A Statistical Analysis of the Magnitude
`and Composition of Drug Promotion
`in the United States in I998
`
`Jun Ma, MD, PhD, RD,l Randall S. Stafford, MD, PhD,l
`Iain M. Cockburn, PhD,’ and Stan N. Finkelstein, MD3
`‘Stanford Centerfor Research in Disease Prevention, Stanford University, Palo Alto, CaliJornia,
`2School of Manag ement, Boston University, Boston, and 3Program on the Pharmaceutical
`Industry, Massachusetts Institute of Technology, Cambridge, Massachusetts
`
`ABSTRACT
`
`Background: Although pharmaceutical
`and other factors
`industry marketing
`may influence physician decisions regarding medication prescribing
`in the United
`States, little information
`is available about the composition of promotional
`efforts
`by promotional mode and medication class.
`Objectives: The aims of this study were to determine
`the magnitude of ex-
`penditures
`for common modes of promotion
`and to delineate patterns of pro-
`motional strategies for particular classes of medications.
`Methods: Nationally representative data on expenditures
`(in US $) for the 250
`most promoted medications
`in the United States in 1998 were available from an
`independent
`pharmaceutical market research company
`for the 5 most commonly
`used modes of promotion. Key patterns of drug promotion were identified by de-
`scriptive statistics, a cluster analysis of expenditures
`by class, and an analysis of
`expenditure
`concentration.
`Results: In 1998, the pharmaceutical
`industry spent $12,724 million promot-
`ing its products
`in the United States, of which 85.9% was accounted
`for by the
`top 250 drugs and 51.6% by the top 50 drugs. Direct-to-consumer
`(DTC) adver-
`tising was more concentrated
`on a small subset of medications
`than was promo-
`tion to professionals. Overall, 1998 expenditures were dominated by free drug
`samples provided
`to physicians (equivalent
`retail cost of $6602 million) and office
`promotion
`($3537 million), followed by DTC advertising
`($1337 million), hospi-
`
`Accepted for publication February 2 7, 2003.
`Printed
`in the USA. Reproduction
`in whole or part is not permitted.
`
`Copyright@ 2003 Excerpta Medm,Inc.
`
`0149-2918/03/$19.00
`
`1503
`
`MYLAN - EXHIBIT 1075
`Mylan Laboratories Limited v. Aventis Pharma S.A.
`IPR2016-00712
`
`

`

`CLINICAL THERAPEUTICS@
`
`($540 million).
`in medical journals
`($705 million), and advertising
`tal promotion
`to office physi-
`Four distinct patterns of expenditures were observed: promotion
`cians with little consumer promotion
`(14 drug classes); dual focus on office physi-
`cians and consumer advertising
`(4 drug classes); predominant DTC advertising
`(1 class: smoking-cessation products); and promotion
`to office- and hospital-based
`professionals without consumer advertising
`(1 class: narcotic analgesics).
`Conclusions: The present
`findings reinforce
`the perception
`that the pharrna-
`ceutical industry
`invests heavily in promoting
`its products and demonstrates
`that
`promotional
`expenditures
`are concentrated
`on a small number of medications.
`Although promotion
`to professionals
`remains dominant, DTC advertising has be-
`come key for a subset of common medications.
`(Clin Ther 2003;25:1503-1517)
`Copyright 0 2003 Excerpta Medica, Inc.
`Key words: pharmaceutical
`promotion,
`advertising.
`
`direct-to-consumer
`
`office promotion,
`
`INTRODUCTION
`im-
`Physician decisions regarding medication prescribing can have a considerable
`in-
`pact on health care costs and patient outcomes.1-3 Although pharmaceutical
`dustry marketing and other factors may influence
`these decisions,
`little informa-
`tion is available on the composition of promotional
`efforts by promotional mode
`and medication class.
`the
`drug advertising has been
`The issue of costs and benefits of prescription
`and proponents
`of the drug in-
`subject of ongoing debate between opponents
`dustry and has long been a focal point in policy discussions between
`regulatory
`agencies and the industry4-” Pharmaceutical marketing has been criticized as
`wasteful and excessive and for contributing
`to the overuse, misuse, and mispre-
`scribing of drugs. 5,6 However, marketing may also serve as a key communication
`channel
`for continuing physician education
`regarding pharmaceutical
`products
`and for exposing consumers
`to drug information.
`In addition, promotion may al-
`low more rapid adoption of new drugs that represent
`incremental
`advances over
`past practices.4-6
`into informative adver-
`can be categorized
`promotion
`Overall, pharmaceutical
`advertising.5
`Informative
`advertising
`is a
`repetitive
`tising and noninformative,
`main strategy in market expansion
`targeted at new pharmaceutical
`products, new
`indications
`for established products, and new prescribers and patients. By foster-
`ing habitual patterns of drug selection,
`repetitive advertising
`is aimed largely at
`maintaining market share.
`Among the possible modes of drug promotion by pharmaceutical companies, an-
`other distinction can be made between promotion
`to health care professionals and
`
`1504
`
`

`

`J. Ma et al.
`
`are professional-
`to consumers. Traditional modes of drug promotion
`promotion
`centered and consist of the following: (1) direct contact with office-based physicians
`by pharmaceutical-company
`sales representatives
`(detailing); (2) direct contact with
`physicians and pharmacists
`in hospitals; (3) advertising in medical journals; and (4)
`free drug samples provided
`to physicians. Although promotion
`to health care pro-
`fessionals continues
`to dominate
`the field of pharmaceutical marketing, direct-to-
`consumer
`(DTC) advertising has increased since the early 1990s with more sub-
`stantial recent increases in television advertising (eg, a -/-fold increase to $1.6 billion
`in 2000, compared with 1996).7,8 Although regulatory changes by the US Food and
`Drug Administration
`(FDA) have made DTC advertising more feasible, other fac-
`tors also may have prompted
`recent substantial increases. These factors include pa-
`tient desire for increasing involvement
`in health care decisions,
`the drug industry’s
`intent
`to educate consumers
`in disease management,
`and the diminishing
`social
`stigma associated with some medical conditions.7.” Expenditures on other modes
`of drug promotion,
`such as Internet advertising and sponsorship of scientific meet-
`ings, are relatively small (-10% of total spending).7
`re-
`promotion
`Regardless of its advantages and disadvantages, pharmaceutical
`mains a significant
`feature of the US health care system. Growing academic evi-
`dence documents
`that drug promotion
`influences physicians’ prescribing choices,
`consumers’ behavior,
`the physician/patient
`relationship,
`and health care quality
`and costs.8-11 Unfortunately, many health care professionals may fail to recognize
`the full influence
`that drug promotion
`exerts on them and their patients. The
`scarcity of published
`research on this topic may contribute
`to these attitudes.
`Physicians also may deny being influenced by commercial sources because doing
`so is perceived by many as unflattering.12
`impact,
`To facilitate physicians’ recognition of drug promotion and its potential
`we created a cross-sectional
`portrait of promotional
`expenditures
`for the 250
`most promoted drugs in 1998. Of note, our study included only promotional
`ac-
`tivities inside the United States. Our objectives were to determine
`the magnitude
`of expenditures
`for common modes of promotion
`and to delineate patterns of
`promotional
`strategies for particular classes of medications. We hypothesized
`that
`promotional
`expenditures would be concentrated
`in a subset of drugs and that
`their distribution
`by mode would vary among drug classes. Specifically, office
`promotion and drug samples would account
`for most of the promotional
`expen-
`ditures for drugs in a majority of classes, whereas DTC advertising would focus
`on only a few products.
`
`MATERIALS AND METHODS
`Data Collection
`The 1998 annual expenditures
`motion expenditure were obtained
`
`(in US S) on the top 250 drugs by total pro-
`from IMS Health (Plymouth Meeting, Penn-
`
`1505
`
`

`

`CLINICAL THERAPEUTICS@
`
`pharmaceutical market research company,
`Sylvania). IMS Health, an independent
`publishes an array of data reports on what, when, how, and how much promo-
`tional activity occurs for pharmaceutical
`products. This study used the Integrated
`Share of Voice Report for 1998, which drew data from the Integrated Promotion
`Services reports and the Consumer Media Reports
`to provide a nearly complete
`view of pharmaceutical
`promotion
`to medical professionals and consumers.
`The Integrated Promotion Services report compiled pharmaceutical
`promo-
`tional expenditures
`on detailing
`to office- and hospital-based
`physicians,
`the re-
`tail value* of free samples provided
`to physicians, and expenditures
`for journal
`advertising. To estimate detailing expenditures,
`IMS Health recruited a national
`panel of 3862 office-based physicians (office promotion panel) and 2067 hospital-
`based physicians and 5 14 hospital pharmacy directors (hospital promotion panel)
`who documented
`the amount of time spent with pharmaceutical-company
`sales
`representatives
`each day for 1 month. Participating
`physicians were selected
`through stratified
`random sample design and were representative
`of office- and
`hospital-based
`physicians
`in the continental United States in terms of detailing
`activity
`In addition, pharmaceutical
`companies were surveyed
`to obtain annual
`information
`on expenses directly
`related
`to deploying
`representatives
`for office
`and hospital visits, including salaries, bonuses, automobiles,
`insurance,
`training,
`and fringe benefits. Expenditures
`for detailing were estimated by the product of
`the average minutes reported
`for promoting a product and the estimated average
`expenses on representatives. To estimate
`the retail value of free samples provided
`by pharmaceutical
`companies,
`IMS Health used a panel of 1265 members of the
`front-office staff in practices
`that were subsampled
`from the previously described
`office-promotion
`panel. Therefore,
`the data on free samples were representative
`of samples given to office-based physicians only Panel members
`reported
`the
`quantity of all drug samples either left at the office by pharmaceutical
`represen-
`tatives or mailed to the office during a l-month period. The retail value of these
`samples was then computed using the retail dollar and quantity sales information
`gathered by IMS Health. This estimate of the value of free samples probably ex-
`ceeded
`the true economic cost borne by the pharmaceutical
`companies
`that sup-
`plied these samples.
`in every
`advertisements
`for pharmaceutical
`Expenditures were also estimated
`journal received by practitioners
`in all types of medical practices. Therefore,
`jour-
`nal advertising costs represented universal census figures rather
`than sample es-
`timates. The cost of a journal advertisement was estimated by first applying a ba-
`sic advertisement price and then adjusting upward or downward according
`to its
`design and printing characteristics
`(eg, use of inserts). The basic advertisement
`
`*The retail value of free samples
`sold at retail pharmacies.
`
`represents
`
`the estimated
`
`revenue
`
`that would be generated
`
`if the drugs were
`
`1506
`
`

`

`j. Ma et al.
`
`and adjustments
`advertisement,
`price was based on a l-time black-and-white
`were made according
`to rates and charges found
`in Standard Rate and Data, an
`advertising
`industry publication.
`on DTC adver-
`In Consumer Media Reports, IMS Health tracked expenditures
`tising on television and radio, as well as in magazines and newspapers. Both ad-
`vertisements
`that promoted
`specific drug products and those that did not men-
`tion a drug by name were included. Advertising
`expenditures were estimated
`according
`to the length of every commercial message and the average of current
`cost rates for a particular medium.
`
`Data Processing and Analysis
`on office promo-
`This study was based on national estimates of expenditures
`tion, hospital promotion,
`free samples, DTC advertising,
`and census figures of
`medical journal advertising. As noted, our analyses focused on the 250 most pro-
`moted drugs in 1998. All statistical analyses were conducted using SAS System
`for Windows, version 8 (SAS Institute
`Inc., Cary, North Carolina). The Gini in-
`dex13J4 was calculated
`to measure
`the extent of concentration
`and inequality
`in
`promotion
`expenditure
`among
`the 250 drugs. The greater
`the Gini index,
`the
`more concentrated
`promotional
`efforts were in a limited set of medications.
`The 250 most highly promoted medications were categorized
`into 20 drug
`classes based on pharmacologic
`indications
`and clinical usage. A residual other
`medications category also was defined for individual medications
`that shared lim-
`ited commonality
`in action or usage with other medications and did not warrant
`a separate drug class, given the relatively small promotion
`expenditures. Medica-
`tions in this category comprised a broad range, including anesthetics, antiemet-
`its, antiglaucoma agents, synthetic
`thyroid hormones,
`skin treatments, nutritional
`supplements,
`and other drug classes. The other medications category was not in-
`cluded
`in our analyses by drug class.
`Cluster analysis examined
`the relative reliance on the 5 promotional modes
`across the 20 drug classes. Results illustrate grouping profiles of the drug classes
`that reflect distinctive usage patterns of the 5 promotional modes. The agglom-
`erative hierarchical method
`is superior
`to the nonhierarchical method when the
`number of clusters cannot be specified based on a practical, objective, or theo-
`retical basis. l5 The agglomerative hierarchical
`clustering procedure grouped
`the
`drug classes based on similarity
`in percent share by mode of promotion
`expen-
`diture within a drug class. This method started with each drug class as a single
`cluster; drug classes that were most alike were combined at each hierarchy
`In se-
`lecting final clusters, several statistics
`that are commonly
`used in determining
`the number
`of clusters were evaluated,
`including
`cubic clustering
`criterion,
`pseudo F, and pseudo
`t2. This process
`led us to define 4 clusters
`that captured
`the most important distinctions between drug classes and also were interpretable
`
`1507
`
`

`

`CLINICAL THERAPEUTICS@
`
`based on subject matter knowledge. The selection of 4 clusters was validated by
`a subsequent nonhierarchical
`clustering procedure.
`
`RESULTS
`In 1998, $12,724 million was spent on promoting prescription
`drugs via office
`promotion,
`hospital promotion,
`journal promotion,
`free samples, and DTC ad-
`vertising (Tables I and II). The top 250 drugs accounted
`for 85.9% ($10,926 mil-
`lion) of the total promotional
`expenditures,
`and the top 50 accounted
`for 51.6%
`of the total. Likewise, the top 250 drugs accounted
`for 83.2% of total expenditures
`for office promotion,
`82.8% for hospital promotion,
`76.9% for journal promo-
`tion, 86.1% for free drug samples, and 97.3% for DTC advertising. Overall con-
`centration of promotional
`expenditures
`among
`the top 250 drugs was reflected
`by a Gini index of 0.57. Expenditures were less concentrated
`for office promo-
`tion (Gini index, 0.50) and hospital promotion
`(Gini index, 0.58), compared with
`journal promotion
`(Gini index, 0.64) and free drug samples (Gini index, 0.66).
`DTC advertising was markedly more concentrated, with 20% of the drugs con-
`suming 97.7% of the promotional
`expenditures
`(Gini index, 0.88). Although
`re-
`tail drug sales by individual medication were also substantially concentrated,
`they
`were less concentrated
`than drug promotion
`(Gini index, 0.49).
`Drug classes with the largest 1998 promotion
`expenditures
`include antibiotics
`($1634 million), antihypertensives
`($1323 million), antiallergics ($12 17 million),
`and antidepressants
`($10 13 million), whose combined
`expenditures
`accounted
`for 40.8% of the total market
`(Table I). By class among
`the top 250 drugs,
`the
`greatest mean expenditures per drug were for antacids ($118.0 million per drug),
`antihyperlipidemics
`($111.7 million), antidepressants
`($84.4 million), and an-
`tiallergics ($71.6 million). A mean of $43.7 million per drug was spent on the
`top 250 drugs in 1998. As another
`indication of unequal distribution of promo-
`tional expenditures,
`the top-promoted
`drug alone accounted
`for ~50% of the
`market share in 5 of the 20 therapeutic drug classes (ie, impotence medications,
`smoking-cessation
`products, antivirals, antianxiety medications,
`and osteoporosis
`medications). The top 2 drugs dominated expenditures
`in an additional 8 of the
`20 classes,
`including
`antihyperlipidemics,
`antacids, oral hypoglycemics,
`post-
`menopausal medications, headache medications, urinary
`tract medications,
`anti-
`fungals, and narcotic analgesics.
`in 1998 were dominated by free drug samples
`Drug promotion
`expenditures
`provided
`to physicians
`(retail value, $6602 million) and office promotion
`($3537
`million), representing 5 1.9% and 27.8% of all promotional
`expenditures,
`respec-
`[ 10.5% of all expenditures])
`tively (Table II). DTC advertising
`($1337 million
`ranked
`third, surpassing both hospital
`($705 million
`[5.5%]) and journal pro-
`motion ($540 million [4.3%]). The ranks of drug classes regarding
`the magnitude
`of expenditures were consistent across the modes of office promotion,
`hospital
`
`1508
`
`

`

`PI 7
`a
`II e,
`L-
`
`_
`12.0
`87.4
`75.6
`54.0
`73.5
`49.4
`27. I
`93.6
`34.8
`48.9
`40.3
`40.5
`44.8
`25. I
`44.6
`30.7
`30. I
`24.3
`21.6
`IO.5
`I I.0
`
`iWithin
`‘May not total evenly due to rounding.
`NSAlDs = nonsteroidal anti-inflammatory drugs.
`
`the 250 most highly promoted drugs.
`
`Class byTop Drug, %
`
`US $, millions
`
`Expenditures,* %
`
`in
`
`Expenditures
`Share of
`
`in Class,
`
`per Drug
`Mean Expenditure
`
`Promotional
`Share of Total
`
`US $, millions
`Expenditure,
`
`Total Promotional
`
`in Class
`Medications
`
`No. of
`
`motion were $12,724 million.
`class, based on data from IMS Health (Plymouth Meeting, Pennsylvania). Total market expenditures for pharmaceutical pro-
`in the United States in 1998, according to
`
`Table I. Promotional expenditures among the 250 most highly promoted medications
`
`43.7
`IS.0
`33.0
`23.3
`20.6
`36.7
`19.2
`20.4
`63.7
`42.0
`49.0
`53.3
`29.5
`43.4
`37.6
`I 18.0
`I II.7
`51.7
`84.4
`71.6
`38.9
`46.7
`
`85.9
`5.2
`0.5
`0.7
`0.8
`0.9
`0.9
`I .4
`I .5
`I .7
`I.9
`2.5
`2.5
`2.7
`3.5
`5.6
`6.1
`6.5
`8.0
`9.6
`IO.4
`12.8
`
`10,926
`662
`66
`93
`103
`I IO
`II5
`184
`I91
`210
`245
`320
`325
`347
`451
`708
`782
`827
`1013
`1217
`I323
`I634
`
`250
`44
`2
`4
`5
`3
`6
`9
`3
`5
`5
`6
`I I
`8
`I2
`6
`7
`I6
`I2
`I7
`34
`35
`
`Top 250
`Other medicationst
`Osteoporosis medications
`Antianxiety medications
`Antivirals
`Smoking-cessation products
`Narcotic analgesics
`Contraceptives
`Impotence medications
`Antifungals
`Urinary
`Headache medications
`Postmenopausal medications
`Oral hypoglycemics
`NSAlDs
`Antacids
`Antihyperlipidemics
`Asthma medications
`Antidepressants
`Antiallergics
`Antihypertensives
`Antibiotics
`
`tract medications
`
`Class
`
`

`

`CLINICAL THERAPEUTICS@
`
`in the United States in
`Table II. Distribution of total medication-promotion expenditures
`1998, according to mode of promotion, based on data from
`IMS Health
`(Plymouth Meeting, Pennsylvania).
`
`Class
`
`Antibiotics
`Antihypertensives
`Antiallergics
`Antidepressants
`Asthma medications
`Antihyperlipidemics
`Antaclds
`NSAlDs
`Oral hypoglycemics
`Postmenopausal medications
`Headache medications
`Urinary
`tract medications
`Antifungals
`Impotence medications
`Contraceptives
`Narcotic analgesics
`Smoking-cessation products
`Antiwals
`Antianxiety medications
`Osteoporosis medications
`Other medicationst
`Top 250
`Total market
`
`Share of Expenditures,* %
`
`Office
`Promotion
`
`Hospital
`Promotion
`
`Journal
`Promotion
`
`Free Drug
`Samples
`
`DTC
`Advertising
`
`33.3
`32.3
`20.7
`23.2
`23.9
`24.7
`19.1
`35. I
`32.9
`29.9
`25.6
`22.9
`26.5
`16.6
`21.7
`41.7
`16.1
`28.7
`23.2
`42.6
`26.8
`26.9
`27.8
`
`8.3
`5.2
`3.0
`5.5
`3.4
`4.6
`2.9
`4.2
`5.7
`4.4
`4.0
`3.7
`4.8
`7.9
`4. I
`I I.3
`3.6
`4.8
`6.9
`4.6
`9.7
`5.3
`5.5
`
`3.0
`6.7
`I .9
`3.1
`3.1
`3.4
`I .7
`3.7
`6.2
`3.6
`2.9
`6. I
`2.9
`7.4
`2. I
`12.7
`I .5
`I .7
`I .9
`0.2
`6.5
`3.8
`4.3
`
`54. I
`55.8
`40.4
`64.2
`67.3
`53.3
`69.2
`57.0
`46.8
`32. I
`46.4
`18.7
`48.4
`52.3
`54.9
`34.3
`6.0
`22.3
`44.5
`52.5
`43.6
`52.0
`5 I .9
`
`I .4
`0.0
`34.0
`4. I
`2.3
`14.0
`7.0
`0.0
`8.4
`30.0
`21.2
`48.6
`17.4
`15.8
`17.3
`0.0
`72.8
`42.4
`23.4
`0. I
`13.4
`I I.9
`IO.5
`
`DTC = direct to consumer; NSAlDs = nonsteroldal anti-Inflammatory drugs
`“May not total evenly due to rounding.
`Within
`the 250 most highly promoted drugs.
`
`and free samples. The leading drug classes for
`journal promotion,
`promotion,
`expenditures were antibiotics, antihypertensives,
`antiallergics, anti-
`promotional
`asthma medications,
`antihyperlipidemics,
`antacids, nonsteroidal
`depressants,
`anti-inflammatory
`drugs (NSAIDs), and oral hypoglycemics
`(Table I). DTC ad-
`vertising expenditures were concentrated
`on smoking-cessation
`products, urinary
`tract medications,
`antivirals, antiallergics, postmenopausal medications,
`and an-
`tianxiety medications
`(Table II). With the exceptions of urinary
`tract medications
`and smoking-cessation
`products, >60% of annual expenditures
`per drug class
`
`1510
`
`

`

`J. Ma et al.
`
`and free drug samples. Combined hos-
`for by office promotion
`were accounted
`ranged from 4.6% to 11.9% of an-
`pital and journal promotional
`expenditures
`nual expenditures
`for all drug classes, with the exceptions of narcotic analgesics
`(24.0%) and impotence medications
`(15.3%). DTC advertising accounted
`for a
`sizable proportion
`of the annual promotion
`for several drug classes, including
`smoking-cessation
`products
`(72.8%), urinary
`tract medications
`(48.6%), and an-
`tivirals (42.4%). The pattern of promotional expenditures
`in the other medications
`category paralleled
`the overall distribution of total expenditures
`among the 5 pro-
`motional modes.
`The patterns of association between drug classes and modes of promotion used
`may be further appreciated with cluster analysis. Table III enumerates
`the 4 final
`clusters
`identified
`and their respective composite drug classes. The office-&us
`cluster included 14 of the 20 drug classes. The office/DTC-focus cluster included
`antiallergics, antivirals, urinary
`tract medications,
`and postmenopausal medica-
`tions. Defining
`their own clusters were smoking-cessation medications
`(DTC Jo-
`cus) and narcotic analgesics (general professional focus). The figure shows the dis-
`tribution patterns of annual expenditures
`among
`the 5 promotional modes for
`each of these 4 clusters. For the office-focus cluster, most promotional
`expendi-
`tures were for office promotion
`(36%) and free drug samples provided
`to office-
`based physicians
`(44%). The office/DTC-focus
`cluster had similar expenditures
`for professional-targeted
`promotional
`activity
`to those observed
`in the office-
`focus cluster, but represented
`a larger proportional
`role of DTC advertising
`(29%
`in the office/DTC-focus
`cluster vs 7% in the office-focus cluster). The DTC-focus
`cluster was characterized by an 83% expenditure
`share for DTC advertising, with
`
`Table III. Cluster-analysis results denoting patterns between drug classes and promo-
`tional modes
`in the United States in 1998, based on data from
`IMS Health
`(Plymouth Meeting, Pennsylvania).
`
`Cluster
`
`ORce focus
`
`Drug Classes
`
`Antibiotics, antidepressants, antifungal5 oral hypoglycemics, antihyper-
`Ilpidemics, asthma medications, antihypertensives, antianxiety medi-
`cations, NSAIDs, contraceptives, osteoporosis medications, headache
`medications, antacids, impotence medications
`
`Offce/DTC
`
`focus
`
`Antiallergics, antiviral5 urinary tract medications, postmenopausal
`medications
`
`DTC focus
`
`Smoking-cessation products
`
`General professional
`
`focus
`
`Narcotic analgesics
`
`NSAlDs = nonsteroidal anti-inflammatory drugs; DTC = direct to consumer
`
`1511
`
`

`

`??Office promotion
`0 Hospital promotion
`??Journal promotion
`Free samples
`Ld DTC promotion
`
`CLINICAL THERAPEUTICS@
`
`IOO-
`
`90-
`
`80-
`
`70-
`
`8
`
`6()_
`
`50-
`
`40 -
`
`30-
`
`20 -
`
`IO-
`
`o-l--
`
`Office Focus
`
`OficelDTC
`Focus
`
`’
`
`DTC Focus
`
`’
`
`General
`Professional Focus
`
`’
`
`Distribution patterns of annual expenditures for 5 promotional modes for each
`of 4 identified clusters among the 250 most highly promoted drugs in the United
`States in 1998, based on data from IMS Health (Plymouth Meeting, Pennsylvania).
`DTC = direct to consumer.
`
`clus-
`little use of other promotional modes. Finally, the general/professional-focus
`for
`ter distinguished
`itself from the others with higher proportional
`expenditures
`office, hospital, and journal promotion;
`lower expenditures
`for free samples; and
`no expenditures
`for DTC advertising.
`
`DISCUSSION
`in the United States are substantial
`promotion
`Expenditures
`for pharmaceutical
`and are channeled
`through a variety of promotional
`strategies aimed at both pro-
`fessionals and consumers. As we hypothesized,
`the particular mix of promotional
`strategies varied among therapeutic
`classes of medications
`in 1998.
`Expenditures
`for acquiring prescription
`drugs represent
`the fastest growing
`component of the US health care cost-9.4%
`($100 billion)
`in 1999 with a pro-
`jected increase of 12.6% per year in the current decade.16 As key contributors
`to
`increasing drug prices and volumes, new brand-name drugs, physician-prescribing
`
`1512
`
`

`

`J. Ma et al.
`
`ef-
`by drug promotional
`are all influenced
`and patient demands
`behaviors,
`forts.1,4,5,g,11,16,17 Traditionally, drug promotion
`has targeted health care profes-
`sionals via strategies such as detailing,
`journal advertising, and provision of free
`drug samples. *,17 However
`consumer-oriented
`advertising has become
`increas-
`ingly common since the early 1990s. This rising trend accelerated after the FDA
`published guidelines
`for broadcast advertising
`in 1997.’ As shown in the present
`study, total expenditures
`for both professional-oriented
`and consumer-directed
`promotions
`in 1998 were $12,724 million;
`this number
`increased by 24% to
`$15.7 billion by 2000.7 As previously noted,
`these estimates
`included
`the retail
`value of drug samples, which was probably higher
`than the actual costs of pro-
`ducing and distributing
`the samples. Nonetheless,
`the magnitude of drug pro-
`motion by the pharmaceutical
`industry was considerable.
`It has also been noted
`that the intensity of drug promotion
`in relation
`to drug sales remained steady at
`-14% of total sales between 1996 and 2000.’ Although expenditures
`for the mar-
`keting of pharmaceutical
`products are large in absolute
`terms,
`they also exceed
`the relative investment
`in promotion observed
`in most other US industries,
`rank-
`ing 34th of the 200 industries with the largest advertising expenditures.16
`The US pharmaceutical market consists of >lO,OOO drug products, with a sub-
`stantial number of new entrants every year. However,
`the promotional
`efforts of
`the pharmaceutical
`industry are concentrated
`in a relatively small set of medica-
`tions. The 250 most promoted drugs accounted
`for 85.9% of the overall promo-
`tional expenditures
`in 1998, whereas
`the top 50 drugs accounted
`for 51.6% of
`the total market. A total of $5187 million was spent to promote antibiotics, an-
`tihypertensives,
`antiallergics, and antidepressants
`The large marketing expenses
`associated with
`these 4 drug classes are partially explained by the number of
`drugs in each class, especially for antibiotics
`(35 drugs) and antihypertensives
`(34
`drugs). When ranked according
`to promotional
`intensity-measured
`as mean ex-
`penditure per drug within
`the class-antacids,
`antihyperlipidemics,
`and antide-
`pressants
`represented
`the greatest per-drug expenditures. The most highly pro-
`moted drug classes correspond with the classes that encompass
`the best-selling
`drugs.l
`In addition,
`the highly concentrated
`distribution
`of promotional
`expen-
`ditures exceeded
`the concentrated
`distribution
`of sales (eg, the 50 top-selling
`drugs accounted
`for 44.4% of the total retail sales market in 2OOl).l In the pres-
`ent study, a high concentration
`of expenditures was evident within
`the top 250
`individual
`drugs, as well as within
`the 20 therapeutic
`drug classes. Of note,
`the reported Gini indices underestimate
`the true degree of market concentra-
`tion based on inclusion of all drugs (not just
`the 250 most highly promoted
`drqs).
`empirical evidence supporting
`Leffler5 and Hurwitz and Caves17 documented
`a dual role of pharmaceutical
`promotion:
`informative
`and persuasive.
`In other
`words, promotion not only informs physicians about
`the existence and charac-
`
`1513
`
`

`

`CLINICAL THERAPEUTICS@
`
`to
`teristics of new drug products, but also produces goodwill assets and loyalty
`brand names. Brand loyalty (or, equivalently, high costs of acquiring new infor-
`mation or of learning by experience) may generate persistence
`in prescribing pat-
`terns, and may be an important
`factor influencing physician
`responsiveness
`to
`promotional
`efforts. It has been documented
`that drug promotion
`can influence
`physicians’ prescribing behaviors, and physicians differ in the degree
`to which
`their prescribing behaviors are responsive
`to pharmaceutical
`promotion.10-12,18
`Physicians who prescribe a more concentrated portfolio of drugs tend to prescribe
`drugs with high levels of advertising and popularity. l8 Although
`it is unknown
`whether drug promotion
`has an overall positive or negative
`impact on patient
`outcomes,
`this impact might be expected
`to differ by drug class. Knowledge of
`the fundamental
`functions of pharmaceutical
`promotion
`and the most promoted
`drug classes can help enhance physicians’ ability to analyze promotional materi-
`als conveyed by drug companies
`from an evidence-based
`perspective.
`Physi-
`cians-especially
`concentrated
`prescribers-should
`be proactive
`in acquiring
`in-
`formation
`on drug products
`as well as alternative
`therapies,
`given
`that drug
`promotion
`is provided
`for commercial goals rather
`than for strictly educational
`purposes.
`pro-
`Drugs can be promoted via various avenues, with professional-oriented
`motion
`traditionally dominating
`the drug industry’s marketing efforts.5 Although
`this dominance has been sustained, DTC advertising has attracted
`rapidly grow-
`ing investments
`from drug manufacturers,
`especially after the 1997 publication
`of FDA guidelines on broadcast advertising.7-9 Our data show that, in 1998, ex-
`penditures
`for professional-directed
`promotion were 9 times as high as those for
`DTC advertising ($11,387 million vs $1337 million). In agreement with Rosenthal
`et a1,7 we found
`that unlike promotion
`to health care professionals-which
`is
`used for nearly all promoted pharmaceutical
`products-DTC
`advertising has been
`adopted as a major marketing
`strategy for only a few products. The concentra-
`tion of expenditures
`for DTC advertising was quantified by a Gini index of 0.88,
`compared with Gini indices in the range of 0.50 to 0.66 noted for the other 4 pro-
`motional modes. Expenditures
`for DTC advertising
`in 1998 were concentrated on
`smoking-cessation medications, urinary
`tract medications, postmenopausal
`treat-
`ment, antiallergics, and antihyperlipidemics.
`DTC advertising
`is a recent market-
`ing trend fueled by consumers’ desire to be more involved
`in their own health
`care, drug manufacturers’ willingness
`to promote
`their products
`to an expanded
`audience, and the presence of less restrictive
`regulatory guidelines.
`In turn,
`in-
`creases in DTC advertising may reinforce
`increasing consumer
`involvement
`and
`the destigmatization
`of selected clinical conditions. The existence and growing
`popularity of DTC advertising are not accepted without controversy; nonetheless,
`the rising trend of advertising directly
`to consumers
`is likely to continue.9 By
`being cognizant
`of drug classes
`that are promoted
`directly
`to consumers,
`
`1514
`
`

`

`J. Ma et al.
`
`to address
`
`drug requests and prepare
`
`physicians can anticipate patient-initiated
`them directly
`for different
`strategies are emphasized
`Likewise, knowing what promotional
`drug categories may help improve
`the physicians
`ability to recognize
`the in