November/December 2012, Vol 5, No 7 by Dalia Buffery

THE PEER-REVIEWED FORUM FOR EVIDENCE IN BENEFIT DESIGN ™ NOVEMBER/DECEMBER 2012

VOLUME 5, NUMBER 7

FOR PAYERS, PURCHASERS, POLICYMAKERS, AND OTHER HEALTHCARE STAKEHOLDERS

PERSPECTIVES

Getting Back to Reality: The Election, the Fiscal Cliff, and the ACA Joseph R. Antos, PhD

Interesting Times Ahead: Payers’ Innovation in the Era of Healthcare Reform Albert Tzeel, MD, MHSA, FACPE EDITORIAL

The Use of Medicines in the United States: A Detailed Review David B. Nash, MD, MBA ™

CLINICAL

National Burden of Preventable Adverse Drug Events Associated with Inpatient Injectable Medications: Healthcare and Medical Professional Liability Costs Betsy J. Lahue, MPH; Bruce Pyenson, FSA, MAAA; Kosuke Iwasaki, FIAJ, MAAA, MBA; Helen E. Blumen, MD, MBA; Susan Forray, FCAS, MAAA; Jeffrey M. Rothschild, MD, MPH Stakeholder Perspective by Jaan Sidorov, MD, MHSA BUSINESS

Current and Future Use of HEOR Data in Healthcare Decision-Making in the United States and in Emerging Markets Anke-Peggy Holtorf, PhD, MBA; Diana Brixner, RPh, PhD; Brandon Bellows, PharmD; Abdulkadir Keskinaslan, MD, MBA, MPH; Joseph Dye, RPh, PhD; Gary Oderda, PharmD, MPH Stakeholder Perspective by Douglas S. Burgoyne, PharmD

Hematologic Complications, Healthcare Utilization, and Costs in Commercially Insured Patients with Myelodysplastic Syndrome Receiving Supportive Care Annette Powers, PharmD, MBA; Claudio Faria, PharmD, MPH; Michael S. Broder, MD, MSHS; Eunice Chang, PhD; Dasha Cherepanov, PhD Stakeholder Perspective by Jeffrey A. Bourret, MS, RPh, FASHP Industry Trends

The Policy-Driven Health Plan: A Roadmap for Value-Based Reimbursement

How will you use your potential savings? Imagine what up to 67% overall savings on insulin can do for your hospital. At the foundation of our new GPO contract are benefits including: t Incentive-based pricing t Access to diabetes education resources for patients and staff t One partner for a portfolio of insulin, prefilled pens, and safety needles

To learn more, follow these simple instructions for direct access from your Smartphone: Step 1: Download the free mobile app on your phone browser at http://gettag.mobi Step 2: Launch the app and scan the tag to the left Or visit our Contract Partners page under the Partnering tab at www.novonordisk-us.com.

ÂŠ 2012 Novo Nordisk

Printed in the U.S.A.

0512-00009322-1

July 2012

NOVEMBER/DECEMBER 2012

VOLUME 5, NUMBER 7

THE PEER-REVIEWED FORUM FOR EVIDENCE IN BENEFIT DESIGN ™

™

FOR PAYERS, PURCHASERS, POLICYMAKERS, AND OTHER HEALTHCARE STAKEHOLDERS

TABLE OF CONTENTS

PERSPECTIVES

403 Getting Back to Reality: The Election, the Fiscal Cliff, and the ACA Joseph R. Antos, PhD 408 Interesting Times Ahead: Payers’ Innovation in the Era of Healthcare Reform Albert Tzeel, MD, MHSA, FACPE CLINICAL

413 National Burden of Preventable Adverse Drug Events Associated with Inpatient Injectable Medications: Healthcare and Medical Professional Liability Costs Betsy J. Lahue, MPH; Bruce Pyenson, FSA, MAAA; Kosuke Iwasaki, FIAJ, MAAA, MBA; Helen E. Blumen, MD, MBA; Susan Forray, FCAS, MAAA; Jeffrey M. Rothschild, MD, MPH 421 Stakeholder Perspective: Injectable Sticker Shock: A Call to Action Jaan Sidorov, MD, MHSA EDITORIAL

423 The Use of Medicines in the United States: A Detailed Review David B. Nash, MD, MBA

Continued on page 400

Publisher Nicholas Englezos nick@engagehc.com 732-992-1884 Editorial Director Dalia Buffery dalia@engagehc.com 732-992-1889 Associate Publisher Maurice Nogueira maurice@engagehc.com 732-992-1895 Associate Editor Lara J. Lorton lara@engagehc.com 732-992-1892 Editorial Assistant Jennifer Brandt jbrandt@the-lynx-group.com 732-992-1536 Executive Vice President Engage Managed Markets Chuck Collins ccollins@engagehc.com 732-992-1894 National Accounts Manager Zach Ceretelle zach@engagehc.com 732-992-1898 Senior Production Manager Lynn Hamilton Quality Control Director Barbara Marino Business Manager Blanche Marchitto Founding Editor-in-Chief Robert E. Henry

Mission Statement American Health & Drug Benefits is founded on the concept that health and drug benefits have undergone a transformation: the econometric value of a drug is of equal importance to clinical outcomes as it is to serving as the basis for securing coverage in formularies and benefit designs. Because benefit designs are greatly affected by clinical, business, and policy conditions, this journal offers a forum for stakeholder integration and collaboration toward the improvement of healthcare.

American Health & Drug Benefits is included in the following indexing and database services: EMBASE/Elsevier Bibliographic Database SCOPUS/Elsevier Bibliographic Database Cumulative Index to Nursing and Allied Health Literature (CINAHL) EBSCO research databases Standard Periodical Directory

This publication further provides benefit design decision makers the integrated industry information they require to devise formularies and benefit designs that stand up to today’s special healthcare delivery and business needs.

MEMBER: Committee on Publication Ethics (COPE)

Contact Information: For subscription information and editorial queries, please contact: editorial@engagehc.com; tel: 732-992-1892; fax: 732-992-1881

Vol 5, No 7

November/December 2012

www.AHDBonline.com

American Health & Drug Benefits

399

NOVEMBER/DECEMBER 2012

VOLUME 5, NUMBER 7

THE PEER-REVIEWED FORUM FOR EVIDENCE IN BENEFIT DESIGN ™

™

FOR PAYERS, PURCHASERS, POLICYMAKERS, AND OTHER HEALTHCARE STAKEHOLDERS

TABLE OF CONTENTS

(Continued)

BUSINESS

428 Current and Future Use of HEOR Data in Healthcare Decision-Making in the United States and in Emerging Markets Anke-Peggy Holtorf, PhD, MBA; Diana Brixner, RPh, PhD; Brandon Bellows, PharmD; Abdulkadir Keskinaslan, MD, MBA, MPH; Joseph Dye, RPh, PhD; Gary Oderda, PharmD, MPH 438 Stakeholder Perspective: Health Economic Outcomes Research Data Key in Coverage Decisions of New Medications Douglas S. Burgoyne, PharmD 455 Hematologic Complications, Healthcare Utilization, and Costs in Commercially Insured Patients with Myelodysplastic Syndrome Receiving Supportive Care Annette Powers, PharmD, MBA; Claudio Faria, PharmD, MPH; Michael S. Broder, MD, MSHS; Eunice Chang, PhD; Dasha Cherepanov, PhD 465 Stakeholder Perspective: Reconsidering the Management of Younger Patients with Myelodysplastic Syndrome Jeffrey A. Bourret, MS, RPh, FASHP

American Health & Drug Benefits, ISSN 1942-2962 (print); ISSN 1942-2970 (online), is published 8 times a year by Engage Healthcare Communications, LLC, 1249 South River Rd, Suite 202A, Cranbury, NJ 08512. Copyright © 2012 by Engage Healthcare Communications, LLC. All rights reserved. American Health & Drug Benefits and The Peer-Reviewed Forum for Evidence in Benefit Design are trademarks of Engage Healthcare Communications, LLC. No part of this publication may be reproduced or transmitted in any form or by any means now or hereafter known, electronic or mechanical, including photocopy, recording, or any informational storage and retrieval system, without written permission from the Publisher. Printed in the United States of America. Address all editorial correspondence to: editorial@engagehc.com Telephone: 732-992-1892 Fax: 732-992-1881 American Health & Drug Benefits 1249 South River Rd, Suite 202A, Cranbury, NJ 08512 The ideas and opinions expressed in American Health & Drug Benefits do not necessarily reflect those of the Editorial Board, the Editors, or the Publisher. Publication of an advertisement or other product mentioned in American Health & Drug Benefits should not be construed as an endorsement of the product or the manufacturer’s claims. Readers are encouraged to contact the manufacturers about any features or limitations of products mentioned. Neither the Editors nor the Publisher assume any responsibility for any injury and/or damage to persons or property arising out of or related to any use of the material mentioned in this publication.

DEPARTMENTS

442 2012 Peer Reviewers INDUSTRY TRENDS 446 The Policy-Driven Health Plan: A Roadmap for Value-Based Reimbursement James Evans WEB EXCLUSIVE Annual Index 2012

For permission to reuse material from American Health & Drug Benefits (ISSN 1942-2962), please access www. copyright.com <http://www.copyright. com/> or contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood Drive, Danvers, MA 01923, 978-750-8400. POSTMASTER: CORRESPONDENCE REGARDING SUBSCRIPTIONS OR CHANGE OF ADDRESS should be directed to CIRCULATION DIRECTOR, American Health & Drug Benefits, 1249 South River Rd, Suite 202A, Cranbury, NJ 08512. Fax: 732-992-1881. YEARLY SUBSCRIPTION RATES: One year: $99.00 USD; Two years: $149.00 USD; Three years: $199.00 USD.

Vol 5, No 7

November/December 2012

www.AHDBonline.com

American Health & Drug Benefits

400

Just a Click Away! Care Management Central®

Care Management Solutions offers resources designed to help S HealthCoach4Me treat the whole patient and inspire health and wellness. Resources are only a click away: Care Management Central® (CMC), an easy-to-use interactive Web site where professionals can build care management toolkits to help address specific healthcare goals across multiple disease states HealthCoach4MeSM, an interactive health coaching Web site designed to help consumers achieve their most important health goals

Get to know Health Coaching Essentials™ on CMC:

• A series of online interactive learning modules that address a variety of health coaching topics • Based on the premise that health behavior change comes from the patient, not an outside authority • Created to help healthcare professionals communicate with and motivate patients to make healthy behavior changes • Helps motivate patients to make challenging lifestyle changes to address chronic health conditions • Designed for individual or group learning

Find Health Coaching Essentials™ and other care management resources and tools on the Care Management Central ® Web site. For more information, please contact your GlaxoSmithKline account representative.

For healthcare professionals

For consumers

www.CareManagementCentral.com www.HealthCoach4Me.com

Printed in USA.

HM3372R1

April 2012

EDITORIAL BOARD EDITOR-IN-CHIEF

David B. Nash, MD, MBA Dean, the Dr Raymond C. and Doris N. Grandon Professor, Jefferson School of Population Health Philadelphia, PA DEPUTY EDITORS

Joseph D. Jackson, PhD Program Director, Applied Health Economics and Outcomes Research, Jefferson University School of Population Health, Philadelphia Laura T. Pizzi, PharmD, MPH, RPh Associate Professor, Dept. of Pharmacy Practice, Jefferson School of Pharmacy, Philadelphia AGING AND WELLNESS

Eric G. Tangalos, MD, FACP, AGSF, CMD Professor of Medicine Mayo Clinic, Rochester, MN CANCER RESEARCH

Al B. Benson, III, MD, FACP Professor of Medicine, Associate Director for Clinical Investigations Robert H. Lurie Comprehensive Cancer Center Northwestern University, IL Past Chair, NCCN Board of Directors Samuel M. Silver, MD, PhD, FASCO Professor of Internal Medicine Hematology/Oncology Assistant Dean for Research Associate Director, Faculty Group Practice University of Michigan Medical School EMPLOYERS

Arthur F. Shinn, PharmD, FASCP President, Managed Pharmacy Consultants, LLC, Lake Worth, FL F. Randy Vogenberg, RPh, PhD Principal, Institute for Integrated Healthcare and Bentteligence, Sharon, MA ENDOCRINOLOGY

James V. Felicetta, MD Chairman, Dept. of Medicine Carl T. Hayden Veterans Affairs Medical Center, Phoenix, AZ Quang Nguyen, DO, FACP, FACE Adjunct Associate Professor Endocrinology Touro University Nevada, College of Osteopathic Medicine

Joseph Couto, PharmD, MBA Clinical Program Manager Cigna Corporation, Bloomfield, CT Steve Miff, PhD Senior Vice President VHA, Inc., Irving, TX Kavita V. Nair, PhD Associate Professor, School of Pharmacy University of Colorado at Denver, CO Gary M. Owens, MD President, Gary Owens Associates Glen Mills, PA Andrew M. Peterson, PharmD, PhD Dean, Mayes School of Healthcare Business and Policy, Associate Professor, University of the Sciences, Philadelphia, PA Sarah A. Priddy, PhD Director, Competitive Health Analytics Humana, Louisville, KY Timothy S. Regan, BPharm, RPh, CPh Executive Director, Strategic Accounts Xcenda, Palm Harbor, FL Vincent J. Willey, PharmD Associate Professor, Philadelphia School of Pharmacy, University of the Sciences Philadelphia, PA David W. Wright, MPH President, Institute for Interactive Patient Care Bethesda, MD HEALTH & VALUE PROMOTION

Craig Deligdish, MD Hematologist/Oncologist Oncology Resource Networks, Orlando, FL Thomas G. McCarter, MD, FACP Chief Clinical Officer Executive Health Resources, PA Albert Tzeel, MD, MHSA, FACPE National Medical Director HumanaOne, Waukesha, WI MANAGED MARKETS

Jeffrey A. Bourret, RPh, MS, FASHP Senior Director, Medical Lead, Payer and Specialty Channel Strategy, Medical Affairs Pfizer Specialty Care Business Unit, PA Richard B. Weininger, MD Chairman, CareCore National, LLC Bluffton, SC

EPIDEMIOLOGY RESEARCH

Joshua N. Liberman, PhD Executive Director, Research, Development & Dissemination Sutter Health, Concord, CA GOVERNMENT

Kevin B. “Kip” Piper, MA, FACHE President, Health Results Group, LLC Washington, DC HEALTH INFORMATION TECHNOLOGY Kelly Huang, PhD President, HealthTronics, Inc. Austin, TX J. B. Jones, PhD, MBA Research Investigator, Geisinger Health System, Danville, PA Victor J. Strecher, PhD, MPH Professor and Director for Innovation and Social Entrepreneurship University of Michigan, School of Public Health and Medicine, Ann Arbor HEALTH OUTCOMES RESEARCH

Diana Brixner, RPh, PhD Professor & Chair, Dept. of Pharmacotherapy Executive Director, Outcomes Research Center, Director of Outcomes Personalized Health Care Program, University of Utah, Salt Lake City

402

PATIENT ADVOCACY

William E. Fassett, BSPharm, MBA, PhD, FAPhA Professor of Pharmacy Law & Ethics Dept. of Pharmacotherapy, College of Pharmacy Washington State University, Spokane, WA Mike Pucci Sr VP Commercial Operations and Business Development, PhytoChem Pharmaceuticals Lake Gaston, NC PERSONALIZED MEDICINE

Emma Kurnat-Thoma, PhD, MS, RN Washington, DC Amalia M. Issa, PhD, MPH Professor and Chair Department of Health Policy and Public Health Director, Program in Personalized Medicine & Targeted Therapeutics University of the Sciences, Philadelphia PHARMACOECONOMICS

Josh Feldstein President & CEO CAVA, The Center for Applied Value Analysis, Inc., Norwalk, CT Jeff Jianfei Guo, BPharm, MS, PhD Professor of Pharmacoeconomics & Pharmacoepidemiology, College of Pharmacy, Univ of Cincinnati, Medical Center, OH

American Health & Drug Benefits

www.AHDBonline.com

PHARMACY BENEFIT DESIGN

Joel V. Brill, MD, AGAF, CHCQM Chief Medical Officer, Predictive Health, Phoenix, AZ Leslie S. Fish, PharmD Vice President of Clinical Programs Fallon Community Health Plan, MA John Hornberger, MD, MS Cedar Associates, LLC CHP/PCOR Adjunct Associate, Menlo Park, CA Michael S. Jacobs, RPh Vice President, National Accounts Truveris, Inc., New York, NY Matthew Mitchell, PharmD, MBA Manager, Pharmacy Services SelectHealth, Salt Lake City, UT Paul Anthony Polansky, BSPharm, MBA Senior Field Scientist, Health Outcomes and PharmacoEconomics (HOPE) Endo Health Solutions, Chadds Ford, PA Christina A. Stasiuk, DO, FACOI Senior Medical Director Cigna, Philadelphia, PA Scott R. Taylor, BSPharm, MBA Executive Director, Industry Relations Geisinger Health System, Danville, PA POLICY & PUBLIC HEALTH

Joseph R. Antos, PhD Wilson H. Taylor Scholar in Health Care Retirement Policy, American Enterprise Institute Washington, DC Robert W. Dubois, MD, PhD Chief Science Officer National Pharmaceutical Council, Washington, DC Jack E. Fincham, PhD, RPh Professor of Pharmacy, Practice and Administration School of Pharmacy, University of Missouri Kansas City, MO Walid F. Gellad, MD, MPH Assistant Professor of Medicine, University of Pittsburgh, Staff Physician, Pittsburgh VA Medical Center, Adjunct Scientist, RAND Health Paul Pomerantz, MBA Executive Director Drug Information Association, Horsham, PA J. Warren Salmon, PhD Professor of Health Policy & Administration School of Public Health University of Illinois at Chicago Raymond L. Singer, MD, MMM, CPE, FACS Chief, Division of Cardiothoracic Surgery Vice Chair, Department of Surgery for Quality & Patient Safety and Outreach Lehigh Valley Health Network, PA RESEARCH & DEVELOPMENT

Frank Casty, MD, FACP Chief Medical Officer Senior VP, Clinical Development Medical Science Endo Pharmaceuticals, Chadds Ford, PA Michael F. Murphy, MD, PhD Chief Medical Officer and Scientific Officer Worldwide Clinical Trials King of Prussia, PA SPECIALTY PHARMACY

Atheer A. Kaddis, PharmD Senior Vice President Managed Markets/Clinical Services Diplomat Specialty Pharmacy, Flint, MI James T. Kenney, Jr, RPh, MBA Pharmacy Operations Manager Harvard Pilgrim Health Care Wellesley, MA Michael Kleinrock Director, Research Development IMS Institute for Healthcare Informatics Collegeville, PA

November/December 2012

Vol 5, No 7

PERSPECTIVE

Getting Back to Reality: The Election, the Fiscal Cliff, and the ACA Joseph R. Antos, PhD Wilson H. Taylor Scholar in Health Care and Retirement Policy American Enterprise Institute, Washington, DC

resident Obama’s victory in November resolves whether the Affordable Care Act (ACA) will be implemented. In general terms, it will. Although the Republican-controlled House of Representatives may pass another resolution to abolish the ACA, that threat is not credible, with Democrats in charge in the Senate. But that does not settle what will actually become of the law. Action or inaction by the states, reaction from insurers and healthcare providers, responses from consumers, and budget-cutting policies adopted by Congress to ease the country off the fiscal cliff will determine the future shape of reform and its impact on the health sector.

States in the Driver’s Seat The biggest challenge facing the second-term Obama White House comes from the states. Perhaps surprisingly, the federal government does not have the power to reshape the healthcare sector on its own. The ACA relies on the states to act as Washington’s agent in implementing the health insurance exchanges, enforcing new insurance regulations, and expanding eligibility for Medicaid. The ACA’s drafters recognized that not all states shared their vision of healthcare reform. Consequently, the legislation includes a combination of steep penalties to force state compliance and federal fallbacks in the event that some states do not comply. Those provisions are unlikely to be effective in obtaining full state cooperation. The Supreme Court decision in June restored the states’ ability to manage their Medicaid programs by declaring unconstitutional the threat that states failing to expand eligibility would lose all of their federal Medicaid funding.1 Republican governors continue to question whether it is in their best interest to create their own insurance exchanges.2 The fallback federal exchange that is supposed to fill in when states have not acted has yet to show signs of life, and there is an ongoing debate over whether the federal insurance exchange can distribute insurance subsidies. Even when states are willing to take on the complex new tasks laid out by the ACA, many are unlikely to meet the deadlines set forth in the law. For the most part, this is not a case of political recalcitrance.

Vol 5, No 7

November/December 2012

Insurance Exchanges Creating an insurance exchange is a complex task that only a few states have previously attempted. A top executive with Maryland Health Benefit Exchange, which has been under active development for 2 years, doubts that states beginning now to implement an exchange will be ready to start enrolling beneficiaries by October 2013.3 Even Massachusetts faces challenges, because its exchange system was developed before enactment of the ACA.4 To comply with the new federal requirements, Massachusetts will have to disassemble parts of its current system. That may prove as difficult as starting from scratch. With 30 Republican state governors next year, as well as some Democratic governors questioning the wisdom of proceeding with full implementation of the ACA, the Obama administration is beginning to show signs that it will accommodate state concerns to some extent. On November 9, Health and Human Services (HHS) Secretary Kathleen Sebelius sent a letter to all governors, extending the deadline for submitting a plan for a state exchange (known as a “blueprint”) from November 16 to December 14, 2012.5,6 On November 15, HHS Secretary Sebelius moved the deadline for states’ declaring their intention to establish an exchange to December 14, 2012, as well, responding to a request from the Republican Governors Association.7,8 These are the first of many compromises to come as states—including those governed by Democrats—recognize that they are in the driver’s seat of healthcare reform. States that fail to meet the federal deadlines for implementing insurance exchanges face no penalty other than a possible delay by the HHS in approving their plans. States that implement exchanges late, or not at all, risk losing insurance subsidies for some of their currently uninsured citizens, but even that is uncertain. The Obama administration insists that the federally facilitated exchange will distribute subsidies, despite clear language in the ACA to the contrary.9 Having created a new entitlement program, the administration intends to spend the money. That raises some interesting possibilities. Could a state create its own insurance exchange but not implement all

www.AHDBonline.com

American Health & Drug Benefits

403

PERSPECTIVE

of the insurance market regulations imposed by Washington?10 If so, would individuals purchasing coverage in such an exchange still be eligible for the subsidies? What punitive actions could the administration take against states that do not vigorously enforce ACA rules? Considering the political consequences of withholding financial support from otherwise deserving families who need health insurance, the federal government probably would seek some avenue other than the insurance subsidies to rein in contrary states. That assumes the administration thinks this is a fight it could win. The alternative approach—not establishing an exchange and leaving it to the federal government—will be attractive to some states, because it avoids any implicit alignment between the state and Washington politics. It also eliminates the need to resolve technical problems that are inevitable with such an enterprise, does not prevent a state from creating an exchange later, and blame for any problems can be shifted to the federal government. This wait-and-see approach is likely how a dozen or more states will proceed.

Medicaid Thanks to the Supreme Court’s decision, the states are more clearly in control with regard to expanding Medicaid eligibility. Faced with fiscal problems, many states are likely to seek ways to maximize federal contributions for health insurance—or at least minimize their own. Although expanding Medicaid to everyone with incomes up to 138% of the federal poverty line will ensure full federal payments for the expansion population for several years, the matching rate eventually drops to 90% of the cost of benefits. There is no guarantee that Congress would not reduce that matching rate in the future.11 States will also bear additional costs as individuals “come out of the woodwork” to enroll in Medicaid. The federal matching rate does not increase for anyone newly enrolling in Medicaid if they were previously eligible. Given the risks, states would be better off financially to take advantage of the subsidies in the exchanges that are available to persons with a household income between 100% and 400% of the federal poverty line.12 One approach would be to limit the Medicaid expansion to incomes up to 100% of the federal poverty line and encourage those with higher incomes to enroll in the exchanges. As an added inducement, states could supplement the exchange subsidies with a few hundred dollars of their own funds for each person who would otherwise have enrolled in Medicaid, reducing their premium cost to zero. States have flexibility in setting eligibility, and the

404

American Health & Drug Benefits

administration has gone out of its way to make that point. In an August meeting of the National Conference of State Legislatures, federal Medicaid director Cindy Mann said that states could expand Medicaid to new populations under the ACA and later drop the coverage.13,14 This gesture of acceptance (if not support) raises the possibility that the Obama administration may be willing to negotiate more favorable terms with states that choose to expand coverage. States want more discretion in how they operate their Medicaid programs. This was a central issue in the Republican proposal to provide states with block grants for Medicaid rather than continuing to require them to seek federal permission for even modest changes in states’ policies and management procedures. States will make their “best deal” with the federal government on exchanges, Medicaid, and other aspects of the ACA rather than simply falling into line with the administration’s policies. Even if a Republican House blocks changes that could make the rules easier to live with, the administration will use the ample authority of the ACA to grant waivers, make exceptions, and otherwise look the other way when circumstances and politics dictate. As a result, the dream of some Democrats for a nationally uniform health insurance system will be replaced with the reality of a multitude of systems with varying degrees of regulatory control, much as we see today.

The Budget Knife Comes Out A second major threat to the ACA is purely a product of Washington politics. Unless the lame duck Congress acts by the end of the year, we will fall off the fiscal cliff. That means starting in January 2013 with a large increase in tax rates, large across-the-board reductions in federal spending through a sequester, and a cap on federal borrowing that limits the government’s ability to run up larger deficits. Three months later, the federal government’s authorization to spend money for discretionary programs expires. Funding could grind to a halt for education, infrastructure development, defense, and other activities but would continue largely unrestrained for the major entitlements—Social Security, Medicare, and Medicaid. The Congressional Budget Office warns that without legislative action, the combination of abruptly higher taxes and abruptly lower spending will trigger another recession.15 That message seems to have gotten through to both political parties, and negotiations are under way between congressional leaders and the president. The bargain that will eventually be struck will probably be a combination of tax increases focused on those with high incomes and an agreement in principle to reform entitlement spending. The bad news for the healthcare sector is that the

www.AHDBonline.com

November/December 2012

Vol 5, No 7

PERSPECTIVE

sequester imposes a much gentler cut in federal payments than is likely under the deal that is expected to ease us off the fiscal cliff. Medicare is scheduled for a 2% reduction under the sequester, and Medicaid spending would not be cut. The president’s 2013 budget that was released in February 2012 gives a good indication of where the administration stands on this issue. That document proposed to double the Medicare cuts, with $292 billion in savings through 2022, and cut federal Medicaid payments by $52 billion.16 Those figures ignore the impending 27% cut in Medicare physician fees under the sustainable growth rate formula. The 10-year cost of replacing the scheduled reductions with a freeze amounts to $271 billion.17 Congress will almost certainly delay any cut in physician fees for next year, pushing that cost forward to 2014. Filling the larger budget hole that includes a physician payment fix will force Congress to accept a mix of Medicare fee reductions (other than for physicians) and increases in beneficiary costs. Rather than inventing new ways to cut provider payments, the fiscal compromise will probably accelerate implementation of the “productivity adjustments” and other fee-reduction provisions imposed by the ACA. Medicare premiums are likely to rise (perhaps to an average of 35% of the cost of Part B and Part D), with more of the cost paid by higher income beneficiaries.18 Other initiatives are more speculative, but possible. Medicare could shed its historical division into Part A and Part B, which no longer has any usefulness (if it ever did). Combining Medicare’s parts into a single comprehensive health benefit would allow restructuring the current and complex cost-sharing requirements that make no sense. A single, all-encompassing deductible, a simple uniform copayment or coinsurance structure across all services, and the addition of a cap on catastrophic expenses would convert the program into modern insurance. Medicare Part B premium would become a premium for the entire benefit, with appropriate adjustment to the rate to avoid overburdening beneficiaries. These are stopgap measures at best. Structural reforms are needed that shift Medicare from an openended entitlement program to a budgeted approach. Premium support remains a viable but dormant political idea. Scare stories about “vouchers” during the campaign did not create a wave of senior backlash against Republicans, but there is also no wave of senior enthusiasm for major program reforms. For the time being, Congress is likely to adopt more modest reforms intended to improve the competitive bidding process for Medicare Advantage and to give accountable care organizations more control over and accountability for the delivery and cost of their services.

Vol 5, No 7

November/December 2012

Another target for budget cutting is the ACA’s subsidy for insurance offered on the exchanges, totaling $1 trillion through 2022.19 The Obama administration will be reluctant to modify the subsidy structure, particularly if they believe that the Supreme Court’s decision to make the health insurance mandate a tax significantly weakens the pressure on individuals to purchase coverage. They may reason that a generous subsidy is needed under that circumstance to maintain a stable insurance market. Nonetheless, reducing the subsidy (perhaps by cutting its generosity and limiting its availability to individuals with incomes below 250% of the poverty line rather than the current 400%) offers substantial budget savings. Those savings may be easier for the public to accept than cuts in existing programs that already have well-established and politically vocal constituents. There is also a chance that the deficit fighters could reduce the tax subsidy for employer-sponsored health insurance. The ACA includes a “Cadillac tax” on employer coverage that is deemed too expensive. The Obama administration has agreed that our current tax subsidy is inefficient, even though it endorsed an inefficient way to address this problem. The tax is imposed on insurers, but obviously most of it will be passed on, in the form of higher premiums, to workers, who will receive a tax subsidy on the higher amount. The political objective was to avoid admitting that the tax would be paid by those who will, in fact, pay it. Broad tax reform could make sense out of this by directly limiting the amount of insurance premium that can be excluded from personal income tax.

Will It Work? The Healthcare Sector and Consumers React Passage of the ACA in 2010 triggered the adoption of massive changes in the practices of health insurers and, to a lesser extent, healthcare providers. Despite the uncertainty of the election, the healthcare industry has taken as given that the basic structure of the ACA would remain intact. With the reelection of President Obama, and the torrent of proposed regulations and other guidance that has ensued in recent weeks, implementation efforts continue with renewed urgency. The question is—will it work? The third threat to the ACA is the response of the market to the policy ambitions of the Obama administration. Although their criticism has been muted, insurers and health plans have warned that the overlapping requirements that are meant to protect consumers will drive up costs, drive away customers, and destabilize the individual insurance market. The self-imposed problems are numerous. For example, partial community rating limits the premiums charged to

www.AHDBonline.com

American Health & Drug Benefits

405

PERSPECTIVE

older people to no more than 3 times the rate charged young purchasers. In most states, the variation in age rating is 5 or 6 to 1. The result is lower premiums for older consumers, offset by higher premiums for the young. The mandate to buy health insurance was added to force young people to purchase high-priced coverage, but the ACA explicitly limits enforcement to a slap on the wrist. The essential benefits package was intended to ensure good coverage, but that will drive up premiums further, making insurance unaffordable for many and increasing the cost of federal subsidies. The political desire to halt what one federal official calls “some of the worst insurance industry practices”20 will collide with the real-world challenges of operating an insurance market. Older people use more healthcare. Consumers decide to purchase insurance based on their assessment of their own need for healthcare services. Medical underwriting (which sets premiums and benefit restrictions based on the person’s health) was instituted to avoid disastrous losses that could result from enrolling high users, without charging commensurately high rates. The ACA eliminates these mechanisms, which are subject to abuse, but does not adequately address the underlying problem of market incentives and risk selection. Health insurers will do their best to succeed in the new environment. Millions of people will seek to purchase health insurance through the exchanges. But insurance costs will rise, and the exchanges will not provide that easy one-stop-shopping experience that proponents of the ACA imagined. Those most eager to find coverage they can afford are likely to be disappointed. The next 2 years are critical to the president’s healthcare legislation. Although there is general agreement that the ACA, like all major legislation, has flaws that should be fixed, no fixes are likely in 2013. The president is already viewed with suspicion by his liberal supporters, who are disappointed that healthcare reform did not produce national health insurance, and who fear that he may be ready to lay a heavy hand on their favorite social insurance programs.21 Republicans will not support legislation that corrects errors in the ACA, which they would view as capitulation. Unless the 2 sides can agree on structural reforms in Medicare and substantial reductions in federal spending, the ACA will limp into 2014 unchanged. Shortly after the election, Speaker John Boehner said that “Obamacare is the law of the land.”22 That was not just a statement of the obvious. The ACA with all its warts has set the government’s healthcare agenda, and citizens, states, employers, and the healthcare sector will have to deal with it. That will be much harder if we fail to reenergize the American economy and rein in our

406

American Health & Drug Benefits

burgeoning federal debt. Unless President Obama’s successor is extraordinarily fortunate, he or she will revisit these thorny issues 4 years from now. ■

References 1. Kaiser Family Foundation. A guide to the Supreme Court’s decision on the ACA’s Medicaid expansion. August 2012. www.kff.org/healthreform/upload/8347.pdf. Accessed November 25, 2012. 2. Republican Governors Public Policy Committee. Letter to President Barack Obama on the ACA. November 14, 2012. www.blogs.roanoke.com/politics/files/2012/ 11/2012-11-14-POTUS_Letter.pdf. Accessed November 25, 2012. 3. Daly R. Setting up exchanges may be tough for latecomers, experts say. Modern Healthcare. November 14, 2012. www.modernhealthcare.com/article/20121114/ NEWS/311149967. Accessed November 25, 2012. 4. Kaiser Family Foundation. Massachusetts health care reform: six years later. May 2012. www.kff.org/healthreform/upload/8311.pdf. Accessed November 25, 2012. 5. Sebelius K. Letter to governors on exchange blueprint deadline. November 9, 2012. www.healthcare.gov/law/resources/letters/exchange-blueprint-letter.pdf. Accessed November 25, 2012. 6. Centers for Medicare & Medicaid Services. Blueprint for approval of affordable state-based and state partnership insurance exchanges. www.cciio.cms.gov/resources/ files/hie-blueprint-081312.pdf. Accessed November 25, 2012. 7. Sebelius K. Letter to governors on exchange declaration deadline. November 15, 2012. www.healthcare.gov/law/resources/letters/exchange-declaration-deadline.pdf. Accessed November 25, 2012. 8. Republican Governors Public Policy Committee. Letter to President Barack Obama on the ACA. November 14, 2012. www.blogs.roanoke.com/politics/files/ 2012/11/2012-11-14-POTUS_Letter.pdf. Accessed November 25, 2012. 9. Pear R. Brawling over health care moves to rules on exchanges. New York Times. July 7, 2012. www.nytimes.com/2012/07/08/us/critics-of-health-care-law-prepare-tobattle-over-insurance-exchange-subsidies.html. Accessed November 25, 2012. 10. Roy A. What states should build instead of Obamacare’s health insurance exchanges. November 19, 2012. www.forbes.com/sites/aroy/2012/11/19/what-statesshould-build-instead-of-obamacares-health-insurance-exchanges/. Accessed November 25, 2012. 11. Haislmaier EF, Gonshorowski D. State lawmaker’s guide to evaluating Medicaid expansion projections. Heritage Foundation. September 7, 2012. www.heritage.org/ research/reports/2012/09/state-lawmakers-guide-to-evaluating-medicaid-expansionprojections. Accessed November 25, 2012. 12. Antos JR. Healthcare reform after SCOTUS: hard decisions needed to avoid health sector meltdown. Am Health Drug Benefits. 2012;5(5):273-276. 13. Kaiser Health News. Medicaid official outlines state flexibility in health law’s Medicaid Expansion. August 7, 2012. www.kaiserhealthnews.org/daily-reports/2012/ august/07/health-law-implementation.aspx. Accessed November 25, 2012. 14. Pear R. Administration advises states to expand Medicaid or risk losing federal money. New York Times. October 2, 2012. www.nytimes.com/2012/10/02/us/us-advisesstates-to-expand-medicaid-or-risk-losing-funds.html. Accessed November 25, 2012. 15. Congressional Budget Office. Economic effects of policies contributing to fiscal tightening in 2013. November 2012. www.cbo.gov/sites/default/files/cbofiles/ attachments/11-08-12-FiscalTightening.pdf. Accessed November 25, 2012. 16. Office of Management and Budget. Budget of the United States government, fiscal year 2013. Table S-9, Mandatory and Receipt Proposals. February 13, 2012. www.whitehouse.gov/sites/default/files/omb/budget/fy2013/assets/tables.pdf. Accessed November 25, 2012. 17. Congressional Budget Office. Medicare’s payments to physicians: the budgetary impact of alternative policies relative to CBO’s March 2012 baseline. July 2012. www. cbo.gov/sites/default/files/cbofiles/attachments/43502-SGR%20Options2012.pdf. Accessed November 25, 2012. 18. The Debt Reduction Task Force. Restoring America’s future, bipartisan policy center. November 17, 2010. www.bipartisanpolicy.org/library/report/restoring-americasfuture. Accessed November 25, 2012. 19. Congressional Budget Office. Estimates for the insurance coverage provisions of the Affordable Care Act updated for the recent Supreme Court decision. July 2012. www.cbo.gov/sites/default/files/cbofiles/attachments/43472-07-24-2012CoverageEstimates.pdf. Accessed November 25, 2012. 20. Pugh T. Insurers’ duties under health care law taking shape. McClatchy Newspapers. November 20, 2012. www.mcclatchydc.com/2012/11/20/175202/insurers-dutiesunder-health-care.html. Accessed November 25, 2012. 21. Delaney A. AARP warns lawmakers about potential “fiscal cliff” deal. Huffington Post. November 8, 2012. www.huffingtonpost.com/2012/11/08/aarp-warns-lawmakersabou_n_2091960.html. Accessed November 25, 2012. 22. Kaiser Health News. Boehner backpedals after calling Obamacare “law of the land.” November 9, 2012. www.kaiserhealthnews.org/Daily-Reports/2012/November/ 09/Boehner-and-the-health-law.aspx. Accessed November 25, 2012.

www.AHDBonline.com

November/December 2012

Vol 5, No 7

Tell your patients about NovoTwist®, the first and only single-twist needle attachment on the market.

GOOD DESIGN

The benefits of NovoTwist® are:

Winner of the US Good Design™ award1

• Less time consuming for patients to attach • Features an audible and tactile confirmatory click for correct attachment • Available in 30G (8mm) and 32G Tip (5mm) needles Available for use with FlexPen® and other compatible Novo Nordisk devices.*

For more information, visit myflexpen.com or call 1-800-727-6500

* Designed to be used with Levemir® FlexPen®, NovoLog® FlexPen®, NovoLog® Mix 70/30 FlexPen®, and other compatible Novo Nordisk delivery devices. Please refer to the delivery device user manual to see if NovoTwist® can be used with your device. Also refer to the user manual for information on assembly and injection. Needles are sold separately and may require a prescription in some states. Reference: 1. http://www.chi-athenaeum.org/gdesign/2010/medical/index.html

Printed in the U.S.A.

0311-00002127-1

April 2011

PERSPECTIVE

Interesting Times Ahead: Payers’ Innovation in the Era of Healthcare Reform Albert Tzeel, MD, MHSA, FACPE National Medical Director, HumanaOne, Waukesha, WI

here is an old saying, attributed to the Chinese but never confirmed to be theirs, “May you live in interesting times.” Although this phrase is thought to be a curse, it is really more of a double entendre. It is also quite applicable to what health plans are currently experiencing. The Supreme Court’s upholding of the Affordable Care Act (ACA) as law and the reelection of President Obama have ensured that interesting times lie ahead for many entities, let alone health plans. In response to these interesting times, health plans have taken steps to confront challenges, redefine risks, and modulate their business model. This article provides my individual perspective on where we are going.

Business Agility In confronting challenges, the key for health plans lies in business agility. Agility allows them to provide the internal resources to address the main aspects of the ACA law according to the various timetables put forth by the federal government and, specifically, the US Department of Health and Human Services. Health plans had already begun addressing the actuarial aspects of the ACA when they found themselves needing to prepare for medical loss ratio targets. Moving forward, plans have now made key assessments as to whether they will participate in various state health insurance exchanges and, if so, how will they market themselves to the millions of newly insured persons who will seek insurance coverage. This does not even take into account how plans will actually manage the health, or illness, of those newly insured millions for whom, of course, they are also planning. As Ray Stata, the founder and current Chairman of the Board of Analog Devices, once said, “I came to the conclusion long ago that limits to innovation have less to do with technology or creativity than organizational agility. Inspired individuals can only do so much.”1 For payers, then, business agility will, by necessity, permeate The opinions expressed in this article are solely those of the author and do not necessarily reflect the opinions of the author’s employer or of any other entity mentioned herein.

408

American Health & Drug Benefits

the organization: it becomes inherent not only in marketing, but also in finance, clinical, business intelligence, and operations. As the late quality guru W. Edwards Deming opined, “It is not necessary to change. Survival is not mandatory.”

Redefining Risks As a sidelight to survival, health plans also need to redefine risks. In the prereform era, payers had significant opportunity to mitigate financial risk, especially for their fully insured populations, and for their individual medical plan policies. (Self-funded groups have the ability to set their own parameters for coverage such that, in providing administrative services only, the payers followed the client’s intentions for coverage. Self-funded plans are also impacted by the ACA, but they are beyond the scope of this discussion.) In that era, health plans could decline to cover a group or an individual or could apply coverage rules to control losses (eg, preexisting clauses). Moreover, if a payer accepted a group or an individual and experienced significant financial losses, that payer could alleviate its risk through reinsurance or by raising rates on premiums. The ACA eliminates the ability to decline an applicant, as well as the ability to rider medical conditions or apply preexisting limitation rules. Add that to the medical loss ratios mentioned above, and payers could find themselves in a vicious cycle of mounting losses, compounded with an inability to restrain them. Redefining risks, then, requires 2 key changes in a payer’s way of thinking—the first addresses philosophy, the second addresses operations. Rethinking Health Insurance Philosophically, health plans need to redefine themselves. Many health plans have already either done so or are in the process of doing so. UnitedHealthcare, for example, is “committed to the delivery of quality care and its continual improvement.”2 WellPoint strives to “improve the lives of the people we serve and the health of our communities.”3 Humana’s business decisions are geared toward improving “the health and well-being of

www.AHDBonline.com

November/December 2012

Vol 5, No 7

PERSPECTIVE

our members, our associates, the communities we serve, and our planet.”4 Of note, in these 3 examples, there is no mention of benefits or of cost. There is, however, specific mention of improving community health.

Payer–Physician Relationships And that is where the second change comes in— payer relationships with physicians. The old adage that “all healthcare is local” certainly comes into play here. If the overarching goal for payers is to improve community health, then that starts with the physicians who deliver the care. The need for payers to redefine risk for themselves complements a new era in payer–physician relations, in which payers will allow physicians more flexibility to care for members/patients in exchange for physicians’ accepting accountability for the outcomes achieved. Every large payer is involved in some sort of arrangement with accountable care organizations (ACOs); Aetna even has an entire unit devoted to this concept.5 ACOs allow physicians to accept some modicum of risk in exchange for shared savings based on outcomes. Although ACOs may still be in their infancy, as they continue to proliferate, to some extent they will certainly serve as a vehicle for payers to redefine some of their own risk. Evolution of the Business Model In further meeting their new obligations for agility while redefining risks, payers will also need to modulate their business model. There are as many ways to change business models as there are models. Most of these entail using various levels of information, such as that found in medical/pharmacy claims, in demographics, or in health risk assessments. To that end, payers appear to be functioning, to some extent, as healthcare infomediaries— entities that profile consumer purchasing and utilization patterns and then customize services to meet that consumer’s needs.6 Payers are also tying in consumer/member utilization patterns with provider practice patterns in more sophisticated ways. Whereas in the past payers worked to automate the transactive nature of physician workflow through sites such as Availity, payers now use these sites for health information exchange.7 Some payers have even taken the grander step of purchasing their own health information exchanges: UnitedHealthcare, Humana, and Aetna have advanced abilities to exchange information through their purchases of Axolotl, Certify Data Systems, and Medicity, respectively. In addition, with a greater involvement in health information exchange, payers are engaged in broader analytics. While searching for where relationships in the data lead to prediction of diagnosis, utilization, or out-

Vol 5, No 7

November/December 2012

come, payers have taken “number crunching” to new levels.8 Moreover, to add even more numbers “to crunch,” a coming deluge of “big data” will continue to keep payers busy.9 As payers collect physical health information off of their members’ personal tracking devices, the possibilities for addressing payers’ concerns regarding cost, outcomes, and improving community health will promote further tweaking and adjustments in the business model. And as seen by payer purchases, among other examples, we can expect these business model adjustments to be made quickly.

Common Goals Driving Innovation Even though I have provided a personal opinion on the challenges of agility, the redefinition of risk, and the need for business model evolution, I have hardly touched on major aspects of the ACA. Nevertheless, it is clear that when it comes to a payer perspective, we live in interesting times, indeed. Would our present “interesting times” define a coming crisis for health plans? I believe it would not. Many people ask if the Chinese symbol for “crisis” is made up of the symbols for “danger” and “opportunity,” as is often quoted. Literally, the answer is yes.10 However, from a pragmatic perspective, one can better define “crisis” as “opportunity in a time of danger.” Are payers operating in a potentially dangerous period, given the unknowns and the uncertainty inherent in deadlines and commitments? We could say maybe. However, with the ingenuity they have shown as a function of their agility and business acumen, I believe that payers have and will continue to embrace what they do to achieve their ultimate and common goals of promoting financial security for their members, improved health for their populations, and a healthier way of life for the communities they serve. ■ References 1. Stata R. Creating Minds.org. Creative quotes and quotations on business. www. creatingminds.org/quotes/business.htm. Accessed December 2, 2012. 2. UnitedHealthcare. About Us. www.uhc.com/about_us.htm. Accessed December 2, 2012. 3. WellPoint. About WellPoint. www.wellpoint.com/AboutWellPoint/MissionValues/ index.htm. Accessed December 2, 2012. 4. Humana. About Humana. www.humana.com/resources/about/. Accessed December 2, 2012. 5. Accountable Care Solutions from Aetna. www.aetnaacs.com/. Accessed December 2, 2012. 6. Gartner Inc. HI (Healthcare Infomediary). www.gartner.com/it-glossary/hi-healthcareinfomediary/. Accessed December 2, 2012. 7. Availity. About Availity. www.availity.com/about-availity/. Accessed December 2, 2012. 8. Tibken S. Numbers, numbers and more numbers. Wall Street J. April 16, 2012. www. online.wsj.com/article/SB10001424052702304692804577285821129341442.html. Accessed December 2, 2012. 9. Hernandez D. Big data is transforming healthcare. Wired Science. October 16, 2012. www.wired.com/wiredscience/2012/10/big-data-is-transforming-healthcare/. Accessed December 2, 2012. 10. Living Chinese Symbols. Chinese symbol Crisis. www.living-chinese-symbols.com/ chinese-symbol-crisis.html. Accessed December 2, 2012.

www.AHDBonline.com

American Health & Drug Benefits

409

The median age of patients in the VISTA† trial was 71 years (range: 48-91).

Indication and Important Safety Information for VELCADE® (bortezomib) INDICATION VELCADE (bortezomib) is indicated for the treatment of patients with multiple myeloma. CONTRAINDICATIONS VELCADE is contraindicated in patients with hypersensitivity (not including local reactions) to bortezomib, boron, or mannitol, including anaphylactic reactions. VELCADE is contraindicated for intrathecal administration. WARNINGS, PRECAUTIONS, AND DRUG INTERACTIONS ▼ Peripheral neuropathy: Manage with dose modification or discontinuation. Patients with preexisting severe neuropathy should be treated with VELCADE only after careful risk-benefit assessment. ▼ Hypotension: Use caution when treating patients taking antihypertensives, with a history of syncope, or with dehydration.

▼ Cardiac toxicity: Worsening of and development of cardiac failure have occurred. Closely monitor patients with existing heart disease or risk factors for heart disease. ▼ Pulmonary toxicity: Acute respiratory syndromes have occurred. Monitor closely for new or worsening symptoms. ▼ Posterior reversible encephalopathy syndrome: Consider MRI imaging for onset of visual or neurological symptoms; discontinue VELCADE if suspected. ▼ Gastrointestinal toxicity: Nausea, diarrhea, constipation, and vomiting may require use of antiemetic and antidiarrheal medications or fluid replacement. ▼ Thrombocytopenia or Neutropenia: Monitor complete blood counts regularly throughout treatment. ▼ Tumor lysis syndrome: Closely monitor patients with high tumor burden. ▼ Hepatic toxicity: Monitor hepatic enzymes during treatment.

In treating multiple myeloma

What is the value of ® VELCADE (bortezomib)? ▼ Overall survival advantage ▼ Defined length of therapy ▼ Medication cost IF YOU DEFINE VALUE AS AN OVERALL SURVIVAL ADVANTAGE: VELCADE (bortezomib) combination delivered a >13-month overall survival advantage At 5-year median follow-up, VELCADE+MP* provided a median overall survival of 56.4 months vs 43.1 months with MP alone (HR=0.695 [95% CI, 0.57-0.85]; p<0.05)† At 3-year median follow-up, VELCADE+MP provided an overall survival advantage over MP that was not regained with subsequent therapies

IF YOU DEFINE VALUE AS DEFINED LENGTH OF THERAPY: Results achieved using VELCADE twice-weekly followed by weekly dosing for a median of 50 weeks (54 planned)1

IF YOU DEFINE VALUE AS MEDICATION COST: Medication cost is an important factor when considering overall drug spend. The Wholesale Acquisition Cost for VELCADE is $1,506 per 3.5-mg vial as of July 2012 When determining the value of a prescription drug regimen, it may be worth considering medication cost, length of therapy, and dosing regimens. This list is not all-inclusive; there are additional factors to consider when determining value for a given regimen

▼ Embryo-fetal risk: Women should avoid becoming pregnant while being treated with VELCADE. Advise pregnant women of potential embryo-fetal harm. ▼ Closely monitor patients receiving VELCADE in combination with strong CYP3A4 inhibitors. Avoid concomitant use of strong CYP3A4 inducers. ADVERSE REACTIONS Most commonly reported adverse reactions (incidence ≥20%) in clinical studies include nausea, diarrhea, thrombocytopenia, neutropenia, peripheral neuropathy, fatigue, neuralgia, anemia, leukopenia, constipation, vomiting, lymphopenia, rash, pyrexia, and anorexia. Please see Brief Summary for VELCADE on the next page of this advertisement. For Reimbursement Assistance, call 1-866-VELCADE (835-2233), Option 2, or visit VELCADEHCP.com.

Reference: 1. Mateos M-V, Richardson PG, Schlag R, et al. Bortezomib plus melphalan and prednisone compared with melphalan and prednisone in previously untreated multiple myeloma: updated follow-up and impact of subsequent therapy in the phase III VISTA trial. J Clin Oncol. 2010;28(13):2259-2266. *Melphalan+prednisone. † VISTA TRIAL: a randomized, open-label, international phase 3 trial (N=682) evaluating the efficacy and safety of VELCADE administered intravenously in combination with MP vs MP in previously untreated multiple myeloma. The primary endpoint was TTP. Secondary endpoints were CR, ORR, PFS, and overall survival. At a prespecified interim analysis (median follow-up 16.3 months), VELCADE+MP resulted in significantly superior results for TTP (median 20.7 months with VELCADE+MP vs 15.0 months with MP [p=0.000002]), PFS, overall survival, and ORR. Further enrollment was halted and patients receiving MP were offered VELCADE in addition. Updated analysis was performed.

Brief Summary INDICATIONS: VELCADE® (bortezomib) for Injection is indicated for the treatment of patients with multiple myeloma. VELCADE for Injection is indicated for the treatment of patients with mantle cell lymphoma who have received at least 1 prior therapy. CONTRAINDICATIONS: VELCADE is contraindicated in patients with hypersensitivity (not including local reactions) to bortezomib, boron, or mannitol, including anaphylactic reactions. VELCADE is contraindicated for intrathecal administration. WARNINGS AND PRECAUTIONS: Peripheral Neuropathy: VELCADE treatment causes a peripheral neuropathy that is predominantly sensory; however, cases of severe sensory and motor peripheral neuropathy have been reported. Patients with pre-existing symptoms (numbness, pain, or a burning feeling in the feet or hands) and/or signs of peripheral neuropathy may experience worsening peripheral neuropathy (including ≥Grade 3) during treatment with VELCADE. Patients should be monitored for symptoms of neuropathy, such as a burning sensation, hyperesthesia, hypoesthesia, paresthesia, discomfort, neuropathic pain or weakness. In the Phase 3 relapsed multiple myeloma trial comparing VELCADE subcutaneous vs intravenous, the incidence of Grade ≥2 peripheral neuropathy events was 24% for subcutaneous and 39% for intravenous. Grade ≥3 peripheral neuropathy occurred in 6% of patients in the subcutaneous treatment group, compared with 15% in the intravenous treatment group. Starting VELCADE subcutaneously may be considered for patients with pre-existing or at high risk of peripheral neuropathy. Patients experiencing new or worsening peripheral neuropathy during VELCADE therapy may require a decrease in the dose and/or a less dose-intense schedule. In the VELCADE vs dexamethasone phase 3 relapsed multiple myeloma study, improvement in or resolution of peripheral neuropathy was reported in 48% of patients with ≥Grade 2 peripheral neuropathy following dose adjustment or interruption. Improvement in or resolution of peripheral neuropathy was reported in 73% of patients who discontinued due to Grade 2 neuropathy or who had ≥Grade 3 peripheral neuropathy in the phase 2 multiple myeloma studies. The long-term outcome of peripheral neuropathy has not been studied in mantle cell lymphoma. Hypotension: The incidence of hypotension (postural, orthostatic, and hypotension NOS) was 8%. These events are observed throughout therapy. Caution should be used when treating patients with a history of syncope, patients receiving medications known to be associated with hypotension, and patients who are dehydrated. Management of orthostatic/postural hypotension may include adjustment of antihypertensive medications, hydration, and administration of mineralocorticoids and/or sympathomimetics. Cardiac Toxicity: Acute development or exacerbation of congestive heart failure and new onset of decreased left ventricular ejection fraction have occurred during VELCADE therapy, including reports in patients with no risk factors for decreased left ventricular ejection fraction. Patients with risk factors for, or existing, heart disease should be closely monitored. In the relapsed multiple myeloma study of VELCADE vs dexamethasone, the incidence of any treatment-related cardiac disorder was 8% and 5% in the VELCADE and dexamethasone groups, respectively. The incidence of adverse reactions suggestive of heart failure (acute pulmonary edema, pulmonary edema, cardiac failure, congestive cardiac failure, cardiogenic shock) was ≤1% for each individual reaction in the VELCADE group. In the dexamethasone group, the incidence was ≤1% for cardiac failure and congestive cardiac failure; there were no reported reactions of acute pulmonary edema, pulmonary edema, or cardiogenic shock. There have been isolated cases of QT-interval prolongation in clinical studies; causality has not been established. Pulmonary Toxicity: Acute Respiratory Distress Syndrome (ARDS) and acute diffuse infiltrative pulmonary disease of unknown etiology, such as pneumonitis, interstitial pneumonia, and lung infiltration have occurred in patients receiving VELCADE. Some of these events have been fatal. In a clinical trial, the first two patients given high-dose cytarabine (2 g/m2 per day) by continuous infusion with daunorubicin and VELCADE for relapsed acute myelogenous leukemia died of ARDS early in the course of therapy. There have been reports of pulmonary hypertension associated with VELCADE administration in the absence of left heart failure or significant pulmonary disease. In the event of new or worsening cardiopulmonary symptoms, consider interrupting VELCADE until a prompt, comprehensive, diagnostic evaluation is conducted. Posterior Reversible Encephalopathy Syndrome (PRES): Posterior Reversible Encephalopathy Syndrome (PRES; formerly termed Reversible Posterior Leukoencephalopathy Syndrome (RPLS)) has occurred in patients receiving VELCADE. PRES is a rare, reversible, neurological disorder, which can present with seizure, hypertension, headache, lethargy, confusion, blindness, and other visual and neurological disturbances. Brain imaging, preferably MRI (Magnetic Resonance Imaging), is used to confirm the diagnosis. In patients developing PRES, discontinue VELCADE. The safety of reinitiating VELCADE therapy in patients previously experiencing PRES is not known. Gastrointestinal Toxicity: VELCADE treatment can cause nausea, diarrhea, constipation, and vomiting, sometimes requiring use of antiemetic and antidiarrheal medications. Ileus can occur. Fluid and electrolyte replacement should be administered to prevent dehydration. Interrupt VELCADE for severe symptoms. Thrombocytopenia/Neutropenia: VELCADE is associated with thrombocytopenia and neutropenia that follow a cyclical pattern, with nadirs occurring following the last dose of each cycle and typically recovering prior to initiation of the subsequent cycle. The cyclical pattern of platelet and neutrophil decreases and recovery remained consistent over the 8 cycles of twice-weekly dosing, and there was no evidence of cumulative thrombocytopenia or neutropenia. The mean platelet count nadir measured was approximately 40% of baseline. The severity of thrombocytopenia was related to pretreatment platelet count. In the relapsed multiple myeloma study of VELCADE vs dexamethasone, the incidence of bleeding (≥Grade 3) was 2% on the VELCADE arm and <1% on the dexamethasone arm. Complete blood counts (CBC) should be monitored frequently during treatment with VELCADE. Platelet counts should be monitored prior to each dose of VELCADE. Patients experiencing thrombocytopenia may require change in the dose and schedule of VELCADE. Gastrointestinal and intracerebral hemorrhage has been reported in association with VELCADE. Transfusions may be considered. Tumor Lysis Syndrome: Tumor lysis syndrome has been reported with VELCADE therapy. Patients at risk of tumor lysis syndrome are those with high tumor burden prior to treatment. Monitor patients closely and take appropriate precautions. Hepatic Toxicity: Cases of acute liver failure have been reported in patients receiving multiple concomitant medications and with serious underlying medical conditions. Other reported hepatic reactions include hepatitis, increases in liver enzymes, and hyperbilirubinemia. Interrupt VELCADE therapy to assess reversibility. There is limited re-challenge information in these patients.

Embryo-fetal: Pregnancy Category D. Women of reproductive potential should avoid becoming pregnant while being treated with VELCADE. Bortezomib administered to rabbits during organogenesis at a dose approximately 0.5 times the clinical dose of 1.3 mg/m2 based on body surface area caused post-implantation loss and a decreased number of live fetuses. ADVERSE EVENT DATA: Safety data from phase 2 and 3 studies of single-agent VELCADE 1.3 mg/m2/dose administered intravenously twice weekly for 2 weeks followed by a 10-day rest period in 1163 patients with previously-treated multiple myeloma (N=1008) and previously-treated mantle cell lymphoma (N=155) were integrated and tabulated. In these studies, the safety profile of VELCADE was similar in patients with multiple myeloma and mantle cell lymphoma. In the integrated analysis, the most commonly reported (≥10%) adverse reactions were nausea (49%), diarrhea NOS (46%), fatigue (41%), peripheral neuropathies NEC (38%), thrombocytopenia (32%), vomiting NOS (28%), constipation (25%), pyrexia (21%), anorexia (20%), anemia NOS (18%), headache NOS (15%), neutropenia (15%), rash NOS (13%), paresthesia (13%), dizziness (excl vertigo 11%), and weakness (11%). Eleven percent (11%) of patients experienced at least 1 episode of ≥Grade 4 toxicity, most commonly thrombocytopenia (4%) and neutropenia (2%). A total of 26% of patients experienced a serious adverse reaction during the studies. The most commonly reported serious adverse reactions included diarrhea, vomiting, and pyrexia (3% each), nausea, dehydration, and thrombocytopenia (2% each), and pneumonia, dyspnea, peripheral neuropathies NEC, and herpes zoster (1% each). In the phase 3 VELCADE+melphalan and prednisone study in previously untreated multiple myeloma, the safety profile of VELCADE administered intravenously in combination with melphalan/prednisone is consistent with the known safety profiles of both VELCADE and melphalan/prednisone. The most commonly reported adverse reactions in this study (VELCADE+melphalan/prednisone vs melphalan/prednisone) were thrombocytopenia (48% vs 42%), neutropenia (47% vs 42%), peripheral neuropathy (46% vs 1%), nausea (39% vs 21%), diarrhea (35% vs 6%), neuralgia (34% vs <1%), anemia (32% vs 46%), leukopenia (32% vs 28%), vomiting (26% vs 12%), fatigue (25% vs 14%), lymphopenia (23% vs 15%), constipation (23% vs 4%), anorexia (19% vs 6%), asthenia (16% vs 7%), pyrexia (16% vs 6%), paresthesia (12% vs 1%), herpes zoster (11% vs 3%), rash (11% vs 2%), abdominal pain upper (10% vs 6%), and insomnia (10% vs 6%). In the phase 3 VELCADE subcutaneous vs intravenous study in relapsed multiple myeloma, safety data were similar between the two treatment groups. The most commonly reported adverse reactions in this study were peripheral neuropathy NEC (37% vs 50%), thrombocytopenia (30% vs 34%), neutropenia (23% vs 27%), neuralgia (23% vs 23%), anemia (19% vs 23%), diarrhea (19% vs 28%), leukopenia (18% vs 20%), nausea (16% vs 14%), pyrexia (12% vs 8%), vomiting (9% vs 11%), asthenia (7% vs 16%), and fatigue (7% vs 15%). The incidence of serious adverse reactions was similar for the subcutaneous treatment group (20%) and the intravenous treatment group (19%). The most commonly reported SARs were pneumonia and pyrexia (2% each) in the subcutaneous treatment group and pneumonia, diarrhea, and peripheral sensory neuropathy (3% each) in the intravenous treatment group. DRUG INTERACTIONS: Bortezomib is a substrate of cytochrome P450 enzyme 3A4, 2C19 and 1A2. Co-administration of ketoconazole, a strong CYP3A4 inhibitor, increased the exposure of bortezomib by 35% in 12 patients. Monitor patients for signs of bortezomib toxicity and consider a bortezomib dose reduction if bortezomib must be given in combination with strong CYP3A4 inhibitors (eg, ketoconazole, ritonavir). Co-administration of omeprazole, a strong inhibitor of CYP2C19, had no effect on the exposure of bortezomib in 17 patients. Co-administration of rifampin, a strong CYP3A4 inducer, is expected to decrease the exposure of bortezomib by at least 45%. Because the drug interaction study (n=6) was not designed to exert the maximum effect of rifampin on bortezomib PK, decreases greater than 45% may occur. Efficacy may be reduced when VELCADE is used in combination with strong CYP3A4 inducers; therefore, concomitant use of strong CYP3A4 inducers is not recommended in patients receiving VELCADE. St. John’s wort (Hypericum perforatum) may decrease bortezomib exposure unpredictably and should be avoided. Co-administration of dexamethasone, a weak CYP3A4 inducer, had no effect on the exposure of bortezomib in 7 patients. Co-administration of melphalan-prednisone increased the exposure of bortezomib by 17% in 21 patients. However, this increase is unlikely to be clinically relevant. USE IN SPECIFIC POPULATIONS: Nursing Mothers: It is not known whether bortezomib is excreted in human milk. Because many drugs are excreted in human milk and because of the potential for serious adverse reactions in nursing infants from VELCADE, 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 VELCADE in children has not been established. Geriatric Use: No overall differences in safety or effectiveness were observed between patients ≥age 65 and younger patients receiving VELCADE; but greater sensitivity of some older individuals cannot be ruled out. Patients with Renal Impairment: The pharmacokinetics of VELCADE are not influenced by the degree of renal impairment. Therefore, dosing adjustments of VELCADE are not necessary for patients with renal insufficiency. Since dialysis may reduce VELCADE concentrations, VELCADE should be administered after the dialysis procedure. For information concerning dosing of melphalan in patients with renal impairment, see manufacturer’s prescribing information. Patients with Hepatic Impairment: The exposure of bortezomib is increased in patients with moderate and severe hepatic impairment. Starting dose should be reduced in those patients. Patients with Diabetes: During clinical trials, hypoglycemia and hyperglycemia were reported in diabetic patients receiving oral hypoglycemics. Patients on oral antidiabetic agents receiving VELCADE treatment may require close monitoring of their blood glucose levels and adjustment of the dose of their antidiabetic medication. Please see full Prescribing Information for VELCADE at VELCADEHCP.com.

VELCADE, MILLENNIUM and are registered trademarks of Millennium Pharmaceuticals, Inc. Other trademarks are property of their respective owners. Millennium Pharmaceuticals, Inc., Cambridge, MA 02139 Copyright © 2012, Millennium Pharmaceuticals, Inc. All rights reserved. Printed in USA

V-12-0306

11/12

CLINICAL

Original research

National Burden of Preventable Adverse Drug Events Associated with Inpatient Injectable Medications: Healthcare and Medical Professional Liability Costs Betsy J. lahue, MPh; Bruce Pyenson, Fsa, Maaa; Kosuke iwasaki, FiaJ, Maaa, MBa; helen e. Blumen, MD, MBa; susan Forray, Fcas, Maaa; Jeffrey M. rothschild, MD, MPh Background: Harmful medication errors, or preventable adverse drug events (ADEs), are a prominent quality and cost issue in healthcare. Injectable medications are important therapeutic agents, but they are associated with a greater potential for serious harm than oral medications. The national burden of preventable ADEs associated with inpatient injectable medications and the associated medical professional liability (MPL) costs have not been previously described in the literature. Objective: To quantify the economic burden of preventable ADEs related to inpatient injectable medications in the United States. Methods: Medical error data (MedMarx 2009-2011) were utilized to derive the distribution of errors by injectable medication types. Hospital data (Premier 2010-2011) identified the numbers and the types of injections per hospitalization. US payer claims (2009-2010 MarketScan Commercial and Medicare 5% Sample) were used to calculate the incremental cost of ADEs by payer and by diagnosis-related group (DRG). The incremental cost of ADEs was defined as inclusive of the time of inpatient admission and the following 4 months. Actuarial calculations, assumptions based on published literature, and DRG proportions from 17 state discharge databases were used to derive the probability of preventable ADEs per hospitalization and their annual costs. MPL costs were assessed from state- and national-level industry reports, premium rates, and from closed claims databases between 1990 and 2011. The 2010 American Hospital Association database was used for hospital-level statistics. All costs were adjusted to 2013 dollars. Results: Based on this medication-level analysis of reported harmful errors and the frequency of inpatient administrations with actuarial projections, we estimate that preventable ADEs associated with injectable medications impact 1.2 million hospitalizations annually. Using a matched cohort analysis of healthcare claims as a basis for evaluating incremental costs, we estimate that inpatient preventable ADEs associated with injectable medications increase the annual US payer costs by $2.7 billion to $5.1 billion, averaging $600,000 in extra costs per hospital. Across categories of injectable drugs, insulin had the highest risk per administration for a preventable ADE, although errors in the higher-volume categories of anti-infective, narcotic/analgesic, anticoagulant/thrombolytic and anxiolytic/sedative injectable medications harmed more patients. Our analysis of liability claims estimates that MPL associated with injectable medications totals $300 million to $610 million annually, with an average cost of $72,000 per US hospital. Conclusion: The incremental healthcare and MPL costs of preventable ADEs resulting from inpatient injectable medications are substantial. The data in this study strongly support the clinical and business cases of investing in efforts to prevent errors related to injectable medications.

Stakeholder Perspective, page 421

Am Health Drug Benefits. 2012;5(7):413-422 www.AHDBonline.com Disclosures are at end of text

Ms Lahue is Vice President, Health Economics and Outcomes Research, Becton, Dickinson and Company, Franklin Lakes, NJ; Mr Pyenson is Principal, Consulting Actuary, Mr Iwasaki is Consulting Actuary, Dr Blumen is Principal, Senior Health Consultant, and Ms Forray is Consulting Actuary at Milliman Inc, New York, NY; and Dr Rothschild is Associate Professor of Medicine, Division of General Medicine, Brigham and Womenâ&#x20AC;&#x2122;s Hospital, Harvard Medical School, Boston, MA.

Vol 5, No 7

November/December 2012

www.AHDBonline.com

American Health & Drug Benefits

413

CLINICAL

reventable medication errors have emerged as a prominent cost and quality issue in the United States, and are estimated to impact more than 7 million patients, contribute to 7000 deaths, and cost almost $21 billion in direct medical costs across all care settings annually.1,2 Adverse drug events (ADEs) are harms that result from medication use; when these harms result from a medication error, they are known as “preventable ADEs.”3 The inpatient hospital setting is particularly resource-intensive in terms of care delivered and exposure to potential harms and errors.4,5 In 2007, the Institute of Medicine (IOM) estimated that 1 medication error occurred per patient per day in hospital care.4 In 2008, the US Department of Health and Human Services (HHS) estimated that approximately 1 of every 7 (13.5%) hospitalized Medicare patients experienced permanent harm from a medical error, and that 37% of these inpatient injuries were associated with medications.5 In addition, the study investigators estimated that 50% of these ADEs were preventable.5 The majority of hospitalized patients receive medications, which means that a high volume of doses are prescribed and are administered daily in the inpatient setting. A study in a 735-bed academic medical center estimated that approximately 16,000 medication doses were administered daily.6 This study and others report that up to 1 of 5 medication doses are associated with an error, and that between 3% and 7% of these errors are potentially harmful to patients.6,7 Furthermore, many of the medications used in the inpatient setting are delivered by injectable routes; these injectable medications have among the highest risk for error and the most severe harms.8 In a study of inpatient ADEs, including life-threatening ADEs, 50% of the medications that were implicated were injectable, including antihypertensives, insulin, and anticoagulants.9 Similarly, studies in the inpatient intensive care unit setting, where medications delivered by infusion are common, have reported that a patient’s risk for a medication error is approximately 10%, with 1 in 100 errors causing harm that requires life-saving treatment.8,10 In addition to the clinical harms caused by preventable ADEs, healthcare stakeholders incur the economic consequences as well. When a patient experiences a preventable ADE, there may be direct medical costs to payers, such as an extended inpatient stay, use of additional medications, and physician visits in an outpatient setting to restore the patient’s health. There are also indirect costs, which may include missed work, reduced quality of life, and disability for the patient, as well as possible uncompensated expenses for the healthcare provider. In a 1997 study, preventable ADEs were estimated to add $4685 in adjusted, postevent costs to an inpatient hospi-

414

American Health & Drug Benefits

KEY POINTS ➤

➤

Half of all adverse drug events (ADEs) are a result of medication errors and are therefore preventable. Injectable medications are among those at highest risk for error and can be associated with life-threatening events. This is the first analysis of the national burden of medication errors associated with inpatient injectable medications. The results show that preventable ADEs associated with injectable medications impact more than 1 million patients in the inpatient setting. Injectable-related preventable ADEs cause an increase of $2.7 billion to $5.1 billion in annual costs to US healthcare payers, with an average of $600,000 in extra annual cost per hospital. Furthermore, the analysis of liability claims shows a cost burden of $300 million to $610 million annually in medical professional liability, with an average cost of $72,000 per hospital. Reducing injectable medication errors and the associated preventable ADEs can improve quality of care for patients and reduce unnecessary cost for payers, hospitals, and physicians. The study’s broad approach to costs, including the 4 months after discharge and medical professional liability costs, is aligned with healthcare reform initiatives in the United States, where payers are introducing new payment structures that consider patient outcomes beyond the inpatient stay.

talization, amounting to an additional $2.8 million in annual costs per hospital.11 Citing articles by Bates and colleagues and Classen and colleagues, the IOM estimates that preventable ADEs affect up to 450,000 hospitalized patients and add $3.5 billion in extra costs to hospitals annually.4,11,12 Lawsuits and administrative actions related to preventable ADEs also increase costs for healthcare stakeholders. Provider costs related to medical professional liability (MPL), once called “medical malpractice,” are substantial. A previous study of MPL claims estimated that 73% of ADE-related cases were preventable; although legal defense costs were similar for inpatient and outpatient ADEs, the legal settlement costs were greatest for inpatient ADEs, which averaged $376,500 per MPL case.13 We conducted a comprehensive analysis of US payer and MPL costs for preventable ADEs related to injectable medications in the inpatient setting. We chose to focus this study on preventable ADEs resulting from injectable medications for several reasons, includ-

www.AHDBonline.com

November/December 2012

Vol 5, No 7

Preventable ADEs Associated with Inpatient Injectable Medications

ing their frequent use, their high risk for error, and their potential for targeted prevention strategies in the inpatient setting.14

Methods In this study we used a healthcare payer perspective to analyze the probability of preventable ADEs and associated medical costs related to inpatient injectable medications and projected the national number of ADEs and their costs. In addition, this analysis took an MPL insurer perspective in analyzing medication-related facility and professional insurance claims to generate national-level MPL costs related to preventable ADEs. Definitions The definitions of the terms that are used in this study are listed in Table 1. Data Sources and Methodology The formulas for each calculation and further information on each source that was used are detailed in the Appendix (available at www.AHDBonline.com). A medication error reporting system database with records noting each type of medication, standardized categories of error and clinical consequences, and setting of care was used to determine the distribution of ADEs for each injectable medication (Quantros MedMarx 2009-2011). The Premier National Database 2010-2011, which contains discharge data from 160 hospitals with detailed medication orders per patient record, was used to examine the frequency of each type of injection per discharge, by diagnosis-related group (DRG). Payer administrative claims databases (Medicare 5% Analytic Sample 2009-2010 and Thomson Reuters MarketScan, 2009-2010), which include all sites of service, were used to select ADE cases and controls to calculate the incremental healthcare costs for preventable ADEs. The distribution of hospitalizations by payer and by DRG was derived from 2010 data, which were comprised of 17 state-level inpatient claims sets (from Arizona, California, Florida, Illinois, Iowa, Maryland, Massachusetts, New Jersey, New York, Oklahoma, Rhode Island, Texas, Utah, Vermont, Virginia, Washington, and Wisconsin), and were extrapolated to match the US total number of annual hospitalizations of approximately 37 million. Five sources were used to estimate relevant MPL costs for inpatient preventable ADEs, including (1) a national database of closed MPL claims for 1990-2011 (National Practitioner Data Bank, HHS); (2) 2007-2011 premium rate filings for publicly available MPL state-level insurance (8 states, including California, Florida, Louisiana, Massachusetts, North Carolina, Ohio, Pennsylvania,

Vol 5, No 7

November/December 2012

Table 1 Adverse Drug Events Definitions Used in This Study Term

Definition

ADE

An injury due to medication; not all are due to errors or are preventable. For example, there may be no warning that a patient will have an allergic reaction to a medication

Medication error

Any error that occurs during the medication use process

Preventable ADE

When an ADE coexists with a medication error, it is considered a preventable ADE

Medical professional liability

Formerly called medical malpractice, medical professional liability costs include claims, administrative costs, insurer profit, and legal fees. The MPL of hospitals and professionals (eg, physicians) are often administered separately

ADE indicates adverse drug event.

and Vermont); (3) the 2010 American Hospital Association hospital-level survey on operational and financial statistics; (4) an industry survey of rate relativities for 201015; and (5) a detailed hospital and professional MPL closed claims database for 1994-2009 (Florida’s Closed Claim Database). The Florida Closed Claim Database is the only publicly available database that allows separation of hospital and physician liabilities for inpatient cases associated with medications.

Cost Analysis: Injectable Medication ADEs and Matched Controls Matched cohorts of inpatient cases and controls were analyzed to calculate the incremental costs resulting from an injectable ADE. For the cases, patients with International Classification of Diseases, Ninth Revision (ICD-9) diagnosis codes identifying likely inpatient medication errors—such as a wrong dose or improper administration (ICD-9 codes E850.xx-E853.xx) and poisoning from an overdose or a wrong substance that was given or taken in error (ICD-9 codes 960.xx-979.xx)— were selected from the 2 sets of payer administrative claims. To ensure that our cases included only patients with a preventable ADE occurring during the admission, we selected surgical DRG cases with error codes, because surgical admissions would unlikely be caused by an outpatient preventable ADE. The detailed claims for each ADE case, including the full list of patient-level diag-

www.AHDBonline.com

American Health & Drug Benefits

415

CLINICAL

Probability of Preventable Adverse Drug Events per Injectable Drug Administered and Administrations Table 2 of Injectable Drug per Admission Probability of preventable ADE per administration of injectable medicationa, % (95% CI)

Injectable drugs administered per admissionb, N (95% CI)

Total admissions receiving injectable medication, in millions, N (95% CI)

Insulin

1.16 (0.43-1.89)

0.74 (0.74-0.74)

6.7 (6.7-6.7)

Cardiovascular

0.50 (0.19-0.82)

0.51 (0.51-0.51)

5.3 (5.3-5.3)

Narcotic/analgesic

0.33 (0.12-0.53)

2.29 (2.28-2.29)

15.4 (15.4-15.5)

Anticoagulant/thrombolytic

0.26 (0.10-0.43)

1.92 (1.92-1.92)

14.2 (14.2-14.2)

Electrolytes/minerals

0.25 (0.09-0.40)

0.85 (0.84-0.85)

5.9 (5.9-5.9)

Anxiolytic/sedative

0.22 (0.08-0.35)

0.63 (0.63-0.63)

12.0 (12.0-12.0)

Anti-infective

0.15 (0.06-0.25)

2.93 (2.93-2.93)

18.5 (18.5-18.5)

Other

0.11 (0.04-0.19)

5.53 (5.52-5.53)

25.7 (25.7-25.7)

Mean

0.25 (0.09-0.40)

15.39 (15.38-15.40)

31.4 (31.4-31.4)

Injectable drug group

Estimated from Premier Database and Quantros MedMarx. Weighted average of all hospitalizations, by Medicare severity diagnosis-related group in the United States. ADEs indicates adverse drug events; CI, confidence interval. b

noses and procedures coded during the admission, were independently reviewed by 2 physicians to confirm the reasonability of the ADE assignment. The same databases were used to identify the controls, who were matched by discharge DRG and by preadmission patient costs that were within 3% of ADE case costs. The incremental cost of a preventable ADE was calculated as the cost difference between cases and matched controls within 4 months of the index hospitalization, inclusive of physician services during the hospitalization and any postdischarge care. This matched cohort analysis was applied for Medicare and for commercial patients separately; per-patient Medicaid costs were estimated as 80% of Medicare costs. All costs were inflated to 2013 dollars using the annual trend rate of 5% for commercial insurers or 4% for other insurers.

Medical Professional Liability Analysis The portion of MPL attributable to inpatient medication errors was developed from the National Practitioner Data Bank data, using cases associated with nurses as a proxy for hospital inpatient site of service, which were not available. We applied this same attributable portion to estimate the per-bed inpatient medication MPL costs, using premium rate documentation that was filed by state-level MPL insurers. Regional MPL estimates were summed to calculate the national figure. The estimated MPL cost includes the facility inpatient (ie, the hospi-

416

American Health & Drug Benefits

talâ&#x20AC;&#x2122;s liability) as well as professional (ie, physician and other clinician liability) costs.

Statistical Considerations Our analysis applied findings from the HHS study that the probability of a preventable medication error during an inpatient stay was 10.07% and that 50% of ADEs were preventable. The preventable ADEs associated with injectable medications were modeled as 87% (from injectable proportions of ADEs reported in the medication errors database) of the estimated preventable ADE total admissions, payer costs, and MPL costs. The means and the 95% confidence intervals were calculated on each result. A sensitivity analysis was conducted by varying the probability of an inpatient medication error per admission and the portion of such errors that were preventable to illustrate the variability of results. The year-to-year statistical variation in the national closed claim cost during the period between 2004 and 2011 was used to generate an estimated range of national MPL costs resulting from preventable ADEs. Results The mean probability for a preventable ADE per administration of an injectable medication was 0.25% (Table 2). Of the injectable medications, insulin had the highest probability for a preventable ADE (1.16%), followed by cardiovascular medications (0.5%), narcotic/

www.AHDBonline.com

November/December 2012

Vol 5, No 7

Preventable ADEs Associated with Inpatient Injectable Medications

Table 3 Diagnosis-Related Group Families with Greatest Contributions to Preventable Adverse Drug Events

Description (MS DRGs included in each family)

Proportion of preventable ADEs resulting from injectable medication, %

Probability of a preventable ADE resulting from injectable medication Proportion of administered at hospitalizations, % the hospital, %

Septicemia or severe sepsis without mechanical ventilation 96+ hrs (871-872)

3.75

1.94

6.3

Major small and large bowel procedures (329-331)

2.71

0.86

10.4

Extracorporeal membrane oxygenation or tracheostomy with mechanical ventilation 96+ hrs (003)

2.57

0.14

59.8

Heart failure and shock (291-293)

2.56

2.18

3.9

Respiratory system diagnosis with ventilator support (207-208)

2.44

0.60

13.3

Major joint replacement or reattachment of lower extremity (469-470)

2.42

2.40

3.3

Cesarean section (765-766)

2.27

3.57

2.1

Coronary bypass (231-236)

2.15

0.41

17.2

Simple pneumonia and pleurisy (193-195)

2.04

2.25

3.0

Esophagitis, gastroenteritis, and miscellaneous digestive disorders (391-392)

1.84

2.52

2.4

Chronic obstructive pulmonary disease (190-192)

1.73

1.85

3.1

Cardiac valve and other major cardiothoracic procedure (216-221)

1.63

0.28

18.9

Infectious and parasitic diseases with operating room procedure (853-855)

1.59

0.29

18.1

Percutaneous cardiovascular procedure with stent (246-249)

1.51

1.26

3.9

Rehabilitation (945-946)

1.48

1.32

3.7

Diabetes (637-639)

1.48

0.91

5.3

Cellulitis (602-603)

1.47

3.3

Tracheostomy with mechanical ventilation 96+ hrs (004)

1.46

0.12

38.8

Vaginal delivery (774-775)

1.45

6.75

0.7

Renal failure (682-684)

1.41

1.15

4.0

ADEs indicates adverse drug events; MS DRGs, Medicare severity diagnosis-related groups.

Vol 5, No 7

November/December 2012

www.AHDBonline.com

American Health & Drug Benefits

417

CLINICAL

Estimated National Incidence and Incremental Cost Table 4 of Preventable Adverse Drug Events Related to Injectable Medications Annual number of inpatient hospitalizations in the United Statesa

37 million

Probability of avoidable ADE from injectable medications per hospitalization

3.3% (95% CI, 2.7%-3.9%)

Annual number of hospitalizations with preventable ADEs resulting from injectable medications in the United States

1.2 million (95% CI, 1.0 million1.4 million)

Incremental cost of preventable ADE from injectable medications per hospitalizationb

$3100 (95% CI, $2700-$3600)

Annual incremental cost of preventable ADEs resulting from injectable medicationsb

$3.8 billion (95% CI, $2.7 billion$5.1 billion)

Average annual inpatient preventable ADE cost resulting from injectable medication per hospital (6268 hospitals)b

$600,000

American Hospital Association. AHA Hospital Statistics, 2012 edition. 2012. www.aha.org/research/rc/stat-studies/fast-facts. shtml. Accessed September 17, 2012. b Costs are expressed in projected 2013 dollars. ADEs indicates adverse drug events; CI, confidence interval.

Figure Costs of Preventable ADEs from Inpatient Injectable Medications, by Payer

Medicaid 6%

Other 10%

Medicare 27%

Commercial health plan 57%

ADEs indicates adverse drug events.

418

American Health & Drug Benefits

analgesic medications (0.33%), and anticoagulant/thrombolytic medications (0.26%). The mean number of administrations of injectable medications per inpatient hospitalization was 15.39, with anti-infectives (2.93) and narcotic/analgesics (2.29) as the leading classes of administered doses. The probability of a preventable ADE from an injectable medication per patient varied by the type of hospitalization, with surgical DRGs carrying a 6.4% probability and medical DRGs carrying a 3.3% probability, whereas patients with other hospitalization types (eg, obstetrics, psychiatry) were significantly less likely to have a preventable ADE from an injectable medication. The 20 DRG families with the highest probability of preventable ADEs from injectable medications per hospitalization are listed in Table 3. A total of 303 preventable ADE cases and 37,513 control matches were identified for the incremental cost analysis. The national incidence and incremental cost of preventable ADEs from injectable medications is shown in Table 4. Preventable ADEs from the administration of injectable medications were estimated to occur in 1.2 million (95% confidence interval, 1.0 million-1.4 million) inpatient hospitalizations annually in the United States. The incremental cost of the preventable ADEs from injectable medications averaged $3100 per admission. The incremental annual cost for preventable ADEs resulting from injectable medications was estimated to be between $2.7 billion and $5.1 billion, which averages $600,000 of payer costs per hospital (Table 4). The annual payer cost burdens are illustrated in the Figure, which shows that the majority (57%) of costs for preventable ADEs is paid by commercial health plans. We estimate the 2013 MPL costs that are associated with inpatient injectable ADEs to be between $300 million and $610 million (Table 5). Although costs vary by region, type, and size of hospital, the total medication-related MPL costs average an annual $72,000 per US hospital.

Discussion Our study of inpatient injectable medications estimated that there are 1.2 million hospitalizations with preventable ADEs annually, contributing incremental direct medical costs of between $2.7 billion and $5.1 billion annually to US payers. This study is among the first to include the postdischarge costs in estimating the impact of inpatient preventable harms. In addition, we estimated that $300 million to $610 million is spent annually in medical liability for inpatient ADEs resulting from injectable medications (Table 5). Previous studies on medication errors have estimated a 0.06% risk for a preventable ADE per dose and have

www.AHDBonline.com

November/December 2012

Vol 5, No 7

Preventable ADEs Associated with Inpatient Injectable Medications

concluded that a hospital with 6 million annual doses could attribute 4000 preventable ADEs annually to medication administration errors.9 Our study showed that the average hospitalized patient is receiving more than 15 injections, with a probability of a preventable ADE of approximately 1 of every 400 injections (0.25%). A hospital with 10,000 injectable doses daily could expect 25 daily preventable ADEs, or more than 9000 preventable ADEs annually resulting from injectable medications.

Estimated Medical Professional Liability Costs Table 5 Associated with Inpatient Adverse Drug Events

Research Implications Our study builds on the 2010 HHS study that reported estimates of numbers and costs of inpatient preventable ADEs in Medicare beneficiaries.5 We expanded to a national estimate but narrowed the focus to injectable medications that may be targeted for prevention strategies. The current study is unique in its inclusion of allpayer national cost estimates encompassing nonfacility inpatient costs, such as physician costs, costs to payers other than traditional Medicare (ie, Medicare Advantage, commercial, and Medicaid), and postdischarge costs for 4 months. This broader vantage is aligned with healthcare reform initiatives in the United States, where payers are introducing new payment structures that consider patient outcomes beyond the inpatient stay. With preventable ADEs related to injectable medication errors adding $2.7 billion to $5.1 billion in extra costs, this issue is similar in magnitude to other foci for healthcare reform, such as reducing bedsore pressure ulcers and hospital-acquired infections. In a study on adverse event costs related to medical errors, Van Den Bos and colleagues documented that pressure ulcers and postoperative infections were the most common preventable events, accounting for approximately $3.27 billion each in annual costs.16 With increased awareness of the healthcare waste related to preventable medical errors, a number of initiatives have been introduced to address this issue, including streamlined provider communication initiatives, payment reform (eg, Medicare’s “never event”) initiative, and reduced payments for hospital-acquired infections.16 Injectable medication errors may be similarly targeted with prevention strategies. The inpatient medication use process includes several steps—prescribing, transcribing/documentation, dispensing (including medication preparation), administering, and monitoring. Errors in the early steps of the medication system can be prevented by verifications done by pharmacists and nurses. Some interventions developed through systematic approaches to error prevention that dramatically reduce the frequency of medication errors and that also impact injectable medications include computerized physician order entry with decision support,17 automated medica-

Vol 5, No 7

November/December 2012

Number of inpatient hospital licensed bedsa

931,000

Inpatient hospital annual MPL cost per licensed bedb

$3700

Total annual inpatient hospital MPL cost

$3.48 billion

Estimated percentage of hospital MPL cost attributable to ADEs resulting from injectable medicationsc

6.5%

Ratio of physician plus facility MPL cost to facility MPL costd

Annual inpatient ADE MPL cost resulting from injectable medication

$450 million (95% CI, $300 million$610 million)

Average annual inpatient ADE MPL cost associated with injectable medication per hospital (total of 6268 hospitals)b

$72,000

a American Hospital Association. AHA Hospital Statistics, 2012 edition. 2012. www.aha.org/research/rc/stat-studies/fast-facts. shtml. Accessed September 17, 2012. Adjusted to remove nonacute beds. b Authors’ analysis of hospital MPL premium rates. c Authors’ analysis of National Practitioner Data Bank. d Authors’ analysis of the Florida Department of Insurance Closed Claim Database. ADEs indicates adverse drug events; CI, confidence interval; MPL, medical professional liability.

tion dispensing systems,18 and bar-coded medication administration.6 However, a large proportion of medication errors (56%-62%) occur during the administration step of the medication use process.3,19

With preventable ADEs related to injectable medication errors adding $2.7 billion to $5.1 billion in extra costs, this issue is similar in magnitude to other foci for healthcare reform. Errors in administration are the most difficult to detect, because they occur in the last step of the medication use process, usually by the bedside nurse, and often without additional oversight.20 Several technologies specifically targeting the administration of injectable medications include smart infusion

www.AHDBonline.com

American Health & Drug Benefits

419

CLINICAL

pumps,21 prefilled syringes,8 and in-line sensors to ensure the correct medication and the correct concentration.22 Our data analysis covers both high-risk clinical conditions and medications. For example, although 2% of national hospitalizations are for septicemia/sepsis, these DRGs make up almost 4% of the preventable ADE cases; 6% of septicemia/sepsis hospitalizations are predicted to experience a harmful medication error (Table 3). The recognition that the DRG families in Table 3 are associated with hospitalizations that are not only intense utilizers of clinical resources but are also at increased risk for medication errors may sharpen hospitals’ focus on improvement opportunities.

Our results are consistent with other estimates, and our intermediate results for particular types of patients or injectable medications could be validated through intensive studies, such as direct observations or patient chart audits. Our results are consistent with other estimates, and our intermediate results for particular types of patients or injectable medications could be validated through intensive studies, such as direct observations or patient chart audits. Sensitivity analyses confirm our linear projections, with each 1% increase or decrease in either the probability of an inpatient medication error per admission or the portion of such errors that were preventable resulting in an increase or a decrease in the estimated costs by 1%. For example, based on the HHS study,5 we assumed that 50% of ADEs were preventable. If this assumption were reduced to 40%, then the average annual incremental cost for preventable ADEs resulting from injectable medication would be reduced by 20%; our midpoint estimate of $3.8 billion would thereby be reduced to $3.0 billion.

Limitations In this study we relied on multiple administrative claims data with known inherent limitations, such as coding inaccuracies and lack of clinical detail.23 In addition, although it is possible that patients could appear in multiple databases that we used, this would not cause double counting, because each source was used to develop distinct assumptions. Our analysis also relied on several assumptions from previous work on preventable ADEs.5 We focused on ADEs from injectable medications, and we mapped reported errors for injectable medications to Medicare

420

American Health & Drug Benefits

severity DRG, but we estimated costs from ADEs caused by all medications and modeled the attribution to injectable medications. We did not include in our calculation patients in observation beds or patients in the emergency department who were not admitted to the hospital. A potential limitation related to our use of a medication error reporting database is whether there is bias in the reporting of adverse events, as would be the case if reported events tended to be more serious. However, in healthcare settings, there is strong evidence that event severity has little relationship to whether the event is reported.24 This assumption invokes the “causal continuum hypothesis” used in other industries, such as in aviation, which suggests that the characteristics and circumstances of reported errors are similar to those of unreported errors.25 Finally, our MPL estimate relied on historical financial information sources and sources that were not assembled for the purpose of quantifying costs associated with MPL. We assumed that all MPL costs were associated with preventable ADEs, which ignores the possibility that some MPL awards may not be associated with preventable errors (or even unpreventable errors). Certainly, for both healthcare and MPL costs, other methodologies and additional data sources could produce different estimates.

Conclusions The approach taken in this study may serve as a basis for further studies of medication errors for injectables. The results of this study highlight potential areas for future prospective research on high-frequency and high-risk medications, such as insulin, narcotics, and anxiolytics. Inpatient preventable ADEs are often caused by injectable medications, potentially affecting more than 1 million US patient hospitalizations and adding $2.7 billion to $5.1 billion of extra direct medical costs to payers and hundreds of millions of dollars in extra liability costs to hospitals and physicians, annually. Reducing the risk for injectable medication errors is a clear target area for improving acute patient quality of care and for reducing unnecessary costs in the US healthcare system. ■ Study Funding Funding for this study was provided by Becton, Dickinson and Company. Author Disclosure Statement Ms Lahue is an employee of and Mr Pyenson, Mr Iwasaki, Dr Blumen, Ms Forray, and Dr Rothschild are consultants to Becton, Dickinson and Company.

References 1. New England Health Institute. Preventing medication errors: a $21 billion opportunity. www.nehi.net/bendthecurve/sup/documents/Medication_Errors_%20Brief.pdf.

www.AHDBonline.com

November/December 2012

Vol 5, No 7

Preventable ADEs Associated with Inpatient Injectable Medications

Accessed September 17, 2012. 2. Committee on Quality of Health Care in America, Institute of Medicine. Errors in health care: a leading cause of death and injury. In: Kohn LT, Corrigan JM, Donaldson MS, eds. To Err Is Human: Building a Safer Health System. Washington, DC: National Academies Press; 2000. 3. Leape LL, Bates DW, Cullen DJ, et al. Systems analysis of adverse drug events. ADE Prevention Study Group. JAMA. 1995;274:35-43. 4. Committee on Identifying and Preventing Medication Errors, Institute of Medicine. Medication errors: incidence and cost. In: Preventing Medication Errors: Quality Chasm Series. Washington, DC: National Academies Press; 2007. 5. Levinson DR. Adverse events in hospitals: national incidence among Medicare beneficiaries. US Dept of Health and Human Services, Office of Inspector General. November 2010. OEI-06-09-00090. https://oig.hhs.gov/oei/reports/oei-06-0900090.pdf. Accessed November 15, 2012. 6. Poon EG, Keohane CA, Yoon CS, et al. Effect of bar-code technology on the safety of medication administration. N Engl J Med. 2010;362:1698-1707. 7. Barker KN, Flynn EA, Pepper GA, et al. Medication errors observed in 36 health care facilities. Arch Intern Med. 2002;162:1897-1903. 8. Adapa RM, Mani V, Murray LJ, et al. Errors during the preparation of drug infusions: a randomized controlled trial. Br J Anaesth. 2012;109:729-734. 9. Kale A, Keohane CA, Maviglia S, et al. Adverse drug events caused by serious medication administration errors. BMJ Qual Saf. 2012;21:933-938. 10. Osmon S, Harris CB, Dunagan WC, et al. Reporting of medical errors: an intensive care unit experience. Crit Care Med. 2004;32:727-733. 11. Bates DW, Spell N, Cullen DJ, et al. The costs of adverse drug events in hospitalized patients. Adverse Drug Events Prevention Study Group. JAMA. 1997;277:307-311. 12. Classen DC, Pestotnik SL, Evans RS, Burke JP. Computerized surveillance of adverse drug events in hospital patients. JAMA. 1991;266:2847-2851. 13. Rothschild JM, Federico FA, Gandhi TK, et al. Analysis of medication-related malpractice claims: causes, preventability, and costs. Arch Intern Med. 2002;162: 2414-2420.

14. Sanborn MD, Moody ML, Harder KA, et al. Second Consensus Development Conference on the Safety of Intravenous Drug Delivery Systems—2008. Am J Health Syst Pharm. 2009;66:185-192. 15. Medical Liability Monitor annual rate survey. Medical Liability Monitor. 2011; 36:7-43. 16. Van Den Bos J, Rustagi K, Gray T, et al. The $17.1 billion problem: the annual cost of measurable medical errors. Health Aff (Millwood). 2011;30:596-603. 17. Bates DW, Teich JM, Lee J, et al. The impact of computerized physician order entry on medication error prevention. J Am Med Inform Assoc. 1999;6:313-321. 18. Chapuis C, Roustit M, Bal G, et al. Automated drug dispensing system reduces medication errors in an intensive care setting. Crit Care Med. 2010;38:2275-2281. 19. Beyea SC, Hicks RW, Becker SC. Medication errors in the OR—a secondary analysis of Medmarx. AORN J. 2003;77:122,125-129,132-134. 20. Bates DW, Cullen DJ, Laird N, et al. Incidence of adverse drug events and potential adverse drug events. Implications for prevention. ADE Prevention Study Group. JAMA. 1995;274:29-34. 21. Rothschild JM, Keohane CA, Cook EF, et al. A controlled trial of smart infusion pumps to improve medication safety in critically ill patients. Crit Care Med. 2005;33: 533-540. 22. SEA Medical Systems. Technology. www.seamedical.com/?pg=products. Accessed September 17, 2012. 23. Meddings JA, Reichert H, Rogers MA, et al. Effect of nonpayment for hospitalacquired, catheter-associated urinary tract infection: a statewide analysis. Ann Intern Med. 2012;157:305-312. 24. Levinson DR. Hospital incident reporting systems do not capture most patient harm. US Dept of Health and Human Services, Office of Inspector General. January 2012. Report No OEI-06-09-00091. http://psnet.ahrq.gov/resource.aspx? resourceID=23842. Accessed November 24, 2012. 25. Cure L, Zayas-Castro J, Fabri P. Clustering-based methodology for analyzing nearmiss reports and identifying risks in healthcare delivery. J Biomed Inform. 2011;44: 738-748.

STAKEHOLDER PERSPECTIVE

Injectable Sticker Shock: A Call to Action By Jaan sidorov, MD, Mhsa Consultant, Sidorov Health Solutions, and Chair, Board of Directors, NORCAL Mutual Insurance Company, a professional medical liability carrier, Harrisburg, PA

As the US healthcare system continues its relentless march toward consuming 20% of the nation’s gross domestic product, providers, policymakers, regulators, and other stakeholders are becoming acutely aware of errors. In response, new jargon, such as “never events,” “hospital- acquired conditions,” “avoidable admissions,” and “near misses,” has sprung up in research journals, as well as on the evening news, in newspaper front pages, and in online media. Front and center in this new lexicon is “preventable medication errors.” Although experts may quibble over the precise definitions of “preventable” and “error,” there is no underestimating the growing national impatience with the seeming inability of the healthcare system to deliver interventions to a patient with the same ease of use and accuracy level as an automated teller machine (ATM).1 POLICYMAKERS/PAYERS: In this issue of American Health & Drug Benefits, Betsy Lahue and her impressive,

Vol 5, No 7

November/December 2012

multidisciplinary team of economists, actuaries, and clinicians have helped us better understand the extent of the problem, by illuminating the clinical and economic burdens associated with injectable medication errors.2 By drawing on multiple databases, what they have uncovered is truly staggering: with 1 of every 400 inpatient medication injections leading to an unnecessary error, a typical hospital may be dealing with as many as 25 adverse events daily. Although most of these adverse events can be safely managed by the superbly trained providers who staff our healthcare system, the nation’s financial toll of up to $5 billion plus additional hundreds of millions of dollars in professional liability expense is more evidence of lax attention to patient safety and the drag on our nation’s economy. The authors are the first to point out that their research methodology may not be perfect. Health insurance claims data are notoriously inadequate indicators of actual misContinued

www.AHDBonline.com

American Health & Drug Benefits

421

CLINICAL

STAKEHOLDER PERSPECTIVE (Continued) takes, mapping events across multiple databases is fraught with error, some of the key information was self-reported, and multiple assumptions were used to derive the observed associations. These shortcomings could be used by critics to undermine this study’s ultimate conclusions, but the approach used by Lahue and colleagues—outside of prospective, multicenter, time-consuming, and unaffordable observational studies—is the best we have. Waiting for a more perfect methodology to catch up with this critically important scientific question is not only irresponsible, but is unlikely to be countenanced by our patients, their families, and their elected representatives. HOSPITALS/RESEARCHERS: The good news is that this information is actionable. Now that this challenge has been identified, hospital leaders and their governing boards can use this study’s findings to craft new safety programs and contrast the benchmark data with their own evolving internal metrics. Furthermore, other researchers will be able to draw on Lahue and colleagues’ methodology to better understand progress against this national baseline. As quality-linked payment approaches continue to grow in number and in sophistication, avoidable errors associated with injectable medications will be able to become an important focus that will ultimately translate into lives saved and reduced healthcare costs.

MEDICAL LIABILITY CARRIERS: In addition, our nation’s medical professional liability (MPL) carriers are unlikely to remain idle on issues like this. As health systems become increasingly integrated and complex, their liability needs are shifting considerably. The usual approach to risk transfer will be replaced by health system–MPL carrier relationships that provide liability coverage, while simultaneously minimizing threats to patient safety. This includes tailored insurance arrangements, innovative risk management initiatives, shared databases, early warning systems, and approaches to prompt resolution of claims. Hospitals grappling with medication errors ignore this resource at their peril. PATIENTS: Will we ever get to the injectable medication error rate that rivals the perfection of an ATM? Although that may be a reach, an error rate of 1 in 400 transactions would be considered unacceptable. Knowing that that is the standard, and armed with a good idea of the extent of the problem, it is time to get to work. Our patients expect nothing less. 1. Conaboy C. BostonGlobe.com. Institute of Medicine report points to backwards structure of health care industry. September 6, 2012. www.boston.com/whitecoatnotes/2012/09/06/institute-medicine-report-points-backwards-structure-health-careindustry/28q3DDsmvUTw0286iPX3jO/story.html. Accessed December 5, 2012. 2. Lahue BJ, Pyenson B, Iwasaki K, et al. National burden of preventable adverse drug events associated with inpatient injectable medications: healthcare and medical professional liability costs. Am Health Drug Benefits. 2012;5(7):413-422.

Request your SUBSCRIPTION to AmeRicAn HeAltH & DRug Benefits

❑ ❑

YES! I would like to receive American Health & Drug Benefits ® as well as related educational supplements. NO. Please discontinue my subscription.

Signature (Required)

Specialty

Date (Required)

Address

Name

City/State/Zip

Company

E-mail

Title

Phone Please provide all information indicated, including date and signature. INCOMPLETE CARDS WILL NOT BE PROCESSED.

Fax to: 732.992.1881

422

American Health & Drug Benefits

www.AHDBonline.com

November/December 2012

Vol 5, No 7

EDITORIAL

The Use of Medicines in the United States: A Detailed Review David B. Nash, MD, MBA, Editor-in-Chief, American Health & Drug Benefits Jefferson School of Population Health, Philadelphia, PA

udos to our colleagues at the IMS Institute for Healthcare Informatics in Parsippany, NJ, for their 2012 report on the use of medicines in the United States in 2011.1 Developed as a public service by the IMS Institute, without sponsorship from the pharmaceutical industry or from the government, this report is well worth a detailed review. I would like to share with you some of the highlights, along with some background information for each of the key conclusions of the report. According to Michael Kleinrock, Director of Research Development at the IMS Institute, breakthrough therapies, innovation in disease treatments, and changes in the consumption of medicines transformed the US healthcare market in 2011.1 At the core of this interesting report are 5 takeaway messages1: 1. Major transformations have become available in 2011 in the treatment options for diseases involving a few thousand to several million patients 2. Overall per-capita drug utilization declined in 2011 as office visits and non–emergency department hospital admissions dropped and older patients reduced the use of retail drugs 3. Patients with health insurance spent $49 billion out of pocket (OOP) for prescription drugs, a decrease of $1.8 billion from OOP spending in 2010; this decline “was largely related to the introduction of subsidies for Medicare Part D beneficiaries in the ‘donut hole’ ” 4. Total healthcare system spending on drugs in 2011 reached $320 billion, increasing on a per-capita basis from 2010 by only 0.5%; the declining use of branded drugs and increased availability of generic agents offset the increase in drug prices and spending on new and innovative medicines 5. Nearly one third of the total drug spending was concentrated in just 6 clinical areas—cancer, asthma, chronic obstructive pulmonary disease (COPD), dyslipidemia, diabetes, and mental health. It is worth expanding a little on each of these 5 takeaway messages. Regarding transformation in treatment, 2011 was characterized by the introduction of 34 new molecular entities (NMEs), the greatest number of NMEs launched in the past 10 years. Key breakthrough therapies became available, for the first time, to treat several types of cancer, multiple sclerosis, hepatitis C, and cardiovascular conditions.1 Regarding the utilization of healthcare and medicines, it is fascinating that per-capita retail prescription use, which averaged 11.33 prescriptions per person in 2011, has gone

Vol 5, No 7

November/December 2012

down from 11.46 prescriptions per person in 2010. Per-capita prescription use declined in 41 states and fell by more than 3% in 10 states. Patients with health insurance spent $1.8 billion less OOP for medicines in 2011 compared with 2010, as the average copay declined, especially for seniors participating in the Medicare Part D program. The majority of OOP costs were incurred by patients with commercial health insurance. The average copay for nearly 75% of all prescriptions was $10 or less, but as much as $40 on average for brand-name drugs was covered by commercial health plans.1 Total spending on medicines, on a real per-capita basis, increased by 0.5% as declining use of branded drugs and greater availability of lower-cost generics offset the price increases of and higher spending on new medicines. Generics reached 80% of dispensed prescriptions; spending in this segment grew by $5.6 billion in 2011. Finally, overall spending on medicines continued to be concentrated on traditional, small-molecule tablets dispensed through retail pharmacies, even as growth in these segments was outpaced by biologics, specialty drugs, injectables, and institutional channels, which accounted for as much as 30% of total drug spending.1 Regarding the major therapeutic areas, spending on cancer drugs was $23.2 billion in 2011, up by 4.2%, as a result of innovative new targeted therapies, and this was offset by patent expirations. Spending for respiratory treatments reached $21 billion, up by $1.7 billion, with antiasthmatic products accounting for more than 50%. Overall, 7.4 million patients were regularly taking medicines for asthma or COPD. Nearly 20 million Americans regularly used lipidlowering medicines, up by 160,000 persons from 2010, while spending increased by $1.4 billion. Finally, 11 million patients were treated with antidiabetes medicines; spending in this sector grew by $1.9 billion, driven by insulins and by further uptake of new-generation therapies.1 In summary, this comprehensive annual report (from a very reliable source) provides a great deal of new information, organized in a user-friendly manner. To me, one message stands out, namely, that the pharmaceutical industry represents an impressive innovative engine, despite the ongoing political drama surrounding the question of how health insurance reform will actually evolve. As always, I am interested in your views and your comments. You can reach me by e-mail at david.nash@ jefferson. edu or via the journal at editorial@engagehc.com. Reference 1. IMS Institute for Healthcare Informatics. The use of medicines in the United States: review of 2011. April 2012. www.imshealth.com/ims/Global/Content/Insights/IMS%20Institute%20for%20 Healthcare%20Informatics/IHII_Medicines_in_ U.S_Report_2011.pdf. Accessed April 11, 2012.

www.AHDBonline.com

American Health & Drug Benefits

423

BUSINESS

Original research

Current and Future Use of HEOR Data in Healthcare Decision-Making in the United States and in Emerging Markets anke-Peggy holtorf, PhD, MBa; Diana Brixner, rPh, PhD; Brandon Bellows, PharmD; abdulkadir Keskinaslan, MD, MBa, MPh; Joseph Dye, rPh, PhD; gary Oderda, PharmD, MPh

Diana Brixner

Stakeholder Perspective, page 438

Am Health Drug Benefits. 2012;5(7):428-438 www.AHDBonline.com Disclosures are at end of text

Background: Health economics and outcomes research (HEOR) is a growing field that provides important information for making healthcare coverage and access decisions. However, there is no standard process for incorporating HEOR into the decision-making process, and the current use of HEOR by healthcare payers remains unknown. Objectives: To examine how HEOR data are being used by healthcare payers, including managed care organizations today, and how the use of such data is expected to change in the future in relation to access and reimbursement decision-making. Methods: The Managed Care Survey (MCS) and the Pharmacy & Therapeutics (P&T) Committee Survey (PTS) were distributed to decision makers in the United States. A total of 72 managed care decision makers responded to the MCS and 30 P&T Committee members responded to the PTS from US healthcare organizations that cover from tens of thousands to millions of lives. The goal of these surveys was to understand the current use of HEOR data, perceived barriers and limitations in the use of HEOR, and the expectations for future use, and how these and other factors affect formulary decisions. An international perspective was gained by modifying the MCS based on feedback received at a European conference, and a pilot version was given to individuals in emerging markets across Asia, Latin America, and the Middle East and Africa. Results: The majority of US respondents to the MCS (74%; N = 53) and to the PTS (77%; N = 23) indicate that HEOR is currently being used in their decision-making process; but the majority of respondents to the MCS (66%; N = 48) also state that quality assessment is limited (quality assessment was not addressed in the PTS). In addition, the majority of respondents to the MCS (82%; N = 59) expect the use of HEOR to increase in the future. Safety and efficacy were reported in the PTS to be the most important factors in the P&T Committee decision-making process, followed by head-to-head comparisons, and cost. The current use of HEOR in Asia, Latin America, and the Middle East and Africa varied widely across respondents. Conclusion: This study provides an important benchmark of HEOR use in the United States before the implementation of healthcare reform. Between the years 2010 and 2011, HEOR data were used to varying extents across global regions, but their use in the future is likely to increase in relation to access and reimbursement decisions.

ith increasing access and utilization of healthcare, resources become more restricted, and prioritization in healthcare becomes unavoidable. Health economics and outcomes research (HEOR) is a discipline that is used to complement traditional clinical development information (ie, efficacy,

safety, quality) to guide decision makers regarding patient access to specific drugs and services. HEOR has advanced considerably in methodology and in quantity over the past several decades. HEOR can provide data to help healthcare payers determine if treatments work in the populations they serve, and how much of the

Dr Holtorf is Managing Director, Health Outcomes Strategies, Basel, Switzerland; Dr Brixner is Chair, Department of Pharmacotherapy, and Executive Director, Pharmacotherapy Outcomes Research Center, College of Pharmacy, University of Utah, Salt Lake City; Dr Bellows is Research Associate, Pharmacotherapy Outcomes Research Center, University of Utah, Salt Lake City; Dr Keskinaslan is Market Pricing Director, Novartis Pharma AG, Basel, Switzerland; Dr Dye is Clinical Research Consultant, Competitive Health Analytics, a Humana Company, Atlanta, GA; Dr Oderda is Professor and Director, Utah Medicaid Drug Regimen Review Center, and Director, Pharmacotherapy Outcomes Research Center, University of Utah, Salt Lake City.

428

American Health & Drug Benefits

www.AHDBonline.com

November/December 2012

Vol 5, No 7

Use of HEOR in Healthcare Decision-Making

drug or treatment cost should be reimbursed by the healthcare system. As a key part of the request for comparative effectiveness evidence, the increased use of HEOR data can be expected in future decision-making processes.1,2 In addition, a greater emphasis has recently been placed on positioning the patient at the center of healthcare decisions. Outcomes research plays an increasingly important role in this, because it can provide data on specific populations and treatment combinations that are used. Understanding how these data are used in decision-making in the United States and globally can direct future efforts in this area. Currently, several global reimbursement agencies formally ask for HEOR information for their standard assessment process, including the National Institute for Health and Clinical Excellence in the United Kingdom, some of the Spanish health technology assessment (HTA) agencies, the Korean Health Insurance Review Agency, and the Health Intervention and Technology Assessment Program in Thailand.3-5 However, healthcare payers in the United States do not currently have a standardized process for requesting or for using HEOR data. In the United States, HEOR data may come primarily from pharmaceutical companies via the Academy of Managed Care Pharmacy (AMCP) dossier format. Consequently, the pharmaceutical industry invests heavily in HEOR studies alongside clinical trials and continues to collect clinical, humanistic, and economic real-world data throughout the life cycle of a therapy.3 Discussions between healthcare payers and academic health economists suggest a need for this information by decision makers, but there is a lack of standardization regarding how such information is integrated into the current processes for drug (and other technology) evaluations.6,7 Therefore, it remains unclear how healthcare payers in the United States currently use HEOR, and whether the use of such evidence will change in the future. The objective of this article is to describe the current and expected future use of HEOR data by healthcare payers, and to examine how pharmaceutical drug and manufacturer attributes are used in the decision-making process. This article summarizes the results of 2 surveys administered to individuals in formulary and reimbursement decision-making positions in the United States, as well as an adaptation of one of these surveys and workshops performed in other parts of the world.

Methods In 2010, US payers and formulary decision makers were asked to complete the Managed Care Survey (MCS) to understand the degree of the current use of HEOR data,

Vol 5, No 7

November/December 2012

KEY POINTS ➤

➤

Health economics and outcomes research (HEOR) is used to complement traditional clinical development information in guiding healthcare coverage and access decisions for specific therapies. This study is based on 2 surveys designed to evaluate the use of HEOR data by managed care organizations and formulary decision makers who provide health insurance to between tens of thousands and millions of members. Among the 72 organizations responding to the Managed Care Survey, 73% use HEOR data regularly in their decision-making processes. Of the 30 decision makers responding to the Pharmacy & Therapeutics Committee Survey, 77% use HEOR as a standard element in their drug therapy review process. Based on these surveys, the majority of decision makers in the United States use HEOR evidence when making formulary and coverage decisions, and this use is expected to increase in the future. However, actual formulary drug placements show that the lowest-priced branded drugs are often not in a preferred tier, indicating that the role that cost plays in the formulary decision-making process is complicated and difficult to isolate from other factors (eg, rebates).

the barriers and limitations perceived, and the expectations toward the future use of these data. The MCS was an internet-based survey developed by the investigators and distributed personally or via the AMCP membership to pharmacy decision makers in pharmacy benefit management (PBM) organizations, health plans, managed care organizations, and Medicaid/Medicare in the United States in April 2010 through the AMCP membership. Around the same time, a second survey, the Pharmacy & Therapeutics (P&T) Committee Survey (PTS), was developed by the investigators sent to P&T Committees in the United States with the objective of exploring managed care pharmacists’ perceptions of factors affecting formulary decisions. The PTS was also an internet-based survey developed by the investigators and was distributed to 176 managed care pharmacists in January 2010. Partial results of the MCS were presented in a workshop at the 12th Biennial European Meeting of the Society for Medical Decision Making (SMDM), with the objective of discussing an extension of such a survey in the European environment.8 Findings from this workshop were subsequently used to adapt the MCS to emerging markets. The revised global MCS was then

www.AHDBonline.com

American Health & Drug Benefits

429

BUSINESS

Table 1 Hypothetical Formulary Drug Placement Scenarios: The Pharmacy & Therapeutics Committee Survey Drug and manufacturer attributes

Drug A: lipid-lowering drug manufactured by small pharmaceutical company

Drug B: lipid-lowering drug manufactured by a Fortune 100 pharmaceutical company

National market share

40%

Net ingredient cost

$50 monthly

$80 monthly

Contract

Flat rate: up to 20% rebates

Market share–driven contract: up to 10% rebates

Safety (relative to others in class)

Equal or better

Efficacy (relative to others in class)

Equal or better

Dosing (relative to others in class)

Equal or better

Drug and manufacturer attributes

Drug A: PPI drug manufactured by small pharmaceutical company

Drug B: PPI drug manufactured by a Fortune 100 pharmaceutical company

National market share

30%

Net ingredient cost

$75 monthly

$125 monthly

Contract

Flat rate: up to 20% rebates

Market share–driven contract: up to 20% rebates

Safety (relative to others in class)

Equal or better

Efficacy (relative to others in class)

Equal or better

Dosing (relative to others in class)

Equal or better

PPI indicates proton-pump inhibitor.

used at healthcare decision maker workshops in Latin America, Asia, and the Middle East and Africa.

Survey Descriptions The objective of the MCS was to understand how HEOR data are used in decision-making today, and how their use may change in the future. There were 9 demographic questions and 24 additional questions around the use of HEOR and decision priorities. Individuals were asked if they currently use HEOR in their organizations “always,” “often,” “sometimes,” “rarely,” or “never,” and if they expected to increase the use of HEOR in the future (ie, “definitely,” “probably,” “to a limited degree,” or did not). They were asked a single yes or no question as to whether measures were in place to ensure quality of the HEOR data. If an organization used quality measures, it was asked what standards were used. To determine the barriers to the use of HEOR data, the organizations were asked what would be required to increase the use of HEOR data in decision-making. Participants were also asked about their current and expected future use of outcomes-based contracting. To

430

American Health & Drug Benefits

determine what drives the use of outcomes-based contracting, they were asked what the highest perceived value of such contracting was. Finally, the participants were asked to rank the current priorities for decisionmaking and the expected future priorities. The objectives of the PTS were to evaluate P&T pharmacists’ perceptions of the importance and the performance of 13 drug and manufacturer attributes used in the formulary decision-making process and to compare these perceptions to actual formulary decisions. P&T pharmacists were asked a single yes or no question to determine if they currently use HEOR data in formulary decision-making. Pharmacists were also asked a single yes or no question to determine if formulary decisions were based on the lowest net ingredient cost if data indicated equivalent safety and efficacy. To determine the importance of the product and manufacturer attributes, participants were asked to rate the importance of specific attributes in their decisionmaking and in formulary placements. To determine the organization’s performance in using these attributes, the individuals were asked to rate how

www.AHDBonline.com

November/December 2012

Vol 5, No 7

Use of HEOR in Healthcare Decision-Making

Table 2 Characteristics of the 72 Respondents in the Managed Care Survey Respondents, N

Minimum covered members, N

Maximum covered members, in millions, N

Median, in millions, N

HMO

8000

0.5

IHCS

6200

18.7

0.27

PBM

100

1.0

PPO

50,000

3.68

0.5

0.2

Total

100

0.5

Organization

IHCS indicates integrated healthcare system; PBM, pharmacy benefit management; PPO, preferred provider organization; VA, Veterans Affairs.

Figure 1 Respondent Characteristics in the Pharmacy & Therapeutics Committee Survey

Current rolea 25

Professional degreeb

Years at current organization

Organization type

Respondents, N

0 er ort ole mb upp No r e s T m P&T P&

D yr yrs 0 yrs 0 yrs arm <1 1-5 1 >1 Ph 6-

h RP

D Ph

r he Ot

O MC

M PB

S ed IHC nsur i le fS

r he Ot

Response rate, 17% (30:176) a

Six responses not eligible; 1 incomplete. May select more than 1. IHCS indicates integrated healthcare system; MCO, managed care organization; PBM, pharmacy benefit management; PPO, preferred provider organization; P&T, Pharmacy & Therapeutics Committee.

well the organization has performed in using the attributes when making formulary decisions. The drug attributes that were assessed included current drug market share, efficacy, head-to-head comparative data, net ingredient cost, outcomes data, drug differentiation, drug superiority, and safety. The manufacturer attributes that were assessed included the ability of the manufacturer to drive market share, cus-

Vol 5, No 7

November/December 2012

tomer programs, manufacturer relationship, rebates, and the size of the manufacturer. Finally, to examine drug placement when data appear to demonstrate equal safety and efficacy, pharmacists were asked to choose between hypothetical lipid-lowering and proton-pump inhibitor (PPI) drugs to place on a formulary (Table 1). The drug placement choices were then compared with actual formulary decisions for simi-

www.AHDBonline.com

American Health & Drug Benefits

431

BUSINESS

Use of HEOR and Importance of Drug Cost in Formulary Decision-Making as Assessed by Managed Figure 2 Care Formulary Decision Makers in 2 Independent Surveys in 2010

Formulary decision-making, %

Pharmacy & Therapeutics Committee Survey

Managed Care Survey Yes No

100 86.7%

81.9%

76.7%

73.6%

60 40 26.4%

23.3% 20

18.1%

13.3%

Final decision is based on price if products seem equally safe and effective

Review of HEOR data is part of the P&T process

HEOR is used in formulary decisions today

HEOR will be used more in decision in the future

HEOR indicates health economics and outcomes research; P&T, Pharmacy & Therapeutics Committee.

lar drugs in organizations with published formularies. The survey responses are reported using frequencies and percentages.

Results Description of Survey Respondents Table 2 outlines MCS respondent characteristics in the United States, and Figure 1 shows PTS respondent characteristics in the United States. The 72 respondents to the MCS represent organizations with national (54%), regional (32%), and local coverage (14%). Managed care organization (MCO) membership varied from tens of thousands of lives to millions of lives, with a median of 500,000 lives. A total of 4 completed surveys were excluded, because the respondents came from the pharmaceutical industry. Of the 72 respondents, 22 were from HMOs, 12 from integrated healthcare systems (IHCSs), 26 from PBMs, 8 from preferred provider organizations, and 4 from Veterans Affairs organizations. Of the 30 respondents to the PTS, 20 came from MCOs, 8 from PBMs, 1 from an IHCS, and 1 from an organization type not listed on the survey. In addition, 23 respondents indicated that they were P&T Committee members, 18 identified themselves in P&T support functions (it is possible to be a member of a P&T Committee and a supporter simultaneously), and 6 respondents had no direct role in a P&T Committee. The SMDM workshop discussion included 28 European experts involved in various aspects of healthcare decision-making with either an academic or health authority background. Participants were from

432

American Health & Drug Benefits

Austria, France, Germany, Poland, Spain, Switzerland, and the United Kingdom. Respondents in the Asian, Middle Eastern and African, and Latin American surveys were all participants in workshops on future trends in healthcare, including pharmacists and experts with medical, public health, or health economic education who were working in healthcare or healthcare decision-making.

Core Surveys in the United States Current and Expected Future Use of HEOR Although 73.6% of the respondents in the MCS state that HEOR data are used regularly, only 5% indicate that the use of HEOR is included in the bylaws of their organizations. The majority (approximately 82%) of the respondents expect an increased use of HEOR in the future (Figure 2). In the PTS, approximately 77% indicate that HEOR is considered a standard part of the P&T review process, and, at the same time, approximately 87% state that drugs with a lower cost are given priority when efficacy and safety are equal. However, quality assessment for HEOR data is limited. In addition, 68% of the MCS respondents indicate that there is no quality standard for HEOR being used by their organization (Table 3). Among those using quality assessments, this is done by using in-house checklists or by relying on the expertise of internal outcomes teams, external review teams, or the P&T Committee. When asked what changes would be needed in their organizations to make better use of HEOR data, 54% of the MCS respondents see a need for a more clear defini-

www.AHDBonline.com

November/December 2012

Vol 5, No 7

Use of HEOR in Healthcare Decision-Making

Respondents’ Rating of Importance and Own Figure 3 Performance Related to 13 Decision Factorsa

Question: Do you have a system in place for quality assessment for HEOR evidence? Organization

Respondent, N

Yes, N (%)

No, N (%)

100

Strengths

“Over and above”

Net ingredient cost

5 (25)

15 (75)

IHCS

2 (20)

7 (80)

PBM

12 (60)

6 (40)

PPO

1 (13)

7 (87)

4 a

Total

0 (0)

4 (100)

20 (32)

39 (68)

Good/very good performance, %

HMO

Only 62 survey respondents replied to this question. HEOR indicates health economics and outcomes research; IHCS, integrated healthcare system; PBM, pharmacy benefit manager; PPO, preferred provider organization; VA, Veterans Affairs.

PPI drugs, N (%)

Drug A (lowest net ingredient cost)

23 (77)

25 (83)

Drug B

7 (23)

5 (17)

80 70

Rebates 60

Current market share Manufacturer relationship Ability of manufacturer to drive market share Customer programs Size of manufacturer

50 40 30 20

Product superiority Outcomes data Product differentiation Head-to-head comparisons

10 0

Drug Choices in Hypothetical Drug Placement Table 4 Scenarios in the Pharmacy & Therapeutics Committee Surveya Lipid-lowering drugs, N (%)

y fet cy Sa ffica E

Importanceb

Performance c

HEOR Quality Assessment in the Managed Table 3 Care Survey

Opportunities 0

Gaps 40

100

Very/extremely important, % a

Gridlines represent median splits. Percentage of respondents indicating the factor was very or extremely important. c Percentage of respondents indicating their performance was good or very good. b

Hypothetical drug

Least expensive net ingredient drug available on preferred tiers Not available Available

14 (93)

11 (73)

1 (7)

4 (27)

Only 15 of 23 formularies for fibrates and 15 of 25 formularies for PPIs could be examined for those who chose lowest net ingredient cost in the hypothetical scenario. PPI indicates proton-pump inhibitor.

tion of HEOR data requirements, 47% see a need for increased HEOR competency in the decision-making committee, 42% suggest a need for improved in-house data analysis, and 36% see a need for increased prospective data collection. In addition, 39% indicate that there should be more frequent reevaluation of decisions based on HEOR evidence.

Vol 5, No 7

November/December 2012

Importance of Product and Manufacturer Attributes for Decision-Making In the PTS, the safety and the efficacy of the drug are reported to be very or extremely important by all of the respondents to this survey (Figure 3). In addition, net ingredient cost and head-to-head comparisons are considered very or extremely important by 90% of participants, drug superiority by 87%, and outcomes data by 83%. Similarly, 100% of respondents also indicate they have good or very good performance in formulary decisions with regard to safety and efficacy. Performance using net ingredient cost is rated as good or very good by 90%, rebates by 67%, drug superiority by 67%, and outcomes data by 63%. Of note, although rebates have high performance (67%), they have low importance (43%), whereas head-to-head trial data have low performance (47%) and high importance (90%). Cost Priorities In the PTS, with hypothetical medication placement scenarios—when safety, efficacy, and dosing were proposed to be similar—the medication with the lowest net

www.AHDBonline.com

American Health & Drug Benefits

433

BUSINESS

ingredient cost was chosen most often: 77% for lipid-lowering and 83% for PPI drugs (Table 4). However, when actual published formularies were examined, only 7% had the lipid-lowering drug and 27% had the PPI with the lowest net ingredient cost available on a preferred tier. In the MCS, the highest priorities in the current decision-making process include direct medical costs, targetCurrent and Expected Future Use of HEOR in Figure 4 Contracting, as Assessed in the Managed Care Survey Current use of HEOR in contracting

Only exceptionally 29%

No 33%

Expected future use To a limited degree 18% No 10%

Probably 32%

Certainly 40%

Yes, increasingly 38%

HEOR indicates health economics and outcomes research.

ed treatment algorithms, and risk of off-label drug use. However, the anticipated future priorities demonstrate a shift toward patient-centered outcomes, in which patient quality of life, patient group opinions, and indirect medical costs are given higher priority.

Outcomes-Based Contracting In an outcomes-based contract, the supplier is paid for the realization of a defined set of health outcomes, business results, or key performance indicators (eg, treatment response within 2 months after initiation of treatment, as measured by defined end points or symptom reduction by a defined percentage). Only 38% of organizations surveyed in the MCS currently use outcomes-based contracting, whereas 72% anticipate (ie, probable or certain) use in the future (Figure 4). The highest perceived value of outcomes-based contracting is reported to be higher flexibility for access to innovative technologies, reduced financial risk, reduced clinical risk, the opportunity for a trial period, and giving more responsibility to the manufacturer for the outcomes. The barriers to implementing outcomes-based contracting include duration of time to outcomes, the definition of unquestionable end points, and the challenge of individual contracts within drug classes (Figure 5).

Figure 5 Future Outcome-Based Contracting: The Managed Care Survey Question: Which major hurdles for the integration of outcomes-based contracting into your process do you perceive? HMO (N = 22) IHCS (N = 12) PBM (N = 26) PPO (N = 8) VA (N = 4) Total (N = 72)

100

Responses, %

80 60 40 20 0

ts th in or cy of th/ wi ers ith ult ien wi ers ng ost ity ersâ&#x20AC;&#x2122; ren t i l s w i a a s c r n e c i p sp es nt su ctu tio vid rab th cts iff ith to pa ith o ran roc eme l is nufa ica f pro tra re d w t le e p a n n m b f a s g u o a d g co a w m no ue Le m a l c es ko /tim se na on ed dat ac ua lass iss on om catio sti ea ma i c L r d t e C i y u r c t i u ra d u iv c qu In bil ed Re o Du Ind drug Un Lia e om c t ou

ts oin p d en

IHCS indicates integrated healthcare system; PBM, pharmacy benefit management; PPO, preferred provider organization; VA, Veterans Affairs.

434

American Health & Drug Benefits

www.AHDBonline.com

November/December 2012

Vol 5, No 7

Use of HEOR in Healthcare Decision-Making

Figure 6 How Much HTA Is Used in Formulary Decision-Making in Latin America, Asia, and the Middle East and Africa? 100%

Latin America Asia Middle East and Africa

80% 71% 60% 50% 40%

20% 13%

23% 14%

11%

Always

14%

13%

30%

28%

25%

Often

Sometimes

Rarely

Never

NOTE: The respondents were from Latin America (N = 61), Asia (N = 9), and the Middle East and Africa (N = 9). HTA indicates health technology assessment.

Global Survey Extensions at Select Workshops Europe Questions from the MCS were presented at a workshop at the European SMDM in 2011. The interest among the European participants in the results from the United States was very high and raised discussion as to what each question means in relation to decision-making in European countries, such as France, Germany, Poland, Austria, and the United Kingdom. In European countries, there is often a gap between national technology assessment and regional or local decision-making regarding drug formularies. The range of how countries in Europe are using HEOR data is broad, with the potential for considerable variations even within countries. Although some countries use clearly defined and explicit decision processes, including health economic analysis (eg, the United Kingdom, Netherlands, Poland), others may have less emphasis on health economic analysis and more emphasis on low price contracting after the primary decision has been made regarding the clinical benefit of a new therapy (eg, Germany). Because of the high diversity of the healthcare systems and the reimbursement decision processâ&#x20AC;&#x201D;and because of the level of payer decisions (ie, national, regional, local)â&#x20AC;&#x201D;the MCS would have to be adapted to the healthcare environment in each of the European countries that are included in an extended study. Pilot Surveys in Asia, Latin America, and the Middle East and Africa The use of HEOR varied across all regions. Only 9%

Vol 5, No 7

November/December 2012

of respondents in Latin America and 11% of respondents in Asia said that HEOR data are a mandatory part of the formulary or reimbursement decision-making. In addition, 24% of respondents in Latin America and 44% in Asia said that HEOR data are used, but are not mandatory. Furthermore, 15% of respondents in Latin America said that HEOR evidence was never used. In the Middle East and Africa, there is some awareness of HEOR but only marginal use of it in making formulary decisions: approximately 33% of respondents saw no use of HEOR in their environment. In addition, 40% of respondents noted that it is used, but it is definitely not mandatory to use such information. Assessment of health technology is a process used to examine the health and economic outcomes of implementing healthcare practices or treatments, including medications.9,10 HTA is used in Asia more than in Latin America or the Middle East and Africa (Figure 6). In Asia, 38% of respondents claimed that decisions are always or often based on HTA, but only 19% said the same in Latin America. Compared with only 13% in Asia who use HTA rarely, in Latin America 30% use it rarely and 23% said they do not use HTA methods at all. In the Middle East and Africa, 71% of respondents indicated that HTA is used rarely and 14% said it was never used. Outcomes-based contracting is not used by 53% of respondents in Asia and by 44% in Latin America; and 12% and 11%, respectively, said that outcomes research is used for contracting. In the Middle East and Africa, outcomes-based contracting is done rarely today, but its use is expected to increase.

www.AHDBonline.com

American Health & Drug Benefits

435

BUSINESS

Discussion Core Surveys in the United States Based on the results of the MCS and PTS, the majority of US healthcare decision makers use HEOR evidence when making formulary and coverage decisions, and the use of HEOR for these decisions is expected to increase in the future. However, the surveys also indicate that the majority of these decision makers do not have a quality assurance process in place for reviewing HEOR data. In addition, nearly half of the respondents indicated that clearer definitions of HEOR data requirements and increased competency of decision makers in interpreting HEOR data were needed to make better use of them. This may indicate a need for P&T Committees to include members with expertise in the evaluation and application of HEOR or to provide additional training programs for their current members.

Based on the results of the MCS and PTS, the majority of US healthcare decision makers use HEOR evidence when making formulary and coverage decisions. However, the majority of these decision makers do not have a quality assurance process in place for reviewing HEOR data. The hypothetical product placement scenarios from the PTS suggest that decision makers make decisions based on cost when other clinical factors are considered equal; however, actual product placements showed the lowest-priced branded products were not in a preferred position in the majority of formularies. This may indicate that the role of cost in the decision-making process is complicated and difficult to isolate from other factors, such as rebates. Decision makers view outcomes-based contracting as a way to provide better access to new technologies and to reduce clinical and financial risks; however, the current use of outcomes-based contracting appears to be limited. These survey results provide an important benchmark for the use of HEOR before the healthcare reform legislation of the Patient Protection and Affordable Care Act (ACA). In 2009, the American Recovery and Reinvestment Act (ARRA) allocated $1.1 billion for comparative effectiveness research and the ACA created the Patient-Centered Outcomes Research Institute (PCORI) to provide a steady stream of funding for outcomes research. The ARRA and the creation of PCORI indicate a shift toward greater support

436

American Health & Drug Benefits

of HEOR research by the US government. This will certainly lead to an increase in the availability of HEOR data to healthcare payers and will help to provide them with data that better represent the populations they serve. Because the pharmaceutical industry currently provides most of the funding for HEOR, this shift in funding may also provide HEOR data for different populations and for disease states that were previously not examined.

Global Survey Extensions at Selected Workshops Europe Observations at the SMDM meeting in Austria and at the subsequent Health Technology Assessment International conference in Dublin suggest that this project comes at a time when HTA organizations in many countries are beginning to address the question of how the work and research on HTA is used by decision makers. Because HEOR often is a major part of HTA, the analysis of the use of HEOR in decision-making seems to be timely and relevant. In terms of healthcare priorities, the discussions with a broad range of European participants (eg, Germany, France, Switzerland, Poland, Austria) indicated that cost containment is a key priority. It was considered to have a higher impact on formulary or reimbursement decisions than patient centricity. Asia, Latin America, and the Middle East and Africa Healthcare policy priorities reported by the respondents ranged from universal coverage and improving access to healthcare services (Latin America, Asia, Middle East and Africa) to improving quality (Middle East and Africa), better health outcomes (specifically Oman in the Middle East and Africa region), and cost containment (Asia and Egypt). Such priorities vary from country to country; for example, in the Middle East and Africa region, the healthcare priorities perceived in Oman differed completely from those seen in Egypt. Some Asian countries have developed systematic procedures for pricing and reimbursement. Many Latin American, Asian, or Middle Eastern and African formulary or reimbursement decision makers cannot yet make appropriate use of HEOR data, or they do not have access to local HTAs. However, this is expected to change. For example, at the 2012 Second Saudi International Pharmacoeconomics Conference, the Minister of Health of Saudi Arabia, Dr Abdullah Al-Rabea, recently announced plans to establish a national pharmacoeconomic and outcomes research center in an attempt to curb public healthcare expenditures.11 Despite greater incorporation of HEOR and HTA in Asia, both Asian and Latin American respondents intend to use HEOR data to a comparable extent for

www.AHDBonline.com

November/December 2012

Vol 5, No 7

Use of HEOR in Healthcare Decision-Making

contracting. From the payer perspective, outcomesbased contracts lower the clinical and economic risks but require commitment to system changes and a partnership approach, as well as alignment of objectives. For the supplier, there is a higher risk than with an activity- or drug-based payment scheme, but there is also a potentially greater reward and a similar requirement for the alignment of objectives. Implementation of outcomes-based contracts usually requires more complex management, performance measurement, and remuneration rules. Despite the attractiveness of only paying for the expected results, it is usually much easier for all parties involved to pay on an input-based (ie, time and materials contracts) or outputbased (ie, fixed price per service or product) contract.

Limitations Neither of the 2 surveys (MCS or PTS) was applied to a systematically selected audience, and many types of healthcare decision makers may not have been included in these surveys. Therefore, the respondents may only represent selected viewpoints, which may differ from the average decision maker in the United States. Furthermore, the global expansion of the survey was limited to a small number of decision makers participating at the workshops. The reporting of the results by region (ie, Asia, Latin America, and the Middle East) does not reveal the specifics of each country, and there may be considerable differences between the participating countries. Conclusion The healthcare policy priorities reported by the respondents across all regions ranged from cost containment, via more patient-centric policies, to universal access. The difference in using HEOR for decision-making across various healthcare systems may be related to the priorities and can be expected to change over time. Once a country has achieved near-universal coverage, for example, the focus may then switch to another priority, such as improvement of health outcomes, cost containment, or toward maximizing overall value. The current data were quantitative in nature and should provide direction toward future qualitative surveys to obtain more detail and reason behind the varia-

tions among countries in the use of HEOR in decisionmaking. This study provides an important benchmark of

This study provides an important benchmark of how HEOR is used in the United States before healthcare reform, and future research should investigate how the use of HEOR changes afterward. how HEOR is used in the United States before healthcare reform, and future research should investigate how the use of HEOR changes afterward. â&#x2013; Study Funding Novartis contributed to the funding of the surveys. Author Disclosure Statement Dr Holtorf receives travel grants from Novartis. Dr Brixner received a grant from Novartis for the survey. Dr Keskinaslan is an employee of Novartis. Dr Oderda is a consultant to Novartis. Dr Dye is a consultant to Polyglot Systems, Inc, and is an employee of Humana. Dr Bellows has nothing to disclose.

References 1. Garrison LP Jr, Neumann PJ, Radensky P, Walcoff SD. A flexible approach to evidentiary standards for comparative effectiveness research. Health Aff (Millwood). 2010;29:1812-1817. 2. Chambers JD, Neumann PJ. US healthcare reform: implications for health economics and outcomes research. Expert Rev Pharmacoecon Outcomes Res. 2010;10:215-216. 3. van Nooten F, Holmstrom S, Green J, et al. Health economics and outcomes research within drug development: challenges and opportunities for reimbursement and market access within biopharma research. Drug Discov Today. 2012;17:615-622. 4. Doherty J, Kamae I, Lee KK, et al. What is next for pharmacoeconomics and outcomes research in Asia? Value Health. 2004;7:118-132. 5. Tantivess S, Teerawattananon Y, Mills A. Strengthening cost-effectiveness analysis in Thailand through the establishment of the health intervention and technology assessment program. Pharmacoeconomics. 2009;27:931-945. 6. Holtorf AP, Watkins JB, Mullins CD, Brixner D. Incorporating observational data into the formulary decision-making processâ&#x20AC;&#x201D;summary of a roundtable discussion. J Manag Care Pharm. 2008;14:302-308. 7. Brixner DI, Holtorf AP, Neumann PJ, et al. Standardizing quality assessment of observational studies for decision making in health care. J Manag Care Pharm. 2009; 15:275-283. 8. Biskupiak J, Brixner DI, Holtorf AP. Current and future use of pharmacoeconomic and pharmacotherapy outcomes research data in decision making. Abstract presented at the 12th Biennial European Meeting of the Society for Medical Decision Making; May 30-June 2, 2010; Hall/Innsbruck, Austria. Abstract 22. 9. Leung MY, Halpern MT, West ND. Pharmaceutical technology assessment: perspectives from payers. J Manag Care Pharm. 2012;18:256-264. 10. Gallio D, Berto P. Health technology assessment (HTA): definition, role and use in the changing healthcare environment. Eur Ann Allergy Clin Immunol. 2007;39 Spec No:7-11. 11. Rasooldeen R. Move under way to set up pharmacoeconomics center. April 29, 2012. www.arabnews.com/node/412095. Accessed November 13, 2012.

Stakeholder Perspective on page 438

Vol 5, No 7

November/December 2012

www.AHDBonline.com

American Health & Drug Benefits

437

BUSINESS

STAKEHOLDER PERSPECTIVE

Health Economics and Outcomes Research Data Key in Coverage Decisions of New Medications By Douglas s. Burgoyne, PharmD President, VRx Pharmacy Services, LLC, Salt Lake City, Utah

PAYERS: Healthcare decision makers have used economic data to evaluate the cost-effectiveness of drugs for the past decade, at the very least. However, a formal process involving health economics, and not just drug cost, is emerging in the United States and around the world. This original research by Holtorf and colleagues shows that health economics and outcomes research (HEOR) will become a key factor in evaluating new chemical and biologic entities in the future. Pharmacy and Therapeutics (P&T) Committees have always placed the highest value on medication safety and effectiveness. Clearly, the job of all healthcare practitioners, whether on the frontlines seeing patients or in managed care organizations making coverage decisions, is to protect the health and welfare of their patients/members and provide medications that do what they claim to do. As Holtorf and colleagues demonstrate in this study, net cost is the next most important factor when considering a new medication for formulary coverage. It is interestingâ&#x20AC;&#x201D; although not surprising to long-time managed care pharmacistsâ&#x20AC;&#x201D;that rebates are lower on the importance scale for most decision makers, as this study shows. The results of this study reinforce the fact that P&T Committees would prefer lower wholesale acquisition cost pricing and better upfront drug pricing than higher rebates. With the advent of the Affordable Care Act and its spawn, the Patient-Centered Outcomes Research Institute, more HEOR will be conducted and published by pharmaceutical manufacturers and by other researchers.

438

American Health & Drug Benefits

This oncoming wave of new research will undoubtedly provide compelling outcomes data that manufacturers will use in an attempt to get every new drug added to the formulary. P&T Committees must therefore become more sophisticated in considering the quality of the HEOR data being presented, and use it appropriately in their evaluations of new drugs. Not all HEOR data are quality research. And more HEOR does not equal better HEOR. Healthcare stakeholders reported in this survey-based research that they do not have a formal process to assess the quality of HEOR evidence. This is a warning sign. Healthcare decision makers must increase the rigor of their evaluation of the HEOR evidence being used by their organizations to continue making the best decisions possible for their organizations and the population they serve. PATIENTS: Patients must also become more sophisticated in their review of medications. Over the past 10 years, patients have become key decision makers in their own healthcare. High-deductible health plans, health savings accounts, and the explosion of drug information available on the internet have made patients their own healthcare decision makers. All of the new HEOR data that will be available to patients with a few clicks of the mouse may lead to confusion and misinterpretation. Patients, too, must learn to critically evaluate to the best of their ability the data being presented to them, and then go to their trusted healthcare providers to help them interpret the evidence.

www.AHDBonline.com

November/December 2012

Vol 5, No 7

2012 PEER REVIEWERS

THANK YOU AHDB PEER REVIEWERS The Editors and Publishers of American Health & Drug Benefits (AHDB) wish to extend a heartfelt Thank You to all the people who participated in the Journalâ&#x20AC;&#x2122;s peer-review process during 2012. Your diligence and expert comments help to ensure the continued high quality of the articles published in AHDB. We are grateful for the time and effort you so generously give freely to the review process by applying your expertise and knowledge in the spirit of true collaboration that typifies the editorial mission of AHDB. We look forward to the continued collaboration with all of you in the coming year.

442

Al B. Benson, III, MD, FACP

Steve Miff, PhD

Jeffrey A. Bourret, RPh, MS, FASHP

Matthew Mitchell, PharmD, MBA

Joel V. Brill, MD, AGAF, CHCQM

Michael F. Murphy, MD, PhD

Frank Casty, MD, FACP

Quang Nguyen, DO, FACP, FACE

Michael T. Einodshofer, RPh, MBA

Gary M. Owens, MD

Josh Feldstein

Laura T. Pizzi, PharmD, MPH, RPh

Jack E. Fincham, PhD, RPh

Paul Anthony Polansky, BSPharm, MBA

Leslie S. Fish, PharmD

Sarah A. Priddy, PhD

Walid F. Gellad, MD, MPH

Kamakshi V. Rao, PharmD, BCOP, CPP

Jeff Jianfei Guo, BPharm, MS, PhD

Donald Reese, PharmD, MBA

Joseph D. Jackson, PhD

Timothy S. Regan, BPharm, RPh, CPh

Michael S. Jacobs, RPh

Jaan Sidorov, MD

Jeffrey Januska, PharmD

Samuel M. Silver, MD, PhD, FASCO

J. B. Jones, PhD, MBA

Christina A. Stasiuk, DO, FACOI

Atheer A. Kaddis, PharmD

Eric G. Tangalos, MD, FACP, AGSF, CMD

James T. Kenney, Jr, RPh, MBA

Albert Tzeel, MD, MHSA, FACPE

Steven T. Kmucha, MD, JD, FACS

F. Randy Vogenberg, RPh, PhD

Joshua N. Liberman, PhD

Dominic Vu, PharmD

Annesha W. Lovett, PharmD, MS, PhD

Vincent J. Willey, PharmD

Robert Mancini, PharmD

David W. Wright, MPH

American Health & Drug Benefits

www.AHDBonline.com

November/December 2012

Vol 5, No 7

For your patients with type 2 diabetes who need more than A1C control, choose Levemir ® (insulin detemir [rDNA origin] injection)

24/7 GLUCOSE CONTROL MORE

Karen’s doctor said taking Levemir ® (insulin detemir [rDNA origin] injection) once-daily may get her the control she needs & more Low rates of hypoglycemia In 1 study, approximately 45% of patients in each treatment arm achieved A1C <7% with no hypoglycemic events within the last 4 weeks of observation.1 t A single major hypoglycemic event was reported in the 70-90 mg/dL group; no major hypoglycemic events in the 80-110 mg/dL group t Minor hypoglycemia rates were 5.09 (70-90 mg/dL) and 3.16 (80-110 mg/dL) per patient-year*

From a 20-week, randomized, controlled, multicenter, open-label, parallel-group, treat-to-target trial using a self-titration algorithm in insulin-naïve patients with type 2 diabetes, A1C ≥7% and ≤9% on OAD therapy randomized to Levemir® and OAD (1:1) to 2 different fasting plasma glucose (FPG) titration targets (70-90 mg/dL [n=121] or 80-110 mg/dL [n=122]). At study end, in the 80-110 mg/dL group, 55% of patients achieved goal (A1C <7%) with A1C decrease of 0.9%. The mean A1C was 7%.1

Covered on more than 90% of managed care plans2† hypoglycemia usually reflects the time action profile of the administered insulin formulations. Glucose monitoring is essential for all patients receiving insulin therapy. Any changes to an insulin regimen should be made cautiously and only under medical supervision. Needles and Levemir ® FlexPen® must not be shared. Severe, life-threatening, generalized allergy, including anaphylaxis, can occur with insulin products, including Levemir ®. Adverse reactions associated with Levemir ® include hypoglycemia, allergic reactions, injection site reactions, lipodystrophy, rash and pruritus. Careful glucose monitoring and dose adjustments of insulin, including Levemir ®, may be necessary in patients with renal or hepatic impairment. Levemir ® has not been studied in children with type 2 diabetes, and in children with type 1 diabetes under the age of six.

Indications and Usage Levemir ® (insulin detemir [rDNA origin] injection) is indicated to improve glycemic control in adults and children with diabetes mellitus. Important Limitations of Use: Levemir ®isnotrecommendedforthetreatmentof diabetic ketoacidosis. Intravenous rapid-acting or short-acting insulin is the preferred treatment for this condition.

Important Safety Information Levemir ® is contraindicated in patients hypersensitive to insulin detemir or one of its excipients. Do not dilute or mix Levemir® with any other insulin solution, or use in insulin infusion pumps. Do not administer Levemir® intravenously or intramuscularly because severe hypoglycemia can occur. Hypoglycemia is the most common adverse reaction of insulin therapy, including Levemir®. The timing of

Please see brief summary of Prescribing Information on adjacent page. Needles are sold separately and may require a prescription in some states. *Minor=SMPG <56 mg/dL and not requiring third-party assistance.

On your iPhone®

Scan the QR code to download the NovoDose™ app to know how to optimally dose Levemir®

†

Intended as a guide. Lower acquisition costs alone do not necessarily reﬂect a cost advantage in the outcome of the condition treated because other variables affect relative costs. Formulary status is subject to change.

References: 1. Blonde L, Merilainen M, Karwe V, Raskin P; TITRATE™ Study Group. Patient-directed titration for achieving glycaemic goals using a once-daily basal insulin analogue: an assessment of two different fasting plasma glucose targets - the TITRATE™ study. Diabetes Obes Metab. 2009;11(6):623-631. 2. Data on ﬁle. Novo Nordisk Inc, Princeton, NJ. iPhone ® is a registered trademark of Apple, Inc. FlexPen® and Levemir ® are registered trademarks and NovoDose™ is a trademark of Novo Nordisk A/S. © 2012 Novo Nordisk Printed in the U.S.A. 0911-00005042-1 April 2012

LEVEMIR® (insulin detemir [rDNA origin] injection) Rx ONLY BRIEF SUMMARY. Please consult package insert for full prescribing information. INDICATIONS AND USAGE: LEVEMIR® is indicated to improve glycemic control in adults and children with diabetes mellitus. Important Limitations of Use: LEVEMIR® is not recommended for the treatment of diabetic ketoacidosis. Intravenous rapid-acting or short-acting insulin is the preferred treatment for this condition. CONTRAINDICATIONS: LEVEMIR® is contraindicated in patients with hypersensitivity to LEVEMIR® or any of its excipients. Reactions have included anaphylaxis. WARNINGS AND PRECAUTIONS: Dosage adjustment and monitoring: Glucose monitoring is essential for all patients receiving insulin therapy. Changes to an insulin regimen should be made cautiously and only under medical supervision. Changes in insulin strength, manufacturer, type, or method of administration may result in the need for a change in the insulin dose or an adjustment of concomitant anti-diabetic treatment. As with all insulin preparations, the time course of action for LEVEMIR® may vary in different individuals or at different times in the same individual and is dependent on many conditions, including the local blood supply, local temperature, and physical activity. Administration: LEVEMIR® should only be administered subcutaneously. Do not administer LEVEMIR® intravenously or intramuscularly. The intended duration of activity of LEVEMIR® is dependent on injection into subcutaneous tissue. Intravenous or intramuscular administration of the usual subcutaneous dose could result in severe hypoglycemia. Do not use LEVEMIR® in insulin infusion pumps. Do not dilute or mix LEVEMIR® with any other insulin or solution. If LEVEMIR® is diluted or mixed, the pharmacokinetic or pharmacodynamic profile (e.g., onset of action, time to peak effect) of LEVEMIR® and the mixed insulin may be altered in an unpredictable manner. Hypoglycemia: Hypoglycemia is the most common adverse reaction of insulin therapy, including LEVEMIR®. The risk of hypoglycemia increases with intensive glycemic control. Patients must be educated to recognize and manage hypoglycemia. Severe hypoglycemia can lead to unconsciousness or convulsions and may result in temporary or permanent impairment of brain function or death. Severe hypoglycemia requiring the assistance of another person or parenteral glucose infusion, or glucagon administration has been observed in clinical trials with insulin, including trials with LEVEMIR®. The timing of hypoglycemia usually reflects the time-action profile of the administered insulin formulations. Other factors such as changes in food intake (e.g., amount of food or timing of meals), exercise, and concomitant medications may also alter the risk of hypoglycemia. The prolonged effect of subcutaneous LEVEMIR® may delay recovery from hypoglycemia. As with all insulins, use caution in patients with hypoglycemia unawareness and in patients who may be predisposed to hypoglycemia (e.g., the pediatric population and patients who fast or have erratic food intake). The patient’s ability to concentrate and react may be impaired as a result of hypoglycemia. This may present a risk in situations where these abilities are especially important, such as driving or operating other machinery. Early warning symptoms of hypoglycemia may be different or less pronounced under certain conditions, such as longstanding diabetes, diabetic neuropathy, use of medications such as beta-blockers, or intensified glycemic control. These situations may result in severe hypoglycemia (and, possibly, loss of consciousness) prior to the patient’s awareness of hypoglycemia. Hypersensitivity and allergic reactions: Severe, life-threatening, generalized allergy, including anaphylaxis, can occur with insulin products, including LEVEMIR®. Renal Impairment: No difference was observed in the pharmacokinetics of insulin detemir between non-diabetic individuals with renal impairment and healthy volunteers. However, some studies with human insulin have shown increased circulating insulin concentrations in patients with renal impairment. Careful glucose monitoring and dose adjustments of insulin, including LEVEMIR®, may be necessary in patients with renal impairment. Hepatic Impairment: Nondiabetic individuals with severe hepatic impairment had lower systemic exposures to insulin detemir compared to healthy volunteers. However, some studies with human insulin have shown increased circulating insulin concentrations in patients with liver impairment. Careful glucose monitoring and dose adjustments of insulin, including LEVEMIR®, may be necessary in patients with hepatic impairment. Drug interactions: Some medications may alter insulin requirements and subsequently increase the risk for hypoglycemia or hyperglycemia. ADVERSE REACTIONS: The following adverse reactions are discussed elsewhere: Hypoglycemia; Hypersensitivity and allergic reactions. Clinical trial experience: Because clinical trials are conducted under widely varying designs, the adverse reaction rates reported in one clinical trial may not be easily compared to those rates reported in another clinical trial, and may not reflect the rates actually observed in clinical practice. The frequencies of adverse reactions (excluding hypoglycemia) reported during LEVEMIR® clinical trials in patients with type 1 diabetes mellitus and

type 2 diabetes mellitus are listed in Tables 1-4 below. See Tables 5 and 6 for the hypoglycemia findings. Table 1: Adverse reactions (excluding hypoglycemia) in two pooled clinical trials of 16 weeks and 24 weeks duration in adults with type 1 diabetes (adverse reactions with incidence ≥ 5%)

Upper respiratory tract infection Headache Pharyngitis Influenza-like illness Abdominal Pain

LEVEMIR®, % (n = 767) 26.1 22.6 9.5 7.8 6.0

NPH, % (n = 388) 21.4 22.7 8.0 7.0 2.6

Table 2: Adverse reactions (excluding hypoglycemia) in a 26-week trial comparing insulin aspart + LEVEMIR® to insulin aspart + insulin glargine in adults with type 1 diabetes (adverse reactions with incidence ≥ 5%)

Upper respiratory tract infection Headache Back pain Influenza-like illness Gastroenteritis Bronchitis

LEVEMIR®, % (n = 161) 26.7 14.3 8.1 6.2 5.6 5.0

Glargine, % (n = 159) 32.1 19.5 6.3 8.2 4.4 1.9

Table 3: Adverse reactions (excluding hypoglycemia) in two pooled clinical trials of 22 weeks and 24 weeks duration in adults with type 2 diabetes (adverse reactions with incidence ≥ 5%)

Upper respiratory tract infection Headache

LEVEMIR®, % (n = 432) 12.5 6.5

NPH, % (n = 437) 11.2 5.3

Table 4: Adverse reactions (excluding hypoglycemia) in a 26-week clinical trial of children and adolescents with type 1 diabetes (adverse reactions with incidence ≥ 5%)

Upper respiratory tract infection Headache Pharyngitis Gastroenteritis Influenza-like illness Abdominal pain Pyrexia Cough Viral infection Nausea Rhinitis Vomiting

LEVEMIR®, % (n = 232) 35.8 31.0 17.2 16.8 13.8 13.4 10.3 8.2 7.3 6.5 6.5 6.5

NPH, % (n = 115) 42.6 32.2 20.9 11.3 20.9 13.0 6.1 4.3 7.8 7.0 3.5 10.4

Hypoglycemia: Hypoglycemia is the most commonly observed adverse reaction in patients using insulin, including LEVEMIR®. Tables 5 and 6 summarize the incidence of severe and non-severe hypoglycemia in the LEVEMIR® clinical trials. Severe hypoglycemia was defined as an event with symptoms consistent with hypoglycemia requiring assistance of another person and associated with either a blood glucose below 50 mg/ dL or prompt recovery after oral carbohydrate, intravenous glucose or glucagon administration. Non-severe hypoglycemia was defined as an asymptomatic or symptomatic plasma glucose < 56 mg/dL (<50 mg/dL in Study A and C) that was self-treated by the patient. The rates of hypoglycemia in the LEVEMIR® clinical trials (see Section 14 for a description of the study designs) were comparable between LEVEMIR®-treated patients and non-LEVEMIR®-treated patients (see Tables 5 and 6).

Table 5: Hypoglycemia in Patients with Type 1 Diabetes Study A Type 1 Diabetes Adults 16 weeks In combination with insulin aspart Twice-Daily Twice-Daily NPH LEVEMIR® Severe hypo- Percent of patients 10.6 8.7 with at least 1 event glycemia (14/132) (24/276) (n/total N) Event/patient/year 0.52 0.43 Non-severe Percent of patients 88.0 89.4 hypoglycemia (n/total N) (243/276) (118/132) Event/patient/year 26.4 37.5

Study B Type 1 Diabetes Adults 26 weeks In combination with insulin aspart Twice-Daily Once-Daily LEVEMIR® Glargine

Study C Type 1 Diabetes Adults 24 weeks In combination with regular insulin Once-Daily Once-Daily NPH LEVEMIR®

Study D Type 1 Diabetes Pediatrics 26 weeks In combination with insulin aspart Once- or Twice Once- or Twice Daily LEVEMIR® Daily NPH

5.0 (8/161)

10.1 (16/159)

7.5 (37/491)

10.2 (26/256)

15.9 (37/232)

20.0 (23/115)

0.13 82.0 (132/161) 20.2

0.31 77.4 (123/159) 21.8

0.35 88.4 (434/491) 31.1

0.32 87.9 (225/256) 33.4

0.91 93.1 (216/232) 31.6

0.99 95.7 (110/115) 37.0

Table 6: Hypoglycemia in Patients with Type 2 Diabetes

Severe hypo- Percent of patients with at least 1 event (n/total N) glycemia Event/patient/year Non-severe Percent of patients hypoglycemia (n/total N) Event/patient/year

Study E Type 2 Diabetes Adults 24 weeks In combination with oral agents Twice-Daily NPH Twice-Daily LEVEMIR® 2.5 0.4 (6/238) (1/237) 0.01 0.08 40.5 64.3 (96/237) (153/238) 3.5 6.9

Insulin Initiation and Intensification of Glucose Control: Intensification or rapid improvement in glucose control has been associated with a transitory, reversible ophthalmologic refraction disorder, worsening of diabetic retinopathy, and acute painful peripheral neuropathy. However, long-term glycemic control decreases the risk of diabetic retinopathy and neuropathy. Lipodystrophy: Long-term use of insulin, including LEVEMIR®, can cause lipodystrophy at the site of repeated insulin injections. Lipodystrophy includes lipohypertrophy (thickening of adipose tissue) and lipoatrophy (thinning of adipose tissue), and may affect insulin adsorption. Rotate insulin injection sites within the same region to reduce the risk of lipodystrophy. Weight Gain: Weight gain can occur with insulin therapy, including LEVEMIR®, and has been attributed to the anabolic effects of insulin and the decrease in glucosuria. Peripheral Edema: Insulin, including LEVEMIR®, may cause sodium retention and edema, particularly if previously poor metabolic control is improved by intensified insulin therapy. Allergic Reactions: Local Allergy: As with any insulin therapy, patients taking LEVEMIR ® may experience injection site reactions, including localized erythema, pain, pruritis, urticaria, edema, and inflammation. In clinical studies in adults, three patients treated with LEVEMIR® reported injection site pain (0.25%) compared to one patient treated with NPH insulin (0.12%). The reports of pain at the injection site did not result in discontinuation of therapy. Rotation of the injection site within a given area from one injection to the next may help to reduce or prevent these reactions. In some instances, these reactions may be related to factors other than insulin, such as irritants in a skin cleansing agent or poor injection technique. Most minor reactions to insulin usually resolve in a few days to a few weeks. Systemic Allergy: Severe, life-threatening, generalized allergy, including anaphylaxis, generalized skin reactions, angioedema, bronchospasm, hypotension, and shock may occur with any insulin, including LEVEMIR®, and may be life-threatening. Antibody Production: All insulin products can elicit the formation of insulin antibodies. These insulin antibodies may increase or decrease the efficacy of insulin and may require adjustment of the insulin dose. In phase 3 clinical trials of LEVEMIR®, antibody development has been observed with no apparent impact on glycemic control. Postmarketing experience: The following adverse reactions have been identified during post approval use of LEVEMIR®. Because these reactions are reported voluntarily from a population of uncertain size, it is not always possible to reliably estimate their frequency or establish a causal relationship to drug exposure. Medication errors have been reported during post-approval use of LEVEMIR® in which other insulins, particularly rapid-acting or short-acting insulins, have been accidentally administered instead of LEVEMIR®. To avoid medication errors between LEVEMIR® and other insulins, patients should be instructed always to verify the insulin label before each injection.

Study F Type 2 Diabetes Adults 22 weeks In combination with insulin aspart Once- or Twice Daily LEVEMIR® Once- or Twice Daily NPH 4.0 1.5 (8/199) (3/195) 0.04 0.13 32.3 32.2 (63/195) (64/199) 1.6 2.0

More detailed information is available upon request.

For information about LEVEMIR® contact: Novo Nordisk Inc., 100 College Road West Princeton, NJ 08540 1-800-727-6500 www.novonordisk-us.com Manufactured by: Novo Nordisk A/S DK-2880 Bagsvaerd, Denmark Revised: 1/2012 Novo Nordisk®, Levemir®, NovoLog®, FlexPen®, and NovoFine® are registered trademarks of Novo Nordisk A/S. LEVEMIR® is covered by US Patent Nos. 5,750,497, 5,866,538, 6,011,007, 6,869,930 and other patents pending. FlexPen® is covered by US Patent Nos. 6,582,404, 6,004,297, 6,235,400 and other patents pending. © 2005-2012 Novo Nordisk 0212-00007333-1 2/2012

INDUSTRY TRENDS

The Policy-Driven Health Plan: A Road Map for Value-Based Reimbursement By James Evans Vice President of Financial and Network Management McKesson Health Solutions, Newton, MA

alue-based reimbursement represents a fundamental shift in the way health plans pay providers for care. Rather than creating incentives for providers to deliver high-quantity care—a downside of the fee-for-service (FFS) model—valuebased reimbursement aims at creating incentives for providers to achieve high-quality care.1,2 Because valuebased reimbursement is focused on driving healthcare value, improved population-based outcomes are an expected result.1,2 The challenge for health plans lies in the details. How do you implement value-based reimbursement? How do you get to the point at which you are reimbursing providers on the basis of value? Establishing a valuebased, policy-driven philosophy within the health plan can be a critical success factor.

Reimbursement Reform: Better, but More Complex, Solutions To deliver improved, more affordable healthcare solutions, health plans today are busy evaluating new benefit designs, piloting new care models, and weighing new reimbursement strategies. These efforts can be seen in initiatives such as patient-centered medical homes, accountable care organizations, and a variety of valuebased reimbursement strategies, including blended payments, bundled payments, partial capitation, and more. New reimbursement strategies typically require more complex, multifaceted contractual arrangements between the health plan and the care provider to achieve the desired outcome. That in itself poses challenges. The core systems in place today in most health plans were not designed to accommodate these payment and reimbursement innovations. Many of today’s claims systems, for example, will not readily incorporate the complex logic required to determine which provider contract and reimbursement methodology should apply to a given claim. It is also worth noting that manual claims editing and payment processes—the fallback process for claims that legacy systems cannot handle easily—are incompatible

446

American Health & Drug Benefits

with value-based reimbursement. To scale any type of reimbursement reform will require the efficiency and accuracy of technology. If aligned and integrated properly, automated solutions can ensure that medical, payment, and provider contract policies are applied consistently and effectively to ensure the greatest value.

A Road Map for Creating a Policy-Driven Environment The nature of a value-based reimbursement system involves a set of logical choices that ultimately align a reimbursement plan with a benefit plan, a care provider, and a medical event. Given the number of potential permutations, the process of mapping these elements properly can be extremely complex. To manage this complexity, health plans should strive for a policy-driven environment. This entails using a set of coordinated and aligned policies for payments, benefits, and medical events, each of which is developed to reinforce each other and is executed consistently across the systems involved with processing claims and payments. Through the use of these policies, it becomes possible to reimburse correctly, consistently, and in a manner that drives value. The following 4 steps are crucial in this process: • Establishing reimbursement policies • Aligning network design • Creating the benefits to complement the reimbursement model • Setting up contracts. Step 1. Establish Reimbursement Policies The path toward a policy-driven environment begins with policy design. A plan must establish policies for payments, benefits, and medical events. Consider the scenario depicted in Figure 1, involving a health plan male member with several medical conditions—diabetes, hypertension, and mildly elevated cholesterol. The health plan defines a set of benefits for members. To facilitate reimbursement, the health plan defines a set of policies that can be applied automatically to pay the care providers. In this case, the policies include:

www.AHDBonline.com

November/December 2012

Vol 5, No 7

INDUSTRY TRENDS

• Global payment for diabetes services provided within a medical home • FFS payments for diabetic testing supplies • FFS payments if the member is covered under preferred provider organization (PPO) and medical home plans • Partial capitation for referrals to a cardiologist for treatment of hypertension. This, however, is just the first step in a multistep process. Once the reimbursement policies have been defined, care networks need alignment with the reimbursement policies.

Step 2. Align Network Design Consider the “hybrid PPO” strategy modeled in Figure 2. It offers subscribers a variety of options for care, and each option is associated with certain networks and provider reimbursement policies. As seen in Figure 2, the benefit and policies in the top row are designed to support steerage to designated networks in the second row. Members’ out-of-pocket (OOP) costs are an important consideration when defining benefit options; they can act as a lever, steering members to the network that is most appropriate for the care they need. Similarly, health plans must design the supporting networks to deliver a variety of care options to meet the needs of both the members and the providers within those networks.

Figure 1 Creating the Reimbursement Policies

A 70-year-old man with diabetes, hypertension, and mildly elevated cholesterol levels

Diabetes medical home with enhanced reimbursement for specific services, and reimbursement for nonphysician and nonoffice services

• Global payment for diabetes services provided within the medical home

• FFS for diabetes testing supplies • FFS if member is covered under PPO and medical home

Reimbursement policies

• Partial capitation for referrals to cardiologist for hypertension

FFS indicates fee for service; PPO, preferred provider organization.

Figure 2 Aligning Supporting Networks with Reimbursement Policies

Benefit plan: hybrid PPO Benefit-based member incentives

15% penalty for PPO network

➤

Diabetes ACO network

Narrow network

General PPO market ➤

prospective budget • Global budget target (50%-90% of budget) • Performance target (10% of budget)

➤

• Shared-savings model based on

Alignment with reimbursement

$0 copay in quality network

➤

Steerage to supporting networks

10% reduction if in chronic care ACO

• FFS for specialty services • Capitation for primary services • Acute episodes • FFS + for preventive services

• FFS for all services • Higher price for members in exchange for choice • Low margin for health plan

ACO indicates accountable care organization; FFS, fee for service; PPO, preferred provider organization.

Vol 5, No 7

November/December 2012

www.AHDBonline.com

American Health & Drug Benefits

447

INDUSTRY TRENDS

Figure 3 Episodes of Care Associated with a Total Knee Replacement Procedure

Primary care provider

Specialist

Magnetic resonance imaging

Total knee replacement

Primary care provider

Physical therapist

➤ Episodes

Figure 4 Different Providers, Members, and Contracts

Contract 2 Hospital A • Shared savings (ie, ACO) • Membership panel • PCP services contracted from Central Bucks

Contract 1 Family Practice • Capitation for HMO • FFS for other • Pay for performance • Leverages orthopedic services from Abington

Contract 3 Hospital B • FFS episodic/ bundled for total knee replacement

Mary: 35-year-old woman Benefit plan: HMO Services: ear infection management

Sally: 50-year-old woman Program: medical home program, ACO panel Services: foot examination

Susan: 60-year-old woman Program: medical home program, ACO panel Services: insulin sensitivity test, vertigo consult

ACO indicates accountable care organization; FFS, fee for service; PCP, primary care provider.

In the third row of the figure, the policies defined in Step 1 are aligned with the care networks themselves. With a policy-driven approach, plans can associate different reimbursement policies with various networks quite easily. In addition, just as the OOP costs can act as an incentive to drive members toward certain optimized networks, the reimbursement policies can act as an incentive to drive providers toward certain networks. Each ensures that members and providers are moving toward an interaction that emphasizes a high-quality outcome at an affordable price.

Step 3. Create the Benefits to Complement the Reimbursement Model Consider the individual care events or encounters

448

American Health & Drug Benefits

that constitute a total knee replacement episode, as presented in Figure 3. In a traditional FFS model, a member pays something (a copay or a portion up to a deductible amount) at each encounter—for the primary care provider (PCP) visit, the specialist, and so on. In a value-based reimbursement approach, however, a health plan would develop member benefit models that would complement the different reimbursement models. These must be adjusted to ensure that the member does not have to pay for each visit within that episode of care. The reduced OOP expenses can act as incentives to draw members into various programs. Examples include: • FFS with a PPO: a standard deductible, with the member paying a set percentage after meeting that deductible • Consumer-directed health plan, with episode bundling: a health reimbursement arrangement covers the first $100; after that, care is covered at 100% • Episode PPO: a $100 flat fee for an entire episode of care.

Step 4. Set Up Contracts Finally, plans must set up contracts with varying types of care providers, as depicted in Figure 4. Depending on the network alignments set up in Step 2, the benefit models created in Step 3, and the contracts set up in Step 4, members could receive care in any of the 3 settings depicted in Figure 4. The way a provider is reimbursed then depends on which provider the member actually visits, and what type of service the member requires. For example, a member may visit a family practice; that provider’s reimbursements are outlined in Contract 1. However, the PCP may refer the member to hospital B for a total knee replacement procedure, in which case the provider’s reimbursement becomes part of the bundled payment governed by Contract 3. Conversely, the PCP may refer the member to an orthopedic surgeon at hospital B for something other than total knee replacement. According to Contract 3, that would indicate a FFS reimbursement strategy rather than the bundled payment method assigned to the total knee replacement procedure (including surgery, followon physical therapy, and more, as outlined in Figure 3).

www.AHDBonline.com

November/December 2012

Vol 5, No 7

INDUSTRY TRENDS

Figure 5 Creating and Refining the Policy-Driven Environment for Value-Based Reimbursement

➤

Reimbursement program design

• Design programs compatible with delivery reforms • Model based on proposed incentives • Create incentives for member participation

➤

Provider collaboration tools

Contract negotiation

• Negotiate contracts with select providers • Align networks around established objectives

➤

• Make data available to providers to optimize • Provide feedback on performance and quality • Improve provider tools to manage their own data

➤

Optimizing the Policy-Driven Environment Given the complexity of this environment—with its range of members, products, services, sites, providers, and contracts—a policy-driven approach to value-based reimbursement requires technologies that can work with very sophisticated selection criteria. The systems supporting a policy-based approach must be able to identify and act on the details associated with members, products, networks, contracts, and the care provided. Available technologies can provide the sophisticated services required to enable this kind of policy-driven environment, based on reimbursement program design tools that integrate with contract negotiation tools, which then integrate with reimbursement program execution tools, reporting tools, and provider collaboration tools. Working together, as depicted in Figure 5, these systems create a feedback loop that can continuously monitor, manage, and refine the processes supporting the health plan’s core activities. In deploying the technology that can enable a policy-driven environment, a plan must remember to integrate and align these systems to: • Reduce manual interpretation and intervention • Synchronize medical policy, payment policy, network design, and benefit design • Apply hybrid and overlapping reimbursement policies • Adjust reimbursement rates based on member attribution to products, programs, and providers • Model the impact of new network and payment models. If these requirements are not considered when deploying the technology to enable a policy-driven envi-

Vol 5, No 7

November/December 2012

• Adopt systems that automate contract and reimbursement initiatives at scale • Create differentiated payment for testing

Reimbursement program execution

Reporting and monitoring ➤

• Allow for visibility (dashboards, etc) to increase feedback speed • Align analytics solutions to rapidly gauge success

ronment, a plan will constrain its ability to gain the full spectrum of benefits that arise from this environment.

Facilitating Provider Transparency One component of the policy-driven environment depicted in Figure 5 worth calling out on its own is a set of provider collaboration tools. Health plans need tools that make it easy for providers to access information about policies and claims. If a plan wants to drive provider behavior toward value, the providers need to understand how their decisions about care affect their reimbursement. Exposing information about policies, claims editing rationales, and which claims have been paid (and which have not, and why), can help accomplish this.

A policy-driven approach to value-based reimbursement requires technologies that can work with very sophisticated selection criteria. Conversely, these same provider collaboration tools can create a way for providers to interact more easily with a plan’s core system. A provider portal, for example, would provide the insight they desire into policies and claims processing, and would also enable them to update information about their practices and their specialties. Enabling the providers to ensure that their information is up to date can streamline claims processing, while lessening the information collection burden that would

www.AHDBonline.com

American Health & Drug Benefits

449

INDUSTRY TRENDS

otherwise fall on the shoulders of personnel within the health plan itself.

The Advantages of a Policy-Driven Environment Ultimately, a policy-driven environment is critical to the realization of value-based reimbursement. By using the tools to enable a policy-driven environment, you can: • Automate the entire reimbursement lifecycle, from reimbursement policy design to contracting, claim editing, pricing, and optimization • Reduce manual data entry by automating the propagation of provider data from enrollment to contracting, and by automating the loading of executed contracts into the reimbursement engine • Increase the autoadjudication rate of claims by improving the integrity of data for claims processing • Increase payment accuracy per contract intent, because the claims editing and pricing processes can access network, provider, and contract data in detail, facilitating accurate price alignment • Automate the alignment of product, network, and reimbursement designs to enforce referential integrity and the rule-driven implementation of policies • Support provider- and contract-level variations in policies, gaining the flexibility to institute excep-

tions—a change in episode definition, for example— without losing the ability to codify and electronically transmit a claim that will be adjudicated properly within the system • Decrease information technology integration costs by creating a single point of integration for claims editing, provider selection, episode bundling, and pricing. With the ability to create and manage products, contracts, and reimbursements more efficiently and effectively, a policy-driven environment will enable the health plan to create the necessary conditions for steering members and providers toward the highest quality of care, delivered at the most affordable price. Such an environment provides integrated mechanisms for creating and refining incentives, for mapping benefits to members, providers to networks, and reimbursements to contracts. In the end, based on our experience in the healthcare industry, this translates to better outcomes for everyone. ■ Author disclosure statement Mr Evans has no conflict of interest to report.

References 1. Porter ME. A strategy for health care reform—toward a value-based system. N Engl J Med. 2009;361:109-112. 2. Porter ME. What is value in health care? N Engl J Med. 2010;363:2477-2481.

Information for Authors Manuscripts submitted to American Health & Drug Benefits (AHDB) must be original and must not have been published previously, either in print or in electronic form. Manuscripts cannot be submitted elsewhere while under consideration by AHDB, and must adhere to the format described in the full Information for Authors available at www.AHDBonline.com. All manuscripts undergo peer review, and acceptance is based on that review. If accepted, authors will be notified of any recommended revisions. Routine editorial changes are made to conform to house style, following the AMA Manual of Style, 10th ed. (New York, NY: Oxford University Press, 2007). Time from submission to publication is generally 3 to 5 months. COPYRIGHT/DISCLOSURE Authors must sign a Copyright Transfer Form, assigning all copyrights to Engage Healthcare Communications, LLC, publisher of AHDB, as well as a Financial Disclosure Form, disclosing any financial interests or potential conflict of interest involving the materials discussed in the manuscript. MANUSCRIPT FORMAT See complete Information for Authors at www.AHDBonline.com. PERMISSIONS Authors must secure a written permission from the original publisher for any previously published (online or in print) table or figure. Provide the source with each element. Submit the manuscript electronically to: editorial@engagehc.com. For assistance, call 732-992-1536.

450

American Health & Drug Benefits

www.AHDBonline.com

November/December 2012

Vol 5, No 7

NOW AVAILABLE

KYPROLIS.com

BUSINESS

Original research

Hematologic Complications, Healthcare Utilization, and Costs in Commercially Insured Patients with Myelodysplastic Syndrome Receiving Supportive Care annette Powers, PharmD, MBa; claudio Faria, PharmD, MPh; Michael s. Broder, MD, Mshs; eunice chang, PhD; Dasha cherepanov, PhD Background: Myelodysplastic syndrome (MDS) is rare in people aged <50 years. Most patients with this disorder experience progressive worsening of blood cytopenias, with an increasing need for transfusion. The more advanced and severe the disorder, the greater the risk that it will progress to acute myeloid leukemia. Therapy is typically based on the patient’s risk category, age, and performance status. Supportive care alone is a major option for lower-risk, older patients with MDS or those with comorbidities. The only potentially curative treatment option is hematopoietic stem-cell transplantation, which is typically used to treat high-risk, younger patients. Objective: To describe and compare the hematologic complications, healthcare utilization, and costs of supportive care in patients with MDS aged <50 years and in older patients aged ≥50 years. Methods: Using the i3/Ingenix LabRx claims database, this retrospective study included patients who were continuously enrolled (ie, 6 months preindex through 1 year postindex) in the study and who had an initial claim of MDS (index date) between February 1, 2007, and July 31, 2008. Patients treated with hypomethylating agents or thalidomide analogues were excluded. Claims included information on office visits, medical procedures, hospitalizations, drug use, and tests performed. The hematologic complications, costs, and utilization analyses were stratified by age into 2 age-groups—patients aged <50 years and those aged ≥50 years. The MDS-related diagnoses, utilization, and costs were analyzed postindex. The data used in this study spanned the period from August 1, 2006, to July 31, 2009. Results: We identified 1133 newly diagnosed patients with MDS who received supportive care only during the study period; of these, 19.5% were younger than age 50 years. These younger patients included more females (62.0% vs 52.5%; P = .011) and had fewer comorbidities (mean Charlson comorbidy index, 1.2 vs 2.4; P <.001) and physician office visits than those aged ≥50 years. Postindex, compared with the older patients, the younger patients had less use of erythropoietin therapy and fewer transfusions, anemia diagnoses, and potential complications of neutropenia and pneumonia diagnoses; however, more diagnoses of neutropenia and of decreased white blood cell counts were seen in the younger patients than in the older patients (P ≤.034 for all comparisons). Furthermore, younger patients had fewer mean office visits in the postindex period than older patients (17.5 vs 24.2, respectively; P <.001) and fewer hospitalizations (32.1% vs 44.6%, respectively; P = .004), but they had a longer (although not statistically significant) mean length of hospital stay (21 vs 14 days, respectively; P = .131). Mean total healthcare charges were $96,277 (median, $21,287) in younger patients compared with $84,102 (median, $39,402) in older patients, although this difference, too, was not significant. Conclusions: MDS is associated with frequent and prolonged hospitalizations, frequent outpatient visits, and high costs in younger and in older patients who are receiving supportive care. Although this study shows that younger patients aged <50 years do not have significantly higher costs overall, a small proportion may have a higher healthcare utilization and cost-related burden of MDS than patients aged ≥50 years.

Stakeholder Perspective, page 465

Am Health Drug Benefits. 2012;5(7):455-465 www.AHDBonline.com Disclosures are at end of text

Dr Powers is Senior Director of Health Economics and Outcomes Research, and Dr Faria is Director of Health Economics and Outcomes Research, Eisai Inc, Woodcliff Lake, NJ. Dr Broder is President, Dr Chang is Chief Statistician, and Dr Cherepanov is Associate Director, Outcomes Research, Partnership for Health Analytic Research, LLC, Beverly Hills, CA.

Vol 5, No 7

November/December 2012

www.AHDBonline.com

American Health & Drug Benefits

455

BUSINESS

yelodysplastic syndrome (MDS) encompasses a heterogeneous group of clonal disorders of hematopoiesis and is characterized by dysplastic morphology of marrow and blood cells, ineffective hematopoiesis, and peripheral blood cytopenias.1,2 Most patients with MDS experience progressive worsening of blood cytopenias, with an increasing need for transfusion.2 These patients also have an increasing number of potentially fatal infections and hemorrhagic complications.2 The more advanced and severe the MDS is, the greater the risk that the disease will progress to acute myeloid leukemia (AML).3 The disease may be classified into 1 of 5 subtypes—refractory anemia, refractory anemia with ringed sideroblasts (RARS), refractory anemia with excess of blasts (RAEB), RAEB in transformation (RAEB-T), or chronic myelomonocytic leukemia.3 Approximately 5% to 15% of the relatively lower-risk patients with refractory anemia/RARS transform to AML; by contrast, 40% to 50% of the high-risk patients with RAEB/RAEB-T transform to AML.3 The therapeutic options that are tailored for specific MDS subgroups are typically based on factors such as the patient’s risk category, age, and performance status.3,4 The National Comprehensive Cancer Network (NCCN) Clinical Practice Guidelines in Oncology recommend that all patients with MDS receive supportive care,3 which includes blood transfusions, erythropoietin with or without granulocyte colony-stimulating factor, iron chelation therapy, and prophylactic antibiotics.4,5 Other therapies indicated for the treatment of patients with MDS include the thalidomide analogue lenalidomide and the hypomethylating agents decitabine and 5azacytidine.3,4 The only potentially curative treatment option is hematopoietic stem-cell transplantation, which is typically used to treat younger, high-risk patients.3,4 Supportive care alone remains a leading option for the treatment of lower-risk, older patients with MDS or those with comorbidities.3,4 Data on the distribution of MDS in the general population are inconsistent, possibly because of misdiagnoses and/or underreporting of the disease.6,7 The most recent estimates of the annual incidence of MDS in the United States range from 3.3 to 5.0 per 100,000 persons.3,7,8 Some studies indicate that the median age of patients with MDS is approximately 65 years, whereas others note that more than 70% of cases occur in patients aged ≥70 years in the United States.3,6,9 The incidence of MDS in individuals aged ≥70 years is between 22 and 45 per 100,000 persons and increases with age.3,6,9-11 Less than 10% of patients with MDS are aged <50 years; therefore, little is known about this disease in this younger age-group, particularly among patients who receive supportive care only.6,11,12 Some data suggest that

456

American Health & Drug Benefits

KEY POINTS ➤

➤

The more advanced and severe the myelodysplastic syndrome (MDS) is, the greater the risk of progression to acute myeloid leukemia. Therapy is currently based on risk category, age, and performance status. In the United States, the majority of newly diagnosed patients with MDS receive only supportive care, although for younger patients at high-risk, hematopoietic stem-cell transplantation is potentially the only curative option. This analysis compares the hematologic complications, healthcare utilization, and cost of care between patients with MDS aged <50 years and those aged ≥50 years who receive supportive care only. Although the younger patients had fewer office visits, they had longer mean length of hospital stay than the older group (21 vs 14 days, respectively). Mean total healthcare charges were $96,277 in younger patients compared with $84,102 in older patients. Based on this study, approximately 20% of patients with MDS are under age 50 years. The results of this study suggest that a small proportion of younger patients with MDS who receive supportive care only may have a higher healthcare utilization and cost-related burden of MDS than older patients with this condition.

younger patients with MDS have less aggressive disease.12,13 We compared hematologic complications, healthcare utilization, and costs in patients aged <50 years and in those aged ≥50 years who were newly diagnosed with MDS and received supportive care only.

Methods This study was a retrospective cohort analysis using data from the i3/Ingenix LabRx database, which is a Health Insurance Portability and Accountability Act– compliant administrative claims database of 8 million to 10 million covered lives from all major regions of the United States. The database contains deidentified adjudicated pharmacy and medical claims submitted for payment by providers, healthcare facilities, and pharmacies. Claims included information on physician office visits, medical procedures, hospitalizations, drugs dispensed, and on the tests that were performed. In this database, charges are reported, but paid claims and costs are not (although charges and costs are conceptually different, we refer to charges as costs in the discussion of the results, for convenience). Data used in this study spanned the period from August 1, 2006, to July 31, 2009.

www.AHDBonline.com

November/December 2012

Vol 5, No 7

Burden of MDS in Patients Receiving Supportive Care

Table 1 List of CPT©, GPI, HCPCS, and ICD-9-CM Codes by Study Measure Study measure

CPT©, GPI, HCPCS, or ICD-9-CM code

Acute myeloid leukemia

ICD-9-CM: 205.0x, 205.2x-205.9x, 206.0x, 206.2x-206.9x, 207.0x, 207.2, 208.0x, 208.2x-208.9x

Anemia Anemia diagnosis

ICD-9-CM: 281.9, 283.9, 284.8, 284.9, 285.0, 285.2x, 285.9, V78.1

Use of erythropoietin

HCPCS: J0885, J0881, Q0137, Q0136, J0880; GPI: 82-40-10

Use of iron chelation therapy

Deferoxamine (GPI: 93-00-00-20-10) or deferasirox (GPI: 93-10-00-25-00)

Bone marrow biopsy

ICD-9-CM: 41.31, 41.38, 41.98; CPT: 38220, 38221

Neutropenia Neutropenia

ICD-9-CM: 288.00, 288.04, 288.09, 289.4

Febrile neutropenia diagnosis

ICD-9-CM: 780.61

Decreased white blood cell count

ICD-9-CM: 288.5x

Use of granulocyte colony-stimulating factors

HCPCS: J1440, J1441, J2505; GPI: 82-40-15

Pancytopenia

ICD-9-CM: 284.1

Potential complications of neutropenia Pneumonia

ICD-9-CM: 481, 482.xx, 485, 486

Unspecified fever

ICD-9-CM: 780.60

Thrombocytopenia

ICD-9-CM: 287.3x, 287.5

Transfusions

HCPCS: P9010, P9016, P9021, P9022, P9038, P9039, P9040, P9057, P9058, P9019, P9020, P9031, P9032, P9033, P9034, P9035, P9036, P9037, P9052, P9053, P9055; CPT: 36430; ICD-9-CM procedure code: 99.0x

CPT indicates Current Procedural Terminology©; GPI, generic product identifier; HCPCS, Healthcare Common Procedure Coding System; ICD-9-CM, International Classification of Diseases, Ninth Revision, Clinical Modification.

Study Population This study included patients with a first diagnosis of MDS between February 1, 2007, and July 31, 2008 (ie, the identification period). MDS was identified using International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) diagnosis codes of 238.72 through 238.75. The first date of a medical claim with an MDS diagnosis in any diagnosis field in the identification period was defined as the index date. Patients were followed for 1 year after the index date. To examine a more homogeneous group of patients with MDS in our final analytic cohort, we included newly diagnosed patients with MDS who received supportive care only; these patients had no claims for hypomethylating agents

Vol 5, No 7

November/December 2012

or for thalidomide analogues (ie, decitabine, 5-azacytidine, or lenalidomide) in the postindex period. Patients were excluded from the study if they (1) had a diagnosis of MDS in the 6-month preindex period, (2) had a diagnosis of AML (ICD-9-CM 205.0x, 205.2x205.9x, 206.0x, 206.2x-206.9x, 207.0x, 207.2, 208.0x, 208.2x-208.9x) in the 6-month preindex period, or (3) were not continuously enrolled in the 6-month preindex and the 1-year postindex periods.

Measures Baseline variables in the study were patient demographics, bone marrow biopsy, number of physician office visits, number of emergency department visits and hospi-

www.AHDBonline.com

American Health & Drug Benefits

457

BUSINESS

Study Cohort Selection: Newly Diagnosed MDS Patients Figure of All Ages Who Received Supportive Care Only 3327 patients with MDS diagnosis in identification period (2/1/2007-7/31/2008)

➤

–748 patients not newly diagnosed

➤

N = 2579

➤

–164 patients with AML diagnosis in preindex period

➤

N = 2415

➤

–1206 patients not continuously enrolled in preindex and postindex periods

➤

N = 1209 Newly diagnosed patients –76 patients with ➤

MDS treatment in postindex period

➤

N = 1133 Newly diagnosed patients, supportive care only ➤

➤

221 patients aged <50 yrs

912 patients

aged ≥50 yrs

talizations, length of stay among patients with hospitalizations, and total healthcare charges. We also calculated the adapted Charlson comorbidity index at baseline, which is a clinical comorbidity index designed to be used with select ICD-9-CM diagnoses and procedure codes.14,15 The primary outcomes were AML diagnosis and mean number of days to first AML diagnosis (among patients with AML diagnoses). Other outcomes included number of transfusions, number of anemia diagnoses, number of neutropenia diagnoses, potential complications of neutropenia, number of thrombocytopenia diagnoses, number of pancytopenia diagnoses, and number of

458

American Health & Drug Benefits

decreased white blood cell count diagnoses.2 We also calculated the number of physician office visits, hospitalizations, and emergency department visits; the length of stay among patients with hospitalizations; and the total healthcare charges. The MDS-related charges were estimated by adding charges from medical claims with a primary diagnosis related to MDS or to AML and charges from pharmacy claims for the treatment of MDS. Table 1 lists the codes used to derive the study measures.

Analyses All pharmacy and inpatient and outpatient medical claims were reviewed in the 6-month preindex period to derive the baseline variables and in the 1-year postindex period to derive the study outcomes (bone marrow biopsies were identified in the preindex and postindex periods). Preindex and postindex analyses were stratified by 2 age cohort groups: patients aged <50 years and patients aged ≥50 years. We report means, medians, and standard deviations (SDs) for continuous variables, whereas patient counts and percentages are reported for categorical variables. Appropriate statistical tests (ie, t-tests for continuous variables and chi-square tests for categorical variables) were used to compare study measures across age cohorts. All statistical analyses were performed using SAS version 9.2 (SAS Institute, Cary, NC). Results Cohort Selection We identified 3327 patients with an MDS diagnosis in the identification period (between February 1, 2007, and July 31, 2008). Of these patients, 748 were not newly diagnosed, 164 had an AML diagnosis in the preindex period, and 1206 patients were not continuously enrolled in both the preindex and postindex periods. After exclusion of these 2118 patients, there were 1209 newly diagnosed patients. For our final cohort of newly diagnosed patients with supportive care only, 76 patients who were treated with hypomethylating agents and thalidomide analogue in the postindex period were removed from the data, resulting in the final analytic sample size of 1133 patients (Figure). Among these 1133 patients with newly diagnosed MDS, 123 (10.9%) had a first diagnosis of low-grade MDS lesions (ICD-9-CM code 238.72), 36 (3.2%) had a diagnosis of high-grade MDS lesions (238.73), 18 (1.6%) had a diagnosis of MDS with 5q deletion (238.74), and 956 (84.4%) patients had a diagnosis of MDS unspecified (238.75). There were no differences in these distributions between the 2 age cohorts (P = .141). Baseline Patient Characteristics At baseline, the mean age in this cohort was 62.9 years

www.AHDBonline.com

November/December 2012

Vol 5, No 7

Burden of MDS in Patients Receiving Supportive Care

Demographics, Comorbidities, Bone Marrow Biopsy, and Healthcare Utilization and Charges in the 6-Month Table 2 Preindex Period in Patients Receiving Supportive Care Patients aged <50 yrs (N = 221)

Patients aged â&#x2030;Ľ50 yrs (N = 912)

Total (N = 1133)

P value

Mean age, yrs (SD)

39.1 (9.4)

68.7 (10.8)

62.9 (15.8)

N/A

Female, N (%)

137 (62.0)

479 (52.5)

616 (54.4)

.011a .036a

Region, N (%) East

36 (16.3)

96 (10.5)

132 (11.7)

Midwest

46 (20.8)

198 (21.7)

244 (21.5)

South

114 (51.6)

465 (51.0)

579 (51.1)

West

25 (11.3)

153 (16.8)

178 (15.7)

1.2 (2.1)

2.4 (2.7)

2.1 (2.6)

<.001a

Bone marrow biopsy,b N (%)

113 (51.1)

413 (45.3)

526 (46.4)

.118

Mean physician office visits, N (SD; median)

8.2 (7.8; 6)

10.5 (9.1; 9)

10.1 (9.0; 8)

<.001a

164 (74.2)

646 (70.8)

810 (71.5)

.52

41 (18.6)

171 (18.8)

212 (18.7)

9 (4.1)

48 (5.3)

57 (5.0)

7 (3.2)

47 (5.2)

54 (4.8)

57 (7.8; 7.9)

266 (9.0, 15.0)

323 (8.8; 14.0)

.417

208 (94.1)

870 (95.4)

1078 (95.1)

.264

7 (3.2)

14 (1.5)

21 (1.9)

6 (2.7)

28 (3.1)

34 (3.0)

Mean total healthcare charges, $ (SD; median)

30,177 (53,550; 9622)

31,832 (64,658; 12,248)

31,509 (62,627; 12,034)

.693

Mean medical charges, $ (SD; median)

28,248 (52,264; 8413)

29,581 (63,660; 9790)

29,321 (61,584; 9616)

.745

1929 (4451; 428)

2251 (4714; 1041)

2188 (4664; 895)

.358

Charlson comorbidity index, mean (SD)

Hospitalizations, N (%)

Length of stay among patients with hospitalizations, days (mean; SD) Emergency department visits, N (%)

Mean pharmacy charges, $ (SD; median) a

P <.05. Identified in the preindex or postindex period. N/A indicates not applicable; SD, standard deviation. b

(SD, 15.8), with 19.5% (N = 221) of the sample aged <50 years and 80.5% (N = 912) of the sample aged â&#x2030;Ľ50 years (Table 2). Mean ages within the 2 cohorts were 39.1 years (SD, 9.4) and 68.7 years (SD, 10.8) in the younger and older age-groups, respectively. There was a significant dif-

Vol 5, No 7

November/December 2012

ference between the 2 age cohorts in the proportion of females (62% vs 52.5% in the younger vs the older agegroups, respectively; P = .011) and in the distribution across US census regions (P = .036). Based on the mean Charlson comorbidity index, the

www.AHDBonline.com

American Health & Drug Benefits

459

BUSINESS

group aged <50 years had fewer comorbid conditions than the group aged ≥50 years (1.2 vs 2.4, respectively; P <.001). There was no significant difference between the 2 age cohorts in the proportion of bone marrow biopsies (51.1% vs 45.3% in the younger and older agegroups, respectively).

Preindex Healthcare Utilization and Costs In terms of baseline (preindex) healthcare utilization and costs (Table 2), no significant differences were seen between the 2 age cohorts, except for the mean number of physician office visits. There were fewer physician office visits in the younger age-group than in the older age-group (mean, 8.2 vs 10.5, respectively; P <.001). The younger and older groups had a similar proportion of hospitalizations (25.9% vs 29.3% for ≥1 hospitalizations; P = .52) and a similar mean length of hospital stay (7.8 vs 9.0 days; P = .417). A similar proportion of younger and older patients had at least 1 emergency department visit (5.9% vs 4.6%; P = .264). The mean total 6-month preindex healthcare costs were $30,177 (SD, $53,550; median, $9622) in the younger patients and $31,832 (SD, $64,658; median, $12,248) in the older patients (P = .693). Postindex MDS-Related Diagnoses, Healthcare Utilization, and Costs As shown in Table 3, in the year after MDS diagnosis, the crude incidence of AML diagnosis was similar in the 2 age-groups—9% of patients aged <50 years versus 5.7% of patients aged ≥50 years (P = .067). There was no significant difference in the mean number of days to first AML diagnosis in patients who were diagnosed with AML between the 2 age-groups (43.8 vs 74.3 days; P = .214). The younger patients aged <50 years had proportionally fewer transfusions than those aged ≥50 years (10.8% vs 14.9% had ≥1 transfusions; P = .034). The younger patients also had a significantly lower proportion of anemia diagnoses (46.6% vs 68.1%; P <.001) and significantly less erythropoietin use (10.9% vs 28.9%; P <.001) than the older patients. The proportion of patients with iron chelation therapy use was similar in the 2 groups (1.4% vs 0.8%, respectively; P = .4). The proportion of neutropenia diagnoses was significantly higher in the younger group than in the older group (24.0% vs 17.1%, respectively; P =.018), but the difference in the use of granulocyte colony-stimulating factor was not significant (8.6% vs 6.5%, respectively; P = .262). Furthermore, fewer potential complications of neutropenia were seen in the younger age-group than in the older age-group (7.2% vs 14.1%, respectively; P = .006) and significantly fewer pneumonia diagnoses were

460

American Health & Drug Benefits

observed in the younger age-group (5.4% vs 12.4%, respectively; P = .003). The number of unspecified fever diagnoses was similar between the 2 age-groups (2.7% for the younger group vs 3.4% for the older group; P = .608), and the use of outpatient pharmacy intravenous antibiotics was also similar (0.5% vs 0.3%, respectively; P = .781). The proportion of thrombocytopenia diagnoses was numerically but insignificantly higher in the younger group than in the older group—25.3% versus 22.3%, as was pancytopenia diagnoses (13.1% vs 12.6%, respectively); decreased white blood cell count diagnoses were the only significant difference, with 13.6% in the younger group and 6.5% in the older group (P <.001). There were no significant differences in MDS-related costs between the 2 age cohorts, although the mean costs were higher in the younger age-group—$35,888 (SD, $139,081; median, $2626) versus $25,435 (SD, $81,866; median, $4717) for the older group (P = .284). Overall, patients aged <50 years had significantly less erythropoietin use (P <.001) and significantly fewer transfusions (P = .034), anemia diagnoses (P <.001), complications of neutropenia (P = .006), and pneumonia diagnoses (P = .003) than patients aged ≥50 years; however, there was a higher percentage of neutropenia (P = .018) and decreased white blood cell count diagnoses in younger patients than in older patients (P <.001).

Postindex Overall Healthcare Utilization and Costs A significant difference was seen in overall healthcare utilization in the 1-year postindex period (Table 4). Patients aged <50 years had a significantly lower mean number of physician office visits than patients aged ≥50 years (17.5 vs 24.2; P <.001), and a lower proportion of younger patients had at least 1 hospitalization (32.1% vs 44.6%; P < .004). However, the mean length of stay among patients with hospitalizations was longer in the younger age-group than in the older age-group (21 vs 14 days), although this difference was not statistically significant (P = .131). The proportion of patients that had at least 1 emergency department visit was similar in the 2 age-groups (8.6% vs 8.5%; P = .74). As shown in Table 4, there was no significant difference in mean total healthcare costs in the postindex period between the 2 age cohorts (P = .473), but there was a numerical difference, with a mean cost of $96,277 (SD, $240,854; median, $21,287) in younger patients compared with a mean cost of $84,102 (SD, $149,877; median, $39,402) in older patients. The mean total healthcare costs were primarily driven by mean medical charges among younger and older patients: $91,435 (SD, $237,723; median, $18,526) for younger patients versus $78,612 (SD, $146,631; median,

www.AHDBonline.com

November/December 2012

Vol 5, No 7

Burden of MDS in Patients Receiving Supportive Care

MDS-Related Diagnoses, Healthcare Utilization and Charges in the 1-Year Postindex Period in Table 3 Patients Receiving Supportive Care Patients aged <50 yrs (N = 221)

Patients aged ≥50 yrs (N = 912)

Total (N = 1133)

P value

AML diagnosis, N (%)

20 (9.0)

52 (5.7)

72 (6.4)

.067

Mean days to first AML diagnosis among patients with MDS diagnosis (SD; median)

43.8 (78.8; 4)

74.3 (96.9; 22)

65.8 (92.7; 16)

.214 .034a

Number of transfusions, N (%) 0

197 (89.1)

776 (85.1)

973 (85.9)

8 (3.6)

80 (8.8)

88 (7.8)

16 (7.2)

56 (6.1)

72 (6.4)

103 (46.6)

621 (68.1)

724 (63.9)

<.001a

24 (10.9)

264 (28.9)

288 (25.4)

<.001a

3 (1.4)

7 (0.8)

10 (0.9)

53 (24.0)

156 (17.1)

209 (18.4)

.018a

Decreased white blood cell count diagnosis, N (%)

30 (13.6)

59 (6.5)

89 (7.9)

<.001a

Use of granulocyte colony-stimulating factor, N (%)

19 (8.6)

59 (6.5)

78 (6.9)

.262

16 (7.2)

129 (14.1)

145 (12.8)

.006a

Pneumonia diagnosis, N (%)

12 (5.4)

113 (12.4)

125 (11.0)

.003a

Unspecified fever diagnosis, N (%)

6 (2.7)

31 (3.4)

37 (3.3)

.608

Use of outpatient pharmacy intravenous antibiotics, N (%)

1 (0.5)

3 (0.3)

4 (0.4)

.781

Thrombocytopenia diagnosis, N (%)

56 (25.3)

203 (22.3)

259 (22.9)

.328

Pancytopenia diagnosis, N (%)

29 (13.1)

115 (12.6)

144 (12.7)

.837

Anemia diagnosis, N (%) Erythropoietin use, N (%) Iron chelation therapy use, N (%) Neutropenia diagnosis, N (%)

Potential complications of neutropenia, N (%)

Mean MDS-related charges, $b (SD; median)

35,888 25,435 27,474 (139,081; 2626) (81,866; 4717) (95,761; 4485)

.284

P <.05. MDS-related costs include charges on medical claims with a primary diagnosis of conditions listed in this table and charges on pharmacy claims for medications listed in this table. AML indicates acute myeloid leukemia; MDS, myelodysplastic syndrome; SD, standard deviation.

$32,782) for older patients. Although the mean postindex total healthcare charges and medical charges were higher in the younger group than in the older group, the medians of these charges were higher in the older group.

Discussion Based on an analysis of a commercial claims database, our study indicates that MDS is associated with frequent and prolonged hospitalizations, frequent outpatient visits, and high charges in younger and in older patients who

Vol 5, No 7

November/December 2012

are receiving supportive care. Although MDS is often referred to as a “disease of the elderly,”3,9,10 this study shows that a substantial percentage of patients with MDS are not elderly, with up to 20% of patients aged <50 years. The greater representation of younger patients in our study allowed us to examine and demonstrate that younger patients may have higher healthcare utilization and higher costs on average. Although this study was not designed to examine the underuse of diagnostic tests or of treatments, we found evidence of low use of bone

www.AHDBonline.com

American Health & Drug Benefits

461

BUSINESS

Overall Healthcare Utilization and Total Charges in the 1-Year Postindex Period in Patients Table 4 Receiving Supportive Care

Mean physician office visits (SD; median)

Patients aged <50 yrs (N = 221)

Patients aged â&#x2030;Ľ50 yrs (N = 912)

Total (N = 1133)

P value

17.5 (16.9; 12)

24.2 (16.3; 21)

22.9 (16.6; 20)

<.001a .004a

Hospitalizations, N (%) 0

150 (67.9)

505 (55.4)

655 (57.8)

37 (16.7)

188 (20.6)

225 (19.9)

12 (5.4)

104 (11.4)

116 (10.2)

22 (10.0)

115 (12.6)

137 (12.1)

71 (21; 37.3)

407 (14; 21.9)

478 (15; 24.9)

Length of stay among patients with hospitalizations, days (mean; SD)

.131 .74

Emergency department visits, N (%) 0

202 (91.4)

835 (91.6)

1037 (91.5)

11 (5.0)

37 (4.1)

48 (4.2)

8 (3.6)

40 (4.4)

48 (4.2)

Mean total healthcare charges, $ (SD; median)

96,277 (240,854; 21,287)

84,102 (149,877; 39,402)

86,477 (171,391; 34,690)

.473

Mean medical charges, $ (SD; median)

91,435 (237,723; 18,526)

78,612 (146,631, 32,782)

81,113 (168,261; 29,912)

.443

4841 (8407; 1314)

5490 (8573; 3048)

5363 (8541; 2749)

.311

Mean pharmacy charges, $ (SD; median) a

P <.05. SD indicates standard deviation.

marrow biopsy (Table 2) and potential undertreatment with hypomethylating agents or with thalidomide analogues (Figure). We estimated that almost 20% of patients in this commercial plan population were aged <50 years, which is almost twice the previously reported prevalence of MDS in that age-group.11,12 In a recent study, Cogle and colleagues demonstrated that cancer registries may have a high number of uncaptured cases of MDS, possibly because of misdiagnoses and/or underreporting of the disease, and that the annual incidence of MDS may be as high as 75 per 100,000 persons aged â&#x2030;Ľ65 years.7 Hence, our findings indicate that the incidence of MDS in the younger age-group (ie, <50 years) may be higher than expected, which highlights the importance of continuing to examine the impact of MDS in this age-group. One possible reason for underreporting of MDS is the low use of diagnostic tests. Our study indicates that cases

462

American Health & Drug Benefits

of MDS may be insufficiently diagnosed, because only approximately half (46.4%) of patients newly diagnosed with MDS who are receiving supportive care have a claim for a bone marrow biopsy (Table 2), an estimate that could be considered low, given that the NCCN guidelines recommend using this procedure.3 In addition, our results show that physicians may not follow other aspects of treatment guidelines, which is evidenced by the relatively low use of thalidomide analogues and hypomethylating agents in our study sample (Figure). Although the NCCN guidelines support the treatment of MDS with thalidomide analogues and hypomethylating agents,3 most newly diagnosed MDS patients in our study received supportive care only (1133 of 1209 total patients = 93.7%). Only 76 (6.2%) of newly diagnosed MDS patients in our study received treatment with decitabine, 5-azacytidine, or lenalidomide in the postindex period. We also found only 12 (1.1%) patients who were receiving allogeneic stem-cell

www.AHDBonline.com

November/December 2012

Vol 5, No 7

Burden of MDS in Patients Receiving Supportive Care

transplant treatment in our study (ICD-9-CM codes 41.05 and 41.08; results not shown). These findings support the results previously reported by Van Bennekom and colleagues on the patterns of treatment among patients with recently diagnosed MDS in a national, disease-based, observational registry between 2006 and 2008.16 Van Bennekom and colleagues reported that only 24% of patients who were recently diagnosed with MDS had received disease-modifying treatments since diagnosis, including 5-azacytidine (9%), decitabine (7%), lenalidomide (6%), or multiple agents (2%), compared with 58% of patients who received supportive therapy.16 Consistent with previous studies,16,17 our results emphasize that most newly diagnosed commercially insured patients with MDS in the United States receive supportive therapy after their initial diagnosis, whereas relatively few receive other therapies. More appropriate treatment for MDS may therefore reduce the burden associated with this condition, such as progression to transfusion dependence that often occurs with supportive care.18 Despite receiving supportive care only, the average total annual healthcare charges for patients in our study were high (>$86,000), with higher mean costs for younger patients ($96,277 vs $84,102; P = .473). There was no evidence that the higher total healthcare costs in younger patients were associated with age-related differences in baseline comorbidity; the mean Charlson comorbidity index was 1.2 in patients aged <50 years compared with 2.4 in patients aged ≥50 years (P <.001). Similarly, this postindex difference in the charges between the 2 age-groups was not associated with baseline mean total healthcare charges ($30,177 in the group aged <50 years vs $31,832 in the group aged ≥50 years; P = .693) or with healthcare utilization, because both were numerically higher in the group of older patients. Similarly, the higher total annual healthcare charges in the younger patient cohort are likely not to be primarily driven by differences in MDS-related diagnoses and healthcare utilization, because anemia was more common in the older patients (aged ≥50 years) than in the younger patients (aged <50 years), as were erythropoietin use, blood transfusions, complications of neutropenia, and diagnoses of pneumonia (all significant differences), whereas diagnoses of neutropenia and decreased white blood cell count were more common in the younger patients than in the older patients. MDS-related costs made up approximately 32% of total healthcare charges, with numerically higher mean costs in the younger age-group. This study included only claims with specific primary diagnoses as MDSrelated charges; a more expansive definition would likely have resulted in a greater proportion of costs being related to MDS.

Vol 5, No 7

November/December 2012

Our study shows that although the mean healthcare costs are greater in the younger age-group, the median MDS-related and the total healthcare costs show an opposite trend, with median healthcare costs being lower in younger patients than in older patients. That is, 50% of younger patients have total healthcare costs of ≥$21,287 (and medical costs of ≥$18,526), and 50% of the older patients have costs of ≥$39,402 (and medical costs of ≥$32,782). One explanation for this finding may be the skewness of the total healthcare costs distribution. These results indicate that the distribution of healthcare costs in the 2 cohorts are skewed toward lower charges, especially in the younger age-group, with a few patients in this group accumulating the highest costs. Neutropenia, a complication strongly associated with increased hospitalization,19 was significantly more prevalent in patients aged <50 years (24%) than in patients aged ≥50 years (17.1%; P = .018). It may be that the younger patients are using more expensive services than older patients, given the longer mean length of stay among younger hospitalized patients (21 days) compared with older patients (14 days; P = .131). Hence, a small group of very expensive younger patients may be considerably increasing the mean MDS-related costs and therefore the total healthcare costs. Supportive care of MDS typically includes red blood cell transfusions, a treatment that most patients with MDS become dependent on given the noncurative nature of the disease.9,18,20,21 Studies have shown that transfusion dependence not only negatively affects morbidity and mortality, but also significantly increases costs in patients with MDS compared with patients with transfusion independence.18,21 For instance, Frytak and colleagues compared the economic burden of patients with MDS who are aged ≥55 years and with either transfusion independence or dependence, and found that the MDS transfusion-dependent cohort had significantly higher mean annual costs (pharmacy, $4457 vs $2926; medical, $50,663 vs $17,469; total, $51,066 vs $19,811 per patient annually).21 Studies have shown that transfusion requirements may be greater in elderly patients than in younger patients.20,21 Although Frytak and colleagues examined only patients aged ≥55 years, the MDS transfusion-dependent patients were significantly older than the MDS transfusion-independent patients.21 We found that a significantly smaller proportion of younger patients than older ones had ≥1 transfusions, and this difference may have contributed to the lower median healthcare charges in the younger patients. The younger cohort in our study also had a higher proportion of females, a population that, in general, may have less transfusion dependence21; this sex differential in our

www.AHDBonline.com

American Health & Drug Benefits

463

BUSINESS

study cohorts may be another factor associated with the lower median healthcare charges in the younger group.

Limitations The use of insurance claims data for research presents unique challenges.22 Healthcare claims are collected for billing purposes, and they lack detail on measures of disease severity, such as the International Prognostic Scoring System, which is designed for evaluating prognosis in MDS.3,23 In addition, our study included patients with commercial insurance, so patients with Medicare were underrepresented. Therefore, we could not further stratify the agegroup of those ≥50 years in our study to perform additional age-group comparisons. Our results may therefore not be representative of the general MDS population, because different populations may have various outcomes. We were also unable to examine other subgroups, because of the small sample sizes (eg, patients receiving allogeneic stem-cell transplant, and those receiving pharmacologic therapy with hypomethylating agents or with thalidomide analogues). In our previous study of 1209 patients newly diagnosed with MDS—a sample that included all treatment groups—we found that mean total healthcare costs were $100,809 (SD, $188,311; median, $40,975), only $14,332 greater than the total healthcare costs reported in the current study of supportive care patients ($86,477; Table 4).24 We also did not examine whether the newly diagnosed patients with MDS in our study could have had AML before MDS in the postindex period, or whether some patients had other clonal or nonclonal diagnoses that are common in a hematologic practice, such as autoimmune disease or toxic injury to the marrow. Furthermore, because of our study’s relatively short follow-up period, we were unable to establish causal relationships. A small sample size could have limited our detection of significant differences (eg, differences in healthcare charges by age). Other limitations that are particular to claims data analyses could have impacted the utilization and the cost results in this study. We were unable to estimate inpatient antibiotic use, because inpatient claims data only contain diagnoses and procedure codes and not information on medication use. Although we reviewed all inpatient and outpatient claims to identify transfusions, inpatient claims in the i3/Ingenix LabRx database include a maximum of 3 procedure codes; thus, inpatient transfusions may have been missed. Similarly, we examined healthcare charges in the newly diagnosed MDS population, and therefore our results may differ from other studies that examined costs or paid amounts for claims associated with MDS. Additional suf-

464

American Health & Drug Benefits

ficiently powered longitudinal studies that account for severity of disease, that are conducted in various MDS populations, and that use various data sources are warranted.

Conclusions Our study indicates that MDS is associated with frequent and prolonged hospitalizations, frequent outpatient visits, and high healthcare charges in both younger and older patients receiving supportive care. Although MDS is considered a disease of the elderly, the results of this study suggest that a small proportion of patients aged <50 years may have this disease and may have a much higher healthcare utilization and cost-related burden of MDS than patients aged ≥50 years, possibly because of the longer length of stay among hospitalized younger patients. This study highlights the importance of conducting further studies to better elucidate the characteristics of patients with early-onset MDS. ■ Study funding Funding for this study was provided by Eisai Inc. Author Disclosure Statement Dr Powers and Dr Faria are employees of Eisai, and Dr Broder, Dr Chang, and Dr Cherepanov are employees of Partnership for Health Analytic Research.

References 1. Greenberg P. The myelodysplastic syndromes. In: Hoffman R, Benz E, Shattil S, et al, eds. Hematology: Basic Principles and Practice. 3rd ed. New York, NY: Churchill Livingstone; 2000:1106-1129. 2. De Roos AJ, Deeg HJ, Onstad L, et al. Incidence of myelodysplastic syndromes within a nonprofit healthcare system in western Washington state, 2005-2006. Am J Hematol. 2010;85:765-770. 3. Greenberg P, Attar E, Bennett JM, et al. NCCN Clinical Practice Guidelines in Oncology: myelodysplastic syndromes. J Natl Compr Canc Netw. 2011;9:30-56. 4. Bryan J, Jabbour E, Prescott H, et al. Current and future management options for myelodysplastic syndromes. Drugs. 2010;70:1381-1394. 5. Atallah E, Garcia-Manero G. Treatment strategies in myelodysplastic syndromes. Cancer Invest. 2008;26:208-216. 6. Rollison DE, Howlader N, Smith MT, et al. Epidemiology of myelodysplastic syndromes and chronic myeloproliferative disorders in the United States, 2001-2004, using data from the NAACCR and SEER programs. Blood. 2008;112:45-52. 7. Cogle CR, Craig BM, Rollison DE, et al. Incidence of the myelodysplastic syndromes using a novel claims-based algorithm: high number of uncaptured cases by cancer registries. Blood. 2011;117:7121-7125. 8. Hatoum HT, Lin SJ, Buchner D, et al. Use of hypomethylating agents and associated care in patients with myelodysplastic syndromes: a claims database study. Curr Med Res Opin. 2011;27:1255-1262. 9. Pan F, Peng S, Fleurence R, et al. Economic analysis of decitabine versus best supportive care in the treatment of intermediate- and high-risk myelodysplastic syndromes from a US payer perspective. Clin Ther. 2010;32:2444-2450. 10. Germing U, Strupp C, Kündgen A, et al. No increase in age-specific incidence of myelodysplastic syndromes. Haematologica. 2004;89:905-910. 11. Ma X, Does M, Raza A, et al. Myelodysplastic syndromes: incidence and survival in the United States. Cancer. 2007;109:1536-1542. 12. Kuendgen A, Strupp C, Aivado M, et al. Myelodysplastic syndromes in patients younger than age 50. J Clin Oncol. 2006;24:5358-5365. 13. Cutler CS, Lee SJ, Greenberg P, et al. A decision analysis of allogeneic bone marrow transplantation for the myelodysplastic syndromes: delayed transplantation for low-risk myelodysplasia is associated with improved outcome. Blood. 2004;104:579-585. 14. Charlson ME, Pompei P, Ales KL, et al. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40:373-383. 15. Deyo RA, Cherkin DC, Ciol MA. Adapting a clinical comorbidity index for use with ICD-9-CM administrative databases. J Clin Epidemiol. 1992;45:613-619. 16. Van Bennekom CM, Abel G, Anderson T, et al. Patterns of treatment among patients with recently-diagnosed myelodysplastic syndromes in a national registry, 2006-2008. Blood. 2008;112:Abstract 876. 17. Powers A, Stein K, Knoth RL, et al. Health care utilization and costs in pa-

www.AHDBonline.com

November/December 2012

Vol 5, No 7

Burden of MDS in Patients Receiving Supportive Care

tients with early onset myelodysplastic syndrome in a commercially insured population. J Clin Oncol. 2011;29(suppl 15):Abstract 6552. 18. Kühne F, Mittendorf T, Germing U, et al. Cost of transfusion-dependent myelodysplastic syndrome (MDS) from a German payer’s perspective. Ann Hematol. 2010;89:1239-1247. 19. Lindquist KJ, Danese MD, Mikhael J, et al. Health care utilization and mortality among elderly patients with myelodysplastic syndromes. Ann Oncol. 2011;22:1181-1188. 20. Gupta P, LeRoy SC, Luikart SD, et al. Long-term blood product transfusion support for patients with myelodysplastic syndromes (MDS): cost analysis and complications. Leuk Res. 1999;23:953-959. 21. Frytak JR, Henk HJ, De Castro CM, et al. Estimation of economic costs associ-

ated with transfusion dependence in adults with MDS. Curr Med Res Opin. 2009;25: 1941-1951. 22. Tyree PT, Lind BK, Lafferty WE. Challenges of using medical insurance claims data for utilization analysis. Am J Med Qual. 2006;21:269-275. 23. Greenberg P, Cox C, LeBeau MM, et al. International scoring system for evaluating prognosis in myelodysplastic syndromes. Blood. 1997;89:2079-2088. 24. Powers A, Stein K, Knoth R, et al. Hematologic complications and high costs associated with patients with myelodysplastic syndrome in a commercially insured population. Abstract presented at: 2011 International MASCC/ISOO Symposium; June 23-25, 2011; Athens, Greece. Support Care Cancer. 2011;19:(suppl 2):S146. Abstract 221.

STAKEHOLDER PERSPECTIVE

Reconsidering the Management of Younger Patients with Myelodysplastic Syndrome Jeffrey a. Bourret, Ms, rPh, FashP Senior Director, Medical Affairs, Pfizer Specialty Care, Collegeville, PA

PAYERS: Myelodysplastic syndrome (MDS) encompasses a heterogeneous group of myeloid disorders that increase the risk for progression to acute myelogenous leukemia (AML), which is associated with significant morbidity and mortality. MDS is more prevalent in older than in younger persons and in those who had been exposed to chemotherapy. Therapy is based on patient risks, needs for transfusion, and bone marrow biopsies; increasingly, genetic profiling has been used to assess risk. The goals of therapy vary with the risk profile. Reductions in transfusions and slowing the progression to high-risk disease or to AML are the goals in low-risk patients. Prolonging survival is the goal in high-risk patients. Treatments include growth factors, aggressive chemotherapy, stem-cell transplantation, lenalidomide, and the hypomethylating agents.1 Research has shown that patients respond to specific treatments based on their risk profile with lenalidomide, demonstrating effectiveness in patients with lower-risk disease, anemia, and genetic alterations, as well as in high-risk patients who respond to 5-azacitidine and to decitabine. Supportive care with iron chelation therapy and prophylactic antibiotics is extensively used.1 The small patient population, multiple treatment options, and recent research demonstrating different effectiveness of treatments for MDS based on risk warrant database analyses such as described in the present article by Powers and colleagues to help payers better understand the complications, healthcare utilization, and costs of treatment in their members with MDS. This study generates insights on the younger, lower-risk patient population, providing evidence to suggest opportunities to

Vol 5, No 7

November/December 2012

improve compliance with the National Comprehensive Cancer Network (NCCN) guidelines for MDS2 and advancements in MDS treatment based on latest research that indicate a need for more than supportive care in low-risk patients. PATIENTS: This article highlights the important differences in disease complications and risk-based treatments for MDS and the goals of reducing transfusion dependence and progression to AML in younger, lowerrisk patients. The data suggest potential underuse of bone marrow biopsies in the diagnosis of MDS, as well as underuse of lenalidomide and hypomethylating agents in low-risk patients as described in the NCCN guidelines2 in favor of supportive care only. PROVIDERS: The approach to treatment of MDS has changed over the years, focusing on therapy based on patient risk for disease progression and the use of genetics for the diagnosis. Diagnostic bone marrow biopsies are critical for treatment decisions, yet data presented by Powers and colleagues suggest suboptimal use. This analysis reflects the potential overuse of supportive care only in younger patients, overlooking the NCCN’s recommendations for lenalidomide and hypomethylating agents for MDS management2; however, because the database used in this reaserch lacks the detailed laboratory tests routinely used in clinical practice to make treatment-based decisions, further study is warranted to arrive at firm conclusions. 1. Garcia-Manero G. Myelodysplastic syndromes: 2012 update on diagnosis, risk-stratification, and management. Am J Hematol. 2012;87:692-701. 2. National Comprehensive Cancer Network. Myelodysplastic syndromes. J Natl Compr Canc Netw. 2011;9:30-56.

www.AHDBonline.com

American Health & Drug Benefits

465

“Managing patients with myeloma means staying current.”

Ira Klein, MD, MBA, FACP Chief of Staff to the Chief Medical Officer Aetna Hartford, CT

Value-Based Care in Myeloma !&%0!,- !2 &/-%0! %(.!,0%!1( *!,-*! .%0!- ,!& .! .) )-. +/ &%.3 ( !-- %--/!*! % & -! .%)(- "), -! &%(% % (0 ( ! *, .% ! (/,-!- ( *$ ,' %-.- 1%&& &-) ") /- )( .$! /(%+/! $ &&!(#!%( .$! ' ( #!'!(. )" '/&.%*&! '3!&)'

www.ValueBasedMyeloma.com Value-Based Care in Myeloma %-

*/ &%

.%)( )" (# #! ! &.$

)''/(%

.%)(-

'!' !, )" $! 3(2

,)/* 5

&& ,%#$.- ,!-!,0!

THIRD ANNUAL

Association for Value-Based Cancer Care Conference Influencing the Patient-Impact Factor May 2-5, 2013 • Westin Diplomat • Hollywood, Florida CONFERENCE CO-CHAIRS

AGENDA* THURSDAY, MAY 2, 2013 8:00 am - 5:00 pm

Registration

FRIDAY, MAY 3, 2013

Craig K. Deligdish, MD Hematologist/Oncologist Oncology Resource Networks

Gary M. Owens, MD President Gary Owens Associates

Burt Zweigenhaft, BS President and CEO OncoMed

PROGRAM OVERVIEW Following on the success of our Second Annual Conference, AVBCC will be coming to Hollywood, Florida, on May 2-5, 2013. We continue to be guided by the expertise of leaders in these fields providing attendees with a thorough understanding of the evolution of the value equation as it relates to cancer therapies. Our goal is to be able to assist them in implementing, improving, and sustaining their organizations and institutions, while improving access for patients and ultimately quality patient care.

7:00 am - 8:00 am

Simultaneous Symposia/Product Theaters

8:15 am - 9:15 am

Session 1: Welcome, Introductions, and Opening Remarks Conference Co-Chairs - Craig K. Deligdish, MD; Gary M. Owens, MD; Burt Zweigenhaft, BS

9:15 am - 10:15 am

Keynote Address

10:15 am - 10:30 am

Break

10:30 am - 11:45 am

Session 2: Trends in Treatment Decision-Making: Pathways and Stakeholder Collaborations Roy A. Beveridge, MD; Michael Kolodziej, MD

12:00 pm - 1:00 pm

Exclusive Lunch Symposium/Product Theater

1:15 pm - 2:00 pm

Session 3: Cost of Cure: When, How, and How Much? John Fox, MD; John Hennessy

2:00 pm - 2:45 pm

Session 4: Where Is Oncology Care Headed in the Future? Jayson Slotnick, JD, MPH (Moderator); Barbara L. McAneny, MD

Upon completion of this activity, the participant will be able to: • Discuss the current trends and challenges facing all stakeholders in optimizing value in cancer care delivery. • Define the barriers associated with cost, quality, and access as they relate to healthcare reform and what solutions are currently being considered. • Compare and contrast the different approaches/tools providers and payers are utilizing to manage and deliver care collaboratively. • Examine the current trends in personalized care and companion diagnostics. • Analyze the patient issues around cost, quality, and access to care.

2:45 pm - 3:30 pm

Session 5: What Will the Cancer Delivery System Look Like in 2015? Linda Bosserman, MD, FACP; John D. Sprandio, MD

3:30 pm - 3:45 pm

Break

3:45 pm - 4:30 pm

Session 6: Employers and Oncology Care F. Randy Vogenberg, PhD, RPh (Moderator); Bridget Eber, PharmD; Patricia Goldsmith; Darin Hinderman

4:30 pm - 5:15 pm

Session 7: The Role of Government in the Future of Oncology Care Jayson Slotnick, JD, MPH

TARGET AUDIENCE

5:15 pm - 5:45 pm

Summary/Wrap-Up of Day 1

This conference is intended for medical oncologists, practice managers/administrators, and managed care professionals. Stakeholders in a position to impact cancer patient care, such as advanced practice nurses, pharmacists, and medical directors, are also invited to join this exciting forum.

6:00 pm - 8:00 pm

Cocktail Reception in the Exhibit Hall

SATURDAY, MAY 4, 2013 7:00 am - 8:00 am

Simultaneous Symposia/Product Theaters

DESIGNATION OF CREDIT STATEMENTS

8:15 am - 8:30 am

Opening Remarks

8:30 am - 9:15 am

Session 8: Advanced Care Directives: Palliative Care, Hospice, Ethics J. Russell Hoverman, MD, PhD Thomas J. Smith, MD, FACP, FASCO

9:15 am - 10:00 am

Session 9: Medicaid: A Healthcare Delivery System Review Matthew Brow

LEARNING OBJECTIVES

SPONSORS This activity is jointly sponsored by Medical Learning Institute Inc, the Association for Value-Based Cancer Care, Inc., Center of Excellence Media, LLC, and Core Principle Solutions, LLC.

COMMERCIAL SUPPORT ACKNOWLEDGMENT

10:00 am - 10:15 am

Break

Grant requests are currently being reviewed by numerous supporters. Support will be acknowledged prior to the start of the educational activities.

10:15 am - 11:00 am

Session 10: Payer, Government, and Industry Insights: Balancing Cost and Quality

11:00 am - 11:45 am

Session 11: National Coalition for Cancer Survivorship: Medication Nonadherence Issues Pat McKercher

12:00 pm - 1:00 pm

Exclusive Lunch Symposium/Product Theater

1:15 pm - 3:00 pm

Session 12: Meet the Experts Networking Roundtable Session

3:00 pm - 3:45 pm

Session 13: Personalized Medicine, Companion Diagnostics, Molecular Profiling, Genome Sequencing—The Impact on Cost, Treatment, and the Value Proposition Mark S. Boguski, MD, PhD

3:45 pm - 4:15 pm

Summary/Wrap-Up of Day 2

4:30 pm - 6:30 pm

Cocktail Reception in the Exhibit Hall

PHYSICIAN CREDIT DESIGNATION The Medical Learning Institute Inc designates this live activity for a maximum of 17.25 AMA PRA Category 1 Credits™. Physicians should claim only the credit commensurate with the extent of their participation in the activity. This activity has been planned and implemented in accordance with the Essential Areas and policies of the Accreditation Council for Continuing Medical Education through the joint sponsorship of the Medical Learning Institute Inc and the Center of Excellence Media, LLC. The Medical Learning Institute Inc is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians.

REGISTERED NURSE DESIGNATION Medical Learning Institute Inc. Provider approved by the California Board of Registered Nursing, Provider Number 15106, for 17.25 contact hours.

SUNDAY, MAY 5, 2013 7:00 am - 8:00 am

Simultaneous Symposia/Product Theaters

REGISTERED PHARMACY DESIGNATION

8:15 am - 8:30 am

Opening Remarks

The Medical Learning Institute Inc is accredited by the Accreditation Council for Pharmacy Education as a provider of continuing pharmacy education. Completion of this knowledge-based activity provides for 17.25 contact hours (1.725 CEUs) of continuing pharmacy education credit. The Universal Activity Number for this activity is (To be determined).

8:30 am - 9:15 am

Session 14: Cancer Rehabilitation: The Next Frontier in Survivorship Care Julie Silver, MD

9:15 am - 10:00 am

Session 15: Current and Future Considerations for the Oncology Practice Manager Dawn Holcombe, MBA, FACMPE, ACHE; Leonard Natelson

10:00 am - 10:15 am

Break

10:15 am - 11:00 am

Session 16: Access to Drugs—Shortages, Biosimilars Douglas Burgoyne, PharmD; James T. Kenney, Jr., RPh, MBA

11:00 am - 11:45 am

Session 17: Perspectives from Large Oncology Group Practices—Successes, Issues, and Challenges

11:45 am - 12:00 pm

Summary and Conclusion of Conference

CONFERENCE REGISTRATION Discounted Pricing Available!

$375.00 until January 15, 2013 $475.00 until March 15, 2013 $675.00 after March 15, 2013

www.regonline.com/avbcc2013

*Agenda is subject to change.

CALL FOR PAPERS American Health & Drug Benefits offers an open forum for all healthcare participants to exchange ideas and present their data, innovations, and initiatives to facilitate patient-centered healthcare and benefit design models that meet the needs of all stakeholders—Distributors, Employers, Manufacturers, Patients, Payers, Providers, Purchasers, Regulators, and Researchers. Readers are invited to submit articles that aim at improving the quality of patient care and patient well-being while reducing or controlling costs, enhancing the health of communities and patient populations, as well as other topics relevant to benefit design with specific implications to policymakers, payers, and employers.

Areas of High Interest: • Health Information Exchange • Health Plan Initiatives • Health Outcomes • Innovations in Healthcare • Literature Reviews • Managed Care • Medicare/Medicaid • Patient Outcomes

• Adherence Concerns • Benefit Design • Case Studies • Comorbidities and Cost Issues • Cost-Effectiveness Comparisons • Decision-Making Tools • Ethics in Medicine • Health Economics Research

• Pharmacoeconomics • Pharmacogenomics • Policy Issues • Prevention Initiatives • Reimbursement Strategies • Social Media and Health • Survey Results • Value-Based Healthcare

Clinical Topics of High Interest: AGING—With the aging of the US population there is a growing need for early implementation of outcomes-based preventive and therapeutic strategies for older people. ALLERGIES—Allergies, such as allergic or seasonal rhinitis, affect millions of Americans daily, resulting in a significant economic burden and human cost. Undertreatment and lack of adherence are common obstacles to patient management. ARTHRITIS—Musculoskeletal conditions are on the increase, yet many patients are undiagnosed and untreated. Comparing new and emerging therapies is a key target for improving patient outcomes and reducing costs. CANCER CARE—The growing focus on biologic agents dictates an enhanced study of these therapeutic options, including reimbursement policies and cost management. CARDIOVASCULAR DISEASE—Original, outcomesbased research on appropriate therapies, cost comparisons, emerging prevention strategies, and best practices will enhance readers’ decision-making.

DIABETES, OBESITY—The growing epidemics of these metabolic conditions mandate a thorough examination of best therapies, adherence issues, access, and prevention strategies. GASTROINTESTINAL CONDITIONS—Recognizing GI conditions, such as hepatitis C, Crohn’s disease, or inflammatory bowel disorder, remains a challenge. INFECTIOUS DISEASES—The spread of common and emerging pathogens within the hospital and in the community is a serious concern requiring increased vigilance. MENTAL DISORDERS—Depression, bipolar disorder, and schizophrenia exert a huge financial and human burden on individuals, employers, and payers. Topics of interest include comparative effectiveness analyses, best practices, and reimbursement.

PAIN MANAGEMENT—Chronic pain is associated with many complicated medical disorders and an enormous economic burden, yet pain medications are still underused.

Manuscripts should follow the Manuscript Instructions for Authors (available at www.AHDBonline.com). Submit articles to editorial@engagehc.com. For more information, call 732-992-1892.

468

American Health & Drug Benefits

www.AHDBonline.com

November/December 2012

Vol 5, No 7

RJ Health Systems The Creators of ReimbursementCodes.com

RJ Health Systems â&#x20AC;&#x201D; the pharmaceutical specialists that healthcare professionals have turned to since 1983 for their drug information. We work with over 170 Payor organizations that touch approximately 110 million lives. RJ Health Systems has established and continuously maintains a Library of Drug Intelligence that provides the most comprehensive, trusted, and up-to-date coding and reimbursement information in the industry. ReimbursementCodes.com incorporates the CMS HCPCS and AMA CPT Drug codes into a system that crosswalks each drug code with the drugâ&#x20AC;&#x2122;s NDC and brand/generic name. Please visit www.rjhealthsystems.com to learn more about our products and services listed below:

>> >> >> >> >> >> >> >> >>

ReimbursementCodes.com Drug Diagnosis Coding CMAC PartBorPartD.com NDC Conversion Database Min/Max Dosing Advisories Notations Clinical Consulting

rjhealthsystems.com

Victoza® (liraglutide [rDNA origin] injection) Rx Only BRIEF SUMMARY. Please consult package insert for full prescribing information. WARNING: RISK OF THYROID C-CELL TUMORS: Liraglutide causes dose-dependent and treatment-duration-dependent thyroid C-cell tumors at clinically relevant exposures in both genders of rats and mice. It is unknown whether Victoza® causes thyroid C-cell tumors, including medullary thyroid carcinoma (MTC), in humans, as human relevance could not be ruled out by clinical or nonclinical studies. Victoza® is contraindicated in patients with a personal or family history of MTC and in patients with Multiple Endocrine Neoplasia syndrome type 2 (MEN 2). Based on the findings in rodents, monitoring with serum calcitonin or thyroid ultrasound was performed during clinical trials, but this may have increased the number of unnecessary thyroid surgeries. It is unknown whether monitoring with serum calcitonin or thyroid ultrasound will mitigate human risk of thyroid C-cell tumors. Patients should be counseled regarding the risk and symptoms of thyroid tumors [see Contraindications and Warnings and Precautions]. INDICATIONS AND USAGE: Victoza is indicated as an adjunct to diet and exercise to improve glycemic control in adults with type 2 diabetes mellitus. Important Limitations of Use: Because of the uncertain relevance of the rodent thyroid C-cell tumor findings to humans, prescribe Victoza only to patients for whom the potential benefits are considered to outweigh the potential risk. Victoza is not recommended as first-line therapy for patients who have inadequate glycemic control on diet and exercise. In clinical trials of Victoza, there were more cases of pancreatitis with Victoza than with comparators. Victoza has not been studied sufficiently in patients with a history of pancreatitis to determine whether these patients are at increased risk for pancreatitis while using Victoza. Use with caution in patients with a history of pancreatitis. Victoza is not a substitute for insulin. Victoza should not be used in patients with type 1 diabetes mellitus or for the treatment of diabetic ketoacidosis, as it would not be effective in these settings. The concurrent use of Victoza and prandial insulin has not been studied. CONTRAINDICATIONS: Do not use in patients with a personal or family history of medullary thyroid carcinoma (MTC) or in patients with Multiple Endocrine Neoplasia syndrome type 2 (MEN 2). Do not use in patients with a prior serious hypersensitivity reaction to Victoza or to any of the product components. WARNINGS AND PRECAUTIONS: Risk of Thyroid C-cell Tumors: Liraglutide causes dose-dependent and treatment-duration-dependent thyroid C-cell tumors (adenomas and/or carcinomas) at clinically relevant exposures in both genders of rats and mice. Malignant thyroid C-cell carcinomas were detected in rats and mice. A statistically significant increase in cancer was observed in rats receiving liraglutide at 8-times clinical exposure compared to controls. It is unknown whether Victoza® will cause thyroid C-cell tumors, including medullary thyroid carcinoma (MTC), in humans, as the human relevance of liraglutideinduced rodent thyroid C-cell tumors could not be determined by clinical or nonclinical studies [see Boxed Warning, Contraindications]. In the clinical trials, there have been 6 reported cases of thyroid C-cell hyperplasia among Victoza®-treated patients and 2 cases in comparator-treated patients (1.3 vs. 1.0 cases per 1000 patient-years). One comparator-treated patient with MTC had pre-treatment serum calcitonin concentrations >1000 ng/L suggesting pre-existing disease. All of these cases were diagnosed after thyroidectomy, which was prompted by abnormal results on routine, protocol-specified measurements of serum calcitonin. Five of the six Victoza®-treated patients had elevated calcitonin concentrations at baseline and throughout the trial. One Victoza® and one non-Victoza®-treated patient developed elevated calcitonin concentrations while on treatment. Calcitonin, a biological marker of MTC, was measured throughout the clinical development program. The serum calcitonin assay used in the Victoza® clinical trials had a lower limit of quantification (LLOQ) of 0.7 ng/L and the upper limit of the reference range was 5.0 ng/L for women and 8.4 ng/L for men. At Weeks 26 and 52 in the clinical trials, adjusted mean serum calcitonin concentrations were higher in Victoza®-treated patients compared to placebo-treated patients but not compared to patients receiving active comparator. At these timepoints, the adjusted mean serum calcitonin values (~1.0 ng/L) were just above the LLOQ with between-group differences in adjusted mean serum calcitonin values of approximately 0.1 ng/L or less. Among patients with pre-treatment serum calcitonin below the upper limit of the reference range, shifts to above the upper limit of the reference range which persisted in subsequent measurements occurred most frequently among patients treated with Victoza® 1.8 mg/day. In trials with on-treatment serum calcitonin measurements out to 5-6 months, 1.9% of patients treated with Victoza® 1.8 mg/day developed new and persistent calcitonin elevations above the upper limit of the reference range compared to 0.8-1.1% of patients treated with control medication or the 0.6 and 1.2 mg doses of Victoza®. In trials with on-treatment serum calcitonin measurements out to 12 months, 1.3% of patients treated with Victoza® 1.8 mg/day had new and persistent elevations of calcitonin from below or within the reference range to above the upper limit of the reference range, compared to 0.6%, 0% and 1.0% of patients treated with Victoza® 1.2 mg, placebo and active control, respectively. Otherwise, Victoza® did not produce consistent dose-dependent or time-dependent increases in serum calcitonin. Patients with MTC usually have calcitonin values >50 ng/L. In Victoza® clinical trials, among patients with pre-treatment serum calcitonin <50 ng/L, one Victoza®-treated patient and no comparator-treated patients developed serum calcitonin >50 ng/L. The Victoza®-treated patient who developed serum calcitonin >50 ng/L had an elevated pre-treatment serum calcitonin of 10.7 ng/L that increased to 30.7 ng/L at Week 12 and 53.5 ng/L at the end of the 6-month trial. Follow-up serum calcitonin was 22.3 ng/L more than 2.5 years after the last dose of Victoza®. The largest increase in serum calcitonin in a comparator-treated patient was seen with glimepiride in a patient whose serum calcitonin increased from 19.3 ng/L at baseline to 44.8 ng/L at Week 65 and 38.1 ng/L at Week 104. Among patients who began with serum calcitonin <20 ng/L, calcitonin elevations to >20 ng/L occurred in 0.7% of Victoza®-treated patients, 0.3% of placebo-treated patients, and 0.5% of active-comparator-treated patients, with an incidence of 1.1% among patients treated with 1.8 mg/ day of Victoza®. The clinical significance of these findings is unknown. Counsel patients regarding the risk for MTC and the symptoms of thyroid tumors (e.g. a mass in the neck, dysphagia, dyspnea or persistent hoarseness). It is unknown whether monitoring with serum calcitonin or thyroid ultrasound will mitigate the potential risk of MTC, and such monitoring may increase the risk of unnecessary procedures, due to low test specificity for serum calcitonin and a high background incidence of thyroid disease. Patients with thyroid nodules noted on physical examination or neck imaging obtained for other reasons should be referred to an endocrinologist for further evaluation. Although routine monitoring of serum calcitonin is of uncertain value in patients treated with Victoza®, if serum calcitonin is measured and found to be elevated, the patient should be referred to an endocrinologist for further evaluation. Pancreatitis: In clinical trials of Victoza®, there have been 13 cases of pancreatitis among Victoza®-treated patients and 1 case in a comparator (glimepiride) treated patient (2.7 vs. 0.5 cases per 1000 patient-years). Nine of the 13 cases with Victoza® were reported as acute pancreatitis and four were reported as chronic pancreatitis. In one case in a Victoza®-treated patient, pancreatitis, with necrosis, was observed and led to death; however clinical causality could not be established. Some patients had other risk factors for pancreatitis, such as a history of cholelithiasis or alcohol abuse. There are no conclusive data establishing a risk of pancreatitis with Victoza® treatment. After initiation of Victoza®, and after dose increases, observe patients carefully for signs and symptoms of pancreatitis (including persistent severe abdominal pain, sometimes radiating to the back and which may or may not be accompanied by vomiting). If pancreatitis is suspected, Victoza® and other potentially suspect medications should be discontinued promptly, confirmatory tests should be performed and appropriate management should be initiated. If pancreatitis is confirmed, Victoza® should not be restarted. Use with caution in patients with a history of pancreatitis. Use with Medications Known to Cause Hypoglycemia: Patients receiving Victoza® in combination with an insulin secretagogue (e.g., sulfonylurea) or insulin may have an increased risk of hypoglycemia. The risk of hypoglycemia may be

lowered by a reduction in the dose of sulfonylurea (or other concomitantly administered insulin secretagogues) or insulin [see Adverse Reactions]. Renal Impairment: Victoza ® has not been found to be directly nephrotoxic in animal studies or clinical trials. There have been postmarketing reports of acute renal failure and worsening of chronic renal failure, which may sometimes require hemodialysis in Victoza®-treated patients [see Adverse Reactions]. Some of these events were reported in patients without known underlying renal disease. A majority of the reported events occurred in patients who had experienced nausea, vomiting, diarrhea, or dehydration [see Adverse Reactions]. Some of the reported events occurred in patients receiving one or more medications known to affect renal function or hydration status. Altered renal function has been reversed in many of the reported cases with supportive treatment and discontinuation of potentially causative agents, including Victoza®. Use caution when initiating or escalating doses of Victoza® in patients with renal impairment. Hypersensitivity Reactions: There have been postmarketing reports of serious hypersensitivity reactions (e.g., anaphylactic reactions and angioedema) in patients treated with Victoza®. If a hypersensitivity reaction occurs, the patient should discontinue Victoza® and other suspect medications and promptly seek medical advice. Angioedema has also been reported with other GLP-1 receptor agonists. Use caution in a patient with a history of angioedema with another GLP-1 receptor agonist because it is unknown whether such patients will be predisposed to angioedema with Victoza®. Macrovascular Outcomes: There have been no clinical studies establishing conclusive evidence of macrovascular risk reduction with Victoza® or any other antidiabetic drug. ADVERSE REACTIONS: Clinical Trials Experience: 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. The safety of Victoza® has been evaluated in 8 clinical trials: A double-blind 52-week monotherapy trial compared Victoza® 1.2 mg daily, Victoza® 1.8 mg daily, and glimepiride 8 mg daily; A double-blind 26 week add-on to metformin trial compared Victoza® 0.6 mg once-daily, Victoza® 1.2 mg once-daily, Victoza® 1.8 mg once-daily, placebo, and glimepiride 4 mg once-daily; A double-blind 26 week add-on to glimepiride trial compared Victoza® 0.6 mg daily, Victoza® 1.2 mg once-daily, Victoza® 1.8 mg oncedaily, placebo, and rosiglitazone 4 mg once-daily; A 26 week add-on to metformin + glimepiride trial, compared double-blind Victoza® 1.8 mg once-daily, double-blind placebo, and open-label insulin glargine once-daily; A double-blind 26-week add-on to metformin + rosiglitazone trial compared Victoza® 1.2 mg once-daily, Victoza® 1.8 mg once-daily and placebo; An open-label 26-week add-on to metformin and/or sulfonylurea trial compared Victoza® 1.8 mg once-daily and exenatide 10 mcg twice-daily; An open-label 26-week add-on to metformin trial compared Victoza® 1.2 mg once-daily, Victoza® 1.8 mg once-daily, and sitagliptin 100 mg once-daily; An open-label 26-week trial compared insulin detemir as add-on to Victoza® 1.8 mg + metformin to continued treatment with Victoza® + metformin alone. Withdrawals: The incidence of withdrawal due to adverse events was 7.8% for Victoza®-treated patients and 3.4% for comparator-treated patients in the five double-blind controlled trials of 26 weeks duration or longer. This difference was driven by withdrawals due to gastrointestinal adverse reactions, which occurred in 5.0% of Victoza®-treated patients and 0.5% of comparator-treated patients. In these five trials, the most common adverse reactions leading to withdrawal for Victoza®-treated patients were nausea (2.8% versus 0% for comparator) and vomiting (1.5% versus 0.1% for comparator). Withdrawal due to gastrointestinal adverse events mainly occurred during the first 2-3 months of the trials. Common adverse reactions: Tables 1, 2, 3 and 4 summarize common adverse reactions (hypoglycemia is discussed separately) reported in seven of the eight controlled trials of 26 weeks duration or longer. Most of these adverse reactions were gastrointestinal in nature. In the five double-blind clinical trials of 26 weeks duration or longer, gastrointestinal adverse reactions were reported in 41% of Victoza®-treated patients and were dose-related. Gastrointestinal adverse reactions occurred in 17% of comparator-treated patients. Common adverse reactions that occurred at a higher incidence among Victoza®-treated patients included nausea, vomiting, diarrhea, dyspepsia and constipation. In the five double-blind and three open-label clinical trials of 26 weeks duration or longer, the percentage of patients who reported nausea declined over time. In the five double-blind trials approximately 13% of Victoza®-treated patients and 2% of comparator-treated patients reported nausea during the first 2 weeks of treatment. In the 26-week open-label trial comparing Victoza® to exenatide, both in combination with metformin and/or sulfonylurea, gastrointestinal adverse reactions were reported at a similar incidence in the Victoza® and exenatide treatment groups (Table 3). In the 26-week open-label trial comparing Victoza® 1.2 mg, Victoza® 1.8 mg and sitagliptin 100 mg, all in combination with metformin, gastrointestinal adverse reactions were reported at a higher incidence with Victoza® than sitagliptin (Table 4). In the remaining 26-week trial, all patients received Victoza® 1.8 mg + metformin during a 12-week run-in period. During the run-in period, 167 patients (17% of enrolled total) withdrew from the trial: 76 (46% of withdrawals) of these patients doing so because of gastrointestinal adverse reactions and 15 (9% of withdrawals) doing so due to other adverse events. Only those patients who completed the run-in period with inadequate glycemic control were randomized to 26 weeks of add-on therapy with insulin detemir or continued, unchanged treatment with Victoza® 1.8 mg + metformin. During this randomized 26-week period, diarrhea was the only adverse reaction reported in ≥5% of patients treated with Victoza® 1.8 mg + metformin + insulin detemir (11.7%) and greater than in patients treated with Victoza® 1.8 mg and metformin alone (6.9%). Table 1: Adverse reactions reported in ≥5% of Victoza®-treated patients in a 52-week monotherapy trial All Victoza® N = 497 Glimepiride N = 248 (%) (%) Adverse Reaction Nausea 28.4 8.5 Diarrhea 17.1 8.9 Vomiting 10.9 3.6 Constipation 9.9 4.8 Headache 9.1 9.3 Table 2: Adverse reactions reported in ≥5% of Victoza®-treated patients and occurring more frequently with Victoza® compared to placebo: 26-week combination therapy trials Add-on to Metformin Trial All Victoza® + Metformin Placebo + Metformin Glimepiride + Metformin N = 724 N = 121 N = 242 (%) (%) (%) Adverse Reaction Nausea 15.2 4.1 3.3 Diarrhea 10.9 4.1 3.7 Headache 9.0 6.6 9.5 Vomiting 6.5 0.8 0.4 Add-on to Glimepiride Trial Placebo + Glimepiride Rosiglitazone + All Victoza® + Glimepiride N = 695 N = 114 Glimepiride N = 231 (%) (%) (%) Adverse Reaction Nausea 7.5 1.8 2.6 Diarrhea 7.2 1.8 2.2 Constipation 5.3 0.9 1.7 Dyspepsia 5.2 0.9 2.6

Add-on to Metformin + Glimepiride Victoza® 1.8 + Metformin Placebo + Metformin + Glargine + Metformin + + Glimepiride N = 230 Glimepiride N = 114 Glimepiride N = 232 (%) (%) (%) Adverse Reaction Nausea 13.9 3.5 1.3 Diarrhea 10.0 5.3 1.3 Headache 9.6 7.9 5.6 Dyspepsia 6.5 0.9 1.7 Vomiting 6.5 3.5 0.4 Add-on to Metformin + Rosiglitazone ® Placebo + Metformin + Rosiglitazone All Victoza + Metformin + Rosiglitazone N = 355 N = 175 (%) (%) Adverse Reaction Nausea 34.6 8.6 Diarrhea 14.1 6.3 Vomiting 12.4 2.9 Headache 8.2 4.6 Constipation 5.1 1.1 Table 3: Adverse Reactions reported in ≥5% of Victoza®-treated patients in a 26-Week Open-Label Trial versus Exenatide Victoza® 1.8 mg once daily + Exenatide 10 mcg twice daily + metformin and/or sulfonylurea metformin and/or sulfonylurea N = 235 N = 232 (%) (%) Adverse Reaction Nausea 25.5 28.0 Diarrhea 12.3 12.1 Headache 8.9 10.3 Dyspepsia 8.9 4.7 Vomiting 6.0 9.9 Constipation 5.1 2.6 Table 4: Adverse Reactions in ≥5% of Victoza®-treated patients in a 26-Week Open-Label Trial versus Sitagliptin All Victoza® + metformin Sitagliptin 100 mg/day + N = 439 metformin N = 219 (%) (%) Adverse Reaction Nausea 23.9 4.6 Headache 10.3 10.0 Diarrhea 9.3 4.6 Vomiting 8.7 4.1 Immunogenicity: Consistent with the potentially immunogenic properties of protein and peptide pharmaceuticals, patients treated with Victoza® may develop anti-liraglutide antibodies. Approximately 50-70% of Victoza®-treated patients in the five double-blind clinical trials of 26 weeks duration or longer were tested for the presence of anti-liraglutide antibodies at the end of treatment. Low titers (concentrations not requiring dilution of serum) of anti-liraglutide antibodies were detected in 8.6% of these Victoza®-treated patients. Sampling was not performed uniformly across all patients in the clinical trials, and this may have resulted in an underestimate of the actual percentage of patients who developed antibodies. Cross-reacting antiliraglutide antibodies to native glucagon-like peptide-1 (GLP-1) occurred in 6.9% of the Victoza®-treated patients in the double-blind 52-week monotherapy trial and in 4.8% of the Victoza®-treated patients in the double-blind 26-week add-on combination therapy trials. These cross-reacting antibodies were not tested for neutralizing effect against native GLP-1, and thus the potential for clinically significant neutralization of native GLP-1 was not assessed. Antibodies that had a neutralizing effect on liraglutide in an in vitro assay occurred in 2.3% of the Victoza®-treated patients in the double-blind 52-week monotherapy trial and in 1.0% of the Victoza®-treated patients in the double-blind 26-week add-on combination therapy trials. Among Victoza®-treated patients who developed anti-liraglutide antibodies, the most common category of adverse events was that of infections, which occurred among 40% of these patients compared to 36%, 34% and 35% of antibody-negative Victoza®-treated, placebo-treated and active-control-treated patients, respectively. The specific infections which occurred with greater frequency among Victoza®-treated antibody-positive patients were primarily nonserious upper respiratory tract infections, which occurred among 11% of Victoza®-treated antibody-positive patients; and among 7%, 7% and 5% of antibody-negative Victoza®-treated, placebo-treated and active-control-treated patients, respectively. Among Victoza®-treated antibody-negative patients, the most common category of adverse events was that of gastrointestinal events, which occurred in 43%, 18% and 19% of antibody-negative Victoza®-treated, placebo-treated and active-control-treated patients, respectively. Antibody formation was not associated with reduced efficacy of Victoza® when comparing mean HbA1c of all antibody-positive and all antibody-negative patients. However, the 3 patients with the highest titers of anti-liraglutide antibodies had no reduction in HbA1c with Victoza® treatment. In the five double-blind clinical trials of Victoza®, events from a composite of adverse events potentially related to immunogenicity (e.g. urticaria, angioedema) occurred among 0.8% of Victoza®-treated patients and among 0.4% of comparator-treated patients. Urticaria accounted for approximately one-half of the events in this composite for Victoza®-treated patients. Patients who developed anti-liraglutide antibodies were not more likely to develop events from the immunogenicity events composite than were patients who did not develop anti-liraglutide antibodies. Injection site reactions: Injection site reactions (e.g., injection site rash, erythema) were reported in approximately 2% of Victoza®-treated patients in the five double-blind clinical trials of at least 26 weeks duration. Less than 0.2% of Victoza®-treated patients discontinued due to injection site reactions. Papillary thyroid carcinoma: In clinical trials of Victoza ®, there were 7 reported cases of papillary thyroid carcinoma in patients treated with Victoza® and 1 case in a comparator-treated patient (1.5 vs. 0.5 cases per 1000 patient-years). Most of these papillary thyroid carcinomas were <1 cm in greatest diameter and were diagnosed in surgical pathology specimens after thyroidectomy prompted by findings on protocol-specified screening with serum calcitonin or thyroid ultrasound. Hypoglycemia :In the eight clinical trials of at least 26 weeks duration, hypoglycemia requiring the assistance of another person for treatment occurred in 11 Victoza®-treated patients (2.3 cases per 1000 patient-years) and in two exenatidetreated patients. Of these 11 Victoza®-treated patients, six patients were concomitantly using metformin and a sulfonylurea, one was concomitantly using a sulfonylurea, two were concomitantly using metformin (blood glucose values were 65 and 94 mg/dL) and two were using Victoza® as monotherapy (one of these patients was undergoing an intravenous glucose tolerance test and the other was receiving insulin as treatment during a hospital stay). For these two patients on Victoza® monotherapy, the insulin treatment was the likely explanation for the hypoglycemia. In the 26-week open-label trial comparing Victoza® to sitagliptin, the incidence of hypoglycemic events defined as symptoms accompanied by a fingerstick glucose <56 mg/ dL was comparable among the treatment groups (approximately 5%).

Table 5: Incidence (%) and Rate (episodes/patient year) of Hypoglycemia in the 52-Week Monotherapy Trial and in the 26-Week Combination Therapy Trials Victoza® Treatment Active Comparator Placebo Comparator None Monotherapy Victoza® (N = 497) Glimepiride (N = 248) Patient not able to 0 0 — self−treat Patient able to self−treat 9.7 (0.24) 25.0 (1.66) — Not classified 1.2 (0.03) 2.4 (0.04) — Glimepiride + Placebo + Metformin Add-on to Metformin Victoza® + Metformin (N = 724) Metformin (N = 121) (N = 242) Patient not able to 0.1 (0.001) 0 0 self−treat Patient able to self−treat 3.6 (0.05) 22.3 (0.87) 2.5 (0.06) Continued Victoza® None Add-on to Victoza® + Insulin detemir + ® Metformin Victoza + Metformin + Metformin alone (N = 163) (N = 158*) Patient not able to 0 0 — self−treat Patient able to self−treat 9.2 (0.29) 1.3 (0.03) — Add-on to Victoza® + Glimepiride Rosiglitazone + Placebo + Glimepiride (N = 695) Glimepiride (N = 231) (N = 114) Glimepiride Patient not able to 0.1 (0.003) 0 0 self−treat Patient able to self−treat 7.5 (0.38) 4.3 (0.12) 2.6 (0.17) Not classified 0.9 (0.05) 0.9 (0.02) 0 ® Placebo + Metformin Add-on to Metformin Victoza + Metformin + Rosiglitazone None + Rosiglitazone + Rosiglitazone (N = 355) (N = 175) Patient not able to 0 — 0 self−treat Patient able to self−treat 7.9 (0.49) — 4.6 (0.15) Not classified 0.6 (0.01) — 1.1 (0.03) ® Add-on to Metformin Victoza + Metformin Insulin glargine Placebo + Metformin + Glimepiride + Metformin + + Glimepiride + Glimepiride (N = 230) Glimepiride (N = 232) (N = 114) Patient not able to 2.2 (0.06) 0 0 self−treat Patient able to self−treat 27.4 (1.16) 28.9 (1.29) 16.7 (0.95) Not classified 0 1.7 (0.04) 0 *One patient is an outlier and was excluded due to 25 hypoglycemic episodes that the patient was able to self-treat. This patient had a history of frequent hypoglycemia prior to the study. In a pooled analysis of clinical trials, the incidence rate (per 1,000 patient-years) for malignant neoplasms (based on investigator-reported events, medical history, pathology reports, and surgical reports from both blinded and open-label study periods) was 10.9 for Victoza®, 6.3 for placebo, and 7.2 for active comparator. After excluding papillary thyroid carcinoma events [see Adverse Reactions], no particular cancer cell type predominated. Seven malignant neoplasm events were reported beyond 1 year of exposure to study medication, six events among Victoza®-treated patients (4 colon, 1 prostate and 1 nasopharyngeal), no events with placebo and one event with active comparator (colon). Causality has not been established. Laboratory Tests: In the five clinical trials of at least 26 weeks duration, mildly elevated serum bilirubin concentrations (elevations to no more than twice the upper limit of the reference range) occurred in 4.0% of Victoza®-treated patients, 2.1% of placebo-treated patients and 3.5% of active-comparator-treated patients. This finding was not accompanied by abnormalities in other liver tests. The significance of this isolated finding is unknown. Vital signs: Victoza® did not have adverse effects on blood pressure. Mean increases from baseline in heart rate of 2 to 3 beats per minute have been observed with Victoza® compared to placebo. The long-term clinical effects of the increase in pulse rate have not been established [see Warnings and Precautions]. Post-Marketing Experience: The following additional adverse reactions have been reported during post-approval use of Victoza®. Because these events are reported voluntarily from a population of uncertain size, it is generally not possible to reliably estimate their frequency or establish a causal relationship to drug exposure: Dehydration resulting from nausea, vomiting and diarrhea [see Warnings and Precautions]; Increased serum creatinine, acute renal failure or worsening of chronic renal failure, sometimes requiring hemodialysis [see Warnings and Precautions]; Angioedema and anaphylactic reactions [see Contraindications, Warnings and Precautions] OVERDOSAGE: In a clinical trial, one patient with type 2 diabetes experienced a single overdose of Victoza® 17.4 mg subcutaneous (10 times the maximum recommended dose). Effects of the overdose included severe nausea and vomiting requiring hospitalization. No hypoglycemia was reported. The patient recovered without complications. In the event of overdosage, appropriate supportive treatment should be initiated according to the patient’s clinical signs and symptoms. More detailed information is available upon request. For information about Victoza® contact: Novo Nordisk Inc., 100 College Road West, Princeton, New Jersey 08540, 1−877-484-2869 Date of Issue: April 6, 2012 Version: 4 Manufactured by: Novo Nordisk A/S, DK-2880 Bagsvaerd, Denmark Victoza® is a registered trademark of Novo Nordisk A/S. Victoza® is covered by US Patent Nos. 6,268,343; 6,458,924; and 7,235,627 and other patents pending. Victoza® Pen is covered by US Patent Nos. 6,004,297; 6,235,004; 6,582,404 and other patents pending. © 2010-2012 Novo Nordisk 0512-00009479-1 6/2012

Help adult patients with type 2 diabetes gain greater access

Powerful reductions in A1C from -0.8% to -1.5%*

Low rate of hypoglycemia

Flexible dosing any time of day, independent of meals

May reduce weight

VictozaCare™ helps patients stay on track with ongoing support

—Victoza is not indicated for the management of obesity, and weight change was a secondary end point in clinical trials

—Patients enrolled in VictozaCare™ were more adherent to Victoza® than those not enrolled†

To see how Victoza® works for your patients, visit VictozaPro.com.

Indications and Usage Victoza® (liraglutide [rDNA origin] injection) is indicated as an adjunct to diet and exercise to improve glycemic control in adults with type 2 diabetes mellitus. Because of the uncertain relevance of the rodent thyroid C-cell tumor findings to humans, prescribe Victoza® only to patients for whom the potential benefits are considered to outweigh the potential risk. Victoza® is not recommended as first-line therapy for patients who have inadequate glycemic control on diet and exercise. In clinical trials of Victoza®, there were more cases of pancreatitis with Victoza® than with comparators. Victoza® has not been studied sufficiently in patients with a history of pancreatitis to determine whether these patients are at increased risk for pancreatitis while using Victoza®. Use with caution in patients with a history of pancreatitis. Victoza® is not a substitute for insulin. Victoza® should not be used in patients with type 1 diabetes mellitus or for the treatment of diabetic ketoacidosis, as it would not be effective in these settings. Victoza® has not been studied in combination with prandial insulin.

Important Safety Information Liraglutide causes dose-dependent and treatment-duration-dependent thyroid C-cell tumors at clinically relevant exposures in both genders of rats and mice. It is unknown whether Victoza® causes thyroid C-cell tumors, including medullary thyroid carcinoma (MTC), in humans, as human relevance could not be ruled out by clinical or nonclinical studies. Victoza® is contraindicated in patients with a personal or family history of MTC and in patients with Multiple Endocrine Neoplasia syndrome type 2 (MEN 2). Based on the findings in rodents, monitoring with serum calcitonin or thyroid ultrasound was performed during clinical trials, but this may have increased the number of unnecessary thyroid surgeries. It is unknown whether monitoring with serum calcitonin or thyroid ultrasound will mitigate human risk of thyroid C-cell tumors. Patients should be counseled regarding the risk and symptoms of thyroid tumors. Do not use in patients with a prior serious hypersensitivity reaction to Victoza® or to any of the product components. Victoza® is a registered trademark and VictozaCare™ is a trademark of Novo Nordisk A/S.

If pancreatitis is suspected, Victoza® (liraglutide [rDNA origin] injection) should be discontinued. Victoza® should not be re-initiated if pancreatitis is confirmed. When Victoza® is used with an insulin secretagogue (e.g. a sulfonylurea) or insulin serious hypoglycemia can occur. Consider lowering the dose of the insulin secretagogue or insulin to reduce the risk of hypoglycemia. Renal impairment has been reported post-marketing, usually in association with nausea, vomiting, diarrhea, or dehydration which may sometimes require hemodialysis. Use caution when initiating or escalating doses of Victoza® in patients with renal impairment. Serious hypersensitivity reactions (e.g. anaphylaxis and angioedema) have been reported during post marketing use of Victoza®. If symptoms of hypersensitivity reactions occur, patients must stop taking Victoza® and seek medical advice promptly. There have been no studies establishing conclusive evidence of macrovascular risk reduction with Victoza® or any other antidiabetic drug. The most common adverse reactions, reported in ≥5% of patients treated with Victoza® and more commonly than in patients treated with placebo, are headache, nausea, diarrhea, and anti-liraglutide antibody formation. Immunogenicityrelated events, including urticaria, were more common among Victoza®-treated patients (0.8%) than among comparator-treated patients (0.4%) in clinical trials. Victoza® has not been studied in type 2 diabetes patients below 18 years of age and is not recommended for use in pediatric patients. There is limited data in patients with renal or hepatic impairment. *Victoza® 1.2 mg and 1.8 mg when used alone or in combination with OADs. † Crossix ScoreBoard™ Report, May 2012. Adherence measured by number of actual Victoza® prescriptions filled for existing Victoza® patients enrolled in VictozaCare™ versus a match-pair control group not enrolled in VictozaCare™ through first 17 months of enrollment.1 Reference: 1. Crossix Solutions Inc. Crossix ScoreBoard™, May 2012.