`
`l~l __________ sP_ E_ c _rA_ L_A_R_ r _rc_ L_E _ _ __ _ __ _ _ _.II
`
`A Decade of Direct-to-Consumer Advertising
`of Prescription Drugs
`
`Julie M. Donohue, Ph.D., Marisa Cevasco, B.A., and Meredith B. Rosenthal, Ph.D.
`
`ABSTRACT
`
`BACKGROUND
`Evidence suggests that direct-to-consumer advertising of prescription drugs increases
`pharmaceutical sales and both helps to avert underuse of medicines and leads to
`potential overuse. Concern about such advertising has increased recently owing to
`the withdrawal from the market of heavily advertised drugs found to carry serious
`risks. Moreover, the Food and Drug Administration (FDA) has been criticized for its
`weak enforcement of laws regulating such advertising.
`
`METHODS
`We examined industry-wide trends in spending by pharmaceutical companies on di(cid:173)
`rect-to-consumer advertising and promotion to physicians during the past decade. We
`characterized the drugs for which such advertising is used and assessed the timing
`of advertising after a drug is introduced. Finally, we examined trends in the FDA's
`regulation of drug advertising.
`
`RESULTS
`Total spending on pharmaceutical promotion grew from $11.4 billion in 1996 to $29.9
`billion in 2005. Although during that time spending on direct-to-consumer advertis(cid:173)
`ing increased by 330%, it made up only 14% of total promotional expenditures in 2005.
`Direct-to-consumer campaigns generally begin within a year after the approval of a
`product by the FDA. In the context of regulatory changes requiring legal review be(cid:173)
`fore issuing letters, the number ofletters sent by the FDA to pharmaceutical manu(cid:173)
`facturers regarding violations of drug-advertising regulations fell from 142 in 1997
`to only 21 in 2006.
`
`CONCLUSIONS
`Spending on direct-to-consumer advertising has continued to increase in recent years
`in spite of the criticisms leveled against it. Our findings suggest that calls for a mora(cid:173)
`torium on such advertising for new drugs would represent a dramatic departure from
`current practices.
`
`From the Department of Health Policy
`and Management, University of Pittsburgh
`Graduate School of Public Health, Pitts(cid:173)
`burgh (J.M.D.); the Department of Health
`Policy and Management, Harvard School
`of Public Health, Boston (M.C., M.B.R.);
`and Vanderbilt School of Medicine, Nash(cid:173)
`ville (M.C.). Address reprint requests to
`Dr. Donohue at the Department of Health
`Policy and Management, University of
`Pittsburgh Graduate School of Public
`Health, Crabtree Hall A613, 130 DeSoto St.,
`Pittsburgh, PA 15261, or at jdonohue@
`pitt.edu.
`
`N Engl J Med 2007;357:673-81.
`Copyright © 2007 Massachusetts Medical Society.
`
`N ENGLJ MED 357;7 WWW.NEJM.0RG AUGUST 16, 2007
`
`673
`
`The New England Journal of Medicine
`Downloaded from nejm.org on May 5, 2015. For personal use only. No other uses without permission.
`Copyright© 2007 Massachusetts Medical Society. All rights reserved.
`
`Sanofi Exhibit 2210.001
`Mylan v. Sanofi
`IPR2018-01675
`
`
`
`The NEW ENGLAND JOURNAL of MEDICINE
`
`IT HAS BEEN 10 YEARS SINCE A CHANGE IN A
`
`policy of the Food and Drug Administration
`(FDA) allowed direct-to-consumer advertising
`of prescription drugs on television. Such advertis(cid:173)
`ing has been criticized for encouraging inappro(cid:173)
`priate use of medications and driving up drug
`spending. 1 ,2 Concern that such advertising may
`lead to increased use of expensive medications was
`amplified by the introduction of a prescription(cid:173)
`drug benefit in Medicare in 2006 (Part D). Studies
`of the effect of advertising on prescribing prac(cid:173)
`tices have shown that such advertising increases
`classwide sales, helps to avert underuse of medi(cid:173)
`cines to treat chronic conditions, and leads to some
`overuse of prescription drugs. 3 - 5
`Direct-to-consumer advertising has also been
`controversial in light of postmarketing revelations
`regarding problems with drug safety. Specifically,
`clinical trials that are required for drug approval
`are typically not designed to detect rare but sig(cid:173)
`nificant adverse effects, and contemporary meth(cid:173)
`ods of postmarketing surveillance often fail to
`connect adverse events that have a high rate of
`background prevalence with the use of particu(cid:173)
`lar drugs. After the market withdrawal of Vioxx
`(rofecoxib), a drug heavily promoted to consum(cid:173)
`ers,6 critics called for the FDA to place limits on
`direct-to-consumer advertising, particularly for
`new drugs,7 a view that was reiterated in a recent
`report by the Institute of Medicine on the safety
`of medicines. 8
`Finally, the Government Accountability Office
`(GAO)9 and others10 have criticized the FDA's en(cid:173)
`forcement of regulations governing direct-to-con(cid:173)
`sumer advertising. Criticism has focused specifi(cid:173)
`cally on the adequacy of the FDA's review of
`pharmaceutical advertisements, as well as the level
`and speed of enforcement actions taken subse(cid:173)
`quent to review.
`Since direct-to-consumer advertising has a sig(cid:173)
`nificant effect on demand for prescription drugs,
`it is important to understand the evolution of such
`advertising and its regulation. Although one study
`reported that spending for such advertising in(cid:173)
`creased by a factor of 3 from 1996 to 2000,11 little
`is known about trends in spending and other
`forms of pharmaceutical promotion in recent
`years. In our study, we examined recent trends in
`the industry's use of direct-to-consumer advertis(cid:173)
`ing (as opposed to other forms of promotion), as(cid:173)
`sessed the timing of advertising campaigns rela(cid:173)
`tive to the introduction of drugs in order to shed
`
`light on safety issues, and examined trends in the
`FDA's regulation of drug advertising during the
`past decade.
`
`METHODS
`
`DATA COLLECTION
`We obtained data on industry-wide and product(cid:173)
`specific promotional expenditures from three mar(cid:173)
`ket-research firms that track advertising spend(cid:173)
`ing and specialize in forms of promotion for the
`pharmaceutical industry; we also obtained infor(cid:173)
`mation from researchers and staff members at the
`FDA and other government agencies. These data
`have been widely used in studies of trends in and
`the effects of direct-to-consumer advertising.3,s,1 1-14
`Data on expenditures for such advertising were
`collected by TNS Media, which tracks local and
`national advertising campaigns at 44 television
`networks (including cable), 658 magazines, 202
`newspapers, the Internet, and several network and
`local radio stations. Data are representative of ma(cid:173)
`jor media markets.
`We obtained publicly available data on promo(cid:173)
`tion to health professionals from 1996 to 2005
`from IMS Health, an independent medical-infor(cid:173)
`mation company. For the industry as a whole, we
`report on three major components of spending on
`promotion to professionals: visits of pharmaceu(cid:173)
`tical sales representatives to physicians in office(cid:173)
`based and hospital practices ("detailing"), free
`samples dispensed to physicians, and advertising
`in professional journals. IMS Health derives spend(cid:173)
`ing estimates on detailing from a nationally rep(cid:173)
`resentative panel of office-based physicians and
`hospital pharmacy directors who track their con(cid:173)
`tacts with sales representatives. IMS Health ob(cid:173)
`tains data on spending on free samples from a
`panel of approximately 1200 office staff members
`in medical practices, sampled from the practices
`of the office-based physicians who are on the de(cid:173)
`tailing panel. To estimate spending on advertising
`in professional journals, IMS Health tracks adver(cid:173)
`tisements placed in approximately 400 medical
`journals and adds estimates of printing costs to
`the publisher's charge for the advertisements.
`We obtained data on industry-wide sales from
`published reports on the basis of an annual sur(cid:173)
`vey conducted by the Pharmaceutical Research and
`Manufacturers of America (PhRMA). We pur(cid:173)
`chased data on promotional expenditures in 2005
`for products in specific classes from Verispan,
`
`674
`
`N ENGLJ MED 357;7 WWW.NEJM.0RG AUGUST 16, 2007
`
`The New England Journal of Medicine
`Downloaded from nejm.org on May 5, 2015. For personal use only. No other uses without permission.
`Copyright© 2007 Massachusetts Medical Society. All rights reserved.
`
`Sanofi Exhibit 2210.002
`Mylan v. Sanofi
`IPR2018-01675
`
`
`
`A DECADE OF DIRECT-TO-CONSUMER PRESCRIPTION-DRUG ADVERTISING
`
`another independent medical-information com(cid:173)
`pany, and from TNS Media. For the 10 therapeu(cid:173)
`tic drug classes that had the highest U.S. sales in
`2004, we obtained data on the five forms of phar(cid:173)
`maceutical promotion that are tracked by Verispan:
`direct-to-consumer advertising, detailing, advertis(cid:173)
`ing in professional journals, meetings and educa(cid:173)
`tional events for physicians, and online pharma(cid:173)
`ceutical promotion to physicians. Data regarding
`spending on advertising are collected by TNS Me(cid:173)
`dia, as described previously. To track detailing,
`Verispan surveys approximately 13,000 office(cid:173)
`based and hospital-based physicians and residents,
`nurse practitioners, and physician assistants who
`track their encounters with pharmaceutical sales
`representatives. The panel is geographically rep(cid:173)
`resentative and includes members of 31 clinical
`specialties.
`Verispan produces estimates of industry ex(cid:173)
`penditures on professional meetings and events
`through a survey of more than 3500 office-based
`physicians representing 19 specialties who report
`on the events sponsored by pharmaceutical com(cid:173)
`panies that they attend. This panel of physicians
`is also asked to report on online pharmaceutical(cid:173)
`promotion activity, which includes digital (Inter(cid:173)
`net and video) promotion and continuing medical
`education modules. Verispan audits approximate(cid:173)
`ly 600 medical journals and tabloids and calculates
`spending on the basis of each journal's rate-card
`information and premium-factor costs.
`Finally, we obtained data on the number of FDA
`enforcement actions related to pharmaceutical
`promotion from 1997 to 2006 from the FDA,
`which posts the regulatory letters sent to pharma(cid:173)
`ceutical companies on its Web site (www.fda.gov/
`cder/ddmac/lawsregs.htrn). FDA approval dates for
`specific products were obtained from the Orange
`Book of approved drug products with therapeutic
`equivalence evaluations.15 We obtained data on
`start dates for advertising campaigns through a
`series of Internet searches (with specific sources
`available from the authors).
`
`bution of promotional spending by type for the 10
`leading classes of drugs in terms of dollar sales in
`the United States. In addition, we examined the
`level and timing (relative to a drug's FDA approval)
`of spending on advertising for the 20 drugs with
`the highest spending for direct-to-consumer ad(cid:173)
`vertising in 2005.
`To characterize the nature of FDA enforcement
`related to advertising spending over time, we ex(cid:173)
`amined the numbers of enforcement letters related
`to promotion in each year and further calculated
`the percentage of promotion-related enforcement
`actions that were for advertising campaigns (as
`opposed to promotional materials aimed at health
`professionals). Finally, we examined the content of
`the notices of violation to determine the type of
`violation (e.g., false or misleading claims about the
`effectiveness or risks of drugs) and calculated the
`proportion related to each type.
`
`RESULTS
`
`INDUSTRY-WIDE TRENDS IN PROMOTION
`Total real spending on promotion grew from $11.4
`billion to $29.9 billion from 1996 to 2005, at an
`average annual rate of 10.6% (Table 1). The per(cid:173)
`centage of sales spent on promotion for the indus(cid:173)
`try as a whole increased from 14.2% in 1996 to
`18.2% in 2005. In the past 9 years, spending on
`direct-to-consumer advertising and free samples
`has risen as a share of total promotion, whereas
`investments in detailing and advertising in profes(cid:173)
`sional journals have fallen as a share of the total.
`Real spending on direct-to-consumer adver(cid:173)
`tising increased by 330% from 1996 to 2005 (Table
`1). After a brief slowdown in spending on adver(cid:173)
`tising in 2000 and 2001, spending grew at an aver(cid:173)
`age annual rate of 14.3% from 2002 to 2005. Yet,
`promotion to professionals still outweighs spend(cid:173)
`ing on direct-to-consumer advertising. In 2005,
`only 14% of total industry expenditures on phar(cid:173)
`maceutical promotion were devoted to such adver(cid:173)
`tising.
`
`DATA ANALYSES
`We conducted descriptive analyses. Data on pro(cid:173)
`motional spending were adjusted to 2005 dollars
`with the use of the Consumer Price Index. We ex(cid:173)
`amined spending on all forms of promotion rela(cid:173)
`tive to sales to determine whether the intensity of
`pharmaceutical promotional spending has changed
`during the past decade. We examined the distri-
`
`ROLE OF ADVERTISING FOR TOP-SELLING DRUGS
`In 2005, 8 of the 10 top drug classes in terms of
`dollar sales had at least one product with advertis(cid:173)
`ing spending (Table 2). The importance of direct(cid:173)
`to-consumer advertising varied substantially across
`the top classes. Manufacturers of proton-pump in(cid:173)
`hibitors, 3-hydroxy-3-methylglutaryl coenzyme A
`(HMG-CoA) reductase inhibitors (statins), and
`
`N ENGLJ MED 357;7 WWW.NEJM.0RG AUGUST 16, 2007
`
`675
`
`The New England Journal of Medicine
`Downloaded from nejm.org on May 5, 2015. For personal use only. No other uses without permission.
`Copyright© 2007 Massachusetts Medical Society. All rights reserved.
`
`Sanofi Exhibit 2210.003
`Mylan v. Sanofi
`IPR2018-01675
`
`
`
`The NEW ENGLAND JOURNAL of MEDICINE
`
`Table 1. Annual Spending on Direct-to-Consumer Advertising and Promotion to Health Professionals, 1996---2005.*
`
`Variable
`
`Direct-to-consumer
`advertising
`
`1996
`
`1997
`
`1998
`
`1999
`
`2000
`
`2001
`
`2002
`
`2003
`
`2004
`
`2005
`
`Annual Spending
`
`Total spending (millions of$)
`
`985
`
`1,301
`
`1,578
`
`2,166
`
`2,798
`
`2,954
`
`2,864
`
`3,478
`
`4,160
`
`4,237
`
`Percentage of sales
`
`Professional promotion
`
`Total spending (millions of$)
`
`1.2
`
`1.5
`
`1.6
`
`1.8
`
`2.1
`
`2.0
`
`1.9
`
`2.2
`
`2.5
`
`2.6
`
`Detailing
`
`3,747
`
`4,093
`
`4,861
`
`5,064
`
`5,447
`
`6,055
`
`6,731
`
`7,364
`
`7,585
`
`6,777
`
`Journal advertising
`
`571
`
`621
`
`597
`
`551
`
`549
`
`469
`
`474
`
`476
`
`516
`
`429
`
`4.9
`
`4.4
`
`Percentage of sales
`
`Free samples
`
`Total retail value (millions
`of$)
`
`Percentage of sales
`
`Total promotion
`
`5.4
`
`5.4
`
`5.6
`
`4.7
`
`4.6
`
`4.5
`
`4.8
`
`5.0
`
`6,104
`
`7,358
`
`7,910
`
`8,476
`
`9,021
`
`11,539
`
`12,928
`
`14,362
`
`16,404
`
`18,438
`
`7.6
`
`8.4
`
`8.1
`
`7.1
`
`6.9
`
`8.0
`
`8.6
`
`9.1
`
`9.9
`
`11.2
`
`Total spending (millions of$)
`
`11,407
`
`13,373
`
`14,946
`
`16,257
`
`17,815
`
`21,018
`
`22,997
`
`25,680
`
`28,664
`
`29,881
`
`Percentage of sales
`
`14.2
`
`15.3
`
`15.3
`
`13.7
`
`13.6
`
`14.6
`
`15.2
`
`16.3
`
`17.2
`
`18.2
`
`'' Data on promotional spending are from I MS Health (www.imshealth.com); data on sales are from PhRMA's annual report. All data were
`adjusted to 2005 dollars, according to the Consumer Price Index. Spending on free samples for 2005 was estimated on the basis of growth
`and spending rates from the previous 3 years.
`
`erythropoietin medications spent 34%, 34%, and
`31% of their total marketing budget, respectively,
`on direct-to-consumer advertising in 2005. The
`manufacturers of several drugs in these classes
`invested in advertising campaigns (Table 2). Spend(cid:173)
`ing for the advertising of antidepressant agents,
`seizure-disorder medications, and antipsychotic
`agents was lower than that for proton-pump in(cid:173)
`hibitors, statins, and erythropoietin medications
`as a proportion of the total marketing budget.
`The remaining 4 of the top 10 drug classes placed
`little emphasis on consumers in their promo(cid:173)
`tional strategies. None of the angiotensin II an(cid:173)
`tagonists used direct-to-consumer advertising in
`2005. Among manufacturers of calcium-channel
`blockers, only non-product-specific or "disease
`awareness" ads were purchased. In 2005, manu(cid:173)
`facturers used direct-to-consumer advertising for
`only one of the cyclooxygenase-2 inhibitors (of
`which celecoxib was the only remaining product)
`and one of the angiotensin-converting-enzyme in(cid:173)
`hibitors. Since data on the retail value of free sam(cid:173)
`ples that are dispensed for these drug classes were
`not available, the overall promotion-to-sales ratios
`probably provide a conservative estimate.
`
`LEVEL AND TIMING OF EXPENDITURES
`Spending on direct-to-consumer advertising con(cid:173)
`tinued to be concentrated among a relatively small
`number of brands. The 20 drugs with the highest
`spending made up 54.4% of total industry spend(cid:173)
`ing on advertising in 2005 (Table 3). Drugs that are
`advertised to consumers are predominantly new
`drugs used to treat chronic conditions. Ten of the
`top 20 drugs, as ranked by advertising spending,
`were introduced in 2000 or later. Notably, nearly
`all (17 of 20) advertising campaigns for the most
`heavily advertised drugs began within a year after
`FDA approval of the drug.
`
`FDA ENFORCEMENT OF REGULATIONS
`The number ofletters sent by the FDA to pharma(cid:173)
`ceutical manufacturers notifying them that they
`had violated regulations for prescription-drug ad(cid:173)
`vertising fell from 142 in 1997 to only 21 in 2006
`(Fig. 1). During the same period, the proportion of
`promotion-related regulatory letters citing prob(cid:173)
`lems with direct-to-consumer advertisements (as
`opposed to promotional material aimed at health
`professionals) increased from 15.5% of all letters
`in 1997 to 33.3% in 2006. And during the years
`
`676
`
`N ENGLJ MED 357;7 WWW.NEJM.0RG AUGUST 16, 2007
`
`The New England Journal of Medicine
`Downloaded from nejm.org on May 5, 2015. For personal use only. No other uses without permission.
`Copyright© 2007 Massachusetts Medical Society. All rights reserved.
`
`Sanofi Exhibit 2210.004
`Mylan v. Sanofi
`IPR2018-01675
`
`
`
`A DECADE OF DIRECT-TO-CONSUMER PRESCRIPTION-DRUG ADVERTISING
`
`Table 2. U.S. Sales Revenues and Promotional Spending for Leading Therapeutic Classes of Drugs, According to Dollar Sales in 2005.*
`
`Variable
`
`Total
`U.S. Sales Promotional Percentage
`Revenues
`Spending
`of Sales
`
`Type of Promotion
`
`Direct-to-
`Consumer
`Advertising Detailing
`
`Online
`Professional
`Promotion to
`Meetings
`Journal
`and Events Advertising Physicians
`
`No. of
`Drugs in
`Class with
`Direct-to-
`Consumer
`Advertising
`
`millions of dollars
`
`HMG-CoA reduc-
`tase inhibitors
`
`Proton-pump
`inhibitors
`
`SSRls or SN Rls
`
`Anti psychotic
`agents
`
`16,000
`
`12,900
`
`12,500
`
`10,500
`
`859
`
`884
`
`1018
`
`513
`
`Erythropoietin
`
`8,700
`
`100
`
`5
`
`7
`
`8
`
`5
`
`1
`
`34
`
`34
`
`12
`
`10
`
`31
`
`percent
`11
`
`7
`
`15
`
`21
`
`12
`
`52
`
`57
`
`68
`
`64
`
`45
`
`2
`
`1
`
`4
`
`3
`
`7
`
`1
`
`1
`
`1
`
`2
`
`5
`
`4
`
`4
`
`6
`
`4
`
`2
`
`Seizure-disorder
`agents
`
`Angiotensin 11
`antagonists
`
`Calcium-channel
`blockers
`
`ACE inhibitors
`
`COX-2 inhibitors
`
`8,000
`
`5,000
`
`4,600
`
`3,800
`
`1,800
`
`348
`
`598
`
`94
`
`251
`
`299
`
`4
`
`12
`
`2
`
`7
`
`17
`
`12
`
`0
`
`1
`
`2
`
`4
`
`65
`
`78
`
`79
`
`71
`
`78
`
`16
`
`19
`
`18
`
`24
`
`16
`
`5
`
`2
`
`1
`
`2
`
`1
`
`2
`
`1
`
`1
`
`1
`
`1
`
`3
`
`0
`
`0
`
`1
`
`1
`
`'' Data on direct-to-consumer advertising are from TNS Media; data on detailing, professional meetings and events, journal advertising, and
`on line promotions to physicians are from Verispan; and data on sales revenues are from I MS Health. Leading therapeutic classes of drugs
`were identified on the basis of publicly available I MS Health rankings of therapeutic classes according to spending for 2004. Values for se(cid:173)
`lective serotonin-reuptake inhibitors (SSRls) and selective norepinephrine-reuptake inhibitors (SNRls) match the classification scheme used
`by Verispan, which was the source of our data on promotions. Values in the far right-hand column refer to product-specific advertising only.
`H MG-CoA denotes 3-hydroxy-3-methylglutaryl coenzyme A, ACE angiotensin-converting enzyme, and COX-2 cyclooxygenase-2.
`
`2003-2004, nearly half of the FDA's promotion(cid:173)
`related regulatory letters were focused on direct(cid:173)
`to-consumer advertisements. From 1997 to 2006,
`nearly 84% of regulatory letters regarding direct(cid:173)
`to-consumer advertising cited advertisements for
`either minimizing risks (e.g., minimizing or omit(cid:173)
`ting information on side effects), exaggerating ef-.
`fectiveness (e.g., portraying the indication too
`broadly or making unsubstantiated claims of su(cid:173)
`periority over other drugs), or both.
`For example, the FDA found that Eli Lilly's tele(cid:173)
`vision broadcast advertisement for Strattera (ato(cid:173)
`moxetine) was false or misleading because it in(cid:173)
`adequately communicated the indication for the
`drug (attention-deficit-hyperactivity disorder) by
`means of competing visuals, graphics, and music
`presented concurrently. Similarly, serious risk dis(cid:173)
`closures were minimized for Strattera, the FDA
`said, by the distracting visuals and graphics (e.g.,
`
`erratic camera movement, quick scene changes,
`and visual changes in point of view). In another
`case, the FDA said Pfizer's print advertisement for
`Zoloft (sertraline) was false or misleading because
`it omitted important information relating to the
`risk of suicidality in patients, a risk stated on the
`product's label at the time the advertisement ran.
`
`DISCUSSION
`
`Spending on direct-to-consumer advertising has
`continued to increase recently in absolute terms
`and as a percentage of pharmaceutical sales in spite
`of pressure on manufacturers to curtail such ad(cid:173)
`vertising.8 Promotion to physicians continues to be
`the dominant marketing strategy, but there are
`some drugs in a majority of the top-selling classes
`that are promoted by such advertising. Driven by
`increases in direct-to-consumer advertising, total
`
`N ENGLJ MED 357;7 WWW.NEJM.0RG AUGUST 16, 2007
`
`677
`
`The New England Journal of Medicine
`Downloaded from nejm.org on May 5, 2015. For personal use only. No other uses without permission.
`Copyright© 2007 Massachusetts Medical Society. All rights reserved.
`
`Sanofi Exhibit 2210.005
`Mylan v. Sanofi
`IPR2018-01675
`
`
`
`The NEW ENGLAND JOURNAL of MEDICINE
`
`Table 3. Top 20 Pharmaceutical Products in Terms of Spending on Direct-to-Consumer Advertising in 2005.*
`
`Drug
`
`Company
`
`Therapeutic Category
`
`Spending"j"
`
`FDA
`Approval
`Date:j:
`
`Year That
`Campaign
`Started~
`
`Nexium (esomeprazole) AstraZeneca
`
`Proton-pump inhibitor
`
`Lunesta (eszopiclone)
`
`Sepracor
`
`Hypnotic-sedative
`
`Vytorin (ezetimibe-
`simvastatin)
`
`Merck/Schering-
`Plough
`
`Crestor (rosuvastatin)
`
`AstraZeneca
`
`Cholesterol absorption
`blocker-HM G-CoA
`reductase inhibitor
`
`HM G-CoA red uctase
`inhibitor
`
`millions of
`dollars
`
`224
`
`214
`
`155
`
`Feb. 2000
`
`Dec. 2004
`
`July 2004
`
`2001
`
`2005
`
`2004
`
`144
`
`Aug. 2003
`
`2004
`
`Advair (fluticasone and
`sal meterol)
`
`GlaxoSmithKline
`
`Co rticoste roid-/3-ad re ner-
`gic-receptor agon ist
`
`137
`
`Aug. 2000
`
`2001
`
`Nasonex (mometasone) Schering-Plough
`
`Corticosteroid
`
`Flonase (fluticasone)
`
`GlaxoSmithKline
`
`Corticosteroid
`
`Lamisil (terbinafine)
`
`Novartis
`
`Allylamine antifungal
`
`Plavix (clopidogrel)
`
`Bristol-Myers Squibb/
`Sanofi
`
`Pl ate let-aggregation
`antagonist
`
`124
`
`111
`
`110
`
`110
`
`Dec. 1997
`
`Oct. 1994
`
`May 1996
`
`Nov. 1997
`
`1998
`
`1995
`
`1997
`
`2001
`
`Lilly ICOS
`
`PDE5 inhibitor
`
`110
`
`Nov. 2003
`
`2004
`
`Cialis (tadalafil)
`
`Wellbutrin XL
`(bupropion)
`
`GlaxoSmithKline
`
`Singulair (montelukast) Merck
`
`Lipitor (atorvastatin)
`
`Pfizer
`
`Dopamine reuptake
`inhibitor-SN RI
`
`Leukotriene D4-receptor
`antagonist
`
`HM G-CoA red uctase
`inhibitor
`
`Ambien (zolpidem)
`
`Sanofi-Aventis
`
`Hypnotic-sedative
`
`Humira (adalimumab)
`
`Abbott
`
`Monoclonal antibody
`
`lmitrex (sumatriptan)
`
`GlaxoSmithKline
`
`Viagra (sildenafil)
`
`Pfizer
`
`Neulasta (pegfilgrastim) Amgen
`
`Vascular 5-HTl-receptor
`agonist
`
`PDE5 inhibitor
`
`G-CSF analogue
`
`Valtrex (valacyclovir)
`
`GlaxoSmithKline
`
`DNA polymerase inhibitor
`
`Prevacid (lansoprazole)
`
`TAP
`
`Proton-pump inhibitor
`
`108
`
`Aug. 2003
`
`2004
`
`105
`
`Feb. 1998
`
`1998
`
`93
`
`88
`
`88
`
`82
`
`80
`
`74
`
`72
`
`71
`
`Dec. 1996
`
`1998
`
`Sept. 2005
`
`Dec. 2002
`
`Aug. 1997
`
`March 1998
`
`Jan. 2002
`
`June 1995
`
`May 1995
`
`2005
`
`2003
`
`1998
`
`1998
`
`2002
`
`1996
`
`2000
`
`'' HMG-CoA denotes 3-hydroxy-3-methylglutaryl coenzyme A, SNRI selective norepinephrine-reuptake inhibitor, 5-HTl
`5-hydroxytryptamine 1, PDE5 phosphodiesterase type 5, and G-CSF granulocyte colony-stimulating factor.
`"j" Data are from Arnold. 16
`:j: Approval dates are from the Electronic Orange Book. 15
`§ Starting dates for direct-to-consumer campaigns were obtained through Internet searches. A detailed source list is
`available from the authors.
`
`promotion as a percentage of sales has increased
`substantially during the past 5 years, leading some
`observers to worry that consumers must bear these
`increased costs in the form of higher prices. Eco(cid:173)
`nomic theory and evidence suggest that changes
`in marketing costs are unlikely to have a direct
`effect on pharmaceutical prices, which largely
`reflect perceptions of product value held by con-
`
`sumers, physicians, and payers.17 Of course, it is
`possible that advertising reduces the price respon(cid:173)
`siveness of demand and thus leads manufactur(cid:173)
`ers to increase prices, but the empirical evidence
`on this point is mixed. 18 ,19
`Advertising campaigns generally begin within
`a year after the introduction of a pharmaceutical
`product, which raises questions about the extent
`
`678
`
`N ENGLJ MED 357;7 WWW.NEJM.0RG AUGUST 16, 2007
`
`The New England Journal of Medicine
`Downloaded from nejm.org on May 5, 2015. For personal use only. No other uses without permission.
`Copyright© 2007 Massachusetts Medical Society. All rights reserved.
`
`Sanofi Exhibit 2210.006
`Mylan v. Sanofi
`IPR2018-01675
`
`
`
`A DECADE OF DIRECT-TO-CONSUMER PRESCRIPTION-DRUG ADVERTISING
`
`D Letters related to promotion
`18
`
`.t. Letters related to direct-to-consumer advertising
`60
`
`160
`j -~ 140
`3 o 12
`~~
`·;; d:
`.. 0
`i i 0 "la
`
`d-;;
`Z co:
`
`4
`
`•
`
`...
`
`...
`
`• •
`
`...
`
`...
`
`...
`
`...
`
`30
`
`0. 0
`
`0 ..
`., C
`50 11:!!!
`"la 1i
`40 &i -fi
`.. c(
`b b
`'Sl E
`......
`:I
`...I
`0 C
`:.~
`20
`.. s 5 '
`10 ul! b .!:
`20
`o...._.......,...____.__,_.__.__,,...___._.,.....____._..,......._.. ......... _
`_
`_
`.....,_.__.......,...___.......,.....___ 0
`1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
`
`Figure 1. Trend s in FDA Enforcement of Regulations Regarding Direct-to-Consumer Advertising, 1997- 2006.
`Data are from regulatory letters posted on the Web site of the Division of Drug Marketing Advertising and Commu(cid:173)
`nication of the FDA (www.fda.gov/ cder/ ddmac/ laws regs .htm) .
`
`to which advertising increases the use of drugs
`with unknown safety profiles. At least one phar(cid:173)
`maceutical manufacturer (Bristol-Myers Squibb)
`recently announced a voluntary moratorium on
`direct-to-consumer advertising for drugs in the
`first year after FDA approval. And PhRMA, the
`industry trade group, has recommended that man(cid:173)
`ufacturers delay sue campaigns for new drugs
`until after health professionals have been uffi(cid:173)
`ciently educated, although no details have been
`provided on how long a period was deemed nec(cid:173)
`essary. 20 Finally, in a recent study of drug safety,
`the Institute of Medicine recommended that the
`FDA restrict advertising for newer prescription
`drugs .8 Our data show that a mandatory waiting
`period on advertising for new drugs would rep(cid:173)
`resent a dramatic departure from current indus(cid:173)
`try practices.
`The number of regulatory actions taken by the
`FDA against companies marketing prescription
`drugs w consumers as f.lllen dramatically in re(cid:173)
`cent years. This decline may reflect either better
`industry compliance with advertising regulations
`or a worsening of FDA oversight. 21 Although a
`systematic assessment of the compliance of phar(cid:173)
`maceutical advertisers with advertising regulations
`is beyond the scope of this article, some insights
`into this issue can be gained from examination of
`policy changes and staffing levels within the FDA
`over the period of our study. Three observations
`from such an examination suggest that the FDA's
`capacity to enforce advertising regulations has
`weakened in recent years.
`
`First, in 2002 the Secretary of Health and Hu(cid:173)
`man Services began requiring that all draft FDA
`regulawry letters, including letters related to ad(cid:173)
`vertising violations, be reviewed and approved by
`the FDA's Office of Chief Counsel before they are
`issued. A GAO report found that this legal review
`has led to a reduction in the number of letters is(cid:173)
`sued, as well as to delays such that FDA warning
`letters are frequently sent out long after the false
`or misleading advertising campaign has run its
`coorse.n Notably, the number of regulatory letters
`sent by the FDA in 2002 was less than half that
`in 2001 (28 vs. 68) (Fig. 1).
`A second indication of weakening FDA over(cid:173)
`sight of direct-to-consumer advertising in recent
`years is that the number of staff members who
`are dedicated to reviewing advertisements has re(cid:173)
`mained relatively stable, whereas the use of such
`advenising has grown substantially. In 2002, three
`FDA staff members were dedicated to reviewing
`direct-to-consumer advertisements.22 In 2004, four
`staffers were reviewing such advertisements, even
`thoug spending on this form of advertising (and
`probably the volume of ads to review) had in(cid:173)
`creased by 45%, from $2.9 billion to $4.2 billion
`(Table 1). 23
`Finally, consistent with the hypothesis that
`staffing has not kept pace with the number of
`prescription-drug advertisements, the proportion
`of broadcast advertisements that underwent FDA
`review before airing declined from 640/o in 1999
`to only 320/o in 2004. 23 Thus, even if manufactur(cid:173)
`ers were to increase submission of advertisements
`
`N ENGLJ MED }57';1 WWW . N EJM . ORG AUGUST 16, 2007
`
`679
`
`The New England Journal of Medicine
`Downloaded from nejm.org on May 5, 2015. For personal use only. No other us~ without permission.
`Copyright 0 2007 Massachusetts Medical Society. All rights reserved.
`
`Sanofi Exhibit 2210.007
`Mylan v. Sanofi
`IPR2018-01675
`
`
`
`The NEW ENGLAND JOURNAL of MEDICINE
`
`to the FDA, the agency has said that "current FDA
`resourcing for this work would probably result in
`delayed reviews ... and discourage [manufactur(cid:173)
`ers] from submitting the materials for prior FDA
`review."23
`Our study has some key limitations. We ob(cid:173)
`tained data on industry sales from PhRMA, which
`includes in its annual reports sales data only for
`its members. Ideally, we would include sales of
`all branded drugs sold by prescription, including
`pharmaceutical and biologic agents, and exclude
`sales of generic drugs (because generic drugs
`typically are not promoted). PhRMA sales data
`may include some generic sales (if a member re(cid:173)
`ports both branded and generic sales) and typi(cid:173)
`cally exclude sales of biologic agents, which are
`manufactured by companies that belong to an(cid:173)
`other trade group (Biotechnology Industry Orga(cid:173)
`nization). As a result, the sales figures may under(cid:173)
`estimate total dollar sales for the industry. We
`provide data on spending on free samples valued
`at their approximate retail price, which is how they
`typically are valued in industry promotional audits.
`Thus, the value of free samples we present prob(cid:173)
`ably overstates the opportunity cost to manufac(cid:173)
`turers, which would lie somewhere between the
`marginal cost of production and the retail value.
`
`Since 2000, direct-to-consumer advertising of
`prescription drugs has continued to grow both in
`absolute dollars and relative to other forms of pro(cid:173)
`motion. Although the evidence base is growing,
`there are few data to support an assessment of the
`balance of the costs and benefits of such adver(cid:173)
`tising.24 The debate over whether and how direct(cid:173)
`to-consumer advertising should be more tightly
`regulated takes place against a backdrop of grow(cid:173)
`ing concern about the growth of health care
`spending, particularly in the Medicare program.
`Gaining a better understanding of the effects of
`direct-to