In 2005, the Association of American Medical Colleges (AAMC) called on U.S. medical schools to increase class sizes by 15% to 30% to address a projected shortage of physicians during the coming decades.1 Many medical schools have responded, despite concerns about available resources to meet the needs of a larger class size,1,2 and despite the lack of an associated increase in the number of residency training positions, which is a necessary step to link increasing medical school class size with an increased supply of physicians.2 Furthermore, plans are under way to establish as many as 17 new allopathic medical schools in the United States.3 The expansion of the number of U.S. medical students is occurring even though prior projections related to the physician workforce have been incorrect,4 medical students now face staggering amounts of debt (more than four times the debt incurred by students in the 1980s),5 and recent pressures to cut or, in some cases, eliminate, federal support of medical education are mounting.6,7
Nearly 5,000 more U.S. medical students, a 15% increase in class size, may seem modest. But, this expansion could have a substantial impact on medical student education, particularly during the clinical years.8 In this study, we sought to identify the effect of increasing class size on the internal medicine clerkship, a core educational experience for medical students.
In May 2006, the Clerkship Directors in Internal Medicine (CDIM), the national organization representing those who teach internal medicine to medical students, conducted its annual, confidential survey of its 110 institutional members from U.S. and Canadian medical schools. (Not every accredited medical school has an institutional member of CDIM.) In addition to demographic information (age, gender, academic rank, academic promotion track, and medical school class size), we (P.H., T.I., S.D.) asked whether the number of medical students at each participating institution changed during the three years before the survey (2003–2006). If the number of students had changed, we asked whether the number increased or decreased, and by how many students. We also asked internal medicine clerkship directors (CDs) whether they anticipated a change in the number of medical students at their institutions during the next three years (through 2009). If the respondents answered yes, we asked whether this change would be an increase or decrease, and by how many students. We asked all respondents to identify what increase in class size they believed their institution/clerkship could accommodate.
We also asked how increasing medical school class size by either 15% or 30% would impact the logistics of running the internal medicine clerkship. Specifically, we asked for each category of increased class size (15% and 30%) how many of the following would need to be added:
* inpatient clerkship sites;
* students per inpatient site;
* ambulatory clerkship sites; and
* students per ambulatory clerkship site.
Next, we posed a series of statements specifically regarding a projected increase in class size of 10% to 15% (the initial increase in class size recommended by the AAMC1), and we asked respondents to note their level of agreement (one = strongly agree, two = agree, three = disagree, four = strongly disagree) with the following statements, all beginning, “If my medical school class size increases by 15% …
* I will have more resources to administer the clerkship.”
* I will have more teachers available to teach ambulatory students.”
* I will have more teachers available to teach inpatient students.”
* I will have more protected time to administer the clerkship.”
* it will be more difficult to recruit or find teachers to teach the Introduction to Clinical Medicine courses.”
* I expect the core medicine clerkship will be shortened to accommodate more students across clerkships.”
Using the same four-point scale, we asked CDs to rate their agreement with the following statements that addressed their attitudes about the underlying reasons behind the proposal to increase class size:
* Increasing medical student class size is an effective means to address the projected shortfall of physicians in the United States.
* The projected shortfall of physicians will require increasing the numbers of international students accepted into U.S. medical schools.
* The projected shortfall of physicians will require increasing the numbers of international students accepted into U.S. residency programs.
* The projected shortfall of physicians will require increasing the numbers of “physician extenders” (e.g., physician assistants and nurse practitioners).
Finally, we provided an opportunity for a free-text response to the question, “What are other challenges that increasing class size would pose for you and your institution?” Using standard qualitative methods and an iterative process, two of us (P.H. and S.D.) reviewed the free-text responses, identified emerging themes, categorized comments into these themes, revised prior coding (if necessary), and resolved any disagreement by consensus.
The survey was developed by the authors, reviewed and modified by the CDIM Research Committee, and approved by the CDIM Council. The survey was distributed in May 2006 and administered through the CDIM Web site. Nonresponders were contacted up to two times. The Uniformed Services University of the Health Sciences institutional review board reviewed and approved the survey as an exempt study. Statistical analysis included descriptive statistics and two-tailed independent-samples t test for continuous variables.
Eighty-three of the 110 respondents returned questionnaires, for an overall response rate of 76% (but every respondent did not answer every question.) The demographic profiles of the respondents remained similar to those of prior surveys of CDIM institutional members: mean age was 44.9 (±7.0) years; 55% of the respondents were male; 34% were assistant professors, 47% were associate professors, and 19% were full professors; and 84% of respondents were in the clinician–educator track. The mean clerkship length was 10.6 weeks (median 12 weeks, standard deviation [SD] 2.4).9,10
Recent and projected class size changes
The median medical school class size was 140 students (mean 140.3, range 40–250). For the three years preceding the survey, 42 respondents (51%) reported no change in their medical school class size; 40 (48%) reported an increase in class size; and one (1%) noted a decrease in class size. The mean increase in class size reported during this time frame was 13.9 students (SD 9.5, range 2–40), which represents a 9.0% (SD 5.4, range 1.3%–24%) increase in class size for those reporting an increase. For the next three years (through 2009), according to the survey, 35 CDs (43%) projected no change in class size, 47 (57%) projected an increase in class size, and none of the respondents expected class size to decrease. The projected mean increase in class size was 17.4 students (SD 10.6, range 4–50), which represents a 13.8% change (SD 9.2, range 3%–48%) for those reporting an expected increase. Overall, 30 CDs (36%) responded that their schools were increasing class size in the three years both before and after the survey, for a total of six years of increasing class size. For these medical schools, the mean increase in class size would be 27.9 students (SD 13.5, range 7–50), which represents a 21% change (SD 11.3, range 4.4%–50%). Internal medicine CDs identified that a feasible increase in class size would be approximately 19 students per class, which represents a 13.4% increase (SD 19.4, range 0–100) based on a median class size of 140.
We wished to understand whether there were differences in the characteristics of the medical schools that were undertaking class size increases. There were no differences in medical school class size between medical schools that did expand and those that did not expand (132 students versus 148 students, P = .13, two-tailed t test) their class sizes from 2003 to 2006. Similarly, there were no differences in medical school class sizes between medical schools that planned to expand and those that did not plan to expand (137 students versus 144 students, P = .52, two-tailed t test) their class sizes from 2006 to 2009. Finally, there were no differences in medical school class sizes between the schools that were expanding during the entire six years compared with those that were not (142 students versus 130 students, P = .81, two-tailed t test).
Impact of 15% or 30% increase in class size
Table 1 summarizes the CDs’ responses to how a 15% or 30% increase in class size would impact the internal medicine clerkship with regard to inpatient and ambulatory sites. Eight respondents answered these questions by stating the total number of students for the academic year (not per rotation) that would be added across the sites. For these eight responses, we divided that number by the number of internal medicine clerkship rotations per academic year (determined by the length of the clerkship reported; for example, a 12-week clerkship would have four clerkship blocks each academic year). Responses demonstrate that increasing class size will not only result in increasing the number of inpatient or ambulatory students per site but will also likely mean that CDs will have to seek new training sites for students. For ambulatory rotations, this increase could represent a substantial expansion in the number of sites, likely because CDs send students to work with community-based faculty, where the relationship is often one teacher for one student.11
Table 2 summarizes the responses about whether internal medicine CDs agreed or disagreed with statements about increasing medical school class size. According to the survey results, internal medicine CDs are concerned that an increased number of students will not be matched by an increase in resources or protected time to administer the clerkship. As a result, respondents report that they believe recruiting teachers, both for the clerkship and for clinical skills courses (such as those that introduce clinical medicine to students), will be more difficult.
Qualitative analysis of free-text comments
Twenty-nine CDs (35%) answered the free-text question, “What are other challenges that increasing class size would pose for you and your institution?” On the basis of the qualitative analysis, two overall themes emerged (Table 3). First, CDs reported concerns about having the necessary resources to maintain a high-quality clerkship if the number of students increases. These concerns included having adequate space during the preclinical and clinical years; receiving increased funding/resources to support an increased class size; and ensuring the adequacy of training sites. Examples of different respondents’ comments from the theme about resources are below:
Lack of space
We have limited space even during the preclinical years.
We can no longer accommodate the class in a single auditorium—ever!
Adequacy of clinical training sites
Finding/retaining ambulatory sites will be the biggest challenge.
May need to add more students onto teams.
Additional sites will likely be added.
Lack of resources
I do not anticipate … that the internal medicine clerkship, other third-year clerkship[s], or other required rotations will be given additional resources to support the new students—no new faculty, no new clinicians, no more administrative staff or support, no more protected time.
The second theme revealed that internal medicine CDs are concerned that expanding the number of students will have a negative impact on the educational experience. Comments addressed general concerns about the overall educational experience, the impact on students, and the impact on faculty.
Overall educational experience
The chief challenge would be to maintain the necessary clinical experience for each student that rotates through the clerkship.
We would have to put students to work with attendings and residents who aren’t as good teachers and on services that aren’t as good learning experiences.
Teaching clinical skills courses will be more difficult and preclinical courses will likely have to be altered to accommodate more students.
The increase in class size comes simultaneously with a push to shorten hospital stays and shift to outpatient and ambulatory settings for the delivery of medical care in our institution. This decreases the amount of time and settings available for inpatient internal medicine, which forms the core of medical student teaching in medicine.
Impact on students in preclinical and clinical years
Dilution of clinical experience for all students by increasing number of students at each site without increasing numbers of patients.
Would make getting to know the students as individuals more difficult.
Impact on faculty
For the impact on faculty, CDs raised concerns about faculty development, increasing the burden on faculty, lack of incentives for teaching, and difficulty recruiting teachers. Examples of comments regarding faculty are below.
Lack of incentives/poor morale:
We just increased our class size by mandate of the dean and it was a very unpopular decision with faculty and students who fear that if the school becomes much bigger it will lose some of its warmth and collegiality.
Although we have a ‘clinician–teacher’ track, being a teacher is still a low-status position; this is the only track where faculty cannot achieve tenure, and our department chair’s interest and support for education is superficial at best.
We get by on the fact that we have a core group of truly dedicated faculty who will teach because they believe it is the right thing to do, but when you get right down to it, money is what counts in the institution.
Faculty physician morale/job satisfaction may also be impacted negatively—i.e., being asked to teach even more students with no additional compensation (financial and/or time).
Lack of mentoring and development:
Expanding to other cities … without housestaff is unacceptable for mentoring.
Difficulty recruiting teachers:
Finding more teachers (e.g., in my school, adding just two students per class) precipitated a complete overhaul to [an] Introduction to Clinical Medicine course.
Even an increase of one to two students per block on the clerkship can mean recruiting 5 to 10 more ambulatory teachers.
Increased burden on faculty:
Adding more students to the mix, the few people who do teach will be asked to do more, and they will be able to give even less attention to the students they have now.
These concerns about the impact of increasing medical school class size on faculty were prominent themes among CDs, accounting for nearly 25% of all comments made.
As our survey shows, U.S. medical schools are already increasing class size as one strategy for addressing the projected shortfall of physicians. As this expansion continues, 17 new medical schools may be established during the next decade.2,8 This two-pronged strategy is occurring despite concerns cited by medical school deans and other academic leaders about the difficulties of coping with increased class sizes, particularly in trying to educate a larger group of students with a current shortage of faculty and clinical training sites, reduced patient volume, problems with providing faculty development, limited clinical teaching time, and pressures on graduate medical education training.8
With this current CDIM survey, we sought the opinions of core educational leaders to understand the realities of adapting undergraduate medical education to more students. Most of the respondents reported that their medical schools were already increasing class size, by an average of 9.0% (2003–2006); many were expecting projected increases in class size (2006–2009), by an average of 13.8%; and, for those medical schools expected to expand during the period of 2003 to 2009, the increase represented a 21% increase in class size. This increase in class size is at—or larger than—what most internal medicine CDs believed their schools could reasonably accommodate.
Clearly, for most internal medicine CDs, increasing class size will necessitate increasing the number of students at both inpatient and ambulatory clerkship sites Furthermore, many respondents believe they will need to add additional clerkship sites to meet the educational needs of these students. Although the ambulatory numbers cited might seem an overestimation, it is likely that the responding CDs send students to work individually with a community-based faculty member; thus, just one more student per clerkship rotation means one more ambulatory teacher and one more ambulatory site. For those ambulatory clerkships that use a multidisciplinary model for ambulatory education (i.e., one student working with a variety of attendings over several weeks or several months), increasing class size would mean that several faculty members would have to be recruited for each additional student.11 However, the current CDIM survey did not address the structure of the ambulatory clerkship. Nevertheless, increasing medical school class size may have a disproportionate impact on ambulatory education, and this deserves close attention.
The responses to the statements about resources and support noted in Table 2, and the responses to the open-ended questions in Table 3, reveal that internal medicine CDs are worried about not simply the logistical challenge of accommodating a larger number of students, but also the wider impact on student education, faculty morale, faculty development, and recruiting an adequate number of qualified teachers. These concerns are further amplified by the belief of CDs that they will not be supported by increased funding or protected time—even though medical schools will have more resources (because of greater tuition, realized by increasing class size). Interestingly, the concerns cited by the internal medicine CDs are essentially identical to those identified by the medical school deans when they considered the impact of an increase in class size before the expansion occurred: space for preclinical teaching, faculty recruitment, adequacy of the number of clinical teaching sites, adequacy of the patient population, and a lack of residency positions.8
CDs noted that the increase in class size has been larger and more extensive than what had been predicted. In 2003, reported changes in class size that were then occurring yielded a net growth in class size of 2.1%. When considering all responses from medical school deans about actual and possible changes in medical school class size, a projected increase of 7.6% was thought possible.8 Responses to the AAMC’s 2005 survey of U.S. medical schools1 indicated a 5.4% class size increase for existing medical schools by 2015. The CDIM survey found that class size has expanded by 9% for schools that have already increased class size and will expand by 14% to 21% during a three-year (2006–2009) or six-year period (2003–2009) for those schools that plan to increase class size. Also, in contrast to the prior study,8 the CDIM survey found no differences in class size between those medical schools planning to expand and those not planning to expand. The differences between this study and the 2003 survey of deans are likely attributable to a more accurate representation of what is occurring rather than what might have been envisioned, a better response rate on the current survey compared with the deans’ survey (response rate = 58%),8 and the fact that the pressure to expand medical school class size is probably reaching more medical schools. This rapid and greater-than-expected expansion is a likely contributing factor to the June 2007 revision of the Liaison Committee on Medical Education Educational Resources directive that now addresses medical schools that are increasing class size beyond 10% to 15%.12
The fundamental concern is that the expansion of medical school class size has been quite aggressive and may have occurred without adequate assessment of the breadth of the impact on student education. On the basis of the average medical school tuition, fees, and health insurance in 2006–2007 ($21,000 [resident] to $40,000 [nonresident] for public schools and approximately $38,000 for private schools),5 increases in class size are likely to generate sizable amounts of money per institution; it is important that the distribution of these new tuition dollars be transparent and substantially dedicated to meeting the needs of educating larger class sizes.
There are limitations to this study. First, we surveyed only internal medicine CDs, but our findings are similar to those in a survey conducted by the AAMC in 2007 of multiple stakeholders at six medical schools that were expanding class size.13 Further, as internal medicine clerkships typically consist of both inpatient and ambulatory rotations,11 it is likely that the concerns about student education and the impact on faculty recruitment and development would extend to other disciplines. Second, we did not explicitly ask CDs whether they saw any benefits to increasing class size. CDs may agree that increasing class size is one way to address the projected physician shortage, but clearly they also have concerns. Nevertheless, we did not explicitly inquire about, and therefore probably failed to capture, perceived advantages of increasing class size. Third, this was a cross-sectional survey conducted at a time when class size was increasing. A follow-up survey, determining whether internal medicine CDs’ concerns about increasing class size have been realized and reporting how they have coped with these changes, is important. Fourth, estimating the increased numbers of students per site and the number of new sites needed was somewhat difficult because some CDs interpreted the pertinent question as asking for the total number of students per year, whereas others interpreted it as asking for the number of students per clerkship rotation. We tried to adjust for this by modifying the responses that were clearly an estimate of the annual impact on the clerkship, but we recognize that the estimates may be an overestimation of the impact of changes in class size.
To address a projected shortfall in the number of physicians in the United States during the next few decades, medical schools are increasing class size. This expansion may be an incomplete response to the identified need for more physicians. Increasing class size clearly has the potential to adversely affect student education, faculty recruitment, and faculty morale. The leaders of academic medicine must pay close attention to outcomes of student education after increasing class size. Ideally, local decisions to increase class size will involve, and be informed by, all who are concerned with student education.
The views expressed in this paper are those of the authors and do not represent the views of the United States Air Force, the Uniformed Services University, the Department of Defense, other federal agencies, the American Society of Nephrology, or the Clerkship Directors in Internal Medicine.
The data reported in this paper are the property of the Clerkship Directors in Internal Medicine and are used with permission.