`
` At the Intersection of Health, Health Care and Policy
`
`Cite this article as:
`Murray Aitken, Ernst R. Berndt and David M. Cutler
`Prescription Drug Spending Trends In The United States: Looking Beyond The Turning
`Point
`Health Affairs
`, 28, no.1 (2009):w151-w160
`(published online December 16, 2008; 10.1377/hlthaff.28.1.w151)
`
`The online version of this article, along with updated information and services, is
`available at:
`http://content.healthaffairs.org/content/28/1/w151.full.html
`
`
`
`For Reprints, Links & Permissions:
` http://healthaffairs.org/1340_reprints.php
`
`E-mail Alerts :
` http://content.healthaffairs.org/subscriptions/etoc.dtl
`
`To Subscribe:
` http://content.healthaffairs.org/subscriptions/online.shtml
`
`
`
`Health Affairs
`is published monthly by Project HOPE at 7500 Old Georgetown Road, Suite 600,
`Bethesda, MD 20814-6133. Copyright ©
` by Project HOPE - The People-to-People Health
`2009
`Foundation. As provided by United States copyright law (Title 17, U.S. Code), no part of
`Health Affairs
`may be reproduced, displayed, or transmitted in any form or by any means, electronic or mechanical,
`including photocopying or by information storage or retrieval systems, without prior written permission
`from the Publisher. All rights reserved.
`
`Not for commercial use or unauthorized distribution
`
`Exhibit 1086
`ARGENTUM
`IPR2017-01053
`
`
`
`Downloaded from Downloaded from content.healthaffairs.orgcontent.healthaffairs.org by Health AffairsHealth Affairs on April 18, 2014 on April 18, 2014 by
`
`
`
`
`
`
`
`
`at NAVIGANTat NAVIGANT
`
`
`
`
`
`
`
`000001
`
`

`

`U . S . T r e n d s
`
`Prescription Drug Spending
`Trends In The United States:
`Looking Beyond The Turning
`Point
`
`The drug spending trends observed in the 1980s, 1990s, and the first
`few years of this decade have changed dramatically in the past five
`years—bringing both opportunity and threat.
`
`by Murray Aitken, Ernst R. Berndt, and David M. Cutler
`
`ABSTRACT: Annual growth in real prescription drug spending averaged 9.9 percent during
`1997–2007 but has slowed since 2003, falling to 1.6 percent in 2007. More patent expira-
`tions, increased generic penetration, and reduced new product innovations have contrib-
`uted to this turning point. We document trends and identify underlying components: de-
`clines in the role of blockbuster drugs, increased importance of biologics and vaccines
`relative to traditional pharmaceuticals, and a changing medication mix away from those
`prescribed principally by primary care physicians toward those mostly prescribed by special-
`ists. We conclude with policy implications. [Health Affairs 28, no. 1 (2009): w151–w160
`(published online 16 December 2008; 10.1377/hlthaff.28.1.w151)]
`
`Ad j u s t e d f o r i n f l at i o n, u. s . s p e n d i n g o n prescription drugs grew
`
`9.9 percent annually between 1997 and 2007—tripling in total real spend-
`ing.1 Since 2003, however, growth rates have declined rapidly, and in 2007
`spending grew but 1.6 percent—the slowest since 1974, the only decline on record
`(Exhibit 1).
`Although comparable 2007 national data on other health-sector spending are
`not yet available, prescription drug spending growth is likely to be lower than any
`other major medical care sector. Whereas prescription drug costs were once the
`bane of payers, that concern has now been replaced by worries about hospital
`care, imaging, and professional services.2
`What accounts for the decline in the growth of overall drug spending? Do re-
`
`Murray Aitken is senior vice president, Healthcare Insight, at IMS Health in Norwalk, Connecticut. Ernst Berndt
`(eberndt@mit.edu) is the Louis E. Seley Professor in Applied Economics, Alfred P. Sloan School of Management, at
`the Massachusetts Institute of Technology in Cambridge. David Cutler is a professor of economics at Harvard
`University in Cambridge.
`
`H E A LT H A F F A I R S ~ We b E x c l u s i v e
`
`w 1 5 1
`
`DOI 10.1377/hlthaff.28.1.w151 ©2008 Project HOPE–The People-to-People Health Foundation, Inc.
`
`Downloaded from
`
`content.healthaffairs.org
` by
`at NAVIGANT
`
`Health Affairs
`
` on April 18, 2014
`
`000002
`
`

`

`D r u g S p e n d i n g
`
`EXHIBIT 1
`Size And Growth Of The U.S. Retail Pharmaceutical Market, 1997–2007
`Billions of dollars in sales
`250
`
`Percent change
`25
`
`20
`
`15
`
`10
`
`5 0
`
`Percent change
`
`200
`
`150
`
`100
`
`50
`
`0
`
`1997
`
`1998
`
`1999
`
`2000
`
`2001
`
`2002
`
`2003
`
`2004
`
`2005
`
`2006
`
`2007
`
`SOURCE: IMS Health, National Sales Perspectives, December 2007 (sales deflated by implicit gross domestic product deflator,
`$2000).
`NOTES: Dollar figures (bars) relate to the left-hand axis. Percent change (line) relates to the right-hand axis.
`y
`y
`
`cent trends suggest a new era of low growth? What are the policy implications of
`the turning point? We explore these issues here. Our data come from the National
`Sales Perspectives (NSP), which audits sales of pharmaceutical products from
`wholesalers to pharmacies and other outlets, and the National Prescription Audit
`(NPA), which tracks prescriptions dispensed by pharmacists; both are produced
`by IMS Health.
`Components Underlying Changing Trends In Prescription
`Drug Sales
`Underlying the trends in overall drug sales are several dynamics driven by
`changes in “blockbuster” drugs; shifts in the mix of medications between primary
`care and specialist drugs; and changes in the mix among traditional chemical-
`based pharmaceuticals, biologics, and vaccines.
`n Blockbuster drugs. The number of blockbuster drugs—those selling in ex-
`cess of $1 billion (real 2000 dollars) in the United States—increased more than
`eightfold between 1997 and 2006, from six to fifty-two (Exhibit 2).3 Concomitantly,
`spending on blockbusters increased from about 12 percent of all sales in 1996 to al-
`most half of all sales in 2006, accounting for three-quarters of prescription drug
`spending growth over the same time period.
`In 2007, for the first time, the number of billion-dollar products fell—from fifty-
`two to forty-eight—and their share of all sales also fell slightly, to 44 percent. As
`more blockbusters go off patent and fewer new ones are developed, the share of
`sales attributable to blockbuster molecules will likely decline still further.
`n Primary care and specialist drugs. A marked change has occurred in the mix
`of medications away from those prescribed principally by primary care physicians
`
`w 1 5 2
`
`1 6 D e c e m b e r 2 0 0 8
`
`Downloaded from
`
`content.healthaffairs.org
` by
`at NAVIGANT
`
`Health Affairs
`
` on April 18, 2014
`
`000003
`
`

`

`U . S . T r e n d s
`
`EXHIBIT 2
`Number Of Blockbuster Drugs, 1996–2007
`Number
`50
`
`40
`
`30
`
`20
`
`10
`
`0
`
`1996
`
`1997
`
`1998
`
`1999
`
`2000
`
`2001
`
`2002
`
`2003
`
`2004
`
`2005
`
`2006
`
`2007
`
`SOURCE: IMS Health, Market Insights Analysis, December 2007.
`NOTE: Blockbuster drugs are those exceeding $1 billion in sales per year in 2000 dollars.
`
`and toward those prescribed mostly by specialists.4 In 2007, the five leading primary
`care–driven therapeutic classes (by dollars) were the lipid regulators, acid pump in-
`hibitors, respiratory agents, antidepressants, and oral antidiabetics. Together they
`accounted for 22 percent of drug spending. Primary care drugs as a whole accounted
`for 55 percent of all sales. Leading specialist drug therapeutic classes included
`oncologics, antipsychotics, anti-epileptics, erythropoetins, and autoimmune agents.
`These five categories accounted for 20 percent of all drug spending, while specialist
`drugs as a whole accounted for 45 percent.
`Notably, real spending growth in primary care–driven drugs fell steadily be-
`tween 2003 and 2005, from 6.4 percent in 2003 to –0.8 percent in 2005, increasing
`temporarily to 1.9 percent in 2006, but then declining by 3.7 percent in 2007 (Ex-
`hibit 3). In sharp contrast, specialist-driven real drug spending grew 17.5 percent
`in 2003, slowed to 7.7 percent in 2005 and then rebounded to 9.5 percent in 2006
`and 8.9 percent in 2007. The reduction in overall prescription drug sales growth is
`therefore due entirely to slower growth and even declines in sales of primary care
`drug classes.
`n Pharmaceuticals, biologics, and vaccines. Changes are also apparent in the
`mix among traditional pharmaceuticals, biologics, and vaccines. Traditional phar-
`maceuticals are “small molecule” drugs, in contrast to larger-protein biologics (de-
`fined as medications manufactured via recombinant DNA technology) and vaccines.
`The most significant biologic molecules are oncologics; they are significant for their
`targeted approach to slowing cancer progression and for their high cost of treat-
`ment: According to IMS Health data, Avastin (colorectal cancer) cost on average
`$42,960, Herceptin (breast cancer) cost $27,990, and Tykerb (breast cancer) cost
`$16,575 per course of treatment in 2007.
`The price of oncologics has been increasing over time. The most expensive drug
`
`H E A LT H A F F A I R S ~ We b E x c l u s i v e
`
`w 1 5 3
`
`Downloaded from
`
`content.healthaffairs.org
` by
`at NAVIGANT
`
`Health Affairs
`
` on April 18, 2014
`
`000004
`
`

`

`EXHIBIT 3
`Size And Growth Of The U.S. Primary Care–Driven And Specialist-Driven Prescribing
`Markets, 1997–2007
`Billions of dollars
`250
`
`Percent change
`40
`
`Spending, primary care–driven
`Spending, specialist-driven
`Growth, specialist-driven
`Growth, primary care–driven
`
`D r u g S p e n d i n g
`
`200
`
`150
`
`100
`
`50
`
`0
`
`30
`
`20
`
`10
`
`0
`
`–10
`
`1997
`
`1998
`
`1999
`
`2000
`
`2001
`
`2002
`
`2003
`
`2004
`
`2005
`
`2006
`
`2007
`
`SOURCE: IMS Health, National Sales Perspectives, December 2007 (sales deflated by implicit gross domestic product deflator,
`$2000).
`NOTES: Dollar figures (bars) relate to the left-hand axis. Percent change (lines) relates to the right-hand axis.
`y
`y
`
`in the early 1990s was Taxol (used for treating breast cancer), which sold for
`$4,000 per year. The cost of Avastin today is ten times higher.
`Vaccines, once a neglected sector, have recently become much more important.
`Prevnar, a conjugate pneumococcal vaccine, and Gardasil, for prevention of cervi-
`cal cancer, are the first two blockbuster vaccines, with the current private-sector
`price being $84 and $125 per dose, respectively, for the three-dose regimen.5
`The decomposition among pharmaceuticals, biologics, and vaccines corre-
`sponds as well to drugs that are mostly self-administered (small-molecule phar-
`maceutical tablets and capsules) versus therapies primarily administered by
`health care providers (biologics and vaccines, injected or infused).
`Between 2002 and 2007, real spending on biologics grew at an average annual
`rate of 16 percent, while vaccine spending grew 19.3 percent annually. In compari-
`son, sales of traditional small-molecule drugs grew only 3.7 percent annually.
`Overall, biologics’ share of the market rose from 9 percent in 2002 to 15 percent in
`2007, while vaccine sales grew from less than 1 percent in 2002 to 2 percent in
`2007. Molecule types are also related to the specialty of the prescribing physician.
`Almost all biologics are prescribed by specialists and a sizable portion of spending
`in specialty-driven biologics is for oncology products. Thus, the growth of bio-
`logics and the shift to specialty-physician therapies are intimately related.
`Causes Of Change
`Underlying these trends in sales are dramatic changes in pharmaceutical inno-
`vation, along with a transformed market environment.
`n Pharmaceutical innovation. Despite remarkable advances in our under-
`
`w 1 5 4
`
`1 6 D e c e m b e r 2 0 0 8
`
`Downloaded from
`
`content.healthaffairs.org
` by
`at NAVIGANT
`
`Health Affairs
`
` on April 18, 2014
`
`000005
`
`

`

`U . S . T r e n d s
`
`standing of biology and genetics over the past decade, recent years have seen a
`marked decline in the number of new molecular entities (NMEs) approved by the
`Food and Drug Administration (FDA). According to the FDA, between 1999 and
`2001 the average total number of such new product approvals was about thirty-five
`per year (six biologics and twenty-nine pharmaceuticals), whereas between 2005
`and 2007 this number fell to about twenty (three biologics and seventeen pharma-
`ceuticals).
`With smaller numbers of new product approvals, the vintage composition of
`drugs sold has matured and has become increasingly vulnerable to generic entry.
`Based on IMS Health NPS data, we calculate that products introduced within the
`prior five years accounted for 34 percent of total drug sales in 1999. That share has
`declined steadily since then, to just 19 percent of total sales in 2007. Meanwhile,
`the value of brand-name products at risk of same-molecule generic penetration
`has almost doubled, from an average of about $9 billion per year between 2002
`and 2005 to about $16 billion in 2006–07. The list of drugs losing patent protec-
`tion in recent years has been substantial: Norvasc (value: $2.6 billion), Lotrel ($1.5
`billion), and Flonase ($1.2 billion). Moreover, drugs likely to come off patent pro-
`tection soon include Cozaar in 2010; Lipitor, Plavix, and Seroquel in 2011; and
`Diovan, Viagra, and Evista in 2012.
`n The changing environment for sales. Drugs having patent protection and
`extensive market power continue to command high prices. But in therapy classes
`where there are multiple treatment options, competition has increased—across
`branded molecules, and between branded and generics. Employers and the pharma-
`ceutical benefit management (PBM) firms with which they contract have increas-
`ingly moved to more sophisticated formularies in an effort to limit spending. IMS
`NPA data indicate that a typical formulary now charges $6 for generic medications,
`$29 for preferred branded drugs, and $40 or more for nonpreferred branded drugs.
`This tier structure creates enormous incentives for consumers to take generic medi-
`cations. Medicare Part D has contributed to this trend, with most plans having at
`least three-tier copayment formularies and many having a fourth tier incorporating
`sizable coinsurance payments. All told, the generic share of total prescriptions
`increased from 51 percent in 2002 to 67 percent in 2007 (Exhibit 4).
`Even within Medicare Part D’s short history, the total retail prescription vol-
`ume share dispensed as generic has steadily increased, from 59 percent in January
`2006 to 68 percent in December 2007. Generic penetration has also become more
`rapid. According to IMS NPS data, in 2002 branded products retained 28 percent
`of their prescription volume twelve months after patent expiry. In 2007 that figure
`dropped to 14 percent.
`The increased extent and speed of generic penetration has resulted in substan-
`tial cost savings for purchasers. The daily cost of drug therapy across all products
`in that class fell 32 percent for lipid regulators in the year after generic entry, 32
`percent for bisphosphonates, 42 percent for selective serotonin reuptake inhibi-
`
`H E A LT H A F F A I R S ~ We b E x c l u s i v e
`
`w 1 5 5
`
`Downloaded from
`
`content.healthaffairs.org
` by
`at NAVIGANT
`
`Health Affairs
`
` on April 18, 2014
`
`000006
`
`

`

`D r u g S p e n d i n g
`
`EXHIBIT 4
`Brand-Name And Generic Drugs’ Share Of Total Retail Dispensed Prescription Drugs,
`1998–2007
`
`Percent of all prescriptions
`100
`
`Unbranded generics
`Brands
`
`80
`
`60
`
`40
`
`20
`
`0
`
`1998
`
`1999
`
`2000
`
`2001
`
`2002
`
`2003
`
`2004
`
`2005
`
`2006
`
`2007
`
`SOURCE: IMS Health, National Prescription Audit, December 2007.
`
`tors (SSRIs), and 20 percent for calcium-channel blockers.
`To quantify the expenditure impact of increased generic penetration, we simu-
`lated spending if the generic efficiency rate (for all molecules for which a generic
`version is available, the proportion of units dispensed as generics) had stayed at its
`actual 2003 rate (77.3 percent) instead of increasing to 86.4 percent, which it did
`by 2007.6 Cumulative pharmaceutical spending would have been 13.5 percent
`greater (22 percent higher in 2007 alone).
`n Statins and Lipitor: overturning conventional wisdom. It has long been con-
`ventional wisdom that after a drug loses patent protection and generic entry occurs,
`the total branded plus unbranded number of prescriptions/extended units for the
`same molecule decreases, mainly because promotional spending by the brand-name
`company drops around the time of patent expiration.7 For the first time in recent
`history, this conventional wisdom has been overturned by cholesterol-lowering
`“statin” drugs.
`No pharmaceutical shows dramatic changes in the market better than Lipitor,
`the best-selling statin from Pfizer. Lipitor was the ultimate blockbuster. At its
`peak in 2006, it had an average price of $2.79 per day and generated $8.6 billion in
`U.S. sales ($13.6 billion internationally).
`Lipitor is but one statin drug. The second leading seller was Zocor, an earlier
`entrant. Zocor (generic: simvastatin) lost patent protection and faced generic
`competition beginning 23 June 2006. As a result of provisions of the Hatch-
`Waxman Act known as Paragraph IV certifications as well as subsequent judicial
`rulings against Merck (the manufacturer of Zocor), the generic company Teva was
`awarded exclusive rights to market the 10 mg, 20 mg, and 40 mg versions of
`simvastatin for 180 days after Merck’s patent expiration. Similarly, Ranbaxy ob-
`
`w 1 5 6
`
`1 6 D e c e m b e r 2 0 0 8
`
`Downloaded from
`
`content.healthaffairs.org
` by
`at NAVIGANT
`
`Health Affairs
`
` on April 18, 2014
`
`000007
`
`

`

`U . S . T r e n d s
`
`tained exclusive rights to market the 80 mg version. On its own, Merck reached an
`agreement in early 2006 with the generic firm Dr. Reddy to produce an “autho-
`rized generic” in all strength versions following Zocor’s loss of patent protection.8
`Although a limited amount of generic simvastatin entry occurred after 23 June
`2006, unfettered generic entry occurred 180 days later, in late December 2006.
`Other statins have gone off patent as well. Mevacor (lovastatin) lost its patent pro-
`tection in 2001, and Pravachol (pravastatin) went off patent 25 April 2006. Ge-
`neric versions of both came out rapidly thereafter.
`Although some controversy still exists, general consensus among the medical
`community is that for most patients, the various statins are equally effective and
`safe, and thus are therapeutically substitutable. An exception is at very high dos-
`ages, where Lipitor is believed to be more effective.9 Since brand-name Lipitor was
`still patent-protected in early 2007, whereas much less costly generic versions of
`Pravachol (pravastatin) and Zocor (simvastatin) were now on the market, payers,
`insurers, and PBMs were given incentives to switch patients on Lipitor to
`pravastatin or simvastatin. This typically took the form of moving Lipitor to the
`highest copayment tier and placing the two generics in the lowest tier.
`For Pravachol and generic pravastatin, the total brand plus generic number of
`prescriptions since loss of patent protection increased only slightly (Exhibit 5).
`After Zocor lost patent protection, however, total monthly Zocor plus generic
`simvastatin prescriptions boomed, from 2.8 million in June 2006 to 4.8 million in
`December 2007. Sales of prescription Zocor plummeted, but those of generic
`
`EXHIBIT 5
`Monthly Prescribing Trends In The Statin Therapeutic Class, Brand-Name And Generic,
`January 2006–December 2007
`Millions prescribed
`
`Lipitor
`
`Zocor + simvastatin
`
`Zocor
`
`Simvastatin
`
`Pravachol
`
`Pravastatin
`
`567
`
`4
`
`3
`
`2
`
`1
`
`0
`
`1/2006
`
`7/2006
`
`1/2007
`
`7/2007
`
`SOURCE: IMS Health, National Prescription Audit, December 2007.
`
`H E A LT H A F F A I R S ~ We b E x c l u s i v e
`
`w 1 5 7
`
`Downloaded from
`
`content.healthaffairs.org
` by
`at NAVIGANT
`
`Health Affairs
`
` on April 18, 2014
`
`000008
`
`

`

`D r u g S p e n d i n g
`
`simvastatin grew dramatically.
`The new sales have come directly from previous Lipitor users or people who
`would have started on Lipitor. In 2007 the number of prescriptions of Lipitor fell
`12 percent, including 26 percent in new starts. Sales of Lipitor have declined the
`most at lower dosages—10 mg and 20 mg per day—and have held steady only for
`80 mg doses. Between 2006 and 2007, domestic sales of Lipitor fell 6.5 percent be-
`low 2006 levels. The Lipitor experience is the first instance in which generic ver-
`sions of one molecule have substituted so significantly for brand-name versions of
`a different molecule.
`Policy Implications
`These changing trends and their underlying causes have several implications for
`policy. First, they are consistent with the view that for payers and consumers, the
`health spending prospects are more optimistic than many fear. Costs of prescrip-
`tion pharmaceuticals—an important segment of health care—are rising very
`slowly or even falling. Unless the situation changes unexpectedly in the near fu-
`ture, this trend will continue. Current projections have not taken this reduced
`spending growth into account. For example, the Centers for Medicare and Medic-
`aid Services (CMS) recently projected 8.5 percent pharmaceutical spending
`growth in 2006, 6.7 percent in 2007, 6.8 percent in 2008, and an average annual
`growth rate of 8.2 percent between 2006 and 2017.10 Our data suggest that these
`forecasts are too high.
`Second, the converse of this implies difficult times for the pharmaceutical in-
`dustry, particularly for traditional small-molecule manufacturers. Expected fu-
`ture revenue is one factor affecting pharmaceutical research and development
`(R&D) and innovation.11 Slow sales growth is likely to put pressure on research
`budgets and marketing costs and to create incentives for mergers. The existence of
`fewer, larger entities with tighter research budgets may stifle or limit investment
`in innovation and the ongoing prospects for improved therapeutics’ reaching the
`market. Unless biopharmaceutical R&D productivity improves or results in an in-
`creased proportion of “blockbuster” molecules affecting large populations (the
`latter an unlikely outcome, given recent trends), reduced revenues are likely to
`constrain future rates of new product innovation.
`Third, our simulation results document that sizable cost savings can be at-
`tained by increasing generic efficiency rates; greater use of generics when available
`since 2003 has resulted in 22 percent lower pharmaceutical spending in 2007.
`More generally, our results suggest that the design of drug cost sharing is ex-
`tremely important. For the large number of drugs for which there is competition
`across branded molecules, and especially between branded and generic drugs,
`out-of-pocket costs have a major influence on what patients consume, and per-
`haps on their health outcomes.12 Used judiciously, cost-sharing instruments can
`be employed by governments, PBMs, and private payers in future attempts to limit
`
`w 1 5 8
`
`1 6 D e c e m b e r 2 0 0 8
`
`Downloaded from
`
`content.healthaffairs.org
` by
`at NAVIGANT
`
`Health Affairs
`
` on April 18, 2014
`
`000009
`
`

`

`U . S . T r e n d s
`
`“Is the reduction in new blockbusters the result of technological
`wells running dry, or the implication of increased regulation?”
`
`growth in pharmaceutical spending.
`A major exception to this rule may be drugs that have extensive market power
`in therapeutic segments where not taking the medication can result in death. In
`such segments—oncology in particular—recent launch prices have been high and
`increasing over time. Pressures to address such costs will increase, but govern-
`ments do not have competition as an effective response lever. This has led some
`policy analysts to recommend that in those situations, Medicare should establish
`a temporary administered-price mechanism.13 How Medicare can best deal with
`the pricing of truly unique and innovative life-saving new drugs is likely to be-
`come an issue generating considerable controversy.
`Fourth, our results are likely to increase pressures placed on agencies such as
`the FDA. Is the reduction in new blockbusters the result of technological wells
`running dry, or the implication of increased regulatory stringency? Should the
`FDA be doing something about this, or has it done all it should or can do? Should
`industry focus on “niche busters” rather than “blockbusters,” searching for more
`stratified medicines?14 These questions have always lingered in the background of
`pharmaceutical policy, and they may soon come to the forefront.
`Fifth, the focus in the past few years on reducing rates of growth of drug spend-
`ing is now coinciding with the loss of exclusivity for a substantial portion of that
`spending. This suggests that further efforts to limit the uptake of new therapies
`through extension of formulary design to tier four, switching to coinsurance
`rather than copayments, or reducing the effective period of exclusivity for prod-
`ucts might not be necessary. Moreover, the long-term impacts of these measures
`on both the cost and the quality of health outcomes remain unknown.
`What is clear, however, is that the prescription drug spending trends observed
`in the 1980s, 1990s, and the first few years of this decade have changed dramati-
`cally in the past five years, and that when one looks beyond the recent turning
`point, the growth, size, and composition of prescription drug spending is likely to
`be dramatically different, raising both policy opportunities and dangers.
`
`Ernst Berndt acknowledges research support from the Merck Pharmaceutical Policy Program; David Cutler, from
`the National Institute on Aging.
`
`H E A LT H A F F A I R S ~ We b E x c l u s i v e
`
`w 1 5 9
`
`Downloaded from
`
`content.healthaffairs.org
` by
`at NAVIGANT
`
`Health Affairs
`
` on April 18, 2014
`
`000010
`
`

`

`D r u g S p e n d i n g
`
`NOTES
`1. All inflation adjustments use the gross domestic product (GDP) deflator.
`2. Mark E. Miller, executive director, Medicare Payment Advisory Commission, “MedPAC Recommenda-
`tions on Imaging Services,” Statement before the House Ways and Means Subcommittee on Health, 17
`March 2005, http://www.medpac.gov/publications/congressional_testimony/031705_TestimonyImaging-
`Hou.pdf (accessed 15 August 2008).
`3. Statistics are sometimes presented using nominal data—the number of drugs with sales of more than a
`billion dollars using prices from that year. By this metric, the increase in blockbusters was tenfold—six to
`sixty.
`4. This classification is based on prescription shares by prescriber type, not dollar shares.
`5. Centers for Disease Control and Prevention, “CDC Vaccine Price List,” updated 8 October 2008, http://
`www.cdc.gov/vaccines/programs/vfc/cdc-vac-price-list.htm (accessed 4 November 2008).
`6. We assume an unchanged actual market size in units and unchanged actual brand-generic average prices
`per extended unit (tablet, capsule, and so on).
`7. R. Caves, M. Whinston, and M. Hurwitz, “Patent Expiration, Entry, and Competition in the U.S. Pharma-
`ceutical Industry,” Brookings Papers on Economic Activity 1 (1991): 1–48; E. Berndt, I. Cockburn, and Z.
`Griliches, “Pharmaceutical Innovations and Market Dynamics: Tracking Effects on Price Indexes for Anti-
`depressant Drugs,” Brookings Papers on Economic Activity: Microeconomic 2 (1996): 133–188; and E. Berndt, M.
`Kyle, and D. Ling, “The Long Shadow of Patent Expiration: Generic Entry and Rx-to-OTC Switches,” in
`Scanner Data and Price Indexes, ed. R. Feenstra and M. Shapiro (Chicago: University of Chicago Press, 2003),
`229–267.
`8. On authorized generics and Paragraph IV certifications, see E.R. Berndt et al., “Authorized Generic Drugs,
`Price Competition, and Consumers’ Welfare,” Health Affairs 26, no. 3 (2007): 790–799. Information regard-
`ing authorized generics for Zocor is found in “Recent Key Generic Launches—A Focus on Zocor,” RxWatch
`(Summer 2006): 1.
`9. P. Jones et al., “Comparative Dose Efficacy Study of Atorvastatin versus Simvastatin, Pravastatin,
`Lovastatin, and Fluvastatin in Patients with Hypercholesterolemia (the CURVES Study),” American Journal
`of Cardiology 81, no. 5 (1998): 582–587; and “Drug Comparisons—HMG-CoA Reductase Inhibitors,” Drug
`Digest, 2007, http://www.drugdigest.org/DD/PrintablePages/Comparisons/1,20038,37-15,00.html (ac-
`cessed 4 November 2008).
`10. See Exhibit 2 in S. Keehan et al., “Health Spending Projections through 2017: The Baby-Boom Generation
`Is Coming to Medicare,” Health Affairs 27, no. 2 (2008): w145–w155 (published online 26 February 2008;
`10.1377/hlthaff.27.1.w145); A. Catlin et al., “National Health Spending in 2006: A Year of Change for Pre-
`scription Drugs,” Health Affairs 27, no. 1 (2008): 14–29; and Congressional Budget Office, The Long-Term Out-
`look for Health Care Spending, Pub. no. 3085 (Washington: CBO, November 2007).
`11. D. Acemoglu and J. Linn, “Market Size in Innovation: Theory and Evidence from the Pharmaceutical In-
`dustry,” Quarterly Journal of Economics 119, no. 3 (2004): 1049–1090.
`12. D. Goldman, G. Joyce, and Y. Zheng, “Prescription Drug Cost Sharing: Associations with Medication and
`Medical Utilization and Spending and Health,” Journal of the American Medical Association 298, no. 1 (2007):
`61–69.
`13. R. Frank and J. Newhouse, “Mending the Medicare Prescription Drug Benefit: Improving Consumer
`Choices and Restructuring Purchasing,” Hamilton Project Discussion Paper no. 2007-03 (Washington:
`Brookings Institution, April 2007).
`14. M.R. Trusheim, E.R. Berndt, and F.L. Douglas, “Stratified Medicine: Strategic and Economic Implications
`of Combining Drugs and Clinical Biomarkers,” Nature Reviews: Drug Discovery 6, no. 4 (2007): 287–293.
`
`w 1 6 0
`
`1 6 D e c e m b e r 2 0 0 8
`
`Downloaded from
`
`content.healthaffairs.org
` by
`at NAVIGANT
`
`Health Affairs
`
` on April 18, 2014
`
`000011
`
`