Industry begins to establish relationships with physicians during training, as they develop their practice habits. Almost all third-year medical students (97%) have eaten food provided by drug companies, and 94% have accepted a gift.1 These relationships with industry affect residents' prescribing; for example, pharmaceutical company sponsorship of psychiatry resident conferences in one academic medical center correlated with patients' use of corresponding antidepressants in the outpatient clinic.2
The provision of drug samples is an effective marketing tool.3 Numerous studies have shown that access to samples influences prescribing choices.4–13 (See Groves et al14 for a comprehensive review of studies on samples.) The only intervention that studies have shown to decrease prescribing of advertised drugs is abolishing the sample closet,4,13,14 and even that is not consistently effective.4,6
In 2002, the Accreditation Council for Graduate Medical Education (ACGME) issued a guideline on industry-GME relationships, recognizing “the inherent conflict of values between industry and the medical profession” and requiring training programs to educate residents about interacting with pharmaceutical companies.15 In 2006, a group of leading educators issued a call to academic medical centers to curtail interactions with industry.16 Recently, the Association of American Medical Colleges, the Institute of Medicine, and the Josiah Macy Jr. Foundation recommended restricting relationships between physicians and industry.17–19
Literature is sparse regarding the proportion of residency programs that allow trainees to interact with industry representatives, but some recent studies shed light on practices in specific disciplines. A survey of internal medicine residencies found that 55.9% of responding programs accepted support from pharmaceutical companies.20 A review of national surveys of residency programs found that 89% to 93% of internal medicine and emergency medicine residencies allowed some interaction with industry21; the types and extent of interactions were not described. A survey of surgery residencies demonstrated that 65% had drug lunches at least occasionally.22 In another survey, 87% of ophthalmology residents reported meeting with pharmaceutical representatives; more than 90% had accepted gifts, most commonly dinners, pens, books, or lunches.23 The most recent national survey on family medicine residencies' relationships with drug companies seems to be a 1992 study in which more than 90% of programs allowed gifts and sample distribution.24
This study grew out of a joint effort of PharmedOut—a Georgetown University Medical Center project that advances evidence-based prescribing and educates health care professionals about pharmaceutical marketing practices —and the PharmFree project of the American Medical Student Association (AMSA). We noticed that family medicine residencies dominated a list that the two groups compiled of residency programs that disallow pharmaceutical industry interactions. We decided, therefore, to conduct a nationwide survey of family medicine residencies to determine the extent and type of industry interactions with trainees and to identify “pharma-free” residencies that avoided industry influence.
We identified the 460 accredited U.S. family medicine residency programs using the Web sites of the ACGME and the American Academy of Family Physicians (AAFP) in May 2008. We categorized residency programs by type, according to categories described by the AAFP: community-based, unaffiliated; community-based, medical-school-affiliated; community-based, medical-school-administered; medical-school-based; and military.25 We obtained program contact information from the ACGME, the AAFP, and the Web sites of individual programs.
We developed a brief survey that invited family medicine residency program directors/coordinators to answer four yes/no questions:
1. Does the residency allow gifts from industry or industry-supported food?
2. Are drug samples accepted?
3. Are industry representatives allowed access to medical students and/or residents at the family medicine center?
4. Are any industry-sponsored residency activities allowed?
We also provided space for additional comments, following the prompt: “Any additional comments or clarifications?”
We decided not to define the terms activities and access in the hope that participants would interpret them as broadly and inclusively as possible. We used the general term industry rather than specifying pharmaceutical industry so that the language would not exclude interactions with purveyors of medical devices, biologics, and dermatologic products. In correspondence accompanying the survey, we informed participants that program names would be identified but that their individual identities would be kept anonymous. We offered to send the results of the survey to all interested participants.
We made up to six attempts to contact each residency program during June to September 2008. We first e-mailed the survey to each program's residency director or residency coordinator. If a program did not respond to the initial e-mail within a week, we resent the survey and a reminder via e-mail and U.S. mail (on AMSA letterhead, accompanied by a stamped, self-addressed envelope). Three weeks later, we sent another e-mail that included a reminder and the survey. Finally, we attempted to contact the nonresponding program by telephone up to three times. If we reached a relevant party, we offered to complete the survey by telephone and transcribed responses verbatim. We offered respondents interviewed by telephone the opportunity to review transcribed copies of their responses to ensure accuracy.
The Georgetown University institutional review board exempted the survey from review.
We tabulated responses to the four questions and performed a qualitative analysis of the written and verbal additional comments using a template-organizing style.26 One author (R.L.T.) coded the comments to identify themes. Two additional authors (A.F.B., S.R.B.) reviewed the comments and coding. Themes and the analyses of themes were established by consensus.
We designated as “pharma-free” family medicine residencies that answered “No” to all four questions; these programs allowed no industry gifts, food, samples, interaction with students or residents, or sponsorship of residency activities. We classified residencies that allowed any interaction between industry representatives and trainees, even if within the context of an educational intervention on industry interactions, as permitting industry access to trainees. For example, if a respondent stated “No” to the question, “Are industry representatives allowed access to medical students and/or residents at the family medicine center?” but commented that representatives only interacted with trainees as part of an educational exercise, we coded that response as a “yes.” We coded as “unclear” responses left blank and some others (e.g., “Not officially,” “Not typically,” “No food; limited to no gifts”). In our quantitative analysis, we coded unclear responses as permitting industry interaction. The difference between community-based unaffiliated or medical-school-affiliated programs and medical-school-based or -administered programs was assessed by chi-square test.
Of the 460 family medicine programs contacted, 286 (62.2%) responded. We received two responses from 11 institutions; in five of those cases, the duplicate surveys contained conflicting answers. We recorded only one answer per question per program. When conflicting answers were received, the least restrictive answer, usually “yes,” was coded after author consensus was reached.
Responding programs were located across 48 states and the District of Columbia; we did not receive responses from any programs in Rhode Island or Delaware. We received 22.7% (65) of the responses via e-mail, 52.4% (150) by mail, 25.2% (72) by telephone, and 3.5% (10) by fax (numbers do not add up to 100% because 11 institutions responded twice). The response rates were similar across program types: 62.2% (186/299) of community-based unaffiliated and medical-school-affiliated programs, 62.3% (91/146) of medical-school-based and medical-school-administered residencies, and 60% (9/15) of military programs responded.
About half of the 286 responding residency programs refused free drug samples (148, or 52.1%) or disallowed industry gifts or industry-supported food (137, or 48.6%). More than two-thirds of programs (193, or 68.5%) forbade industry-sponsored residency activities, whereas fewer than half (124, or 44.1%) denied industry representatives access to trainees at the family medicine center (see Figure 1).
The pharma-free criteria were met by 26.2% (75) of the 286 programs; these were spread across 21 states (see Supplemental Digital Table 1, http://links.lww.com/ACADMED/A48). Among community-based unaffiliated and medical-school-affiliated residencies, 25.8% (48/186) were pharma-free. Similarly, 28.6% (26/91) of medical-school-based and medical-school-administered programs were pharma-free. The difference between program types was not significant (P > .05) by chi-square test. Among military programs, 11.1% (1/9) were pharma-free.
Among the 211 responding programs that permitted industry interaction, 145 (68.7%) permitted gifts or food, 129 (61.1%) accepted drug samples, 150 (71.1%) allowed industry representatives access to students and residents in the family medicine center, and 80 (37.9%) allowed industry-sponsored residency activities.
Our qualitative analysis of respondents' additional written and verbal comments (Table 1) revealed four main themes:
* Trainees can and should be taught how to manage interactions with industry representatives.
* There are challenges inherent in attempts to limit industry interactions.
* Residents often participate in sponsored events after hours.
* Programs recently changed plans or practices regarding interaction with industry representatives.
Our results show that most family medicine residencies allow some interaction with industry. However, more than two-thirds of the programs limit industry relationships by forbidding industry access to trainees, and more than one-quarter are pharma-free—that is, they allow no industry gifts, food, interaction with medical students or residents, samples, or sponsorship of residency activities. This constitutes a major change from 1992, when 90% of family medicine residencies allowed industry support.24 Our data complement Loertscher and colleagues'20 recent finding that 55.6% of internal medicine residencies reported industry support in 2007 compared with 88.6% in 1990. Together, these results suggest a possible trend toward limiting industry interaction with trainees.
In our study, residency programs based at medical schools were no more likely than community-based residencies to be free of pharmaceutical influence. Although Yale, Stanford, and other prominent academic medical centers have received national media attention for limiting industry gifts, lesser-known academic medical centers and community-based residencies have also quietly severed industry relationships. Our survey shows that community-based residencies are in the forefront of the movement to restrict industry relationships, which is important because the majority of family medicine residencies (88%) are not located in large university-medical school complexes.25
Many respondents noted that their institution's policies concerning industry relationships had changed recently or were in the process of changing. This may reflect heightened federal and state scrutiny of financial arrangements between physicians and industry. Faced with the threat of increased regulation, industry has begun to voluntarily set limits on some promotional activities; for example, the Pharmaceutical Research and Manufacturers of America issued guidelines that took effect in January 2009 barring the distribution of pens, mugs, and other “noneducational” gifts to physicians.27
This study has several limitations. Although a majority (62.2%) of family medicine residencies responded, 37.8% did not. Responding programs may differ from nonresponding residencies. It is possible that programs that allow interaction with pharmaceutical companies may have been less likely to respond to the survey. If every nonresponding program were assumed to permit industry relationships, the percentage of pharma-free family medicine residencies would be reduced from 26.2% to 16.3%.
It is also possible that some completed questionnaires did not accurately reflect programs' practices. Although questionnaires were addressed to residency directors or coordinators, other individuals sometimes completed the surveys.
Further, there is little consensus about how best to measure pharmaceutical influence in residency programs. We kept our survey deliberately brief to maximize responses, so we focused on what we believed to be the most educationally relevant areas of interaction. We chose not to ask potentially relevant questions about faculty involvement in speakers' bureaus, advisory boards, or industry consultancies, faculty or program involvement in industry-sponsored research, or faculty agreements to permit industry representatives to ghost-write articles. Additionally, because we were more interested in practices than policies, we did not attempt to document the presence of formal institutional policies regarding interactions with industry.
As we noted in the Method section, we did not define the terms industry, activities (“Are any industry-sponsored residency activities allowed?”), and access (“Are industry representatives allowed access to medical students and/or residents at the family medicine center?”), to permit respondents to interpret them as broadly and inclusively as possible. It is possible that some respondents may have misinterpreted these questions, although no confusion was apparent in answers or comments.
Finally, some residencies that may classify themselves as pharma-free were not so classified in our study. We classified residencies that allowed any interaction of industry representatives with trainees, even if that interaction occurred only within the context of an educational intervention on industry interactions, as permitting industry access. As noted in the Method section, programs that noted that drug representatives only interacted with trainees as part of an educational exercise were counted as allowing access even if they responded “No” to the question, “Are industry representatives allowed access to medical students and/or residents at the family medicine center?”
This decision was based on two factors. First, most published curricula regarding relationships between residents and industry are taught without the participation of representatives.28 Second, some of us (A.F.B., A.M.B., J.S.) have observed educational sessions using pharmaceutical representatives that encouraged trainees to view them as potentially reliable sources of drug information. The representatives who participated in training exercises were perceived to be “good”; any negative experiences with sales tactics were attributed to “bad” representatives. Yet, industry representatives who participate in this type of exercise do so with the permission of their employers, who would not agree unless their participation encouraged rather than discouraged future interactions.
Further, there is no evidence that pharmaceutical representatives provide objective information. To the contrary, studies have shown that the information they provide favors targeted drugs29 and that many assertions they make are inconsistent with FDA-approved prescribing information.30
Respondents in our survey noted challenges in teaching residents how to interact with representatives, especially because residents may do so after hours. Some residency programs include educational curricula regarding marketing,28,31 and the ACGME has asserted that such training is essential.15 However, the effectiveness of formal curricula may be compromised by the “hidden curriculum.”1,32 Most physicians (94%) have some relationship with industry; most frequently, they accept samples and food or beverages at work.31 The example set by trusted mentors and colleagues who accept industry gifts, food, and consulting opportunities can be expected to exert more influence than required curricula. Additionally, there is no evidence that “training” residents to interact with drug representatives mitigates the effect of persuasive selling tactics. Including industry representatives in educational interventions can be expected to foster future industry relationships to the detriment of rational prescribing.
Residency programs may also face financial challenges associated with restricting industry relationships. Several respondents commented on this issue; one mentioned that residents “wanted the lunches, but not from the drug reps. Cost of lunches comes out of residency budget.” Indeed, food has been shown to increase attendance at educational sessions.33 Residency programs should balance the potential harm of industry-sponsored education against the potential benefit of “free” food. A follow-up study addressing how pharma-free programs have dealt with this challenge may help programs that wish to eschew industry funding.
Several studies have examined the effects of restricting drug company interactions in residency. One survey of family medicine residencies found that residents in programs with some restrictions were less likely to perceive educational information provided by pharmaceutical representatives as being reliable and were less likely to view gifts as appropriate.34 In another study, the scores of residents passing American Board of Internal Medicine examinations were inversely related to their programs' acceptance of industry funds.20 Data on the long-term effects of restrictive policies in residencies are limited; one study that examined posttraining attitudes showed that graduates from an internal medicine residency with a strict interaction policy were less likely to view information from pharmaceutical representatives as helpful.35
Existing policies have not, apparently, dissuaded residents from permitting drug representatives access to their offices. Residents from programs that restrict such access to trainees are just as likely as residents from programs without restrictions to interact with drug representatives after graduation.35,36 Again, receiving conflicting or confusing messages from training programs and mentors may help explain this discrepancy. Family physicians meet with pharmaceutical representatives 16 times per month—far more often than internists (10), cardiologists (9), pediatricians (8), surgeons (4), and anesthesiologists (2).37 This difference most likely reflects industry allocation of sales forces and may reflect the market share of pharmaceuticals controlled by family physicians.
Our findings raise several questions. Are family medicine residencies more likely than other residencies to limit industry interactions? Are graduates of pharma-free residencies less likely than graduates of other programs to accept industry samples, gifts, or financial arrangements? Are they more likely to prescribe drugs according to best evidence than are graduates of residencies with no restrictions on industry interactions? Whereas academic medical centers may tackle this issue using a “top-down” approach—that is, under a directive from the dean38—it is worth exploring where the “call to action” originates within community-based residencies that restrict industry-trainee relationships.
Further research is needed to explore how family medicine residencies' practices compare with those of programs in other specialties and how these practices change over time. Applying our survey tool and technique to a range of residencies would provide an important comparison. Regular surveys of residencies could help track trends in policies and practices.
Most U.S. family medicine residencies allow interaction between the pharmaceutical industry and trainees. However, a sizeable minority of programs are free of pharmaceutical influence. Given the evidence of the effects of marketing on physicians and learners,2–13 medical educators should consider carefully whether interaction between learners and pharmaceutical sales representatives is appropriate in an academic environment. Reliable information cannot be extracted from industry marketing pitches, and seemingly collegial relationships may undermine physicians' critical evaluation of the risks and benefits of new medicines. In the interest of rational prescribing, perhaps physicians-in-training should be taught how to refuse to interact with industry and to decline samples, gifts, and industry-funded services.
The authors thank Johanna Gusman, MS, for assisting with data collection and Tony Scialli, MD, for statistical analysis and contributions to study design.
Ms. Bell was the project manager of PharmedOut, a Georgetown University Medical Center project that advances evidence-based prescribing and educates health care professionals about pharmaceutical marketing practices. Dr. Fugh-Berman is director of PharmedOut and serves as a paid expert witness on behalf of plaintiffs in litigation regarding pharmaceutical marketing practices.
The Georgetown University institutional review board exempted the survey from review.
Fugh-Berman A, Brown SR, Trippett R, et al. A national survey of drug company interaction in family medicine residencies. Oral presentation. Research forum at the Society of Teachers of Family Medicine Annual Spring Conference, Vancouver, British Columbia, Canada, April 27, 2010.
1Sierles FS, Brodkey AC, Cleary LM, et al. Medical students' exposure to and attitudes about drug company interactions: A national survey. JAMA. 2005;294:1034–1042.
2Huang FY, Weiss DS, Fenimore PG, et al. The association of pharmaceutical company promotional spending with resident physician prescribing behavior. Acad Psychiatry. 2005;29:500–501.
3Tsang J, Rudychev I. The sample equation. Med Mark Media. 2006;41:53–58.
4Adair RF, Holmgren LR. Do drug samples influence resident prescribing behavior? A randomized controlled trial. Am J Med. 2005;118:881–884.
5Chew LD, O'Young TS, Hazlet TK, Bradley KA, Maynard C, Lessler DS. A physician survey of the effect of drug sample availability on physicians' behavior. J Gen Intern Med. 2000;15:478–483.
6Brewer D. The effect of drug sampling policies on residents prescribing. Fam Med. 1998;30:482–486.
7Roughead EE, Harvey KJ, Gilbert AL. Commercial detailing techniques used by pharmaceutical representatives to influence prescribing. Aust N Z J Med. 1998;28:306–310.
8Morelli D, Koenigsberg MR. Sample medication dispensing in a residency practice. J Fam Pract. 1992;34:42–48.
9Spiller LD, Wymer WW. Physicians' perceptions and uses of commercial drug information sources: An examination of pharmaceutical marketing to physicians. Health Mark Q. 2001;19:91–106.
10Shaughnessy AF, Slawson DC, Bennett JH. Drug samples. J Fam Pract. 1993;37:533.
11Strang D, Gagnon M, Molloy W, et al. National survey on the attitudes of Canadian physicians towards drug-detailing by pharmaceutical representatives. Ann R Coll Physicians Surg Can. 1996;29:474–478.
13Boltri JM, Gordon ER, Vogel RL. Effect of antihypertensive samples on physician prescribing patterns. Fam Med. 2002;34:729–731.
14Groves KEM, Sketris I, Tett SE. Prescription drug samples—Does this marketing strategy counteract policies for quality use of medicines? J Clin Pharm Ther. 2003;28:259–271.
16Brennan TA, Rothman DJ, Blank L, et al. Health industry practices that create conflicts of interest: A policy proposal for academic medical centers. JAMA. 2006;295:429–433.
20Loertscher LL, Halvorsen AJ, Beasley BW, Holmboe ES, Kolars JC, McDonald FS. Pharmaceutical industry support and residency education: A survey of internal medicine program directors. Arch Intern Med. 2010;170:356–362.
21Zipkin DA, Steinman MA. Interactions between pharmaceutical representatives and doctors in training: A thematic review. J Gen Intern Med. 2005;20:777–786.
22Nakayama DK, Bozeman AP. Industry support of graduate medical education in surgery. Am Surg. 2009;75:395–400.
23Wang Y, Adelman RA. A study of interactions between pharmaceutical representatives and ophthalmology trainees. Am J Ophthalmol. 2009;148:619–622.
24Brotzman GL, Mark DH. Policies regulating the activities of pharmaceutical representatives in residency programs. J Fam Pract. 1992;34:54–57.
26Crabtree BF, Miller WL, eds. Doing Qualitative Research. 2nd ed. Thousand Oaks, Calif: SAGE Publications; 1999.
28Montague BT, Fortin AH, Rosenbaum J. A systematic review of curricula on relationships between residents and the pharmaceutical industry. Med Educ. 2008;42:301–308.
29Lexchin J. What information do physicians receive from pharmaceutical representatives? Can Fam Physician. 1997;43:941–945.
30Sernyak M, Rosenheck R. Experience of VA psychiatrists with pharmaceutical detailing of antipsychotic medications. Psychiatr Serv. 2007;58:1292–1296.
31Shaughnessy AF, Slawson DC, Bennett JH. Teaching information mastery: Evaluating information provided by pharmaceutical representatives. Fam Med. 1995;27:581–585.
32Steinman MA, Shlipak MG, McPhee SJ. Of principles and pens: Attitudes and practices of medicine housestaff toward pharmaceutical industry promotions. Am J Med. 2001;110:551–557.
33Segovis CM, Mueller PS, Rethlefsen ML, et al. If you feed them, they will come: A prospective study of the effects of complimentary food on attendance and physician attitudes at medical grand rounds at an academic medical center. BMC Med Educ. 2007;7:22–27.
34Brotzman GL, Mark DH. The effect of resident attitudes of regulatory policies regarding pharmaceutical representative activities. J Gen Intern Med. 1993;8:130–134.
35McCormick BB, Tomlinson G, Brill-Edwards P, Detsky AS. Effect of restricting contact between pharmaceutical company representatives and internal medicine residents on posttraining attitudes and behavior. JAMA. 2001;286:1994–1999.
36Ferguson RP, Rhim E, Belizaire W, Egede L, Carter K, Lansdale T. Encounters with pharmaceutical sales representatives among practicing internists. Am J Med. 1999;107:149–152.
37Campbell EG, Gruen RL, Mountford J, et al. A national survey of physician-industry relationships. N Engl J Med. 2007;356:1742–1750.
38Rothman DJ, Chimonas S. New developments in managing physician-industry relationships. JAMA. 2008;300:1067–1073.