White, Casey B. PhD; Haftel, Hilary M. MD, MHPE; Purkiss, Joel A. PhD; Schigelone, Amy S. PhD; Hammoud, Maya M. MD
On July 1, 2003, new regulations by the Accreditation Council for Graduate Medical Education (ACGME) that restricted medical residents' work schedules to 80 hours a week went into effect throughout the United States.1 In anticipation of the restriction, the ACGME issued a statement on potential effects of these regulations on residents, program directors and faculty, teaching institutions, and medical specialties.2
Effects on residents included positive impact on their personal lives (including less fatigue and more time for family), but also concerns about reduced operative and patient care experiences, and reduced teaching time and teaching responsibilities. Effects on teaching institutions emphasized their responsibility in creating an appropriate learning environment to promote safe patient care and high-quality learning. Effects on specialties focused mainly on how the programs would have to adapt to enforce the work-hour restrictions.
There was discussion and debate about the ramifications of adopting the restrictions. Those in favor argued that they would reduce resident fatigue and help them to provide safer patient care,3 and that residents would have better balance between their personal and professional lives and more time for independent study and research.4 Those opposed argued that patient care would be compromised because continuity would be disrupted by frequent house officer transitions required as shifts changed, and that medical education would be compromised as well.5,6
Since the implementation, studies have been published on the actual effects of the restrictions. Positive effects include increased resident satisfaction with less fatigue,7,8 an increase in sleep hours and a reduction in attention failures,9 and an improvement in residents' attitudes toward their overall educational experience.10 However, studies have also documented concerns about decreased resident experiences in surgery fields,8,11 decreased quality of patient care related to decreased continuity in patient care,12–14 and minimal evidence of improvement in patient care.15
Work-hour restrictions have also been found to have effects on medical students' education. On the positive side, there is some evidence that medical students' interest in surgery has increased since the introduction of work-hour limitations.16 However, medical students are worried about the effect of the work-hour regulations on their learning. They worry that less time spent in the hospital means busier residents with less time to teach, and that the students themselves will be spending long hours in the hospital without the benefit of residents' teaching.17
Their concerns are well grounded. In many disciplines, most hospital-based education is done by residents18 who spend substantial amounts of time teaching medical students and junior residents.19 Brown reported that residents spend between 20% and 25% of their time teaching,20 and that the majority of medical students are getting over 50% of their teaching from house officers and fellows in many of the required clerkships.21
Given the changes that have been anticipated and reported with regard to effects of residents' work-hour restrictions on medical education, we undertook this study to document the effects at our medical school, the University of Michigan Medical School. Our study goals were to examine effects of the work-hour restrictions on medical students' experiences as measured by their perceptions of the quality of their experiences during the third-year clinical clerkships. We hypothesized that the work-hour restrictions would have negative effects on medical students' perceptions of their education in the third-year clerkships.
Data were collected during academic year 2002–03 and again in 2003–04 from third-year medical students who completed the required clinical clerkships during each of those years and who volunteered for this study. The students in the class of 2002–03 completed the clerkships prior to the adoption of restricted work hours for house officers. The students in the class of 2003–04 completed clerkships during the same year in which restricted work hours for house officers were adopted. At the times the two classes responded to our questionnaires (which were given to them as part of an ongoing curriculum evaluation program), we had not planned or discussed our study on the effects of restricted work hours upon residents.
There were virtually no differences between the cohorts in entering credentials. For example, in comparing the earlier to the later class, the average scores on the Medical College Admission Test were 11.1 and 11.2, respectively; the average overall grade-point averages (GPAs) were 3.64 and 3.68, respectively; and the average science GPAs for each class was the same: 3.60. Also, there were no important differences between the classes in their performance or the content and assessment measures and standards in the courses in the first and second years of medical school, and no real differences in their performance, goals, objectives, or expectations in the clinical clerkships. Also with regard to clerkships, there were no real differences in the qualifications or caliber of house officers, their preparation for teaching, or their teaching methods.
We collected data from each cohort of students, from questionnaires they had completed at the end of each of their clinical clerkships. The questionnaires have been used for over ten years to assess students' satisfaction with their clerkship experiences. Third-year students complete a series of seven required, discipline-based clerkships in internal medicine (12 weeks), surgery (8 weeks), obstetrics–gynecology (6 weeks), pediatrics (6 weeks), family medicine (4 weeks), psychiatry (4 weeks), and neurology (4 weeks). Four clerkships were chosen for this study: internal medicine, obstetrics–gynecology, pediatrics, and surgery. Our decision was based on the fact that house officers shared significant responsibility for teaching on these clerkships, and so we anticipated that medical students would perceive a more significant impact of restricted work hours on their experiences. Eleven core questions pertaining to accessibility, quality, and time spent on specific activities were asked on the clerkship evaluations; these questions remained consistent for each of the two classes.
Descriptive statistics were obtained for all items described above. Independent-sample t-tests were performed on all items, comparing the 2002–03 cohort of students with the 2003–04 cohort. Items from the clerkship evaluations were analyzed separately for each of the four clerkships examined (internal medicine, obstetrics–gynecology, pediatrics, and surgery).
The students in the two classes studied completed their questionnaires on a voluntary basis. Response rates for the clerkship evaluations varied by clerkship, ranging from a high of 100% (n = 168) in the 2003–04 cohort for obstetrics–gynecology, to a low of less than 50% (n = 81) in the 2002–03 cohort for surgery.
In several specific areas, third-year students in 2003–04 were less satisfied with their clerkship experiences than were third-year students in 2002–03. Table 1 organizes the results by clerkship (columns) and by three broad groupings of responses (rows): accessibility and quality, accessibility influences on student experiences, and time spent in student tasks and responsibilities. Of the core 11 questionnaire items, two significant negative results were reported for internal medicine, seven for obstetrics–gynecology, and seven for surgery. For pediatrics, three significant positive results were reported.
Students from the 2003–04 cohort who completed the internal medicine clerkship reported a significant decrease in the quality of feedback they received compared to the perceptions of feedback quality by the previous cohort, and they spent significantly more time on unproductive activities (“scut”). There were also notable downward trends between the two cohorts in accessibility of house officers and time spent in independent study, but notable upward trends in observation of clinical skills and quality of the clerkships overall. The trends were not statistically significant.
Students completing the obstetrics–gynecology clerkship in the 2003–04 cohort reported significantly less access to faculty and house officers, and significantly lower quality of house officer teaching, feedback, and the clerkship overall. Students in that cohort also reported a significant decrease in observation of their clinical skills and in their ability to manage patient problems. Finally, there were notable downward trends in the perceptions of that cohort compared with the reports of the earlier cohort regarding quality of faculty teaching and clarity of expectations. The trends were not statistically significant.
Students completing the surgery clerkship in the 2003–04 cohort reported significantly less access to faculty and significantly lower quality of feedback and of the clerkship overall. They also reported a significant decrease in their ability to manage patient problems as well as significant decreases in the clarity of expectations and time spent in independent study, as well as a significant increase in time spent on unproductive activities. There were notable downward trends in the perceptions of that cohort compared with the reports of the earlier cohort regarding quality of faculty teaching and observation of clinical skills and, interestingly, a notable upward trend in the quality of house officer teaching. The trends were not statistically significant.
Reversed patterns of significance were observed in the pediatrics clerkship, where students in the 2003–04 cohort reported significantly more access to faculty and significantly higher quality of faculty teaching and feedback. There was also a notable upward trend between the earlier cohort and the later cohort in the quality of the clerkship overall, although the trend was not statistically significant.
This study investigated two cohorts of third-year medical students' assessments of their clerkships utilizing a standard clerkship evaluation. There were no differences between the cohorts in the content or assessments in the first and second years of medical school, and no differences in their performance. Also, there were no differences in the goals and objectives in the clinical clerkships, and no differences in their performance. However, despite these similarities and the similarities in academic credentials when they entered medical school noted earlier, the experiences of these two cohorts during the third-year clerkships were significantly different in several areas. We hypothesized that these differences were due at least in part to the adoption of house officer work-hour restrictions.
Decreases in students' satisfaction between the two cohorts were clearly more prevalent and significant in the two surgery-oriented clerkships, obstetrics–gynecology and surgery. Because the operating room is an essential learning environment in these disciplines, a disproportionate amount of a resident's (limited) time is now spent there rather than on the floors working with students. Faculty are not filling this gap, because they are spending a lot more of their time performing patient-related activities previously handled by residents such as writing orders, reviewing data, talking with families, and consulting with nursing staff.8,11,14 This limited access to faculty was most likely the major contributor to students' negative perceptions of the quality of critical components of the two clerkships, including faculty teaching (noteworthy if not significant for both clerkships) and the clerkship overall (significant for both clerkships). Limited access to faculty also meant limited observation of their clinical skills (significant for students in the obstetrics–gynecology clerkship), which negatively influenced students' perceptions of their ability to manage patient problems (significant for both clerkships).
Interestingly, the 2003–04 surgery clerkship students' perceptions of their accessibility to house officers remained stable and their view of the quality of house officer teaching actually improved (although not significantly), whereas the students from that cohort in the obstetrics–gynecology clerkship had significantly more negative views of these two aspects of their experience. A closer review of the data revealed that—at 3.96 (accessibility) and 3.68 (quality of teaching)—students' perceptions in the obstetrics–gynecology clerkship were well above student perceptions on surgery (3.67 and 3.29, respectively) before the change in resident work hours. (The just-stated data are mean scores on a five-point scale where 1 was poor and 5 was excellent.) It might be that because accessibility was “very good” and quality of teaching was almost “very good,” the change in residents' hours had a more significant and negative impact on student experiences in obstetrics–gynecology. Also, surgery hired additional physician assistants to offset, at least in part, the change in residents' schedules. Both the residents and students could therefore spend more time in the operating room, a site where a large proportion of teaching could occur, allowing the surgery residents to remain at least somewhat accessible to the students and—given the fact that students saw faculty less as the new regulations were adopted—even improved the quality of residents' teaching in the eyes of the students.
Students want to learn all they can from faculty experts on the clinical clerkships, but they also want opportunities to impress those who will make major decisions in the assessment of their performance, and those who they will ask to write letters recommending them for residency placement. To succeed, they want to know exactly what is expected of them, as outlined clearly by the faculty. The changes in residents' work hours and the limitation of students' access to faculty negatively influenced students' perceptions of the clarity of expectations for their performance in the surgery and obstetrics–gynecology clerkships. For the surgery clerkship, this was a significant change from the year before.
Students reported significant changes across the surgery-oriented and nonsurgery-oriented clerkships in the quality of feedback (the only positive change was from the students in the pediatrics clerkship, discussed later in this report). Even though students' reports of the quality of the internal medicine clerkship remained mostly stable through the change in resident work hours, quality of feedback took a significant downturn here, as it did for obstetrics–gynecology and surgery. The significant decrease in the quality of feedback reported by the internal medicine and obstetrics–gynecology students can probably be connected at least in part with the downward trend in accessibility of residents. This, however, is not a new phenomenon. Long before work-hour regulations were adopted, medical schools recognized a problem with the quantity and quality of feedback provided to clinical medical students. The problem was rooted in increased performance expectations of physicians directly connected with activities that generated revenue, which had negative effects on the time available to teach. More than 10 years ago, Stillman and colleagues, in a study focused on evaluating the clinical skills of students in four medical schools, reported that 35% of clinical students had never been observed doing a complete history and physical examination; another 22% reported having been observed once.22 A few years later, Irby reported that learning encounters between physicians and patients typically ranged from three to six minutes and contained little teaching and virtually no feedback.23 More recently, Elnicki and colleagues reported that although studies about clinical teaching had consistently found feedback to be an essential element of student learning, clinicians found it to be a difficult aspect of medical education.24 Not unexpectedly, this longstanding problem seems to have been exacerbated by the new work-hour restrictions, resulting in the significant decreases in the quality of feedback reported by the 2003–04 cohort.
Students in that cohort also reported significantly more time spent in unproductive activities in the internal medicine and surgery clerkships and less time spent in independent study. A matter of concern would be that students on those two clerkships have been picking up at least some of the work not perceived as having educational value (i.e., “scut” work) no longer done by the residents, and that work has been reducing (significantly for those in the surgery clerkship) the time they have to spend on more academic tasks, such as independent study.
The significant exception to the negative changes reported by the 2003–04 cohort was from the students in the pediatrics clerkship. Students from that cohort in that clerkship said they had significant increases in accessibility to faculty, quality of faculty teaching, and quality of feedback, and there was a clear positive trend in their views of the quality of the clerkship. These upward swings were because the Department of Pediatrics expanded and formalized their pediatrics hospitalist program when work-hour restrictions were instituted. These physicians are now present in the hospital 24 hours a day and are charged with supervising and educating the residents and students on the general inpatient services. Pediatrics also added resident assistants to each inpatient team, which redirected some of the “scut” activities away from house officers and students. These changes have significantly improved the learning environment in the pediatrics clerkship despite the work-hour restrictions.
Across three major clerkships, students in the 2003–04 cohort reported experiences that were more negative—significantly in many cases—than did students in the 2002–03 cohort. We believe the negative and the positive (for pediatrics) changes reported by students can be directly attributed to the adoption of restricted work hours for house officers, and these changes were influenced by how programs prepared (or did not prepare) for the change. Authors of a recently published study described positive changes in 2003–04, compared with 2002–03, as reported by two cohorts of medical students.6 However, that study combined poor and fair ratings and excellent and outstanding ratings across all of the survey items, had only one clerkship questionnaire per student (rather than having students complete a questionnaire for each of the clerkships as they finished the clerkship) and— most importantly—did not report results by clerkship. Thus, differences among the individual clerkships were not described, compared, or discussed, and accommodations or changes made that presumably resulted in positive reports by students were only briefly described and not substantiated.
Our study highlights important outcomes in students' perceptions related to preparations made (or not) by specific clerkships as restricted resident work-hour regulations were adopted. Pediatric hospitalists who focused on teaching in the inpatient setting not only filled the teaching gap left by busy residents on reduced hours and busy faculty picking up leftover work, they actually addressed an existing problem among all of the clinical clerkships—the need for more feedback to students. This model should be carefully considered by all nonsurgical clerkships across U.S. medical schools. Internal medicine hospitalists focused on picking up overflow admissions, which helped alleviate some of the residents' burden, particularly later in the day, in turn allowing residents to spend at least some time with medical students that would have been spent on paperwork and other responsibilities related to admitting patients. Likewise, physician assistants hired by the Department of Surgery clerkship also helped that clerkship to maintain some level of contact between residents and medical students. Unfortunately, the medicine hospitalists and surgery physician assistants were not assigned to pick up “scut” work left over by residents, and this burden appears to have initially fallen to the medical students, at least on those two clerkships. Finally, the obstetrics–gynecology clerkship spent the year after adoption of the new regulations making changes designed to improve clinical students' experiences, including adoption of “night float” and consideration of adding “laborists” (i.e., labor and delivery hospitalists) to the medical staff. Overall, subsequent changes in some areas of the clerkships led to increases the next year—in some cases significant increases—in students' scores that reflected more positive perceptions of their clerkship experiences.
We conducted this study using existing data that had been gathered as one component of an ongoing curriculum evaluation program. Because students were not specifically asked to use the change in house officer work hours as a context for their responses, we are making inferences that connect the changes in student perceptions of their experiences with changes in house officer work hours. These inferences are based on consistency between the two cohorts across all of the other variables we considered, including students' entering credentials and academic performance, house officer qualifications, preparation for teaching and teaching methods used, and consistency across the courses, sequences, and clerkships they all completed.
Also, this study was conducted at one medical school with class sizes of approximately 170 students. Studies of the effects of restricted work hours at other medical schools or across a larger group of medical schools might yield different results.
Findings from this study point to the power of the interconnections between and among faculty, residents, and medical students, all of whom are major stakeholders in medical education. The findings also point to the differences in experiences between the surgery-oriented and nonsurgery-oriented clerkships. Without access to the faculty, and even with stable access to residents (as reported by the students in the surgery clerkship), students reported that their clinical experiences were negatively and in many cases significantly affected. Very few clinicians ever believed that restricted work hours for residents would have no impact on resident or medical student education, and even some of the medical students were concerned enough to conduct their own study.17 Yet not all clinical clerkships were prepared. The clerkships that planned in advance, and comprehensively, for this major change remained the most stable (at least in terms of students' perceptions of their clinical education) as the new regulations were adopted.
Results yielded from this study raise interesting questions for future research. What are faculty members' and house officers' perceptions of how restricted work hours are affecting their medical student teaching? How are clinical departments addressing the effects of restricted work hours on the quality of medical student education? How do changes they are making vary between surgery-oriented and nonsurgery-oriented disciplines? Are students' perceptions of their clinical education changing as the disciplines make changes to assure the quality of their educational programs, and what is the nature of these changing perceptions?
1 Philibert I, Friedmann P, Williams WT. ACGME Work Group on Resident Duty Hours. Accreditation Council for Graduate Medical Education. New requirements for resident duty hours. JAMA. 2002;288:1112–14.
3 Gaba DM, Howard SK. Fatigue among clinicians and the safety of patients. N Engl J Med. 2002;347:1249–55.
4 Steinbrook R. The debate over residents’ work hours. N Engl J Med. 2002;347:1296–1302.
5 Barden CB, Specht MC, McCarter MD. Effects of limited work hours on surgical training. J Am Coll Surg. 2002;195:531–38.
6 Jagsi R, Shapiro J, Weinstein DF. Perceived impact of resident work hour limitations on medical student clerkships: a survey study. Acad Med. 2005;80:752–57.
7 Lund KJ, Alvero R, Teal SB. Resident job satisfaction: will 80 hours make a difference? Am J Obstet Gynecol. 2004;191:1805–10.
8 Chung R, Ahmed N, Chen P. Meeting the 80-hour work week requirement: what did we cut? Curr Surg. 2004;61:609–11.
9 Lockley SW, Cronin JW, Evans EE, et al. Effect of reducing interns’ weekly work hours on sleep and attentional failures. N Engl J Med. 2004;35:1829–37.
10 Peterson CM, Gerstle R, Bhalla S, Menias CO, Jost RG. Results of the 2004 survey of the American Association of Academic Chief Residents in Radiology. Acad Radiol. 2005;12:373–78.
11 Kort KC, Pavone LA, Jensen E, Haque E, Newman N, Kittur D. Resident perceptions of the impact of work-hour restrictions on health care delivery and surgical education: time for transformational change. Surgery. 2004;136:861–71.
12 Cohen-Gadol AA, Piepgras DG, Krishnamurthy S, Fessler RD. Resident duty hours reform: results of a national survey of the program directors and residents in neurosurgery training programs. Neurosurgery. 2005;56:398–403.
13 Fletcher KE, Saint S, Mangrulkar RS. Balancing continuity of care with residents’ limited work hours: defining the implications. Acad Med. 2005;80:39–43.
14 Bailit JL, Blanchard MH. The effect of house staff working hours on the quality of obstetric and gynecologic care. Obstet Gynecol. 2004;103:613–16.
15 Winslow ER, Berger L, Klingensmith ME. Has the 80-hour work week increased faculty hours? Curr Surg. 2004;61:602–8.
16 Miller G, Bamboat ZM, Allen F, et al. Impact of mandatory resident work hour limitations on medical students’ interest in surgery. J Am Coll Surg. 2004;199:615–19.
17 Chen J, Kamath R, Lam A, et al.How will changes in resident work hours affect medical students? Curr Surg. 2003;60:268–70.
18 De SK, Henke PK, Ailawadi G, Dimick JB, Colletti LM. Attending, house officer, and medical student perceptions about teaching in the third-year medical school general surgery clerkship. J Am Coll Surg. 2004;199:932–42.
19 Lapalio LR. Time study of students and house staff on a university medical service. J Med Educ. 1981;56:61–64.
20 Brown RS. House staff attitudes towards teaching. J Med Educ. 1970;45:156–59.
21 Association of American Medical Colleges (AAMC) Medical School Graduation Questionnaires, 1978–2003.Washington, DC: Association of American Medical Colleges, 2003.
22 Stillman PL, Regan MB, Swanson DB, et al. An assessment of clinical skills of fourth-year students at four New England medical schools. Acad Med. 1991;66:320–26.
23 Irby D. Teaching and learning in ambulatory care settings: a thematic review of the literature. Acad Med. 1995;70:3898–931.
24 Elnicki D, Kolarik R, Bardella I. Third-year medical students’ perceptions of effective teaching behaviors in a multidisciplinary ambulatory clerkship. Acad Med. 2003;78:815–19.