Dubovsky, Steven L. MD; Kaye, David L. MD; Pristach, Cynthia A. MD; DelRegno, Paula MD; Pessar, Linda MD; Stiles, Keith
With each new revelation in the media of the potential influence of “big pharma” on research, practice, and teaching in academic health centers, pressure mounts on medical schools and teaching hospitals to limit the intrusion of industry.1 Such concerns seem warranted. A recent national survey of physicians in six specialties found that 94% of respondents had some sort of relationship with industry.2 Drug companies spend as much as $21 billion per year promoting their products,3,4 primarily to physicians.4 Almost half of practicing physicians and a third of residents acknowledge that pharmaceutical representatives are moderately to very important in influencing their prescribing habits,3 and physicians who believe that they are not influenced by pharmaceutical representatives' gifts with “no strings attached,” especially if they are only small trinkets like a pen or mug, still experience a sense of obligation to reciprocate.4 Accepting drug samples, attending industry-sponsored symposia, and even simply seeing industry representatives have all been found to increase prescribing of the representative's product and requesting that the product be added to the physician's hospital formulary.5
In response to these kinds of findings, medical schools have developed increasingly stringent policies for reducing industry influence on the medical profession and its trainees. Most have advocated isolating faculty, students, and residents from marketing on the basis of the assumption that, because physicians usually do not recognize its impact, such influence can only be avoided by banning any situation in which it could occur. The assumption that a deeply enmeshed culture can be altered just by promulgating new rules may not be entirely correct, and prohibiting exposure to marketing on one's own time may raise First Amendment issues. Even if such limits could be fully implemented, restricting all interactions with industry would not necessarily promote internal change or provide for teaching the next generation of physicians how to interpret and respond to the many forms of marketing they will inevitably encounter in practice. A truly effective policy should integrate the urgent need to define the boundaries between industry and medicine with the academic missions of the medical school.
We describe the evolution in a clinical department of a comprehensive policy for industry relations that applies equally to full-time and volunteer faculty and to residents, that is educational as well as prescriptive, and that attempts to maintain an ethical partnership with industry. We approached the policy with the understanding that it was possible to focus the attention of physicians on their commitment to independent decision making, though it was not possible to restrict anything that might influence them. We disagreed with suggestions that, because research shows physicians are unaware of the influence marketing has on them, they are incapable of learning new approaches to industry relations, making it necessary to legislate behavior with regulations that are impossible to enforce. This orientation is not specific to psychiatry, but because psychiatrists have frequently been cited in examples of industry influence,6 we offer our experience with developing and implementing a new policy for industry relations within an academic psychiatry department as an illustration of issues that have risen to general awareness.
Development of a Policy
Like most state medical schools, the University at Buffalo supports the majority of its activities with grants and clinical activity. With several areas of extremely strong extramural funding, a large number of clinical investigations were supported by industry-sponsored projects. The presence of a separate practice plan for each clinical department inhibited economies of scale and group contracting, thereby limiting overages of clinical funds available to support the research mission. Limited resources and a long history of reliance on industry to support a variety of activities spurred the evolution of an institutional culture involving unfettered access of industry representatives to faculty, residents, and support staff; industry support of travel and textbooks for residents; regular industry-sponsored meals in hospitals and clinics, free for residents and faculty and accompanied by promotional presentations by representatives; unmonitored and frequent after-hour promotional dinners; recruitment of lecturers from speakers' bureaus; and industry sponsorship of research that bore no relationship to academic activities. The pervasive presence of industry sponsorship created among faculty and trainees a sense of entitlement to food, educational and personal items, and access to national experts. The recipients of industry largesse generally claimed that it did not influence their prescribing patterns and attitudes toward various products because they believed they did not pay attention to promotional activities.
In the absence of a consistent institutional policy for interactions with industry representatives, leadership in the Department of Psychiatry began to pursue a change in this orientation in 2005. The University at Buffalo's Department of Psychiatry was a moderately sized department with 55 full-time faculty, 120 volunteer faculty, 35 adult psychiatry residents, 6 to 8 child psychiatry residents, and 1 to 2 geropsychiatry fellows. Clinical income of the department was robust, and industry-sponsored clinical trials were used to support aspects of the research infrastructure and to support exploratory academic research. Nevertheless, residents, faculty, and community practitioners felt dependent on industry funding for most of their activities to a degree that was disturbing to departmental leadership. On the basis of previous research in this area,7 we decided to develop a realistic yet comprehensive approach to governing relationships with industry.
The first step in the process was to raise awareness of the problem in the faculty through a process that involved no industry financial or material support of any kind. A series of faculty retreats supported by departmental funds were held at our primary clinical site and at an off-site location. Data on the impact of industry marketing on practice were discussed, and a new standing committee on industry relations made up of faculty and residents was assembled. The committee was charged with developing a comprehensive policy for industry interactions and then with continually reviewing, modifying, and ensuring adherence to the policy. The committee continues to meet regularly to review progress and discuss modifications.
Because we planned to create a fundamental shift in the relationship between industry and academia, we wanted to involve representatives from industry in the process to ensure their understanding of and commitment to a productive reform. Once the committee had developed a new policy, we began to invite a representative of each manufacturer that interacts regularly with the department to a portion of each meeting to address problems that arose in implementation of the policy. We thought that a true collaboration could be developed with firm, enforceable boundaries. One industry representative (K.S.) who felt that such collaboration was in the best interests of both parties participated as an author of this article.
Although the dialogue between industry representatives and the committee members has continued, industry representatives did not influence the policy as it emerged and evolved, nor did they provide input into its documentation (including this manuscript). Each phase of the development of the policy was discussed in faculty meetings, and any changes required an endorsement from the faculty. Through continued dialogue, a policy emerged that reflected the changing will of the faculty rather than one that was imposed on it. The entire policy was codified before national groups or our own institution released their own suggested approaches to industry relations, and therefore it was not a response to external recommendations. The policy, a complete version of which is available at http://www.smbs.buffalo.edu/psychiatry/main/index.html, applies to voluntary as well as full-time faculty and provides objective, achievable, and measurable standards for compliance.
Consistent with most such policies that have emerged recently, our guidelines for interactions with industry eliminated industry-sponsored meals of all kinds as well as all gifts that had no patient-care value. The policy states that items with potential use in patient care, such as unbranded informational brochures, must be given to a faculty “industry liaison” for distribution as determined by the department. Representatives no longer have any direct access to residents, and they must make appointments to see faculty members. All support of educational activities must be in the form of unrestricted educational grants, and no recommendations of speakers by industry representatives can be accepted.
Further discussion among the faculty led to additional points that have not been endorsed elsewhere. For example, so that we may teach residents how to interpret promotional material they will undoubtedly encounter in practice, pharmaceutical representatives are invited to present peer-reviewed information about new research and new products at appropriate points in the pharmacology curriculum. Each presentation is discussed by a faculty member who points out strengths and weaknesses in methodology and the appropriate place of the product in the treatment of the disorder under consideration; the faculty member may point out that the research presented does not support the advertised application of the product. To extend this education, a departmental “pharma symposium” was developed at which research supported by two different manufacturers is presented by representatives or senior scientific liaisons chosen by their companies. Each of these presentations is then discussed by a faculty member, followed by a general discussion led by the department chair and the vice chair for research. General discussions focus on the implications of the material for research and practice, including how to interpret the statistical and clinical analysis of the results. Lunch at the symposium is provided by the department. An annual grand rounds presentation addresses conflicts of interest and relationships with industry.
As the policy developed, we encountered much debate about faculty giving promotional talks (defined as any presentation directly sponsored by industry, including all mealtime lectures, or as use of slides or other material or cases provided by industry) and participating in speakers' bureaus. A consensus emerged that restricting and policing the participation of faculty members in legal activities on their own time not only was impractical but also implied that intelligent adults are not capable of changing their behavior in the light of new data. At the same time, all agreed that promotional activities, by definition, are not scientific presentations and should not carry any academic imprimatur. Toward this end, faculty titles and university affiliations cannot be used to advertise promotional events, and faculty must provide a disclaimer at the beginning of any such presentation stating that the department and the university do not endorse it. Faculty and residents are discouraged, but not prohibited, from attending such events after hours, but residents must report their attendance to the residency director, both as a means of getting residents to think twice about their participation and to determine whether their level of participation is high enough to warrant additional coursework on the impact of promotional talks.
Although the faculty felt that industry-supported, investigator-initiated research is an important source of funding for preliminary studies and that participation in multicenter trials is a source of support for the research infrastructure, some degree of oversight was necessary over the quality of industry-sponsored clinical trials and the use of funds generated by these trials. Without duplicating the efforts of our institutional review board, an existing departmental committee charged with supporting faculty research was given the charge of reviewing protocols and budgets to make sure that industry-supported research does not involve development of a product that has no potential for meaningful advantage over existing products and to ensure that funds generated by the research are used to support independent research rather than additional faculty income. Faculty chose to extend the usual disclosure recommendations to include disclosure of industry sponsorship to patients to whom a physician recommends any product of a company that has supported the faculty member's research or has paid the faculty member for any industry-related activity (e.g., a presentation or lecture) in the past year.
Response to the New Policy
Because the new policy evolved over the course of about 18 months, we did not gather systematic pre- and postintervention data on resident and faculty attitudes and behavior. As a result, our observations of apparent reactions to the changing policies are anecdotal, and we could not control for intervening variables such as changes in industry itself. However, in a department of this size, it was possible to conduct extensive discussions with all constituencies to obtain consistent impressions of reactions to the new policy.
The decision to eliminate all industry-sponsored meals engendered substantial initial concern. Burdened with large debts, residents appreciated not having to pay for food during industry-sponsored meals, and some faculty members were worried that eliminating meals might adversely affect resident morale or recruitment. Some residents did, in fact, object. For example, when presentations by representatives that had been accompanied by a meal were moved into coursework with faculty member discussion, residents were angry that all they were hearing was marketing. When asked what they thought they were hearing when representatives presented their pitches over free pizza, the answer was “lunch,” and some resident evaluations of these courses ended with “want food.” Without accompanying food, residents were more acutely aware that representatives' presentations were not clinically meaningful. Increasing skepticism of the value of these presentations led some residents to wonder whether the reason why industry presentations were included in the curriculum was that the department was being paid to do so. We addressed these concerns by ensuring that discussions with representatives occurred at logical points in the curriculum (e.g., presentations on new antidepressants during the curriculum on treatment of depression), incorporating more structured teaching about the forms marketing takes, and supplying food paid for by the department but not associated with industry presentations. These measures received positive comments from residents. However, in extensive discussions in curriculum committee meetings, residents were unanimous in their opinion that the only utility of the industry representatives' presentations was for them to learn about marketing, and once they had mastered this concept, further representative presentations were unnecessary.
Although residents are not prohibited from attending promotional events, their attendance at these sessions has declined substantially since the institution of the new educational policy. In addition, their perception of industry influence has sharpened. At a regular lecture by a volunteer faculty member, residents noticed that the lecturer emphasized a particular medication, and they asked the instructor if he had a relationship with the manufacturer of this particular medication. The lecturer admitted to being on the company's speakers' bureau. In response to an anonymous questionnaire survey, residents unanimously described the role of the industry representative as a marketer and not a source of scientific information, and the majority felt that gifts from industry could influence their prescribing behavior. Recruitment into the residency from our own and other medical schools has increased since implementation of the policy. Enhanced recruitment into our residency in all likelihood has been the result of multiple factors, including continuing to develop an appealing and useful educational program for medical students and residents, but the new industry policy at least did not seem to create ill will in existing residents that could have influenced applicants to the residency.
Adherence to complex protocols, especially by voluntary faculty, can be difficult to achieve and harder to verify. However, we have observed that, because they have been involved in the development of the protocol, faculty members have been willing to monitor themselves. On several occasions, faculty members have reported to the industry relations committee that they have seen advertisements for promotional talks that contain the university affiliation of the speaker, usually a volunteer faculty member. In these cases, the representative and the faculty member are reminded that departmental policy prohibits such advertising, and in every instance the material has been withdrawn and the required disclaimer has been issued. We were initially concerned that some members of the volunteer faculty might resign if they were required to adhere to the new protocol, but this has turned out to be unwarranted, and several volunteer faculty members have taken pains to ensure that their faculty titles are not used at promotional events. However, whereas full-time and voluntary faculty members have decreased or discontinued participation in promotional meals, local medical staff and professional groups continue to rely on industry representatives to supply meals at their meetings, which are, as required by Pharmaceutical Research and Manufacturing Association guidelines, always accompanied by a promotional talk. Although full-time faculty members have not been concerned about having to buy their own meals, it has taken considerable effort even to begin discussing reconsideration of this practice in other venues.
To address the response of industry to the new policy, one of us (K.S.) conducted an anonymous survey of the seven local representatives of pharmaceutical companies (other than his own) who most frequently visit the department. Of this group, all three representatives from smaller companies and one of the four from larger companies thought that the new policy equalized opportunities to present data, whereas previously companies with bigger marketing budgets had more opportunities to interact with physicians. Not surprisingly, the other representatives of larger companies preferred the previous state of affairs because their extensive use of meals and gifts and their larger numbers of representatives gave them a competitive advantage, and because they felt it to be more effective to see physicians unannounced than to make appointments. Two of the representatives reported that they had been pressured by their superiors to find ways to continue the previous marketing techniques. Although a formal survey might have resulted in more comprehensive replies, answers to a survey administered by the department might not have been as honest. It is a matter of opinion whether equalizing involvement of companies is beneficial, but pressure on representatives to continue previous marketing techniques is worth noting.
A recent proposal for reform of the relationship between academic health centers and industry provided a long list of activities that should be proscribed because these activities have the potential to influence physician behavior in a way that could supersede the needs of the patient.4 In addition to elimination of gifts, meals, ghostwritten publications, and payment for attendance at lectures and conferences, the authors urged prohibition of direct industry support of continuing medical education, participation in speakers' bureaus, and accepting drug samples. Paid consultation was considered acceptable if a specific work product was identified, but research grants would only be permitted if they were unrestricted, thereby preventing participation of academic faculty in industry-sponsored registration and product development studies. The suggestion has further been made that faculty should not be allowed to attend promotional events on their own time, a prohibition that raises interesting constitutional and monitoring issues. Extensive rules for full-time faculty in this and other proposals have not been accompanied by guidelines for volunteer faculty or residents.
All of the guidelines that have been proposed by various groups (Table 1)4,6,8,9 involve a variety of prohibitions and directives. However, the likelihood that they will be adhered to is not known, nor is their impact on the field. Actively practicing physicians in Florida who responded to a survey said they would be more likely to comply with AMA guidelines (Table 1) if important organizations or individuals expected it of them, even though they did not think that the guidelines would improve anything.10 A review of two studies of the effect of new regulations on industry relationships and 10 studies of the impact of educational interventions on the ability of residents and medical students to assess industry interactions concluded that residents in programs with more limitations on their interactions with industry representatives were less likely than residents in unrestricted programs to consider information from representatives to be beneficial and to find gifts, samples, meals, and honoraria to be appropriate; the more contact residents in unrestricted programs had with representatives, the more they found information from representatives to be useful.11
The educational interventions that were reviewed in the latter report11 included seminars, workshops, practice sessions, role-playing, vignettes, lectures, and debates. Most of these interventions consisted of one class or a small series of classes without changes in overall relationships with industry in the departments, and pre-post evaluations found a decrease in trainees' thinking that interactions with representatives and acceptance of gifts were appropriate.11 The educational sessions did not focus on evaluating promotional material and interactions, and changes in these factors were not assessed. The only assessment of long-term effects of any of these programs found that two months after an educational intervention, participants were less likely to accept gifts of no educational value.
Global restrictions of industry-academic interactions are appealing because they seem to control influences that faculty have not controlled themselves and because they respond to escalating public demands for accountability. However, unless pharmaceutical companies are completely prohibited from marketing their products to anyone or from interacting in any way with physicians and patients—an outcome that seems unlikely in a free society—students who do not learn how to interact with and critically interpret information from industry (or any other potential source of conflict of interest) may be more likely to be influenced in practice.11 Our own experience combining a comprehensive education program with limits on, but not absolute prohibitions of, interactions with industry seems to promote a more enduring sense of medical professionalism. Involving faculty and residents in formulating parameters of the interaction results in greater investment in the process, whereas rules that are imposed from the outside inspire less commitment on the part of those being regulated.
In the absence of more extensive data, it is, of course, impossible to assert that the process we have described has a positive impact on the capacity of academicians and their students to eliminate inappropriate influences of marketing on teaching, research, and practice. However, our experience suggests that it may be possible to involve faculty more directly in developing enforceable policies directed toward a professional partnership with industry that maintains necessary boundaries between the partners. The only costs to the department were the minimal expense of meals at resident meetings and the pharma symposium. We may be more effective at protecting our patients from overenthusiastic marketing and at promoting lifelong learning by teaching students how to identify and interpret marketing than by delegating to others the responsibility for deciding what to do about the marketing to which we will continue to be exposed so long as we live in a marketplace economy.
Funding/Support: Dr. Dubovsky has received research support from Sumitomo, Otsuka, Pfizer, and the Tower Foundation. Doctor Kaye has received research support from Pfizer.
Other disclosures: None.
Ethical approval: Not applicable.
1 Monastersky R. Hidden payments to university researchers draw new fire. Chron High Educ. October 31, 2008:1.
2 Campbell EG, Gruen RL, Mountford J, Miller LG, Cleary PD, Blumenthal D. A national survey of physician-industry relationships. N Engl J Med. 2007;356:1742–1750.
3 Ziegler MG, Lew P, Singer BC. The accuracy of drug information from pharmaceutical sales representatives. JAMA. 1995;273:1296–1298.
4 Brennan 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.
5 Somerset M, Weiss MK, Fahey T. Dramaturgical study of meetings between general practitioners and representatives of pharmaceutical companies. BMJ. 2001;323:1481–1484.
6 Kaplan A. Conflicts grow over conflicts-of-interest policies and practices. Psychiatr Times. August 1, 2008;25(9):3.
7 Dubovsky SL, Dubovsky AN. Psychotropic Drug Prescriber's Survival Guide: Ethical Prescribing in the Age of Big Pharma. New York, NY: W.W. Norton; 2007.
8 Research A-AACoFCoIiHS. Protecting Patients, Preserving Integrity, Advancing Health: Accelerating the Implementation of COI Policies in Human Subjects Research. Washington, DC: American Association of Medical Colleges; 2008.
9 American College of Obstetricians and Gynecologists Committee on Ethics. ACOG Committee Opinion No. 401: Relationships with industry. Obstet Gynecol. 2008;111:799–804.
10 Pinto SL, Lipowski E, Segal R, Kimberlin C, Algina J. Physicians' intent to comply with the American Medical Association's guidelines on gifts from the pharmaceutical industry. J Med Ethics. 2007;33:313–319.
11 Carroll AE, Vreeman RC, Buddenbaum J, Inui TS. To what extent do educational interventions impact medical trainees' attitudes and behaviors regarding industry-trainee and industry-physician relationships? Pediatrics. 2007;120:e1528–e1535.