Four reviews have documented the extent of the physician-pharmaceutical industry interaction and its impact on physician prescribing practices (1—4). These reviews show that continuing medical education events sponsored by pharmaceutical companies, funding for travel or lodging for educational symposia and information from and presentations by pharmaceutical representatives were all associated with an increase in practicing physicians’ arbitrarily prescribing a sponsor’s medication. More recently, a retrospective study showed that psychiatric residents preferentially prescribe companies’ medications shortly after sales visits by that company’s pharmaceutical representative (5).
Organized medicine, the federal government, and the pharmaceutical industry itself have recognized the conflicts of interest between physicians and the pharmaceutical industry and have developed guidelines. The American Medical Association (AMA) published ethical guidelines on accepting gifts from pharmaceutical representatives, stating that gifts "should primarily entail a benefit to patients," "should not be of substantial value," and should "serve a genuine educational function" (6). The American College of Physicians has published the following in their Ethics Manual: "The acceptance of individual gifts, hospitality, trips, and subsidies of all types from the health care industry by an individual physician is strongly discouraged." Pharmaceutical Research and Manufacturers of America (PhRMA) has published a Code on Interactions with Healthcare Professionals which limits types and expenses of gifts as well as the context in which they are given with the stated goal of avoiding any compromise of the clinician’s independence of decision making (7). Recently, the Department of Health and Human Services published a compliance program mandating the PhRMA code for pharmaceutical companies, resulting in substantial fines for violations (8).
Physician training programs offer a critical opportunity to disseminate knowledge about the relationship between physicians and the pharmaceutical industry and to encourage appropriate behavior with pharmaceutical representatives. Based on the literature, unchecked access to residents by pharmaceutical representatives is very common (9—12). Residency programs have tried banning pharmaceutical representatives, restricting access and gifts and educating residents. Policies that define and restrict interactions between internal medicine residents and pharmaceutical representatives have been shown to increase residents’ awareness of ethical issues concerning this interaction (13). Educational seminars, depending on the content and persuasiveness, have also been shown to affect family practice and internal medicine residents’ attitudes about their own vulnerability to marketing strategies (14, 15).
However, no studies have determined whether changes in gift accepting behavior could be affected by a one-time educational intervention. Moreover, those studies demonstrating changes in attitude after an educational intervention have not included comparison groups. The purpose of this study was to determine whether an educational intervention would change both residents’ attitudes toward pharmaceutical representatives and their gift-accepting behavior.
A controlled trial of an educational intervention on residents’ attitudes and behavior toward pharmaceutical representatives was conducted in a large, multisite, university-affiliated psychiatry residency program. The residents were divided into two groups based on a natural geographic division within the program: one-half of the residents were rotating at a Veterans Affairs Medical Center and the other half were at a community mental health center and two private facilities several miles away. Therefore, the residents were not randomized. The VA residents received the educational intervention while the residents at the state and private facilities served as the comparison group.
After obtaining informed consent, the questionnaire was administered at a central meeting to all residents at each site on October 4 and 5, 2001 to the comparison group and on October 18, 2001 to the intervention group. Residents were randomly assigned a code which was printed on the questionnaire at baseline and follow up. In order to facilitate anonymity, all data were entered and analyzed by code and never linked to residents’ names. The same questionnaire was mailed to all residents two months after the educational intervention. All questionnaires included in the analysis were completed and returned within the three month rotation block, before any resident rotated to another site. The study was approved by the West Haven VA Human Investigation Committee and exempted from review by Yale University Human Investigation Committee.
All residents rotating on psychiatry services at the time of the study were invited to participate (N=48). Only residents who completed the baseline and follow up surveys and attended the seminar (for the intervention group) were included in the analyses of attitude and behavior changes (N=32).
The goal of the seminar was twofold: 1) to facilitate a discussion of the pros and cons of receiving information from, socializing with, and accepting gifts from pharmaceutical representatives, and 2) to disseminate ethical guidelines and evidence of the impact of detailing and gifts on physicians.
The 1.5-hour educational intervention was conducted at the VA facility by two faculty, including the associate residency director, with many years of teaching experience. The first half consisted of reading and discussing brief editorials "Gifts to Physicians from the Pharmaceutical Industry" and "Pharmaceutical Promotions—a Free Lunch?" from prominent journals (16, 17). Viewpoints for and against industry contributions were presented by the co-leaders and then discussed by the residents. The co-leaders presented for discussion the research data showing that interactions with pharmaceutical representatives and receiving gifts negatively affect prescribing practices of physicians. The amount of money the pharmaceutical industry spends on marketing to physicians was reviewed. The group discussed the American Medical Association and American College of Physicians written ethical guidelines for accepting gifts from pharmaceutical representatives.
The second half of the seminar consisted of four fictional, humorous vignettes of typical interactions between psychiatrists and pharmaceutical representatives. The residents were asked how they would handle each situation and a discussion of the ethical implications was facilitated. The vignettes illustrated common rational and irrational marketing strategies utilized by pharmaceutical representatives in their interactions with physicians, as described in "Separating the Wheat from the Chaff" (18), which was distributed to the residents at the end of the seminar. This article details the fallacies of logic and gimmicks described in marketing textbooks and variably used by the pharmaceutical industry.
To assess residents’ attitudes toward pharmaceutical representatives, a modified version of the questionnaire developed and validated by McKinney et al. (10) was administered. This widely used questionnaire assesses three broad categories of information: demographic data, quantity of interactions and gifts, and personal judgments about the influence and usefulness of interacting with pharmaceutical representatives. The latter were measured with a 5-point Likert scale for each one of 21 statements about pharmaceutical representatives and typical gifts they offer. Factor subscores of attitude were calculated based on the factor analysis McKinney et al. performed to identify the underlying constructs: education by pharmaceutical representatives, influence of pharmaceutical representatives, and adequacy of ethics training of residents about their interaction with the pharmaceutical industry (10). Subscores for gifts were also calculated based on the categories of cost, educational value, and patient care benefit. The questionnaire was piloted the previous year with eight graduating residents, leading to minor modifications of a few items.
Simple frequencies were used to assess demographic characteristics of the residents, including level of training, age, sex and number of hours worked per week. Mean scores on the survey instrument subscales were calculated. T-tests were used to test for significant differences on attitude and behavior factor scores before and after the educational intervention. Analysis of covariance was used to examine the significance of interactions between group (intervention vs. control) and time (pre vs. post). An alpha level of 0.5 was used as the criterion of significance. All analyses were conducted using SAS, Version 8.
At the time of the study, there were 48 residents in the residency program. Overall, 43 residents (90%) completed the baseline survey. Four of the five residents who did not respond were in the intervention group. The response rate to the baseline survey in the intervention group was 18 out of 22 residents (82%). The response rate from the comparison group was 25 of 26 residents (96%). Resident demographics are shown in t1.
Overall, 14 of the 22 residents (64%) in the intervention group completed the baseline survey, attended the seminar, and completed the follow up survey. In the comparison group, 18 of 26 residents (69%) completed the baseline and the follow up surveys. There were no significant differences in age or gender between these residents and the residents who only completed the baseline survey.
Attitudes of Residents Toward Pharmaceutical Representatives and Their Gifts
At baseline, residents tended to believe that they were immune to the influence of pharmaceutical representatives and their gifts, but that other physicians were influenced (t2). Residents believed that pharmaceutical representatives use marketing techniques and that the information they present is of questionable validity, but they did not think pharmaceutical representatives should be banned. The mean subscore on a Likert scale of 1—5 for the influence factor was 2.65 (SD=0.55). A lower score indicates that residents believe pharmaceutical representatives do not influence physicians. The mean subscore for the education factor was 3.22 (SD=0.76). Residents wanted more training in interacting with pharmaceutical representatives but did not want those interactions restricted or monitored. At baseline, 81% of all residents stated they did not have sufficient training during residency regarding how they should interact with pharmaceutical representatives.
Residents’ attitudes toward gifts from pharmaceutical representatives were assessed at baseline. The most acceptable gift to residents was drug samples for patients (t3). The majority of residents indicated that educational gifts of moderate expense were appropriate to accept. Residents believed that expensive gifts with no educational value were inappropriate (t3).
At the 2-month follow up, there were no significant changes in either group in the factor subscores of residents’ attitudes toward pharmaceutical representatives and their gifts, nor were there any group by time interactions.
Behavior of Residents With Pharmaceutical Representatives
At baseline, residents reported that they had interactions with a pharmaceutical representative on average once a month (range 0—15 times per month). The two groups differed at baseline: the intervention group averaged 2.16 interactions per month, while the comparison group averaged only 0.92 interactions per month. Only 14% of the residents had no contact with a pharmaceutical representative over the two months. The most common gifts accepted were meals while attending educational and administrative meetings, social meals, and miscellaneous office supplies. Interestingly, more residents had attended a social dinner at a restaurant than thought this gift was appropriate (t3). Although residents considered drug samples for patients and medical textbooks the most appropriate gifts, few had accepted such gifts during the period studied.
Altogether, 53% of residents had attended at least one presentation by a pharmaceutical representative in the previous two months. Most of these residents (N=13, 59%) listened quietly to the presentation. A quarter of them reported that they ignored the presentation, working on other things like progress notes. A minority (9%) of those who attended such a presentation reported that they questioned the representative about the reliability of the information presented.
The educational intervention significantly reduced residents’ self-reported acceptance of miscellaneous office supplies and gifts with no educational value. The intervention group reduced their acceptance of miscellaneous office supplies by 35% (F = 17.28, p=0.0001). They reduced their acceptance of gifts with no educational value by 20% (F = 4.83, p= 0.032). The educational intervention did not change resident behavior in acceptance of educational gifts, modestly priced gifts, or gifts with benefit to patients.
It is widely accepted that interactions between the pharmaceutical industry and physicians raise conflicts of interests which should be addressed during residency training. While there have been studies of residents’ attitudes toward pharmaceutical representatives and the effect of education on these attitudes, there have been no published controlled studies of an educational intervention on attitudes and behavior of residents. Educational intervention studies seek to establish whether a curriculum increases competency, justifying wide dissemination. In this study, gift-accepting behavior of residents who attended an educational seminar significantly changed compared to that of a comparison group who did not attend the seminar. Specifically, the intervention group accepted significantly fewer office supplies and noneducational gifts. The residents were taught that companies attempt to influence physicians by using arbitrary appeals like reminders branded on office supplies. Both of these changes in behavior concur with the marketing techniques and ethical guidelines which were taught during the seminar.
Although residents accepted fewer office supplies and noneducational gifts, their attitudes toward accepting these gifts did not change. Residents were neutral about the appropriateness of accepting office supplies both before and after the seminar. They also remained neutral about the appropriateness of accepting gifts with no educational value. One explanation is that residents changed their behavior out of concern of what others would think rather than a change in their own beliefs. A question posed during the seminar was "What would patients think of my accepting this gift?" Residents may have become more aware of how their behavior looked to others. Furthermore, the seminar, which recommended limiting interactions with pharmaceutical representatives, was taught by the associate residency director. The residents may have been concerned with his opinion of them and may have changed their behavior in accordance with their beliefs about his stance on the issue.
At baseline, the majority of residents responded that physicians in general are influenced by pharmaceutical representatives but that they themselves are not influenced by pharmaceutical representatives or their gifts. This finding is consistent with studies showing that residents and physicians from other specialties also believe they are not influenced by gifts from pharmaceutical companies (9, 15, 19). However, psychiatry residents were more likely than internal medicine residents (10) to believe that discussions and gifts do not influence their own prescribing practices.
At baseline, residents tended to believe that pharmaceutical representatives use marketing techniques and do not provide accurate or useful information, but that they should not be banned nor should the program restrict their interactions with residents. Compared to other studies, psychiatry residents were more likely than internal medicine residents to oppose banning pharmaceutical representatives. The residents in this study also did not think a faculty member’s presence at pharmaceutical representative presentations would be useful. However, consistent with other studies, residents in this study think they need more training in interacting with pharmaceutical representatives (9—11). Overall, residents seemed to value highly their interactions with pharmaceutical representatives, but not because they considered this interaction educational. Observations of pharmaceutical representatives complimenting and giving gifts to psychiatry residents—who endure long, exhausting call nights—suggests that residents may value their contact with pharmaceutical representatives because representatives make them feel appreciated and important.
Limits of the study include the low number of pharmaceutical representative-resident interactions during the study period as well as a small sample. However, effect size as represented by percent changes in scores over time were substantial for the statistically significant findings. A second limitation is that disproportionate number of PGY-2s did not attend the seminar or complete the follow up survey. It was not possible to control for changes in policy, educational curriculum, or attitudes toward pharmaceutical representatives that occurred in the control sites.
The addition of this research to the two published studies of the effect of formal education on residents’ attitudes (14, 15) suggests that to change residents’ attitudes, the curriculum may require role modeling and experiential learning, wherein faculty model and discuss interactions with pharmaceutical representatives. Shaughnessy et al. (14) used an extended course which included presentations by pharmaceutical representatives and residents’ formal critique of those presentations supervised by faculty. They showed that residents were more likely to believe that discussions and gifts affect their own prescribing practices after the course. However, there was no comparison group. Hopper et al. (15) used a shorter didactic and discussion format with little effect on residents’ insight about the influence of pharmaceutical representatives. Therefore, residents may learn more about the influence of pharmaceutical representatives when they are engaged in experiential learning rather than didactics or hypothetical cases alone.
One-time educational interventions may have significant but limited impact on the type and quantity of gifts that psychiatric residents accept from pharmaceutical representatives. While targeted gift-accepting behaviors changed, attitudes did not change. As our study has underscored, there are many determinants of behavior other than residents conscious attitudes about those behaviors. Therefore, it would be useful in further research to confirm the influence of pharmaceutical representatives on residents’ prescribing practices and to study the impact of an educational intervention on this influence.