision analysis model. Am J Health Syst Pharm. 2008;65(15):1435-42. 37. Naci H, de Lissovoy G, Hollenbeak C, et al. Historical clinical and economic consequences of anemia management in patients with end-stage renal disease on dialysis using erythropoietin stimulating agents versus routine blood transfusions: a retrospective cost-effectiveness analysis. J Med Econ. 2012;15(2):293-304. 38. Grabner M, Johnson W, Abdulhalim AM, Kuznik A, Mullins CD. The value of atorvastatin over the product life cycle in the United States. Clin Ther. 2011;33(10):1433-43. 39. Crespin DJ, Federspiel JJ, Biddle AK, Jonas DE, Rossi JS. Ticagrelor versus genotype-driven antiplatelet therapy for secondary prevention after acute coronary syndrome: a cost-effectiveness analysis. Value Health. 2011;14(4):483-91. 40. Broder MS, Chang EY, Bentley TG, Juday T, Uy J. Cost effectiveness of atazanavir-ritonavir versus lopinavir-ritonavir in treatment-naive human immunodeficiency virus-infected patients in the United States. J Med Econ. 2011;14(2):167-78. 41. Lee WC, Conner C, Hammer M. Results of a model analysis of the costeffectiveness of liraglutide versus exenatide added to metformin, glimepiride, or both for the treatment of type 2 diabetes in the United States. Clin Ther. 2010;32(10):1756-67. 42. Brogan A, Mauskopf J, Talbird SE, Smets E. US cost effectiveness of darunavir/ritonavir 600/100 mg bid in treatment-experienced, HIV-infected adults with evidence of protease inhibitor resistance included in the TITAN Trial. Pharmacoeconomics. 2010;28(Suppl 1):129-46. 53. Simpson KN, Roberts G, Hicks CB, Finnern HW. Cost-effectiveness of tipranavir in treatment-experienced HIV patients in the United States. HIV Clin Trials. 2008;9(4):225-37. 54. Kirbach S, Simpson K, Nietert PJ, Mintzer J. A Markov model of the cost effectiveness of olanzapine treatment for agitation and psychosis in Alzheimer’s disease. Clin Drug Investig. 2008;28(5):291-303. 55. Minshall ME, Oglesby AK, Wintle ME, Valentine WJ, Roze S, Palmer AJ. Estimating the long-term cost-effectiveness of exenatide in the United States: an adjunctive treatment for type 2 diabetes mellitus. Value Health. 2008;11(1):22-33. 56. Loyd M, Rublee D, Jacobs P. An economic model of long-term use of celecoxib in patients with osteoarthritis. BMC Gastroenterol. 2007;7:25. 57. Moore S, Tumeh J, Wojtanowski S, Flowers C. Cost-effectiveness of aprepitant for the prevention of chemotherapy-induced nausea and vomiting associated with highly emetogenic chemotherapy. Value Health. 2007;10(1):23-31. 58. Watkins JB, Minshall ME, Sullivan SD. Application of economic analyses in U.S. managed care formulary decisions: a private payer’s experience. J Manag Care Pharm. 2006;12(9):726-35. Available at: http://www.amcp.org/ WorkArea/DownloadAsset.aspx?id=7483. 59. Wu EQ, Birnbaum HG, Mareva MN, et al. Cost-effectiveness of duloxetine versus routine treatment for U.S. patients with diabetic peripheral neuropathic pain. J Pain. 2006;7(6):399-407. 43. Patel JJ, Mendes MA, Bounthavong M, Christopher ML, Boggie D, Morreale AP. Cost-utility analysis of bevacizumab versus ranibizumab in neovascular age-related macular degeneration using a Markov model. J Eval Clin Pract. 2012;18(2):247-55. 60. Noyes K, Dick AW, Holloway RG; Parkinson Study Group. Pramipexole and levodopa in early Parkinson’s disease: dynamic changes in cost effectiveness. Pharmacoeconomics. 2005;23(12):1257-70. 44. Ohsfeldt RL, Gandhi SK, Smolen LJ, et al. Cost effectiveness of rosuvastatin in patients at risk of cardiovascular disease based on findings from the JUPITER trial. J Med Econ. 2010;13(3):428-37. 61. Chhatwal J, Ferrante SA, Brass C, et al. Cost-effectiveness of boceprevir in patients previously treated for chronic hepatitis C genotype 1 infection in the United States. Value Health. 2013;16(6):973-86. 45. Anis AH, Bansback N, Sizto S, Gupta SR, Willian MK, Feldman SR. Economic evaluation of biologic therapies for the treatment of moderate to severe psoriasis in the United States. J Dermatolog Treat. 2011;22(2):65-74. 62. Kuan RK, Janssen R, Heyward W, Bennett S, Nordyke R. Cost-effectiveness of hepatitis B vaccination using HEPLISAV in selected adult populations compared to Engerix-B(R) vaccine. Vaccine. 2013;31(37):4024-32. 1092 Journal of Managed Care & Specialty Pharmacy JMCP November 2014 Vol. 20, No. 11 www.amcp.org br i e f r e p o rt A Case Study in Generic Drug Use: Should There Be Risk Adjustment in Incentive Payments for the Use of Generic Medications? Surrey M. Walton, PhD; Christine Rash, PharmD; Bruce L. Lambert, PhD; and William L. Galanter, MD, PhD ABSTRACT BACKGROUND: Encouraging generic drug use has reduced health care costs for payers and consumers, but the availability of therapeutically interchangeable medications or generic medications of choice is not equal across disease states. The extent to which systems of care are able to substitute with generics is not well understood. OBJECTIVES: To (a) define and measure the maximum generic rate (MGR) of currently prescribed drugs within an academic medical group in and (b) illustrate differences across drugs associated with selected underlying diseases. METHODS: Prescription claims data were examined from an academic medical group in Chicago, Illinois. Based on pharmacologic and therapeutic criteria, drugs were classified into 2 categories—potentially substitutable and not potentially substitutable—based on whether the drugs are branded forms of the same chemical entities that are available as generics or are therapeutically interchangeable with other medications that have different chemical compositions but the same mechanisms of action and potential efficacy. A medication was considered potentially substitutable if it (a) did not have a narrow therapeutic index as defined by the FDA; (b) did not belong to 1 of 6 protected classes of drugs in the Medicare D provisions; (c) was substitutable with a generic medication containing the same chemical entity; or (d) was therapeutically interchangeable with a therapeutically equivalent medication. MGR was defined as the percentage of prescriptions that could potentially be prescribed in generic form. This rate was examined overall and across drugs known to be associated with illustrative diseases including hypertension, diabetes mellitus, and obstructive lung diseases. RESULTS: The MGR ranged from 100% for drugs used in hypertension to 26.7% for drugs used in obstructive lung diseases. The MGR was 83.6%. CONCLUSIONS: Payers wishing to promote generic substitution should incorporate the potential for substitution of clinically appropriate generic medications as part of incentives for generic utilization to avoid unintended consequences of using a fixed target rate. A practical methodology for determining an MGR is offered. J Manag Care Pharm. 2014;20(11):1093-99 Copyright © 2014, Academy of Managed Care Pharmacy. All rights reserved. What is already known about this subject •Use of generic medications can reduce costs with potential savings estimates ranging around $300 per patient without harming patients. •Current incentives towards generic medication use tend to use a flat rate not conditional on the existing use of drugs or underlying patient population. What this study adds •This study provides a method for generating target generic rates that control for the existing medications in use and, thereby, the underlying patient population. •This study provides novel estimates of the potential for generic substitution for groups of drugs associated with common conditions. www.amcp.org Vol. 20, No. 11 E ncouraging the use of generic drugs has helped reduce costs and can continue to reduce health care costs for payers and consumers with a recent simulation estimate suggesting savings of close to $300 per patient among Medicare Part D enrollees.1-4 Recent opportunities and related efforts to increase the use of generic drugs have been key driving factors moderating pharmacy cost trends over the past 5 years.5,6 Various health care payers have employed specific payment initiatives to drive down costs by incentivizing providers to increase the percentage of generic drugs used in their patient populations.4 Currently, payers often reward independent physician associations (IPAs), individual physicians, and accountable care organizations (ACOs) that show improved prescribing patterns in terms of approaching predetermined benchmarks for generic prescribing. Benchmarks for generic dispensing rates (GDR) are typically derived using formulas created by insurance companies on an annual basis based on past performance of “top performing” IPAs.7,8 One concern with strategies designed to achieve a flat-rate benchmark for generic utilization is that available treatments for some patient populations are branded only, such as insulin products for insulin-requiring patients with diabetes mellitus and asthma medications in the form of inhaler devices. Furthermore, the use of medical supplies (e.g., blood glucose monitors, glucose test strips, glucose testing lancets) tend to be branded only, and orders for such equipment often count as prescriptions under some payers’ financial incentive measures. In addition, in certain clinical scenarios that require medications with a narrow therapeutic index (NTI), switching patients from brand to generic medication may be clinically inappropriate and potentially harmful (e.g., switching a patient on Tegretol tablets to carbamazepine tablets for management of seizure disorder). Hence, the prevalence of conditions being treated and pre-existing drug choices may have a significant impact on the overall ability for an IPA, ACO, or individual physician to achieve a particular percentage of generic use and therefore to reach the benchmark for financial incentive. For example, 1 study claimed that 87% of variation in generic prescription rates was due to differences in case-mix.9 To our knowledge, only 1 previous study has examined the maximum potential cost savings as a result of generic substitution. Kunisawa et al. (2013) examined 9 million dispensing records in Japan from January to March 2010 and defined maximum potential quantity-based shares as the quantity of generic drugs dispensed plus the quantity of branded drugs that could have been replaced by generic drugs divided by the quantity of all drugs dispensed.10 A “substitution index” was November 2014 JMCP Journal of Managed Care & Specialty Pharmacy 1093 A Case Study in Generic Drug Use: Should There Be Risk Adjustment in Incentive Payments for the Use of Generic Medications? defined as the proportion of potentially substitutable drugs (brand-name medications that could have a generic alternative dispensed) that were actually prescribed as generic. Study results showed a maximum potential quantity-based share of 50.1% and a maximum possible cost savings of 16.5% with generic medications.10 Kunisawa et al. concluded that comparisons based on quantity-based share may misrepresent actual medication use, and a substitution index accounting for maximum potential quantity-based share may be a fairer measure and encourage more realistic goals for generic medication use.10 Payers commonly offer financial incentives to providers for achieving benchmark rates of generic drug use.11,12 Providers and payers should understand the extent to which patient case mix affects the ability to prescribe generic medications. For example, if an IPA sees a high volume of patients with insulindependent type 2 diabetes mellitus, they will be forced, due to the nature of the disease state, to prescribe more brand-name products for which therapeutic alternatives do not exist (i.e., glucometers, insulin, glucose testing strips, glucose testing lancets, insulin syringes) compared with an IPA that has a low volume of insulin-dependent type 2 diabetic patients. As previously noted, the ability to prescribe generic drugs depends on the availability of generic alternatives, but the availability of generic alternatives varies significantly as a function of case mix (i.e., as a function of patient and disease-related factors).9 A clearer understanding of the availability of generic alternatives for a given population would allow providers and payers to negotiate more rational financial incentives for generic prescribing. For incentives to be rational in the sense intended here, incentives must be based on the potential for generic substitution or therapeutic interchange within a medical group or IPA. Given the potential for differences in the ability of clinicians to substitute for alternatives (because of differences in case mix), uniform benchmarks, common in contemporary contracts between providers and payers, are inappropriate. To date, insufficient research exists regarding the extent of variance in opportunities for generic prescribing. No one has quantified, for example, what proportion of drugs are available as generic or have a therapeutic alternative among medications used often in common medically treated chronic conditions such as diabetes, hypertension, and chronic respiratory diseases such as chronic obstructive pulmonary disease (COPD) and asthma. Examining variation in the availability of alternative generic drugs would provide insight to payers who create fixed GDR benchmarks for contracted IPAs, ACOs, and individual physicians based on internally derived formulas. We provide relatively simple mechanisms for quantifying the extent of generic substitution possible based on commonly available information in prescription claims data. Although this is an inexact science, such information should allow incentive benchmarks to be adjusted, leveling the playing field and providing equal opportunity for provider groups to achieve benchmarks when generic alternatives are available. Rational incentives should not penalize providers’ use of branded products when no clinically acceptable generic substitute or generic therapeutic alternative is available to treat a patient’s condition. 1094 Journal of Managed Care & Specialty Pharmacy JMCP November 2014 The purpose of this study was to define and measure, based on prescription claims data, the generic conversion potential of medication therapy of a midsized academic medical group. Medications that cannot be converted to generic alternative were also characterized by the reason for the inability to produce a generic conversion. ■■ Methods This study focused on identifying branded drugs being used when there were available alternatives of clinically appropriate generic drugs based on a retrospective analysis of prescription claims data from January 1, 2012, through December 31, 2012. Drugs in the claims data were classified based on an algorithm for coding medications as substitutable or not, which is described in further detail below. For illustrative purposes, drugs were further examined across groups of drugs known to be used in treating 4 common chronic diseases (see Table 1) as follows: hypertension, diabetes mellitus, COPD, and asthma. The groups of drugs were then characterized in terms of the proportion of medications that could be generic, as well as the burden that unavoidable branded medications for these diseases put on the overall GDR. Due to the large overlap of identical inhaled treatment options between COPD and asthma (i.e., Advair Diskus is a brand-only inhaled medication used for the treatment of COPD and asthma), these 2 disease states were subsequently combined and analyzed as 1 group. In addition, a descriptive measure of potential generic use within a disease, called the maximum generic rate (MGR), was constructed and analyzed. The prescription data represented patients covered by a single pharmacy benefit manager (PBM). The physician group was composed of 3 large academic clinics: General Internal Medicine, Family Medicine, and Pediatrics, along with multiple community clinics including a federally qualified health center. The patients were racially diverse with significant numbers of African-American, Caucasian, and Hispanic patients. The insurance was not a Medicare plan, so the age distribution was younger than a cohort of all patients. Since all the patients were covered by insurance, indigent patients were not represented in the cohort. The analyses were based on the number of prescriptions for all medications or supplies. Branded medications were coded as potentially substitutable (having an available therapeutically equivalent generic or available therapeutic alternative) or not potentially generic (no available therapeutically equivalent generic and no available therapeutic alternative that is nonbranded). The characterization was performed by one of the authors, who is a general internist and the chair of the medical center’s pharmacy and therapeutics committee, and another author, who is a fellowship-trained ambulatory care clinical pharmacist, using the following algorithm (see Figure 1): 1.Was the medication a branded form of the same chemical entity that is available as a generic with the same delivery system? If yes, the medication was considered a potentially substitutable generic prescription. For instance, if Lipitor was prescribed and filled as the brand name product, it is considered potentially substitutable, since atorvastatin calcium is an available nonbranded, therapeutically equivalent medication. Vol. 20, No. 11 www.amcp.org A Case Study in Generic Drug Use: Should There Be Risk Adjustment in Incentive Payments for the Use of Generic Medications? TABLE 1 Medications Assigned to COPD/Asthma, Hypertension, and Diabetes H acebutolol H alfuzosin H aliskiren/amlodipine/hydrochlorothiazide H aliskiren-amlodipine H aliskiren-hydrochlorothiazide H aliskiren-valsartan H amiloride-hydrochlorothiazide H amlodipine H amlodipine/hydrochlorothiazide/olmesartan H amlodipine/hydrochlorothiazide/valsartan H amlodipine-atorvastatin H amlodipine-benazepril H amlodipine-olmesartan H amlodipine-telmisartan H amlodipine-valsartan H atenolol H atenolol-chlorthalidone H azilsartan H azilsartan-chlorthalidone H benazepril H benazepril-hydrochlorothiazide H bendroflumethiazide H bendroflumethiazide-nadolol H betaxolol H bisoprolol H bisoprolol-hydrochlorothiazide H candesartan H candesartan-hydrochlorothiazide H captopril H captopril-hydrochlorothiazide H chlorothiazide H clonidine H diltiazem H doxazosin H enalapril H enalapril-hydrochlorothiazide H eprosartan H eprosartan-hydrochlorothiazide H felodipine H fosinopril H fosinopril-hydrochlorothiazide H hydrALAZINE H hydrALAZINE-hydrochlorothiazide H hydrochlorothiazide H hydrochlorothiazide-irbesartan H hydrochlorothiazide-lisinopril H hydrochlorothiazide-losartan C/A = COPD/asthma; DM = diabetes mellitus; H = hypertension. H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H DM DM DM DM DM DM DM hydrochlorothiazide-methyldopa hydrochlorothiazide-metoprolol hydrochlorothiazide-moexipril hydrochlorothiazide-olmesartan hydrochlorothiazide-propranolol hydrochlorothiazide-quinapril hydrochlorothiazide-spironolactone hydrochlorothiazide-telmisartan hydrochlorothiazide-triamterene hydrochlorothiazide-valsartan irbesartan isradipine labetalol lisinopril losartan methyclothiazide methyldopa metoprolol minoxidil moexipril nadolol nicardipine NIFEdipine nisoldipine olmesartan perindopril pindolol prazosin propranolol quinapril ramipril telmisartan terazosin timolol trandolapril trandolapril-verapamil triamterene valsartan verapamil chlorpropamide exenatide glimepiride-pioglitazone glimepiride-rosiglitazone glipizide glipizide-metformin glyburide 2.Exceptions were made for NTI agents as defined in the 1988 NTI list from the U.S. Food and Drug Administration (FDA).13 An example is Dilantin. We also used more recent lists from Health Canada and the North Carolina Board of Pharmacy to produce the final NTI list: aminophylline carbamazepine, cyclosporine, digoxin, ethosuximide www.amcp.org Vol. 20, No. 11 DM DM DM DM DM DM DM DM DM DM DM DM DM DM DM DM DM DM DM DM DM DM DM DM DM DM C/A C/A C/A C/A C/A C/A C/A C/A C/A C/A C/A C/A C/A C/A C/A C/A C/A C/A C/A C/A glyburide-metformin insulin aspart insulin aspart protamine insulin aspart-insulin aspart protamine insulin detemir insulin glargine insulin glulisine insulin isophane insulin isophane-insulin regular insulin lispro insulin lispro-insulin lispro protamine insulin regular linagliptin linagliptin-metformin metformin metformin-pioglitazone metformin-rosiglitazone metformin-saxagliptin metformin-sitagliptin nateglinide pioglitazone repaglinide rosiglitazone saxagliptin simvastatin-sitagliptin sitagliptin albuterol albuterol-ipratropium arformoterol beclomethasone budesonide budesonide-formoterol ciclesonide flunisolide fluticasone fluticasone-salmeterol formoterol formoterol-mometasone indacaterol ipratropium levalbuterol mometasone montelukast pirbuterol salmeterol tiotropium flecainide, levothyroxine sodium, lithium carbonate, phenytoin, primidone, procainamide hydrochloride, quinidine, sirolimus, tacrolimus, theophylline, valproic acid, and warfarin sodium.14,15 Medications that did not meet the first criterion were considered not potentially substitutable if they were in 1 of the 6 protected classes of drugs as defined in the November 2014 JMCP Journal of Managed Care & Specialty Pharmacy 1095 A Case Study in Generic Drug Use: Should There Be Risk Adjustment in Incentive Payments for the Use of Generic Medications? FIGURE 1 Flow Diagram of Drug Categorization with Illustrative Examples Lipitor Prozac Coreg CR Potential Generic Crestor Abilify Dilantin Brand #1 A branded form of the same chemical entity available as a generic with the same delivery system? #2 A narrow therapeutic index medication? #3 A member of the Medicare 6 classes of concern? #4 A brand of the same chemical entity available as a generic with clincally insignificant differences in delivery system? #5 Branded member of a class that is thought to have class effect benefits and minimal risk of harm in switching? Medicare D provisions.17 These include immunosuppressant, antidepressant, antipsychotic, anticonvulsant, antiretroviral, and antineoplastic classes. For example, Abilify does not have a generic and is an antipsychotic. 3.For medications that were not NTI, or in 1 of the 6 protected classes, it was next determined if the medication was a branded form of the same chemical entity available as a generic product with a delivery system that differed from the branded drug in clinically insignificant ways, for example, Coreg CR. A switch from a daily to a twice daily regimen was considered acceptable (i.e., a therapeutic alternative), but conversion to a 3 or more times a day drug was not considered acceptable due to the increased risk of nonadherence.17-19 4.For medications still remaining, the reviewer determined if the medication was a branded member of a class that is thought clinically to have class effect benefits and minimal risk of harm in switching (i.e., a therapeutic alternative). For example, the use of brand-only Crestor 10 mg for the treatment of hyperlipidemia. This would be considered a potentially generic prescription due to the availability of generic atorvastatin calcium at higher, equipotent doses. This algorithm required clinical judgment. Two independent coders coded each drug and had zero disagreements. 1096 Journal of Managed Care & Specialty Pharmacy JMCP November 2014 All medications and prescriptions were categorized for further analysis according to the reason for nonsubstitutability: NTI; 6 protected classes; branded durable medical equipment, supply, or device; and unnecessary branded formulation. For illustrative purposes, medications were selected and categorized further as being predominantly indicated for diabetes (DM), hypertension (HTN), or COPD/asthma (see Table 1). MGR was defined as the highest proportion of generic medication use possible. Maximum Generic Rate: 1 – not potentially substitutable rate Another useful concept employed was the brand burden of a disease. This is the not potentially substitutable rate for a specific disease or, more specifically, for the set of prescriptions inferred to be used for the disease in our analysis. So for either a disease or, in our case, a group of medications inferred to be for a disease, we defined the following: Brand Burden(disease) = 1 – Maximum Generic Rate(disease) The brand burden was determined for the diseases HTN, DM, and COPD/asthma. ■■ Results There were 99,336 prescriptions or supplies filled during 1 year of observation, with 76.1% being generic. Table 2 shows Vol. 20, No. 11 www.amcp.org A Case Study in Generic Drug Use: Should There Be Risk Adjustment in Incentive Payments for the Use of Generic Medications? TABLE 2 TABLE 3 Classification of Nongeneric Medications % of % Classification N Subtype of Total Potentially substitutable 7,466 31.4 Brand available as generic 4,320 57.9 18.2 Therapeutic alternative available 3,146 42.1 13.2 Not potentially generic 16,315 68.6 Narrow therapeutic index 487 3.0 2.0 2,320 14.2 9.8 Six protected classesa No generic substitute or therapeutic 13,508 82.8 56.8 alternative generic available Total 23,781 100.0 a Six protected classes include immunosuppressants, antidepressants, antipsychotics, anticonvulsants, antiretrovirals, and antineoplastics. the classification of the 23,781 nongeneric prescriptions into 5 mutually exclusive types. More than two-thirds (68.6%) of the prescriptions were classified as not potentially generic. The inclusion of medications from the 6 protected drug classes only contributed to 9.8% of the nongeneric prescriptions, with the inclusion of NTI medication being less at 2.0%. The majority of branded products were prescribed in clinical contexts where there was no generic alternative available. The majority of the potentially generic prescriptions were found to be for branded versions of generically available medications. The distribution across disease categories of the brand-only medications and durable medical equipment and supplies is shown in Table 3. DM (30.9%) and COPD/asthma (23.1%) accounted for more than half of the prescriptions that could not be switched to a generic alternative. Roughly 1 in 8 of the prescriptions were for supplies used for the treatment of DM (e.g., syringes, test strips, etc.). The largest single category included medications used to treat diseases other than DM or COPD/ asthma. This category contained no prescriptions for drugs used to treat HTN. The 6 protected classes from Medicare Part D, combined with NTI drugs, accounted for roughly 15% of prescriptions that could not clinically appropriately be written for generic alternatives. Table 4 shows potential generic substitutability (generic, potentially substitutable, or not potentially substitutable) overall and by the selected disease categories (HTN, DM, COPD/ asthma). Table 4 also includes the MGR overall and by selected disease category. Potentially substitutable branded prescribing was highest for COPD/asthma at 14.7% and lowest for DM at 3.0%. The average for all prescriptions was 7.5%. The not potentially substitutable rate was 16.4% overall. The specific not potentially substitutable rate for each disease, or brand burden, was as high as 73% for COPD/asthma and 0% for HTN. The MGRs ranged from 100% for HTN to 26.7% for COPD/asthma, while the average for all prescriptions was 83.6%. ■■ Discussion Generic substitution remains a valuable tool in improving efficiency in the health care system. However, our findings illus- www.amcp.org Vol. 20, No. 11 Distribution of Branded-Only Medications and DME by Disease Type Percentage of Branded-Only Drug Category Medications or Equipment DM 30.9% (13.3% DME, 17.6% medications) COPD/asthma 23.1% Medications for other diagnoses 31% Six protected classes 11.6% Narrow therapeutic index 2.9% Non-DM DME 0.5% COPD = chronic obstructive pulmonary disease; DM = diabetes mellitus; DME = durable medical equipment. trate the importance of considering the underlying distribution of drugs, itself impacted by underlying case mix, in the formation of policy surrounding generic drug use. This is of particular concern if strong financial incentives are tied to achieving fixed benchmarks. Failure to adjust can lead to perverse incentives that may encourage institutions either to select patients whose drug therapy can be handled exclusively by generics (e.g., HTN) and/or to provide less than optimal clinical care by prescribing generic drugs when a branded alternative would be the true drug of choice. This project attempts to address the question of how to set an optimal target for generic drug use. Here, we offer a relatively low cost method inferred from the existing prescribing data. Adjustment along these lines at an institutional or IPA level would promote feasible generic substitution but not penalize groups that had a disadvantageous patient mix (e.g., groups with large numbers of patients with COPD/asthma and/or DM) or incentivize inappropriate medication switching when the drug is unavailable as a generic. There are other ways to perform a case mix adjustment. For example, one might use financial claims data to determine the mix of diseases in a given cohort of patients. Each disease could have an ideal MGR, and across all diseases, an overall weighted MGR could be developed. Our data suggest sizable differences in the maximum feasible generic prescribing rate by disease, with asthma and other chronic obstructive lung diseases having an MGR a quarter of that of HTN, 26.7% vs. 100%. Although this method of case adjustment is what is often done for length of stay, cost of care, and other markers, we believe the better method is to determine, for each medication, whether there was a potentially acceptable generic substitute or therapeutic alternative available as a generic, as we have done. In this manner, imperfect knowledge of the diseases and the best MGR for each disease is not required. In addition, the development of the MGR allows an assignment of each branded medication as having a clinically acceptable generic substitute or generic therapeutic alternative or not, something that can be shared with prescribers. Initially, making this determination for each medication is not trivial, but once performed, maintenance of this database would require changes only for newly available generic formulations, newly available medications and new evidence-based November 2014 JMCP Journal of Managed Care & Specialty Pharmacy 1097 A Case Study in Generic Drug Use: Should There Be Risk Adjustment in Incentive Payments for the Use of Generic Medications? TABLE 4 Distribution of Prescriptions by Hypertension, Diabetes Mellitus, COPD/Asthma, and Generic Status COPD/ HTN DM Asthma All n (%) n (%) n (%) N (%) Generic 16,916 3,169 616 75,555 (94.9) (37.3) (12.0) (76.1) Potentially 903 259 756 7,466 substitutable (5.1) (3.1) (14.7) (7.5) Not potentially 0 5,064 3,766 16,315 substitutable (0) (59.6) (73.3) (16.4) Total sample size 17,819 8,492 5,138 99,336 100.0% 40.4% 26.7% 83.6% Maximum generic rate a a Maximum generic rate is the highest possible percentage of prescriptions for a given disease that could have been written for generic alternatives. COPD = chronic obstructive pulmonary disease; DM = diabetes mellitus; HTN = hypertension. indications, whether FDA approved or off-label. Further, as electronic medical records capture better and better data, if shared with PBMs, this type of model can always be improved. One of the remarkable findings was the large differences in the brand burden by disease. This is clearly related to factors that will change over time, such as the guideline-based preferred therapy for a disease and new evidence-based brand-only additions to therapy and generic availability. In addition, the particular order of medication given for a chronic disease in a patient over time may affect the use of generic medications. Nonetheless, the concept of a brand burden for a disease becomes clear when looking at our 3 diseases of interest. In our cohort, HTN has no brand burden. Treating HTN pharmacologically or taking new patients with HTN into an IPA should not affect nongeneric medication use if medications are selected carefully. On the other hand, DM has a large brand burden. Moreover, for patients on insulin, not only the insulin itself but also the syringes, lancets, test strips, and glucometers produce multiple branded prescriptions with no generic alternatives. Consequently, a traditional fixed generic rate incentive would produce a misaligned incentive not to use insulin. Ideally, there should be no incentive in place that would lead prescribers to consider delaying the start of insulin therapy. For diseases such as asthma and COPD, most medical therapy is also presently branded, and the brand burden we found was 73.3%. Here, clinicians should not be disincentivized to diagnose lung disease or treat with branded inhalers, since this, in fact, would represent the best clinical, evidence-based practice. Using the concept of the MGR, the performance of an IPA, clinician, or health system would be based on the difference between their MGRs and their actual generic rates. In the example analyzed in this article, the medical therapy for HTN is not ideal. While the MGR is 100%, the actual generic rate is 94.9%, and there is still a 5% improvement that can be made within drugs used to treat this common disease. Overall, the MGR is 83.6%, and the actual generic rate is 76.1%, so it follows that the incentive should be to move from 76.1% to 83.6% 1098 Journal of Managed Care & Specialty Pharmacy JMCP November 2014 by better use of generic medications. Moreover, if the MGR is used, there are no incentives to decrease the proportion of patients with DM, asthma, and COPD or to cut down on insulin or inhaler use. Limitations The physician group used for this study is predominantly urban within an academic teaching center. Prescribing decisions may not represent the norm for IPAs as a whole. Patient population is also limited to a commercially insured health maintenance organization population in a large urban setting. It is not likely representative of the average patient who has medication therapy managed by a PBM, which now may encompass Medicaid, Medicare, and commercial insurers. There are inherent limitations in our coding of drugs based solely on prescription claims data. First, there is potential for error via the underlying assumptions that a medication is being used for a specific indication without patient-level information. However, this might not be such a significant weakness, since often a generic substitution exists, independent of the exact indication. For instance, whether the use of a branded angiotensin receptor blocker is for nephropathy, heart failure, or HTN, an acceptable generic substitution is available. This is likely more often the case than the presence of a generic substitution only being available for one indication but not another. Again, in practical use of this method, periodic clinical updating of whether or not a drug has a generic substitute would be helpful and recommended. Second, in this analysis, we chose to use a somewhat arbitrary NTI list because the FDA has not produced a recent NTI list, and there are many to choose from. Hence, we included a medication such as levothyroxine as an NTI medication, but a more aggressive PBM may have left this out. The same could be said for the 6 protected classes used in Medicare D formularies. The exclusion of NTI and protected classes of medications contributed to about 1 in 6 of the not potentially substitutable prescriptions. Thus, it is clear that if others use our method and produce an MGR for a disease or cohort of patients, the results are likely to change some, but the MGR method will be consistent across the prescribers and IPAs being managed for a given PBM. Third, the method requires clinical judgment, and we based our results on consensus between only 2 experienced clinicians. Although we recognize that this is imperfect in design and that the number of clinicians involved in decision making was low, we felt it was sufficient for the purpose of introducing the concept. Further, the general method used here is relatively easily adapted to other settings, and if a PBM were to adopt this methodology, it could certainly choose to use more than 2 clinicians. ■■ Conclusions Substantial efficiencies can be gained by substituting generic drugs for branded drugs, but this can only be done when the drugs used for a particular condition are actually available in clinically acceptable generic form. The availability of generic alternatives varies substantially across prescriptions currently Vol. 20, No. 11 www.amcp.org A Case Study in Generic Drug Use: Should There Be Risk Adjustment in Incentive Payments for the Use of Generic Medications? being used. To illustrate, essentially all drugs for HTN are available as generic, or have a generic alternative, but most drugs for obstructive lung disease do not have a generic alternative. Hence, the maximum rate of generic substitution is determined by the underlying patient population. Because incentives ignore drug and disease-specific variation in the availability of generic alternatives, continued use of uniform generic prescription incentive rates may encourage patient cherry-picking or inappropriate prescribing. Rational incentives for generic drug use should vary according to the underlying patient case-mix. As a solution, we proposed using a per medication methodology and determining an MGR. 3. Balaban D, Dhalla I, Law M, Bell C. Private expenditures on brand name prescription drugs after generic entry. Appl Health Econ Health Policy. 2013;11(5):523-29. 4. Gilman BH, Kautter J. Impact of multitiered copayments on the use and cost of prescription drugs among Medicare beneficiaries. Health Serv Res. 2008;43(2):478-95. 5. Hartman M, Martin A, Nuccio O, Catlin A; National Health Expenditure Accounts Team. Health spending growth at a historic low in 2008. Health Aff (Millwood). 2010;29(1):147-55. 6. Martin AB, Hartman M, Whittle L, Catlin A; National Health Expenditure Accounts Team. National health spending in 2012: rate of health spending growth remained low for the fourth consecutive year. Health Aff (Millwood). 2014;33(1):67-77. 7. Share DA, Mason MH. Michigan’s Physician Group Incentive Program offers a regional model for incremental ‘fee for value’ payment reform. Health Aff (Millwood). 2012;31(9):1993-2001. Authors SURREY M. WALTON, PhD, is Associate Professor, Department of Pharmacy Systems, Outcomes, and Policy, and CHRISTINE RASH, PharmD, is Clinical Pharmacist and Internal Medicine/ Managed Care Clinical Assistant Professor, Department of Pharmacy Practice, College of Pharmacy, University of Illinois Chicago. BRUCE L. LAMBERT, PhD, is Professor, Department of Communication Studies, and Director, Center for Education and Research on Therapeutics, Northwestern University, Chicago, Illinois, and WILLIAM L. GALANTER, MD, PHD, is Assistant Professor, Departments of Medicine; Pharmacy Practice; and Pharmacy Systems, Outcomes, and Policy, University of Illinois Chicago, and Medical Director, Center for Education and Research on Therapeutics, Northwestern University, Chicago, Illinois. AUTHOR CORRESPONDENCE: Surrey M. Walton, PhD, Department of Pharmacy Systems, Outcomes, and Policy, College of Pharmacy, University of Illinois Chicago, 833 S. Wood St., (M/C 871) Rm. 287, Chicago IL 60612. Tel.: 312.413.2775; E-mail: [email protected]. DISCLOSURES None of the authors received any funding directly for this study; however, Galanter and Lambert are supported by grant number U19HS021093 from the Agency for Healthcare Research and Quality. The content is solely the responsibility of the authors and does not necessarily represent the official views of the Agency for Healthcare Research and Quality. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Study concept and design were primarily contributed by Galanter, Walton, and Lambert, with assistance from Rash. Rash and Galanter collected the data, which was interpreted by Walton, Rash, and Galanter, with assistance from Lambert. The manuscript was written by Walton, Lambert, Rash, and Galanter and revised by Rash, Galanter, Walton, and Lambert. References 1. Duru OK, Ettner S, Turk N, et al. Potential savings associated with drug substitution in Medicare Part D: the Translating Research into Action for Diabetes (TRIAD) Study. J Gen Intern Med. 2014;2(1):230-36. 2. Aitken M, Berndt ER, Cutler DM. Prescription drug spending trends in the United States: looking beyond the turning point. Health Aff (Millwood). 2009;28(1):w151-w160. www.amcp.org Vol. 20, No. 11 8. Conrad DA, Perry L. Quality-based financial incentives in health care: can we improve quality by paying for it? Annu Rev Public Health. 2009;30(1):357-71. 9. Wosinska M, Huckman RS. Generic dispensing and substitution in mail and retail pharmacies. Health Aff (Millwood). 2004;Suppl Web Exclusives:W4-409-16. 10. Kunisawa S, Otsubo T, Lee J, Imanaka Y. Improving the assessment of prescribing: use of a ‘substitution index’. J Health Serv Res Policy. 2013;18(3):138-43. 11. Dylst P, Vulto A, Simoens S. Demand-side policies to encourage the use of generic medicines: an overview. Expert Rev Pharmacoecon Outcomes Res. 2013;13(1):59-72. 12. Hoadley JF, Merrell K, Hargrave E, Summer L. In Medicare Part D plans, low or zero copays and other features to encourage the use of generic statins work, could save billions. Health Aff (Millwood). 2012;31(10):2266-75. 13. Center for Drug Evaluation and Research. Guidance for industry: bioavailability and bioequivalence studies for orally administered drug products—general considerations. March 2003. Available at: http://www.fda.gov/ downloads/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/ ucm070124.pdf. Accessed September 15, 2014. 14. North Carolina Department of Health and Human Services. North Carolina Medicaid Pharmacy Newsletter. Number 115, March 2004, page 6. Available at: http://www.ncdhhs.gov/dma/pharmnews/0304pharm.pdf. Accessed September 23, 2014. 15. Health Canada. Guidance document: comparative bioavailability standards: formulations used for systemic effects. Ottawa, Ontario. February 2012. Available at: http://www.hc-sc.gc.ca/dhp-mps/alt_formats/pdf/prodpharma/applic-demande/guide-ld/bio/gd_standards_ld_normes-eng.pdf. Accessed September 23, 2014. 16. U.S. Center for Medicare and Medicaid Services. Medicare prescription drug benefit manual. Chapter 6: Part D drugs and formulary requirements. 2010. Available at: https://www.cms.gov/Medicare/Prescription-DrugCoverage/PrescriptionDrugCovContra/downloads/Chapter6.pdf. Accessed September 15, 2014. 17. Wells KE, Peterson EL, Ahmedani BK, Williams LK. Real-world effects of once vs greater daily inhaled corticosteroid dosing on medication adherence. Ann Allergy Asthma Immunol. 2013;111(3):216-20. 18. Coleman CI, Roberts MS, Sobieraj DM, Lee S, Alam T, Kaur R. Effect of dosing frequency on chronic cardiovascular disease medication adherence. Curr Med Res Opin. 2012;28(5):669-80. 19. Coleman CL, Limone B, Sobieraj DM, et al. Dosing frequency and medication adherence in chronic disease. J Manag Care Pharm. 2012;18(7):52739. Available at: http://www.amcp.org/WorkArea/DownloadAsset. aspx?id=15582. November 2014 JMCP Journal of Managed Care & Specialty Pharmacy 1099 ART-related diarrhea Positively in Control As unique as ART-related diarrhea is to HIV patients, so is its treatment. With its novel mechanism of action, Fulyzaq® (crofelemer) is the only treatment proven effective for the relief of ART-related diarrhea.1 In clinical studies, the most common adverse reactions were upper respiratory tract infection, bronchitis, cough, flatulence, and increased bilirubin. Indication FulyzAq® (crofelemer) is an antidiarrheal indicated for the symptomatic relief of noninfectious diarrhea in adult patients with HIV/AIDS on antiretroviral therapy. Important Safety Information about FULYZAQ FulyzAq® (crofelemer) delayed-release tablets should not be used for the treatment of infectious diarrhea. Rule out infectious etiologies of diarrhea before starting FulyzAq. If infectious etiologies are not considered, and FulyzAq is initiated based on a presumptive diagnosis of noninfectious diarrhea, then there is a risk that patients with infectious etiologies will not receive the appropriate treatments, and their disease may worsen. Based on animal data, FulyzAq may cause fetal harm. Safety and effectiveness of FulyzAq have not been established in patients less than 18 years of age. In clinical studies, the most common adverse reactions (occurring in ≥3% of patients and at a rate greater than placebo) were upper respiratory tract infection, bronchitis, cough, flatulence, and increased bilirubin. Please see brief summary for FULYZAQ [FUHL-ih-zack] on the following page and complete Prescribing Information at www.Fulyzaq.com. Reference: 1. Fulyzaq [prescribing information]. Raleigh, NC: Salix Pharmaceuticals, Inc; 2013. Ful2-0214 T:6.875” The following is a brief summary only. See complete Prescribing Information at www.Fulyzaq.com or request complete Prescribing Information by calling 1-800-508-0024. INDICATIONS AND USAGE FULYZAQ is an anti-diarrheal indicated for the symptomatic relief of non-infectious diarrhea in adult patients with HIV/AIDS on anti-retroviral therapy. CONTRAINDICATIONS None. WARNINGS AND PRECAUTIONS Risks of Treatment in Patients with Infectious Diarrhea If infectious etiologies are not considered, and FULYZAQ is initiated based on a presumptive diagnosis of non-infectious diarrhea, then there is a risk that patients with infectious etiologies will not receive the appropriate treatments, and their disease may worsen. Before starting FULYZAQ, rule out infectious etiologies of diarrhea. FULYZAQ is not indicated for the treatment of infectious diarrhea. ADVERSE REACTIONS Clinical Trials Experience Table 1: Adverse Reactions Occurring in at Least 2% of Patients in the 125 mg Twice Daily Group Adverse Reaction Upper respiratory tract infection Bronchitis Cough Flatulence Increased bilirubin Nausea Back pain Arthralgia Urinary tract infection Nasopharyngitis Musculoskeletal pain Hemorrhoids Giardiasis Anxiety Increased alanine aminotransferase Abdominal distension *Twice daily Crofelemer 125 mg BID* N = 229 n (%) Placebo 13 (5.7) 4 (1.5) 9 (3.9) 8 (3.5) 7 (3.1) 7 (3.1) 6 (2.6) 6 (2.6) 6 (2.6) 5 (2.2) 5 (2.2) 5 (2.2) 5 (2.2) 5 (2.2) 5 (2.2) 5 (2.2) 0 3 (1.1) 3 (1.1) 3 (1.1) 4 (1.5) 4 (1.5) 0 2 (0.7) 2 (0.7) 1 (0.4) 0 0 1 (0.4) 3 (1.1) 5 (2.2) 1 (0.4) N = 274 n (%) DRUG INTERACTIONS Drug Interaction Potential In vitro studies have shown that crofelemer has the potential to inhibit cytochrome P450 isoenzyme 3A and transporters MRP2 and OATP1A2 at concentrations expected in the gut. Due to the minimal absorption of crofelemer, it is unlikely to inhibit cytochrome P450 isoenzymes 1A2, 2A6, 2B6, 2C9, 2C19, 2D6, 2E1 and CYP3A4 systemically. Nelfinavir, Zidovudine, and Lamivudine FULYZAQ administration did not have a clinically relevant interaction with nelfinavir, zidovudine, or lamivudine in a drug-drug interaction trial. USE IN SPECIFIC POPULATIONS Pregnancy Pregnancy Category C Reproduction studies performed with crofelemer in rats at oral doses up to 177 times the recommended daily human dose of 4.2 mg/kg revealed no evidence of impaired fertility or harm to the fetus. In pregnant rabbits, crofelemer at an oral dose of about 96 times the recommended daily human dose of 4.2 mg/kg, caused abortions and resorptions of fetuses. However, it is not clear whether these effects are related to the maternal toxicity observed. A pre- and postnatal development study performed with crofelemer in rats at oral doses of up to 177 times the recommended daily human dose of 4.2 mg/kg revealed no evidence of adverse pre- and postnatal effects in offspring. There are, however, no adequate, well-controlled studies in pregnant women. Because animal reproduction studies are not always predictive of human response, this drug should be used during pregnancy only if clearly needed. and patients with baseline CD4 cell counts greater than or equal to 404 cells/μL (N=289). The safety profile of crofelemer was similar in patients with baseline HIV viral loads less than 400 copies/mL (N = 412) and patients with baseline HIV viral loads greater than or equal to 400 copies/mL (N = 278). PATIENT COUNSELING INFORMATION Instruct patients that FULYZAQ tablets may be taken with or without food. Instruct patients that FULYZAQ tablets should not be crushed or chewed. Tablets should be swallowed whole. You are encouraged to report negative side effects of prescription drugs to the FDA. Visit www.fda.gov/medwatch or call 1-800-FDA-1088. To report adverse events, a product complaint, or for additional information, call: 1-800-508-0024. Rx Only Manufactured by Patheon, Inc. for Salix Pharmaceuticals, Inc. 8510 Colonnade Center Drive, Raleigh, NC 27615 www.salix.com Copyright © Salix Pharmaceuticals, Inc. US Patent Nos. 7,341,744 and 7,323,195. FUL-RALAB36-022014 FULYZAQ is distributed by Salix Pharmaceuticals, Inc. under license from Napo Pharmaceuticals, Inc. The botanical drug substance of FULYZAQ is extracted from Croton lechleri (the botanical raw material) that is harvested from the wild in South America. Nursing Mothers It is not known whether crofelemer is excreted in human milk. Because many drugs are excreted in human milk and because of the potential for adverse reactions in nursing infants from FULYZAQ, a decision should be made whether to discontinue nursing or to discontinue the drug, taking into account the importance of the drug to the mother. Pediatric Use The safety and effectiveness of FULYZAQ have not been established in pediatric patients less than 18 years of age. Geriatric Use Clinical studies with crofelemer did not include sufficient numbers of patients aged 65 and over to determine whether they respond differently than younger patients. Use in Patients with Low CD4 Counts and High Viral Loads No dose modifications are recommended with respect to CD4 cell count and HIV viral load, based on the findings in subgroups of patients defined by CD4 cell count and HIV viral load. The safety profile of crofelemer was similar in patients with baseline CD4 cell count less than 404 cells/μL (lower limit of normal range) (N=388) CROF14CDNY2660_Fulyzaq_BS_6.875x9.875_r6.indd 1 CROF14CDNY2659_Journal_AD_Update_2014_A_Size_BS_r5.indd 1 3/10/14 3/6/14 12:25 3:06 PM PM T:9.875” Because clinical trials are conducted under widely varying conditions, adverse reaction rates observed in the clinical trials of a drug cannot be directly compared to rates in the clinical trials of another drug and may not reflect the rates observed in practice. A total of 696 HIV-positive patients in three placebocontrolled trials received FULYZAQ for a mean duration of 78 days. Of the total population across the three trials, 229 patients received a dose of 125 mg twice a day for a mean duration of 141 days, 69 patients received a dose of 250 mg twice a day for a mean duration of 139 days, 102 patients received a dose of 250 mg four times a day for a mean duration of 14 days, 54 patients received a dose of 500 mg twice a day for a mean duration of 146 days, and 242 patients received a dose of 500 mg four times a day for a mean duration of 14 days. Adverse reactions for FULYZAQ that occurred in at least 2% of patients and at a higher incidence than placebo are provided in Table 1. Adverse reactions that occurred in between 1% and 2% of patients taking a 250 mg daily dose of FULYZAQ were abdominal pain, acne, increased aspartate aminotransferase, increased conjugated bilirubin, increased unconjugated blood bilirubin, constipation, depression, dermatitis, dizziness, dry mouth, dyspepsia, gastroenteritis, herpes zoster, nephrolithiasis, pain in extremity, pollakiuria, procedural pain, seasonal allergy, sinusitis and decreased white blood cell count. Adverse reactions were similar in patients who received doses greater than 250 mg daily. RESEARCH Price Elasticity and Medication Use: Cost Sharing Across Multiple Clinical Conditions Justin Gatwood, PhD, MPH; Teresa B. Gibson, PhD; Michael E. Chernew, PhD; Amanda M. Farr, MPH; Emily Vogtmann, PhD, MPH; and A. Mark Fendrick, MD ABSTRACT BACKGROUND: To address the impact that out-of-pocket prices may have on medication use, it is vital to understand how the demand for medications may be affected when patients are faced with changes in the price to acquire treatment and how price responsiveness differs across medication classes. OBJECTIVE: To examine the impact of cost-sharing changes on the demand for 8 classes of prescription medications. METHODS: This was a retrospective database analysis of 11,550,363 commercially insured enrollees within the 2005-2009 MarketScan Database. Patient cost sharing, expressed as a price index for each medication class, was the main explanatory variable to examine the price elasticity of demand. Negative binomial fixed effect models were estimated to examine medication fills. The elasticity estimates reflect how use changes over time as a function of changes in copayments. RESULTS: Model estimates revealed that price elasticity of demand ranged from -0.015 to -0.157 within the 8 categories of medications (P < 0.01 for 7 of 8 categories). The price elasticity of demand for smoking deterrents was largest (-0.157, P < 0.0001), while demand for antiplatelet agents was not responsive to price (P > 0.05). CONCLUSIONS: The price elasticity of demand varied considerably by medication class, suggesting that the influence of cost sharing on medication use may be related to characteristics inherent to each medication class or underlying condition. J Manag Care Pharm. 2014;20(11):1102-07 Copyright © 2014, Academy of Managed Care Pharmacy. All rights reserved. What is already known about this subject •Cost sharing for prescription medications has risen in the past decade, placing a larger share of the treatment cost on the enrollee. •In some medication classes, increases in consumer cost sharing have been linked to lower levels of medication utilization. What this study adds •Price elasticity of demand varied considerably by medication class among 8 categories of prescription medications. •Results suggest the need for plan designers and practitioners to be sensitive to changing levels of patient cost sharing when providing guidance on medication use. 1102 Journal of Managed Care & Specialty Pharmacy JMCP November 2014 B etween 2000 and 2009, the average copayment for generic (tier 1), preferred brand (tier 2), and nonpreferred brand (tier 3) medications increased 25%, 80%, and 59%, respectively.1 By 2010, the average drug copayment was $10 for a generic prescription, $29 for preferred brand medications, and $49 for nonpreferred brand medications.2 Many studies have found that higher prescription drug cost-sharing amounts are associated with lower levels of medication utilization. For example, Goldman et al. (2004) found that doubling copayment amounts resulted in a 25%-45% decrease in the days of medication supplied, depending on the drug class.3 Across multiple studies, the price elasticity of demand for medications (i.e., the percentage of change in the quantity demanded of a good in response to a 1% change in its price) has been observed to be inelastic and in a range from -0.032 to -0.60—every percentage increase in a medication’s price would result in a 0.032%-0.60% decrease in the amount demanded (often fills per person or the number of days with medication on hand).4-7 Moreover, without respect to medication class, price elasticity of demand for medications to treat chronic conditions was shown to be -0.23 in Medicare patients.8 While such evidence primarily emphasizes the effects that differential cost sharing can have on medication use, it further serves to highlight differences between drug classes and conditions, suggesting that price responsiveness varies by medication class. The variation in price responsiveness may be rooted in the basic determinants of demand for prescription drugs: a derived demand based on the individual’s overall demand for health— the ultimate good being “purchased.” We would expect to see changes in its demand due to several factors, including increases in its own price, as reflected in the level of cost sharing faced by the patient to purchase the medication, and the availability and price of close substitutes or complements.9 In the latter case, if a medication class has less expensive over-thecounter (OTC) substitutes, then the price elasticity of prescription alternatives are likely to be higher than other classes without OTC alternatives, so that when faced with a price increase, the patient may substitute away from prescription medications. Similarly, if a high percentage of generic medications exist within the medication class, then patients may substitute away from brand-name alternatives. However, when considering the role of medication in the production of health, we must consider the varied intention Vol. 20, No. 11 www.amcp.org Price Elasticity and Medication Use: Cost Sharing Across Multiple Clinical Conditions inherent to different classes of medications and their related effects. While numerous medications may be given to resolve acute symptoms or disease, many are prescribed to manage ongoing symptoms or as preventive measures intended to reduce the impact of risk factors or symptoms known to be associated with particular conditions. In the case of prevention, the benefit understood by the patient and the physician may differ dramatically, impacting the resulting value placed on the medication by the patient, who is the consumer.10 In determining this value and making a purchasing decision, the patient must then weigh the benefits and likelihood of obtaining better health in the future against the present cost of the medication; in essence, the patient is determining individual and immediate necessity for the prescribed treatment. Observations to date incorporating the perception of necessity have demonstrated that there is mixed evidence on price elasticities for medications considered to be essential and those that are “discretionary,” with a few studies finding differences in the price responsiveness between essential and discretionary medications and others finding no clear pattern.3,5,7 Additional evidence on price responsiveness across medication classes is warranted to better understand how patients behave when faced with differential cost sharing and how this pattern may differ by medication class. The purpose of this study was to further investigate the relationship between cost sharing and medication utilization patterns across multiple conditions among commercially insured adults. Unlike most previous studies, we analyzed price effects for each medication class across an enrolled population of 11.5 million adults, estimating the collective effects of price changes on medication utilization within the entirety of enrollment. This approach adopted a payer view of the impact of changes in cost sharing, the net effect of which is a combination of behaviors: initiation, discontinuation, and adherence to a medication class. This is in contrast to a patient cohort approach, which focuses on discontinuation and adherence rates for those who have already chosen to initiate a medication class. Results of this study adds to the scant literature on differences in the price elasticity of demand for medications across 8 classes of pharmacotherapy for the treatment of a variety of conditions and will inform efforts to determine cost sharing levels for medication classes. ■■ Methods Data Source This analysis was based on data from the 2005-2009 Truven Health MarketScan Commercial Claims and Encounters Research Database. This database contains health insurance claims for inpatient, outpatient, and outpatient prescription drug services of millions of employees of over 100 medium and large-sized firms in the United States. The data have been statistically de-identified and conform to the Health Insurance www.amcp.org Vol. 20, No. 11 Portability and Accountability Act of 1996 (HIPAA); neither informed consent nor institutional review board (IRB) approval were necessary. Study Population Enrollees were eligible for inclusion in the study if they were between the ages of 18 and 64, were continuously enrolled for at least 7 continuous calendar quarters between January 1, 2006, and September 30, 2009, and had no evidence of pregnancy throughout the study period. Enrollees were included in any calendar quarter if they were enrolled throughout the quarter. Prescription Fills Medication categories of interest included smoking deterrents, nonsteroidal anti-inflammatory drugs or opioids (NSAIDs/ opioids), proton pump inhibitors (PPIs), anticonvulsants, 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitors (statins), bisphosphonates, thyroid hormone, and antiplatelet agents (see Appendix, available in online article). These medications were chosen because they are commonly used and because many are high-cost categories and are classes for which cost-related nonadherence has been identified. The number of prescription fills for each medication was calculated per quarter throughout the study window based on paid claims for each patient. The number of fills per claim was based on the days supply field on the prescription drug claims: a claim with days supply of 30 or less was considered to be 1 fill, while claims with days supply greater than 30 were standardized to a 30-day fill. For example, if a patient had one 12-day fill for a specific drug during the quarter, that patient would have 1 prescription fill, whereas if a patient had one 90-day fill for a specific drug, that patient would have 3 prescription fills during the quarter. Explanatory Variables The main explanatory variable was patient cost sharing—for each medication class, this was measured using a cost-sharing index created for each employer/plan combination within the database.11 The cost-sharing index was based on the average cost-sharing amount (i.e., copayment, coinsurance) per prescription (standardized to a 30-day supply) in an employer/ plan for brand and generic drugs in each medication class. The price index is a market basket approach providing an aggregate measure of plan-level cost sharing weighted for utilization. The index aggregated the brand and generic copayments using weights developed from the overall proportion of utilization of brand and generic drugs within each medication class during the study time frame, with the weights for each class summing to 1. For example, the weights for statins were 0.55 for brand name and 0.45 for generic. Such aggregation of cost sharing into levels for each employer/plan combination reduces any November 2014 JMCP Journal of Managed Care & Specialty Pharmacy 1103 Price Elasticity and Medication Use: Cost Sharing Across Multiple Clinical Conditions effects of potential selection bias related to actual, individuallevel cost sharing and related choices. Sociodemographic variables included age in years, gender, urban or rural residence (based on metropolitan statistical area), median household income (from the 2000 census files based on ZIP codes), and U.S. census region (Northeast, North Central, West, and South). The number of general practitioners and specialists per capita were also included (based on county of residence) from the Area Resource File. The relationship of the enrolled to the employee (self, spouse, or dependent) was indicated. Two health status controls were also included. The Deyo Charlson Comorbidity Index (CCI) was calculated in the year prior to the index date—the date of first fill for each medication. This version of the CCI accounts for the effects of comorbid conditions in the analysis of claims data.12 The number of psychiatric diagnostic groupings (PDG) were also measured during the pre-index period of this study. PDG measures the presence of psychiatric or substance dependent conditions. There are 12 possible PDGs, which are aggregated from ICD9-CM (International Classification of Diseases, Ninth Revision, Clinical Modification) diagnosis codes.13 Examples include alcohol and other substance use disorders, depression, bipolar disorder, post-traumatic stress disorders, and schizophrenia. Analysis As a descriptive analysis of each medication class, patient characteristics and quarterly utilization measures were calculated in the quarter of first fill for each medication. For a multivariable analysis of price elasticity, we created a panel dataset with employer/plan as the cross sectional unit and calendar quarter as the unit of time. In each employer/plan combination, quarterly prescription fills and enrollee counts were summarized, and average characteristics within the plan (e.g., average age, percent female) were calculated. In these plans, cost-sharing levels were assessed across all plan members. The key variable of interest, cost sharing, was measured at the plan level, and we focused on the price responsiveness to these cost-sharing changes. Such an analysis approximates the payer view and includes nonusers of medications in the denominator. Although we did not formally examine cost sharing and initiation of use of a medication class, this per enrollee analysis included the impact of differential rates of initiation. We estimated negative binomial fixed effect models with a fixed effect for each employer/plan combination to account for time invariant employer/plan-level characteristics. We also included quarterly time dummy variables to account for common time trends and exposure-adjusted for enrollment counts in the employer/plan.14 To estimate the impact of cost sharing on the demand for medications, we utilized a fixed effects panel data measuring changes in utilization as they related to changes in cost sharing 1104 Journal of Managed Care & Specialty Pharmacy JMCP November 2014 TABLE 1 Sample Characteristics Total Enrollment Characteristic N = 11,550,464 42.03(12.69) Age in years, mean (SD) Female gender, % 53.1 Census region, % Northeast 13.8 North Central 25.9 South 38.8 West 20.7 Urban, % 85.8 Employee relationship, % Employee 58.9 Spouse 30.2 Child/other 10.9 Plan type, % PPO 52.4 HMO 21.3 POS 14.2 Other 12.2 49.98(18.08) Income (000s), mean (SD)a 2.96(1.36) General practitioners per capita, mean (SD) 9.40(7.26) Specialists per capita, mean (SD) 0.21(0.71) Charlson Comorbidity Index, mean (SD) 0.14(0.46) Psychiatric diagnostic grouping, mean (SD) a Average median household income by ZIP code of residence. HMO =health maintenance organization; POS = point of service; PPO = preferred provider organization; SD = standard deviation. in a given employer/plan from 1 year to the next. By estimating a unique intercept for each employer/plan, the fixed effect design accounted for any time-invariant employer/plan characteristics that were correlated with cost sharing and medication use. The model also accounted for time-variant characteristics that might affect cost sharing over time equally across all employer/plans by including a separate intercept for each quarter. Price responsiveness was measured by calculating the price elasticity: the percentage of change in the utilization measure (fills) with a 1% percentage of change in price. ■■ Results Demographics Across the entire population, the mean age was approximately 42 years, and slightly more than half of the included enrollees were female. The majority of subjects were employees (versus spouse or child), and a larger proportion resided in the southern United States, reflecting the underlying composition of the convenience sample, and in an urban setting. Also, most subjects were insured under a preferred provider organization, and the average median income by ZIP code of residence was approximately $50,000 (Table 1). The demographic characteristics of individuals utilizing each medication class varied: mean ages ranged from 45.5 Vol. 20, No. 11 www.amcp.org Price Elasticity and Medication Use: Cost Sharing Across Multiple Clinical Conditions TABLE 2 Medication Use and Costs by Class Sample Population Fills Medication Cost a Drug Class (n) Mean (SD) Mean (SD) Statins 1,123,236 2.64(1.31) 204.14(158.40) Bisphosphonates 200,404 2.70(1.30) 196.13(106.39) Thyroid hormone 438,603 2.96(1.42) 31.29(25.20) Antiplatelet agents 116,799 2.70(1.40) 330.83(187.35) NSAIDs/opioids 8,249,187 1.65(1.32) 42.67(227.05) PPIs 994,424 2.23(1.30) 267.72(210.51) Anticonvulsants 555,401 2.21(1.66) 210.09(336.77) Smoking deterrents 200,522 1.68(1.00) 172.81(111.20) a Average spend by quarter. NSAID = nonsteroidal anti-inflammatory drug; PPI = proton pump inhibitor; SD = standard deviation. to 55.7, and patients were mostly female in all but 2 classes (statins and antiplatelet agents). Average median household income was similar across all drug classes, ranging from $45,550 to $50,400. Aside from patients on antiplatelet agents, comorbidity indices were relatively similar (not shown). Medication Use and Cost Patient cost sharing (the price index) varied from $6.92 for thyroid hormones to $23.48 for antiplatelet agents. Average spending (quarterly) for medications in each class ranged considerably by drug class—those taking thyroid hormone reported an average expenditure of $31.29, while those on antiplatelet agents had an average expenditure of $330.38. Quarterly spending generally increased with increasing comorbidity indices, while higher levels of generic medication use were associated with lower cost sharing. The vast majority of fills for antiplatelet agents (84.4%) and nearly all fills for smoking deterrents (95.3%) were for branded medications; however, most fills for NSAIDs/opioids were for generic products (88.4%; Table 2). Between the first and last quarter of the study period for each medication class, the percentage of change in average cost sharing ranged from 1.1% (biphosphonates) to 70.9% (smoking deterrents) for generic medications and from 2.0% (NSAIDs/opioids) to 45.7% (smoking deterrents) for brand medications. Elasticity of Demand Estimated price elasticity of demand per enrollee is shown graphically in Figure 1. The effects on prescription drug fills at a per enrollee level ranged from -0.015 to -0.157 (P < 0.01 for 7 of the 8 medication classes). Smoking deterrents were observed to have the largest price responsiveness across all drug classes (-0.157, P < 0.001) followed by PPIs and bisphosphonates; NSAIDs/opioids were observed to be the least responsive to price across all medication classes (-0.015, P < 0.05). Demand for antiplatelet agents was not observed to be responsive to price (P > 0.05). www.amcp.org Vol. 20, No. 11 Price Index ($) 12.11 15.71 6.92 16.43 8.05 15.30 12.24 23.48 Generic Use (%) 45.5 24.5 48.4 15.6 88.4 36.7 62.4 4.7 ■■ Discussion Relatively few studies have investigated the effect of differential cost sharing on price elasticity of demand for a variety of classes of prescription medication. Our results suggest that changes in the demand for particular medications, as evidenced by subsequent fills, can be expected when patients face increases in their required cost sharing. While the magnitude of effects varied by medication class, statistically significant changes, with 1 exception, were seen for each of the included medication classes, suggesting that differential cost sharing has the potential to impact ongoing medication use regardless of therapeutic class. Compared with 2 recent analyses examining the impact of cost sharing on the use of medications in several defined therapeutic classes indicated to treat chronic conditions, the values for price elasticity of demand in our study were noticeably smaller. Goldman et al. observed that across 8 therapeutic classes, the decrease in medication use due to a doubling in copayment (100% change in price) ranged from 25%-45%.3 Moreover, they reported findings that are double our own, where a doubling in copayment was associated with an inelastic response for antiplatelets to a 15.7% decrease for smoking deterrents. Unlike Goldman et al., we controlled for unobserved employer/plan traits by including employer/plan fixed effects, so our elasticities reflect changes over time in employers as a function of how much they raised copayments without relying on differences between employers. Some variation in results may be due to the medication classes examined by each study, since only 4 classes were represented in both analyses; however, wide disparities in the use of medications were observed between classes included in both studies. Differences may also be due to the benefit structure examined by each study as the availability of medications for these conditions and the number of lower-priced generic substitutes would have changed considerably between the years of data used by each investigation, allowing for more patients to remain on their original therapies rather than making abrupt cost-related dis- November 2014 JMCP Journal of Managed Care & Specialty Pharmacy 1105 Price Elasticity and Medication Use: Cost Sharing Across Multiple Clinical Conditions FIGURE 1 Price Elasticity of Demand by Drug Class -0.2 -0.15 -0.1 -0.05 0 -0.015 -0.018 -0.032 -0.051 NSAIDs/opiodsa Antiplatelets Thyroid hormonea Anticonvulsantsa Statinsa -0.064 Bisphosphonatesa -0.066 -0.087 PPIa Smoking deterrentsa -0.157 Note: Bars represent price elasticity of demand for each drug class. a P < 0.01, otherwise P > 0.05. NSAID = nonsteroidal anti-inflammatory drug; PPI = proton pump inhibitor. continuations. Considering the years of data used by Goldman et al. (1997-2000) and our own study (2005-2009), significant differences in market structure would be expected; therefore, our study provides an updated snapshot of consumer reaction to cost-sharing changes in a more recent market that likely offered more generic and OTC options. More recently, Landsman et al. (2005) published findings on medication price elasticity of demand across 8 therapeutic classes that similarly exceeded our own.6 Partial explanation for these differences may be due to the limited overlap of studied medication classes between the 2 studies. However, disparities existed for estimates of the price elasticity of demand when considering the 2 classes (statins and NSAIDs/opioids) included in both investigations. Differences in results are also likely due to the methods employed by each study. Our investigation focused on longitudinal changes in cost sharing over time and the corresponding changes in utilization, including plans in each quarter that did and did not change cost-sharing amounts, allowing for a contemporaneous comparison. Comparatively, Landsman et al. examined 3 health plans with a defined benefit change in prescription drug tier structure, attributing the entire drop in demand after the price increase to the price increase.6 Additionally, and similar to what was mentioned previously, our study examined changes in price elasticity from 2005 to 2009, providing an update on consumer behavior and a snapshot of a more recent market from the 1999-2001 data employed by Landsman et al. 1106 Journal of Managed Care & Specialty Pharmacy JMCP November 2014 The current findings may provide guidance on how patients are likely to alter their medication-taking behaviors when changes are made to the prices they face for particular classes of drugs. For medications with close OTC substitutes, we may have expected to see more dramatic differences in use—such as that observed with PPIs—as patients substitute away from more expensive prescription products. Such a phenomenon was seen in an earlier study where a doubling of copayments led to a 45% and 44% reduction in days supplied for NSAIDs and antihistamines, respectively.3 However, this would fail to explain the relatively small change in NSAIDs/opioids we observed, a category of medications for which some therapeutic equivalent exists over the counter. For this class of drugs, this may be a reflection of the severity of the underlying disease or pain being addressed in our population, suggesting a level of treatment that cannot be adequately managed by OTC substitutes. In terms of benefit design, high value classes with the largest amount of price responsiveness are those that may cause the greatest concern. These classes may be the best candidates for payers to reduce cost-sharing amounts or to implement value-based approaches. Limitations Several limitations impacted the results of our study. The analysis was driven by the use of administrative claims data that is an indirect measure of medication use and may not Vol. 20, No. 11 www.amcp.org Price Elasticity and Medication Use: Cost Sharing Across Multiple Clinical Conditions accurately reflect actual use patterns; however, since these data are generated based on health care claims, they do reflect actual purchasing patterns, which was the focus of this research. The database is also limited in how health status may be measured and accounted for, changes in which may lead to alterations in medication use. Additionally, since multiple medications may have been taken by those included, the resulting fill behavior may be the result of changes in the total cost burden faced by the patient rather than merely the consequence of change realized in each drug class. This may have affected some of the reported values for price elasticity of demand. However, the large sample size employed and range of medication classes included in this analysis are strengths of this study, contributions beyond what has been performed previously in this area of research. Also, this study was limited to only those individuals with commercial health coverage; therefore, results may not be generalizable to patients with other insurance or without coverage. ■■ Conclusions These results of this study suggest that variation in price elasticity of demand exists between classes of medication in response to changes in cost sharing. Considering these results, payers should be wary of the potential effects that changes in cost sharing may have on subsequent medication use, regardless of the role that each medication may play in the patient’s treatment, since some deterrence is likely when cost sharing increases. Of particular note are high value drug classes where patients are highly responsive to price—consideration for the lowering of copayments in these classes should be made to avoid potential interruptions in therapy. Therefore, connecting the patient with the present and future value of therapy at the time of initiation and beyond should be encouraged to optimize treatment effectiveness. DISCLOSURES This project was funded by Pfizer, Inc. All opinions expressed are those of the authors. This study was a Best Poster Finalist at the International Society for Pharmacoeconomics and Outcomes Research (ISPOR) 18th Annual International Meeting, New Orleans, LA, May 22, 2013. Study concept and design were contributed by Gibson, Chernew, Vogtmann, and Fendrick. Data were collected by Gibson, Farr, and Vogtmann, and interpreted by Gibson, Vogtmann, and Gatwood, with assistance from Chernew, Farr, and Fendrick. The manuscript was primarily written and revised by Gatwood, with assistance from Gibson and the other authors. REFERENCES 1. Kaiser Family Foundation, Health Research and Educational Trust. Employer Health Benefits 2011 Summary of Findings. Menlo Park, CA: Kaiser Family Foundation; 2011. Available at: http://www.nahu.org/meetings/ capitol/2012/attendees/jumpdrive/2011%20Employee%20Benefits%20 Survey%20by%20KFF.pdf. Accessed September 11, 2014. 2. Kaiser Family Foundation. Prescription drug trends. May 2010. Available at: http://kaiserfamilyfoundation.files.wordpress.com/2013/01/3057-08.pdf. Accessed September 11, 2014. 3. Goldman DP, Joyce GF, Escarce JJ, et al. Pharmacy benefits and the use of drugs by the chronically ill. JAMA. 2004;291(91):2344-50. 4. Smith DG. The effects of copayments and generic substitution on the use and costs of prescription drugs. Inquiry. 1993;30(2):189-98. 5. Gibson TB, McLaughlin CG, Smith DG. A copayment increase for prescription drugs: the long-term and short-term effects on use and expenditures. Inquiry. 2005;42(3):293-310. 6. Landsman PB, Yu W, Liu X, Teutsch SM, Berger ML. Impact of 3-tier pharmacy benefit design and increased consumer cost-sharing on drug utilization. Am J Manag Care. 2005;11(10):621-28. 7. Goldman DP, Joyce GF, Zheng Y. Prescription drug cost sharing: associations with medication and medical utilization and spending and health. JAMA. 2007;298(1):61-69. 8. Gilman BH, Kautter J. Impact of multitiered copayments on the use and cost of prescription drugs among Medicare beneficiaries. Health Serv Res. 2008;43(2):478-95. 9. Leibowitz A, Manning WG, Newhouse JP. The demand for prescription drugs as a function of cost-sharing. Sco Sci Med. 1985;21(10):1063-69. Authors JUSTIN GATWOOD, PhD, MPH, is Assistant Professor, University of Tennessee College of Pharmacy, Memphis, Tennessee. TERESA B. GIBSON, PhD, is Vice President, and AMANDA M. FARR, MPH, is Researcher, Truven Health Analytics, Ann Arbor, Michigan. EMILY VOGTMANN, PhD, MPH, is Cancer Prevention Fellow, National Cancer Institute, Bethesda, Maryland; MICHAEL E. CHERNEW, PhD, is Professor, Harvard Medical School, Cambridge, Massachusetts; and A. MARK FENDRICK, MD, is Professor, University of Michigan Medical School, Ann Arbor, Michigan. AUTHOR CORRESPONDENCE: Justin Gatwood, PhD, MPH, University of Tennessee College of Pharmacy, 881 Madison Ave., Memphis, TN 38103. Tel.: 901.448.7215; E-mail: [email protected]. www.amcp.org Vol. 20, No. 11 10. Dor A, Encinosa W. How does cost-sharing affect drug purchases? Insurance regimes in the private market for prescription drugs. NBER Working Paper No. 10738. September 2004. Available at: http://www.nber. org/papers/w10738. Accessed September 11, 2014. 11. Chernew M, Gibson TB, Yu-Isenberg K, Sokol MC, Rosen AB, Fendrick AM. Effects of increased patient cost sharing on socioeconomic disparities in health care. J Gen Intern Med. 2008;23(8):1131-36. 12. Deyo RA, Cherkin DC, Ciol MA. Adapting a clinical comorbidity index for use with ICD-9-CM administrative databases. J Clin Epid. 1992;45(6):613-19. 13. Ashcraft ML, Fries BE, Nerenz DR, et al. A psychiatric patient classification system: an alternative to diagnosis-related groups. Med Care. 1989;27(5):543. 14. Hilbe JM. Negative Binomial Regression. 2d ed. New York: Cambridge University Press; 2011. November 2014 JMCP Journal of Managed Care & Specialty Pharmacy 1107 Price Elasticity and Medication Use: Cost Sharing Across Multiple Clinical Conditions Appendix Study Category Proton pump inhibitors Biphosphonates Statins Anticonvulsants Thyroid hormone Antiplatelet agents NSAIDs/opioids Smoking deterrents Medication Classes Used to Determine Study Categories Example Medications Common Uses Dexlansoprazole, Esomeprazole Magnesium, Lansoprazole, Lansoprazole/Naproxen, Gastroesophageal Omeprazole, Omeprazole/Sodium Bicarbonate, Pantoprazole Sodium, Rabeprazole reflux disorder Sodium Alendronate Sodium, Alendronate Sodium/Cholecalciferol, Ibandronate Sodium, Osteoporosis Risedronate Sodium, Risedronate Sodium/Calcium Carbonate, Tiludronate Disodium, Etidronate Disodium, Pamidronate Disodium, Zoledronic Acid Atorvastatin Calcium, Cerivastatin Sodium, Fluvastatin Sodium, Lovastatin, Lovastatin/Niacin, Pravastatin Sodium, Rosuvastatin Calcium, Simvastatin, Amlodipine Besylate/Atorvastatin Calcium, Ezetimibe/Simvastatin, Niacin/Lovastatin, Niacin/Simvastatin Carbamazepine, Clonazepam, Diazepam, Divalproex Sodium, Ethosuximide, Ethotoin, Felbamate, Gabapentin, Lacosamide, Lamotrigine, Levetiracetam, Mephenytoin, Methsuximide, Oxcarbazepine, Paramethadione, Phenacemide, Phenobarbital, Phenobarbital Sodium, Phenytoin, Pregabalin, Primidone, Rufinamide, Tiagabine HCl, Topiramate, Trimethadione, Valproate Sodium, Valproic Acid, Vigabatrin, Zonisamide Levothyroxine Sodium, Liothyronine Sodium, Liotrix Dipyridamole, Cilostazol, Clopidogrel Bisulfate, Aspirin/Dipyridamole, Ticlopidine HCl Hyperlipidemia Anxiety disorders, insomnia, muscle relaxant, epilepsy, panic disorders Market Limited generic entry during study period; OTC substitutes in similar classes Some generic substitutes made available during the study period; no in-class OTC options No OTC substitutes; some generic medications available Numerous generic options available; no OTC substitutes Hypothyroidism Generic substitutes available but no OTC options Acute coronary syn- OTC medication available drome, peripheral but limited generic substiartery disease, stroke tutes available Pain, inflammation, Numerous generic and arthritis OTC substitutes available Alfentanil HCl, Buprenorphine HCl/Naloxone HCl, Butorphanol Tartrate, Celecoxib, Codeinea, Diclofenac Sodium, Etodolac, Fenoprofen Calcium, Flurbiprofen, Ibuprofen a, Indomethacin, Ketoprofen, Ketorolac Tromethamine, Meclofenamate Sodium, Mefenamic Acid, Levomethadyl Acetate HCl, Levorphanol Tartrate, Meloxicam, Meperidine HCl a, Methadone HCl, Morhine Sulfatea, Nalbuphine HCl, Nabumetone, Naproxen, Oxaprozin, Pentazocine HCl, Piroxicam, Rofecoxib, Sulindac, Tolmetin Sodium, Valdecoxib, Aspirin a, Salsalate, Propoxyphene HCl a, Tapentadol HCl, Acetaminophen (including combination products), Fentanyl Citrate a, Hydrocodone a, Hydromorphone a, Oxycodone a, Oxymorphone, Propoxyphene Napsylate, Tramadol HCl a, Remifentanil HCl, Bromfenac Sodium Bupropion HCl, Nicotine, Nicotine Polacrilex, Varenicline, Varenicline Tartrate Smoking cessation OTC medications widely available but few generic substitutes a Including available combination products. HCl = hydrochloride; OTC = over the counter. www.amcp.org Vol. 20, No. 11 November 2014 JMCP Journal of Managed Care & Specialty Pharmacy 1107a 2014 2014 MEMBERSHIP MEMBERSHIP APPLICATION APPLICATION MEMBER INFORMATION MEMBER INFORMATION l Mr. l Mr. l Ms. l Ms. l Mrs. l Mrs. [[ l Dr. l Dr. first name last name first name last name please enter the amcp member who referred you for membership (if applicable). please enter the amcp member who referred you for membership (if applicable). referred by referred by DEMOGRAPHIC INFORMATION DEMOGRAPHIC INFORMATION title please tell us: title please tell us: organization name l l l l l l l I. What degrees/designations do you hold? I. What degrees/designations do you hold? BS Pharmacy PharmD organization name organization address organization address city state zip code city state zip code state zip code home address home address city l Company Address l Company Address state city send all mailings to my: send all mailings to my: work telephone fax work telephone fax home telephone cellular telephone home telephone cellular telephone email address (primary) email address (secondary) email address (primary) email address (secondary) l Home Address l Home Address zip code ANNUAL MEMBERSHIP RATES ANNUAL MEMBERSHIP RATES l Active Member . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . $275 per year Activemembership Member .–. must . . . .be. .a.pharmacist, . . . . . . . physician, . . . . . . .nurse, . . . .nurse . . . .practitioner, . . . . . . . .or. physician . . . . . . assistant. . . . . . $275 per year l Active Active membership – must: be required for pharmacists a pharmacist, physician, nurse, nurse practitioner, or physician assistant. required for pharmacists: school/college of pharmacy graduation year state(s) licensed l Associate Member . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . $440 per year l Associate Member . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . $440 per year l Student Pharmacist Member . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . $45 per year l Student Pharmacist Member . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . $45 per year school/college of pharmacy graduation year required: graduation date (month/year) school/college of pharmacy state(s) licensed l Resident/Fellow/Graduate Student Member . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . l Resident/Fellow/Graduate Student Member . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . required: graduation date (month/year) school/college of pharmacy required: completion date (month/year) site required: completion date (month/year) site school/college of pharmacy graduation year state(s) licensed school/college of pharmacy graduation year state(s) licensed METHOD OF PAYMENT METHOD OF PAYMENT $100 per year $100 per year l Check made payable to AMCP for $ ________________ (in US funds drawn on a US bank) Check made payable to AMCP for $ ________________ (in US funds drawn on a US bank) l l Charge $ ________________ to my credit card: l Visa l MasterCard l American Express l Charge $ ________________ to my credit card: l Visa l MasterCard l American Express card number card number expiration date expiration date cardholder printed name cardholder printed name cardholder signature ]] l PharmD l MPH MPH l JD JD l RPh RPh l RN RN l DO Other (specify) l DO _________________________________________________________ MPA BS Pharmacy PhD MPA MBA PhD MD MBA Other (specify) MD _________________________________________________________ II. Which of the following best describes your (check one) best describes your II. employer? Which of the following employer? (check one) ACO/PCMH/Emerging Care Model Medical/Physician Group l l l l l l l l l l l l l Service Medical/Physician Group l MTM Mail Service MTM or Service l PBM Industry PBM or Mail Service l Pharmaceutical Pharmaceutical Industry l Press Press l Research/Analytics Research/Analytics l Retired Retired Pharmacy l Specialty Specialty Pharmacy l Technology/IT Technology/IT l Wholesale/Distribution/ (specify) Medical Other Education GPO l Wholesale/Distribution/ Other (specify) GPO _________________________________________________________ l l l l l l l l l l l l l l l Assistant Director Director/ l Pharmacy Manager Assistant Director l Pharmacy Pharmacy Manager l Pharmacy/Provider Network Management l Pharmacy/Provider Network Management l President/CEO Devel President/CEO l Product/Program Relations Product/Program Devel l Prof./Trade - Outcomes/Clinical Prof./Trade Relations l Research Resident/Fellow - Outcomes/Clinical l Research Resident/Fellow l Staff/Operations Pharmacist l Staff/Operations Pharmacist l Student Other (specify) l Student _________________________________________________________ Adherence Service Provider ACO/PCMH/Emerging Care Model College/University Adherence Service Provider Community Pharmacy College/University Consulting CommunityFirm Pharmacy Government/Military Consulting Firm Health Plan Government/Military Hospital/Health System Health Plan Managed Markets Agency Hospital/Health System Medical Education Managed Markets Agency _________________________________________________________ III. Which of the following best describes your job one) best describes your job III. function(s)? Which of the(check following (check one) function(s)? Academic Faculty/Staff Pharmacy Director/ Case Manager Academic Faculty/Staff Clinical Pharmacist/Coordinator Case Manager Consultant Clinical Pharmacist/Coordinator Contracting/Distribution Consultant C-Suite Member/VP Contracting/Distribution Formulary/Drug Use Mgmt C-Suite Member/VP Legal Affairs/Govt Affairs Formulary/Drug Use Mgmt Marketing/Sales Legal Affairs/Govt Affairs Medical Affairs Marketing/Sales Medical Director/CMO Affairs Not Employed/Retired Medical Director/CMO (specify) Other Not Employed/Retired _________________________________________________________ IV. Indicate your license or eligibility for licensure below. one) IV. (check Indicate your license or eligibility for licensure below. l l l l l (check MD one) Not Applicable MD Nurse Not Applicable Nurse Practitioner Nurse Practitioner l l l l l Other ________________ Physician Assistant Other ________________ Pharmacist Physician Assistant Pharmacy PharmacistTechnician Pharmacy Technician AMCP dues are not deductible as a charitable contribution for U.S. federaldues income be deductible as a business AMCP are tax notpurposes, deductiblebut as may a charitable contribution for U.S. expense. AMCPtax estimates that dues areasnot deductible federal income purposes, but15% mayofbeyour deductible a business because ofAMCP AMCP’s lobbying activities behalf its not members. expense. estimates that 15% ofonyour duesofare deductible AMCP contributes of all Active and on Associate annual dues to because of AMCP’s$5 lobbying activities behalf of its members. the AMCP Foundation. memberships are not transferable AMCP contributes $5 ofAMCP all Active and Associate annual dues to or refundable. the AMCP Foundation. AMCP memberships are not transferable or refundable. cardholder signature Academy of Managed Care Pharmacy | 100 N Pitt Street | Suite 400 | Alexandria, VA 22314 | Tel 703/683-8416 | Fax 703/683-8417 | www.amcp.org Academy of Managed Care Pharmacy | 100 N Pitt Street | Suite 400 | Alexandria, VA 22314 | Tel 703/683-8416 | Fax 703/683-8417 | www.amcp.org Need the latest on managed care pharmacy issues? AMCP AMCP provides provides the the solutions solutions to to your your challenges! challenges! AMCP’s monthly webinar series is offered at no cost to AMCP members — bringing the latest in the issues and experts on managed care AMCP’s monthly seriesThis is offered at no to AMCP members — bringing the latest in the issues and experts on managed care pharmacy right towebinar your desk! is just one of cost the many benefits of AMCP membership! pharmacy right to your desk! This is just one of the many benefits of AMCP membership! n Updates on legislation and regulatory issues pertinent to managed care pharmacy that can affect your business. n Updates on legislation and regulatory issues pertinent to managed care pharmacy that can affect your business. n Topics have included Medicare Part D updates, electronic prior authorization, biosimilars, comparative effectiveness research, n Topics have included Medicare Part D updates, electronic prior authorization, biosimilars, comparative effectiveness research, medication therapy management and much more! medication therapy management and much more! n Miss a webinar or want to reference it later — no problem! AMCP records all webinars and posts them to our Virtual Conference n Miss a webinar or want to reference it later — no problem! AMCP records all webinars and posts them to our Virtual Conference Center. All recordings are available at no cost to AMCP members. Center. All recordings are available at no cost to AMCP members. n Valued at $149 each — webinar attendance pays for Active membership after attending just two webinars! n Valued at $149 each — webinar attendance pays for Active membership after attending just two webinars! AMCP AMCP — — The The Solution Solution to to Your Your Challenges Challenges Join Today at www.amcp.org Join Today at www.amcp.org IN THEIR WORDS AMCP webinars are a must-see for anyone wanting the latest, “ webinars are a must-see for anyone wanting the latest, “ AMCP most collaborative information from top leaders in the health care most collaborative information from top leaders in the health care industry. ” industry. ” Dr. Caroline Atwood Dr. Caroline Atwood AMCP VALUED MEMBER SINCE 2009 AMCP VALUED MEMBER SINCE 2009 Academy of Managed Care Pharmacy | 100 N Pitt Street | Suite 400 | Alexandria, VA 22314 | Tel 703/683-8416 | Fax 703/683-8417 | www.amcp.org Academy of Managed Care Pharmacy | 100 N Pitt Street | Suite 400 | Alexandria, VA 22314 | Tel 703/683-8416 | Fax 703/683-8417 | www.amcp.org RESEARCH Predictors of Treatment Initiation with Tumor Necrosis Factor-α Inhibitors in Patients with Rheumatoid Arthritis Rishi J. Desai, PhD; Jaya K. Rao, MD; Richard A. Hansen, PhD; Gang Fang, PhD; Matthew L. Maciejewski, PhD; and Joel F. Farley, PhD ABSTRACT BACKGROUND: Introduction of biologic disease-modifying antirheumatic drugs (DMARDs) has revolutionized treatment in patients with rheumatoid arthritis (RA). However, due to substantially higher costs of biologics compared with nonbiologics, patients with less insurance generosity may have difficulty affording these agents, which may lead to potential access disparities. OBJECTIVE: To identify factors affecting treatment initiation with tumor necrosis factor (TNF)-α inhibitor biologics in patients with RA. METHODS: Health insurance claims data derived from Truven’s MarketScan Commercial Claims and Encounters and Medicare Supplemental and Coordination of Benefits (2007-2010) were used to conduct a retrospective cohort study. Two separate cohorts of RA patients were identified: (1) monotherapy nonbiologic DMARD users and (2) combination therapy nonbiologic DMARD users. The primary outcome was TNF-α inhibitor initiation 12 months following an index inpatient or outpatient RA visit during 2008-2009. Predictors were measured 12 months pre-index and grouped into predisposing, enabling, or need factors based on Andersen’s Behavior Model. Predisposing variables included age, sex, and geographic location; enabling variables included insurance-related factors such as capitation, payer type, and insurance generosity, which was defined using costsharing information from prescriptions filled by the patients in the previous year; and need variables included disease-related factors such as severity of RA, use of pain control medications, and presence of other comorbidities. Hierarchical logistic regression models were used to derive estimates of the impact of individual predictors. RESULTS: Initiation of TNF-α inhibitors was observed in 10.31% of the monotherapy nonbiologic DMARD users (1,922 of 18,641) and 13.09% of combination nonbiologic DMARD users (983 of 7,508). Among monotherapy nonbiologic DMARD users, initiation with TNF-α inhibitors was associated with the predisposing factors of age (OR = 0.98, 95% CI = 0.97-0.98 for each year increase) and geographic region (Midwest vs. South OR = 0.83, 95% CI = 0.73-0.93; Northeast vs. South OR = 0.77, 95% CI = 0.64-0.92; and West vs. South OR = 0.86, 95% CI = 0.74-0.99); enabling factors of visit to rheumatologists (1 visit vs. no visit OR = 1.22, 95% CI = 1.01-1.46), health insurance type (commercial vs. Medicare supplemental OR = 0.79, 95% CI = 0.66-0.95), and drug benefit generosity (above average vs. poor OR = 1.16, 95% CI = 1.01-1.34 and most generous vs. poor OR = 1.21, 95% CI = 1.05-1.40); and need factors of RA severity (OR = 1.19, 95% CI = 1.14-1.23 for each unit increase in a claims-based RA severity index [CIRAS]), pre-index pain reliever use (steroids OR = 1.81, 95% CI = 1.622.02; nonselective nonsteroidal anti-inflammatory drugs [NSAID] OR = 1.17, 95% CI = 1.05-1.31; COX-2 inhibitors OR = 1.22, 95% CI = 1.05-1.41), and comorbidities (OR = 0.94, 95% CI = 0.90-0.99 for each unit increase in a comorbidity index). Treatment initiation with TNF-α inhibitors among patients with combination therapy nonbiologic DMARDs use at baseline 1110 Journal of Managed Care & Specialty Pharmacy JMCP November 2014 was associated with age (OR = 0.98, 95% CI = 0.97-0.99 for each year increase) and region (Midwest vs. South OR = 0.81, 95% CI = 0.68-0.96). Stronger associations with some of the need factors were observed (CIRAS OR = 1.28, 95% CI = 1.21-1.35 for each unit increase, steroids use OR = 2.05, 95% CI = 1.73-2.42, and nonselective NSAID use OR = 1.36, 95% CI = 1.171.58) in these patients compared with the monotherapy nonbiologic DMARD users. However, unlike the monotherapy DMARD user group, the enabling factors of health insurance type and drug benefit generosity were not found to be associated with TNF-α inhibitor initiation among nonbiologic DMARD combination therapy users. CONCLUSIONS: Potential disparities in the initiation of TNF-α inhibitors among RA patients on monotherapy DMARDs at baseline were noted among older patients, patients in certain geographic region of the United States, and patients with less generous prescription drug benefits. Although future research should examine the impact of these disparities on health outcomes, payers should be aware of the potential for undertreatment among these groups of RA patients when making formulary decisions. J Manag Care Pharm. 2014;20(11):1110-20 Copyright © 2014, Academy of Managed Care Pharmacy. All rights reserved. What is already known about this subject •Biologic agents, indicated for the treatment of rheumatoid arthritis (RA) in patients who do not respond adequately to nonbiologics alone, are substantially more costly compared with nonbiologics. •Prior studies from limited geographic regions of the United States suggest that certain patient characteristics, including lower income, minority race, and higher age, are negatively associated with biologic treatment in RA. What this study adds •This is the largest study based on U.S. commercial and Medicare population evaluating treatment predictors of tumor necrosis factor (TNF)-α inhibitor biologics conducted using a nationally representative sample of commercially insured RA patients. •Among RA patients on monotherapy nonbiologics, insurance generosity was found be a significant predictor of treatment initiation with TNF-α inhibitor biologics. However, among RA patients on combination therapy nonbiologics, the need for treatment, and not enabling characteristics such as insurance generosity, predicted treatment initiation with TNF-α inhibitor biologics. This observation demonstrates potential disparities related to patient cost sharing in the early stages of RA. Vol. 20, No. 11 www.amcp.org Predictors of Treatment Initiation with Tumor Necrosis Factor-α Inhibitors in Patients with Rheumatoid Arthritis R heumatoid arthritis (RA) is an autoimmune disease that affects approximately 1.3 million adults in the United States.1 RA is associated with substantial morbidity and mortality.2-4 Disease-modifying antirheumatic drugs (DMARDs), which are generally classified into nonbiologics and biologics, form the mainstay of RA management. Nonbiologic DMARDs include agents, such as methotrexate, sulfasalazine, hydroxychloroquine, and leflunomide, that halt disease progression by suppressing inflammation. In contrast, biologic DMARDs target specific components of the immune system, such as T cells, B cells, and cytokines (i.e., tumor necrosis factor (TNF)-α and interleukins), that play an important role in the pathogenesis of RA. Currently there are 10 biologics approved for the indication of RA: 5 TNF-α inhibitors (infliximab, etanercept, adalimumab, certolizumab, and golimumab), 2 interleukin inhibitors (tocilizumab and anakinra), a T-cell activation inhibitor (abatacept), a CD-20 activity blocker (rituximab), and a janus kinase inhibitor (tofacitinib). Among the available biologics, TNF-α inhibitors are the most commonly used agents, accounting for approximately 90% of the total biologic use.5 According to the American College of Rheumatology (ACR) recommendations, RA patients with low or moderate disease activity without features of poor prognosis should receive treatment with nonbiologic DMARDs, while RA patients with moderate-to-high disease severity with features of poor prognosis whose RA is not well controlled with nonbiologic DMARDs alone should receive treatment with biologic DMARDs.6,7 Since all the biologics are only available as brands, they are substantially more costly than nonbiologic DMARDs. According to 1 estimate, the total direct costs for biologics are approximately 5-fold greater than nonbiologic DMARDs.2 Given this cost, certain patient subgroups may have difficulty affording treatment, including patients with low income, less generous insurance coverage, and minority race. This is supported by studies that show lower biologic treatment initiation in RA patients with older age, lower income, and minority race.8-10 This disparity may lead to differences in such clinical outcomes as greater disease activity and lower remission rates in these patient subgroups.11 Reduction in overall health services utilization costs through sustained remission in RA is well documented.12 Therefore, from a payer’s point of view, it is very important to understand and address potential disparities in the use of biologics among RA patients in order to control future health care costs. Although previous studies have examined treatment disparities in particular subgroups,8-10 none has used a comprehensive model incorporating the variety of treatment determinants that might predict TNF-α inhibitor use. This study used Andersen’s Behavioral Model (ABM) of health services use to examine TNF-α inhibitor use in a large cohort of commercially insured RA patients from the United States.13 ABM is a theoretical www.amcp.org Vol. 20, No. 11 model that uses a combination of factors grouped into predisposing, enabling, and need factors in order to predict the use of health care services. In addition, this study expanded on the current literature, which includes studies conducted in limited geographic regions of the United States,8-10 by evaluating factors influencing treatment initiation with TNF-α inhibitors in a nationally representative sample of RA patients. We exclusively focus on TNF-α inhibitor biologics because non-TNF biologics are generally reserved for a select group of patients who either fail to respond to a TNF-α inhibitor agent or are at an increased risk of adverse events from TNF-α inhibitors.6 ■■ Methods Study Design and Data Source A retrospective cohort study was designed to evaluate the predictors of TNF-α inhibitor treatment initiation in RA patients who were aged 18 years and older using data from Truven’s MarketScan Commercial Claims And Encounters (CCAE) and Medicare Supplemental and Coordination of Benefits (COB) for the period between January 1, 2007, to December 31, 2010. These databases contain de-identified, person-specific health data including clinical utilization, expenditures, insurance enrollment/plan benefit, inpatient, outpatient, and prescription information. The CCAE contains health care data for nearly 40 million individuals, encompassing employees, their spouses, and their dependents. The COB contains the health care experiences of 3.8 million Medicare-eligible retirees with employersponsored Medicare supplemental plans.14 These patients have coordination of benefits , meaning that in addition to Medicare they have a private insurance plan paid for by their employers and therefore are not typical of the usual Medicare patient population. The Medicare supplemental dataset provided by Truven contains information on Medicare paid and supplemental insurance paid services. Patient Identification and Exclusion Criteria Diagnosis of RA was identified using the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) diagnosis code of 714.0 on at least 2 outpatient or 1 inpatient insurance claims between January 1, 2008, and December 31, 2009. The date of the first claim was defined as the index date. In order to ensure continuous availability of health care data, we required patients to be continuously enrolled in their health plans 12 months pre-index (defined as the baseline period) and 12 months post-index (defined as the follow-up period). Combining diagnosis codes with DMARD prescription fills is known to result in a high positive predictive value (> 85%) in identifying RA from administrative claims.15 Therefore, to improve the specificity of our RA identification algorithm, we further required these patients to have used at least 1 nonbiologic DMARD during the 12 months baseline period. November 2014 JMCP Journal of Managed Care & Specialty Pharmacy 1111 Predictors of Treatment Initiation with Tumor Necrosis Factor-α Inhibitors in Patients with Rheumatoid Arthritis FIGURE 1 Study Design 24 months Continuous enrollment in the health plan Baseline outpatient/inpatient visit with rheumatoid arthritis Prescription filled for a TNF-α inhibitor Index date January 1, 2007 12 months pre-index • Required nonbiologic DMARD monotherapy or combination therapy use • Excluded prevalent biologic DMARD users December 31, 2010 12 months follow-up Identified initiation of a TNF-α inhibitor DMARD = disease-modifying antirheumatic drug; TNF=tumor necrosis factor. Given our focus on TNF-α inhibitor initiation, we excluded patients who used any biologic DMARDs during the 12-month baseline period. To ensure RA-specific TNF-α inhibitor initiation, we excluded patients having psoriatic arthritis or Crohn’s disease (inflammatory conditions for which TNF-α inhibitor treatment is indicated). Further, to ensure that all the included patients were eligible to receive TNF-α inhibitors during follow-up, we excluded patients with a history of tuberculosis, which is a contraindication to TNF-α inhibitor use. Eligible RA patients were followed for 12 months beginning from their index dates to examine initiation of TNF-α inhibitors (Figure 1). To contrast patient characteristics between nonbiologic DMARD users and TNF-α inhibitors users, patients who either did not fill any DMARD prescription or initiated treatment with non-TNF biologics prior to initiating TNF-α inhibitors were excluded during follow-up. Additionally, in order to compare RA patients at different stages of the disease separately, we created 2 separate cohorts based on nonbiologic DMARD use during the baseline period: (1) RA patients on monotherapy nonbiologic DMARDs in the baseline period— this cohort represented RA patients with mild-to-moderate disease activity without features of poor prognosis—and (2) RA patients on combination therapy nonbiologic DMARDs in the baseline period—this cohort represented RA patients with moderate-or-high disease activity with features of poor prognosis. 1112 Journal of Managed Care & Specialty Pharmacy JMCP November 2014 Measures Predictors of biologic treatment initiation were measured during the 12-month baseline period in both cohorts and grouped according to the ABM for health services use.13 ABM posits a process of health care use in which predisposing factors influence the ability (measured through enabling factors) of a person to obtain health care that, when adding the need for treatment, predicts the use of health care services. Predisposing Factors. Predisposing factors included the variables that may influence the likelihood of receiving health care services. Predisposing factors of age and sex from the ABM have successfully predicted some health care services use in RA patients in the past.16 Therefore, we hypothesized that patient demographic factors including age (as a continuous variable), gender (male/female), geographic location (Northeast, Midwest, West, and South), and urban/rural residence (as determined by metropolitan statistical areas [MSA]) may be able to explain the use of TNF-α inhibitors in this population. Enabling Factors. Enabling factors included variables that may influence a patient’s ability to secure health care services. Because of the high cost of TNF-α inhibitors, based on the ABM we hypothesized that RA patients with better means to secure health care may initiate these agents more frequently. We included the following factors as enabling variables to capture a patient’s ability to secure health care services: visit to a rheumatologist as a categorical variable indicating no visit; 1 visit and more than 1 visit in the baseline period; the year of a patient’s Vol. 20, No. 11 www.amcp.org Predictors of Treatment Initiation with Tumor Necrosis Factor-α Inhibitors in Patients with Rheumatoid Arthritis index visit as a binary variable, 2008 or 2009; health plan type as a binary variable indicating capitated plan (included health maintenance organization or capitated point-of-service plans) or noncapitated plan (included basic major medical, comprehensive, exclusive and preferred provider organizations, noncapitated point of service, consumer-driven health plan, or high deductible health plan); type of insurance as a binary variable indicating either Medicare supplemental or commercial insurance; and drug benefit generosity. Drug benefit generosity was approximated by creating a “generosity index” using payment information from the prescriptions filled by patients in the 12-month baseline period.17 This index was calculated as a continuous variable in the range of 0-1 and was defined as the proportion of total drug costs paid by the patient out of pocket as copay or coinsurance. Based on this index, patients were classified into quartiles of drug benefit generosity to facilitate interpretation. The quartiles were termed as poor drug benefit generosity (fourth quartile, > 33% cost shared by the patients), average drug benefit generosity (third quartile, 20%-33% cost shared by the patients), above average drug benefit generosity (second quartile, 10%-20% cost shared by the patients), and most generous drug benefit (lowest out-of-pocket costs, first quartile, < 10% cost shared by the patients). Need Factors. Need factors included health conditions of patients that necessitate the utilization of health services. Since TNF-α inhibitors are reserved for patients whose RA is not well controlled with nonbiologic DMARDs,6 we hypothesized that patients with more severe RA, as captured by the need variables in the ABM, may initiate TNF-α inhibitors more frequently. In this set, we included a continuous measure approximating disease severity (claims-based index of RA severity [CIRAS]) validated in a previous study.18 We also added indicators for baseline steroid use, nonsteroidal anti-inflammatory drug (NSAIDs) use, and COX-2 inhibitor use based on at least 1 dispensing of these agents during the baseline period. The comorbidity profile of patients, which was calculated as a continuous score based on the presence of 20 individual comorbid conditions, was also included in this set.19 The outcome variable of interest was initiation of a TNF-α inhibitor agent. We dichotomized TNF-α inhibitor initiation as present or absent based on pharmacy or medical claims indicating use of these agents during the 12-month period following the index date. The following TNF-α inhibitors were included in this study: adalimumab, certolizumab, etanercept, golimumab, and infliximab. The use of these agents was identified using both the National Drug Code (NDC) numbers from outpatient pharmacy files for filled prescriptions and J codes using outpatient services files for injectable agents administered at physician offices. The following NDCs were used: 00074937402, 00074379901, 00074379902, 00074433902, 00074433906, 00074433907, and 54868482200 for adalimumab; 50474070062 and 50474071079 for certolizumab; 54868478200, 58406042534, 58406042541, 58406045501, www.amcp.org Vol. 20, No. 11 58406045504, 58406043501, 58406043504, 58406044501, and 58406044504 for etanercept; 57894007001 and 57894007002 for golimumab; and 57894003001 for infliximab. The following J codes were used: J0135 for adalimumab, J0718 for certolizumab, J1438 for etanercept, and J1745 for infliximab. Statistical Analyses Descriptive statistics were used to summarize patient characteristics among TNF-α inhibitor initiators and nonbiologic DMARD users. For dichotomous and categorical variables, the results were presented as numbers and proportions. For continuous variables, the results were presented as mean (± standard deviation). The patient factors were then compared between TNF-α inhibitor initiators and nonbiologic DMARD users with standardized differences.20 Standardized differences were used to avoid statistically significant differences that have limited clinical importance between our 2 groups owing to the large sample size. A standardized difference of less than 10 suggests no correlation between the variable in question and the treatment group. To understand the impact of various predictors on the initiation of TNF-α inhibitors while controlling for other variables, hierarchical logistic regression models were used in which the predictors were entered in 3 sets. The dependent variable in these models was a binary indicator for initiation of TNF-α inhibitors. The independent variables were grouped in 3 categories based on ABM: predisposing, enabling, and need variables. Predisposing variables were first included in the model followed by enabling variables and then need variables for both cohorts. Improvement in model fit was assessed using the Akaike information criterion (AIC) after addition of each set of variables. The goodness-of-fit of the logistic regression models were tested using Hosmer-Lemeshow tests. Linear equivalents of the logistic regression analyses were used to derive variance inflation factors (VIF), which were used to check for collinearity among the variables added to the model.21 All analyses were conducted using SAS version 9.2 (SAS Institute, Cary, NC). Sensitivity Analyses In order to evaluate the robustness of our findings, we undertook 2 sets of sensitivity analyses. First, Medicare and commercial enrollees may have different patient characteristics and coverage characteristics. Therefore, in order to evaluate whether our results are sensitive to pooling these patients and studying them as a single group, we fit logistic regression models predicting initiation of TNF-α inhibitors in Medicare and commercial enrollees separately in both cohorts. Second, certain TNF-α inhibitors that are administered at physician offices (most notably infliximab infusion) are likely to be covered under medical benefits, while other agents that are available as a self-injectable kit (e.g., etanercept) are more likely to be covered under pharmacy benefits. Therefore, to check whether our results apply to both physician-administered as November 2014 JMCP Journal of Managed Care & Specialty Pharmacy 1113 Predictors of Treatment Initiation with Tumor Necrosis Factor-α Inhibitors in Patients with Rheumatoid Arthritis FIGURE 2 Study Sample Derivation Patients identified as having RA between January 1, 2008, and December 31, 2009, who had 24 months continuous enrollment in their health plans Included n = 44,709 Total excluded n = 16,408 • Prevalent biologic use (15,246) • History of Chrohn’s disease or psoriatic arthritis (1,094) • History of tuberculosis (68) RA patients on nonbiologic DMARD treatment Included n = 28,201 Total excluded n = 2,052 • No DMARD prescriptions filled during follow-up (1,599) • Initiated non-TNF-α inhibitor biologic during follow-up (453) RA patients on nonbiologic DMARD treatment who continue nonbiologic DMARDs or initiate a TNF-α inhibitor Included n = 26,149 Cohort 1 RA patients on monotherapy nonbiologic DMARDs at baseline (n = 18,641) Cohort 2 RA patients on combination therapy nonbiologic DMARDs at baseline (n = 7,508) DMARD = disease-modifying antirheumatic drug; RA = rheumatoid arthritis; TNF = tumor necrosis factor. well as prescription TNF-α inhibitors, we fit separate logistic regression models predicting initiation of both types of TNF-α inhibitors in both cohorts. ■■ Results Derivation of Study Cohorts Figure 2 shows application of the inclusion and exclusion criteria for this study. We identified 44,709 RA patients who had at least 12 months pre-index and 12 months post-index continuous enrollment in their health plans. After excluding prevalent biologic users (15,246), patients ineligible for TNF-α inhibitor initiation (68), patients with comorbid inflammatory conditions (1,094), patients with no DMARD use during follow-up (1,599), and initiators of non-TNF-biologics (453), a total of 26,149 patients met all our inclusion criteria. These patients were then divided into 2 cohorts. Cohort 1, which included RA patients on monotherapy nonbiologic DMARDs in the baseline period, comprised 18,641 patients, and cohort 2, which included RA patients on combination therapy nonbiologic DMARDs, comprised 7,508 patients. 1114 Journal of Managed Care & Specialty Pharmacy JMCP November 2014 Patient Characteristics A total of 1,922 patients (10.31%) among monotherapy nonbiologic users (cohort 1) and a total of 983 patients (13.09%) among combination therapy nonbiologic users (cohort 2) initiated treatment with a TNF-α inhibitor during the 12-month followup period. Table 1 compares the baseline characteristics of the TNF-α inhibitor initiators with patients who continued treatment with nonbiologic DMARDs during the follow-up period. Comparison of the predisposing variables suggested that the TNF-α inhibitor initiators were younger in both cohorts (mean age: 54 years vs. 62 years, standardized difference (SD) = 63.16 in cohort 1; 53 years vs. 60 years, SD = 57.18 in cohort 2). In both cohorts, patients in the South initiated TNF-α inhibitors more frequently, while patients in the Midwest initiated these agents less frequently. For the enabling variables, a lower proportion of the TNF-α inhibitor initiators had not visited a rheumatologist in the prior year compared with noninitiators only among monotherapy nonbiologic DMARD users (39.23% vs. 46.07%, SD = 13.87). The type of insurance was less frequently Medicare among the TNF-α inhibitor initiators in both cohorts Vol. 20, No. 11 www.amcp.org Predictors of Treatment Initiation with Tumor Necrosis Factor-α Inhibitors in Patients with Rheumatoid Arthritis TABLE 1 Patient Characteristics of TNF-α Inhibitors Initiators and Noninitiators Stratified by Baseline DMARD Treatment, 2007-2010 Cohort 1: Monotherapy Nonbiologic DMARD Users at Baseline (n=18,641) TNF-α Inhibitor TNF-α Inhibitor Initiators Noninitiators (n = 1,922) (n = 16,719) Cohort 2: Combination Therapy Nonbiologic Users at Baseline (n=7,508) TNF-α Inhibitor TNF-α Inhibitor Initiators Noninitiators (n = 983) (n = 6,525) Standardized Standardized Variable n (%) n (%) n (%) n (%) Differencea Differencea Predisposing factors Patient age, in years, mean (SD) 63.2 57.2 54(12.4) 62(13.5) 53(11.9) 60(12.6) Female 5.5 7.4 1,477(76.9) 12,456(74.5) 783(79.7) 4,998(76.6) Metropolitan statistical area 1,567(81.6) 13,596(81.3) 0.6 791(80.5) 5,295(81.2) 1.8 Region Northeast 10.2 5.4 158(8.2) 1,882(11.3) 69(7.0) 552(8.4) North Central 507(26.4) 5,372(32.1) 12.7 278(28.3) 2,326(35.6) 15.9 South 22.2 20.7 959(49.9) 6,509(38.9) 450(45.8) 2,328(35.7) 5.8 3.3 West 298(15.5) 2,956(17.7) 186(18.9) 1,319(20.2) Enabling factors Capitation Noncapitated health plan 0.4 2.2 1,618(84.2) 14,101(84.3) 806(82.0) 5,404(82.8) Capitated health plan 304(15.8) 2,618(15.7) 177(18.0) 1,121(17.2) Visits to rheumatologists No visit in the prior year 13.9 5.9 754(39.2) 7,703(46.1) 398(40.5) 2,833(43.4) 7.9 0.5 At least 1 visit in the prior year 183(9.5) 1,226(7.3) 54(5.5) 366(5.6) More than 1 visit in the prior year 9.3 6.1 985(51.2) 7,790(46.6) 531(54.0) 3,326(50.9) Calendar year of the index visit 2008 21.6 17.8 1,214(63.2) 8,788(52.6) 675(68.7) 3,928(60.2) 2009 708(36.8) 7,931(48.4) 301(31.3) 2,597(39.8) Payer type 54.1 46.8 Commercial 1,647(85.7) 10,508(62.8) 847(86.2) 4,362(66.8) Medicare 275(14.3) 6,211(37.1) 136(13.8) 2,163(33.1) Drug benefit generosityb Most generous 435(22.7) 4,177(25.1) 5.5 199(20.2) 1,706(26.2) 14.0 Better than average 472(24.7) 4,096(24.6) 0.1 263(26.7) 1,686(25.8) 2.0 485(25.3) 4,159(25.0) 262(26.6) 1,611(24.7) Average 0.8 4.5 522(27.3) 4,218(25.3) 259(26.3) 1,518(23.3) Below average 4.4 7.1 Need factors 6.0(1.9) 4.8(1.9) 12.7 6.1(1.9) 4.9(1.8) 15.8 CIRAS,c mean (SD) Comedications of interest COX-2 inhibitors 260(13.5) 1,893(11.3) 6.7 141(14.3) 777(11.9) 7.2 Nonsteroidal anti-inflammatory drugs 573(29.8) 3,630(21.7) 18.6 322(32.8) 1,462(22.4) 23.3 Steroids 1,391(72.4) 9,668(57.8) 30.9 774(78.7) 4,067(62.3) 36.6 0.3(1.1) 0.5(1.4) 12.2 0.4(1.1) 0.5(1.4) 10.6 Combined comorbidity score (CCS)d Selected individual comorbid conditions from CCSd Congestive heart failure 70(3.6) 1,035(6.2) 11.8 27(2.7) 420(6.4) 17.7 Any tumor 65(3.4) 1,254(7.5) 18.2 39(4.0) 469(7.2) 14.1 89(4.6) 1,391(8.3) 46(4.7) 502(7.7) Cardiac arrhythmias 15.0 12.5 Hypertension 687(35.7) 6,976(41.7) 12.3 323(32.9) 2,644(40.5) 15.9 a A standardized difference of 10 (approximately equivalent to P < 0.05) indicates significant imbalance of a baseline covariate. bDrug benefit generosity was classified according to the quartiles of a calculated generosity index. This index was calculated as a continuous variable and defined as the proportion of total drug cost paid by the patient out of pocket. The quartiles were termed as poor drug benefit generosity (highest out-of-pocket costs, fourth quartile), average drug benefit generosity (third quartile), above average drug benefit generosity (second quartile), and most generous drug benefit (lowest out-of-pocket costs, first quartile). cCIRAS: Claims-based index of rheumatoid arthritis severity, which ranged from 0.6 to 10.7, with higher values indicating more severe rheumatoid arthritis. d A composite score indicating a patient’s comorbidity burden after taking into account 20 individual conditions. The score ranged from -2 to 13 in our cohorts, with greater values indicating higher disease burden. From 20 individual conditions, only those with a standardized difference > 10 were shown. DMARD = disease-modifying antirheumatic drug; SD = standard deviation; TNF = tumor necrosis factor. www.amcp.org Vol. 20, No. 11 November 2014 JMCP Journal of Managed Care & Specialty Pharmacy 1115 Predictors of Treatment Initiation with Tumor Necrosis Factor-α Inhibitors in Patients with Rheumatoid Arthritis TABLE 2 Multivariate Predictors of Treatment Initiation with TNF-α Inhibitors in RA Patients, 2007-2010 Predicting Treatment Initiation with TNF-α Inhibitors, OR (95% CI) Cohort 1: Among Cohort 2: Among Monotherapy Combination Nonbiologic Users Nonbiologic Users at Baseline at Baseline Variable Predisposing factors Patient age 0.98 (0.97-0.98) 0.98 (0.97-0.99) Gender Female 1 1 Male 0.99 (0.88-1.11) 0.93 (0.78-1.11) Metropolitan statistical area (MSA) Non-MSA 1 1 MSA 1.09 (0.96-1.24) 0.95 (0.79-1.13) Region South 1 1 North Central 0.83 (0.73-0.94) 0.81 (0.68-0.96) Northeast 0.77 (0.64-0.92) 0.84 (0.63-1.11) West 0.86 (0.74-0.99) 0.94 (0.77-1.14) Enabling factors Capitation Noncapitated health plan 1 1 Capitated plan 0.95 (0.83-1.09) 1.01 (0.84-1.22) Visit to rheumatologist in the pre-index period No visit 1 1 At least 1 visit 1.21 (1.01-1.45) 0.87 (0.64-1.20) More than 1 visit 0.95 (0.85-1.06) 0.75 (0.64-0.89) Calendar year of index visit 2008 1 1 2009 1.09 (0.97-1.22) 1.16 (0.98-1.38) Insurance type Commercial 1 1 Medicare 0.75 (0.62-0.91) 0.90 (0.68-1.19) Drug benefit generositya Poor 1 1 Average 1.03 (0.90-1.18) 0.97 (0.80-1.18) Better than average 1.16 (1.01-1.33) 1.01 (0.83-1.23) Most generous 1.20 (1.04-1.39) 0.89 (0.72-1.10) Need factors 1.19 (1.14-1.23) 1.28 (1.21-1.35) RA severity (CIRAS)b Pain medication use No steroid use 1 1 Steroid use 1.80 (1.62-2.01) 2.04 (1.73-2.41) No COX-2 inhibitor use 1 1 COX-2 inhibitor use 1.22 (1.06-1.41) 1.22 (0.99-1.50) No NSAID use 1 1 NSAID use 1.17 (1.05-1.31) 1.36 (1.17-1.59) 0.94 (0.90-0.98) 0.95 (0.90-1.01) Combined comorbidity scorec a Drug benefit generosity was classified according to the quartiles of a calculated generosity index. This index was calculated as a continuous variable and defined as the proportion of total drug cost paid by the patient out of pocket. The quartiles were termed as poor drug benefit generosity (highest out-of-pocket costs, fourth quartile), average drug benefit generosity (third quartile), above average drug benefit generosity (second quartile), and most generous drug benefit (lowest out-of-pocket costs, first quartile). bCIRAS: Claims-based index of rheumatoid arthritis severity, which ranged from 0.6 to 10.7 with higher values indicating more severe rheumatoid arthritis. c A composite score indicating a patient’s comorbidity burden after taking into account 20 individual conditions. The score ranged from -2 to 13 in our cohorts, with greater values indicating higher disease burden. CI = confidence interval; NSAID = nonsteroidal anti-inflammatory drug; OR = odds ratio; TNF = tumor necrosis factor. 1116 Journal of Managed Care & Specialty Pharmacy JMCP November 2014 (14.31% vs. 37.15%, SD = 54.13 in cohort 1; 13.84% vs. 33.15%, SD = 46.79 in cohort 2). Among need variables, the severity of RA was found to be significantly greater among TNF-α inhibitors in both cohorts (mean CIRAS 5.96 vs. 4.70, SD = 12.67 in cohort 1; 6.12 vs. 4.89, SD = 15.85 in cohort 2). In both cohorts, a higher proportion of patients used NSAIDs during the baseline period compared with the noninitiator groups (29.81% vs. 21.71%, SD = 18.60 in cohort 1; 32.76% vs. 22.41%, SD = 23.32 in cohort 2). Steroid use was also more frequent in the TNF-α initiator groups in both cohorts (72.37% vs. 57.83%, SD = 30.88 in cohort 1; 78.74% vs. 62.33%, SD = 36.59 in cohort 2). Overall comorbidity burden was lower in the TNF-α inhibitor initiator group in both cohorts (mean score 0.30 vs. 0.46, SD = 12.19 in cohort 1; 0.36 vs. 0.49, SD = 10.62 in cohort 2). Predictors of TNF-α Inhibitor Initiation The results of our multivariate models that evaluated the influence of various predictors on treatment initiation with TNF-α inhibitors are presented in Table 2. The goodnessof-fit for both models was found to be adequate (P for Hosmer-Lemeshaw > 0.05), and no evidence for collinearity was observed for the variables added to the model (VIF < 5 for all the variables). Among monotherapy nonbiologic users, the predisposing variables of patient age and geographic region were found to be significant predictors of TNF-α inhibitor initiation. Each year increase in age reduced the odds of TNF-α inhibitor initiation by 2% (odds ratio [OR] = 0.98, 95% confidence interval [CI] = 0.97-0.98). Patients in the Midwest, Northeast, and West regions had significantly lower likelihood of treatment initiation with TNF-α inhibitors compared with patients in the South (OR = 0.83, 95% CI = 0.73-0.93; OR = 0.77, 95% CI = 0.64-0.92; and OR = 0.86, 95% CI = 0.74-0.99, respectively). Of the enabling variables, having Medicare supplemental insurance had lower likelihood of TNF-α inhibitor treatment initiation compared with having commercial insurance (OR = 0.75, 95% CI = 0.62-0.91). On the other hand, patients who visited their rheumatologists once in the prior year had 21% higher odds of initiating a TNF-α inhibitor compared with those who did not visit their rheumatologists at all (OR = 1.21, 95% CI = 1.01-1.45). The drug benefit generosity variable also significantly predicted treatment initiation with a TNF-α inhibitor. Patients with better than average and the most generous drug benefit had 16% and 21% higher odds of initiating a TNF-α inhibitor compared with patients with poor drug benefits (OR = 1.16, 95% CI = 1.01-1.33; OR = 1.20, 95% CI = 1.041.39). All the need variables were found to have an association with TNF-α inhibitor initiation. With each unit increase in RA severity measure (CIRAS), the odds of TNF-α inhibitor initiation increased by 19% (OR = 1.19, 95% CI = 1.15-1.23). Previous use of steroids raised the odds of TNF-α inhibitor initiation by 80% (OR = 1.80, 95% CI = 1.62-2.01); previous use of NSAIDs raised the odds by 17% (OR = 1.17, 95% CI = 1.05- Vol. 20, No. 11 www.amcp.org Predictors of Treatment Initiation with Tumor Necrosis Factor-α Inhibitors in Patients with Rheumatoid Arthritis 1.31); and previous use of COX-2 inhibitors raised the odds by 22% (OR = 1.22, 95% CI = 1.06-1.41). Each unit decrease in the combined comorbidity score lowered the odds of TNF-α inhibitor initiation by 6% (OR = 0.94, 95% CI = 0.90-0.99). Addition of each sets of variables in the model improved the model fit for this cohort (AIC for the model with predisposing factors only = 11,667; predisposing + enabling factors = 11,640; and predisposing + enabling + need factors = 11,386). Some interesting contrasts were observed in the model predicting TNF-α inhibitor initiation among combination therapy nonbiologic users. Similar to monotherapy nonbiologic users, higher age was significantly associated inversely with TNF-α inhibitor initiation (OR = 0.98, 95% CI = 0.97-0.99), and patients in the Midwest had lower odds of initiating treatment with TNF-α inhibitors compared with patients in the South (OR = 0.81, 95% CI = 0.68-0.96). However, none of the enabling factors that predicted TNF-α inhibitor initiation in cohort 1 were found to be significantly associated with TNF-α inhibitor initiation in this cohort. Surprisingly, having visited a rheumatologist more than once in the prior year reduced the odds of TNF-α inhibitor initiation by 24%, compared with having no visit (OR = 0.75, 95% CI = 0.64-0.89). A stronger association between TNF-α inhibitor initiation and RA-related need factors, including CIRAS (OR = 1.28, 95% CI = 1.21-1.35), steroids use (OR = 2.04, 95% CI = 1.73-2.41), and NSAID use (OR = 1.36, 95% CI = 1.17-1.59) was observed in this cohort. Consistent with these observations, it was also noted that the addition of enabling variables to the model did not improve the model fit for this cohort, but addition of need variables improved the model fit (AIC for the model with predisposing factors only = 5,559; predisposing + enabling factors = 5,563; and predisposing + enabling + need factors = 5,364). Sensitivity Analyses Findings In our first sensitivity analysis where we fit separate models predicting TNF-α inhibitors in commercial and Medicare enrollees, no noticeable differences in trends were observed compared with the main analysis in important explanatory variables including age, drug benefit generosity, and RA-related factors including CIRAS and steroid use (Appendix A, available in online article). However, since the estimates in Medicare were based on fewer TNF-α inhibitor initiations (275 in monotherapy and 136 in combination therapy) compared with estimates in commercial enrollees (1,647 in monotherapy and 847 in combination therapy), we observed estimates in Medicare enrollees with wider CIs. Additionally, certain factors, including gender, MSA, and visits to rheumatologists, were observed to have estimates that were numerically inconsistent (meaning on different sides of the null value of 1.0) between the 2 data sources. However, in all instances, the 95% CI for these estimates demonstrated considerable overlap between the 2 data sources. In our second sensitivity analysis where we fit separate logistic regression models predicting initiation of physician-admin- www.amcp.org Vol. 20, No. 11 istered and prescription-filled TNF-α inhibitors, the majority of the findings were similar (Appendix B, available in online article). However, the drug benefit generosity variable was a stronger predictor of TNF-α inhibitors filled at a pharmacy, while it did not predict the initiation of physician-administered TNF-α inhibitors in the monotherapy cohort, unlike our main analysis. ■■ Discussion Findings from the current study provide insights into realworld treatment initiation patterns of TNF-α inhibitors in patients with RA. One of the purposes of our study was to examine potential disparities in treatment using a well-defined conceptual model. As suggested by the ABM, under an equitable health care system, the use of health care services would primarily be driven by need factors. However, we found significant variation in TNF-α inhibitor initiation across patients, with predisposing factors—including age and geographic region—as well as enabling factors—including visit to rheumatologists, drug benefit generosity, and insurance type— playing a role in treatment initiation with TNF-α inhibitors. This is potentially suggestive of inequitable access among RA patients. A recent investigation observed that close to 50% of RA patients did not receive care consistent with the 2008 ACR guidelines.22 Our study identifies some of the potential factors that may be contributing to this worrisome trend. In order to better characterize factors influencing treatment initiation with TNF-α inhibitors, we separately evaluated the effects of various sets of predictors in 2 cohorts of patients with different stages of RA, as suggested by either monotherapy or combination therapy nonbiologic use during the baseline period. We observed that in the cohort of monotherapy nonbiologic users, patients with certain demographics (younger age or residence in the South) and with better means to secure health care (care of rheumatologists or health plans with a higher drug benefit generosity) had higher odds of initiating treatment with TNF-α inhibitors. These results suggest that during the early stages of the disease, potential disparities in access to the costly TNF-α inhibitors may exist. Delay in initiation of timely TNF-α inhibitors may lead to higher probability of radiographic progression and hence reduced quality of life among these patients.23 Prior research has also demonstrated that RA patients with multiple failed nonbiologic DMARDs prior to initiating a TNF-α inhibitor have lower odds for treatment response with TNF-α inhibitors.24 This further emphasizes the importance of timely initiation of TNF-α inhibitors in RA patients. Because of the substantially higher cost of biologics, there is a potential for inequitable access in the use of these agents among RA patients. The coverage of biologics under a higher or specialty formulary tier of pharmacy benefits has become increasingly common.25 Research suggests that this practice has substantially increased the out-of-pocket costs for biologics.25,26 Greater patient cost sharing has been known November 2014 JMCP Journal of Managed Care & Specialty Pharmacy 1117 Predictors of Treatment Initiation with Tumor Necrosis Factor-α Inhibitors in Patients with Rheumatoid Arthritis to delay or reduce the odds of initiation of treatments in various disease conditions,27 including RA.28 In the current study, we reported findings in line with these observations among RA patients who were on monotherapy nonbiologic DMARDs at baseline. Reduction in patient cost sharing may represent a potential strategy for payers to increase the odds of timely biologic initiation. Among RA patients with moderate-or-high disease activity with features of poor prognosis (as approximated by combination nonbiologic DMARDs use at baseline in cohort 2), we observed that need factors mostly explained the initiation of TNF-α inhibitors, and enabling factors, such as insurance generosity and insurance type, played little role. This finding suggests less potential for disparity in TNF-α inhibitor treatment use among commercially insured patients with higher need for treatment. Although our finding of having visited a rheumatologist more than once in the prior year resulting in lower odds of TNF-α inhibitor initiation compared with no visit in this cohort may seem counterintuitive at first glance, we postulate that this may reflect improved RA management under the constant care of a rheumatologist, which may in turn result in lower need for TNF-α inhibitor initiation in these patients. It was interesting to note that TNF-α inhibitor initiators were younger than patients not initiating these treatments. This is consistent with prior studies that evaluated initiation of biologics specifically,8-10 as well as several studies that examined the use of DMARDs as a class.29-31 This inverse association between age and TNF-α inhibitor initiation may be attributed to several factors. It is likely that older patients may be at a higher risk for adverse events of TNF-α inhibitors, owing to a higher burden of comorbid conditions and frailty. Although the literature suggests similar effectiveness of TNF-α inhibitors across different age groups,32,33 our finding of their differential use based on age is concerning because it may reflect less aggressive RA management and possibly uncontrolled RA in older patients. Another factor leading to less aggressive treatment in older RA patients may be physician preference.34 We also observed that patients in the South were more likely to initiate treatment with TNF-α inhibitors. Ours is the first study to document regional variations in the initiation of treatment with TNF-α inhibitors. The regional variation we observed persisted after controlling for other predisposing, enabling, and need factors. Prior research has documented substantial regional variation in prescription medication utilization among Medicare Part D enrollees in the United States, and some of the factors contributing to the geographic variation may include prescriber practice styles, prescriber awareness, and patient preferences.35 The significant association of higher RA severity score and pre-index pain medication use with TNF-α inhibitor treatment initiation is an expected finding because these need variables represent high RA activity. We also observed strong trends towards lower likelihood of TNF-α inhibitor initiation among 1118 Journal of Managed Care & Specialty Pharmacy JMCP November 2014 patients with higher combined comorbidity scores. This finding may reflect the fact that TNF-α inhibitors are contraindicated in a variety of comorbid conditions, including congestive heart failure, multiple sclerosis, and a variety of infections, while nonbiologic DMARDs are not.36 Therefore, it is possible that physicians may avoid TNF-α inhibitor treatment in RA patients with a higher burden of comorbidities. Our study has several unique strengths. First, this is the largest study of its kind conducted in RA patients from all over the United States, who are enrolled in commercial or Medicare supplemental insurance, that provides estimates on the influence of population characteristics on TNF-α inhibitor treatment initiation. Second, because of the availability of diagnoses for various comorbid conditions within the claims, we were able to exclude patients with contraindications and risk-adjust our estimates based on the presence of various comorbidities. Third, we carefully constructed 2 cohorts of RA patients according to their disease progression based on their baseline DMARD use and predicted initiation of TNF-α inhibitors separately in each cohort. This approach ensured the inclusion of homogenous groups of patients in each cohort and provided insights into factors that were differentially associated with TNF-α inhibitor initiation in each of the cohorts. Finally, we conducted extensive sensitivity analyses to evaluate the robustness of our findings. Limitations We also acknowledge several limitations of this study. As with any other study using administrative claims, we were not able to validate the diagnoses of the disease condition (RA). To address this limitation, we used nonbiologic DMARD prescriptions in pharmacy claims along with ICD-9-CM codes on inpatient or outpatient visit to identify RA. Combining DMARD claims with diagnosis codes has been shown to result in a high positive predictive value (86%) for identifying RA in administrative claims in a prior validation study.15 Further, the administrative claims contain very limited information on clinical conditions of RA patients, such as disease activity and swollen joint count. Therefore, we were not able to capture the exact severity of RA in patients in our cohorts. However, as a proxy, we used the validated claims-based index for getting an approximation of RA severity.18 Next, because of the unavailability of information on important patient factors, including race, education, income, and family medical cost burden, our study cannot explain potential disparities in TNF-α inhibitor initiation owing to these factors. Additionally, claims data may have incomplete information on certain variables. For instance, provider type is coded as “unknown” on some physician visits. This may have artificially inflated the number with zero visits to rheumatologists and deflated the number with 1 and more than 1 visit. As a result, the absolute percentage reported in Table 1 may not represent a typical care pattern by Vol. 20, No. 11 www.amcp.org Predictors of Treatment Initiation with Tumor Necrosis Factor-α Inhibitors in Patients with Rheumatoid Arthritis rheumatologists of RA patients with commercial or Medicare supplemental insurance. However, we do not expect the amount of missing information to be related to the initiation of TNF-inhibitors. Therefore, we postulate that our effect estimates in Table 2 are not systematically biased due to this problem. Next, an important factor leading to TNF-α inhibitor initiation may be physician preference, independent of patient need for treatment.37 Since we did not have information about this important variable in our data, our study cannot explain potential variability related to physician preferences. Finally, the insurance claims data only represent employed individuals and their dependents, and the Medicare supplemental data only represent retirees whose insurance are paid by their employers, which somewhat limits the generalizability of our study. ■■ Conclusions Potential disparities in the initiation of TNF-α inhibitors among RA patients on monotherapy DMARDS at baseline were noted among older patients, patients in certain geographic regions of the United States, and patients with less generous prescription drug benefits among commercial and Medicare supplemental participants. Although future research should examine the impact of these disparities on health outcomes, payers should be aware of the potential for undertreatment among these groups of RA patients when making formulary decisions. Among patients on combination therapy DMARDs, little impact of enabling factors, including drug benefit generosity and data type on the initiation of TNF-α inhibitors, was observed. Future research using data that have detailed information on drug benefit structure of health plans of the patients should be considered to confirm our findings. Authors RISHI J. DESAI, PhD, is Postdoctoral Research Fellow, Division of Pharmacoepidemiology and Pharmacoeconomics, Department of Medicine, Harvard Medical School and Brigham & Women’s Hospital, Boston, Massachusetts. GANG FANG, PhD, is Assistant Professor; and JOEL F. FARLEY, PhD, is Associate Professor, Division of Pharmaceutical Outcomes and Policy, University of North Carolina Eshelman School of Pharmacy, Chapel Hill, North Carolina. JAYA K. RAO, MD, is Deputy Editor, Annals of Internal Medicine, Philadelphia, Pennsylvania; RICHARD A. HANSEN, PhD, is Professor, Department of Pharmacy Care Systems, Harrison School of Pharmacy, Auburn University, Auburn, Alabama; and MATTHEW L. MACIEJEWSKI, PhD, is Associate Professor, Division of General Internal Medicine, Department of Medicine, Duke University Medical Center, Durham, North Carolina. AUTHOR CORRESPONDENCE: Joel F. Farley, PhD, Associate Professor, UNC Eshelman School of Pharmacy, CB 7573 Kerr Hall, Rm. 2201, Chapel Hill, NC 27599-7573. Tel.: 919.966.9973; Fax: 919.966.8486; E-mail: [email protected]. www.amcp.org Vol. 20, No. 11 DISCLOSURES This study was not supported by any external funding institution. Farley, Maciejewski, and Hansen have received consulting support for unrelated projects from Daiichi Sankyo and Novartis Pharmaceuticals. Hansen has provided expert testimony for Allergan. Rao reports owning stocks in Pfizer and Eli Lilly. No other authors have any conflict of interest to report. Study concept and design were primarily contributed by Desai and Farley, with assistance from Rao, Hensen, Fang, and Maciejewski. Desai and Farley collected the data, which were interpreted by Maciejewski, Hansen, and Desai, assisted by Rao, Fang, and Farley. The manuscript was written by Desai, Fang, Rao, and Hansen, assisted by Farley and Maciejewski, and revised by Desai, Rao, Hansen, and Fang, assisted by Maciejewski and Farley. References 1. Helmick CG, Felson DT, Lawrence RC, et al. Estimates of the prevalence of arthritis and other rheumatic conditions in the United States. Part I. Arthritis Rheum. 2008;58(1):15-25. 2. Michaud K, Messer J, Choi HK, Wolfe F. Direct medical costs and their predictors in patients with rheumatoid arthritis: a three-year study of 7,527 patients. Arthritis Rheum. 2003;48(10):2750-62. 3. Pugner KM, Scott DI, Holmes JW, Hieke K. The costs of rheumatoid arthritis: an international long-term view. Semin Arthritis Rheum. 2000;29(5):305-20. 4. Cooper NJ. Economic burden of rheumatoid arthritis: a systematic review. Rheumatology (Oxford). 2000;39(1):28-33. 5. McBride S, Sarsour K, White LA, Nelson DR, Chawla AJ, Johnston JA. Biologic disease-modifying drug treatment patterns and associated costs for patients with rheumatoid Arthritis. J Rheumatol. 2011;38(10):2141-49. 6. Singh JA, Furst DE, Bharat A, et al. 2012 Update of the 2008 American College of Rheumatology recommendations for the use of disease modifying antirheumatic drugs and biologic agents in the treatment of rheumatoid arthritis. Arthritis Care Res (Hoboken). 2012;64(5):625-39. 7. Nam JL, Winthrop KL, van Vollenhoven RF, et al. Current evidence for the management of rheumatoid arthritis with biological diseasemodifying antirheumatic drugs: a systematic literature review informing the EULAR recommendations for the management of RA. Ann Rheum Dis. 2010;69(6):976-86. 8. Yelin E, Tonner C, Kim SC, et al. Sociodemographic, disease, health system, and contextual factors affecting the initiation of biologic agents in rheumatoid arthritis: a longitudinal study. Arthritis Care Res (Hoboken). 2014;66(7):980-89. 9. Chu LH, Portugal C, Kawatkar AA, Stohl W, Nichol MB. Racial/ethnic differences in the use of biologic disease-modifying antirheumatic drugs among California Medicaid rheumatoid arthritis patients. Arthritis Care Res (Hoboken). 2013;65(2):299-303. 10. DeWitt EM, Lin L, Glick HA, Anstrom KJ, Schulman KA, Reed SD. Pattern and predictors of the initiation of biologic agents for the treatment of rheumatoid arthritis in the United States: an analysis using a large observational data bank. Clin Ther. 2009;31(8):1871-80. 11. Greenberg JD, Spruill TM, Shan Y, et al. Racial and ethnic disparities in disease activity in patients with rheumatoid arthritis. Am J Med. 2013;126(12):1089-98. 12. Barnabe C, Thanh N, Ohinmaa A, et al. Healthcare service utilisation costs are reduced when rheumatoid arthritis patients achieve sustained remission. Ann Rheum Dis. 2013;72(10):1664-68. 13. Andersen RM. Revisiting the behavioral model and access to medical care: does it matter? J Health Soc Behav. 1995;36(1):1-10. 14. Danielson E. Health research data for the real world: the MarketScan databases. White paper. January 2014. Available at: http://truvenhealth.com/ Portals/0/Users/031/31/31/PH_13434%200314_MarketScan_WP_web.pdf. Accessed September 30, 2014. November 2014 JMCP Journal of Managed Care & Specialty Pharmacy 1119 Predictors of Treatment Initiation with Tumor Necrosis Factor-α Inhibitors in Patients with Rheumatoid Arthritis 15. Kim SY, Servi A, Polinski JM, et al. Validation of rheumatoid arthritis diagnoses in health care utilization data. Arthritis Res Ther. 2011;13(1):R32. 16. Jacobi CE, Triemstra M, Rupp I, Dinant HJ, Van Den Bos GA. Health care utilization among rheumatoid arthritis patients referred to a rheumatology center: unequal needs, unequal care? Arthritis Rheum. 2001;45(4):324-30. 17. Artz MB, Hadsall RS, Schondelmeyer SW. Impact of generosity level of outpatient prescription drug coverage on prescription drug events and expenditure among older persons. Am J Public Health. 2002;92(8):1257-63. 18. Ting G, Schneeweiss S, Scranton R, et al. Development of a health care utilisation data-based index for rheumatoid arthritis severity: a preliminary study. Arthritis Res Ther. 2008;10(4):R95. 19. Gagne JJ, Glynn RJ, Avorn J, Levin R, Schneeweiss S. A combined comorbidity score predicted mortality in elderly patients better than existing scores. J Clin Epidemiol. 2011;64(7):749-59. 20. Austin PC. Using the standardized difference to compare the prevalence of a binary variable between two groups in observational research. Commun Stat Simul Comput. 2009;38(6):1228-34. 27. Solomon MD, Goldman DP, Joyce GF, Escarce JJ. Cost sharing and the initiation of drug therapy for the chronically ill. Arch Intern Med. 2009;169(8):740-48. 28. Karaca-Mandic P, Joyce GF, Goldman DP, Laouri M. Cost sharing, family health care burden, and the use of specialty drugs for rheumatoid arthritis. Health Serv Res. 2010;45(5 Pt 1):1227-50. 29. Bonafede MM, Fox KM, Johnson BH, Watson C, Gandra SR. Factors associated with the initiation of disease-modifying antirheumatic drugs in newly diagnosed rheumatoid arthritis: a retrospective claims database study. Clin Ther. 2012;34(2):457-67. 30. Desai R, Agarwal S, Aparasu R. Drug utilization trends for arthritis and other rheumtic conditions and impact of patients’ age on treatment choice. N C Med J. 2011;72(6):432-38. 31. Tutuncu Z, Reed G, Kremer J, Kavanaugh A. Do patients with olderonset rheumatoid arthritis receive less aggressive treatment? Ann Rheum Dis. 2006;65(9):1226-29. 21. Allison PD. Logistic Regression Using the SAS System: Theory and Application. Wiley-SAS Publishing; 2001. 32. Bathon JM, Fleischmann RM, Van der Heijde D, et al. Safety and efficacy of etanercept treatment in elderly subjects with rheumatoid arthritis. J Rheumatol. 2006;33(2):234-43. 22. Harrold LR, Harrington JT, Curtis JR, et al. Prescribing practices in a U.S. cohort of rheumatoid arthritis patients before and after publication of the American College of Rheumatology treatment recommendations. Arthritis Rheum. 2012;64(3):630-38. 33. Genevay S, Finckh A, Ciurea A, Chamot AM, Kyburz D, Gabay C. Tolerance and effectiveness of anti-tumor necrosis factor alpha therapies in elderly patients with rheumatoid arthritis: a population-based cohort study. Arthritis Rheum. 2007;57(4):679-85. 23. Keystone EC, Breedveld FC, van der Heijde D, et al. Long-term effect of delaying combination therapy with tumor necrosis factor inhibitor in patients with aggressive early rheumatoid arthritis: 10-year efficacy and safety of adalimumab from the randomized controlled PREMIER trial with open-label extension. J Rheumatol. 2014;41(1):5-14. 34. Fraenkel L, Rabidou N, Dhar R. Are rheumatologists’ treatment decisions influenced by patients’ age? Rheumatology. 2006;45(12):1555-57. 24. Hyrich KL, Watson KD, Silman AJ, Symmons DP; British Society for Rheumatology Biologics Register. Predictors of response to anti-TNFalpha therapy among patients with rheumatoid arthritis: results from the British Society for Rheumatology Biologics Register. Rheumatology (Oxford). 2006;45(12):1558-65. 25. Goldman DP, Joyce GF, Lawless G, Crown WH, Willey V. Benefit design and specialty drug use. Health Aff (Millwood). 2006;25(5):1319-31. 26. Polinski JM, Mohr PE, Johnson L. Impact of Medicare Part D on access to and cost sharing for specialty biologic medications for beneficiaries with rheumatoid arthritis. Arthritis Rheum. 2009;61(6):745-54. 1120 Journal of Managed Care & Specialty Pharmacy JMCP November 2014 35. Munson J, Morden N, Goodman D, Valle N, Wennberg J. The Dartmouth atlas of medicare prescription drug use. The Dartmouth Institute for Health Policy & Clinical Practice. October 15, 2013. Available at: http://www.dartmouthatlas.org/downloads/reports/Prescription_Drug_Atlas_101513.pdf. Accessed September 22, 2014. 36. Saag KG, Teng GG, Patkar NM, et al. American College of Rheumatology 2008 recommendations for the use of nonbiologic and biologic diseasemodifying antirheumatic drugs in rheumatoid arthritis. Arthritis Rheum. 2008;59(6):762-84. 37. Curtis JR, Chen L, Harrold LR, Narongroeknawin P, Reed G, Solomon DH. Physician preference motivates the use of anti–tumor necrosis factor therapy independent of clinical disease activity. Arthritis Care Res (Hoboken). 2010;62(1):101-07. Vol. 20, No. 11 www.amcp.org Predictors of Treatment Initiation with Tumor Necrosis Factor-α Inhibitors in Patients with Rheumatoid Arthritis Appendix A Multivariate Predictors of Treatment Initiation with TNF-α Inhibitors in RA Patients by Data Source, 2007-2010 Medicare Data, OR (95% CI) Among Monotherapy Nonbiologic Users at Baseline Among Combination Therapy Nonbiologic Users at Baseline Commercial Data, OR (95% CI) Among Monotherapy Nonbiologic Users at Baseline Among Combination Therapy Nonbiologic Users at Baseline Variables Predisposing factors Patient age 0.94(0.92-0.96) 0.97(0.94-1.00) 0.98(0.97-0.99) 0.99(0.98-0.99) Gender Female 1 1 1 1 Male 0.79(0.59-1.04) 0.78(0.51-1.17) 1.03(0.91-1.17) 0.96(0.80-1.17) Metropolitan statistical area (MSA) Non-MSA 1 1 1 1 MSA 0.84(0.61-1.15) 0.63(0.41-0.98) 1.15(1.00-1.32) 1.02(0.83-1.24) Region South 1 1 1 1 0.56(0.41-0.76) 0.70(0.46-1.08) 0.88(0.77-1.01) 0.84(0.70-1.02) North Central 0.70(0.46-1.06) 1.12(0.61-2.06) 0.78(0.63-0.96) 0.76(0.55-1.04) Northeast 0.98(0.69-1.38) 0.95(0.56-1.61) 0.82(0.70-0.96) 0.94(0.76-1.18) West Enabling factors Capitation Noncapitated health plan 1 1 1 1 Capitated plan 0.84(0.56-1.25) 0.77(0.43-1.37) 0.97(0.84-1.13) 1.08(0.88-1.32) Visit to rheumatologist in the pre-index period No visit 1 1 1 1 1.84(1.15-2.93) 1.48(0.67-3.26) 1.13(0.93-1.37) 0.81(0.57-1.14) At least 1 visit More than 1 visit 1.25(0.95-1.66) 1.07(0.72-1.58) 0.91(0.81-1.03) 0.72(0.60-0.86) Calendar year of the index visit 2008 1 1 1 1 0.88(0.59-1.13) 0.87(0.56-1.35) 1.05(0.92-1.29) 1.11(0.88-1.45) 2009 Drug benefit generositya Poor 1 1 1 1 1.07(0.71-1.61) 0.93(0.50-1.73) 1.02(0.88-1.18) 0.98(0.80-1.21) Average Better than average 1.31(0.88-1.94) 1.07(0.59-1.94) 1.11(0.96-1.29) 1.00(0.81-1.23) Most generous 1.37(0.93-2.02) 1.38(0.79-2.42) 1.16(0.99-1.36) 0.78(0.62-0.99) Need factors RA factors 1.17(1.05-1.30) 1.14(0.97-1.33) 1.19(1.15-1.24) 1.31(1.24-1.39) CIRASb No steroid use 1 1 1 1 Steroid use 1.66(1.26-2.19) 2.24(1.44-3.48) 1.82(1.62-2.05) 2.00(1.67-2.40) No coxib use 1 1 1 1 Coxib use 0.98(0.68-1.41) 1.73(1.09-2.77) 1.27(1.08-1.49) 1.12(0.89-1.41) No NSAID use 1 1 1 1 NSAID use 0.95(0.68-1.33) 1.26(0.80-1.99) 1.19(1.06-1.34) 1.37(1.17-1.62) 0.91(0.84-0.99) 0.95(0.85-1.06) 0.96(0.92-1.01) 0.96(0.90-1.03) Combined comorbidity scorec a Drug benefit generosity was classified according to the quartiles of a calculated generosity index. This index was calculated as a continuous variable and defined as the proportion of total drug cost paid by the patient out of pocket. The quartiles were termed as poor drug benefit generosity (highest out-of-pocket costs, fourth quartile), average drug benefit generosity (third quartile), above average drug benefit generosity (second quartile), and most generous drug benefit (lowest out-of-pocket costs, first quartile). b CIRAS: Claims-based index of rheumatoid arthritis severity, which ranged from 0.6 to 10.7 with higher values indicating more severe rheumatoid arthritis. c A composite score indicating a patient’s comorbidity burden after taking into account 20 individual conditions. The score ranged from -2 to 13 in our cohorts, with greater values indicating higher disease burden. CI = confidence interval; NSAID = nonsteroidal anti-inflammatory drug; OR = odds ratio; RA = rheumatoid arthritis; TNF = tumor necrosis factor. 1120a Journal of Managed Care & Specialty Pharmacy JMCP November 2014 Vol. 20, No. 11 www.amcp.org Predictors of Treatment Initiation with Tumor Necrosis Factor-α Inhibitors in Patients with Rheumatoid Arthritis Appendix B Multivariate Predictors of Treatment Initiation with TNF-α Inhibitors in RA Patients by the Source of Treatment Receipt, 2007-2010 Pharmacy-Filled TNF-α Inhibitors, Physician-Administered TNF-α Inhibitors, OR (95% CI) Among Monotherapy Nonbiologic Users at Baseline OR (95% CI) Among Combination Therapy Nonbiologic Users at Baseline Among Monotherapy Nonbiologic Users at Baseline Among Combination Therapy Nonbiologic Users at Baseline Variables Predisposing factors Patient age 0.97(0.97-0.98) 0.98(0.97-0.99) 1.00(0.98-1.01) 0.98(0.97-1.00) Gender Female 1 1 1 1 Male 1.02(0.9-1.16) 0.93(0.77-1.13) 0.91(0.72-1.15) 0.96(0.68-1.37) Metropolitan statistical area (MSA) Non-MSA 1 1 1 1 1.04(0.91-1.2) 0.92(0.75-1.11) 1.28(0.98-1.68) 1.08(0.73-1.60) MSA Region South 1 1 1 1 0.92(0.81-1.06) 0.85(0.71-1.03) 0.58(0.45-0.74) 0.69(0.48-0.98) North Central 0.84(0.68-1.03) 0.84(0.62-1.15) 0.58(0.40-0.84) 0.81(0.46-1.42) Northeast 0.94(0.8-1.11) 1.00(0.80-1.25) 0.62(0.46-0.83) 0.77(0.50-1.18) West Enabling factors Capitation Noncapitated health plan 1 1 1 1 0.92(0.79-1.07) 0.90(0.73-1.11) 1.12(0.86-1.46) 1.55(1.09-2.20) Capitated plan Visit to rheumatologist in the pre-index period No visit 1 1 1 1 1.13(0.92-1.39) 0.86(0.61-1.22) 1.44(1.02-2.02) 0.92(0.46-1.81) At least 1 visit More than 1 visit 0.96(0.85-1.09) 0.72(0.60-0.86) 0.90(0.71-1.13) 0.91(0.65-1.27) Calendar year of the index visit 2008 1 1 1 1 1.13(1.00-1.28) 1.15(0.95-1.39) 0.95(0.75-1.20) 1.24(0.87-1.75) 2009 Data source Commercial 1 1 1 1 0.57(0.45-0.72) 0.75(0.54-1.03) 1.31(0.90-1.89) 1.53(0.89-2.65) Medicare Drug benefit generositya Poor 1 1 1 1 Average 1.04(0.89-1.21) 0.93(0.75-1.15) 1.01(0.77-1.32) 1.20(0.80-1.78) Better than average 1.18(1.01-1.38) 1.02(0.83-1.27) 1.06(0.80-1.40) 0.99(0.65-1.50) Most generous 1.26(1.07-1.47) 0.89(0.71-1.12) 1.04(0.78-1.38) 0.90(0.58-1.41) Need factors RA factors 1.19(1.14-1.24) 1.29(1.21-1.37) 1.19(1.11-1.28) 1.25(1.11-1.39) CIRASb No steroid use 1 1 1 1 Steroid use 1.74(1.54-1.97) 2.05(1.70-2.46) 2.06(1.64-2.59) 2.04(1.43-2.92) No coxib use 1 1 1 1 Coxib use 1.24(1.06-1.47) 1.09(0.86-1.38) 1.17(0.88-1.57) 1.76(1.20-2.58) No NSAID use 1 1 1 1 NSAID use 1.19(1.05-1.34) 1.33(1.12-1.57) 1.12(0.90-1.40) 1.58(1.15-2.16) 0.96(0.91-1.00) 0.96(0.90-1.02) 0.90(0.83-0.98) 0.94(0.83-1.05) Combined comorbidity scorec a Drug benefit generosity was classified according to the quartiles of a calculated generosity index. This index was calculated as a continuous variable and defined as the proportion of total drug cost paid by the patient out of pocket. The quartiles were termed as poor drug benefit generosity (highest out-of-pocket costs, fourth quartile), average drug benefit generosity (third quartile), above average drug benefit generosity (second quartile), and most generous drug benefit (lowest out-of-pocket costs, first quartile). b CIRAS: Claims-based index of rheumatoid arthritis severity, which ranged from 0.6 to 10.7 with higher values indicating more severe rheumatoid arthritis. c A composite score indicating a patient’s comorbidity burden after taking into account 20 individual conditions. The score ranged from -2 to 13 in our cohorts, with greater values indicating higher disease burden. CI = confidence interval; NSAID = nonsteroidal anti-inflammatory drug; OR = odds ratio; RA = rheumatoid arthritis; TNF = tumor necrosis factor. 1120b Journal of Managed Care & Specialty Pharmacy JMCP November 2014 Vol. 20, No. 11 www.amcp.org A AL MEETIN U G NN & EX APRIL 7–10 SAN DIEGO PO AMCP’S 2 7 TH S AV E T H E DAT E ! 2015 Choose from more than 40 educational programs Attendees include representatives from 23 of the top 25 health plans Review and discuss results one-on-one with authors of over 200 research posters 1111 e e e e Register for AMCP’s 27th Annual Meeting & Expo starting January 13, 2015 at www.amcpmeetings.org Academy of Managed Care Pharmacy | 100 N Pitt Street | Suite 400 | Alexandria, VA 22314 | Tel 703/683-8416 | www.amcp.org RESEARCH North Carolina Medicaid Recipient Management Lock-In Program: The Pharmacist’s Perspective S. Rose Werth, BA; Nidhi Sachdeva, MPH, CHES; Andrew W. Roberts, PharmD; Mariana Garrettson, MPH; Chris Ringwalt, PhD; Leslie A. Moss, MHA, CHES; Theodore Pikoulas, PharmD, BCPP; and Asheley Cockrell Skinner, PhD ABSTRACT BACKGROUND: The misuse and abuse of prescription opioids have become an urgent health issue in North Carolina (NC), particularly among Medicaid patients who suffer high rates of morbidity and mortality due to abuse and overdose. The NC Division of Medical Assistance (DMA) implemented a recipient management lock-in program, which limits identified patients for a 12-month period to 1 prescriber and 1 pharmacy for benzodiazepine, opiate, and certain anxiolytic prescriptions in order to prevent misuse and reduce overutilization of Medicaid benefits. OBJECTIVES: To (a) evaluate pharmacists’ perceptions of the implementation of the NC recipient management lock-in program (MLIP) and (b) determine how the beliefs and attitudes of pharmacists could promote or inhibit its success. METHODS: We conducted 12 structured phone interviews with NC pharmacists serving lock-in patients. Interview responses were analyzed through construct analysis, which identified themes organized into 3 domains: organization and implementation, perceived effectiveness, and acceptability. RESULTS: Most respondents reported a positive experience with the program but expressed doubt concerning its impact on prescription drug abuse. The program successfully utilized the pharmacist role as a gatekeeper of controlled substances, and the procedures of the program required no active effort on pharmacists’ part. However, respondents suggested that the DMA improve communication and outreach to address pharmacists’ lack of knowledge about the program’s purpose and confusion over remediating problems that arise with lock-in patients. The DMA should also address the ways in which the program can interfere with access to health care and treatment, allow patients to see multiple physicians within the same clinic, and clarify procedures for patients whose complex health issues require multiple specialists. CONCLUSIONS: Although possible improvements were identified, the NC MLIP has strong potential for success as it utilizes pharmacists’ medication gate-keeping role, while minimizing the effort required for successful implementation. J Manag Care Pharm. 2014;20(11):1122-28 Copyright © 2014, Academy of Managed Care Pharmacy. All rights reserved. What is already known about this subject •Mortality among Medicaid recipients due to prescription drug abuse is 5 times that of the general population. •North Carolina Medicaid beneficiaries and providers had one of the highest rates of potentially fraudulent purchases of controlled substances in the the nation. 1122 Journal of Managed Care & Specialty Pharmacy JMCP November 2014 •North Carolina implemented a narcotic lock-in program to address controlled substance misuse and reduce cost and utilization among Medicaid beneficiaries. What this study adds •While many respondents reported an overall positive experience with the program, they doubted its impact on prescription drug misuse and abuse. •Educating pharmacists about the program’s purpose and policies as well as addressing potential barriers to patient care may improve the program. O verdose and abuse of prescription drugs has become an increasingly urgent health issue in the United States. Most overdoses from prescribed medications result from nonmedical uses, which refers to both the misuse and abuse of prescription drugs.1 The national death rate due to unintentional poisoning increased by 91% between 1999 and 2009 and by 213% in North Carolina (NC) during the same period.2,3 Opioid analgesics, a class of controlled substances (CS) that includes methadone, codeine, hydrocodone, oxycodone, and morphine, among others, are the driving factor behind this increase.4 In the last decade, prescription painkillers have caused more deaths than either cocaine or heroin.5 The economic burden of prescription drug misuse and abuse is estimated to contribute $25 billion to health care costs, $5 billion to the justice system, and $25.6 billion in workplace losses annually.6 Mortality among Medicaid recipients due to prescription drug abuse is 5 times that of the general population, and opioid nonmedical users are more likely to be covered by Medicaid than by any other insurance program.7,8 In 2009, the U.S. Government Accountability Office reported that NC Medicaid beneficiaries and providers had one of the highest rates of potentially fraudulent and abusive purchases of CS of any state in the nation.9 Efforts to prevent unintentional poisoning are reliant on the effective monitoring of prescription CS. The NC Division of Medical Assistance (DMA) implemented the NC Medicaid Recipient Management Lock-In Program (NC MLIP) to address issues of CS misuse and abuse in the NC Medicaid population. Vol. 20, No. 11 www.amcp.org North Carolina Medicaid Recipient Management Lock-In Program: The Pharmacist’s Perspective The goal of the program is to decrease the nonmedical use of opiates, benzodiazepines, and some anxiolytics by Medicaid recipients and to prevent recipient overutilization of Medicaid benefits.10 The program operates by identifying NC Medicaid recipients with 1 of the following criteria: (a) > 6 prescription claims for an opioid medication in 2 consecutive months; (b) > 6 prescription claims for a benzodiazepine or controlled anxiolytic medications in 2 consecutive months; or (c) > 3 prescribers of these same CS in 2 consecutive months. Providers, NC Medicaid administrators, and administrators at Community Care of North Carolina (CCNC), the state’s managed care organization for Medicaid enrollees, can also nominate patients whom they think should be included in the program based on a subjective assessment of their risk behavior, although in practice, this option is rarely used. Once identified, Medicaid administrators lock patients into 1 prescriber and 1 pharmacy so that the recipient can only receive Medicaid-covered CS prescriptions from the providers to whom they are restricted.10 The patients will be locked in for a 1-year time period after which they will be removed from the program if they no longer meet criteria.10 There are 2,000 to 3,000 patients enrolled at any given time. The MLIP has already demonstrated effectiveness in an analysis conducted by the DMA in 2012, which showed that the program saved $5.2 million over the first year and reduced the number of pain pills and anti-anxiety medications prescribed to MLIP patients by 2.3 million in only 3 months.11 The beliefs and attitudes of providers carrying out health interventions can either hinder or improve program effectiveness.12-16 Pharmacists, as the final intermediary between patient and prescription, are in a position to promote the successful operation of the program and identify opportunities for its improvement, provided they understand and support the process and implementation. The purpose of this study was to assess pharmacists’ attitudes towards the MLIP and their experiences implementing the program by conducting in-depth structured interviews with NC pharmacists serving locked-in Medicaid patients. ■■ Methods Respondents and Setting The sampling frame included all licensed pharmacists in NC. The NC Board of Pharmacy sent a solicitation e-mail to all licensed pharmacists, which included basic study information and a link to an initial screening survey. This survey was used to determine if respondents were practicing pharmacists with locked-in patients and to gauge their willingness to participate in phone interviews. From this screening, a list of 97 willing respondents was generated and prioritized according to the highest number of MLIP patients served. Two project staff members contacted and interviewed pharmacists from the prioritized list until the data were saturated to the point where no www.amcp.org Vol. 20, No. 11 new information or themes emerged. In-depth phone interviews were conducted until thematic saturation was reached at a total of 12 pharmacists. Respondents practiced in a range of settings, including urban and rural pharmacies, independent and chain pharmacies, and clinic and hospital outpatient pharmacies. Structured Interviews Structured phone interviews were conducted using an Institutional Review Board (IRB)-approved topic guide (see Appendix, available in online article). Interview questions were intended to prompt pharmacists to report their experiences as the sole dispenser to a locked-in patient, as well as their experiences with the program, with particular regard to its impact on patient outcomes and access to health care. The interviews lasted 15 to 45 minutes, depending on the length of the respondent’s answers and subsequent discussion. The interviews were audio-recorded, with each respondent’s permission, to aid in subsequent data analysis. Data Management and Analysis Data were drawn from detailed notes that were taken by the interviewers; accuracy was confirmed using the audio recordings. Following thematic content analysis,17 each line or group of lines was coded with an identifier. Two members of the research team coded the first interview separately and created identifiers as they appeared within the data. The identifiers and codes were discussed and combined into 1 uniform codebook with the assistance of a third senior researcher. The remaining interviews were analyzed using the uniform codebook. If new codes arose from subsequent interviews, they were discussed and added to the codebook. The coded text was divided into themes as they emerged within the analysis, with quotes and sections from interviews selected to describe and exemplify each theme. The themes were used to examine pharmacists’ perceptions of 3 domains: (1) program organization and implementation, (2) perceived effectiveness of the program, and (3) the program’s acceptability to pharmacists and their patients. This study was reviewed and approved by the University of North Carolina at Chapel Hill IRB. ■■ Results The screening survey yielded 97 respondents who met all selection criteria and were willing to participate in an interview. Interviews were arranged sequentially, prioritizing respondents who served the highest number of lock-in patients. Thematic saturation was reached with 12 in-depth structured interviews. References to CS indicate only those targeted by the NC MLIP: opiates, benzodiazepines, and certain anxiolytics. Table 1 lists the 12 themes identified within the data, organizing each theme into 1 of the 3 domains. November 2014 JMCP Journal of Managed Care & Specialty Pharmacy 1123 North Carolina Medicaid Recipient Management Lock-In Program: The Pharmacist’s Perspective TABLE 1 Content Analysis, Domains, and Themes Themes Domains Organization/ Implementation Notification Perceived Effectiveness Overall opinion Selection process Communication Program impact Impact on dispensing patterns Proposed changes Understanding of program intent Problem remediation Acceptability Difficulty of implementation Impact on job roles Unintended effects on patients Domain 1: Organization and Implementation Notification. With the exception of 1 respondent, pharmacists received or were aware of letters sent from Medicaid informing them that their pharmacies had become the sole dispenser for a patient enrolled in the MLIP. Pharmacists can refuse a patient who has been locked-in, but none of the respondents had used this option. However, many respondents also depended on the claims process to alert them to locked-in patients. One pharmacist commented: “I have seen letters that come in the mail sometimes informing us of a patient that may or may not be locked in, but we generally rely on the [claims] transmission process of the Medicaid system.” Respondents reported very few problems with the notification process and transition into the program. A few pharmacists reported patients claiming to be unaware of their lock-in status, but the pharmacists were skeptical about these patients’ reports that they had not been formally notified. Selection Process. The majority of respondents stated that they did not believe any of their patients were mistakenly enrolled into the program. On the contrary, many believed that too few of their patients were locked in. One respondent stated: “I’ve seen a lot that have been mistakenly not enrolled in the program. . . . There’s a lot of people falling through the cracks.” Only 3 other respondents described patients as unfit for the program. One pharmacist stated that up to 75% of his lockedin patients did not belong in the MLIP and went further, saying: “I don’t know that we’ve ever had a scenario that I felt like the patient had been appropriately locked in with a prescriber.” Communication with Administrators and Managers. The main form of communication between pharmacists and program officials was Medicaid phone channels. If pharmacists needed to contact Medicaid administrators with questions about the program or problems with patients or claims, they had to call the Medicaid phone line and wait on hold to be redirected to an administrator in charge of the program or leave a message. Respondents expressed frustration about Medicaid’s limited “bankers’ hours.” Many pharmacists reported waiting on hold for long periods of time or waiting for hours for a return phone call. 1124 Journal of Managed Care & Specialty Pharmacy JMCP November 2014 Communications between the DMA and each of the pharmacies did not flow through corresponding channels. Respondents described different protocols within pharmacies for handling messages and information from the state Medicaid program. One pharmacist reported that only 1 person within the pharmacy handled communications from Medicaid. A respondent who had previously served as a pharmacy manager mentioned that managers may not relay information about the program to their technicians. The level of support for the program from pharmacy management varied considerably. Understanding of Program Intent. Respondents received mixed or incomplete messages about the goals and intentions of the MLIP. Ten of 12 respondents stated the explicit desired outcome that the program would ensure patients received their CS prescriptions from 1 prescriber and 1 pharmacist. Other perceived goals of the program included reduction of doctor and pharmacy shopping, minimization of Medicaid costs, prevention of prescription drug misuse and abuse, and prevention of adverse patient outcomes. However, 1 pharmacist expressed frustration that “They don’t really tell us what the goal is. I don’t know. I would guess cost, but it could be anything.” Half of respondents were confused about the boundaries between the MLIP and the NC Medicaid prescription capitation program. The capitation program required patients who filled 11 prescriptions of any drug category within a certain time period to opt-in to a locked-in single pharmacy if they needed another prescription that same month. The capitation program ended several months prior to our interviews. Respondents called it the “over 11” or “opt-in” program, reporting that many pharmacists use the terminology of the 2 programs interchangeably. Some respondents explained that their pharmacy computers used the same indicators for MLIP patients as were used for the previous program. A couple of respondents were unaware that the MLIP was an entirely separate program. Problem Remediation. Respondents experienced difficulty when resolving problems they encountered with the program. A few pharmacists mentioned that drug shortages in the past year required patients to switch to a nonlock-in pharmacy that stocked their prescriptions. Respondents labeled this a “temporary unlock” or “secondary lock.” Two pharmacists discussed their patients’ need for exceptions in the case of holidays, closings, or travel and their wish that they could provide these patients with 3-day supplies in such situations. When encountering issues with prescribers, a number of pharmacists did not know of any readily available solutions, especially for incidents occurring on weekends. In 1 case, a respondent’s patient was discharged from an emergency department (ED) and presented a CS prescription written by the ED doctor: “We can’t [fill it]. There’s no work around. What the patient is required to do is take the provider’s prescription and contact that provider that’s been approved to rewrite the order. To me, if it’s a surgical patient, that could be something that causes them to come back to the hospital. Especially if it’s on the weekend and the provider is not open ‘til Monday. Really a lot of these hiccups come on the weekend.” Vol. 20, No. 11 www.amcp.org North Carolina Medicaid Recipient Management Lock-In Program: The Pharmacist’s Perspective Domain 2: Perceived Effectiveness Overall Opinion. Ten of 12 respondents reported positive experiences with the program but also volunteered flaws or areas for improvement. One pharmacist viewed the program as a temporary fix and said that many patients would “go back to their old lifestyle” after only a year in the program. Those who expressed the most positive opinions recognized the extent of the prescription drug epidemic and appreciated the program’s efforts to mitigate it. Of the pharmacists interviewed, only 2 had explicitly negative evaluations of the program, with 1 calling it “one of the worst programs I’ve ever dealt with.” These pharmacists believed that the MLIP was either ineffective or was actively harming their patients by restricting access to care. The discussion under the next section, “Domain 3: Acceptability” further addresses this effect. come back out on a recurring basis at least a couple times a year just to remind pharmacy providers that there are options for patients.” One pharmacist proposed changes to address cash payments: “[I would like it] if they had a line that would allow health care practitioners out there to submit things for potential abuse situations or diversion situations. When a man comes in locked into 1 store and his prescription is early and he offers to pay several hundred dollars in cash to get the prescription early—that should be reported somewhere.” Other recommended changes included expanding the program to enroll more patients; managing patients using chronic pain specialists; increasing pharmacist awareness and education about the program; locking patients into 1 clinic instead of 1 unique provider; and improving patients’ transition out of the program. Program Impact. With regard to specific program outcomes, 4 pharmacists believed that it reduced doctor shopping; 4 believed that it reduced misuse, abuse, and diversion of drugs; and 3 believed that it reduced Medicaid costs. Two other pharmacists stated that the program somewhat affected these outcomes but was not effective enough. Four respondents explicitly stated that it did not affect any abuse or misuse outcomes. A major issue brought up by 8 of the 12 respondents was that patients may circumvent the program by paying cash for their prescriptions. In a few instances, cash payments were used for prescriptions that respondents viewed as medically necessary even though they had been rejected by Medicaid. However, respondents primarily saw cash payments as a way for misusers and abusers to continue receiving excess prescriptions. One pharmacist stated: “I have a feeling if people are doctor hopping and shopping, they’ll just learn to go to a store they’ve never been to and pay cash. I don’t know how effective [the program] is at keeping controlled substances off the street that shouldn’t be there.” Domain 3: Acceptability Difficulty in Implemention. The MLIP program was not difficult for respondents to implement. Five of the 12 respondents specifically mentioned that their roles were passive or did not require effort outside of their normal duties. Impact on Dispensing Patterns. Seven of the 12 pharmacists stated that the program had not altered how they monitored enrolled patients. The other 5 respondents treated lock-in patients differently, refusing to fill certain prescriptions, checking the state’s Controlled Substance Reporting System before all fills, and verifying prescriptions with physicians. Four respondents believed that the program had increased their monitoring of all patients. One respondent particularly stressed this effect: “I think the idea that knowing that somebody else is watching and that there’s a certain level of accountability changes it for everyone. If it’s successful it should change your thought process across the board.” Proposed Changes. Respondents proposed a wide array of changes or improvements to the program. They most frequently requested improved communication with Medicaid. One respondent proposed changes to the Medicaid newsletter: “I would like to see the [lock-in policies] in the Medicaid newsletter www.amcp.org Vol. 20, No. 11 Impact on Job Roles. Respondents honed in on 3 different roles they take on as pharmacists: as a caregiver, a gatekeeper, and as a business person. Pharmacists must maintain a balance between providing care to their patients and controlling the flow of prescription drugs within a community. With regard to controlling drug flow, 1 pharmacist described this tradeoff: “There’s a wide spectrum. There’s some who feel like they need to be the police. I don’t think that’s our role. I think we need to make the prescribers aware and be aware ourselves. You have to have some dialogue with the patients. If you just turn everyone away they’ll go to multiple prescribers and pharmacies. They will learn how to work the system.” Several pharmacists reported that the program had made their roles easier. One pharmacist expressed gratitude: “Really it makes my life a lot easier being the sole provider. I mean really it does because that takes the question out. So if they’re only seeing 1 doctor and 1 drug store then there’s that much more control.” Aiding their role as caregivers, the program would alert respondents to when a patient needed counseling: “It really affirms in a busy pharmacy world, where you might be filling hundreds of prescriptions, that you need to stop, take a pregnant pause for a minute, and have a conversation.” Other pharmacists mentioned that acting as the sole dispenser allows them to improve adherence and track disease states. Many respondents saw the program as supportive of their business, insofar as the patients locked into their pharmacies increase their patient volume and sales. When asked about rejecting lock-in patients, 1 pharmacist replied: “I’ve never tried not to be locked in [i.e., involved in the lock-in program] because of course we’re trying to increase prescription sales.” For a few November 2014 JMCP Journal of Managed Care & Specialty Pharmacy 1125 North Carolina Medicaid Recipient Management Lock-In Program: The Pharmacist’s Perspective respondents, focusing on the program’s potential to increase volume was an “economic reality” faced by independent pharmacies, which often struggle to make a profit against large chains. Unintended Effects on Patients. Pharmacists were asked how the program affected their patients’ access to care. Six respondents reported that the program could be a barrier to receiving pain treatment, mental health care, or general health care. While 6 respondents stated that the program had no effect on access, 3 of those 6 also described instances where patients had difficulty receiving medications or care at other points in the interview. Some of these problems arose from the type of physician a patient was locked in with: “I think what happened was a lot of patients were identified for the lock-in program because they had received short supplies of opiate pain relievers—either they were postsurgical, or they had an injury, an accident, or something. And, then they were locked in with this short-term—sometimes it was even an ED physician—and then they were locked out of their psychiatrists. And, then it would go the opposite way, too . . . people who had gotten locked in with their psychiatrists, but then their pain doctor was locked out and couldn’t write their medicines.” Other access issues stemmed from the availability of physicians: “They make an appointment for their next visit and the prescriber they’re locked in individually is not here. They see whoever is available to be seen next. They’re not doctor shopping. They’re just seeing whoever is there.” Availability of Medicaid also hindered access: “Any of those types of programs where stuff needs approval or assistance at the Medicaid level, that’s a potential barrier in all of those cases. Not to say those programs aren’t all good, but when you don’t have 24-hour, 7-days-a-week coverage for those types of things, that is a barrier for some patients.” One particularly frustrated respondent summed up this issue as an inherent design flaw: “The program’s not designed to keep the bad people away and the keep the good people getting what they need. It’s just designed to keep the bad people away. It’s hard to do both, but it can be done. . . . The way it was done you ended up with a lot of people who probably would have benefited from being locked into 1 pain provider but were adversely affected on other parts of their treatment.” ■■ Discussion The qualitative data ascertained through in-depth interviews with NC MLIP pharmacists revealed a wide range of experiences with, and opinions about, the lock-in program. While results may not be generalizable to all program pharmacists, its findings can provide insight into the program’s strengths and weaknesses. 1126 Journal of Managed Care & Specialty Pharmacy JMCP November 2014 Organization and Implementation While respondents valued the ease of enrollment, many believed that communication within the program was problematic. One pharmacist stated, “I think it would be a better program if all pharmacists knew exactly what we were trying to do.” The program was initially announced to pharmacies through the DMA bulletin, while several subsequent bulletins issued reminders.10,18,19 CCNC clinical pharmacists have also done education and training with regards to the MLIP, especially with community pharmacies throughout the state. The work culture within some respondents’ pharmacies may have interfered with their ability to receive these messages. If individual staff members handle Medicaid correspondence without communicating with their coworkers, or if pharmacy managers do not discuss the program’s goals and features with their staff, widespread understanding of the program cannot be achieved. The issues that respondents faced with problem remediation may be a result of their confusion about the program’s goals and policies. Only 1 DMA bulletin mentioned how to handle lock-in exceptions. In an emergency situation, a 4-day supply of a CS may be provided to patients by a pharmacy or prescriber to which they are not locked in.10 The process for switching lock-in providers or handling other complications is not clearly available online nor was it described in any DMA bulletins. Only 1 pharmacist mentioned this policy; others merely expressed confusion. Respondents’ dissatisfaction with Medicaid phone channels underscores program shortfalls in communications. Perceived Effectiveness The interviewed pharmacists gave mixed responses when asked whether the program had affected their CS dispensing practices. Some pharmacists reported increased monitoring of locked-in patients, but the majority stated that the program had not affected how they dispensed CS. While interviews revealed both positive and negative aspects of the program, the recurring mention of cash payments for CS prescriptions outside of the program indicates a substantial weakness. While the scale of this problem is unknown, cash payments for CS prescriptions would undoubtedly counteract the programs’ impact, outside of reducing Medicaid expenditures. Also, there has been some question as to whether all pharmacists are upholding the integrity of the program, with some possibly allowing patients to fill a CS with cash after receiving a rejected prescription claim from Medicaid. Fortunately, NC’s prescription drug monitoring program, the Controlled Substances Reporting System, will soon be upgraded to include the method of payment used for all CS prescriptions in the system records. Awareness by pharmacists that a patient is enrolled in Medicaid should make it Vol. 20, No. 11 www.amcp.org North Carolina Medicaid Recipient Management Lock-In Program: The Pharmacist’s Perspective increasingly difficult for Medicaid patients to circumvent the MLIP by paying cash. Several respondents stated that a 12-month lock-in does not allow sufficient time to establish long-term positive effects on recipient behavior. This time period does allow for CCNC clinical pharmacists and care managers to offer education to the patient, connect the patient with any needed substance abuse services, and provide more intensive care management services.20,21 The MLIP also lacks a transition period at the end of recipients’ lock-in. Patients are re-enrolled if they still meet lock-in criteria. However, there is no transitory process or provision of resources for those who no longer qualify, even if their pharmacists believe they are still misusing or abusing CS. Program Acceptability Overall, program acceptability was high among study respondents, in large part because the automated Medicaid claims system minimized pharmacists’ efforts in actively engaging the MLIP. The program also allowed pharmacists to track patient utilization behavior, prompted them to provide targeted patient counseling, and ensured continued patronage. However, the biggest concern to arise from these interviews was that the program may impede patients’ access to care. For example, patients who are locked in with a single provider, as opposed to any provider within a given clinic, may be denied appropriate treatment if their prescribers are unavailable. The program may be particularly harmful to patients with multiple diseases, such as chronic pain and psychiatric comorbidities, since these patients are more likely to see multiple specialists. Particular problems may arise from the program’s definition of CS. The definition applies to opioid analgesics, which are used to treat acute and chronic pain, and to benzodiazepines and anxiolytics, which are most commonly used to treat anxiety disorders. While there is significant correlation between chronic pain and mental health issues, opioid analgesics are usually prescribed by different specialists than benzodiazepines or anxiolytics.22 While allowing patients to receive CS prescriptions from multiple providers may present a health risk,23 locking them into 1 provider may not be the most advantageous method to address complex health needs. For this reason, the DMA may be requested to allow for up to 2 providers for a single patient in situations in which 2 prescribers are being utilized (i.e., benzodiazepine prescribed by a psychiatrist and opioid medication prescribed by a pain specialist). It appears however, that pharmacists were not widely aware of this policy. The MLIP has only 1 set of enrollment criteria for all opiate, benzodiazepines, and controlled anxiolytic users, treating them as a single, homogenous group. This strategy could lead to inappropriate patient enrollment, targeting patients with www.amcp.org Vol. 20, No. 11 broadly specified criteria. Multiple studies have illustrated that prescription drug users comprise diverse subpopulations21-23 that include high school students; the elderly; street drug users; chronic pain, cancer, and acute pain patients; and patients receiving end-of-life care.24-26 The reason patients misuse CS also vary, ranging from poor patient education and undertreatment of pain to dependence and addiction, diversion, and desire for euphoria.24,26 The MLIP’s enrollment criteria may not be effectively capturing these diverse subpopulations, although the ultimate goal for each group of patients would be the same: to reduce use and misuse of prescription medications, reduce unintentional overdoses, and promote the safe and effective use of these dangerous medications that can be addicting to all patients. Limitations The foundational strength of this study was the use of structured interviews, allowing the pharmacists surveyed to express a wide range of thoughts and opinions. The interview guide ensured that researchers covered key topics with pharmacists, while giving them the opportunity to freely express their opinions and concerns and raise new and unforeseen topics for discussion. The primary study limitation was the small sample size, leading to a potential limited generalizability of study results. Additionally, although we sent our request to all members of the NC Board of Pharmacy, the sample consisted primarily of outpatient community pharmacists. ■■ Conclusions This study revealed strengths and shortcomings of the NC MLIP. The program holds promise for success as it utilizes pharmacists’ medication gate-keeping role, while minimizing the effort required from pharmacists for successful implementation. The program can also improve pharmacists’ awareness of prescription drug misuse and abuse among their patients and improve their ability to prevent and combat it. However, more outreach is needed by the DMA in order to educate pharmacists about the MLIP’s policies and underlying purpose. Improved communication between pharmacies and the DMA is also necessary for program success. This is particularly true when issues with locked-in patients arise (e.g., on weekends). Resolving these issues may require new or improved methods of communication. Examples of methods that may be beneficial include extended access to Medicaid phone channels; a direct phone line for problems pharmacies face with locked-in patients; or set protocols for them to follow when common issues arise. The flexibility of the program should also be enhanced, allowing patients to see multiple prescribers within the same clinic. While the NC MLIP addresses an urgent health issue within the NC Medicaid population, further refinements have the potential to substantially enhance its impact. November 2014 JMCP Journal of Managed Care & Specialty Pharmacy 1127 North Carolina Medicaid Recipient Management Lock-In Program: The Pharmacist’s Perspective Authors S. ROSE WERTH, BA, is Program Specialist, Partnership for a DrugFree NC, Durham, North Carolina. NIDHI SACHDEVA, MPH, CHES, is Project Manager; MARIANA GARRETTSON, MPH, is Research Scientist; and CHRIS RINGWALT, PhD, is Senior Scientist, Injury Prevention Research Center, University of North Carolina at Chapel Hill. ANDREW W. ROBERTS, PharmD, is PhD Candidate, Division of Pharmaceutical Outcomes and Policy, UNC Eshelman School of Pharmacy, and Postdoctoral Fellow, Cecil G. Sheps Center for Health Services Research, University of North Carolina at Chapel Hill; LESLIE A. MOSS, MHA, CHES, is Data Coordinator, Division of Medical Assistance, North Carolina Department of Health and Human Services, Raleigh; THEODORE PIKOULAS, PharmD, BCPP, is Associate Director of Behavioral Health Pharmacy Programs, Community Care of North Carolina, Raleigh; and ASHELEY COCKRELL SKINNER, PhD, is Associate Professor of Pediatrics, University of North Carolina at Chapel Hill. AUTHOR CORRESPONDENCE: Asheley Cockrell Skinner, PhD, University of North Carolina at Chapel Hill School of Medicine, 231 MacNider, 229B, CB 7225, Chapel Hill, NC 27599. Tel.: 919.843.9941; E-mail: [email protected]. DISCLOSURES This research was supported by Cooperative Agreement CDC U01 CE00216001, Building Interdisciplinary Research Careers in Women’s Health (BIRWCH) Training Grant NIH K12 HD001441 (Skinner), CTSA Grant UL1TR000083, the UNC Injury Prevention Research Center, and a National Research Service Award Post-Doctoral Traineeship from the Agency for Health Care Research and Quality sponsored by the Cecil G. Sheps Center for Health Services Research, 5 T32 HS000032 (Roberts). The authors have no conflicts of interest to disclose. Study concept and design was primarily contributed by Garrettson, Ringwalt, and Skinner, with assistance from Sachdeva and Werth. Data collection was primarily the responsibility of Werth and Sachdeva, with assistance from Garrettson and Ringwalt, and Pikoulas and Werth interpreted the data, with assistance from the rest of the authors. The manuscript was written by Werth, Roberts, and Ringwalt, assisted by the rest of the authors, and revision was carried out by Moss and Skinner, assisted by the rest of the authors. ACKNOWLEDGMENTS The authors would like to thank the North Carolina Board of Pharmacy for its assistance in recruitment. REFERENCES 1. Substance Abuse and Mental Health Services Administration, Center for Behavioral Health Statistics and Quality. Highlights of the 2010 Drug Abuse Warning Network (DAWN) findings on drug-related emergency department visits. The DAWN Report. July 2, 2012. Rockville, MD. Available at: http://www.samhsa.gov/data/2k12/DAWN096/SR096EDHighlights2010.pdf. Accessed August 5, 2014. 2. Centers for Disease Control and Prevention. Prevent unintentional poisoning. 2012. Available at: http://www.cdc.gov/Features/PoisonPrevention/. Accessed August 5, 2014. 3. Ford MD. Unintentional poisoning in North Carolina: an emerging public health problem. N C Med J. 2010;71(6):542-46. 4. Warner M, Chen LH, Makuc DM. Increase in fatal poisonings involving opioid analgesics in the United States, 1999-2006. NCHS data brief, no 22. Hyattsville, MD: National Center for Health Statistics; 2009. Available at: http://www.stoprxdrugabuse.org/2009_9_30_CDC-_Opioid_ Analgesics_1999-2006_US.pdf. Accessed August 5, 2014. 5. Paulozzi LJ, Budnitz DS, Xi Y. Increasing deaths from opioid analgesics in the United States. Pharmacoepidemiol Drug Saf. 2006;15(9):618-27. 1128 Journal of Managed Care & Specialty Pharmacy JMCP November 2014 6. Birnbaum HG, White AG, Schiller M, Waldman T, Cleveland JM, Roland CL. Societal costs of prescription opioid abuse, dependence, and misuse in the United States. Pain Med. 2011;12(4):657-67. 7. Centers for Disease Control and Prevention. Overdose deaths involving prescription opioids among Medicaid enrollees—Washington, 2004-2007. MMWR Morb Mortal Wkly Rep. 2009;58(42):1171-75. Available at: http://www.cdc.gov/ mmwr/preview/mmwrhtml/mm5842a1.htm. Accessed August 5, 2014. 8. Centers for Disease Control and Prevention. Issue brief: unintentional drug poisoning in the United States. MMWR Morb Mortal Wkly Rep. 2010;59(10):300. Available at: http://www.cdc.gov/homeandrecreationalsafety/pdf/poison-issue-brief.pdf. Accessed August 5, 2014. 9. U.S. Government Accountability Office. Medicaid: fraud and abuse related to controlled substances identified in selected states. GAO report number GAO-09-957. 2009. Available at: http://www.gao.gov/assets/300/294715. html. Accessed August 5, 2014. 10. North Carolina Division of Medical Assistance. Implementation of a recipient management lock-in program. August 2010 Medicaid Bulletin. Available at: http://www.ncdhhs.gov/dma/bulletin/0810bulletin.htm#lock. Accessed August 5, 2014. 11. North Carolina Department of Health and Human Services. 2.3 Million pills off the streets, $5.2 million saved by narcotics lock-in. May 14, 2012. Available at: http://www.ncdhhs.gov/pressrel/2012/2012-05-14_pills_off_ streets.htm. Accessed August 5, 2014. 12. Aarons GA. Measuring provider attitudes toward evidence-based practice: consideration of organizational context and individual differences. Child Adolesc Psychiatr Clin N Am. 2005;14(2):255-71, viii. 13. Aarons GA, Palinkas LA. Implementation of evidence-based practice in child welfare: service provider perspectives. Adm Policy Ment Health. 2007;34(4):411-19. 14. Cornuz J, Ghali WA, Carlantonio D, Pecoud A, Paccaud F. Physicians’ attitudes towards prevention: importance of intervention-specific barriers and physicians’ health habits. Fam Pract. 2000;17(6):535-40. 15. Glisson C, Landsverk J, Schoenwald S, et al. Assessing the organizational social context (OSC) of mental health services: implications for research and practice. Adm Policy Ment Health. 2008;35(1-2):98-113. 16. Kushner RF. Barriers to providing nutrition counseling by physicians: a survey of primary care practitioners. Prev Med. 1995;24(6):546-52. 17. Attride-Stirling J. Thematic networks: an analytic tool for qualitative research. Qual Res. 2001;1(3):385-405. 18. North Carolina Division of Medical Assistance. Recipient management lock-in program emergency fill. October 2010 Medicaid Bulletin. Available at: http://www. ncdhhs.gov/dma/bulletin/1010bulletin.htm#lock. Accessed August 5, 2014. 19. North Carolina Division of Medical Assistance. Recipient opt-in program and monthly prescription limits. February 2013 Medicaid Bulletin. Available at: http:// www.ncdhhs.gov/dma/bulletin/0213bulletin.htm#optin. Accessed August 5, 2014. 20. Jackson CT, Trygstad TK, DeWalt DA, DuBard CA. Transitional care cut hospital readmissions for North Carolina Medicaid patients with complex chronic conditions. Health Aff (Millwood). 2013;32(8):1407-15. 21. Chisholm-Burns MA, Kim Lee J, Spivey CA, et al. US pharmacists’ effect as team members on patient care: systematic review and meta-analyses. Med Care. 2010;48(10):923-33. 22. Gureje O. Psychiatric aspects of pain. Curr Opin Psychiatry. 2007;20(1):42-46. 23. Centers for Disease Control and Prevention. CDC grand rounds: prescription drug overdoses—a U.S. epidemic. MMWR Morb Mortal Wkly Rep. 2012;61(1):10-13 Available at: http://www.cdc.gov/mmwr/preview/mmwrhtml/mm6101a3.htm. Accessed August 5, 2014. 24. Green TC, Black R, Grimes Serrano JM, Budman SH, Butler SF. Typologies of prescription opioid use in a large sample of adults assessed for substance abuse treatment. PLoS One. 2011;6(11):e27244. 25. Ghandour LA, Martins SS, Chilcoat H. Understanding the patterns and distribution of opioid analgesic dependence symptoms using a latent empirical approach. Int J Methods Psychiatr Res. 2008;17(2):89-103. 26. Davis WR, Johnson BD. Prescription opioid use, misuse, and diversion among street drug users in New York City. Drug Alcohol Depend. 2008;92(1-3):267-76. Vol. 20, No. 11 www.amcp.org North Carolina Medicaid Recipient Management Lock-In Program: The Pharmacist’s Perspective Appendix Interview Guide Provider Interview Guide Hello, my name is _____________. I am with the Injury Prevention Research Center at UNC Chapel Hill. I am calling to follow up with the interview about the Medicaid Lock-In Program that we scheduled for today. Is now still a good time for a 20- to 30-minute interview? (If not, reschedule.) Great. Did you receive the fact sheet in the confirmation e-mail? I would like to review that briefly with you before we start. (Go over fact sheet.) What questions, if any, do you have about the study or the fact sheet? Please keep in mind that this interview is completely voluntary; there are no right or wrong answers—so please answer as honestly and best as you can; your name will not be linked to your responses; and you have the option to stop the interview at any time. CONFIRM CONSENT: Are you still interested in participating? Would it be all right to begin the interview now? QUESTIONS My first questions are about the processes involved in becoming a NC MLIP patient’s sole provider: 1.How do you get notified that you have been selected as a sole provider for an enrolled patient? 2.Can you decide whether to accept or not? If so, a. How do you decide whether to accept a patient or not? b.How do you communicate with Medicaid about your decision? c. If you have declined to accept a patient, how did the Medicaid administrator respond? 3.Do you believe that any of the patients you serve have been mistakenly, or inappropriately, enrolled? a. If so, what was the nature of the mistake? 4.What challenges, if any, have you had with any aspect of this process? Now, I have several questions about your experience of treating patients who are enrolled in the NC MLIP: 5.We understand from the online survey that you have been the provider for XX patients in the MLIP, and you identified your overall experience with the program as XX on a scale of 1 to 6. Is this still accurate? a.(For people who have more than 1 patient locked in) What were the most common patient characteristics or circumstances that led to their initial enrollment in the program? 6.Please describe your characterization of your experience as positive and/or negative. Could you tell me more about what makes you characterize your experience that way? 7.If you were going to give advice to another physician/pharmacist who is accepting the role of sole provider, what would you say? 8.If you could make 1 change to the program that would make it easier for providers like you, what would that be? My last questions are about your perceptions of the impact of this program on your patients and on you: 9.How does this program affect a patient’s access to pain care? (Probe: Does the program improve or impair patients’ effective pain management?) 10.How does this program affect a patient’s access to mental health care? (Probe: Does the program improve or impair patients’ mental illness management?) 11.How does this program affect a patient’s access to all other medical care, if at all? 12.How does this program affect a patient’s health? 13.How has this program changed the way that you prescribe/dispense controlled substances to patients who are enrolled in the program? 14.How has this program changed the way that you prescribe/dispense controlled substances to other patients? 15.What else about the implementation or impact of this program would you like to share? 1128a Journal of Managed Care & Specialty Pharmacy JMCP November 2014 Vol. 20, No. 11 www.amcp.org RESEARCH Is There an Association Between the High-Risk Medication Star Ratings and Member Experience CMS Star Ratings Measures? Sara C. Erickson, PharmD; R. Scott Leslie, MPH; and Bimal V. Patel, PharmD, MS ABSTRACT BACKGROUND: Methods to achieve high star ratings for the High-Risk Medication (HRM) measure are thought to result in unintended consequences and to compromise several member experience measures that ultimately put at risk the plan sponsor’s Medicare Part D Centers for Medicare & Medicaid (CMS) star rating. •Achieving high star ratings is associated with increased likelihood of new beneficiary enrollment and enrollment from existing Medicare beneficiaries switching from lower-rated plans. CMS quality bonus payments also provide increased revenue opportunities for plans with high star ratings. OBJECTIVE: To determine if HRM scores are associated with relevant member experience measure scores. What this study adds METHODS: This is a cross-sectional analysis utilizing CMS 2013 and 2014 plan star ratings reports (2011 and 2012 benefit year data) for Medicare Advantage prescription drug (MA-PD) plans and prescription drug plans (PDPs). Medicare contracts with complete data for all measures of interest in 2013 and 2014 star ratings reports were included (N = 443). Bivariate linear regressions were performed for each of 2 independent variables: (1) 2014 HRM score and (2) 2013 to 2014 change in HRM score. Dependent variables were the 2014 scores for “Getting Needed Prescription Drugs,” “Complaints about Drug Plan,” “Rating of Drug Plan,” and “Members Choosing to Leave the Plan.” RESULTS: The bivariate linear regressions demonstrated weak positive associations between the 2014 HRM score and each of the 2014 member experience measures that explained 0.5% to 4% (R 2 ) of variance of these measures. The bivariate regressions for the 2013 to 2014 change in the HRM score and 2014 member experience measures of interest demonstrated associations accounting for 1% to 8% of variance (R 2 ). The greatest associations were observed between each independent variable and the 2014 “Getting Needed Prescription Drugs” score with correlation coefficients of 0.21 and 0.29. CONCLUSIONS: HRM star ratings and change in HRM star ratings are weakly correlated with member experience measures in concurrent measurement periods. Plan sponsors may be more aggressive in HRM utilization management, since it is unlikely to negatively impact CMS summary star ratings. J Manag Care Pharm. 2014;20(11):1129-36 Copyright © 2014, Academy of Managed Care Pharmacy. All rights reserved. What is already known about this subject •The Beers Criteria for potentially inappropriate medications to be used in the elderly and, more recently, the updated 2012 American Geriatric Society (AGS) Beers Criteria, have been adapted into the quality performance measure High-Risk Medications (HRM) used in Centers for Medicare & Medicaid (CMS) star ratings. •CMS has placed an emphasis on the member experience with several star rating measures related to member satisfaction that have a greater total weight contribution to the overall summary star rating than the individual HRM star rating measure. www.amcp.org Vol. 20, No. 11 •The results of this study indicate that the HRM star rating measure score and its improvement, prior to the implementation of the updated 2012 AGS Beers Criteria, have minimal influence on the member experience star ratings measure scores. •This information may be used by plan sponsors in making informed decisions regarding HRM utilization management. T he late Mark H. Beers, MD, a geriatrician, is widely known for his contribution to geriatric care through the development of a list of potentially inappropriate medications (PIMs) used by the elderly. Published in 1991, the “Beers Criteria” of PIMs was developed to reduce inappropriate medication use in nursing home patients.1 The 1997 update expanded the criteria for application to all ambulatory older persons.2 The last update under Dr. Beer’s auspices was published in 2003 and stood as the definitive criteria for nearly a decade.3 The American Geriatric Society (AGS) sponsored the 2012 Beers Criteria update, which included many improvements, such as following the Institute of Medicine standard of evidence and transparency for creating guidelines.4,5 The use of the Beers Criteria expands beyond individual practitioners or institutions of care. The National Committee for Quality Assurance first included a performance measure based on the Beers Criteria known as “Use of HighRisk Medications in the Elderly,” in the 2006 Healthcare Effectiveness Data and Information Set (HEDIS), a tool for health plans to measure care and service performance.6 The High-Risk Medication (HRM) measure was subsequently adapted and endorsed by the Pharmacy Quality Alliance (PQA), endorsed by the National Quality Forum, and adopted by the Centers for Medicare & Medicaid Services (CMS). Consistent with HEDIS and PQA, CMS included the revised 2012 AGS Beers Criteria into the HRM measure specifications for the 2015 star ratings for measuring HRM use in the 2013 benefit year. CMS includes the HRM measure in the Part D star November 2014 JMCP Journal of Managed Care & Specialty Pharmacy 1129 Is There an Association Between the High-Risk Medication Star Ratings and Member Experience CMS Star Ratings Measures? TABLE 1 Measure Label High-risk medication Getting needed prescription drugs Members’ rating of drug plan CMS Star Ratings Measures Included in Analysis Measure Description Percentage of plan members aged 65 years or older who received 2 or more prescription fills for the same drug with a high risk of serious side effects in the elderly Case-mix adjusted percentage of the best possible score the plan earned on how easy it is for members to get the prescription drugs they need using the plan Case-mix adjusted percentage of the best possible score the plan earned from members who rated the prescription drug plan Complaints about the drug plan Data Source Prescription drug event data files submitted by drug plans CAHPS survey questions: 1130 Journal of Managed Care & Specialty Pharmacy JMCP November 2014 1.5 In the last 6 months, how often was it easy to use your health plan to get the medicines your doctor prescribed? In the last 6 months, how often was it easy to use your health plan to fill a prescription at a local pharmacy? In the last 6 months, how often was it easy to use your health plan to fill prescriptions by mail? CAHPS survey question: 1.5 Using any number from 0 to 10, where 0 is the worst prescription drug plan possible and 10 is the best prescription drug plan possible, what number would you use to rate your prescription drug plan? Complaints logged into Complaint Tracking Module 1.5 Total complaints received about the drug plan per 1,000 members adjusted by average contract enrollment Members choosing to The percentage of plan members who chose to leave Disenrollment reason codes in Medicare’s enrollment leave the plan the plan during the benefit year (does not include system members who moved out of the service area) CAHPS = Consumer Assessment of Healthcare Providers and Systems; CMS = Centers for Medicare & Medicaid Services. ratings for Medicare Advantage prescription drug (MA-PD) plans and stand-alone prescription drug plans (PDPs). Star ratings summarize the performance of contracted health and drug plans on various indicators of clinical quality, care access, customer service, and member satisfaction. Contracts receive 1 to 5 stars, with 5 stars representing the highest quality for each individual performance measure, as well as for aggregate groupings of like measures referred to as domains and an overall summary rating. Star ratings are important to contracting plans for several reasons. The Affordable Care Act of 2010 requires quality bonus payments and rebates to be tied to star ratings.7 Also, low-performing contracts risk exclusion from Medicare, since past performance is a consideration in the approval of applications.8 Lastly, CMS star ratings are posted publicly to help beneficiaries choose plans. An association has been recently established between Medicare Advantage plan star ratings and enrollment numbers.9 Reid et al. (2013) found that new enrollees were more likely to enroll in plans with higher star ratings, and continuing enrollees were more likely to switch to plans with higher star ratings.9 The HRM star ratings measure is 1 of 5 triple-weighted medication-related patient safety measures that comprise 19% of the Part C rating or 54% of the Part D rating. CMS has also placed an emphasis on the member experience with several star rating measures related to member satisfaction. The management of HRMs within the prescription benefit may therefore be problematic for Medicare contracts. Low-touch Weighting Value 3 1.5 interventions, such as educational mailings intended to influence prescribing behavior, are not likely to have an immediate or dramatic effect. Requiring prior authorization approval for HRMs or formulary exclusion of HRMs are likely to cause significant member disruption, negatively impacting the member experience. HRMs are often older, generic medications that members have been able to obtain easily prior to turning 65 and enrolling in a Medicare plan. Safe, low-cost alternatives are not always available. Plans with poor HRM star ratings may be hesitant to implement management strategies with the assumption that this may have an unintended negative impact on member satisfaction scores, which have a greater total weight contribution to the overall summary star rating. The purpose of our analysis was to determine the relationship of the HRM score and the change in HRM score to relevant member experience measures. ■■ Methods Study Design This is a cross-sectional analysis utilizing the 2 most recently released CMS star ratings reports (i.e., the 2013 and 2014 star ratings reports that use 2011 and 2012 benefit years, respectively). These reports can be found on the CMS website.10 Data Each October, prior to open enrollment for the upcoming benefit year, the most recent star ratings results are available on the CMS website. The report is named for the coming benefit year. Vol. 20, No. 11 www.amcp.org Is There an Association Between the High-Risk Medication Star Ratings and Member Experience CMS Star Ratings Measures? FIGURE 1 Medicare Contracts Included in 2013-2014 Star Ratings Reports and Included in Analysis Medicare contracts included in 2013 star ratings report (n = 650) Medicare contracts included in both 2013 and 2014 star ratings reports (n = 603) Medicare contracts included in 2014 star ratings report (n = 749) Medicare contracts given overall Part D rating in 2013 report (n = 557) Medicare contracts given overall Part D rating in 2013 and 2014 reports (n = 549) Medicare contracts given overall Part D rating in 2014 report (n = 550) Medicare contracts with complete information for all 5 studied measuresa in 2013 reports (n = 490) Medicare contracts with complete information for all 5 studied measuresa in 2013 and 2014 reports (n = 443) This is the sample for the primary analysis Medicare contracts with complete information for all 5 studied measuresa in 2014 reports (n = 450) Medicare contracts with a 2-star or greater increase in HRM measure from 2013 to 2014 reports (n = 81) This is sample for the subanalysis aContracts were excluded if “No data available,” “Not enough data available,” “Plan too new to be measured,” or “CMS identified issues with this plan’s data” was indicated for any of the following measures: High-Risk Medications, Complaints about Plan, Getting Needed Prescriptions, Rating of Drug Plan, and Members Leaving Plan. HRM = high-risk medication. For example, the star ratings report released in October 2013 is titled “2014 Medicare Part D Star Ratings Data,” since the information is intended to help beneficiaries select plans for the 2014 benefit year. The 2014 star ratings report released in 2013 reflects the performance of each Medicare contract during the 2012 benefit year. The majority of the data was collected throughout the 2012 benefit year, although some measures, such as those obtained from the Consumer Assessment of Healthcare Providers and Systems (CAHPS) survey data, were obtained in early 2013. Each star rating measure has the contract’s calculated score presented, known as the base level. Thresholds are established to assign base-level measure scores a star rating based on a 5-star scale. These scores are rolled up to a domain star rating, and the domain star ratings are rolled up to a Part D summary rating and Part C summary rating (for MA-PD plans). In the 2014 star ratings for the 2012 benefit year, there were 15 Part D star ratings measures and 36 Part C measures that roll up to a total of 9 domains. The Part D score is the highest level for a PDP. MA-PD plans receive an overall rating that summarizes the 51 measures included in Part C and Part D measures. The measures of interest for this analysis included the Patient Safety measure HRM and all measures that were deter- www.amcp.org Vol. 20, No. 11 mined to be potentially negatively impacted by formulary restrictions. Four measures within the domains of “Member Experience with the Drug Plan” and “Member Complaints, Problems Getting Services, and Improvement in the Drug Plan’s Performance” were included as follows: Getting Needed Prescription Drugs; Members’ Rating of Drug Plan; Complaints About the Drug Plan; and Members Choosing to Leave the Plan (Table 1). Inclusion Criteria Medicare contracts with base-level detail and star ratings for all measures of interest in both the 2013 and the 2014 star ratings reports were included. Contracts with “No data available,” “Not enough data available,” “Plan too new to be measured,” or “CMS identified issues with this plan’s data” in place of base data or star ratings for the 5 measures of interest were excluded (Figure 1). High-Risk Medication Measure (Explanatory or Independent Variable) The HRM performance measure for 2013 and 2014 star ratings (2011 and 2012 benefit year, respectively) is reported as the percentage of members aged 65 years and older enrolled during November 2014 JMCP Journal of Managed Care & Specialty Pharmacy 1131 Is There an Association Between the High-Risk Medication Star Ratings and Member Experience CMS Star Ratings Measures? TABLE 2 2013 and 2014 CMS Star Ratings Base Measure Results for Measures of Interest Measure Label High-risk medication Getting needed prescription drugs Members’ rating of drug plan Complaints about the drug plan a Members choosing to leave the plan Measure Detail Data time frame Base data range Base mean (SD) Data time frame Base data range Base mean (SD) Data time frame Base data range Base mean (SD) Data time frame Base data range Base mean (SD) Data time frame Base data range Base mean (SD) 2013 CMS Star Ratings Report January 1, 2011-December 31, 2011 2%-21% 8% (3%) February 15, 2012-May 31, 2012 84%-97% 91% (2%) February 15, 2012-May 31, 2012 77%-95% 85% (3%) January 1, 2012-June 30, 2012 0-2.68 0.33 (0.28) January 1, 2011-December 31, 2011 0%-48% 10% (7%) 2014 CMS Star Ratings Report January 1, 2012-December 31, 2012 1%-18% 6% (3%) February 15, 2013-May 31, 2013 85%-97% 91% (2%) February 15, 2013-May 31, 2013 76%-96% 85% (3%) January 1, 2013-June 30, 2013 0-3.44 0.28 (0.28) January 1, 2012-December 31, 2012 0%-64% 11% (8%) a Reported as complaints per 1,000 members. CMS = Centers for Medicare & Medicaid Services; SD = standard deviation. the benefit year with at least 2 fills for a prescription medication that has a high risk for adverse events in the elderly. The base score is calculated for members aged 65 years and older from prescription drug event claims data files provided by drug plans to CMS throughout the benefit year. Patient Satisfaction Measures (Dependent Variables) Getting Needed Prescription Drugs. Data for the Getting Needed Prescription Drugs performance measure are acquired through the CAHPS survey. Three questions from the CAHPS survey are used to calculate the Getting Needed Prescription Drugs base score: • “In the last 6 months, how often was it easy to use your health plan to get the medicines your doctor prescribed?” • “In the last 6 months, how often was it easy to use your health plan to fill a prescription at a local pharmacy?” • “In the last 6 months, how often was it easy to use your health plan to fill prescriptions by mail?” The CAHPS survey is administered between February 15 through May 31 after the benefit year by CMS-approved vendors by mail and telephone to a random sampling of members. The measure is reported as the percentage of the best possible CAHPS survey score. It is a case-mix adjusted measure that takes into account socioeconomic differences (such as age, education level, and dual eligibility) of enrollees across contracts. Members’ Rating of Drug Plan. The Members’ Rating of Drug Plan performance measure is also a case-mix adjusted percentage of the best possible CAHPS survey score. The measure is based on the score of 1 question within the CAHPS survey: “Using any number from 0 to 10, where 0 is the worst prescription drug plan possible and 10 is the best prescription drug 1132 Journal of Managed Care & Specialty Pharmacy JMCP November 2014 plan possible, what number would you use to rate your prescription drug plan?” The data are collected by CMS-appointed vendors February 15 through May 31 after the benefit year. Complaints about the Drug Plan. The Complaints about the Drug Plan performance measure is reported as the number of complaints received about the drug plan logged into the CMS Complaint Tracking Module (CTM) per 1,000 members. The CTM is a central repository for complaints received in either central or regional offices or 1-800-MEDICARE. Historically, CTM data from the first 6 months following the benefit year is used for the performance measure. Members Choosing to Leave the Plan. The Members Choosing to Leave the Plan performance measure is reported as the percentage of members who chose to disenroll from the plan during the benefit year. This is calculated by taking the number of members who chose to disenroll using disenrollment codes in Medicare’s enrollment system, excluding members who disenrolled due to reasons beyond their control (e.g., relocation out of the service area, disenrollment due to low-income subsidy reassignments, or not meeting requirements for special needs plans), and dividing the number by the number of members enrolled at any time during the benefit year. Statistical Analysis Bivariate regressions were performed for 2 independent variables: (1) 2014 star ratings HRM base score (2012 benefit year) and (2) the change in HRM base score from the 2013 to 2014 star ratings reports (2011 and 2012 benefit years; Table 2). The dependent variables were the member experience base scores in the 2013 and 2014 star ratings reports. Three of the 4 Vol. 20, No. 11 www.amcp.org Is There an Association Between the High-Risk Medication Star Ratings and Member Experience CMS Star Ratings Measures? TABLE 3 Bivariate Linear Regression for All Plans with Complete Information for 5 Measures in 2013 and 2014 (n = 443) High-Risk Medication Measure Member Experience Measure (Independent Variable) (Dependent Variable) Beta Coefficient 0.139 2014 high-risk 2014 getting needed prescription drugs basea medication base 0.066 2014 rating of drug plan basea -1.067 2014 complaints about drug plan basea -0.260 2014 members choosing to leave the plan basea 0.213 2013-2014 change in 2014 getting needed prescription drugs basea high-risk medication -0.113 2014 rating of drug plan basea base -1.372 2014 complaints about drug plan basea -0.574 2014 members choosing to leave the plan basea 2014 high-risk 2013 getting needed prescription drugs baseb 0.551 medication base 0.051 2013 rating of drug plan baseb -1.447 2013 complaints about drug plan baseb 0.155 2013-2014 change in 2013 getting needed prescription drugs baseb high-risk medication -0.146 2013 rating of drug plan baseb base -1.569 2013 complaints about drug plan baseb a Dependent variable has the same benefit year (2012) as the 2014 HRM star rating. bDependent variable has the same measurement year (2012) as the 2014 HRM star rating. HRM = high-risk medication. member experience measures of interest are collected in a time period following the benefit year. It is therefore possible that changes in the formulary management of HRMs could affect member experience measures of the previous benefit year’s star ratings report for members enrolled in the same contract in both years. Regressions were performed for each independent variable and dependent variable pair within the same report year and within the same year of data collection. Regression analyses were also performed on a subgroup of Medicare contracts that achieved a 2-star or greater increase in the HRM measure from 2013 star ratings (2011 benefit year) to 2014 star ratings (2012 benefit year). These contracts may represent those that made significant changes in the management of HRMs to achieve this improvement. If achieving an excellent HRM base score (lower is better) was associated with a negative impact on member experience measures, positive associations would be observed for the 2 measures Getting Needed Prescription Drugs and Rating of Drug Plan (higher is better) and negative associations would be observed for the 2 measures Complaints About the Drug Plan and Members Choosing to Leave the Plan (lower is better). Bivariate linear regression models were performed in Microsoft Excel 2010. A two-tailed P value < 0.05 was considered statistically significant. ■■ Results Medicare contracts included in the 2013 and 2014 star ratings reports (2011 and 2012 benefit years, respectively) with complete information for all 5 measures of interest were included www.amcp.org Vol. 20, No. 11 Standard Error 0.031 0.045 0.427 0.126 0.034 0.050 0.471 0.137 0.035 0.045 0.439 0.038 0.050 0.485 R2 0.04 0.005 0.01 0.01 0.08 0.01 0.02 0.04 0.02 0.003 0.02 0.04 0.02 0.02 P Value < 0.001 0.14 0.013 0.04 < 0.001 0.02 0.004 < 0.001 0.002 0.27 0.001 < 0.001 0.004 0.001 in regression models. A total of 443 contracts were included, which represents 73.5% of the total Medicare contracts present in both years and 80.7% of the total Medicare contracts with summary star ratings for both years (Figure 1). The distribution of organization type was similar for Medicare contracts with complete information for all 5 measures of interest in the 2013 star ratings, Medicare contracts with complete information for all 5 measures of interest in 2014 star ratings, and Medicare contracts meeting both criteria. The most prevalent organization type (83.1%-83.3%) was local coordinated care (CCP) MA-PD plans. Medicare contracts achieving a 2-star or greater improvement in the HRM star ratings between the 2013 and the 2014 star ratings reports numbered 101. Of the 101 contracts, 81 (80.2%) had complete data for all measures of interest. The contracts in this subgroup had a greater proportion of local CCP MA-PD plans (87.7% vs. 83.1%-83.3%) and private feefor-service plans (4.9% vs. 2.0%-2.4%) compared with all plans with complete information for the measures of interest. The bivariate linear regressions demonstrated associations that accounted for 1% to 4% (R 2) of variance between the explanatory and dependent variables between the 2014 star ratings HRM base score and 3 of the 4 2014 star ratings member experience measures: Getting Needed Prescription Drugs (R 2 = 0.04; P < 0.001); Complaints about Drug Plan (R 2 = 0.01; P = 0.013); and Members Choosing to Leave the Plan (R 2 = 0.01; P = 0.04; Table 3). As the data collection is measured after the benefit year for 3 of the 4 member experience measures, bivariate linear regressions were performed for the 2014 star ratings November 2014 JMCP Journal of Managed Care & Specialty Pharmacy 1133 Is There an Association Between the High-Risk Medication Star Ratings and Member Experience CMS Star Ratings Measures? TABLE 4 Bivariate Linear Regression for All Plans with Complete Information for 5 Measures in 2013 and 2014 and a ≥ 2-Star Increase in HRM Measure Stars (n = 81) High-Risk Medication Measure Member Satisfaction Measure (Independent Variable) (Dependent Variable) Beta Coefficient 0.267 2013-2014 change in 2014 getting needed prescription drugs basea high-risk medication 0.205 2014 rating of drug plan basea base 0.356 2014 complaints about drug plan basea -0.078 2014 members choosing to leave the plan basea 0.177 2013-2014 change in 2013 getting needed prescription drugs baseb high-risk medication 0.076 2013 rating of drug plan baseb base -1.829 2013 complaints about drug plan baseb a Dependent variable has the same benefit year (2012) as the 2014 HRM star rating. bDependent variable has the same measurement year (2012) as the 2014 HRM star rating. HRM = high-risk medication. HRM base scores and the 2013 star ratings member experience measures with data collection in the same calendar year. The 2014 HRM measure had associations with the 2013 star ratings Getting Needed Prescription Drugs (R 2 = 0.02; P = 0.002) and Complaints about Drug Plan (R 2 = 0.02; P = 0.001), explaining 2% (R 2) of variance. The 2014 HRM base score was not significantly associated with the 2013 or the 2014 Rating of Drug Plan measure scores. The bivariate regressions using the change in HRM base score from the 2013 to the 2014 star ratings reports as the explanatory variable demonstrated associations accounting for 1% to 8% of variance with 2014 member experience measures of interest: Getting Needed Prescription Drugs (R 2 = 0.08; P < 0.001); Rating of Drug Plan (R 2 = 0.01; P = 0.02); Complaints about Drug Plan (R 2 = 0.02; P = 0.004); and Members Choosing to Leave the Plan (R 2 = 0.01; P < 0.001). Associations between the change in the HRM base score and the 2013 member experience measures of interest with overlapping data collection time frames explain 2% to 4% (R 2) of variance: Getting Needed Prescription Drugs (R 2 = 0.04; P < 0.001); Rating of Drug Plan (R 2 = 0.02; P = 0.004); and Complaints about Drug Plan (R 2 = 0.02; P = 0.001). As with the 2014 HRM base score explanatory variable regressions, the associations between the 2013-2014 change in HRM base score and the Complaints about Drug Plan base score dependent variables (2013 and 2014) and 2014 Members Choosing to Leave the Plan were consistently negative. Also consistent with the 2014 HRM base score explanatory variable regressions, the 2013-2014 change in the HRM base score explanatory variable and Getting Needed Prescriptions Drug measures (2013 and 2014) were consistently positively associated. In contrast to the 2014 HRM base score regressions, which did not demonstrate an association with Rating of Drug Plan scores, the change in HRM measure base scores from 2013 to 2014 had negative associations with both the 2013 and 2014 1134 Journal of Managed Care & Specialty Pharmacy JMCP November 2014 Standard Error 0.112 0.155 2.340 0.500 0.124 0.157 1.397 R2 0.07 0.02 0.0003 0.0003 0.03 0.003 0.02 P Value 0.02 0.19 0.88 0.88 0.16 0.63 0.19 Rating of Drug Plan measure base scores, which explain 1% to 2% of the variance (R 2). Bivariate regressions were performed using the change in HRM base scores for a subgroup of Medicare contracts that achieved a 2-star or greater improvement in the HRM measure (Table 4). The only significant association between the explanatory variable and member experience measures was for the 2014 Getting Needed Prescription Drugs measure (P = 0.02), which explains 7% (R 2) of the variance in the model. ■■ Discussion The results of this analysis indicate that the HRM score and the HRM score change are marginally able to predict member experience star ratings measure scores. The HRM score and the HRM score change had the greatest influence on the Getting Needed Prescription Drugs measure (R 2 = 0.04 and R 2 = 0.08, respectively). However, the magnitude of this association is small. The corresponding correlation coefficients (R) of 0.21 and 0.29 suggest that there are likely limited linear associations between the variables.11 Scatter plots of the data points (plots not shown) for each dependent and independent variable pair show single clusters with 1 to 4 outliers. Bivariate regressions were repeated after excluding outliers. The size of the variance and association did not change appreciably after removing the outliers (data not shown). The only change in the significance testing occurred for the result of the 2014 HRM and the 2014 Complaints about Drug Plan regression, which became nonsignificant after outlier removal. The HRM score was generally more predictive of the concurrent member experience star ratings measures than previous year’s member experience star ratings measures. Since the member experience data are often captured after the benefit year, the effect of HRM formulary restrictions may be diluted over multiple years. The 2013 to 2014 star ratings report change in the HRM score explained more variance in the Vol. 20, No. 11 www.amcp.org Is There an Association Between the High-Risk Medication Star Ratings and Member Experience CMS Star Ratings Measures? models than the 2014 the HRM score. It is intuitive that an increase in HRM score, the more significant of which is likely the result of increased formulary restrictiveness, would be more likely to be associated with a marked impact to member experience measures. The models’ beta-coefficients suggest that HRM score improvement negatively impacts all member satisfaction measures. However, the results indicate an overall minimal contribution to the member experience measures. This is likely due to many factors influencing the overall member experience and the intentionally broad nature of the CAHPS survey questions. That is, there may be several reasons influencing member response to the CAHPS survey questions that may or may not be related to HRM usage. Members may give greater consideration to perceived cost burdens, consisting of pharmacy benefit premiums and prescription copayments, rather than to prescription medication access when rating their prescription drug plans or choosing to leave their plans. It is also possible that the timing of survey administration chosen by CMS results in significant recall bias. Additionally, only a small proportion of members surveyed are likely to have utilized or tried to fill an HRM. The mean prevalence of members meeting HRM criteria for the 2014 star ratings report across all contracts is 6%.10 The degree of restrictiveness and the extent to which drug utilization is managed through prior authorization across the entire formulary may have considerable negative repercussions on the Getting Needed Prescription Drug star rating, compared with strict management of the HRM category alone. Plans with greater HRM dispensing rates may observe a more significant impact on patient satisfaction measures by restricting access to HRMs than plans with smaller HRM prevalence. The subanalysis of Medicare contracts achieving a 2-star or greater increase in the HRM measure yielded 1 significant regression between the change in the HRM score and the 2014 Getting Needed Prescription Drugs that explained 7% of variance. Contracts with a 2-star or greater increase represent those who likely implemented formulary changes between the 2011 and 2012 benefit years to restrict the use of HRMs. It is likely that these improvements were accomplished through formulary restrictions, since educational outreach is generally associated with more gradual improvements. However, the specific utilization management techniques employed by these plans are unknown, and some interventions may have greater negative impact on the Getting Needed Prescription Drugs score than others. The subanalysis may also be limited by the small sample size (n = 81) and unmeasured differences from the entire cohort of contracts. Limitations The applicability of these results may be limited by the changes in the HRM list. Beginning with the 2015 star ratings (2013 www.amcp.org Vol. 20, No. 11 benefit year), the HRM star ratings measure has been updated based on the 2012 AGS Beers Criteria. There are many significant differences between the HRMs listed prior to the 2013 benefit compared with the present version. Many medications were deleted (e.g., cimetidine, diazepam, and phentermine), and several new medications have been added (e.g., glyburide, indomethacin, and guanfacine). Another significant change is the addition of medications with daily dosage or cumulative days’ supply limits in addition to or in place of the 2 fills criterion. Nonbenzodiazepine hypnotics (i.e., esczopiclone, zolpidem, and zaleplon) are included as HRMs when the cumulative days’ supply is greater than 90 days. Future analysis should be conducted to determine the relationship between HRM management strategies, HRM score, and member experience scores in the 2015 star ratings (2013 benefit year) and beyond. Other limitations of this analysis include not being able to control for socioeconomic differences in patient populations across Medicare contracts. This study used publicly available Medicare contract performance results, which do not include demographic data. Also, we did not have access to reasons for disenrollment or to complaint type logged in the CTM, which may explicate the impact of formulary restrictions on these measures. ■■ Conclusions The results of regression analyses suggest that the HRM star rating measure score and its improvement have minimal influence on the member experience star ratings measure scores. Plan sponsors may be more aggressive in the management of HRM utilization to increase patient safety and clinical quality without compromising member experience performance measures. Safety-related formulary restriction implementations intended to improve the HRM star rating score are unlikely to cause decreases in member satisfaction that would negatively impact summary star ratings. Furthermore, increasing the HRM star rating score is likely to improve summary star ratings. Further research is needed to determine the impact of different methods of HRM star rating improvement on member satisfaction using the current AGS Beers Criteria. Authors SARA C. ERICKSON, PharmD, is Health Outcomes Researcher; R. SCOTT LESLIE, MPH, is Health Outcomes Researcher; and BIMAL V. PATEL, PharmD, MS, is Director, Health Outcomes Research, MedImpact Healthcare Systems, Inc., San Diego, California. AUTHOR CORRESPONDENCE: Sara C. Erickson, PharmD, MedImpact Healthcare Systems, Inc., 10181 Scripps Gateway Ct., San Diego, CA 92131. Tel.: 858.790.7494; E-mail: [email protected]. November 2014 JMCP Journal of Managed Care & Specialty Pharmacy 1135 Is There an Association Between the High-Risk Medication Star Ratings and Member Experience CMS Star Ratings Measures? DISCLOSURES This research was conducted by MedImpact Healthcare Systems, Inc., San Diego, California, without external funding. All authors are employed by MedImpact Healthcare Systems, Inc. Study concept and design were contributed by Erickson, Patel, and Leslie. Erickson collected the data, which were interpreted by Erickson, Leslie, and Patel. Erickson wrote the manuscript, which was revised by Leslie and Patel. REFERENCES 1. Beers MH, Ouslander JG, Rollingher I, Reuben DB, Brooks J, Beck JC. Explicit criteria for determining inappropriate medication use in nursing home residents. Arch Intern Med. 1991;151(9):1825-32. 2. Beers MH. Explicit criteria for determining potentially inappropriate medication use by the elderly: an update. Arch Intern Med. 1997;157(14):1531-36. 3. Fick DM, Cooper JW, Wade WE, Waller JL, Maclean JR, Beers MH. Updating the Beers criteria for potentially inappropriate medication use in older adults: results of a US consensus panel of experts. Arch Intern Med. 2003;163:271(22):2716-24. 4. The American Geriatrics Society 2012 Beers Criteria Update Expert Panel. American Geriatrics Society Updated Beers Criteria for Potentially Inappropriate Medication Use in Older Adults. J Am Geriatr Soc. 2012;60(4):616-631. Available at: http://www.ncbi.nlm.nih.gov/pmc/articles/ PMC3571677/. Accessed September 25, 2014. 1136 Journal of Managed Care & Specialty Pharmacy JMCP November 2014 5. Graham R, Mancer M, Wolman DM, Greenfield S, Steinberg E, eds. Clinical Practice Guidelines We Can Trust. Institute of Medicine. Washington, DC: National Academies Press; 2011. Available at: http://www.iom.edu/ Reports/2011/Clinical-Practice-Guidelines-We-Can-Trust.aspx. Accessed September 25, 2014. 6. National Committee for Quality Assurance. HEDIS 2006 Technical Specifications. Item 10284-100-06. Washington, DC: National Committee for Quality Assurance; 2005. 7. Kaiser Family Foundation. Quality ratings of Medicare Advantage Plans: key changes in the health reform law and 2010 enrollment data. Issue Brief. September 2010. Publication #8097. Available at: http://kaiserfamilyfoundation.files.wordpress.com/2013/01/8097.pdf. Accessed September 25, 2014. 8. Tudor, CG. 2012 application cycles past performance review methodology. Memo dated December 12, 2010. Center for Medicare. Available at: http://www.cms.gov/Medicare/Prescription-Drug-Coverage/ PrescriptionDrugCovContra/Downloads/PastPerformanceMethodology_121 010Final.pdf. Accessed September 25, 2014. 9. Reid RO, Deb P, Howell BL, Shrank WH. Association between Medicare Advantage plan star ratings and enrollment. JAMA. 2013;309(3):267-74. 10. Centers for Medicare & Medicaid Services. Parts C & D performance data. 2014 Part C & D Medicare star ratings data. Available at: http://www. cms.gov/Medicare/Prescription-Drug-Coverage/PrescriptionDrugCovGenIn/ PerformanceData.html. Accessed September 25, 2014. 11. Elston RD, Johnson WD. Essentials of Biostatistics. Philadelphia, PA: FA Davis, Co.; 1987. Vol. 20, No. 11 www.amcp.org Make a Difference — Volunteer for AMCP Leadership Today! Involvement will give you numerous opportunities to strengthen your professional network, develop your leadership skills and contribute to AMCP and the future of managed care pharmacy. Call for Candidates: Call for Volunteers: Serve on the AMCP Board of Directors AMCP 2015–2016 Committee Service Help direct the future of AMCP by serving on AMCP’s Board of Directors. The AMCP Committee on Nominations is seeking individuals for the offices of President-Elect, Treasurer and Director (two positions). Candidates must be Pharmacist Members in good standing. Contribute your talents and expertise to the future of managed care pharmacy by volunteering for committee service. Review the descriptions for each of AMCP’s volunteer opportunities and decide what is right for you. Find detailed information about qualifications and requirements for each Committee at www.amcp.org. Deadline Information Completed Candidate Application packets are due November 21, 2014. For the full Candidate packet, go to www.amcp.org/board. ! W O N A C T ine: Deadl 21 November Deadline Information Apply online at www.amcp.org. All applications for consideration are due November 21, 2014. For more information — www.amcp.org RESEARCH Perceptions and Attitudes of Community Pharmacists Towards Generic Medicines Suzanne S. Dunne, BSc (Hons), MSc; Bill Shannon, MD, FRCGP, MICGP; Walter Cullen, MD, MICGP, MRCGP; and Colum P. Dunne, BSc (Hons), MBA, PhD ABSTRACT What is already known about this subject BACKGROUND: Following the enactment of legislation in June 2013, generic substitution and reference pricing of medicines has been introduced, for the first time, in Ireland. This novel study is the first assessment of the perceptions of community pharmacists in Ireland towards generic medicines completed in the period immediately prior to the introduction of generic substitution and reference pricing. •Pharmacist perceptions of, and attitudes towards, generic medicines are a relatively unexplored area internationally, with only 10 publications found in PubMed on this topic since 2003. •No studies of the views of community pharmacists in Ireland towards generic medicines have ever been published. OBJECTIVE: To determine the perceptions towards generic medicines among community pharmacists. What this study adds METHODS: One-to-one semistructured interviews were performed with a convenience sample of 44 community pharmacists (from approximately 4,500 pharmacists in Ireland) recruited from Ireland’s Midwest, South, and Southwest regions. Interviews were transcribed and analysed using NVivo (version 9). RESULTS: 98% of pharmacists believed that generics were of a similar quality to the originator, and 96% stated that they were as effective as the originator. However, a small number demonstrated some reticence regarding generics: 9% believed that generics were not manufactured to the same quality as the originator; 7% stated they would prefer to take an originator medicine themselves; and 7% reported having experienced quality issues with generic medicines. 89% of pharmacists reported receiving patient complaints regarding use of generic medicine, although 64% suggested that this was due to a nocebo effect (i.e., a result of patients’ preconceived notions that generics were inferior). Only a minority (21%) reported that they had attempted to educate patients as to the equivalency of generics. Although 80% were in favor of Ireland’s new legislation promoting the use generic medicines, 46% expressed concerns regarding its practical implementation. CONCLUSIONS: This key stakeholder group had positive attitudes towards generics and the legislation that promotes their use. Concerns regarding patient perception and experience, clinical effectiveness, and manufacturing quality were identified. We propose that interventions supporting implementation of the new legislation should address these concerns. J Manag Care Pharm. 2014;20(11):1138-46 Copyright © 2014, Academy of Managed Care Pharmacy. All rights reserved. 1138 Journal of Managed Care & Specialty Pharmacy JMCP November 2014 •This is the first study of pharmacist perceptions of generic medicines in Ireland and 1 of only 6 other studies on pharmacist perceptions of generics in Europe. •This is only the second qualitative investigation of pharmacist views in Europe—the other being from Sweden and published in 2012. •This study adds to the body of knowledge on pharmacist attitudes towards generics, providing in-depth, qualitative data that can be used as a basis for policy implementation and decision making. I n June 2013, new legislation came into effect in Ireland1— the Health (Pricing and Supply of Medical Goods) Act 2013)2—that introduced generic substitution and reference pricing for the first time in this country. As a result of this new legislation, Irish patients will now have a greater opportunity to receive a generic medicine instead of a brand-name prescription medication. In an effort to ensure that this legislation is successful, pharmacists’ opinions of, and attitudes towards, generic medicines are critical to the changes being implemented—that is, to increase the use of generic medicines in Ireland. Attitudes of Irish pharmacists towards generics have not been published in the past. While assessments of pharmacist perceptions of generic medicines have been carried out in a limited number of other countries, a PubMed search covering the period from January 2003 to January 2014 did not return any peer-reviewed publications on the topic of pharmacist perceptions of generic medicines in Ireland. In fact, only 10 publications since 2003 were found in PubMed on the topic of pharmacist perceptions of, and attitudes towards, generic Vol. 20, No. 11 www.amcp.org Perceptions and Attitudes of Community Pharmacists Towards Generic Medicines medicines, indicating that this is a relatively underexplored area internationally.3-12 With Ireland on the cusp of a major modification in health care practices, there are many potential hurdles to overcome during the introduction of such changes.13 The attitudes and behaviors of health care professionals towards generic medicines are pivotal to the successful implementation of the new legislation. The objective of this novel study was to assess these perceptions among community pharmacists in Ireland in the time leading up to the enactment of the new legislation and to determine what challenges might arise as a result of these stakeholder opinions. ■■ Methods Preparation of Study Instrument The study instrument was developed based on a recently published review of the usage of generic medicines and how policy changes to promote the use of generic medicines may affect health care provision14 and the personal experience of the primary author and study designer (who has over 15 years of quality management and regulatory affairs within the pharmaceutical and biopharmaceutical industry). Questions for the semistructured interview were prepared and validated by cognitive testing, the purpose of which was to ensure that the test questions were understood as intended. The purpose of the interviews was to elucidate perceptions relating to general opinion and understanding of generic medicines; behaviors towards generic medicines (e.g., dispensing behaviors in the case of community pharmacists); opinions as to the historical poor usage of generics in Ireland; beliefs held as to the quality and efficacy of generics and how these compare with proprietary (that is, brand-name) medicines; and knowledge and opinion of the impending legislative change. Cognitive testing was performed with 3 individuals who were first asked the questions to be included in the survey, allowed to provide responses, and after responding were asked what their understanding of the questions were. Amendments were made to questions based on responses from all 3 test participants. The responses of these participants to the interview questions were not included in those finally analysed for this study. The interviews used in the study began after cognitive testing had been completed, and the interview questions had been amended. Sampling, Recruitment, and Interviews A convenience sample of community pharmacists was recruited, and interviews completed and analysed. Pharmacists were approached in person, while in the pharmacy, and invited to participate in the study. A verbal explanation of the study was provided, and an invitation letter was offered. One-to- www.amcp.org Vol. 20, No. 11 TABLE 1 Study Instrument: Questions That Formed the Basis for Semistructured Interviews What is your understanding of what a generic medicine is? What is your understanding of how a generic medicine differs from an originator medicine? What is your understanding of bioequivalence? To the best of your knowledge, what percentage of difference is allowed in terms of bioequivalence between an originator medicine and an equivalent generic product? What is your understanding of why generic medicines are cheaper than originator medicines? What do you believe about how generic medicines compare with brandname medicines? What is your opinion as to why use of generic drugs in Ireland has historically been much lower than other European countries? Have you ever had a patient report that a generic medicine, which you dispensed for them, did not work as effectively as an originator medicine? If yes, what type of medicine(s) have you seen this with? Can you please give some brief details of what the patient reported having experienced? What action did you take in this case? Did you then dispense the originator medicine? If yes, was there any reported lack of efficacy from the substituted originator medicine? Have you ever had a patient report that an originator medicine, which you dispensed for them, did not work as effectively as a generic medicine? If yes, what type of medicine(s) have you seen this with? Can you please give some brief details of what the patient reported having experienced? What action did you take in this case? Are you aware of the government’s plans to introduce reference pricing and generic substitution in Ireland? What is your opinion of this proposed change in Irish legislation? one interviews were carried out with consenting pharmacists between June and October 2012: 34 face to face and 10 via telephone. Interview lengths were as follows: minimum 10 minutes 44 seconds; maximum 36 minutes and 15 seconds; mean 19 minutes 29 seconds. Interviews that were recorded (with the interviewee’s consent) were semistructured and based on the described study instrument (see Table 1). Additional supporting assessment of opinions was completed using a series of structured questions to which participants could select from predefined answers (Table 2). In this instance, a 5-point Likert scale was used with a single response allowed for each question.15 Participants were free to volunteer additional commentary on each question. Furthermore, participants were offered the opportunity to express freely any additional opinions or views at the end of the interview session. Participating pharmacists were located in counties Limerick, Tipperary, Kilkenny, Cork, and Waterford. November 2014 JMCP Journal of Managed Care & Specialty Pharmacy 1139 Perceptions and Attitudes of Community Pharmacists Towards Generic Medicines TABLE 2 Study Instrument: Supporting Structured Questions and Pharmacist Responses Do you strongly agree, agree, neither agree nor disagree, disagree, strongly disagree with the following statements: Pharmacists, N = 44 SA/A a SD/Db Nc n % n % n % Generic medicines are generally of the same quality as originator medicines. 43 97.7 1 2.3 0 0.0 Generic medicines are generally poorer quality than originator medicines. 1 2.3 42 95.5 1 2.3 Generic medicines are generally better quality than originator medicines. 1 2.3 27 61.4 16 36.4 Generic medicines work as effectively as originator medicines. 42 0 0.0 2 4.5 95.5 Generic medicines work better than originator medicines. 0 0.0 35 79.5 9 20.5 Generic medicines don’t work as well as originator medicines. 1 2.3 43 97.7 0 0.0 Generic medicines may be dangerous compared with originator medicines. 2 4.5 39 88.6 3 6.8 Generic medicines are as safe as originator medicines. 44 100.0 0 0.0 0 0.0 Generic medicines are manufactured to the same quality as originator medicines. 35 4 9.1 5 11.4 Generic medicines are manufactured to a poorer quality than originator medicines. 4 9.1 39 88.6 1 2.3 Generic medicines are manufactured to a higher quality than originator medicines. 0 0.0 39 88.6 5 11.4 Generic medicines are cheaper to buy than originator medicines. 41 2 4.5 1 2.3 2.3 43 97.7 0 0.0 2 4.5 1 2.3 6.8 39 88.6 2 4.5 Generic medicines are cheaper because they are of inferior quality to originator medicines. 1 If I were ill, I would be happy to take a generic medicine if my doctor prescribed it for me. 41 If I were ill, I would prefer to take an originator medicine rather than a generic medicine, even if it is more expensive. 3 79.5 93.2 93.2 a Strongly agree/agree. disagree/disagree. c Neutral/no opinion. bStrongly Approval of the design and the implementation of this study was granted by the Research Ethics Committee of the Irish College of General Practitioners. Analysis of Data Using a grounded theory approach,16 interviews were transcribed verbatim and imported into NVivo, version 9 (QSR 1140 Journal of Managed Care & Specialty Pharmacy JMCP November 2014 International, Melbourne, Australia) for analysis. Using an inductive process, transcripts were open coded for themes relating to interviewee opinions, perceptions, and behaviors, including any other emerging themes, and the results were analysed using Nvivo. To facilitate visualization and understanding of the numbers of participants holding the perceptions/behaviors that were coded into specific themes, responses were expressed as a percentage of the total number of participants. Interviews were conducted until saturation of data was observed. Analysis was completed by the primary researcher (SD) and reviewed to ensure reliability and rigor of the analysis by a senior investigator (CD). The coding framework included (but was not limited to) such themes as opinions regarding safety and efficacy; previous experience with use of generics; personal preferences; beliefs regarding historical usage of generics in Ireland; experiences with patient reports regarding generics; prescribing rationales; personal knowledge of, and attitudes towards, generic medicines; and opinions regarding the proposed legislative changes. Ongoing analysis of themes emerging from the interviews was carried out as interviews were completed. When 4 to 5 consecutive interviews did not lead to the emergence of any new themes, it was determined that data saturation had been achieved and interviewing was concluded. ■■ Results Supporting quotations from pharmacists are included in Table 3, as referenced in the text. Analyzing Pharmacist Interviews Forty-four community pharmacists were interviewed. Demographics of the group are available in Table 4. Participating pharmacists were located in counties Limerick, Tipperary, Cork, and Waterford. Opinions Regarding Quality, Efficacy, and Safety of Generics Table 2 shows the analysis of opinions regarding quality, efficacy, and safety of generics. The majority of pharmacists (98%) were of the belief that generic medicines are of the same quality as the originator, with 96% holding the view that they are as efficacious as brand-name products. All of the pharmacists interviewed believed that generics are as safe as the originator. A small number (9%), however, were of the opinion that generics are not manufactured to the same quality as originator medicines and were of the view that generic manufacturing is of a poorer standard. The majority of pharmacists (93%) stated that they would take a generic medicine themselves, with a small number (7%) stating that they would prefer to take the originator rather than an equivalent generic, if offered a choice (reference quotations 1-3, Table 3). Vol. 20, No. 11 www.amcp.org Perceptions and Attitudes of Community Pharmacists Towards Generic Medicines TABLE 3 Supporting Quotations from Pharmacists Quotation Number Quotation 1 To be honest . . . for any decisions that I make, or anything I say to customers, if it was me or any of my family and I was given the option of a generic medicine I would go for it, I would absolutely take it. Female, aged 30-39 years 2 I believe that they are equivalent in therapeutic value, and I would have no hesitation in recommending a generic product over a branded one to a customer. Male, aged 19-29 years 3 I think generic medications are brilliant. To be honest with you, the only downside is people’s perceptions—they think that the brands are better when in reality generics are just as good. Male, aged 18-29 years 4 On paper they should be the same and should act in the same way, but we have had cases where people have come in and said that they didn’t find a generic as effective as the original and they prefer the original . . . and from customers’ queries, some of them don’t find that they’re the same. Female, aged 30-39 years 5 [Generic medicines] work the same therapeutically, but I suppose people just have this notion that if it’s cheaper it can’t be as good—that’s the patients perception of it I think. Female, aged 30-39 years 6 I think, to be honest, any time [a patient has] had a problem with a generic instead of a brand is because they feel that they’re being cheated; they basically feel that they’re getting second best because it’s cheaper. Female, aged 30-39 years 7 I think if [the patient] started on a generic, and it’s what they know, they prefer that; so I think that it’s maybe the change—it’s a change management issue more than anything else. Male, aged 40-49 years 8 One woman I can’t convince that the generic coated aspirin would not have caused her to bleed, it’s totally in her mind; you can’t win those battles. Male, aged 50-64 years 9 [If] you know what a patient is satisfied with, you generally won’t rock their boat. Female, aged over 65 years 10 The first thing I would do is I’d try and explain the situation but . . . at the end of the day, I think quite often with people who are coming in, they’ve made up their mind and there’s really very little you can do at that stage. Male, aged 30-39 years 11 You try to explain to them that it is exactly the same medication and explain to them that it’s the same amount of drug, just called something else, that 500 milligrams of the generic drug is exactly the same, that it’s made in the same way, but then if they’re still going “no, no, no,” we’d give them the original. Female, aged 30-39 years 12 There have been an increasing number of incidences where people have come back and said that the quality of the solid dosage form is significantly poorer, and there is one company who are particularly culpable in this regard, whereby their tablets crumble on punching from a blister pack. Their capsules are virtually impossible to get out of the blister pack. Now that’s not to say that there’s anything wrong with the actual raw ingredient, with the medication within the solid dosage form, but there are significant shortcomings in the way those solid dosage forms are compounded. And I think that if it’s not rectified, it is going to compromise patients’ attitudes towards generic medicines. Male, aged 40-49 years 13 I’ve had a couple of issues with a few [generic] tables—they have disintegrated, over time, and that problem didn’t arise with the original drug . . . but 99% of the time there’s no issue with the quality of [generics]. Female, aged 30-39 years 14 Packaging-wise, you definitely notice a difference with some of the generics, that you wouldn’t have half as much detail on the packaging, the boxes are quite plain. I know some customers will only take [the originator] tablets that actually have the label Monday, Tuesday, Wednesday at the back of them, and a lot of generics won’t have any of that detail on the [foil]. Female, aged 30-39 years 15 I’d like the generic companies to package their stuff better; if it’s packaged shabbily it gives a bad impression. Now I know it’s nothing to do with the effectivity of the substance, but some of them are very poorly packaged. Male, aged over 65 years 16 My grandmother in law . . . was on a generic simvastatin which was changed to another simvastatin which happened to be the same color and shape, with no markings, as her blood pressure tablet, and she ended up taking double blood pressure tablets for about 2 weeks. Male, aged 30-39 years 17 If the demographic of patients you deal with are elderly people, and you know they just don’t like change, they want to stay the same, so you’re kind of on the back foot immediately if you’re trying a new drug. Male, aged 30-39 years 18 Quotation from a non-Irish pharmacist: My experience with the Irish psyche is that they’re very brand oriented. I don’t know why, but they tend to be very brand oriented. And, I think that could be impacting on why they don’t like generics; they like the original brand . . . but as soon as you tell them it’s a copy, it’s a generic, they will think it’s a second-class drug. Male, aged 40-49 years 19 I just think people are very used to getting brands; they think all brands are better. It can be to do with the prescribers; some doctors prescribe a brand because that’s what they’ve always known. Male, aged 18-29 years 20 I think that private patients, paying themselves, don’t mind, but that the people that don’t have to pay are the ones that want to stick to the original brand…. It’s GMS [General Medical Services] patients that have a problem with it, not the private patients. Female, aged 30-39 years 21 I think an awful lot of people have it in their head that the generic isn’t as good. As well, medical card patients have commented a few times, ‘It’s because I’m on a medical card that you’re giving me the cheaper tablet.’ That’s the kind of presumption I think that’s out there—people think that because it’s cheaper, they don’t see generic as equivalent, but as a lesser tablet. Female, aged 30-39 years 22 Get rid of branded generics . . . either it’s a generic or its not; there’s no need for the middle ground of a branded generic. Female, aged 30-39 years www.amcp.org Vol. 20, No. 11 November 2014 JMCP Journal of Managed Care & Specialty Pharmacy 1141 Perceptions and Attitudes of Community Pharmacists Towards Generic Medicines TABLE 3 Supporting Quotations from Pharmacists (continued) Quotation Number Quotation 23 Bring it on—pharmacists have been waiting for it for years; we have absolutely no problem, we’re here ready to go, just give us the guidelines and let us just work on it. I’m all about value, we have to be; as professionals and as people that are actually concerned about people’s health and their finances, we want to give the best value. We’re not in the business of trying to rip people off, so give us that law so we can do what we’re supposed to do—which is look after our customers in every way possible and make them feel better. Female, aged 30-39 years 24 I can envisage that people will have issues because we’ve had issues before with people, and when it does come to it, I’d say there is a certain percentage of the population that won’t be happy with taking the generic one or whatever one is the best price at the time because they’re just comfortable taking their one particular brand and that’s it. Female, aged 30-39 years 25 I’ve had some cases [of patient complaints] where it’s actually, say, the Pfizer atorvastatin generic, comes off the same line as Lipitor; the only thing you can blame is patient perception when it’s exactly the same thing. Even after I told [the patient] they’re the same thing, and you know what they said: “They left some of the good stuff out though.” Male, aged 30-39 years 26 I think maybe a lot of the time “generic” and “cheaper” are put in the same sentence, so people think because it’s cheaper it can’t be as good as the original. Male, aged 30-39 years 27 I did have one particular incident where [a patient] reported [a problem with a generic] and [the medication] was actually exactly the same thing; it was a parallel import as opposed to a generic. They saw the pack was different, and they said [the medicine] didn’t work the same, they didn’t want that, but it was, in every sense, exactly the same medication. Female, aged 30-39 years TABLE 4 Group Pharmacists N = 44 Demographics Gender Age M F 18-29 30-39 40-49 50-64 65+ 23 21 9 17 10 5 3 Pharmacist Experiences with Patient Complaints Regarding Generic Medicines Of the 44 pharmacists, 39 (89%) reported receiving patient complaints associated with use of a generic medicine. Of the 5 pharmacists who did not experienced these complaints, 1 pharmacist did not dispense generics. Pharmacists reported that when patients had issues with generics, the main experiences described were that the generics were not as effective or that the patients experienced altered or increased side effects. Twenty-eight pharmacists (64%) expressed an opinion that at least some of the negative experiences reported by patients were not actual, but rather were caused by a nocebo effect (i.e., patients’ preconceived ideas as to a perceived substandard nature of generics led to them having negative experiences with generics) rather than an actual issue with the medication (reference quotations 4-6, Table 3). Medication types most reported as being problematic included protein pump inhibitors (27%, 12/44), statins (18%, 8/44), inhalers (7%, 3/44), antihypertensives (7%, 3/44), antibiotics (7%, 3/44), antidepressants (5%, 2/44), and analgesics (2%, 1/44). Conversely, 11 pharmacists (25%) stated that a patient had reported an issue with an originator medicine compared with 1142 Journal of Managed Care & Specialty Pharmacy JMCP November 2014 a generic. In most cases, since the patient had received the generic before the originator medication, pharmacists indicated that, in their opinion, the patient’s preference is often for the medicine first encountered and that such issues are more likely to be due to a change having occurred, rather than an actual issue with the medicine (reference quotation 7, Table 3). In the situation where pharmacists received complaints from patients related to use of generic medicine and the patients requested the originator instead, 34 pharmacists (77%) stated that they would accede to the patients’ preferences (reference quotations 8-9, Table 3). Only 9 pharmacists (21%) stated that they would attempt to educate the patient (reference quotations 10-11, Table 3). When asked about the differences between an originator and an equivalent generic, 2 pharmacists (5%) felt that there was no difference. Given that the only requirement for similarity (in terms of ingredients) between an originator product and a generic equivalent is that the same active ingredient be used (excipients may vary) and that generic products are often aesthetically different from the originator, patients can be confused if the differences in appearance and excipient content are not adequately explained to them. Opinions Regarding Low Historic Usage of Generics When asked why usage of generics in Ireland has been low in the past, the main reasons given by pharmacists were as follows: • Lack of generic prescribing (31%, 27/44). The primary reasons given for this opinion were familiarity with trade names on the part of prescribers and their lack of knowledge of the generic names of medicines. Vol. 20, No. 11 www.amcp.org Perceptions and Attitudes of Community Pharmacists Towards Generic Medicines • Lack of government incentive or pressure for generics usage (50%, 22/44). • The influence of the pharmaceutical industry (i.e, proprietary manufacturers) in Ireland (41%, 18/44). • Poor understanding of generics by consumers (41%, 18/44). • Brand consciousness or loyalty on the part of the consumer, including being used to a particular brand and having poor cost consciousness (39%, 17/44). • The nonallowance of generic substitution (32%, 14/44). Pharmacist Perceptions of Quality and Patient Issues with Generic Medicines Three pharmacists (7%) reported having experienced quality issues with generic medicines. Issues reported included crumbling tablets and having difficulty getting tablets out of blister packs. The pharmacists reported that, in their opinion, these issues affect consumer confidence in generic products (reference quotations 12-13, Table 3). Poorer packaging was also mentioned by 4 pharmacists (9%) as being perceived as a negative, and 1 pharmacist (2.3%) stated, anecdotally, that differences between originator and generic packaging can even cause issues for patients (e.g., where an originator brand tablet had the days of the week printed on the foil, serving as a reminder to the patient as to whether that day’s medication had been taken or not, but similar printing was not available with the generics). This led to patient preference for the originator medicine (reference quotations 14-15, Table 3). Nineteen pharmacists (43.2%) also reported the opinion that patients are sometimes resistant to change and that the different aesthetic presentation of generics can cause confusion and medication errors for some patients, particularly the elderly (reference quotations 20-21, Table 3). Consequently, patient education was seen as a necessary step for wider acceptance of generics, and 15 pharmacists (34%) stated that, in their opinion, patients see generics as being a substandard, or lesser, alternative because they are cheaper, which is described as “own-brand syndrome.” Indeed, 16 pharmacists (36%) expressed the opinion that Irish patients hold a significant preference for branded medications (reference quotations 18-19, Table 3). Five pharmacists (11.4%) reported having patients who asked for cheaper generics. This was a minority of cases and tended to be limited to private patients, who, according to the pharmacists, have a better understanding and education regarding generics. Pharmacists additionally made reference to General Medical Services (GMS) patients getting more branded medication than private (i.e., self-paying) patients (reference quotations 20-21, Table 3). In Ireland, the GMS, or medical card, scheme is a means-tested scheme available to www.amcp.org Vol. 20, No. 11 persons who are unable, without undue financial hardship, to arrange general practitioner, medical, or surgical services. Having a medical card entitles holders and their dependents to a number of free services, including prescription medicines (a dispensing charge applies to prescription medicines). In quarter 4 of 2013 approximately 40% of the Irish population were holders of medical cards.17 Furthermore, some pharmacists felt that branding of generics should be disallowed because it is contrary to the intent of having generic medication and made it necessary for them to stock multiple “brands” of the same generic medication (reference quotation 22, Table 3). Opinions Regarding New Legislation All of the pharmacists interviewed were aware of the Irish government’s plan to introduce reference pricing and generic substitution in Ireland. When asked about their opinions about the new legislation, 35 pharmacists (80%) indicated that they felt positive about the legislation or were accepting of it. Twentyfour pharmacists (55%) were of the opinion that it made financial sense and was necessary for the country, although 20 pharmacists (46%) expressed concerns and reported that they anticipated issues with its practical implementation (reference quotations 23-24, Table 3). ■■ Discussion According to a PubMed search in January 2014, Irish pharmacists’ perceptions of generic medicines have not been studied in the past. Internationally, a limited number of assessments have taken place for such countries as New Zealand,9 Portugal,4 South Africa,5 Malaysia,8 France,10 and Sweden6 that included studies on views held regarding specific medication types, such as antiepileptic drugs7 and inhalers.3 Given the major changes currently underway in the Irish health care system (i.e., the introduction for the first time of reference pricing and generic substitution), the opinions and behaviors of this critical stakeholder group have the potential to be pivotal to the success or failure of the changes being implemented. In contrast to other reports of reticent pharmacist views,3,8,9 this study has shown that Irish pharmacists were generally positive towards, and accepting of, generic medicines, with many holding the view that they are as effective as the originator, with the exception of nonsubstitutable situations—such as with Narrow Therapeutic Index drugs— and that differences in presentation can be a source of problems for some patients. Very few pharmacists expressed reticent opinions, but 1 of the primary concerns, as has been reported elsewhere,6 was that confusion caused by differing aesthetic presentations of generic medicines has the potential to be problematic for patients. November 2014 JMCP Journal of Managed Care & Specialty Pharmacy 1143 Perceptions and Attitudes of Community Pharmacists Towards Generic Medicines While a majority of pharmacists were in favor of the new legislation (with references made to the United Kingdom situation: that no clinical issues linked to a much greater use of generic medicines are seen, thus, the same situation could reasonably be expected in Ireland without risk to patients) about half of the pharmacists interviewed (46%, 20/44) expressed concerns as to the practical implementation of associated changes. Concerns included the impact on the running of the pharmacy as well as on patients. Pharmacists felt that they could meet considerable resistance from patients and that they, being at the “coal face,” may need to spend substantial periods of time explaining the new system to patients, if adequate educative interventions are not put in place by either the government or other interested bodies (e.g., the Pharmaceutical Society of Ireland). Indeed, the requirement for education of the general public to improve opinions and, therefore, increase patient acceptance of generics was a recurring theme in this study, as it has been in other studies.4-6,10 Increased public awareness and education were considered to be fundamental to improved acceptance of generics by consumers. In fact, an anecdote told by a pharmacist, regarding how she convinced a patient who was reticent to take a generic version of an inhaler, is indicative of how such an intervention might work. The pharmacist told how she brought out both the generic and the proprietary inhalers and showed both to the patient, pointing out the ingredients of both and showing the patient that they were the same. This practical demonstration of equivalence convinced the patient to try the generic inhaler, and the pharmacist indicated that the patient did not return with any subsequent issues. Such examples should be made use of when designing educational interventions for patients. Patient preference was seen to have a considerable influence on dispensing practices, with many pharmacists (77%, 34/44) acceding to patients’ wishes for brand-name medications, despite the fact that pharmacists believed the majority of issues/complaints from patients regarding generics are not actual, but rather due to the nocebo effect, that is, patients’ prejudices regarding generic medicines (reference quotation 25, Table 3). Pharmacists were of the opinion that this negative patient perception may be based on the fact that generics are less expensive so, therefore, cannot be as good (reference quotation 26, Table 3). Also, pharmacists believed that many negative patient experiences were due to changes in medication and that the first medication that the patient is exposed to will tend to be the preferred option. Therefore, when this is changed, the patient is more likely to experience a problem (reference quotation 27, Table 3). 1144 Journal of Managed Care & Specialty Pharmacy JMCP November 2014 Additionally, generics manufacturers/licence holders may play a role in improving the opinions of consumers regarding their products. One aspect could be to ensure that packaging is of a standard at least equivalent to that of the originator and, where relevant, to ensure that it provides the same facilities for prompting/reminding of patients to take the medication (e.g., the anecdote where a pharmacist stated that continued use of 1 proprietary brand was due to patient preference for the packaging, as the days of the week were printed on the blister pack foil). An argument can be made for regulators approving generic medicines to require that if patient aids are part of the originator packaging, any generic equivalents must provide similar aids in order to obtain a marketing authorization. Moreover, a theme emerged on the topic of branded generics: while generic substitution makes the issue of pharmacists needing stocks of multiple branded generics moot, (that is, unless a “do not substitute” prescription has been written), pharmacists expressed views that branding of generics should not be permitted as, practical aspects aside, branding of generic medications is not in keeping with the intention of provision of generic medicines. Indeed, a recent report from the Irish Economic and Social Research Institute on the costing of generics in Ireland has shown them to be similar to the original branded medication, thereby not resulting in substantial benefit to either the Irish exchequer or consumer.18 Since improved consumer confidence in generics was considered to be one of the major hurdles to be overcome in improving use of generics in Ireland (similarly noted in other studies4-6), the question was posed by pharmacists: How can this information/education be provided in a manner that is easy for patients to access and understand? While 1 pharmacist showed a patient both the originator and generic products side by side to prove their equivalency (in the case of asthma inhalers), the practicality of doing this on a day-to-day, patientto-patient basis is obviously something that busy pharmacists cannot undertake. Provision of educational supports could be facilitated, for example, by use of a novel tool, recently published by our group, based on optimized quality of information and reading ability, for development of websites providing health care information.19 The resulting availability of easy-toread handouts/pamphlets, websites, or similar sources of information may not only provide consumers with the information to dispel myths about generics and, hence, improve their confidence but may also have the dual effect of making the role of the pharmacist easier during a time of upheaval and change. Vol. 20, No. 11 www.amcp.org Perceptions and Attitudes of Community Pharmacists Towards Generic Medicines Limitations A possible limitation of this study could be in the selection of participants. All of the pharmacists interviewed were community pharmacists, whose opinions may differ from pharmacists working in hospitals or other settings. Furthermore, differing interview settings (some participants were interviewed face-toface, and others were interviewed over the telephone) might have influenced the data gathered in this study.20 However, review and comparison of the themes emerging from participants interviewed in different settings did not show any substantial difference in the opinions, perceptions, and behaviors expressed between participants. Moreover, while the authors acknowledge that quantification of qualitative data is sometimes contentious, we chose to adopt this approach in order to best provide easy visualization of results and offer a more comprehensive insight into the patient perspective. The strengths of such an approach have been discussed by Schonfelder in 2011.21 A strength of this study is the number of subjects used for qualitative interview; the number of participants in this study compared favorably with the only other semistructured interview-based studies that could be found in PubMed (i.e., 16 participants were interviewed for an analogous study in Sweden6 and 6 pharmacists (from a total of 15 health care professionals) were interviewed for a similar study in South Africa5). ■■ Conclusions Community pharmacists in Ireland hold positive opinions about usage of generic medicines, yet they have concerns about the practical implementation of reference pricing and generic substitution. Concerns were also raised about the impact on patient acceptance due to the varying appearance of generic medicines and regarding the lack of confidence that they observed in the general public in relation to usage of generic medicines. DISCLOSURES The authors do not have any financial interest, have not received any funding for this study, and do not have any conflicts of interest. This work was supported in part by a scholarship from the Faculty of Education and Health Sciences, University of Limerick, Ireland. S. Dunne was responsible for study design, data collection, and data interpretation and was primarily responsible for the writing of the manuscript, with assistance from C. Dunne. All authors contributed equally to manuscript revision. ACKNOWLEDGMENTS The authors wish to express their sincere thanks to all of the pharmacists who took part in these interviews. REFERENCES 1. An Roinn Sláinte (Department of Health). Commencement of the Health (Pricing and Supply of Medical Goods) Act 2013. June 24, 2013. Available at: http://health.gov.ie/blog/press-release/commencement-of-the-health-pricingand-supply-of-medical-goods-act-2013/. Accessed September 19, 2014. 2. Irish Statute Book. Health (Pricing and Supply of Medical Goods) Act 2013. Available at: http://www.irishstatutebook.ie/2013/en/act/pub/0014/ index.html. Accessed September 15, 2014. 3. Williams AE, Chrystyn H. Survey of pharmacists’ attitudes towards interchangeable use of dry powder inhalers. Pharm World Sci. 2007;29(3):221-27. 4. Quintal C, Mendes P. Underuse of generic medicines in Portugal: an empirical study on the perceptions and attitudes of patients and pharmacists. Health Policy. 2012;104(1):61-68. 5. Patel A, Gauld R, Norris P, Rades T. Quality of generic medicines in South Africa: perceptions versus reality—a qualitative study. BMC Health Serv Res. 2012;12:297. 6. Olsson E, Kälvemark Sporrong S. Pharmacists’ experiences and attitudes regarding generic drugs and generic substitution: two sides of the coin. Int J Pharm Pract. 2012;20(6):377-83. 7. McAuley JW, Chen AY, Elliott JO, Shneker BF. An assessment of patient and pharmacist knowledge of and attitudes toward reporting adverse drug events due to formulation switching in patients with epilepsy. Epilepsy Behav. 2009;14(1):113-17. 8. Chong CP, Hassali MA, Bahari MB, Shafie AA. Exploring community pharmacists’ views on generic medicines: a nationwide study from Malaysia. Int J Clin Pharm. 2011;33(1):124-31. 9. Babar ZU, Grover P, Stewart J, et al. Evaluating pharmacists’ views, knowledge, and perception regarding generic medicines in New Zealand. Res Social Adm Pharm. 2011;7(3):294-305. 10. Allenet B, Barry H. Opinion and behaviour of pharmacists towards the substitution of branded drugs by generic drugs: survey of 1,000 French community pharmacists. Pharm World Sci. 2003;25(5):197-202. Authors SUZANNE S. DUNNE, BSc (Hons), MSc, is PhD Candidate; BILL SHANNON, MD, FRCGP, MICGP, is Director of International Liaison; WALTER CULLEN, MD, MICGP, MRCGP, is Professor of General Practice; and COLUM P. DUNNE, BSc (Hons), MBA, PhD, is Chair of Research, Centre for Interventions in Infection, Inflammation and Immunity (4i), Graduate Entry Medical School, University of Limerick, Limerick, Ireland. AUTHOR CORRESPONDENCE: Suzanne S. Dunne, BSc, MSc, Centre for Interventions in Infection, Inflammation and Immunity (4i), Graduate Entry Medical School, University of Limerick, Limerick, Ireland. E-mail: [email protected]. www.amcp.org Vol. 20, No. 11 11. Auta A, Bala ET, Shalkur D. Generic medicine substitution: a crosssectional survey of the perception of pharmacists in North-Central, Nigeria. Med Princ Pract. 2014;23(1):53-58. 12. Maly J, Dosedel M, Kubena A, Vlcek J. Analysis of pharmacists’ opinions, attitudes and experiences with generic drugs and generic substitution in the Czech Republic. Acta Pol Pharm. 2013;70(5):923-31. 13. Moran M. Proposed model for reference pricing and generic substitution. May 2010. Available at: http://www.webcitation.org/6FD1WRh1e. Accessed September 15, 2014. 14. Dunne S, Shannon B, Dunne C, Cullen W. A review of the differences and similarities between generic drugs and their originator counterparts, including economic benefits associated with usage of generic medicines, using Ireland as a case study. BMC Pharmacol Toxicol. 2013;14:1. November 2014 JMCP Journal of Managed Care & Specialty Pharmacy 1145 Perceptions and Attitudes of Community Pharmacists Towards Generic Medicines 15. Likert R. A technique for the measurement of attitudes. Archives of Psychology. 1932;22(140):5-55. 16. Engward H. Understanding grounded theory. Nurs Stand. 2013;28(7):37-41. 17. Health Service Executive. Health Service managment data report. October 2013. Available at: http://www.hse.ie/eng/services/publications/ corporate/performanceassurancereports/Oct13mdreport.pdf. Accessed September 19, 2014. 18. Brick A, Gorecki PK, Nolan A. Ireland: pharmaceutical prices, prescribing practices and usage of generics in a comparative context. Economic and Social Research Institute Research Series No. 2. June 2013. Available at: http://www.esri.ie/UserFiles/publications/RS32_Pharma.pdf. Accessed September 15, 2014. 1146 Journal of Managed Care & Specialty Pharmacy JMCP November 2014 19. Dunne S, Cummins N, Hannigan A, Shannon B, Dunne C, Cullen W. A method for the design and development of medical or health care information websites to optimize search engine results page rankings on Google. J Med Internet Res. 2013;15(8):e183. 20. Elwood SA, Martin DG. “Placing” interviews: location and scales of power in qualitative research. Prof Geogr. 2000;52(4):649-57. 21. Schönfelder W. CAQDAS and qualitative syllogism logic—NVivo 8 and MAXQDA 10 compared. Forum Qual Soc Res. 2011;12(1):Article 21. Vol. 20, No. 11 www.amcp.org UPCOMING WEBINARS AMCP webinars explore the hottest topics in managed care pharmacy! n Breaking the Link Between Pain Management and Substance Abuse November 5, 2014 n Compounding Basics: Current Federal and State Laws and Regulations November 19, 2014 n /c ale g r o . p c w w w.am | 2:00 pm–3:00 pm ET Health Care Reform Update and Strategies for 2015–2016 December 3, 2014 n | 2:00 pm–3:00 pm ET p U n g i S Today! | 2:00 pm–3:00 pm ET Are you Leveraging your MTM Program to Fit Your MCO’s ACO Strategy? December 10, 2014 | 2:00 pm–3:00 pm ET n AMCP Advocacy 101: What Do I Need to Know? December 17, 2014 | 2:00 pm–3:00 pm ET n AMCP 2015 Legislative & Regulatory Priorities January 14, 2015 | 2:00 pm–3:00 pm ET Recordings of AMCP’s webinars can be found in the Virtual Conference Center on AMCP’s website. Go to www.amcp.org/webinars for more information. ndar ONE HE’S BEEN WAITING FOR IS HERE harvoni.com/hcp NOW APPrOVED One tablet contains ledipasvir and sofosbuvir HARVONI, the HARVONI logo, GILEAD and the GILEAD logo are trademarks of Gilead Sciences, Inc., or its related companies. ©2014 Gilead Sciences, Inc. All rights reserved. GILP0423 10/14