- Email: [email protected]
Prescription drug advertising: Trends and implications
0277-9536185 $3.00 + 0.00 Copyright cc)1985 Pergamon Press Ltd
Sot. SCL Med. Vol. 20. No. 3, pp. 191-197, 1985 Prmted in Great Bntain. All nghts reserved
PRESCRIPTION
DRUG ADVERTISING: IMPLICATIONS
LAWRENCE ‘Department
of
Natural
R. KRUPKA’
and
TRENDS AND
ARTHUR M. VENER”
Science and ‘Department of Social Science, Michigan State University, East Lansing, MI 48824, U.S.A.
Abstract-Prescription drug advertisements which appeared in two leading American medical journals in 1972,1977 and 1982 were analyzed to discover possible trends in advertising. The 5016 ads examined showed that ads for the diuretic-cardiovasculars. especially the beta-adrenergic blocking agents and the slow channel inhibitors, as well as the analgesics, had increased, while ads for the anti-infectives and tranquilizers had diminished. The average amount of space allocated for each ad had increased. On the average. most ads (69 %) depicted neither male nor female patients in their graphics, and a trend of increased neutrality was observed. When the hormones were excluded, an average of 21 y0 of the ads showed male patients and IO 7, showed females. Since a relationship was discerned between the leading drugs advertised and the leading prescriptions filled, it was concluded that advertising does have some effect on the prescribing behavior of practitioners. The findings suggest that great investment in advertising is necessary in order to achieve high levels of sales for such drugs as Valium@ (diazepam) which do not have a clear-cut ameliorative effect on a specific physiological condition. On the other hand, it was suggested that saturation advertising would not significantly enhance the sales of such drugs as Dyazide@ (triamterene and hydrochlorothiazide) because of its well established therapeutic value in the control of hypertension. Ten advertising companies, on the average, had purchased 67 Y0of all advertising space and five had purchased almost half (47 %),The same two pharmaceutical companies were among the top five advertisers and the same five were among the top ten for the three years studied.
INTRODUCTION
In the United States, in 1982, the number of prescriptions dispensed totaled approximately one and a half billion. This represents about a 5 0/0increase from the previous year [l 1. Although 200 different drugs accounted for about 63 ?,, of these prescriptions, a pharmacopoeia of at least 8000 other drugs or drug combinations were available to the practitioner [2]. In one year alone, from 1982 to 1983, approximately 30 new drugs. as well as several hundred new alterations in formulae were placed on the market by the United States pharmaceutical industry [2]. In addition, other drugs had become obsolete, or were shown to be ineffective [3]. It has been estimated that the average viable market life for a prescription drug is approximately 5 years [4.5]. Thus, many of the drugs available to practitioners today were unknown to them during their medical training [6]. As a result, physicians are inundated with information regarding drug therapy. These messages emanate from such professional sources as journal articles, colleagues, meetings, pharmacists and from such commerical sources as detailmen, journal advertisements, mail advertisements and samples [7]. A review of the literature regarding the actual impact of these different sources of drug information on a practitioner’s prescribing behavior is limited and based on a few dated studies [6-111. This is particularly true of research dealing with the influence of advertisements in medical journals. Three studies indicate that l-6:, of physicians first learn of new prescription drugs from journal advertisements [7]. while another study shows that 62 O0first learn of a drug’s existence in this manner [ 121. In a study of 235 American general practitioners and specialists in internal medicine. 73 O0reported that advertising in medical journals was very important or
somewhat important. Aft& detailmen and physicians, this made journal advertisements the third leading source of information regarding new prescription drugs [9]. The phenomenon of prescription drug advertising in medical journals has been and remains a source of controversy. Some researchers and practitioners see a basic conflict between the profit motive and the need for accurate, unbiased drug information [13]. Although in such countries as Australia, Great Britain and the United States, there exists legislation and regulatory agencies which control drug advertising to some degree [14], concern remains regarding the amount and the reliability of the information provided in these advertisements [15-221. Furthermore, the advertising techniques employed by some drug companies in developing nations have been criticized on medical and ethical grounds for failing to disclose deleterious side effects and for exaggerating the effectiveness of their drugs [23-251. Other investigators have been concerned with the tendency of advertisements to promulgate cultural stereotypes. They maintain that prescription drug advertisements attempt to convince the practitioner that ordinary aspects of living are pathological and therefore should be treated with a drug. Such aspects of living include a person who is distressed by the performance of tedious tasks, an individual who is bothersome to his family and a person who is concerned about the future [26]. In this vein, women are depicted as being emotional, complaining and irrational, while men are shown as being nonemotional, rational and stoic [27]. In addition, drug advertisements link cardiovascular conditions, such .as angina, hyperlipidemia and hypertension primarily to men [28]. Other advertisements medicalize the normal aging process and depict 191
LAWRENCE R. KRUPKA and ARTHUR M. VENER
192
the aged as apathetic, debilitated, disruptive, sluggish and out of control [29]. While some practitioners have maintained that medical journals should not permit themselves to become dependent on income generated from drug advertisements [26], others have argued that advertising has an important place in the continuance of therapeutic education [30]. In fact, the position has been taken that prescription drug advertisements are of an absolute necessity because they subsidize the publication of medical journals. One medical journal has decried the fact that the editors were forced to reduce the size of recent issues because of a diminished number of drug advertisements [31]. Readers of the journal were urged to show their appreciation of the drug companies to pharmaceutical detailmen in the expectation that this message would be transmitted back to company officials. Although the research extant does not permit us to make an accurate estimation of the relative impact of journal prescription drug advertising on the prescribing behavior of the practitioner in comparison to other sources of therapeutic information, there is little doubt that it does have some effect. In addition to the evidence cited above, it can be assumed that the drug industry would not invest millions of dollars in these advertisements if their market research had indicated that these advertisements were of limited or no influence. The study renorted, herein, was designed to examine some basic trends and implications of prescription drug advertising over an 1l-year period, 1972-1982. PROCEDURE
The combined prescription drug advertisements of two prestigious American medical journals, The New Englnnd Journal of Medicine (NEJM) and The Journal of the American Medical Association (JAMA), with approximately 500,000 subscribers [32], were examined. The drug advertisements in these journals for the years 1972, 1977 and 1982 were analyzed in terms of the total number of advertisements, the number of pages allocated for each advertisement, and the therapeutic category of the drugs advertised. In addition, a rank ordering was made of the drug companies with regard to the total amount of advertising space they had purchased. The sex of the patients depicted in advertisements was also examined. In order to obtain a tentative estimate of the relative impact of journal advertising on the prescribing behavior of physicians, the leading 15 advertised drugs with regard to space allocation for the years 1972,1977 and 1982 were compared with the actual prescriptions filled during these years as well as for those of 1973 and 1978. This latter information was abstracted from Pharmacy Times [l]. Since for 1983, prescriptions tilled information was not available, a comparison of the prescriptions filled for this year with the 1982 drug advertising data was not possible. At this juncture, several methodological precautions should be mentioned. Prescriptions filled data for 1973 and 1978 were selected for comparative purposes because it was felt that the effects of an advertising campaign begun in one year would also have an effect on prescriptions filled for the following year. In
addition, information obtained by physicians regarding drug therapy is not limited to medical journal advertisements. We believe that the 15 most advertised drugs in the two leading medical journals would be part of a multi-media advertising campaign for these drugs. Another potential problem with utilizing prescriptions filled data is that of patient compliance. In this regard. we have assumed that no one particular drug would necessarily be singled out for patient noncompliance. Rather, other factors such as prescription labeling, differences in personality and educational level and doctor-patient relationships would be of greater importance in predicting patient compliance [33-371. The therapeutic classificatory scheme employed in this study was obtained from Dr[rg Ftrcts trrltl Comparisons: 1983 edition [2]. The particular ses of the patient targeted in advertisements was determined by examining the graphic content. Wording was not utilized. For example, a number of advertisements for the drug Megace@ (megestrol acetate), a palliative treatment for advanced carcinoma of the breast or endometrium, did not depict females in their graphics and therefore were considered to be neutral. i.e. targeted to neither sex. In addition. if individuals of both sexes were depicted in the advertisement, at least two-thirds of them had to be of the same sex in order to be classified as an advertisement aimed at that sex. At times, only a segment of an individual was shown in the advertisement, e.g. a hand. If this hand held a purse. possessed painted fingernails, or had feminine type jewelry and rings, it was classified as being a female targeted advertisement. In those instances when the sex depicted was difficult to ascertain, both authors had to arrive at a consensus. In the very few cases where no consensus was achieved, the advertisement was considered to be neutral. Finally, when neither sex was depicted, the advertisement was also considered to be neutral. RESULTS
Trends by therapeutic category
In 1972, 128 different prescription drugs were advertised. In 1977, 109 were advertised and in 1982. 141. Table 1 demonstrates no clear-cut trend either in the number of advertisements or in the number of pages purchased by the pharmaceutical manufacturers. The only consistent trend which can be deduced from the table is an increase in the average space allocated for each advertisement. In 1972, there were 2.0 pages per ad, in 1977,2.3 and in 1982,2.8. In 1972. only 14 advertisements contained six or more pages, while in 1982, 122 advertisements were of such length. Although the NEJM consistently contained a larger total number of advertisements and a greater number Table 1. Number
of
prescrlpt~on drug adverttsements and space allocatton by year 1977
1972
JAMA NEJM
Totals
1982
Ads
Pages
Ads
Pages
-\ds
Pages
814 1009 1823
1672 1932 3604
651 663 1314
1496 1866 3062
X15 IO64 I x79
2216 2948 5164
Prescription drug advertising: trends and implications of pages for these advertisements, a very similar average increase of pages per advertisement was in evidence for both journals. For the three years studied, the differences between the journals in this regard were miniscule. Table 2 shows that approximately 82 % of all drug advertisements with regard to space were due to three therapeutic categories, i.e. anti-infectives, diureticcardiovasculars and central nervous system (CNS) drugs. To simplify data analysis, only the number of advertising pages were examined. The percentages for the number of advertisements and pages were very similar for all three years. The relative number of pages allocated for the diuretic-cardiovasculars increased from 1972 to 1982, while the relative number of pages for thi: anti-infectives and CNS drugs diminished. In 1972, CNS drugs, which include the anti-depressants, tranquilizers, hypnotics, sedatives and analgesics, accounted for the largest percentage (41%) of advertising pages. In 1982, however, the diureticcardiovasculars accounted for the greatest percentage of advertising pages (41”//,). With the exception of the hormones, the remaining types of drugs accounted for such a relatively low percentage of pages that definitive trends were difficult to ascertain. However, the hormones, primarily antidiabetics, do show a downward trend in the relative number of pages purchased by the pharmaceutical companies. Since the diuretic-cardiovasculars and CNS drugs together comprise over 70% of the total number of pages allocated in 1982, these medications were analyzed in greater detail. Table 3 demonstrates that with regard to the diuretic-cardiovasculars, the cardiac glycosides and peripheral vasodilators were no longer advertised in 1982. Also, much less emphasis with regard to space allocation was given to the diuretics and antihyperlipidemics. On the other hand, a notable increase in the relative amount of advertising for the beta-adrenergic blockers and th& slow channel inhibitors occurred. Table 4 indicates that two types of CNS drugs, analgesics and psychotherapeutics together, accounted for over 90 O/bof the total amount of pages advertised for this therapeutic category in 1982. When the years 1977 and 1982 are compared with that of 1972, a substantial increase in the number of pages advertised is found for the analgesics, while the psychotherapeutics, primarily tranquilizers, showed a 209, decrease. CNS stimulants, anticonvulsants, muscle relaxants and antiparkinsonism Table
2. Space allocation for drug advertisements category and year (in O;)
Ant+infectives Antineoplastics Blood modifiers Central nervous system Diuretic-cardiovasculars Gastromtestinals Hormones Nutrltlonal Respuatory Topic& Miscellaneous Totals ( “,,I .\
by therapeutic
1972 Pages
1977 Pages
1982 Pages
23
13 4
11 1 1 30 41 4 3 2 3 2 2 100 5164
1 41 19 3 9 1 2
34 34 4 6 3 2 -
1
100 .3604
100 3062
193
Table 3. Space allocation for advertised dluremxardlovascular prescription drugs by therapeutic subcategory and year fin :‘,;,J
Antl-angina1 Anti-arrhythmics Antihyperlipidemlcs Antihypertensives Beta-adrenergic blocking agents Cardiac glycosides Duetics Peripheral vasodilators Slow channel Inhibitors Totals ( ;,A) N
1972 Pages
1977 Pages
1982 Pages
14 6 9 23
4 4 10 47
10 5 5 22
7 2 31 8
10
31
24
I2
100 1053
9 100 2120
100 693
agents, although advertised to some extent in 1972, were no longer advertised in 1982. Leading drugs aduertised
An analysis of specific drugs by brand names shows that only one drug, Dalmane@ (flurazepam HCl) was one of the leading 15 drugs advertised for all 3 years. Four drugs, Valium@ (diazepam), Sinequan@ (doxipin HCl), Inderal@ (propranolol HCl) and Tagamet@ (cimetidine) were among the leading 15 drugs advertised for two of the 3 years. Although Valium@ was the most advertised drug in terms of the number of pages for 1972 and 1977, it was not among the leading 15 in 1982. In this year, 29 other drugs had received greater advertising space. As noted in previous tables, we again observe an increase in the amount of advertising space allocated to the diurecticcardiovasculars. In 1982, 9 of the top 15 advertised drugs were of this type, 5 in 1977 and only 2 in 1972. Five of the 9 in 1982 were beta-adrenergic blocking agents. In contrast to the diuretic-cardiovasculars, an apparent downward trend in advertising space for the CNS type drugs was observed, especially when 1977 and 1982 were compared with 1972. In 1982,4 of the leading 15 advertised drugs were of a CNS type, in 1977, there were 3 and in 1972, there were 8. A majority of the CNS drugs in 1972 and 1977 were psychotherapeutics. Only one psychotherapeutic, Sinequan@ (doxepin HCl) was in the top 15 in 1982. Over the years, the leading 15 drugs, combined, accounted for a decreasing proportion of the total number of pages allocated for advertising. In 1972, the top 15 drugs accounted for 45 y0 of the total advertising space, in 1977, 41 y0 and in 1982, 39 % Table 4. Space allocation for advertised central nervous system prescription drugs by therapeutic subcategory and year (m 7;)
Analgesics Anticonvulsants Anti-emetic-antivertigo agents Antiparkinsonism agents Muscle relaxants Psychotherapeutics Sedative-hypnotics Stimulants Totals ( O,)
h-
1972 Pages
1977 Pages
1982 Pages
18 4
42
1 3 2 64 7 1 100 1477
3 1 1 44 9
44 8
100 1062
100 1552
47 1
LAWRENCER. KRUPKA and ARTHUR M. VENER
194
drugs udcertised
Leading
and prescriptions
jilled
pages purchased in 1972 and 1977 and 68”,, in 1981. Five of the companines. Ayerst. CIBA. Hoflmann-LaRoche. Merck. Sharp & Dohme and Pfizer, were among the leading ten advertisers for each ofthe 3 years. In contrast to the leading ten advertisers. the top five, representing about 12 OO.on the average. of the companies which advertised. accounted for almost half (47”,) of the total number of advertising pages purchased. Two manufacturers. Pfizer and Hoffmann-LaRoche were among the top five advertisers for each of the 3 years studied.
A comparison was made between the leading 15 drugs advertised and prescriptions filled for 1972 and 1973; the leading drugs in 1977 with prescriptions filled for 1977 and 1978; and the leading drugs in 1982 with prescriptions filled data for this same year. An apparent association between these variables seems to exist. Approximately one-fifth (21 “,) ofthe leading 15 advertised drugs were also found to be one of the leading 15 drugs with regard to the number of prescriptions filled for the 5 years analyzed. Two-fifths (39 “,) were in the leading 50; three-fifths were in the leading 100 while four-fifths (79 9;) were in the leading 200. A comparison of the leading 200 prescription drugs filled in 1972 and 1973 showed that 10 of the leading 15 drugs advertised in 1972 had advanced their relative ranking with regard to the number of prescriptions filled in 1973. Two of the drugs retained the same relative position. One of these latter drugs was Valium@ (diazepam) which was the mos’t common prescription filled medicine. Only 2 of the leading 15 drugs advertised demonstrated a reduction in rank with regard to the number of prescriptions filled. One drug, Garamycins (gentamicin), one of the 15 most advertised drugs in 1972, was not one of the leading 200 drugs for prescriptions filled either for 1972 or 1973. A comparison of the top 200 prescriptions filled with the 15 leading advertised drugs in 1977 and 1978 showed similar results. Six of the leading 15 advertised drugs in 1977 increased their rank with regard to prescriptions filled in 1978. Two maintained their relative rank and four demonstrated a decrease. Three of the leading 15 drugs advertised in 1977, Apresazide@ (hydrochlorothiazide + hydralazine HCl), Amikin@ (amikacin sulfate) and Megace@ (megestrol acetate) were not one of the leading 200 prescriptions filled for either year. Leading
phurmaceuticul
Sex of‘ patients depicted
adaertisers
An average of 42 different pharmaceutical manufacturers advertised in both journals during 1972, 1977 and 1982. The ten leading advertisers, representing an of 24”; of the companies doing the average advertising, accounted for 66 y;, of the total advertising Table
5. Sex of patients
deplcted
m prescription
drug
Central Ant+mfectives
( “,,)
lY77 Neutral Malr \
and year (III
agents
cardiovasculars
mtestmals
52
61
27
28
loo
categoq
Gastro-
21 I00
443
639
73
59
22
23 IX
( “,,)
by therapeutic Diuretic-
10
Female Totals
advertwments
nervous
71 19
Female TTotals
system
in trduertisrmerzts
Five therapeutic categories, combined. accounted for an average of 92 O0 of the total number of advertisements placed. Table 5 shows the sex of the patients depicted in the advertisements for these five therapeutic categories. On the average, most of the advertisements (69 “i) depict neither male nor female patients in their graphics. Frequently, no human figure. or parts of human figures. were included. When human figures were presented, they were frequently shown in such an amorphous manner that it was not possible to determine which sex they were intended to represent. In their graphics, many advertisements focused their efforts on a display of how effective their products were in the alleviation of medical problems. A trend of increased neutrality of advertisements was discerned. In 1972, an average of 62 “, of the ads were neutral: in 1977, 69% and, in 1982, 77”“. In 1972, an average of 18 y0 of the advertisements showed male patients. in 1977, 18;; and in 1982, 13 “,,. With regard to advertisements showing female patients. the 1972 average was 20 y,;, in 1977, I3 “<,and in 1982,9 OO.When the hormones are excluded, an average of 21 no of the ads depicted male patients and IO “” depicted females. In three types of drugs. the diuretic-cardiovasculars. CNS and gastrointestinals, male patients appear more frequently in advertisements than female patients. Only with hormones do we find a consistently greater number ofadvertisements depicting female patients. In 1972, although more female patients were shown in anti-infective advertisements, for both 1977 and 1982. male oriented advertisements were depicted more frequently for this class ofdrugs. Since special attention
438
57 35 8 100 426
56 3x 100
72
IO0
loo 137
58 9 33
loo
100
178
491
80
65
87
77
I2
24
I2
IX
100
I00
106
216
625
695
II
HOrIIlOIleS
21
II 100
“,,I
I 00
37
100 69
78 21
IO0
7x
100
h7
Prescription
drug advertising:
has been given in the literature to CNS drug use by women. especially with regard to the analgesics and psychotropics. these two subcategories were analyzed separately. Even in this case. we again found a trend towardincreasingsexualneutralityinadvertising. With the exception of the hormones, in only one instance, psychotropic advertisements for 1977, did we observe a clear-cut greater percentage of female oriented advertisements (females 33 “/,. males 14 “,i). DISCUSSION
The observed increase in advertising for the diureticcardiovasculars between 1972 and 1982 is associated with the relatively recent introduction of such effective new compounds as the beta-adrenergic blocking agents and the slow channel inhibitors. The market potential for such drugs continues to be quite large since heart disease is still the leading cause of death among the populations ofdeveloped nations [38]. Since, with age, the heart becomes a poorer pump, the management of older patients with heart related complications represents a substantial part oftheir medical care [39]. The finding that both the CNS and anti-infective drugs showed a decline in advertising emphasis may be associated with the fact that no major therapeutic advances have been made for these two types of therapeutics. Additionally, since there has been an outcry of concern with regard to the deleterious effects of the psychotropics [40]. it is possible that the pharmaceutical companies have decided that their marketing efforts would best be served by emphasizing other products. The fact that most prescription drug advertisements were targeted to neither sex and that this trend has increased. may indicate that the pharmaceutical manufacturers have attempted to defuse accusations of sexual bias in the advertisement of their products. Women, in particular, have been stereotyped in advertisements as needing emotional support through the aid of psychotropics [40-421. Between 1968-1972, Prather and Fidel1 [27] found that 59”i, of the psychotropic drug advertisements showed women while in 1973. McRee et al. [43] found that 64 ‘A of the drugs used in psychiatric advertisements in the Aniericari Jourrral of Psychiatry depicted women. However, in this study. with regard to theanalgesicsand psychotropics. in only one year, 1977, did we find that a greater percentage of the advertisements depicted more female than male patients. Even in this instance, only 33 “(, of the advertisements were female oriented, while the majority (53”,) were neutral. In essence, our findings are supported by Thompson [28], who also found that when ads were targeted to a specific sex. they were oriented more toward males. Only in the case of the hormones. primarily antidiabetics, was a greater percentage of female oriented advertisements found. Even in this case. the majority of advertisements were neutral. Thompson [28] has expressed theconcern that male bias in advertisements might cause physicians to become predisposed to perceive men as needing more medical treatment than women and thus some health problems ofwomen may be overlooked. It is difficult to understand why. in 1977 and 1982. three times the number of analgesic advertisements were male oriented. In thisvein are the pharmaceutical companies
trends
and implications
195
suggesting that men have a lower threshold for pain and therefore need a greater amount and variety of prescribed analgesics? We do not believe this to be the case. Rather, we think that being profit oriented organizations, they simply wish to increase sales of these products. Women have been found to be the prime users of analgesic compounds [41,44-451 and therefore targeting advertisements toward them would probably not dramatically increase sales. Rather, there would be greater potential for increased sales if advertisements were more male oriented. This same motivation may apply to the other therapeutic categories as well. The fact that the average number of pages per advertisement has increased and that a relationship exists between the leading drugs advertised and the leading prescriptions filled implies that prescription drug advertising in professional journals is of financial benefit to pharmaceutical manufacturers. The advertising of Tagamet@ (cimetidine), a histamine H2 antagonist, indicated for the healing of duodenal ulcers, is a case in point. In 1977, a total of 68 pages was allocated for the advertising of this drug. In 1982, the advertising space purchased for this drug was increased by 86 0/0(117 pages). We also found that this product was among the 15 most advertised drugs for 1977 and 1982. There is little doubt that increased advertising of this product was a factor involved in its increased sales. In 1977, Tagamet@ was not among the leading 200 drug prescriptions filled in the United States. In the very next year, it ranked fiftieth. In 1979, it ranked twenty-second, in 1980, fifteenth, in 1981, ninth and in 1982, seventh. During this advertising campaign, not only were the number of advertisements and pages increased, but there was also an increase in the number of identical advertisements in the same weekly issue. For example, the 30 September 1982 issue of the NEJM contained five separate advertisements for Tagamet@ for a total of 12 pages. This saturation type advertising campaign, however, was probably not limited to the two professional journals analyzed. More likely than not, the campaign also included the use of still other medical journals as well as manufacturers’ representatives, the direct mailing of brochures, and information provided at special conferences and professional meetings. Although it is difficult to assess the actual contribution that advertising in the JAMA and the NEJM had on the increased sales of Tagamet@, it is our belief that its effect was not insignificant. Dyazide@ (triamterene and hydrochlorothiazide) is another case in point. In 1972, we found it to be one of the 15 leading advertised drugs. This was not so in 1977 and 1982. Regarding prescriptions filled, it was ranked 44th in 1972 and climbed to be the leading prescribed drug in 1982 in the United States. With little doubt, physicians have found this drug to be very effective in the control of hypertension. As a result, -saturation advertising of this effective medication probably would not have significantly increased its sales and therefore the manufacturer may have decided that it was wiser to invest advertising funds into the marketing of their other products. A question may be raised regarding the marketing of drugs which do not have a clear-cut ameliorative effect on a specific physiological condition. An example of
196
LAWRENCER. KRUPK~ and
such a drug is Valium@ (diazepam), used for the management of anxiety disorders. Anxiety, a feeling of apprehension, uncertainty and fear without apparent stimulus, may or may not be directly linked to definitive physiological changes [46]. Some critics have maintained that the prescribing of Valium@ has implications which go beyond medical practice and is indicative of a mass cultural phenomenon [47]. In fact, Cooperstock and Parnell [40] suggest that Valium@ should be perceived in a manner similar to that of alcohol, i.e. as being a social drug. Considering the ambivalence practitioners may have when prescribing Valium@, as well as the other benzodiazepines, advertising may be an influential factor in their decision to finally prescribe it. In 1972 and again in 1977, Valium@ was given more advertising space than any other prescription drug, but in 1982, it did not appear among the leading 15 advertised drugs. In terms of prescriptions filled, it was the leading drug from 1972 to 1981. Only in 1982, in association with the reduction of advertising, did it decline to a rank of three. It is possible to conclude that, at least for nonspecific therapeutic drugs, greater investment in advertisement campaigns is necessary to maintain high sales levels. In terms of medical ethics, we believe that only effective drugs which meet the requirements of substantial evidence should be advertised. Twenty-six of the drugs advertised in 1972, 1977 and 1982 were reported by Wolfe [3] as being less than effective or ineffective. Because of the United States Food and Drug Agency’s reticence to expedite their removal, these drugs have remained on the market. All were introduced at least 20 or more years ago. The problem here is that the drugs which are advertised may not be the ones which are the most effective. If it is accepted that prescription drug advertising in medical journals does have an impact on prescribing behavior, then the extensiveness of advertising by any one or combination of companies represents another potential ethical problem. The finding that the top ten advertising companies, on the average, accounted for space for the three years 67’;b of all advertising examined, and that five accounted for almost half (47’:;)) of advertising space, portends the hazards typically associated with monopolistic business practices. This problem is further exacerbated by the fact that the same two pharmaceutical companies were among the leading five advertisers for the 3 years analyzed, and the same five were among the leading top ten. Since the cost of advertising in the major medical journals represents a considerable investment of funds, smaller companies may have some difficulty in obtaining exposure for their products. Again, the possibility exists that less effective medications will be advertised. A resolution to this problem is suggested by a 1983 law enacted by the United States Congress regarding ‘orphan drugs’-pharmaceutical products that have been introduced in other countries, but not made available in the United States. These drugs are frequently indicated for relatively rare diseases and therefore are not profitable. The law provides a federal subsidy for pharmaceutical companies, private investigators and non-profit organizations to help in the development and marketing of such drugs[48]. Perhaps federal funds can also be made available for
ARTHUR M. VENER
the advertising campaigns of effective drugs which have been developed by individual investigators or small pharmaceutical companies who might be unable to amass the large amounts of funding which would be required to launch a successful advertising campaign in order to gain sufficient exposure for their products.
REFERETTCES
Anonymous. The top 200 drugs. Pharm. Times April 1972 to 1983 issues. Boyd
J. R. (Ed.)
Drug
Facts
und Comparisons.
Lippincott, St Louis, MO, 1983. Wolfe S. M. Pills that Don’t Work. Warner
6. 7
8.
9.
10. 11.
12.
13. 14.
15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27.
J. B.
Books, New York, 1982. Stolley P. D. and Lasagna L. Prescribing patterns of physicians. J. chron. Dis. 22, 395, 1969. Norwood G. J. and Smith M. C. Market mortality of new products in the pharmaceutical industry. J. .Im. pharm. Ass. 11, 592, 1971. Hemminki E. Review of literature on the factors affecting drug prescribing. Sot. Sci. Med. 7, 51, 1973. Herman C. M. and Rodowskas C. A. Communicating drug information to physicians. J. med. Educ. 51, 189, 1976. Becker M. H., Stolley P. D., Lasagna L., McEvilla J. D. and Sloan L. M. Differential education concerning therapeutics and resultant physician prescribmg patterns. J. med. Educ. 47, 118, 1972. Linn L. S. and Davis M. S. Physicians’orientation toward thelegitimacyofdruguseand theirpreferredsourceofnew drug information. Sot. Sci. Med. 6, 199. 1972. Worthen D. B. Prescribing influences: an overview. Br. J. med. Educ. 7, 109, 1973. Miller R. R. Prescribing habits of physicians: a review of studies prescribing drugs. Drug Intel/. clin. Phnrm. 8. 81, 1974. Eaton G. and Parish P. Sources of drug information used by general practitioners. R. Coil. gen. &act. 26, Suppl. I, 58, 1976. Worthen D. B. Clinical drug information: profit or knowledge oriented. Ethics Sci. Med. 3, 235, 1976. Najman J. M., Siskind V. and Bain C. Prescription drug advertising: medical journal practices under different types of control. Med. J. Aust. 1, 420, 1979. Ditzion B. R. Psychotropic drug advertisements. Ann. intern. Med. 75, 473, 1971. Seidenberg R. Advertising and abuse of drugs. N. Engl. .I. Med. 284, 789, 1971. Kimmey J. R. Drug efficacy andajournal’s responsibility. Am. .I. puhl. Hlth 62, 461. 1972. Simpsin F. 0. Omissions in advertising. N. Z. med. J. 77, 412, 1973. Stimson G. V. Information contained in drug advertisements. Br. med. J. 4, 508. 1975. Holmes J. D. Prescribing information in drug advertisements. N. Z. med. J. 86, 467, 1977. Lefferts G. Journal advertising. Ann. inter,z. Med. 88, 581. 1978. Relman A. S. Who reviews the ads’? N. Engl. J. .&fed. 301, 999. 1979. Silverman M. The Drugging of the Ameruns. Universtty of California Press, Berkeley, 1976. Yudkin J. S. Provision of medicines in a developing country. Luncet I, 801, 1978. Victora C. G. Statistical malpracttce in drug promotton: a case-study from Brazil. Sot. Sci. .Lfed. 16, 707, 1982. Seidenberg R. Drug advertising and perception of mental illness. M&t. Hvgr55, 21, 1971. Prather J. and Fidel1 L. S. Sex differences in the content and style of medical advertisements. Sot. ScI. Med. 9, 23. 1975.
Prescription drug advertising: trends and implications 2X. Thompson E. L. Sexual bias in drug advertisements. Sot. Sci. Med. 13A, 187, 1979. 29. Smith M. C. Portrayal of the elderly in prescription drug advertising. Gerontologist 16, 329, 1976. 30. Robinson R. G. Medical journals and advertising. N. Z. med. J. 87, 139. 1978. 31. Woody E. The paradox of pharmaceutical advertising. .I. med. Ass. Ga. 69, 379. 1980. 32. L’lrich’s Internarional Periodicals Directory, 22nd Edition. R. R. Bowker, New York, 1983. 33. Blackwell B. Patient compliance. N. Engl. J. Med. 289, 249. 1973. 34. Haynes R. B. Sackett D. L., Taylor W., Roberts R. S., Tech M. and Johnson A. L. Manipulation of the therapeutic regimen to improve compliance: conceptions and misconceptions. C/in. Pharmac. Ther. 22, 125, 1977. 35. Boyd J. R., Covington T. R., Stanaszek W. F. and Coussons R. T. Drug defaulting. Am. J. Hosp. Pharm. 31, 485. 1974. 36. Whitney H. A. Educating the public about drugs. Drug Intell. c/in. Pharm. 9, 571. 1975. 37. Caldwell J. R., Cobb S., Dowling M. D. and De Jongh D. The dropout problem in antihypertensive treatment. J. chror~. Dis. 22, 579, 1970. 38. Pelletier K. R. Holistic Medicine. Delacorte Press, New York. 1979. 39. Harris R. Special problems of geriatric patients with
40. 41. 42.
43.
44. 45.
46. 47.
48.
197
heart disease. In Clinical Aspects of Aging (Edited by Reichel W.), p. 45. Williams & Wilkins, Baltimore, MD, 1978. Cooperstock R. and Parnell P. Research on psychotropic drug use. Sot. Sci. Med. 16, 1179, 1982.. Prather J. E. and Fidel! L. S. Drug use and abuse among women: an overview. Int. J. Addict. 13, 873, 1978. Linn L. S. Physician characteristics and attitudes toward legitimate use of psychotherapeutic drugs. J. Hlth WC. Behau. 12, 132, 1971. McRee C., Corder B. F. and Haizlip T. Psychiatrists’ responses to sexual bias in pharmaceutical advertising. Am. J. Psychiat. 131, 1273, 1974. Krupka L. R., Vener A. M., Stewart C. S. and ZaengleinSenger M. M. Patterns of aspirin use among American youth. Int. J. Addict. 13, 911, 1978. Vener A. M., Krupka L. R. and Climo J. J. Drugs (prescription, over-the-counter, social) and tile young adult: use and attitudes. Jnt. J. Addict. 17, 399, 1982. Friel J. P. (Ed.) Dorland’s Illusrrated Medical Dictionary, 25th Edition. W. B. Saunders, Philadelphia, 1974. Bellantvono C., Reggi V., Tognoni G. and Garattini S. Benzodiazepines: clinical pharmacology and therapeutic use. Drugs 19, 195, 1980. Anderson K. Orphan Drugs. Linden Press, New York, 1983.
Sot. SCL Med. Vol. 20. No. 3, pp. 191-197, 1985 Prmted in Great Bntain. All nghts reserved
PRESCRIPTION
DRUG ADVERTISING: IMPLICATIONS
LAWRENCE ‘Department
of
Natural
R. KRUPKA’
and
TRENDS AND
ARTHUR M. VENER”
Science and ‘Department of Social Science, Michigan State University, East Lansing, MI 48824, U.S.A.
Abstract-Prescription drug advertisements which appeared in two leading American medical journals in 1972,1977 and 1982 were analyzed to discover possible trends in advertising. The 5016 ads examined showed that ads for the diuretic-cardiovasculars. especially the beta-adrenergic blocking agents and the slow channel inhibitors, as well as the analgesics, had increased, while ads for the anti-infectives and tranquilizers had diminished. The average amount of space allocated for each ad had increased. On the average. most ads (69 %) depicted neither male nor female patients in their graphics, and a trend of increased neutrality was observed. When the hormones were excluded, an average of 21 y0 of the ads showed male patients and IO 7, showed females. Since a relationship was discerned between the leading drugs advertised and the leading prescriptions filled, it was concluded that advertising does have some effect on the prescribing behavior of practitioners. The findings suggest that great investment in advertising is necessary in order to achieve high levels of sales for such drugs as Valium@ (diazepam) which do not have a clear-cut ameliorative effect on a specific physiological condition. On the other hand, it was suggested that saturation advertising would not significantly enhance the sales of such drugs as Dyazide@ (triamterene and hydrochlorothiazide) because of its well established therapeutic value in the control of hypertension. Ten advertising companies, on the average, had purchased 67 Y0of all advertising space and five had purchased almost half (47 %),The same two pharmaceutical companies were among the top five advertisers and the same five were among the top ten for the three years studied.
INTRODUCTION
In the United States, in 1982, the number of prescriptions dispensed totaled approximately one and a half billion. This represents about a 5 0/0increase from the previous year [l 1. Although 200 different drugs accounted for about 63 ?,, of these prescriptions, a pharmacopoeia of at least 8000 other drugs or drug combinations were available to the practitioner [2]. In one year alone, from 1982 to 1983, approximately 30 new drugs. as well as several hundred new alterations in formulae were placed on the market by the United States pharmaceutical industry [2]. In addition, other drugs had become obsolete, or were shown to be ineffective [3]. It has been estimated that the average viable market life for a prescription drug is approximately 5 years [4.5]. Thus, many of the drugs available to practitioners today were unknown to them during their medical training [6]. As a result, physicians are inundated with information regarding drug therapy. These messages emanate from such professional sources as journal articles, colleagues, meetings, pharmacists and from such commerical sources as detailmen, journal advertisements, mail advertisements and samples [7]. A review of the literature regarding the actual impact of these different sources of drug information on a practitioner’s prescribing behavior is limited and based on a few dated studies [6-111. This is particularly true of research dealing with the influence of advertisements in medical journals. Three studies indicate that l-6:, of physicians first learn of new prescription drugs from journal advertisements [7]. while another study shows that 62 O0first learn of a drug’s existence in this manner [ 121. In a study of 235 American general practitioners and specialists in internal medicine. 73 O0reported that advertising in medical journals was very important or
somewhat important. Aft& detailmen and physicians, this made journal advertisements the third leading source of information regarding new prescription drugs [9]. The phenomenon of prescription drug advertising in medical journals has been and remains a source of controversy. Some researchers and practitioners see a basic conflict between the profit motive and the need for accurate, unbiased drug information [13]. Although in such countries as Australia, Great Britain and the United States, there exists legislation and regulatory agencies which control drug advertising to some degree [14], concern remains regarding the amount and the reliability of the information provided in these advertisements [15-221. Furthermore, the advertising techniques employed by some drug companies in developing nations have been criticized on medical and ethical grounds for failing to disclose deleterious side effects and for exaggerating the effectiveness of their drugs [23-251. Other investigators have been concerned with the tendency of advertisements to promulgate cultural stereotypes. They maintain that prescription drug advertisements attempt to convince the practitioner that ordinary aspects of living are pathological and therefore should be treated with a drug. Such aspects of living include a person who is distressed by the performance of tedious tasks, an individual who is bothersome to his family and a person who is concerned about the future [26]. In this vein, women are depicted as being emotional, complaining and irrational, while men are shown as being nonemotional, rational and stoic [27]. In addition, drug advertisements link cardiovascular conditions, such .as angina, hyperlipidemia and hypertension primarily to men [28]. Other advertisements medicalize the normal aging process and depict 191
LAWRENCE R. KRUPKA and ARTHUR M. VENER
192
the aged as apathetic, debilitated, disruptive, sluggish and out of control [29]. While some practitioners have maintained that medical journals should not permit themselves to become dependent on income generated from drug advertisements [26], others have argued that advertising has an important place in the continuance of therapeutic education [30]. In fact, the position has been taken that prescription drug advertisements are of an absolute necessity because they subsidize the publication of medical journals. One medical journal has decried the fact that the editors were forced to reduce the size of recent issues because of a diminished number of drug advertisements [31]. Readers of the journal were urged to show their appreciation of the drug companies to pharmaceutical detailmen in the expectation that this message would be transmitted back to company officials. Although the research extant does not permit us to make an accurate estimation of the relative impact of journal prescription drug advertising on the prescribing behavior of the practitioner in comparison to other sources of therapeutic information, there is little doubt that it does have some effect. In addition to the evidence cited above, it can be assumed that the drug industry would not invest millions of dollars in these advertisements if their market research had indicated that these advertisements were of limited or no influence. The study renorted, herein, was designed to examine some basic trends and implications of prescription drug advertising over an 1l-year period, 1972-1982. PROCEDURE
The combined prescription drug advertisements of two prestigious American medical journals, The New Englnnd Journal of Medicine (NEJM) and The Journal of the American Medical Association (JAMA), with approximately 500,000 subscribers [32], were examined. The drug advertisements in these journals for the years 1972, 1977 and 1982 were analyzed in terms of the total number of advertisements, the number of pages allocated for each advertisement, and the therapeutic category of the drugs advertised. In addition, a rank ordering was made of the drug companies with regard to the total amount of advertising space they had purchased. The sex of the patients depicted in advertisements was also examined. In order to obtain a tentative estimate of the relative impact of journal advertising on the prescribing behavior of physicians, the leading 15 advertised drugs with regard to space allocation for the years 1972,1977 and 1982 were compared with the actual prescriptions filled during these years as well as for those of 1973 and 1978. This latter information was abstracted from Pharmacy Times [l]. Since for 1983, prescriptions tilled information was not available, a comparison of the prescriptions filled for this year with the 1982 drug advertising data was not possible. At this juncture, several methodological precautions should be mentioned. Prescriptions filled data for 1973 and 1978 were selected for comparative purposes because it was felt that the effects of an advertising campaign begun in one year would also have an effect on prescriptions filled for the following year. In
addition, information obtained by physicians regarding drug therapy is not limited to medical journal advertisements. We believe that the 15 most advertised drugs in the two leading medical journals would be part of a multi-media advertising campaign for these drugs. Another potential problem with utilizing prescriptions filled data is that of patient compliance. In this regard. we have assumed that no one particular drug would necessarily be singled out for patient noncompliance. Rather, other factors such as prescription labeling, differences in personality and educational level and doctor-patient relationships would be of greater importance in predicting patient compliance [33-371. The therapeutic classificatory scheme employed in this study was obtained from Dr[rg Ftrcts trrltl Comparisons: 1983 edition [2]. The particular ses of the patient targeted in advertisements was determined by examining the graphic content. Wording was not utilized. For example, a number of advertisements for the drug Megace@ (megestrol acetate), a palliative treatment for advanced carcinoma of the breast or endometrium, did not depict females in their graphics and therefore were considered to be neutral. i.e. targeted to neither sex. In addition. if individuals of both sexes were depicted in the advertisement, at least two-thirds of them had to be of the same sex in order to be classified as an advertisement aimed at that sex. At times, only a segment of an individual was shown in the advertisement, e.g. a hand. If this hand held a purse. possessed painted fingernails, or had feminine type jewelry and rings, it was classified as being a female targeted advertisement. In those instances when the sex depicted was difficult to ascertain, both authors had to arrive at a consensus. In the very few cases where no consensus was achieved, the advertisement was considered to be neutral. Finally, when neither sex was depicted, the advertisement was also considered to be neutral. RESULTS
Trends by therapeutic category
In 1972, 128 different prescription drugs were advertised. In 1977, 109 were advertised and in 1982. 141. Table 1 demonstrates no clear-cut trend either in the number of advertisements or in the number of pages purchased by the pharmaceutical manufacturers. The only consistent trend which can be deduced from the table is an increase in the average space allocated for each advertisement. In 1972, there were 2.0 pages per ad, in 1977,2.3 and in 1982,2.8. In 1972. only 14 advertisements contained six or more pages, while in 1982, 122 advertisements were of such length. Although the NEJM consistently contained a larger total number of advertisements and a greater number Table 1. Number
of
prescrlpt~on drug adverttsements and space allocatton by year 1977
1972
JAMA NEJM
Totals
1982
Ads
Pages
Ads
Pages
-\ds
Pages
814 1009 1823
1672 1932 3604
651 663 1314
1496 1866 3062
X15 IO64 I x79
2216 2948 5164
Prescription drug advertising: trends and implications of pages for these advertisements, a very similar average increase of pages per advertisement was in evidence for both journals. For the three years studied, the differences between the journals in this regard were miniscule. Table 2 shows that approximately 82 % of all drug advertisements with regard to space were due to three therapeutic categories, i.e. anti-infectives, diureticcardiovasculars and central nervous system (CNS) drugs. To simplify data analysis, only the number of advertising pages were examined. The percentages for the number of advertisements and pages were very similar for all three years. The relative number of pages allocated for the diuretic-cardiovasculars increased from 1972 to 1982, while the relative number of pages for thi: anti-infectives and CNS drugs diminished. In 1972, CNS drugs, which include the anti-depressants, tranquilizers, hypnotics, sedatives and analgesics, accounted for the largest percentage (41%) of advertising pages. In 1982, however, the diureticcardiovasculars accounted for the greatest percentage of advertising pages (41”//,). With the exception of the hormones, the remaining types of drugs accounted for such a relatively low percentage of pages that definitive trends were difficult to ascertain. However, the hormones, primarily antidiabetics, do show a downward trend in the relative number of pages purchased by the pharmaceutical companies. Since the diuretic-cardiovasculars and CNS drugs together comprise over 70% of the total number of pages allocated in 1982, these medications were analyzed in greater detail. Table 3 demonstrates that with regard to the diuretic-cardiovasculars, the cardiac glycosides and peripheral vasodilators were no longer advertised in 1982. Also, much less emphasis with regard to space allocation was given to the diuretics and antihyperlipidemics. On the other hand, a notable increase in the relative amount of advertising for the beta-adrenergic blockers and th& slow channel inhibitors occurred. Table 4 indicates that two types of CNS drugs, analgesics and psychotherapeutics together, accounted for over 90 O/bof the total amount of pages advertised for this therapeutic category in 1982. When the years 1977 and 1982 are compared with that of 1972, a substantial increase in the number of pages advertised is found for the analgesics, while the psychotherapeutics, primarily tranquilizers, showed a 209, decrease. CNS stimulants, anticonvulsants, muscle relaxants and antiparkinsonism Table
2. Space allocation for drug advertisements category and year (in O;)
Ant+infectives Antineoplastics Blood modifiers Central nervous system Diuretic-cardiovasculars Gastromtestinals Hormones Nutrltlonal Respuatory Topic& Miscellaneous Totals ( “,,I .\
by therapeutic
1972 Pages
1977 Pages
1982 Pages
23
13 4
11 1 1 30 41 4 3 2 3 2 2 100 5164
1 41 19 3 9 1 2
34 34 4 6 3 2 -
1
100 .3604
100 3062
193
Table 3. Space allocation for advertised dluremxardlovascular prescription drugs by therapeutic subcategory and year fin :‘,;,J
Antl-angina1 Anti-arrhythmics Antihyperlipidemlcs Antihypertensives Beta-adrenergic blocking agents Cardiac glycosides Duetics Peripheral vasodilators Slow channel Inhibitors Totals ( ;,A) N
1972 Pages
1977 Pages
1982 Pages
14 6 9 23
4 4 10 47
10 5 5 22
7 2 31 8
10
31
24
I2
100 1053
9 100 2120
100 693
agents, although advertised to some extent in 1972, were no longer advertised in 1982. Leading drugs aduertised
An analysis of specific drugs by brand names shows that only one drug, Dalmane@ (flurazepam HCl) was one of the leading 15 drugs advertised for all 3 years. Four drugs, Valium@ (diazepam), Sinequan@ (doxipin HCl), Inderal@ (propranolol HCl) and Tagamet@ (cimetidine) were among the leading 15 drugs advertised for two of the 3 years. Although Valium@ was the most advertised drug in terms of the number of pages for 1972 and 1977, it was not among the leading 15 in 1982. In this year, 29 other drugs had received greater advertising space. As noted in previous tables, we again observe an increase in the amount of advertising space allocated to the diurecticcardiovasculars. In 1982, 9 of the top 15 advertised drugs were of this type, 5 in 1977 and only 2 in 1972. Five of the 9 in 1982 were beta-adrenergic blocking agents. In contrast to the diuretic-cardiovasculars, an apparent downward trend in advertising space for the CNS type drugs was observed, especially when 1977 and 1982 were compared with 1972. In 1982,4 of the leading 15 advertised drugs were of a CNS type, in 1977, there were 3 and in 1972, there were 8. A majority of the CNS drugs in 1972 and 1977 were psychotherapeutics. Only one psychotherapeutic, Sinequan@ (doxepin HCl) was in the top 15 in 1982. Over the years, the leading 15 drugs, combined, accounted for a decreasing proportion of the total number of pages allocated for advertising. In 1972, the top 15 drugs accounted for 45 y0 of the total advertising space, in 1977, 41 y0 and in 1982, 39 % Table 4. Space allocation for advertised central nervous system prescription drugs by therapeutic subcategory and year (m 7;)
Analgesics Anticonvulsants Anti-emetic-antivertigo agents Antiparkinsonism agents Muscle relaxants Psychotherapeutics Sedative-hypnotics Stimulants Totals ( O,)
h-
1972 Pages
1977 Pages
1982 Pages
18 4
42
1 3 2 64 7 1 100 1477
3 1 1 44 9
44 8
100 1062
100 1552
47 1
LAWRENCER. KRUPKA and ARTHUR M. VENER
194
drugs udcertised
Leading
and prescriptions
jilled
pages purchased in 1972 and 1977 and 68”,, in 1981. Five of the companines. Ayerst. CIBA. Hoflmann-LaRoche. Merck. Sharp & Dohme and Pfizer, were among the leading ten advertisers for each ofthe 3 years. In contrast to the leading ten advertisers. the top five, representing about 12 OO.on the average. of the companies which advertised. accounted for almost half (47”,) of the total number of advertising pages purchased. Two manufacturers. Pfizer and Hoffmann-LaRoche were among the top five advertisers for each of the 3 years studied.
A comparison was made between the leading 15 drugs advertised and prescriptions filled for 1972 and 1973; the leading drugs in 1977 with prescriptions filled for 1977 and 1978; and the leading drugs in 1982 with prescriptions filled data for this same year. An apparent association between these variables seems to exist. Approximately one-fifth (21 “,) ofthe leading 15 advertised drugs were also found to be one of the leading 15 drugs with regard to the number of prescriptions filled for the 5 years analyzed. Two-fifths (39 “,) were in the leading 50; three-fifths were in the leading 100 while four-fifths (79 9;) were in the leading 200. A comparison of the leading 200 prescription drugs filled in 1972 and 1973 showed that 10 of the leading 15 drugs advertised in 1972 had advanced their relative ranking with regard to the number of prescriptions filled in 1973. Two of the drugs retained the same relative position. One of these latter drugs was Valium@ (diazepam) which was the mos’t common prescription filled medicine. Only 2 of the leading 15 drugs advertised demonstrated a reduction in rank with regard to the number of prescriptions filled. One drug, Garamycins (gentamicin), one of the 15 most advertised drugs in 1972, was not one of the leading 200 drugs for prescriptions filled either for 1972 or 1973. A comparison of the top 200 prescriptions filled with the 15 leading advertised drugs in 1977 and 1978 showed similar results. Six of the leading 15 advertised drugs in 1977 increased their rank with regard to prescriptions filled in 1978. Two maintained their relative rank and four demonstrated a decrease. Three of the leading 15 drugs advertised in 1977, Apresazide@ (hydrochlorothiazide + hydralazine HCl), Amikin@ (amikacin sulfate) and Megace@ (megestrol acetate) were not one of the leading 200 prescriptions filled for either year. Leading
phurmaceuticul
Sex of‘ patients depicted
adaertisers
An average of 42 different pharmaceutical manufacturers advertised in both journals during 1972, 1977 and 1982. The ten leading advertisers, representing an of 24”; of the companies doing the average advertising, accounted for 66 y;, of the total advertising Table
5. Sex of patients
deplcted
m prescription
drug
Central Ant+mfectives
( “,,)
lY77 Neutral Malr \
and year (III
agents
cardiovasculars
mtestmals
52
61
27
28
loo
categoq
Gastro-
21 I00
443
639
73
59
22
23 IX
( “,,)
by therapeutic Diuretic-
10
Female Totals
advertwments
nervous
71 19
Female TTotals
system
in trduertisrmerzts
Five therapeutic categories, combined. accounted for an average of 92 O0 of the total number of advertisements placed. Table 5 shows the sex of the patients depicted in the advertisements for these five therapeutic categories. On the average, most of the advertisements (69 “i) depict neither male nor female patients in their graphics. Frequently, no human figure. or parts of human figures. were included. When human figures were presented, they were frequently shown in such an amorphous manner that it was not possible to determine which sex they were intended to represent. In their graphics, many advertisements focused their efforts on a display of how effective their products were in the alleviation of medical problems. A trend of increased neutrality of advertisements was discerned. In 1972, an average of 62 “, of the ads were neutral: in 1977, 69% and, in 1982, 77”“. In 1972, an average of 18 y0 of the advertisements showed male patients. in 1977, 18;; and in 1982, 13 “,,. With regard to advertisements showing female patients. the 1972 average was 20 y,;, in 1977, I3 “<,and in 1982,9 OO.When the hormones are excluded, an average of 21 no of the ads depicted male patients and IO “” depicted females. In three types of drugs. the diuretic-cardiovasculars. CNS and gastrointestinals, male patients appear more frequently in advertisements than female patients. Only with hormones do we find a consistently greater number ofadvertisements depicting female patients. In 1972, although more female patients were shown in anti-infective advertisements, for both 1977 and 1982. male oriented advertisements were depicted more frequently for this class ofdrugs. Since special attention
438
57 35 8 100 426
56 3x 100
72
IO0
loo 137
58 9 33
loo
100
178
491
80
65
87
77
I2
24
I2
IX
100
I00
106
216
625
695
II
HOrIIlOIleS
21
II 100
“,,I
I 00
37
100 69
78 21
IO0
7x
100
h7
Prescription
drug advertising:
has been given in the literature to CNS drug use by women. especially with regard to the analgesics and psychotropics. these two subcategories were analyzed separately. Even in this case. we again found a trend towardincreasingsexualneutralityinadvertising. With the exception of the hormones, in only one instance, psychotropic advertisements for 1977, did we observe a clear-cut greater percentage of female oriented advertisements (females 33 “/,. males 14 “,i). DISCUSSION
The observed increase in advertising for the diureticcardiovasculars between 1972 and 1982 is associated with the relatively recent introduction of such effective new compounds as the beta-adrenergic blocking agents and the slow channel inhibitors. The market potential for such drugs continues to be quite large since heart disease is still the leading cause of death among the populations ofdeveloped nations [38]. Since, with age, the heart becomes a poorer pump, the management of older patients with heart related complications represents a substantial part oftheir medical care [39]. The finding that both the CNS and anti-infective drugs showed a decline in advertising emphasis may be associated with the fact that no major therapeutic advances have been made for these two types of therapeutics. Additionally, since there has been an outcry of concern with regard to the deleterious effects of the psychotropics [40]. it is possible that the pharmaceutical companies have decided that their marketing efforts would best be served by emphasizing other products. The fact that most prescription drug advertisements were targeted to neither sex and that this trend has increased. may indicate that the pharmaceutical manufacturers have attempted to defuse accusations of sexual bias in the advertisement of their products. Women, in particular, have been stereotyped in advertisements as needing emotional support through the aid of psychotropics [40-421. Between 1968-1972, Prather and Fidel1 [27] found that 59”i, of the psychotropic drug advertisements showed women while in 1973. McRee et al. [43] found that 64 ‘A of the drugs used in psychiatric advertisements in the Aniericari Jourrral of Psychiatry depicted women. However, in this study. with regard to theanalgesicsand psychotropics. in only one year, 1977, did we find that a greater percentage of the advertisements depicted more female than male patients. Even in this instance, only 33 “(, of the advertisements were female oriented, while the majority (53”,) were neutral. In essence, our findings are supported by Thompson [28], who also found that when ads were targeted to a specific sex. they were oriented more toward males. Only in the case of the hormones. primarily antidiabetics, was a greater percentage of female oriented advertisements found. Even in this case. the majority of advertisements were neutral. Thompson [28] has expressed theconcern that male bias in advertisements might cause physicians to become predisposed to perceive men as needing more medical treatment than women and thus some health problems ofwomen may be overlooked. It is difficult to understand why. in 1977 and 1982. three times the number of analgesic advertisements were male oriented. In thisvein are the pharmaceutical companies
trends
and implications
195
suggesting that men have a lower threshold for pain and therefore need a greater amount and variety of prescribed analgesics? We do not believe this to be the case. Rather, we think that being profit oriented organizations, they simply wish to increase sales of these products. Women have been found to be the prime users of analgesic compounds [41,44-451 and therefore targeting advertisements toward them would probably not dramatically increase sales. Rather, there would be greater potential for increased sales if advertisements were more male oriented. This same motivation may apply to the other therapeutic categories as well. The fact that the average number of pages per advertisement has increased and that a relationship exists between the leading drugs advertised and the leading prescriptions filled implies that prescription drug advertising in professional journals is of financial benefit to pharmaceutical manufacturers. The advertising of Tagamet@ (cimetidine), a histamine H2 antagonist, indicated for the healing of duodenal ulcers, is a case in point. In 1977, a total of 68 pages was allocated for the advertising of this drug. In 1982, the advertising space purchased for this drug was increased by 86 0/0(117 pages). We also found that this product was among the 15 most advertised drugs for 1977 and 1982. There is little doubt that increased advertising of this product was a factor involved in its increased sales. In 1977, Tagamet@ was not among the leading 200 drug prescriptions filled in the United States. In the very next year, it ranked fiftieth. In 1979, it ranked twenty-second, in 1980, fifteenth, in 1981, ninth and in 1982, seventh. During this advertising campaign, not only were the number of advertisements and pages increased, but there was also an increase in the number of identical advertisements in the same weekly issue. For example, the 30 September 1982 issue of the NEJM contained five separate advertisements for Tagamet@ for a total of 12 pages. This saturation type advertising campaign, however, was probably not limited to the two professional journals analyzed. More likely than not, the campaign also included the use of still other medical journals as well as manufacturers’ representatives, the direct mailing of brochures, and information provided at special conferences and professional meetings. Although it is difficult to assess the actual contribution that advertising in the JAMA and the NEJM had on the increased sales of Tagamet@, it is our belief that its effect was not insignificant. Dyazide@ (triamterene and hydrochlorothiazide) is another case in point. In 1972, we found it to be one of the 15 leading advertised drugs. This was not so in 1977 and 1982. Regarding prescriptions filled, it was ranked 44th in 1972 and climbed to be the leading prescribed drug in 1982 in the United States. With little doubt, physicians have found this drug to be very effective in the control of hypertension. As a result, -saturation advertising of this effective medication probably would not have significantly increased its sales and therefore the manufacturer may have decided that it was wiser to invest advertising funds into the marketing of their other products. A question may be raised regarding the marketing of drugs which do not have a clear-cut ameliorative effect on a specific physiological condition. An example of
196
LAWRENCER. KRUPK~ and
such a drug is Valium@ (diazepam), used for the management of anxiety disorders. Anxiety, a feeling of apprehension, uncertainty and fear without apparent stimulus, may or may not be directly linked to definitive physiological changes [46]. Some critics have maintained that the prescribing of Valium@ has implications which go beyond medical practice and is indicative of a mass cultural phenomenon [47]. In fact, Cooperstock and Parnell [40] suggest that Valium@ should be perceived in a manner similar to that of alcohol, i.e. as being a social drug. Considering the ambivalence practitioners may have when prescribing Valium@, as well as the other benzodiazepines, advertising may be an influential factor in their decision to finally prescribe it. In 1972 and again in 1977, Valium@ was given more advertising space than any other prescription drug, but in 1982, it did not appear among the leading 15 advertised drugs. In terms of prescriptions filled, it was the leading drug from 1972 to 1981. Only in 1982, in association with the reduction of advertising, did it decline to a rank of three. It is possible to conclude that, at least for nonspecific therapeutic drugs, greater investment in advertisement campaigns is necessary to maintain high sales levels. In terms of medical ethics, we believe that only effective drugs which meet the requirements of substantial evidence should be advertised. Twenty-six of the drugs advertised in 1972, 1977 and 1982 were reported by Wolfe [3] as being less than effective or ineffective. Because of the United States Food and Drug Agency’s reticence to expedite their removal, these drugs have remained on the market. All were introduced at least 20 or more years ago. The problem here is that the drugs which are advertised may not be the ones which are the most effective. If it is accepted that prescription drug advertising in medical journals does have an impact on prescribing behavior, then the extensiveness of advertising by any one or combination of companies represents another potential ethical problem. The finding that the top ten advertising companies, on the average, accounted for space for the three years 67’;b of all advertising examined, and that five accounted for almost half (47’:;)) of advertising space, portends the hazards typically associated with monopolistic business practices. This problem is further exacerbated by the fact that the same two pharmaceutical companies were among the leading five advertisers for the 3 years analyzed, and the same five were among the leading top ten. Since the cost of advertising in the major medical journals represents a considerable investment of funds, smaller companies may have some difficulty in obtaining exposure for their products. Again, the possibility exists that less effective medications will be advertised. A resolution to this problem is suggested by a 1983 law enacted by the United States Congress regarding ‘orphan drugs’-pharmaceutical products that have been introduced in other countries, but not made available in the United States. These drugs are frequently indicated for relatively rare diseases and therefore are not profitable. The law provides a federal subsidy for pharmaceutical companies, private investigators and non-profit organizations to help in the development and marketing of such drugs[48]. Perhaps federal funds can also be made available for
ARTHUR M. VENER
the advertising campaigns of effective drugs which have been developed by individual investigators or small pharmaceutical companies who might be unable to amass the large amounts of funding which would be required to launch a successful advertising campaign in order to gain sufficient exposure for their products.
REFERETTCES
Anonymous. The top 200 drugs. Pharm. Times April 1972 to 1983 issues. Boyd
J. R. (Ed.)
Drug
Facts
und Comparisons.
Lippincott, St Louis, MO, 1983. Wolfe S. M. Pills that Don’t Work. Warner
6. 7
8.
9.
10. 11.
12.
13. 14.
15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27.
J. B.
Books, New York, 1982. Stolley P. D. and Lasagna L. Prescribing patterns of physicians. J. chron. Dis. 22, 395, 1969. Norwood G. J. and Smith M. C. Market mortality of new products in the pharmaceutical industry. J. .Im. pharm. Ass. 11, 592, 1971. Hemminki E. Review of literature on the factors affecting drug prescribing. Sot. Sci. Med. 7, 51, 1973. Herman C. M. and Rodowskas C. A. Communicating drug information to physicians. J. med. Educ. 51, 189, 1976. Becker M. H., Stolley P. D., Lasagna L., McEvilla J. D. and Sloan L. M. Differential education concerning therapeutics and resultant physician prescribmg patterns. J. med. Educ. 47, 118, 1972. Linn L. S. and Davis M. S. Physicians’orientation toward thelegitimacyofdruguseand theirpreferredsourceofnew drug information. Sot. Sci. Med. 6, 199. 1972. Worthen D. B. Prescribing influences: an overview. Br. J. med. Educ. 7, 109, 1973. Miller R. R. Prescribing habits of physicians: a review of studies prescribing drugs. Drug Intel/. clin. Phnrm. 8. 81, 1974. Eaton G. and Parish P. Sources of drug information used by general practitioners. R. Coil. gen. &act. 26, Suppl. I, 58, 1976. Worthen D. B. Clinical drug information: profit or knowledge oriented. Ethics Sci. Med. 3, 235, 1976. Najman J. M., Siskind V. and Bain C. Prescription drug advertising: medical journal practices under different types of control. Med. J. Aust. 1, 420, 1979. Ditzion B. R. Psychotropic drug advertisements. Ann. intern. Med. 75, 473, 1971. Seidenberg R. Advertising and abuse of drugs. N. Engl. .I. Med. 284, 789, 1971. Kimmey J. R. Drug efficacy andajournal’s responsibility. Am. .I. puhl. Hlth 62, 461. 1972. Simpsin F. 0. Omissions in advertising. N. Z. med. J. 77, 412, 1973. Stimson G. V. Information contained in drug advertisements. Br. med. J. 4, 508. 1975. Holmes J. D. Prescribing information in drug advertisements. N. Z. med. J. 86, 467, 1977. Lefferts G. Journal advertising. Ann. inter,z. Med. 88, 581. 1978. Relman A. S. Who reviews the ads’? N. Engl. J. .&fed. 301, 999. 1979. Silverman M. The Drugging of the Ameruns. Universtty of California Press, Berkeley, 1976. Yudkin J. S. Provision of medicines in a developing country. Luncet I, 801, 1978. Victora C. G. Statistical malpracttce in drug promotton: a case-study from Brazil. Sot. Sci. .Lfed. 16, 707, 1982. Seidenberg R. Drug advertising and perception of mental illness. M&t. Hvgr55, 21, 1971. Prather J. and Fidel1 L. S. Sex differences in the content and style of medical advertisements. Sot. ScI. Med. 9, 23. 1975.
Prescription drug advertising: trends and implications 2X. Thompson E. L. Sexual bias in drug advertisements. Sot. Sci. Med. 13A, 187, 1979. 29. Smith M. C. Portrayal of the elderly in prescription drug advertising. Gerontologist 16, 329, 1976. 30. Robinson R. G. Medical journals and advertising. N. Z. med. J. 87, 139. 1978. 31. Woody E. The paradox of pharmaceutical advertising. .I. med. Ass. Ga. 69, 379. 1980. 32. L’lrich’s Internarional Periodicals Directory, 22nd Edition. R. R. Bowker, New York, 1983. 33. Blackwell B. Patient compliance. N. Engl. J. Med. 289, 249. 1973. 34. Haynes R. B. Sackett D. L., Taylor W., Roberts R. S., Tech M. and Johnson A. L. Manipulation of the therapeutic regimen to improve compliance: conceptions and misconceptions. C/in. Pharmac. Ther. 22, 125, 1977. 35. Boyd J. R., Covington T. R., Stanaszek W. F. and Coussons R. T. Drug defaulting. Am. J. Hosp. Pharm. 31, 485. 1974. 36. Whitney H. A. Educating the public about drugs. Drug Intell. c/in. Pharm. 9, 571. 1975. 37. Caldwell J. R., Cobb S., Dowling M. D. and De Jongh D. The dropout problem in antihypertensive treatment. J. chror~. Dis. 22, 579, 1970. 38. Pelletier K. R. Holistic Medicine. Delacorte Press, New York. 1979. 39. Harris R. Special problems of geriatric patients with
40. 41. 42.
43.
44. 45.
46. 47.
48.
197
heart disease. In Clinical Aspects of Aging (Edited by Reichel W.), p. 45. Williams & Wilkins, Baltimore, MD, 1978. Cooperstock R. and Parnell P. Research on psychotropic drug use. Sot. Sci. Med. 16, 1179, 1982.. Prather J. E. and Fidel! L. S. Drug use and abuse among women: an overview. Int. J. Addict. 13, 873, 1978. Linn L. S. Physician characteristics and attitudes toward legitimate use of psychotherapeutic drugs. J. Hlth WC. Behau. 12, 132, 1971. McRee C., Corder B. F. and Haizlip T. Psychiatrists’ responses to sexual bias in pharmaceutical advertising. Am. J. Psychiat. 131, 1273, 1974. Krupka L. R., Vener A. M., Stewart C. S. and ZaengleinSenger M. M. Patterns of aspirin use among American youth. Int. J. Addict. 13, 911, 1978. Vener A. M., Krupka L. R. and Climo J. J. Drugs (prescription, over-the-counter, social) and tile young adult: use and attitudes. Jnt. J. Addict. 17, 399, 1982. Friel J. P. (Ed.) Dorland’s Illusrrated Medical Dictionary, 25th Edition. W. B. Saunders, Philadelphia, 1974. Bellantvono C., Reggi V., Tognoni G. and Garattini S. Benzodiazepines: clinical pharmacology and therapeutic use. Drugs 19, 195, 1980. Anderson K. Orphan Drugs. Linden Press, New York, 1983.