If you feed them, they will come: A prospective study of the effects of complimentary food on attendance and physician attitudes at medical grand rounds at an academic medical center

Background Evidence suggests that attendance at medical grand rounds at academic medical centers is waning. The present study examined whether attendance at medical grand rounds increased after providing complimentary food to attendees and also assessed attendee attitudes about complimentary food. Methods In this prospective, before-and-after study, attendance at medical grand rounds was monitored from September 25, 2002, to June 2, 2004, using head counts. With unrestricted industry (eg, pharmaceutical) financial support, complimentary food was provided to medical grand rounds attendees beginning June 4, 2003. Attendance was compared during the pre-complimentary food and complimentary food periods. Attitudes about the complimentary food were assessed with use of a survey administered to attendees at the conclusion of the study period. Results The mean (± SD) overall attendance by head counts increased 38.4% from 184.1 ± 90.4 during the pre-complimentary food period to 254.8 ± 60.5 during the complimentary food period (P < .001). At the end of the study period, 70.1% of the attendee survey respondents indicated that they were more likely to attend grand rounds because of complimentary food, 53.6% indicated that their attendance increased as a result of complimentary food, and 53.1% indicated that their attendance would decrease if complimentary food was no longer provided. Notably, 80.3% indicated that food was not a distraction, and 81.7% disagreed that industry representatives had influence over medical grand rounds because of their financial support for the food. Conclusion Providing free food may be an effective strategy for increasing attendance at medical grand rounds.


Background
Medical grand rounds (MGR) is a central teaching activity in US departments of medicine at academic medical centers. However, attendance at MGR by faculty, fellows, and residents appears to be waning even though the perceived quality of MGR is increasing [1][2][3]. This decrease in attendance, coupled with the considerable resources (eg, financial resources and time) invested in this activity, should cause MGR planners and continuing medical education (CME) providers to question why attendance is waning. Moore [4] described a 6-level approach for evaluating the value of CME. Each level is associated with an outcome: participation/attendance, satisfaction, learning, performance, patient health, and population health. The present study focuses on measuring attendance as an outcomes measure. On the basis of Moore's model, increasing attendance is a legitimate CME goal.
One commonly used strategy to increase attendance at MGR is to provide complimentary food [3]. Observational and anecdotal data indicate that complimentary food can increase attendance at meetings and educational events [3,[5][6][7][8]. Notably, needs assessments at our institution identified lack of food as a barrier to attendance at MGR [2]. Complimentary food has also been associated with successful, well-attended internal medicine journal clubs [9].
Nevertheless, providing complimentary food requires substantial financial resources. However, before financial resources are allocated or sought from outside sources to provide free food, a study examining the effects of complimentary food on attendance at MGR should be undertaken. Attendee attitudes should also be assessed to evaluate whether this intervention changes one's attitude toward MGR [8]. To date, no systematic study has examined whether complimentary food increases attendance at MGR or what effect complimentary food has on attendee attitudes toward MGR.
MGR at Mayo Clinic (Rochester, Minnesota) is a 1-hour, noontime, weekly conference sponsored by the Department of Internal Medicine (DOM). The main site for MGR is a large auditorium in the outpatient clinic facility, but the conference is telecast elsewhere on campus. At Mayo Clinic, the principal objective of MGR is to educate DOM faculty, fellows, residents, and students about advances in internal medicine practice, research, and education. Additional objectives of MGR are to provide opportunities for socialization and CME credit. These objectives are similar to those of MGR in departments of medicine elsewhere [1][2][3]10].
We report the results of a prospective, before-and-after study of the effects of providing complimentary food on attendance at MGR. We also report the results of a survey that assessed the attitudes of MGR attendees on the provision of complimentary food. . Attendance counts were eliminated if attendance data were missing (eg, because of a holiday). Head count data were available for 29 corresponding weeks during the pre-complimentary food and complimentary food periods, for a total of 58 events in the analysis. Given the nonnormal distribution of the data, a Wilcoxon/Kruskal-Wallis test (rank sum test) was used to calculate P values. The level of significance was P < .05.

Methods
To assess the attitudes of MGR attendees about complimentary food, a Web-based survey was administered to 943 DOM faculty, fellows, and residents (See Appendix for the survey questions).
Group comparisons of survey response distributions were done using the Pearson χ 2 test. In cases of small cell counts, a hybrid Fisher exact test was used for the comparisons [11]. P values less than .05 were considered to be statistically significant. All analyses except the hybrid χ 2 analysis were conducted using JMP 5.1 software (SAS Institute, Inc, Cary, North Carolina). The hybrid χ 2 tests were run using a Fortran routine (The Fortran Company, Tucson, Arizona) on a UNIX platform (The Open Group, San Francisco, California) [12].
This study was approved by the Mayo Clinic Institutional Review Board.

Effect of providing complimentary food on attendance at MGR
The mean (± SD) overall attendance increased 38.4% from 184.1 ± 90.4 per MGR session during the pre-complimentary food period to 254.8 ± 60.5 per MGR session during the complimentary food period (P < .001). Similar significant results were obtained when head count data for the main auditorium (P < .001) and the main telecast site (P < .001) were examined separately (Table 1). Notably, the size of the DOM increased by only 6% during the study period.

Complimentary food survey
After the study period, a Web-based survey was administered to all 943 DOM faculty, fellows, and residents. Of these, 444 (47.1%) responded (some respondents did not answer all questions; Tables 2, 3, 4, 5).
In response to the question "On average, how frequently do you attend MGR?" a majority of respondents (67.4%) reported that they attended either weekly or monthly. The responses of faculty members to this question were significantly different from the responses of fellows (P = .034) and residents (P < .001). Specifically, fellows and residents indicated that they attended MGR more frequently than did the faculty ( Table 2).
In response to the question "Are you more or less likely to attend MGR because of free food?" 70.1% of the respondents indicated that they were more likely to attend, whereas only 5.2% indicated that they were less likely to attend. The faculty's responses to this question differed significantly from the fellows' responses (P < .001) and the residents' responses (P < .001). For example, only 21.8% of the faculty respondents indicated that they were "much more likely" to attend MGR because of complimentary food, compared with 51.4% of the fellows and 44.2% of the residents (Table 3).
In response to the question "How has your attendance at MGR changed as a result of free food?" 53.6% of the respondents indicated that their attendance had increased, whereas 44.1% indicated that their attendance had not changed at all. The faculty's responses to this question differed significantly from the fellows' responses (P = .003) and the residents' responses (P = .001). Compared with faculty respondents, more fellows and residents indicated that their attendance had increased "much more" and "slightly more" as a result of the complimentary food (Table 4).
In response to the question "How would your attendance at MGR change if food ceased being provided free of charge?" 53.1% of respondents indicated that their attendance would decrease, whereas 44.0% indicated that their attendance would not change. The faculty's responses to this question differed significantly from the fellows' responses (P < .001) and the residents' responses (P < .001). Compared with fellow and resident respondents, a much larger percentage of faculty respondents indicated that their attendance would not change. Compared with faculty respondents, more fellow and resident respondents indicated that their attendance would decrease (Table 5).
When asked if food is a distraction (eg, because of noise) at MGR, most respondents (80.3%) indicated that the food was not a distraction. This response was similar among residents, fellows, and faculty. Among all respondents, 81.7% disagreed with the following statement: "The DOM receives unrestricted support from industry to pay for food at MGR. As a result, industry representatives have influence over MGR."

Discussion
Ours is the first systematic study assessing the effects of complimentary food on attendance at MGR. We found that, compared with attendance during the pre-complimentary food period, MGR attendance during the complimentary food period was significantly greater. These results suggest that providing free food may enhance attendance at MGR. The survey administered at the conclusion of the study period adds to these findings. A majority of respondents indicated that they were more likely to attend MGR because of the complimentary food and that their attendance increased because of it (although, compared with residents and fellow respond- ents, fewer faculty respondents reported that their attendance at MGR increased as a result of the complimentary food).
A number of factors (eg, program content and barriers to attendance) affect physician attendance at CME activities [13]. Survey data indicate that several barriers affect physicians' decisions to attend MGR, such as conflicting meetings, little presenter-attendee interaction, and inconvenient location [3]. Likewise, survey data indicate that some institutions provide complimentary food in an attempt to improve attendance at MGR [3]. Given that MGR occurs at noon at our institution, complimentary food not only removes a barrier (ie, by eliminating the need to choose between seeking food and attending MGR) but also adds an incentive for attending MGR (complimentary food). Some have described incentives as "the cornerstone of modern life" [14], and commonly acknowledged incentives for attending MGR, such as gaining new knowledge and CME credit, may not be sufficient for maintaining attendance. Although some view incentives negatively, incentives can be effective [5,6,8,14]. Therefore, as one examines strategies to increase attendance at MGR, one should consider not only removal of barriers but also the effect of incentives.
Although providing complimentary food may be associated with increased attendance at MGR, it also increases the cost of conducting MGR, which, for many departments, is the most expensive conference to conduct [3]. The cost of providing complimentary food at MGR at our institution is approximately $60,000 per year. To defray these costs, many departments, including ours, have garnered industry (eg, pharmaceutical) financial support [1][2][3]15,16].
Industry support of MGR raises the ethical concern of industry influence over MGR organizers, content, speakers, and attendees [1][2][3]17,18]. This concern can be addressed by using the following guidelines: 1) industry support should be unrestricted; 2) MGR speakers should disclose to attendees any conflicts of interest; 3) industry representatives should not determine MGR content; and 4) presentations at MGR should be unbiased, especially when the industry sponsor's products are discussed [3,19,20]. These guidelines are rigorously followed at our institution. Notably, our MGR attendees did not perceive inappropriate industry influence over the conference. However, measuring the influence of industry support by self-report may be biased. A recent study concluded that physicians' attitudes regarding industry support of CME activities may be biased (ie, those attending industry-sup-  ported activities are less likely to report bias than those attending non-industry-supported activities) [21].
Our study has several limitations. Although we used a prospective, before-and-after design, our study was neither randomized nor blinded. However, such a design would have been impractical. We could not control for or compare the quality of presentations during the 2 study periods, and, therefore, we do not know whether this factor contributed to an increased attendance at the MGR sessions. However, we compared attendance data matched for time of year to minimize bias (eg, related to holidays). Furthermore, advertisement of MGR did not change during the 2 study periods. We did not change the time of day or the day of the week that MGR was held during the study period. In addition, although our survey data suggest that faculty, fellows, and residents may behave differently in response to complimentary food as an incentive for attending MGR, we were unable to break down the attendance data according to attendee training status. Furthermore, we were unable to break down the attendance data by physician versus nonphysician attendees.

Competing interests
The author(s) declare that they have no competing interests.