A Threat to Physician Education
Peter Lurie, MD, MPH
Sidney M. Wolfe, MD
Background: Commercially-oriented Medical Education Services Suppliers (MESSs) are predominantly for-profit companies that provide promotional and nonpromotional/educational services to the medical community on behalf of their clients, a large proportion of whom are pharmaceutical companies. Our objective was to better describe and understand the role that MESSs play in physician education today. Methods: We analyzed the results of a survey, conducted and published by the periodical Medical Marketing & Media, which included general information and the 1998 and 1999 financial information for 80 of the 123 contacted MESSs. Results: The MESS industry's revenue grew 19% between 1998 and 1999, totaling $643 million in U.S. revenue in 1999 for the 42 respondents (53%) who reported financial data. MESS's averaged $15 million in 1999 U.S. revenue (range: $0.35-98 million). A total of $289 million, or 45% of the total reported revenue, was earned through providing grand rounds, symposia, or publications-related activities, with $115 million earned through providing grand rounds-related services. On average, 76% of each MESS's clients were pharmaceutical manufacturers listed in the Physicians Desk Reference. Conclusions: The results suggest that MESS's work on behalf of the pharmaceutical industry to provide at least a billion dollars in physician "education" activities annually. A for-profit, pharmaceutical industry-oriented company is a poor candidate for organizing physician education as there is a clear conflict between the needs of MESSs' clients and stockholders and the need for unbiased information in physician education.
Key Words: physician education; continuing medical education; conflict of interest; pharmaceutical industry; medical education services suppliers.
Commercially-oriented Medical Education Services Suppliers (MESSs) are assuredly the least recognized providers of physician education. However, these for-profit companies are growing rapidly in both prominence and influence, disseminating ostensibly objective medical information on behalf of their clients, which include medical organizations and pharmaceutical manufacturers. In the past, the drug industry more often worked directly with physician member organizations and hospitals to provide educational activities for physicians. Now, MESSs are increasingly acting as intermediaries between the two groups, organizing the "educational" activities, including identifying physicians and other speakers to present the information in forums as diverse as continuing medical education (CME) programming, grand rounds presentations, and clinical symposia. Other services provided by MESSs include strategic publication planning and telecommunication conferences.
To better describe and understand the role that MESSs play today in physician education, we analyzed a survey conducted by the periodical Medical Marketing & Media (MM&M).(1)
MM&M is a monthly business publication owned by CPS Communications, Inc. It circulates to over 12,000 healthcare executives, advertising agencies, and medical marketers, including pharmaceutical manufacturers. Each issue carries a mix of industry news and feature articles on medical marketing and promotion.(2) The results of a survey MM&M mailed to MESSs were published in its December 1999 issue. The survey was intended to furnish readers with a reference source for companies offering medical education services.1
One hundred and twenty-three companies known to provide medical education services were contacted by MM&M. Their survey asked for general information about each company, 1998 and 1999 financial data, and a profile of each company, including clients and specialized services provided. Each MESS also reported the percentage of its 1999 billings derived from 13 specific services listed on the questionnaire: symposia, editorials, professional meetings, publications, videos, audiotapes, interactive computer programs, advisory boards, grand rounds, faculty meetings, investigator meetings, internet web sites, and other. MM&M published 80 (65%) detailed responses to their survey as a list alphabetized by company. A sample response is shown in Appendix A.
All information published in MM&M was entered into a Microsoft Access 97 database. A 1999 U.S. billings revenue for each service provided was calculated as the product of each company's reported 1999 total U.S. billing revenue and each service's reported percentage of the company's U.S. billings. The database was transferred into Stata Statistical Software: Release 6.0 for analysis. Descriptive data are presented as means or medians with ranges. A Spearman's rank correlation coefficient was calculated to assess the relationship between 1999 total U.S. billing revenue and the number of services provided by a MESS. We calculated relative risks to assess the relationship between a company's accreditation status by the Accreditation Council for Continuing Medical Education (ACCME) and the provision of each of the 13 different types of services. The Wilcoxon rank-sum test was used to assess the relationship between 1999 total U.S. billing revenue and ACCME accreditation status and between 1999 total U.S. billing revenue and the provision of the specific services. A p-value < 0.05 (2-sided) was considered statistically significant.
Respondents reported an average of 39 employees (range: 2-321) (Table I). The median year the companies were founded was 1988 (range: 1781-1999), with 17 of the 77 respondents (22%) being founded in 1995 or later. Sixty-six respondents (83%) reported international capabilities.
Sixty-eight respondents (85%) listed a total of 235 different companies or organizations as clients. Ninety-one of those clients (39%) are pharmaceutical manufacturers indexed in the Physicians' Desk Reference (PDR).(3) The average number of clients per MESS was 11 (range: 2-44). However, because many pharmaceutical manufacturers were listed as clients by multiple respondents, an average of 76% of a respondent's clients were pharmaceutical manufacturers (range: 0-100%). Only four respondents (6%) reported a client list of which fewer than 50% were pharmaceutical manufacturers, and only one did not report a pharmaceutical manufacturer as a client. In contrast, 18 respondents (26%) reported a client list of which 90% or more were pharmaceutical manufacturers, including 12 (18% of respondents) which reported that all of their clients were pharmaceutical manufacturers. Twenty pharmaceutical manufacturers were listed as clients by at least 10 (15%) of the 68 respondents. Novartis Pharmaceutical Corporation was listed as a client by 29 respondents (43%), Bristol-Meyers Squibb Company by 24 (35%), Pfizer, Inc. by 23 (34%), Wyeth-Ayerst Laboratories by 23 (34%), and Roche Pharmaceuticals by 21 (31%). MM&M did not publish the MESSs' percentage of revenue derived from each client.
Six of 80 respondents (7.5%) did not provide a breakdown of their 1999 U.S. billings by services provided. Of the remaining 74 respondents, 23 (31%) billed for four or fewer different services, 28 (38%) billed for between five and nine different services, and 23 (31%) billed for ten to thirteen different services (Figure I). The mean number of services provided was 7 (range: 1-13). However, 26 respondents (35%) billed at least 50% of their revenue by providing a single type of service, 16 (22% of the 74 respondents) of which specialized either in providing grand rounds, symposia, or publications-related services.
The most common service provided by the 74 respondents who provided a breakdown of their 1999 U.S. billings was symposia (81%), followed by publications-related services (72%), and advisory boards-related services (68%) (Table II). The third column of Table II in effect displays a profile for how the billings of a "typical" MESS would be organized. Publications-related services average 17% of domestic billings (range: 0-90%), while symposia average 16% (range: 0-80%). Due to specialization within the industry, few MESSs reported a billings by service which resembled this profile.
Only 44 of 80 respondents (55%) provided 1998 U.S. billing data, reporting an average revenue of $13 million (range: $0.25-59 million). For the remainder of this report, we focus on 1999 U.S. billing data, which only 42 of 80 respondents (53%) provided. The average 1999 billing revenue was $15 million (range: $0.35-98 million), for a total of $643 million. The average U.S. billing revenue of respondents thus increased 19% between 1998 and 1999. Six respondents billed for $40 million or more in 1999 (Figure II), accounting for $357 million, 56% of the total 1999 U.S. billing revenue.
Grand rounds-related services provided the greatest amount of 1999 U.S. revenue, $115 million, for an average of $6 million (range: $0.14-49 million) among the 18 respondents who provided this service and reported 1999 U.S. financial data (Table III). Other large sources of billing revenue were symposia ($114 million), advisory boards ($64 million), and publications ($60 million). The five respondents with the largest grand rounds-related revenue accounted for 85% ($98 million) of the total grand rounds revenue (Table IV). The five respondents with the largest symposia-related revenue accounted for 59% ($67 million) of the total symposia revenue. The five respondents with the largest publications-related revenue accounted for 65% ($39 million) of the total publications revenue.
Thirty-four of 80 respondents (43%) reported being accredited by the ACCME, 4 (5%) reported that their accreditation was pending, and the remaining 42 respondents (53%) reported no ACCME accreditation. At least 4 of the top 5 respondents, whether ranked by 1999 U.S. revenue derived from grand rounds-related services, symposia-related services, publications-related services, or total 1999 U.S. billings revenue, reported ACCME accreditation or that their accreditation was pending.
There were no statistically significant associations between ACCME accreditation/ pending accreditation and the provision of any of the 13 different types of services. However, those respondents who reported ACCME accreditation/pending accreditation had statistically significantly higher 1999 U.S. billings revenue than those without accreditation ($23 million vs. $7.4 million, p=0.0035). In addition, 1999 U.S. billing revenue was statistically significantly higher for respondents who provided, compared to those who did not, grand rounds-related services ($29 million vs. $5.2 million, p=0.0001) and symposia-related services ($18 million vs. $2.3 million, p=0.0043). Respondents who provided advisory board-related services, faculty meeting-related services, investigator meeting-related services, and internet web site-related services were also statistically more likely to be larger revenue companies compared to those who did not. The remaining services were not associated with billing revenue. 1999 U.S. billing revenue and the number of services provided by a MESS were strongly correlated (r=0.66).
*Year founded is presented as the median with range.
*Forty-one of 80 respondents (51%) provided both a complete breakdown by percentage of their U.S. billings and their 1999 U.S. billing revenue.
**The total 1999 U.S. billing revenue for all services provided equals $634 million, a smaller tally than the previously reported $643 million, because it includes only 41 respondents, not all 42 who reported 1999 billing revenue (see above).
Our analysis of MM&M's "1999 Review of Medical Education Services Suppliers" reveals the tremendous size and recent growth of this little-known industry. Expenditures on medical education services appear to have grown 19% between 1998 and 1999, totaling $643 million in revenue in 1999. This is certainly an underestimate of the total revenue of the MESS industry. Although MM&M mailed its questionnaire to 123 companies known to supply medical education services, there may be additional MESSs who were not known to MM&M. In addition, 43 MESSs who were mailed a questionnaire by MM&M did not reply and 38 of those who did reply did not provide their 1999 financial data.
There is some reason to believe that smaller companies would be more likely to respond to the MM&M survey, as it would increase their name recognition among potential clients. However, very small MESSs were probably less likely to be known, and thus contacted, by MM&M. On the other hand, it is also possible that the most financially successful MESSs were more likely to report their financial data, a reporting bias that would reduce the underestimate of total revenue. However, both MESSs who did and did not report financial data reported providing an average of seven different services. As our study demonstrated, a strong correlation exists between billing revenue and the number of services provided by each MESS. Thus, it is unlikely that MESS respondents who did not report financial data differ greatly in terms of revenues from those who did report. We therefore feel comfortable concluding that this is a growing industry with a 1999 revenue probably well over $1 billion.
A total of $289 million, or 45% of the total reported revenue, was earned through providing three direct education-related services: grand rounds, symposia, and publications, with $115 million earned through providing grand rounds-related services alone. In terms of revenue, each of these services was dominated by three or four MESSs, all of which reported accreditation by the ACCME or that their accreditation was pending. In addition, 16 respondents specialized in one of these three direct education-related activities, meaning that they earned 50% or more of their revenue by providing either grand rounds, symposia, or publications-related activities.
Our survey demonstrates that 76% of each respondent's clients were pharmaceutical manufacturers listed in the PDR. However, this is probably an underestimate. Of the 144 clients who did not appear in the PDR, at least 8 companies had names that strongly imply they manufacture pharmaceuticals. For instance, Otsuka Pharmaceutical was listed as a client by 4 respondents, while Ligand Pharmaceutical, Inc. was listed by 3 respondents. The absence of these clients from the PDR index means either that they currently have no FDA-approved products or that they have chosen not to the pay the fee required to appear in the PDR.
The dominance of the pharmaceutical industry as the most common type of client and main revenue source of MESSs would be of less consequence if there was evidence that pharmaceutical companies could be trusted to provide unbiased information to the medical community. But the evidence is to the contrary. Bowman demonstrated 14 years ago that drug industry support was associated with an increased number of positive statements by CME speakers about the sponsoring company's drug(4) and with increased prescribing of the drug by attendees after the conference.(5) Similarly, Spingarn, et al. demonstrated an association between housestaff attendance at grand rounds presentations given by a physician who was an employee of a pharmaceutical manufacturer and the subsequent choice of the pharmaceutical manufacturer's drug, both appropriately and inappropriately.(6) In addition, Bero, et al. showed that symposia sponsored by the pharmaceutical industry often have promotional attributes, such as misleading titles and a focus on brand names rather generics, and are frequently not peer-reviewed.(7) Clearly, the drug industry is successful at promoting its products, at times under the guise of unbiased physician education.
To permit for-profit companies to provide physician education is to sanction a clear conflict of interest. Why would one expect a MESS with obligations to its clients, or a pharmaceutical company with obligations to its stockholders, to provide unbiased information? A simple search of several MESS's company web sites confirms that some companies who may present themselves to physicians as providers of unbiased education market themselves to potential clients precisely on the basis of their ability to be biased. For example, Concepts in Professional Education and Communications, a MESS whose client list includes 14 different pharmaceutical manufacturers, states unabashedly:
Promeda, a large MESS that reported $45 million in revenue in 1999, advertises simply, "Putting the science of medicine to work for you. Preparing and building the market through professional education ..."(10) The largest MESS in our survey, Phase V Communications, with clients such as Pfizer and Eli Lilly and which earned nearly $49 million for providing grand rounds-related activities and $10 million for publications-related activities in 1999, explains:
Unfortunately, most physicians do not realize the extent of the involvement of the for-profit industry in physician education. In 1999, ACCME-accredited providers earned in excess of $1.11 billion for CME activities, $388 million (35%) of which was income directly from commercial supporters, which are defined as entities providing funds or resources to a CME sponsor. This is a 28% increase in income from commercial sources between 1998 and 1999.13 Moreover, ACCME-accredited communications, education, and publishing companies (the categories into which MESSs can be classified) earned in excess of $224 million for CME activities in 1999, $150 million (67%) of which was income directly from commercial supporters, a 107% increase in income from commercial sources between 1998 and 1999.(13) However, even beyond ACCME commercial support, the pharmaceutical industry enjoys significant influence, as its money and persuasion continues to permeate virtually all levels of physician education, in the forms of complimentary meals and entertainment, "consultation" fees, and pseudo-CME courses.(14)
In order to minimize this influence, the ACCME,(15) the American Medical Association,(16) the Food and Drug Administration,(17),(18) and several other professional medical organizations have established standards or guidelines to define the appropriate role of the pharmaceutical industry in physician education. The common features of nearly all of the guidelines are that CME sponsors, not the funding source, should be responsible for the content and quality of the CME program, that the funds should be given in the form of an unrestricted educational grant, and that the financial arrangements and any possible conflict of interest should be disclosed.(19)
While the guidelines stress the importance of objectivity and reducing the influence of industry, in practice, MESSs are hired to promote the interests of particular companies. The very purpose for which a pharmaceutical company hires a MESS is at odds with the intent of these standards and guidelines which call for CME sponsors, including ACCME-accredited MESSs, to be responsible for the content and quality of physician education. MESSs may also undermine the disclosure aspects of the standards and guidelines. While many physicians understand the potentially biased nature of pharmaceutical company promotions, few are likely to recognize a MESS, and the information it presents, to be potentially biased. In addition, the ACCME's rules do not address the issue of whether the identity and funds of one or both companies (the pharmaceutical company and the MESS) must be disclosed in print at an educational activity. For example, if a drug company hires a MESS, which then hires targeted physicians to deliver a presentation, will those physicians know that their funding, in essence, comes from a pharmaceutical manufacturer, and disclose this information to their audience?
MESSs present other problems as well. First, the information they disseminate through physician education may be skewed toward diseases and products from which the pharmaceutical industry can profit. Presentations on new, patented drugs will replace presentations on older, generic drugs, while general topics in medicine will tend to be replaced by discussions of profitable drug treatments or diseases for which pharmaceutical manufacturers have developed potential blockbuster products, such as erectile dysfunction. Second, MESSs are hired on the basis of their ability to provide educational programs with greater efficiency compared to hospitals or physician-member organizations, since that is their explicit business. This will only enhance the influence of the pharmaceutical industry and dilute the unbiased, uninfluenced educational programs that might be provided by more objective sources. Last, the educational activities MESSs provide also affect resident physicians in graduate medical training programs. Whenever a resident attends a MESS-sponsored grand rounds lecture or symposium, or receives a MESS-distributed publication, the quality of graduate medical education diminishes.
Disclosure of funding sources is often put forward as the solution to potential conflicts of interest. Unfortunately, disclosure is difficult to enforce and does not necessarily eliminate the industry's influence on doctors.(20) A more practical solution is for hospitals to reject presentations provided by MESSs or the pharmaceutical industry. Hospitals must prioritize physician education and provide sufficient funds to alleviate any reliance on drug industry support. Although reductions in Medicare subsidies for medical education place a strain on hospital budgets, payments for physician education are a small fraction of these reductions and a still smaller fraction of overall hospital budgets.
In turn, the Accreditation Council for Graduate Medical Education (ACGME), the official--and only--organization reviewing and accrediting the over 7,700 U.S. graduate medical training programs, must preserve objective, evidence-based residency training by making prohibitions on MESS and pharmaceutical company "educational" activities for residents as a criterion for accreditation. The quality of graduate medical education must be assured. Both resident and physician education are too important to be provided by a biased source such as a pharmaceutical manufacturer or a MESS acting on its behalf.
1. Anonymous. Medical education services suppliers, 1999 review. Medical Marketing & Media 1999; 34:40-80.
2. CPS Communications, Inc. Medical Marketing & Media. Available at: http://www.cpsnet.com/Pubs/mmm.asp. Downloaded on July, 19 2000.
3. Physicians' Desk Reference, 53rd edition. Montvale, NJ: Medical Economics Company, Inc. 1999.
4. Bowman MA. The impact of drug company funding on the content of continuing medical education. Mobius 1986; 6:66-69.
5. Bowman MA, Pearle DL. Changes in drug prescribing patterns related to commercial company funding of continuing medical education. J Contin Educ Health Prof 1988; 8:13-20.
6. Spingarn RW, Berlin JA, Strom BL. When pharmaceutical manufacturers' employees present grand rounds, what do residents remember? Acad Med 1996; 71:86-88.
7. Bero LA, Galbraith A, Rennie D. The publication of sponsored symposiums in medical journals. NEJM 1992; 327:1135-1140.
9. Concepts in Professional Education and Communications. Why should you invest in medical education? and Influencing those who matter. Available at: http://www.cpeducate.com/pages/CPEintroCont.html . Downloaded on July 19, 2000.
11. Grey Healthcare Group. Pathways to success: medical education: Phase V Communications. Available at: http://www.ghgroup.com/pathway/content.asp?A=2&B=3&pg=11 . Downloaded on July 19, 2000.
12. Grey Healthcare Group. Pathways to success: medical education: Phase V Communications: speaker's bureau. Available at: http://www.ghgroup.com/pathway/content.asp?A=2&B=3&C=1 &pg=12 . Downloaded on July 19, 2000.
13. Accreditation Council for Continuing Medical Education. Income and expense by organization type - 1999. Available at: http://www.accme.org/incoming/ 98_1999_Annual_Data_page3.PDF. Downloaded on July 19, 2000.
14. Tenery RM. Interactions between physicians and the health care technology industry. JAMA 2000; 283:391-393.
15. Accreditation Council for Continuing Medical Education. The ACCME's essential areas and their elements: standards for commercial support of continuing medical education. Available at: http://www.accme.org/incoming/17_system98_essential_areas.pdf . Downloaded on July 19, 2000.
16. Council on Ethical and Judicial Affairs, American Medical Association. Chapter 6: AMA ethical opinions on CME: AMA ethical opinion 9.011: continuing medical education; AMA ethical opinion 8.061: gifts to physicians from industry. Available at: http://www.ama-assn.org/cmeselec/cmeres/cme-6.htm . Downloaded on July 19, 2000.
17. Kessler DA. Drug promotion and scientific exchange: the role of the clinical investigator. NEJM 1991; 325:201-203.
18. Department of Health and Human Services: Food and Drug Administration. Final guidance on industry-supported scientific and educational activities; notice. 62 Federal Register 64073-64100, December 3, 1997.
19. Rosner F. Pharmaceutical industry support for continuing medical education programs: a review of current ethical guidelines. Mt. Sinai J Med 1995; 62:427-430.
20. Bero LA. Accepting commercial sponsorship: disclosure helps--but is not a panacea. BMJ 1999; 319:653-654.