Relationships between academic medical centers and the pharmaceutical industry have been increasingly criticized as creating conflicts of interest that may undermine the scientific rigor and objectivity of medical education and research. Pharmaceutical industry presence has been pervasive in academic medical schools, and access to physicians is promoted by gifts from pharmaceutical representatives (PRs). Sixty percent of department chairs in academic medical centers have personal relationships with industry (1). In 2006, after congressional investigations, a group of influential medical professionals proposed elimination of industry sponsorship of continuing medical education (CME) and prohibition of gifts of any size at academic medical centers (2). Some medical schools began to restrict PR presence and gifts on campus (3–5). At the same time, a special task-force of the American Association of Medical Colleges urged academic medical centers to establish and implement clear policies prohibiting on-site and off-site acceptance of gifts, travel funds, ghostwriting, and, in lectures that were not CME-accredited, refreshments (6). The task-force recommended that medical schools restrict site access by PRs to only nonpatient and nonpublic areas and centrally manage requests for industry-supported CME activities, distribution of pharmaceutical samples, scholarships, and educational funds. Finally, they encouraged development of educational programs to raise awareness of challenges to professionalism that can result from certain types of interactions with industry (6).

There have been many studies examining the knowledge, attitudes, and practices of postgraduate trainees around PRs (7). These studies indicate that contacts with PRs are frequent; the information presented by PRs is biased; and gift-giving effectively results in increased prescription of expensive medications with no clear advantage over generics; yet there is the perception among residents that acceptance of gifts does not influence them. There are fewer studies about the attitudes and interactions between medical students and PRs. In 2003, Sierles and colleagues surveyed medical students’ exposure to and attitudes about drug company interactions at eight medical schools in the United States (8). They found that the average exposure was one gift or sponsored activity per week. Two-thirds of students indicated that gifts would not influence their own prescribing practice, and 58% believed gifts did not affect other physicians’ prescribing practices (8).

In 2006, members of the OHSU House Staff Association appealed to the Dean of the OHSU School of Medicine to ban pharmaceutical presence at OHSU. The Dean responded by forming a task-force to examine the issue. In order to obtain a clearer picture of the exposure of trainees to the pharmaceutical industry, a web-based, campus-wide survey was distributed. The goal of this study is to examine the results of this survey and compare clinical and preclinical medical students’ attitudes and experiences with PRs.

A web-based, anonymous survey was sent to all OHSU medical students in 2007 and re-sent two additional times to nonrespondents. To ensure anonymity, collected demographic information was restricted to sex and age-in-decades. Medical students were divided into preclinical (Years 1 and 2) and clinical (Years 3 and 4) groups. Ten attitudinal questions developed by McKinney and coauthors (9), rated on a 5-point scale (Strongly Agree to Strongly Disagree), measured respondents’ views on the value, educational benefit, and influence of industry in the academic setting, and how prepared students felt to manage relationships with industry. McKinney ensured construct validity by having individuals with expertise in questionnaire design, education, and pharmaceutical detailing review the questionnaire. Four behavioral questions measured the number of pharmaceutical company conferences attended both on and off campus and number of books and small items, such as pens, received as gifts in the previous year. One additional question, developed by the OHSU Task Force, asked whether the student was aware of formal educational offerings for medical students addressing conflicts of interest in relationships with the pharmaceutical industry (Yes, No, Unsure). The OHSU Institutional Review Board approved analysis of the data for the purposes of publication. Results from this survey analyzing the views of psychiatry residents and faculty on interactions with the pharmaceutical industry have been previously published (10).

Logistic regression was used for each individual attitudinal question to assess the relationship between agreement with the question (Agree or Strongly Agree versus Strongly Disagree, Disagree, or Neither Agree Nor Disagree) and the variables year in program (clinical, preclinical), sex, and age (over 30 years, 30 years or under). Results for each question include proportions that agree/strongly agree for each category of the predictors and multivariate p values. Adjusted odds ratios (ORs) and 95% confidence intervals (CIs), are reported only if different between clinical and preclinical students.

Behaviors were described as present or absent, and compared between clinical and preclinical students by chi-square tests. Among the subset of students who either attended a pharmaceutical event or accepted a gift from a pharmaceutical representative, we compared the median number over the last year by Mann-Whitney U. Based on the work of McKinney (9), we developed two factors. Each factor score was obtained by averaging the values of the component questions, with scores ranging from 1 to 5. One factor incorporated five Attitudinal items measuring educational benefits of interacting with PRs (PRs have an important teaching role; PRs provide accurate information on new and established drugs; PRs should be restricted from making presentations on campus [this item score was reversed]; PRs support important speakers). For the Educational Benefit factor, higher scores indicate stronger belief in higher educational benefit of interactions with PRs. The second factor incorporated three items measuring the influence of the PRs (student would maintain contact with PRs if there were no gifts; discussion with or acceptance of gifts from PRs does not affect students’ or others’ prescribing). For the PRs’ Influence factor, lower scores indicate more acknowledgment of the influence of PRs. Stepwise logistic regression (Wald procedure) was used to examine the factor’s ability to predict each of the four behaviors, after adjusting for age, sex, and year in program. Data for these analyses are reported as ORs with 95% CIs.

The survey was sent to 476 students in 2007; 280 responded (59%), but 20 surveys were excluded because the medical student did not indicate year of training or had other missing values. Overall, the views of 50% of preclinical students and 60% of clinical students were analyzed and compared.

Respondents did not believe that PRs have an important teaching role on campus, with only 5% of clinical and 7% of preclinical students endorsing the statement (NS); 17% of clinical students and 18% of preclinical students believed that PRs provided useful and accurate information on new drugs (NS) and 14% of clinical and 15% of preclinical students on established drugs (NS).

Many students appeared aware of the impact that gifts or food from PRs might have on their own prescribing. Only 30% of clinical and 33% of preclinical students reported that they would maintain the same degree of contact with PRs if no food or gifts were distributed (NS). Ten percent of clinical students and 12% of preclinical students believed that accepting gifts had no impact on other physicians’ prescribing (NS). Only 18% of clinical students and 7% of preclinical students indicated they had sufficient training about interacting with pharmaceutical representatives (p=0.03). Despite the lack of positive views, 72% of clinical and 65% of preclinical students agreed that pharmaceutical companies supported important conferences and speakers (NS).

Among clinical students, 34% agreed that the medical school provided formal education in managing conflicts of interest with the pharmaceutical companies, as compared with only 7% of preclinical students. Awareness of formal teaching had no impact on the factors that measured their views on the educational benefit of interactions with PRs (formal education: mean score 2.7 [SD: 0.8] on a 5-point scale; no formal education: mean score 2.7 [SD: 0.7]; NS) or the influence of PRs (formal education, mean score 2.7 [SD: 1.6]; no formal education, mean score 2.9 [SD 0.8]; NS).

Several measures indicated that views on the pharmaceutical industry were more negative among the clinical than the preclinical students; 54% of clinical medical students and 32% of preclinical medical students affirmed that PRs should be restricted from making presentations at OHSU (adjusted OR: 2.31; 95% CI: 1.37–3.9; p=0.002); only 30% of clinical students, versus 50% of preclinical students, agreed that accepting gifts or food had no impact on their prescribing choices (adjusted OR: 0.42; 95% CI: 0.25–0.72; p=0.002). However, despite more negative views, clinical students had substantially more contact with PRs than did preclinical students and were more likely to accept gifts from PRs. Clinical students were more likely to have accepted small gifts (71% of clinical students, 58% of preclinical students; p=0.03) and educational materials such as books (52% of clinical students, 33% of preclinical students; p=0.004). In the previous year, clinical students were more likely to have attended at least one industry-sponsored event, both on campus (84% of clinical students, 40% of preclinical students; p<0.001) and off campus (57% of clinical students, 27% of preclinical students; p<0.001). The median number (interquartile range) of industry-sponsored events attended on campus by clinical students in the previous year was 5 (3–10) and by preclinical students was 2 (2–5; p <0.001).

Views on the educational benefits and influence of pharmaceutical representatives had no impact on whether students attended conferences both on and off campus. For these behaviors, only clinical year was important, with clinical students having 7 times the odds of preclinical students of attending a conference supported by the pharmaceutical industry on campus (OR: 7.3; 95% CI: 3.8–13.9; p<0.001), and three times the odds of attending one off campus (OR: 3.2; 95% CI: 1.8–5.6; p<0.001). However, views on the pharmaceutical industry did affect the likelihood of accepting gifts. For example, for each point-increase on the 5-point scale in views on the educational benefit of interactions with PRs, students had 80% greater odds of accepting a book or other educational material, after adjusting for industry influence (OR: 1.8; 95% CI: 1.2–2.8; p=0.006). For each point-increase on the 5-point measure acknowledging the influence of PRs, students had 30% lower odds of accepting a book or other educational material, after controlling for educational attainment and the influence of industry factor (OR: 0.7; 95% CI: 0.5–1.0; p=0.04). Clinical students had twice the odds of preclinical students of accepting a medical text as a gift (OR: 2.1; 95% CI: 1.2–3.9; p=0.015). For each point-increase on the factor measuring educational benefit of relationships with the pharmaceutical industry and its representatives, medical students had 70% greater odds of accepting small gifts with company logos, such as pens (OR: 1.7; 95% CI: 1.2–2.5; p=0.01).

The results of this survey indicate that, overall, medical students at OHSU do not have positive views about the benefits of interactions with PRs and are dubious of the educational benefits of these relationships. Despite this, they do believe that funds from pharmaceutical companies support important conferences. Furthermore, among the clinical students who have the more negative views, the majority still attend industry-sponsored events, both on and off campus. Those with more negative views on the educational benefit of these interactions and who were more willing to acknowledge PRs’ influence on prescribing were less likely to accept books as gifts from PRs. Those with more negative views on the educational benefit of PR interactions were less likely to accept pens and other small gifts.

Studies of medical students completed several years earlier at other institutions indicate more positive views. In a study of four medical schools in the Midwest, 70% of students felt that acceptance of gifts from the pharmaceutical industry had no influence on physician practice, and 65% of clinical students agreed it was appropriate for medical students to accept these gifts. Clinical students were more permissive than preclinical students (11). The authors concluded that medical students are vulnerable to “unrecognized influence.” Sierles et al., in a 2003 study of eight geographically diverse medical schools, found that 68.8% did not believe that acceptance of gifts would influence their practices, but 57% did not believe gifts would not affect a colleague’s practice (8). Only 17.3% believed that their school should exclude pharmaceutical representatives from meeting with students.

The reasons for the negative views of OHSU medical students about the pharmaceutical industry are unclear. A minority of students acknowledged awareness of formal teaching sessions about conflicts of interest with the pharmaceutical industry, and this awareness had no impact on their attitudes. However, the years immediately preceding this survey have been marked by substantial national upheaval regarding the role of the pharmaceutical industry in academic medical centers. Nationally, the American Medical Student Association (AMSA), which had 68,000 members in 2006, was quite vocal in their concerns regarding relationships with drug companies. In 2001, they developed a comprehensive PharmFree policy, which proposed banning all pharmaceutical advertising and sponsorship, and they launched the No Free Lunch Campaign. By 2005, AMSA launched a counter-detailing initiative. In 2006, they undertook a national survey of medical school policies on relationships with pharmaceutical companies, which resulted in a PharmFree Scorecard in 2007, ranking medical schools based on their policies to limit the access of pharmaceutical companies and representatives at academic teaching centers (12, 13). Approximately half of OHSU medical students belong to AMSA (Molly Osborne, personal communication). After mounting controversy nationally regarding the role of the pharmaceutical industry, the influential paper by Brennan et al. called for the elimination of even small gifts and funds for physician travel and speaker’s bureaus, and modification of industry contribution to CME (2). Also, there was a strong proposal by OHSU house-staff to restrict the influence of pharmaceutical companies. Clinical medical students may have had more exposure to concerns about the pharmaceutical industry through interactions with residents. If so this presents an interesting example of the effects of a hidden or informal curriculum (14). Grande et al. recently examined clinical medical student exposure to promotional items at two U.S. medical schools with differing policies toward pharmaceutical industry marketing (15). In a randomized, controlled manner, clinical medical students were exposed to small, branded items from one of two companies. Differences in unconscious or implicit attitudes toward two brand-name medications were then measured. In the school with less-restrictive policies, there was a significantly more favorable attitude among medical students for the medication that received promotional item exposure. In the school with more-restrictive policies, however, the reverse was seen, with medical students having a less-favorable attitude toward the branded medication (15).

A strength of this study is the high response proportion. Limitations include the fact that it was conducted at only one medical school, and therefore unlikely to be representative of all U.S. medical schools; The data are now 5 years old; the brevity of the questionnaire did not allow us to standardize our definitions; and, because the information is by self-report, it may not reflect actual behaviors. It does suggest that acknowledgment and awareness of conflicts of interest around relationships with PRs can lead to changes in attitudes, which results in a reduced likelihood of acceptance of gifts. However, institutional factors that promote attendance of drug company-sponsored activities may overwhelm ethical concerns.

Support for the survey was through the Office of the Dean, School of Medicine, OHSU. Also, this material is the result of work supported with resources and the use of facilities at the Portland Veterans Affairs Medical Center Health Services Research and Development Research Enhancement Award Program. Biostatistical support was provided by Oregon Clinical and Translational Research Institute, OHSU, Grant ULIRR024140.

The need for written informed consent was waived by the Institutional Review Board at OHSU.

The views expressed in this manuscript are those of the authors and do not necessarily reflect the position of policy or the Department of Veterans Affairs or the U.S. Government.