Advances in technology are changing medical education. Web-based learning is becoming increasingly popular in institutions of higher education (1). Currently there are more than 50 institutions in 16 countries planning the development of an international virtual medical school, which will rely heavily on computer-based learning (2). The growing popularity of computer-based learning is due, in part, to its comparable effectiveness relative to more traditional learning methods. McDonough (3), for example, found that third-year medical students had the same knowledge gain from a solo computer session as compared with a small group face-to-face tutorial. Additionally, a recent study showed that online lectures are an effective way to deliver curriculum on disease screening to senior medical students (4).
Psychiatric knowledge is growing rapidly, and educational resources are increasingly limited (5). An ongoing challenge for psychiatry clerkship directors, who have students rotating at multiple clinical sites, is to assure a comparable quality of educational experience. Differences in patient populations and in the numbers and interest level of onsite faculty have long made it a challenge to assure that students have equitable educational experiences. Such experiences may include traditional lectures, case conferences, clinical rounds, and informal discussions with attending faculty. Each of these unique learning experiences can be effective in imparting knowledge to students with different learning styles (e.g., those who learn best aurally, visually, by discussion, or by "doing") (6). Hence, although some clerkship programs have moved away from traditional lectures, such lectures may continue to be useful for students who learn best through this modality.
Lecture attendance has become more difficult to consistently achieve given that medical care delivery has shifted from an inpatient to an outpatient model over the past decade. Consequently, medical students are often required to complete community rotations in an ambulatory setting to achieve learning goals (4). Many of these students complete their rotations at distant sites, making it difficult to return to campus to attend required lectures.
Drexel University College of Medicine (formerly known as MCP Hahnemann University) is one of the largest medical schools in the nation, admitting approximately 250 students per year. Students rotate in psychiatry services for 6-week blocks. At any given time, up to 35 students rotate in the psychiatry services during a clerkship block. These students are assigned to one of six different sites for their clerkship. The sites are located in Philadelphia (main campus and three local sites within 20 miles of the main campus), Pittsburgh (306 miles), and central New Jersey (68 miles). Currently, the experience at these sites generally includes rotating on inpatient units or hospital-based consultation-liaison services. We continue to search for ways to provide ambulatory experiences for students, which is a major goal because exposure to the ambulatory setting is positively related to student satisfaction (7). The dean has proposed that we develop more sites for clerkships in the near future. As we develop sites, we are challenged to ensure didactic comparability across sites.
Prior clerkship directors attempted to standardize the didactic experience at the various clinical sites by requesting that site directors provide a core set of lectures at their site. While site directors were committed to this mission, it was impossible to assure that the same quantity or quality of lectures would be provided because sites differed as to the number of faculty and their degree of expertise and availability.
During the prior (2001—2002) academic year, 2 full days of lectures were given at the main campus on the first and last days of the clerkship. While these lectures improved our ability to ensure an equitable didactic experience, problems remained. Students in Pittsburgh, for example, were unable to participate due to their distance from the main campus. Furthermore, although students rated highly the quality of lectures, many complained about having 7 hours of lectures at a time. Additionally, attendings at the various sites complained that student absence from clinical duties for 2 days was disruptive to patient care and made it difficult to adequately assess the students. This was especially true for students in Philadelphia, where the 6-week rotation was divided into two separate 3-week placements. Hence, we sought to examine whether the use of web-based lectures could help to standardize students' didactic experience, and to assess student satisfaction with such lectures.
Starting in April of 2002, our medical school began to develop web-based lectures for clerkship students. The clerkship directors from each department worked closely with the department of information technology to design a lecture bank web site accessible to all students. Microsoft software was used to package the audio, video, and slide components of each lecture into synchronized, multimedia presentations. Beginning in the Fall of 2000, all medical students were required to have a computer with Internet access as a condition of their acceptance into our medical school. In addition, rotation sites were required to provide Internet access for students.
The clerkship director requested that taping faculty lectures be permitted. Initially, some faculty members were resistant, as many were concerned about the lack of student interaction with web-based lectures. Some faculty members also expressed concern about updating lectures as new scientific advances are made. In addition, many faculty members used commercial videotapes of patient interviews, presenting significant copyright concerns. We agreed to tape lectures that did not include patient interviews and asked faculty members to tape only the didactic portion of their lectures. Despite these concerns, all 10 faculty members who were asked to participate agreed to do so.
In spite of these difficulties, 11 psychiatry lectures were posted on the web site over a 6-month period. The lectures varied in length from 30 minutes to 68 minutes. Most lectures (n = 7) were PowerPoint presentations that included audio narration. There were also four standalone streaming video presentations. Students received supplemental handouts for each lecture, along with lecture evaluation forms. The clerkship director was responsible for reviewing lecture content prior to release.
For piloting purposes, students who were in the last two rotation blocks of the 2001—2002 clerkship year were encouraged to view the lectures as a supplement to the live didactic sessions held at each clinical site. Starting in July of 2002, viewing the web-based lectures became mandatory for students who were unable to attend live lectures due to their being away from the main campus. We retained 1 full day of live lectures presented on orientation day (day 1) of the clerkship at the main campus. However, we discontinued lectures on day 2, as students were told to view these lectures on the web site. Whether live or web-based, attendance at lectures was a requirement, and attendance was monitored by sign-in sheets for live lectures, but only at the main campus. Students were required to complete an anonymous evaluation form for each lecture viewed. The evaluation was a paper and pencil form that asked students "Overall, how would you rate this session?" (1 = poor, 2 = fair, 3 = good, 4 = very good, 5 = excellent). During the first year of using the web-based lectures, there was no formal mechanism to ensure that a student actually viewed a given lecture.
The department of information technology compiled data on the number of times the psychiatry lecture web site had been accessed (July: 144, August: 257, September: 42, October: 66, November: 54, December: 71). The initial high numbers for July and August likely reflect frequent accesses by faculty and the clerkship director while they were setting up and editing lectures. The numbers for the last 4 months indicate steady use of the web-based lectures, presumably by students.
The clerkship office compiled data on student evaluations of the lectures (F1). Most web-based lectures (83.8%) were rated as "good" or higher, as were most live lectures (90.0%). Overall, however, live lectures were rated higher than web-based lectures (Live: M = 3.63 [SD] = 0.92, n = 491; web-based: M = 3.34 [SD] = 0.87, n = 314; t (803 df) = 4.50, p < 0.001). This analysis compared all web-based and all live lectures, and as such it included some live lectures that were not yet placed on the Web and some web-based lectures that were not presented live. However, six lectures were available to students at the main campus either live or on the Web, providing a more direct comparison. Students were not randomly assigned but were asked to view lectures on the Web that they were unable to attend live. Student satisfaction ratings did not significantly differ between web-based and live lectures (t1). It is noteworthy, however, that the direction of scores favored live lectures for 5 of the 6 topics.
Next, we examined the subset of web-based lectures to find whether student satisfaction varied between those who used PowerPoint slides with audio narration compared to those who used video stream. No significant difference in satisfaction ratings was observed (PowerPoint: M = 3.38 [SD] = 0.85, n = 152; video stream: M = 3.29 [SD] = 0.89, n = 166; t (316 df) = 0.95, p = 0.34).
Finally, we examined whether psychiatry shelf-exam scores were negatively affected by the use of web-based lectures. As surveys were completed anonymously, we were unable to compare students who viewed lectures live versus on the Web. However, we were able to compare exam scores for all students from the first 4 blocks of 2001—2002 to those from the first 4 blocks of 2002—2003 (web-based lectures began in 2002—2003). The scores were actually higher for the 2002—2003 group, suggesting that the use of web-based lectures did not impair shelf-exam performance (M2002—2003 = 75.6 [SD] = 8.2, n = 99, versus M2001—2002 = 72.7 [SD] = 8.2, n = 121; t (218 df) = 2.58, p = 0.01). We examined whether student aptitude might be related to the 2002—2003 class performing better on the shelf exams. Medical College Admission Test (MCAT) scores were not significantly different across the two classes, although a trend favoring the most recent class was found (M2001—2002 = 28.19 [SD] = 3.85, n = 273, versus M2002—2003 = 28.80 [SD] = 3.77, n = 246; t (517 df) = 1.82, p = 0.07).
Web-based lectures can be used to ensure didactic comparability across clerkship sites. Moreover, our generally positive findings concerning student satisfaction with web-based lectures in the psychiatry clerkship are consistent with those reported for other students in different settings. The 84% of students who rated web-based lectures as good or higher is remarkably similar to a study of undergraduates in which 85% of students reported satisfaction with their online learning experiences (8). Our findings are also similar to those from a study in which surgery students rated a computer-assisted instructional program positively (9). Although students rated the web-based lectures positively, the analyses suggested that they preferred the live lectures. Reasons for this preference are unclear, though some students did report technical difficulties accessing the web-based lectures. Students with slower modems reported problems with "breaks" during the lectures, and others complained of problems "seeing the slides." Many of these issues have now been resolved with the support of our information technology program. A more basic and currently unresolved issue is that some students may prefer the interactive nature of the live classroom environment, although we received no comments to this effect.
Students continued to show evidence of learning and, in fact, had improved shelf-exam scores compared with students from the prior year. This is consistent with a previous study, where undergraduate medical students studying anxiety disorders scored higher on objective measures after using a computer-based teaching package as opposed to a live lecture format (10). In addition, another study found that clerkship students obtained the same degree of knowledge as those who viewed live lectures, while spending 50 fewer minutes to complete the lecture (4), which indicates that web-based presentations may be more time efficient for students. However, because the present study used a naturalistic design, it is possible that other factors accounted for the increased shelf scores. Most notably, we changed our grading system in the 2002—2003 year so that students were required to score greater than the 72nd percentile in order to receive honors, a change that may have enhanced student motivation to increase their studying. It is also possible that the 2002—2003 class had greater aptitude, although comparison of their MCAT scores indicated that there was only a nonsignificant trend. Although the present study cannot confirm that web-based lectures increase exam scores, such lectures offer the advantage of giving students unlimited access to the lecture material. Further research is needed to examine whether students access web-based lectures more than once to review the material.
Limitations of the present study deserve mention. First, although the overall sample size was moderately large, the number of students viewing particular lectures for the direct comparison of web-based versus live lectures was modest. This was because only six of 11 web-based lectures were also available as live lectures, and these were taught only at the main campus. Thus, it is possible that some of these direct comparisons may have reached significance with a larger number of students. Second, we were unable to randomly assign students to web-based or live lectures. Third, students viewing web-based lectures may have done so during hours different from those attending live lectures. Although one of the advantages of the web-based lecture is its constant accessibility, it is possible that students viewing the web-based lectures at night, for example, may have been more fatigued, possibly lowering their enjoyment of the lecture and leading to a somewhat lower satisfaction rating. Finally, although we were able to monitor attendance for live lectures given at the main campus, it was not possible to monitor which individuals viewed the web-based lectures and completed evaluations. It is possible that some students never actually viewed the web lectures, yet turned in the required evaluation form. Our department of information technology has informed us that such monitoring will be available in the near future.
Web-based materials are likely to become increasingly important in medical education as the trend toward ambulatory experience becomes more pronounced and clerkship students rotate at sites away from their main clinical campus. Our findings suggest that web-based lectures are a well-received alternative to live lectures from the students' perspective. Nonetheless, future research is needed to examine whether such noninteractive web-based lectures are equally well received by all students and to examine which students best learn from such lectures. For example, prior research has found that medical students who viewed a computer-based lecture actually performed better on a knowledge test than did their peers who attended a live lecture, yet those viewing the computer-based lecture rated their knowledge and skills lower (10). Hence, there may be an adjustment period for students to adapt to web-based lectures, particularly given their lack of interaction with live lecturers and peers. Students have expressed concern that computer-based learning will severely limit interaction with instructors and may be less enjoyable (3, 9). It is possible, though, that computer-assisted lectures may become more interactive in the future, with the addition of online discussions, short quizzes and additional didactic material that target students' areas of weakness in a given topic (11). Finally, we will also need to develop and evaluate a standardized process for updating lectures as scientific advances are made.
Anecdotal reports suggest that faculty members were satisfied with the use of web-based technology. In the future, it would be important to collect standardized ratings of faculty satisfaction. Faculty initially expressed concerns, but their informal reports indicated that they became more accepting of using web-based technology and were generally satisfied with the experience. In addition, cost effectiveness needs to be investigated. The clinical lecture is the responsibility of many clinician-educators. With less time designated for lecture preparation, the quality of the clinical lecture may decline (11). Although faculty may spend extra time posting their lectures to a web site, ultimately, faculty may save time if web-based lectures decrease the need for repeated live lectures throughout the academic year.
Finally, others have called for the development of a nationally standardized curriculum within psychiatry education programs (5). A web-based curriculum developed by experts, posted on the Internet and used in multiple medical schools would standardize medical training and give students greater exposure to leaders in their respective disciplines.
The authors thank Dr. Richard Malone, Dr. Paul Ambrosini, and Dr. Susan McLeer for reading this paper and making suggestions.