Physician–industry interactions are common.1,2 For example, Campbell et al1 reported that physicians met with pharmaceutical representatives twice per month on average in 2009. Physicians, however, are prone to denying that such interactions could influence them. For instance, six studies included in a review published in 2000 showed that the majority of physicians denied that these interactions influenced their prescribing practices.2 More recently, in a focus-group-based study, Chimonas et al3 found that many physicians either denied being influenced by gifts or rationalized accepting gifts from pharmaceutical representatives.
Like physicians, medical students are exposed to drug company interactions, and their attitudes about these interactions parallel physicians’ attitudes.4–6 Specifically, in a 2003 survey of third-year students at eight U.S. medical schools, we found that students were exposed to a gift or sponsored presentation once per week on average, and 68.8% denied that their future prescribing could be influenced by gifts.4 In a 2011 national survey, Austad et al5 found that 56.8% of senior medical students received industry-sponsored gifts. Some of the similarity in students’ and physicians’ behaviors and attitudes likely develops during supervised experiences in clerkships and preceptorships. For example, in our 2003 survey, most students (93.2%) reported being asked or required by a physician to attend at least one drug-company-sponsored lunch.4
The more that students and physicians interact with a drug company’s representatives, the more they are exposed to information that is biased in favor of the company’s products.7–19 Specifically, bias favoring sponsors’ products has been found in journal articles,7–11 advertisements,12–16 pharmaceutical representative visits,17–19 and meeting abstracts20—all of which are major sources of information for students and physicians. A content analysis of two continuing medical education (CME) presentations published in 1986 also documented this bias.21 (No subsequent CME content analyses have been published to the best of our knowledge.) Factors contributing to this bias include withholding negative results and selecting less ill patients in trials.9,11,22,23
Clinicians’ exposure to this information increases the likelihood that they will prescribe the sponsor’s drug, despite evidence favoring other drugs or treatments, or whether the drug is best for the patient.11,24,25 This leads to suboptimal and costlier (e.g., on-patent versus generic) prescribing practices and poorer patient health.11,24,25
Industry–physician ties also lead to conflicts of interest (COIs). Addressing and minimizing COIs is central to professionalism.1 As research about industry–physician interactions has expanded, concerns about COIs have intensified. These concerns have been raised in books,11,26,27 journals,7,28 newspapers,19,29,30 and U.S. Senate investigations,30 as well as by activist groups like NoFreeLunch31 and PharmedOut.32 Organizations including the Institute on Medicine as a Profession (IMAP),28,33,34 the Association of American Medical Colleges (AAMC),35 and the Institute of Medicine (IOM)36 have called for academic health centers to develop rules and teach physicians and students about industry interactions, while the American Medical Student Association (AMSA) has developed a scorecard to rate schools’ COI policies.37 Representatives from the drug companies Pfizer and Eli Lilly participated in the AAMC task force on industry funding of medical education.35 Soon after the task force’s report was released in 2008, the Pharmaceutical Research and Manufacturers of America recommended limiting meals for doctors’ spouses or guests, entertainment or recreational items, and gifts of minimal value like pens and mugs.38 From 1980 to 2004, MEDLINE-indexed publications on COI progressively increased each year from 5 new publications in 1980 to 525 new publications in 2006.39 Similarly, the number of schools with policies about industry–physician and industry–student interactions increased from 10 in 20054,40 to 127 by 2011.37,41
Six studies4,5,42–46 have endeavored to link medical school or residency program rules restricting industry–trainee interactions with concurrent trainee or future physician behaviors. Five of these studies4,42–46 suggested that restrictive rules were associated with less frequent industry contacts or more skeptical attitudes. For example, Epstein et al42 found that after psychiatry residency programs instituted COI policies, their postpolicy graduates were less likely to prescribe heavily promoted antidepressants as compared with their prepolicy graduates. In contrast, Austad et al5 found no relationship between AMSA scorecard ratings and student receipt of gifts. Other studies showed that teaching about drug company interactions led to more skeptical student attitudes.47,48
Campbell et al1 reported a modest but statistically significant decline in industry interactions among U.S. physicians from 2004 to 2009. To examine the impact of increased national attention on COIs and of published recommendations about industry–student interactions, in 2012 we initiated a follow-up study on drug company interactions among third-year students at the eight U.S. medical schools we studied in 2003.4 Although a national survey on students’ exposures to and attitudes about drug company interactions was conducted as recently as 2011,5 to our knowledge this is the first multi-institutional follow-up survey. Our objectives were to ascertain whether changes had occurred in student exposure to and attitudes about drug company interactions, what factors (e.g., restrictive policies, role modeling, site assignments) influenced student exposure and attitudes, and whether exposure and attitudes influenced future plans to interact with drug companies.
This is a 2012 follow-up study conducted at the same eight U.S. medical schools we studied in 20034 (Case Western Reserve University School of Medicine; George Washington University School of Medicine and Health Sciences; Mayo Medical School; Ohio State University College of Medicine; Rosalind Franklin University of Medicine and Science; State University of New York Upstate Medical University; University of California, San Francisco, School of Medicine; and University of Nebraska College of Medicine). As in 2003, we surveyed third-year students exclusively so that students’ reported behaviors would be related solely to their own school culture, as many fourth-year students take electives at other schools. We selected these schools nonrandomly in 2003 on the basis of our access as faculty members to students (i.e., a convenience sample), as well as diversity of ownership (i.e., public and private schools), National Institutes of Health (NIH) funding levels,49 and geographic locations (see Table 1 for school characteristics).
The 2012 77-item, two-page questionnaire was modified from our 2003 64-item questionnaire. (For the full 2012 questionnaire, see Supplemental Digital Appendix 1 at http://links.lww.com/ACADMED/A265.) It included 50 items from the 2003 survey and 27 new items added on the basis of a review of the literature45,50 and our previous findings.4 The 2012 survey was pilot-tested on two groups of senior medical students.
The survey’s cover page or e-mailed invitation stated the study’s purpose and indicated that participation was anonymous and optional. No incentives were offered for participation.
The questionnaire elicited information about age, gender, school, third-year clerkship start date, survey completion date, perception of whether the school had a policy or taught about industry–student interactions, and exposure to and attitudes about drug company interactions. We repeated 2003 survey items which had yielded important findings4 and developed new items to expand or explain those findings. The new items in the 2012 survey elicited information about perceptions of clerkship sites with the most pharmaceutical representative–student interactions; best role models with regard to pharmaceutical representative interactions; plans to interact with pharmaceutical representatives during residency and preferences for residency program restrictions; observations of physicians chatting with pharmaceutical representatives or disclosing industry relationships; and experiences of listening to a pharmaceutical representative discuss his or her company’s drugs or approaching a pharmaceutical representative for information.
At each school, the local author (F.S.S., M.M., G.B., S.S., D.J.L., J.C., L.M.C., W.S.), a clinical educator known by third-year students and clerkship directors, obtained institutional review board exemption and then distributed the survey to all third-year students. Distribution occurred between December 2011 and June 2012. At five schools, the author distributed Scantron forms (Scantron, Eagan, Minnesota) at large meetings with required attendance (e.g., exam, dean’s forum). At three schools, the author e-mailed invitations linked to an online survey. At one of these three schools, the invitation was sent before the author’s class about industry–student interactions. At the second, half of the students were absent when a Scantron form was distributed, so an invitation was e-mailed as well. At the third, which prohibited hard copy distribution, the author e-mailed three successive invitations.
For the eight types of exposure that had occurred frequently in 20034 (see Table 2), we asked students in 2012 to report how many times they had been exposed to drug companies in these ways since starting their clerkships. To identify monthly frequencies of these interactions for each student, we divided total exposures by the reported number of months in clerkships (each student’s survey completion date minus his or her clerkship start date). We compared our 2003 and 2012 findings for each type of interaction.
For students who completed all eight of these items, we summed frequencies to create a monthly exposure index to facilitate statistical associations among exposure, skepticism, and appropriateness indexes and other variables.
For the six types of exposure that had occurred less frequently in 2003,4 we asked students in 2012 whether any exposure occurred, and compared our 2003 and 2012 findings.
Acceptance and skepticism
Students responded to seven items soliciting attitudes about drug company interactions and marketing, using a four-point Likert scale (strongly agree = 1, agree = 2, disagree = 3, and strongly disagree = 4) (see Table 3). The students’ responses reflect accepting or skeptical attitudes. For the first four items, agreement suggests an acceptance of interactions and marketing, and disagreement suggests skepticism. For the other three items, agreement suggests skepticism, and disagreement suggests acceptance. For students completing all seven items, we reversed the scores for the last three items and summed the points to create that student’s skepticism index. To compare our 2012 findings with our 2003 findings, we compared the mean scores as well as the proportions of students who agreed and disagreed with each item.
Students indicated their perception of the appropriateness of eight types of drug company gifts (see Table 4), using a five-point Likert scale (very appropriate = 1, appropriate = 2, neutral = 3, inappropriate = 4, and very inappropriate = 5).45 For each student completing all eight items, we summed the student’s appropriateness scores to create that student’s appropriateness index. To compare our 2012 findings with our 2003 findings, we compared mean scores for each item as well as proportions of students who agreed and disagreed with each item.
Medical school policy
To obtain ratings of each medical school’s policy about drug company–student and drug company–physician relationships, we consulted two sources. One source was AMSA’s 2012 Pharm-Free Scorecard,37 which provided a global rating of the restrictiveness of the policy for each school (A–F, where A = excellent, B = good, C = passable, D = discussions of a forthcoming policy, and F = no policy, no discussions). The other source was IMAP’s 2011 COI Policy Database,33,41 which documented whether each school had a policy that addressed 11 areas (gifts, meals, samples, vendor access, pharmacy committees, CME, consulting, honoraria, travel, scholarships, and ghostwriting) and scored each area as follows: 0 = no policy, 1 = permissive policy, 2 = moderate policy, and 3 = stringent policy. To determine each school’s IMAP score, we summed its scores for the 11 areas. For the eight schools in this study, correlation between AMSA’s and IMAP’s ratings was r = 0.713, P = .047, df = 6. Neither the AMSA nor IMAP source addressed policy implementation or enforcement.
Anticipated interactions during residency
In 2012, we asked students whether they planned to interact with pharmaceutical representatives during residency and if they had a preference for a residency program that restricted interactions.
Clerkship time spent in private outpatient offices
Apart from the student survey, the author at each school asked each third-year clerkship director and the associate dean who oversees the third-year curriculum, “What proportion of your third-year students’ total clerkship time is spent in private outpatient offices?” To compute the numerator for the proportion of time spent by respondents in private outpatient offices, we multiplied the number of respondents at each school by the percentage of time that school’s students spent in private outpatient offices and summed those eight products. The denominator was the total number of survey respondents.
Scantron and online survey data were converted to Microsoft Excel (Microsoft, Redmond, Washington) and imported into SPSS version 19 (IBM Corp., Armonk, New York). Pearson product–moment correlations were used to characterize associations among the various indexes and plans to interact with pharmaceutical representatives during residency. Independent sample t tests (in SPSS) and Z tests for proportions of independent groups (VassarStats, vassarstats.net) were used to compare our 2012 and 2003 results and to compare the 2012 student perceptions about locations having the most pharmaceutical representative interactions with the proportions of students’ time spent in private outpatient offices as reported by the clerkship directors. In the Z tests comparing our 2012 and 2003 results, Pa − Pb was the difference in the proportion of students who responded the same to a particular question in 2012 versus 2003. For one of the nine comparisons of 2003 versus 2012 monthly exposures, and for one of the eight comparisons of 2003 versus 2012 perceptions of gift appropriateness, we used Pearson chi-squared tests. P values were two-tailed, with P < .05 being the level set for statistical significance.
In 2012, 866 of the 1,269 third-year students at the eight included medical schools responded to the survey, for an overall response rate of 68.2%, with a range among schools of 22.7% to 94.5% (Table 1). In 2003, the overall response rate was 826/1,143 (72.3%; range = 30.9%–90.7%), which was significantly higher than the 2012 response rate (Pa − Pb = 0.0402, Z = 2.156, P < .04). In 2012, the mean age of respondents was 26.4 (n = 839, standard deviation [SD] = 3.3, range = 21–46). In 2003, the mean age of respondents was 26.3 (n = 805, SD = 2.9, range = 22–47). Among respondents reporting gender in 2012, 409/826 (49.5%) were women. This proportion was not significantly different from the 2012 proportion of female second- and third-year students at all U.S. medical schools (17,988/37,920 [47.4%]; Pa − Pb = −0.0208, Z = −1.184, P = .24),51 or the proportion of female respondents among the third-year students we surveyed in 2003 (352/782 [45.0%]; Pa − Pb = −0.045, Z = −1.808, P = .07).4 The total number of third-year students at the eight schools increased from 1,143 in 2003 to 1,269 in 2012, reflecting class size increases at Case Western Reserve University School of Medicine, Mayo Medical School, Rosalind Franklin University of Medicine and Science, and the University of Nebraska College of Medicine. The other schools’ class sizes were unchanged.
Exposure to drug company interactions
Each of the eight exposure types that had occurred frequently in 2003 occurred significantly less frequently in 2012, and a significantly lower proportion of students was exposed to each gift and interaction type in 2012 than in 2003 (Table 2). Students’ mean monthly exposures were 1.6 exposures per month in 2012 (n = 845, SD = 2.2, range = 0–23.3), an average of one time every 2.4 weeks, compared with 4.1 exposures per month in 2003, or an average of one time per week. In 2012, 184/860 (21.4%) of the students neither received a gift nor attended a sponsored event, compared with 0/821 (0.0%) in 2003. Despite across-the-board reductions in exposures from 2003 to 2012, most students in 2012 had eaten drug-company-sponsored lunches (548/854 [64.2%]). Also, 361/861 (41.9%) had received a reprint or brochure, and 300/860 (34.9%) had eaten a snack provided by a drug company. In 2012, significantly fewer students (255/860 [29.7%]) were asked or required by a resident or attending physician to attend a lunch or dinner provided by a drug company than were asked in 2003 (762/818 [93.2%]; Pa − Pb = 0.6362, Z = 26.652, P < .001).
For five of the six exposure types that had been less common in 2003, there was significantly less exposure in 2012 than in 2003, as follows: receiving a book donated by a drug company (421/826 [51.0%] in 2003 versus 38/866 [4.4%] in 2012; Pa − Pb = 0.4658, Z = 21.541, P < .001), attending a workshop sponsored by a drug company (214/826 [25.9%] in 2003 versus 38/866 [4.4%] in 2012; Pa − Pb = 12.428, Z = 12.428, P < .001), having a registration fee for a conference paid by a drug company (37/826 [4.5%] in 2003 versus 6/866 [0.7%] in 2012; Pa − Pb = 0.0379, Z = 4.947, P < .001), participating in a market survey conducted by a drug company (29/826 [3.5%] in 2003 versus 5/866 [0.6%] in 2012; Pa − Pb = 0.0293, Z = 4.298, P < .001), and participating in a research project sponsored by a drug company (22/826 [2.7%] in 2003 versus 6/866 [0.7%] in 2012; Pa − Pb = 0.0197, Z = 4.298, P < .001). In 2012, there was less exposure than in 2003 to the sixth item, attending a conference with travel expenses paid for by a drug company, but the difference was not significant (15/826 [1.8%] in 2003 versus 7/866 [0.8%] in 2012; Pa − Pb = .0101, Z = 1.829, P = .07).
Regarding the exposure types we first included in the 2012 survey, 555/863 (64.3%) of the students had listened to a pharmaceutical representative discuss a drug; 636/863 (73.7%) had observed an attending or resident chatting with a pharmaceutical representative; 447/864 (51.7%) had heard a lecturer, supervisor, or teacher disclose a relationship with a drug company; and 149/863 (17.3%) had approached a pharmaceutical representative for information.
Student perceptions about schools’ policies and teaching
In 2012, significantly more students (419/859 [48.7%]) responded affirmatively to “My school has a policy. The policy is that there are limits or restrictions on pharmaceutical representative–student interactions” than responded affirmatively in 2003 (33/822 [4.0%]; Pa − Pb = −0.4476, Z = −20.691, P < .001). Because all but one of the eight schools in 2012 had restrictive policies (AMSA rating of A, B, or C; see Table 1), this means that 310/703 (44.1%) of the students in 2012 were unaware of their school’s restrictive policy. Also, in 2012, significantly more students (530/862 [61.5%]) responded affirmatively to the item asking whether they had received any formal presentations on drug company–doctor or drug company–student relationships than in 2003 (311/821 [37.9%]; Pa − Pb = −0.236, Z = −9.681, P < .001).
Attitudes about drug company interactions and marketing
Significantly more students were skeptical about, and significantly fewer were accepting of, drug company interactions and marketing in 2012 compared with their 2003 counterparts (Table 3). For example, students in 2012 were less likely than students in 2003 to feel entitled to gifts (344/823 [41.8%] versus 533/664 [80.3%], P < .001) and were more apt to feel that gifts could influence them (364/822 [44.3%] versus 252/808 [31.2%], P < .001). Yet, as in 2003,4 many students in 2012 simultaneously held contradictory attitudes. For example, among 630 students who perceived that drug-company-sponsored grand rounds were biased in favor of company products, 419 (66.5%) simultaneously believed that drug-company-sponsored grand rounds were helpful and educational. For each gift type, significantly fewer 2012 than 2003 students believed that accepting the gift was appropriate (Table 4). For seven of the eight types of gifts listed in Table 4, a minority of students in 2012 thought the gift was appropriate. However, the majority (487/833; 58.5%) considered meals appropriate.
Appropriateness and skepticism
The correlation between the skepticism and appropriateness indexes was r = 0.535 (n = 738, P < .001). Thus, the more skeptical students were less likely to think that gifts were appropriate.
Skepticism and exposure
The skepticism and exposure indexes had a significant inverse correlation (r = −0.149, n = 738, P < .001), meaning that the more skeptical students were exposed less often. The strength of this inverse correlation between skepticism and exposure was low because behaviors often contradicted the attitudes reported. For example, among 126 students who considered meals inappropriate, 54 (42.9%) had nonetheless eaten a drug-company-sponsored lunch.
The role models selected most often by the 827 students reporting “best role models” for pharmaceutical representative interactions were attending physicians (548 [66.3%]), followed by residents (201 [24.3%]), deans or department chairpersons (164 [19.8%]), physician friends (163 [19.7%]), physician family members (129 [15.6%]), and fellow students (123 [14.9%]).
Locations of greatest exposure
In 2012, students reported that their greatest pharmaceutical representative exposure occurred in private outpatient offices (545/839; 65.0%), even though clerkship directors indicated that only 159.7/866 (18.4%) of students’ clerkship time was spent in private outpatient offices (Pa − Pb = 0.466, Z = 19.537, P < .001). In addition, in 2012, 160/839 (19.1%) of the students reported that their greatest exposure to pharmaceutical representatives occurred at an affiliated community hospital/medical center, and 134/839 (16.0%) reported that their greatest exposure was at one of the university’s main hospitals/medical centers.
Anticipated future interactions
In 2012, 467/837 students (55.8%) responded affirmatively to “During my residency, I plan to interact with drug reps,” and 410/832 (49.2%) agreed or strongly agreed with “I would prefer a residency that restricts interactions between residents and drug reps.” Correlation between these variables was r = −0.451 (n = 802, P < .001), indicating that students planning to interact with pharmaceutical representatives were less likely to prefer residencies with restrictions.
Relationships between exposure, skepticism, appropriateness, and anticipated interactions
Correlation between students’ monthly exposure indexes and agreement with “During my residency, I plan to interact with drug reps” was r = −0.185 (n = 804, P < .001), indicating that the more often students interacted with pharmaceutical representatives, the more likely they were to plan to interact with them. Correlation between students’ skepticism indexes and agreement with “I plan to interact” was r = 0.478 (n = 739, P < .001), indicating that the more skeptical students were, the less likely they were to plan to interact with pharmaceutical representatives. Correlation between students’ appropriateness indexes and agreement with “I plan to interact” was r = 0.480 (n = 797, P < .001), meaning that the more appropriate they perceived gifts were, the more likely they were to plan to interact with pharmaceutical representatives.
Relationships between student perceptions of school policies, skepticism, appropriateness, and exposures
Students who indicated that their school had a policy limiting or restricting interactions between pharmaceutical representatives and medical students had higher mean skepticism and appropriateness indexes than students who did not perceive that their school had a restrictive policy: skepticism, n = 360, mean (SD) = 18.7 (3.5) versus 17.4 (2.8), t = 5.699, P < .001; appropriateness, n = 393, mean (SD) = 28.7 (6.8) versus 27.1 (7.1), t = 3.221, P = .001. Thus, students who perceived that their school had a restrictive policy were more likely than others to hold skeptical attitudes toward drug company interactions and to think that gifts were inappropriate. However, there was no association between student perceptions of school policy and the extent of their exposure: n = 413, mean (SD) = 11.2 (3.6) versus 11.6 (3.4), t[1,842] = 1.656, P = .10.
In this study, we examined third-year students’ exposure to and attitudes about drug company interactions at eight U.S. medical schools, comparing our 2012 survey findings with our 2003 findings.4 We also investigated factors contributing to student exposure to, attitudes about, and future plans to interact with drug companies.
We found that third-year students were less likely to report exposure to drug company interactions in 2012 than they were in 2003, paralleling Campbell and colleagues’1 findings about declines in physician exposure from 2004 to 2009. Compared with students in 2003, students in 2012 were more skeptical about and less accepting of drug company interactions and marketing, more likely to perceive that they had been taught about pharmaceutical representative interactions, and more likely to perceive that their schools had restrictive policies in place. This is consistent with heightened awareness of and concerns about COIs among the public and in academic medicine. One outcome of these concerns is that in 2012, seven of the eight schools in this study had restrictive policies with AMSA ratings of A, B, or C,37 whereas in 2003 only two of the schools had restrictive policies.4
However, despite the restrictive policies in place at seven schools in 2012, a majority of students still believed that free meals were appropriate. Also, a majority of students—particularly those with less skeptical attitudes—planned to interact with pharmaceutical representatives during residency. Many students’ continuing exposure to and accepting attitudes about drug company interactions suggest that simply publishing rules is insufficient to ensure that students will avoid or be skeptical about interactions. Some of the above findings can be explained by the influence of key role models and students’ low level of awareness of their schools’ restrictive policies.
Another explanation is the dispro portionately large exposure to pharma ceutical representatives occurring in private outpatient offices, which most schools rely upon heavily for outpatient assignments. One reason for the disproportionate exposure in this type of setting may be that organizations such as the AAMC,35 IOM,36 and IMAP41 have focused on developing recommendations regarding policies and instruction about interactions in academic medical centers, perhaps with the expectation that these interventions would eventually spread to volunteer faculty in private outpatient offices.
This study has some limitations. Some generalizability of our study to the total U.S. medical student population is reflected by the diversity of the eight included schools in terms of geographic location, ownership, NIH funding, and IMAP ratings, and by the similarity of the proportion of our 2012 female survey respondents and the 2012 proportion of female second- and third-year students at all U.S. medical schools.51 Nevertheless, as in 2003, we studied only eight schools, which we did not randomly select, and using a convenience sample raises questions about study generalizability. For example, none of our schools is from the South.
In addition, because student responses about prior exposures were often made several months after the exposures occurred, they are susceptible to inaccurate recall. Nevertheless, students’ self-reported exposures were to simple, salient, and relatively recent events. Also, because our study was neither controlled nor experimental, we could not precisely determine the extent to which each type of intervention contributed to the decline in exposure to and increase in skepticism about drug company interactions from 2003 to 2012, or which strategies best influence these outcomes.
Our findings have implications for medical trainees, educators of medical students and residents, deans, compliance officers, and accrediting bodies. Medical schools and residency programs that hope to increase awareness of industry influence, promote skeptical attitudes, and reduce trainees’ current and future interactions with drug companies—in compliance with AAMC,35 IOM,36 IMAP,41 and AMSA37 recommendations—should continue to look for opportunities to influence attitudes and limit exposure among students and their physician role models. To these ends, we believe that the following interventions should be considered and monitored by curriculum committees; course, clerkship, and residency directors; deans; compliance officers; the AMSA; IMAP; the Liaison Committee on Medical Education3; and residency review committees.
Medical schools and residency programs should establish learning objectives and educate trainees about drug-company-sponsored research and marketing, industry–physician and industry–trainee interactions, COIs, and methods of reducing COIs.35,47,48 They should assess fulfillment of these objectives through course (e.g., ethics, professionalism, pharmacology) and National Board of Medical Examiners examinations, anonymous surveys of learners’ behaviors,4,5 the AAMC Medical School Graduation Questionnaire,52 and studies of graduates’ prescribing practices.42,46
Not only should medical schools and residency programs establish restrictive policies about industry–physician and industry–trainee interactions, they should also consider enforcing their policies more than they have previously. Policies and their rationales should be presented, in person and in writing, starting early in trainees’ careers. These policies and rationales should be revisited and enforced throughout students’ and residents’ training and clinical and academic careers. This process should include supervisory reviews of clinician prescribing practices and faculty development programs. In addition, faculty members should disclose industry funding to student as well as physician audiences.53
One approach to the disproportionately high exposure to pharmaceutical representatives in private outpatient offices might be to engage practitioners in these settings in developing and adopting restrictive policies and in teaching and learning about COIs, with the aim of encouraging them to become role models for refusing gifts. As our experience indicates that volunteer faculty may resist these efforts, we recommend that the educational interventions described above be delivered to students before and after their rotations in these settings.
Considering that almost half of the students in our 2012 study indicated that they would prefer their residencies to restrict interactions with pharmaceutical representatives, residency program Web sites should summarize such policies to foster trainee–residency compatibility on this issue.
In summary, from 2003 to 2012, there was a decrease in third-year medical students’ exposure to, and an increase in their skepticism about, drug company interactions. However, given that interactions still occur, particularly in private outpatient offices, and many students plan to interact with pharmaceutical representatives during residency, continued efforts to study and to influence students’ and physician role models’ exposures to and attitudes about drug companies are warranted.
Acknowledgments: The authors thank Rachel Greenlee and Paul Hung for statistical advice.
1. Campbell EG, Rao SR, DesRoches CM, et al. Physician professionalism and changes in physician–industry relationships from 2004 to 2009. Arch Intern Med. 2010;170:1820–1826
2. Wazana A. Physicians and the pharmaceutical industry: is a gift ever just a gift? JAMA. 2000;283:373–380
3. Chimonas S, Brennan TA, Rothman DJ. Physicians and drug representatives: exploring the dynamics of the relationship. J Gen Intern Med. 2007;22:184–190
4. Sierles FS, Brodkey AC, Cleary LM, McCurdy FA, Mintz M, Frank J, et al. Medical students’ exposure to and attitudes about drug company interactions: a national survey. JAMA. 2005;294:1034–1042
5. Austad KE, Avorn J, Franklin JM, Kowal MK, Campbell EG, Kesselheim AS. Changing interactions between physician trainees and the pharmaceutical industry: a national survey. J Gen Intern Med. 2013;28:1064–1071
6. Austad KE, Avorn J, Kesselheim AS. Medical students’ exposure to and attitudes about the pharmaceutical industry: a systematic review. PLoS Med. 2011;8:e1001037
7. Bekelman JE, Li Y, Gross CP. Scope and impact of financial conflicts of interest in biomedical research: a systematic review. JAMA. 2003;289:454–465
8. Lexchin J, Bero LA, Djulbegovic B, Clark O. Pharmaceutical industry sponsorship and research outcome and quality: systematic review. BMJ. 2003;326:1167–1170
9. McGauran N, Wieseler B, Kreis J, Schüler YB, Kölsch H, Kaiser T. Reporting bias in medical research - a narrative review. Trials. 2010;11:37
10. Montori VM, Devereaux PJ, Adhikari NK, Burns KE, Eggert CH, Briel M, et al. Randomized trials stopped early for benefit: a systematic review. JAMA. 2005;294:2203–2209
11. Goldacre B. Bad Pharma: How Drug Companies Mislead Doctors and Harm Patients. 2013 London, England Fourth Estate
12. Spielmans GI, Thielges SA, Dent AL, Greenberg RP. The accuracy of psychiatric medication advertisements in medical journals. J Nerv Ment Dis. 2008;196:267–273
13. Othman N, Vitry A, Roughead EE. Quality of pharmaceutical advertisements in medical journals: a systematic review. PLoS One. 2009;4:e6350
14. Gilbody S, Wilson P, Watt I. Benefits and harms of direct to consumer advertising: a systematic review. Qual Saf Health Care. 2005;14:246–250
15. Frosch DL, Krueger PM, Hornik RC, Cronholm PF, Barg FK. Creating demand for prescription drugs: a content analysis of television direct-to-consumer advertising. Ann Fam Med. 2007;5:6–13
16. Del Signore A, Murr AH, Lustig LR, Platt MP, Jalisi S, Pratt LW, et al. Claim validity of print advertisements found in otolaryngology journals. Arch Otolaryngol Head Neck Surg. 2011;137:746–750
17. Ziegler MG, Lew P, Singer BC. The accuracy of drug information from pharmaceutical sales representatives. JAMA. 1995;273:1296–1298
18. Fugh-Berman A, Ahari S. Following the script: how drug reps make friends and influence doctors. PLoS Med. 2007;4:e150
20. Fries JF, Krishnan E. Equipoise, design bias, and randomized controlled trials: the elusive ethics of new drug development. Arthritis Res Ther. 2004;6:R250–R255
21. Bowman MA. The impact of drug company funding on the content of continuing medical education. Möbius (J Cont Educ Health Prof Health Sci). 1986;6:66–69
22. Turner EH, Matthews AM, Linardatos E, Tell RA, Rosenthal R. Selective publication of antidepressant trials and its influence on apparent efficacy. N Engl J Med. 2008;358:252–260
23. Melander H, Ahlqvist-Rastad J, Meijer G, Beermann B. Evidence b(i)ased medicine–selective reporting from studies sponsored by pharmaceutical industry: review of studies in new drug applications. BMJ. 2003;326:1171–1173
24. Spurling GK, Mansfield PR, Montgomery BD, Lexchin J, Doust J, Othman N, et al. Information from pharmaceutical companies and the quality, quantity, and cost of physicians’ prescribing: a systematic review. PLoS Med. 2010;7:e1000352
25. Gagnon MA, Lexchin J. The cost of pushing pills: a new estimate of pharmaceutical promotion expenditures in the United States. PLoS Med. 2008;5:e1
26. Angell M. The Truth About the Drug Companies: How They Deceive Us and What to Do About It. 2004 New York, NY Random House
27. Brody H. Hooked: Ethics, the Medical Profession, and the Pharmaceutical Industry. 2007 Lanham, MD Rowman & Littlefield
28. Brennan TA, Rothman DJ, Blank L, Blumenthal D, Chimonas SC, Cohen JJ, et al. Health industry practices that create conflicts of interest: a policy proposal for academic medical centers. JAMA. 2006;295:429–433
32. Fugh-Berman AJ, Scialli AR, Bell AM. Why lunch matters: assessing physicians’ perceptions about industry relationships. J Contin Educ Health Prof. 2010;30:197–204
33. Chimonas S, Patterson L, Raveis VH, Rothman DJ. Managing conflicts of interest in clinical care: a national survey of policies at U.S. medical schools. Acad Med. 2011;86:293–299
34. ABIM Foundation, American College of Physicians Foundation, European Federation of Internal Medicine. . Medical professionalism in the new millennium: a physician charter. Ann Intern Med. 2002;136:243–246
35. Association of American Medical Colleges. . Industry Funding of Medical Education: Report of an AAMC Task Force. 2008 Washington, DC Association of American Medical Colleges
36. Institute of Medicine. . Conflict of Interest in Medical Research, Education, and Practice. 2009 Washington, DC National Academies Press
40. Sierles F, Brodkey A, Cleary L, McCurdy FA, Mintz M, Frank J, et al. Relationships between drug company representatives and medical students: medical school policies and attitudes of student affairs deans and third-year medical students. Acad Psychiatry. 2009;33:478–483
41. Chimonas S, Evarts SD, Littlehale SK, Rothman DJ. Managing conflicts of interest in clinical care: the “race to the middle” at U.S. medical schools. Acad Med. 2013;88:1464–1470
42. Epstein AJ, Busch SH, Busch AB, Asch DA, Barry CL. Does exposure to conflict of interest policies in psychiatry residency affect antidepressant prescribing? Med Care. 2013;51:199–203
43. McCormick 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
44. Grande D, Frosch DL, Perkins AW, Kahn BE. Effect of exposure to small pharmaceutical promotional items on treatment preferences. Arch Intern Med. 2009;169:887–893
45. Brotzman GL, Mark DH. The effect on resident attitudes of regulatory policies regarding pharmaceutical representative activities. J Gen Intern Med. 1993;8:130–134
46. King M, Essick C, Bearman P, Ross JS. Medical school gift restriction policies and physician prescribing of newly marketed psychotropic medications: difference-in-differences analysis. BMJ. 2013;346:f264
47. Kao AC, Braddock C 3rd, Clay M, Elliott D, Epstein SK, Filstead W, et al. Effect of educational interventions and medical school policies on medical students’ attitudes toward pharmaceutical marketing practices: a multi-institutional study. Acad Med. 2011;86:1454–1462
48. Wofford JL, Ohl CA. Teaching appropriate interactions with pharmaceutical company representatives: the impact of an innovative workshop on student attitudes. BMC Med Educ. 2005;5:5
50. Sierles FS. How to do research with self-administered surveys. Acad Psychiatry. 2003;27:104–113
51. Barzansky B, Etzel SI. Medical schools in the United States, 2011-2012. JAMA. 2012;308:2257–2263
52. Association of American Medical Colleges. . Medical School Graduation Questionnaire (GQ). www.aamc.org/data/gq
. Accessed January 4, 2015
53. Austad KE, Kesselheim AS. Conflict of interest disclosure in early education of medical students. JAMA. 2011;306:991–992