Patient safety strategies, such as duty hours restrictions (DHRs) for graduate medical education, are anticipated to improve health care quality, but such interventions may also have unintended negative consequences.1 DHRs attempt to reduce fatigue among clinicians in training with the intention of preventing errors and enhancing patient safety. Although DHRs have led to improved resident well-being2–4 and to possible, but not proven, benefits to patient safety,5 measures to ensure their implementation have also been associated with deleterious consequences. These include potential liabilities to patient safety in other areas, such as increased sign-over and more cross-coverage,6 and possible threats to resident education.3,4,7–10 In addition, some attendings believe that the new regulations are generating a new sort of physician, one that is less altruistic, less professional, and less dedicated to patients.7,9,10 Many faculty are also dissatisfied with the amount of work they are now expected to carry out to accommodate resident understaffing as a result of the new regulations.9–13
Within the context of DHRs, a new residency schedule was introduced on the internal medicine clinical teaching units (CTUs) at two University of Toronto-affiliated hospitals in 2009. Instead of the traditional team model in which each team takes call together, the new schedule features a more distributed model with each member of a team taking call on different nights. The new system was implemented to improve patient care, to comply with revised DHRs, and to increase faculty satisfaction; however, there were concerns about the overall effects of the new schedule and the potential for unexpected consequences.
Although there is a developing literature on the effects of DHRs, many related studies rely on self-administered survey data3,8,9,11,14 or focus on only one group of stakeholders.8–11 When training programs undergo significant modifications, performing initial open-ended exploratory analyses across different stakeholder groups may be more informative than collecting survey data from a single cohort. Qualitative methodology permits a more in-depth examination of perspectives and ensures that all experiences, including those that are unexpected, will be captured. Therefore, to determine how the new call structure was experienced by attending physicians, residents, and medical students, and to look carefully for unintended consequences, we performed an in-depth exploratory study using the qualitative method of grounded theory.
In Ontario, Canada, DHRs for graduate medical education were first introduced in 1996. Updated regulations, as of July 2009, stipulate that residents on in-hospital call for a 24-hour period must relinquish new responsibilities and complete handover within two hours post call.15 Residents are also guaranteed two weekends off within every four-week period. Unlike the most recent Accreditation Council for Graduate Medical Education (ACGME) regulations,16 Ontario DHRs currently apply equally across training levels, and there are no explicit weekly duty hours maximums.
The traditional CTU structure
Historically, the internal medicine CTUs at the University of Toronto teaching hospitals consisted of four teams, each made up of an attending physician, a senior resident (postgraduate year two or three), two or three junior residents (postgraduate year one; usually one internal medicine intern and one or two non-internal-medicine interns), and several medical students (third- or fourth-year). Residents typically rotated on internal medicine in four- or eight-week blocks, and attending physicians usually covered CTUs for four weeks at a time.
During time on the CTU, each team took call together on a 1:4 schedule, with a call day starting at 8:00 AM on day 1 and finishing at 12:00 PM on day 2 (see Chart 1). Throughout periods of residents’ absence, including during the postcall afternoon, attending physicians provided coverage. During their four-week blocks, attending physicians spent time in the hospital on the two weekends that their team was on call to learn about and round on new admissions. For the two weekends that their team was off call, attending physicians were also expected to come in to round on their entire team’s patients while the residents and students were off duty.
An informal internal review of this system in 2008–2009 raised three major areas of concern. First, the traditional system posed a possible threat to patient safety: Workload was not predictable or equitable because each team independently admitted all patients every fourth night, so patient numbers could fluctuate markedly depending on referrals, and there was significant cross-coverage by interns. Second, this system did not accommodate the updated 2009 provincial DHRs because large numbers of admissions to one team meant that it was often unfeasible to release residents from duties by 10:00 AM. Third, local reports suggested that the traditional structure promoted dissatisfaction with work–life balance amongst attending physicians while on CTUs, echoing findings in the literature.10,11 This model not only had a negative effect on attending physician morale, but, at a time when interest in general internal medicine as a career is declining amongst trainees,17–19 the traditional system was not fostering a desirable career model for trainees.
The new CTU structure
In the new CTU structure, implemented in January 2009 at one hospital, and March 2009 at the other, members of the team take call on different nights in a 1:4 schedule. The attending physician, senior resident, and students are on call together (as in the old system), but junior residents from the other three teams make up the rest of the on-call (nighttime) team. Each intern provides inpatient coverage for and only admits patients to his or her own team. All members of the on-call (nighttime) team review cases with the on-call senior resident; however, in the morning, nighttime team members review their respective patients with their own daytime team members and attending physician, resulting in a distribution of patients across all (daytime) teams. All postcall residents are relieved of duties as soon as they complete these patient reviews, before 10:00 AM, and return to work the following day as part of their daytime team. The two attending physicians whose teams are on call during a weekend cover admissions to the teams of their two colleagues, whose teams are not on call during that weekend, meaning that each attending has two full weekends each month during which he or she is completely off duty.
Although we anticipated advantages to the new system, it was impossible to predict all potential outcomes. Therefore, we wanted to examine the effects of the new system on students, residents, and attending physicians. We determined that an open-ended exploration across different stakeholder groups was the most suitable approach to ensure that all outcomes and experiences, especially unexpected ones, were captured.
We used the qualitative method of grounded theory to achieve our research objective. Grounded theory is useful when no relevant existing theory or framework exists, and the resulting analysis allows important themes to emerge from participants’ own observations and experiences.20 We chose focus groups as a means for collecting data because they encourage open commentary and permit an in-depth exploration of shared experiences.21 The experience of being on a CTU is a shared one, and focus groups provided the opportunity to facilitate discussion and input among participants. Because different levels of participants (attending physicians, senior residents, junior residents, and students) interact in a hierarchical relationship and also potentially have unique perspectives and particular sensitivities to a new system, we conducted separate focus groups for each level.
We invited attending physicians and residents who had experience of both the old and new internal medicine CTU structures at two University of Toronto–affiliated teaching hospitals to participate. We also invited medical students rotating on the CTUs to participate, but not all of them had experience of the old system. We recruited participants by e-mail, with responses returned anonymously to a research assistant (RA). Participation, which was voluntary, was remunerated with a small gift card to a bookstore. The study received approval by the University of Toronto research ethics board.
Between June and August 2009, we conducted 10 group or individual interviews with a total of 28 participants: attending physicians (n = 8), senior residents (n = 4), junior residents (n = 6; 4 internal medicine, 2 off service), and students (n = 10). Although we intended all discussions to be in focus group format, several residents were unable to participate at the last minute and rescheduled for later sessions. Therefore, three of the sessions only had one participant and were done as individual interviews. The remainder of the groups had between two and six participants. Each focus group lasted approximately 60 minutes and was audio-taped. An experienced RA (O.O.) facilitated each group using a scripted semistructured interview guide (Appendix 1) and follow-up probes as necessary. We developed the interview guide, piloted it on nonparticipant attending physicians and senior residents, and refined it on the basis of their comments. The RA transcribed the audio tapes verbatim and rendered them anonymous before we analyzed them.
We began analysis concurrently with data collection, consistent with grounded theory methodology, to establish that the focus groups were capturing the information that we anticipated and to consider potential new topics to include in interviewing subsequent groups. Each investigator read the initial transcripts during the open coding process and met regularly in a constant comparative process to discuss emerging themes. For the most part, participants had quite similar perspectives, and, therefore, our sample seemed fairly homogenous. However, we used iterative recruitment to add two focus groups, one senior-resident group and one junior-resident group, because these groups were initially smaller than others, and we were concerned about underrepresentation and reaching theoretical saturation amongst these levels.22 We continued to refine and challenge the coding structure in a process involving frequent meetings of our research team over several weeks until no further categories emerged from the data, consensus was reached, and the coding structure appeared stable; at this point, we had achieved theoretical saturation.22 The RA then entered the coding structure into NVivo qualitative data analysis software (QRS International Pty Ltd., Doncaster, Victoria, Australia) and coded all transcripts. During this time, the RA was in regular contact with the researchers to ensure accuracy and consensus in the coding.
Our analysis revealed six major thematic categories. Four of these were independent role-related categories—physician, manager, learner, and teacher—and two were non-role-related themes that influenced the role-related categories—workload and “teamness.” Within each of these thematic categories, participants observed significant trade-offs, noting both positive and negative effects of the change to a new system. A seventh theme, accountability, was a more overarching concept, or meta-theme, that had complex relationships with the other themes.
Physician. Focus group participants of all levels perceived the new system to be better for providing patient care and enhancing safety. The new system facilitated distribution of admissions across all four teams each day, rather than inundating one team every fourth day, so there were no longer highs and lows in admissions. As a result, participants perceived workloads to be more manageable, predictable, and equitable.
Participants also perceived fewer gaps and transitions in care and less cross-coverage by interns because the new system ensured that someone from each team was always in the hospital, and this individual only covered his or her own team’s patients. One junior resident observed that this arrangement made caring for patients on call much safer and easier because “You’ve heard about patients before, so when someone pages you about them overnight, then they are familiar, you know what their problems are, so it’s better.” One of the attending physicians commented that the new system also improved interactions with patients and their families because
There is someone in the hospital who is reasonably knowledgeable about the case, who clearly has the responsibility to talk to that family, whereas in the old system I would tell people not to talk to families.
Overall, participants perceived an increased continuity of care for patients. The increase came, however, at the level of the collective team responsibility for care, and, as one senior resident noted, came at the expense of lost continuity of care at the individual physician level: “I think continuity of care has actually improved. My own continuity of care may have declined, but for patients I think it has improved.” Senior residents were the most vulnerable to this loss.
Manager. Both senior and junior residents perceived that the new system enhanced their management skills. Most junior residents relished their new added responsibility for the team when the senior resident was absent or post call. Senior residents also felt more comfortable and confident in delegating duties to others on the postcall days and, therefore, in being able to leave the hospital by close to 10:00 AM on those days. This was because, as one senior resident observed,
If you know that you are the only one going and there are going to be four or five other people there, it’s much easier to let go of these tasks, and you say, “Here is what you need to do for the day,” and you can trust that it’s going to be done.
Senior residents also appreciated developing skills in triaging patients across teams in the emergency room and in being able to assume care efficiently and effectively for patients admitted to their team daily by interns who were on call. However, acquiring these new skills caused significant stress for some senior residents, and this was underappreciated by attendings. The stress came largely from a lack of clear guidelines from attending physicians about triaging, and, as one senior resident noticed, the new system created pressure to please four attendings rather than one: “In the old system you learned to adjust to the staff’s style and learn the way that they worked. In the new system there are too many staff to be able to adjust this way.”
Learner. Although participants noted benefits for learning managerial skills in the new system, there were other negative effects on learning. All participants, from students to attendings, identified that the most significant negative effect on learning occurred for senior residents. In the new system, approximately two-thirds of patients that senior residents see in the emergency room are admitted to other teams, meaning that they miss out on the educational experience of following these patients throughout their admission. One senior resident observed, “you do lose out on the learning potential … it’s important to have the follow-up to know whether what you did was right….”
Teacher. Although the new system afforded greater time for postcall rounding on each new patient because there were fewer patients admitted, attending physicians felt more pressure to teach mostly around these cases because it was the only time that the whole team was together. In addition, morning teaching rounds were often cut short to accommodate scheduled meetings with other health care professionals.
There were other challenges to teaching in the new system. Because of the staggered call structure, attending physicians felt that they had to teach daily so that no one would miss out. Residents also had more difficulty with some aspects of teaching in the new system. Senior residents had less time to teach their daytime junior residents, and junior residents had less time to teach their daytime students, because they were on staggered schedules and interacted infrequently. Senior residents also sometimes found it difficult to know the learning needs of their on-call or nighttime junior residents because they did not share daytime teaching experiences.
Workload. Overall, participants did not perceive that the amount of work changed in the new system, but they noted that it was more evenly shared. There were trade-offs in this new model, though. For example, although the amount of work was more predictable, there was less flexibility, which affected attending physicians’ schedules and team teaching activities. Also, although participants clearly perceived the postcall day to be more manageable, they noted that the Monday after a weekend when their team was not on call seemed chaotic because their interns had been covering their team’s weekend admissions with another team’s attending. However, one attending physician found the arrangement tolerable: “I am certainly happy to have the weekends off, and I don’t have the postcall 20 patients, 10 of them new.” Lastly, although participants generally considered workload to be well distributed, they perceived deterioration in the efficiency and function of remaining team members during times of understaffing, such as when individuals were on vacation or away at protected academic activities, due to the increase in relative work. Participants perceived that this was more apparent in the new system versus the old.
“Teamness.” One of the greatest concerns before restructuring was the effect that the new staggered schedule might have on the camaraderie considered to be inherent in the old team system. However, residents and students considered the sense of team to be sufficient in the new system. As one junior resident explained, “It’s kind of like having two teams (day team and night team). I actually feel like that’s great; one advantage is that I get to know more people this way.”
Although we did not explicitly label it, a more general concept of accountability emerged from all levels of focus group discussions and was interwoven through the other six themes. Participants perceived that overall accountability increased in the new system, but the manifestation of this was different by level.
Junior residents seemed to experience heightened accountability to their own team, or intrateam accountability. This seemed to be largely a result of junior residents only covering their own team while on call and being responsible for the management of their team during their senior resident’s absence. One attending physician observed that different opportunities for giving and receiving feedback in the new system may have influenced junior residents’ intrateam accountability:
The thing about ownership and cross-coverage is this system creates a lot more feedback. In the old system if there was cross-coverage and things weren’t done, that information never got fed back. Now it’s always fed back, every day.
This feedback seemed to drive further investment in and ownership of patient care.
On the other hand, attending physicians and senior residents seemed to have greater accountability toward colleagues on other teams because they were responsible for triaging patients and admitting patients to other teams in addition to their own. One attending observed that, in the new system,
The way I manage consults has changed a little bit. I feel more of a responsibility to make sure that all the consults are appropriate…. [I]t does mean that as staff we are more responsible to one another about the decisions that we make overnight because it’s that other person who is going to take care of the patient.
Compared with junior residents, senior residents received significantly less direct feedback about their performance. Although they wanted feedback about their admissions to other teams, there was no mechanism for this to routinely take place, so senior residents were left to seek this feedback informally.
Senior residents’ diligent follow-up with residents and attending physicians on other teams highlighted another common thread among many of the discussions: that participants’ greatest priority was providing good patient care. Participants’ responses often reflected a sense of altruism, in particular at the resident level. Although senior residents could have forgotten about patients once they were relegated to other teams, the majority went out of their way to follow up, not only for their own learning but also to ensure that they had cared for the patient well. Similarly, senior residents talked about the difficulty when triaging in reconciling a desire to admit interesting cases to their own team with a desire to avoid “dumping” cases on their colleagues on other teams.
Although this new system was in part developed to accommodate new DHRs, residents reflected that the new system also made it easier to follow DHRs. They talked about leaving at 10:00 AM on postcall days only because they felt safe doing so under the new system. Senior residents described feeling uneasy signing over patients to a cross-covering intern in the old system and staying until 2:00 or 3:00 PM even when they were required to finish by noon because “You tried really hard to clean everything up because you just couldn’t leave that intern looking after things when they don’t know anyone.”
Overall, in considering the trade-offs and negatives identified with the new system, residents favored the new system largely because of its advantages for patients. Even amongst senior residents who lost out on learning opportunities, one observed, “The benefits of having reduced gaps in coverage and the benefit of having more predictable patient admissions and discharges make up for whatever is lost.”
Participants perceived that the new system provided an overall benefit to patient care, though at the expense of several trade-offs. Workload was more predictable and equitable but was less flexible; for example, attending physicians could no longer attend morning clinics or meetings while on service. The postcall day was less hectic, but the postweekend Monday after a team was off was more chaotic. Senior residents sacrificed continuity of patient care on an individual level, as they no longer knew patients from the emergency room through discharge, but there was increased team continuity of care because someone who knew the patient was always in the hospital. Junior residents worked harder on weekends rounding on their own teams, but they did not have to cross-cover patients that they did not know.
Both attending physicians and residents recognized these tensions, yet they overwhelmingly reported that the benefits outweighed what was lost. The residents in particular were altruistic in valuing improved care and safety for patients over their own experience. Recent reports in the literature have begun to document similar findings,23,24 but in general this is contrary to much of what has previously been described. The implementation of ACGME DHRs in 2003 led to a flurry of editorials about and studies of the effects of DHRs on residents’ professionalism. Many described significant worries about an erosion of professionalism, in particular related to decreased continuity of care for patients with a move toward a shift mentality.7,10 Concerns were voiced by faculty across specialties,7,9,10,13 and residents themselves acknowledged less continuity of care.4
In contrast, our study suggests that, at least from participants’ points of view, continuity of care appeared to be enhanced. This was particularly true for interns, who admitted and provided inpatient coverage only to patients on their own team. This enhancement seemed to be a result of staggering team members’ presence in the hospital, which was perceived to increase accountability in patient care and strengthen the concept of “team coverage.” Junior residents’ accountability to their own teams meant that senior residents felt comfortable leaving earlier on their postcall days, and attending physicians felt that it was much easier to identify poorly performing interns. Senior residents and attending physicians also adjusted their performance as they felt more accountable to their counterparts on other teams.
Participants may have perceived an increase in accountability for several reasons. First, the 2009 DHR updates did not involve significant changes to existing DHRs. Duty hours have been restricted in Ontario since 1996, so the trainees and faculty who participated in the study did not experience an abrupt change. Second, in light of constrained duty hours, the new call structure was designed to enhance coverage of patients. Unlike systems in which night floats have been used to increase coverage, our call system included residents who also had daytime responsibility to their patients and team. This model of care has previously been suggested as a mechanism to improve personal accountability.25 Third, rather than just avoiding a “shift mentality,” our system may have actually fostered an increased identity of continuity of patient care on the team level compared with the old system, which still focused primarily on individual physicians’ continuity of patient care. This concept of collective responsibility for patient care and a team approach to continuity has been described previously.23,26,27 An attending confirmed this by commenting, “I think the biggest change with this structure is the mechanism of teambuilding … now [team members] have shared responsibility for the assigned constellation of patients.” Fourth, it may be that residents are genuinely more altruistic and accepting of trade-offs than previous reports suggest if patient care improves. More recent studies, based on ethnographic observations of actual resident behavior, would support this.24
It is important to note that our new system was not beneficial to all aspects of continuity. Senior residents did not have continuity of care for all the patients they saw in the emergency room throughout the patients’ admission. In addition, continuity of care also suffered when residents attended other activities or were on vacation, though this also frequently occurred in the old system. Again, residents appeared to be altruistic and made personal sacrifices to care for patients during these periods, such as missing teaching rounds to cover for absent colleagues. This illustrates the tension that trainees experienced among conflicting professional obligations. Evidence shows that conflicts in duties may drive trainees to behave in perceived unprofessional ways,24 such as missing educational rounds, but that most residents seem to measure thoughtfully how to handle this “fine balance.” Rather than impose rigid rules, there may be a need to allow for more flexibility for our residents to develop and display their developing professional judgment.24,25,28
Although continuity of care seemed to improve in the new system, our findings revealed significant unintended negative consequences on education, especially learning opportunities for senior residents and challenges to timely and cohesive teaching. In the interim, weekly senior resident rounds have been organized to give residents a forum to discuss cases and get feedback from attendings across teams; however, this format does not allow real-time review of cases and immediate feedback. It may be necessary to further change practices to promote educational opportunities while maintaining improved continuity of patient care. Examples might include faculty staying later into the evening,29 doing teaching sessions at less conventional times,30 or changing supervision patterns.16,31
Any proposed changes must be mindful of placing further demands on attending physicians. Our study found that faculty satisfaction improved with relatively small adjustments to a schedule adhering to DHRs—increased predictability (but not volume) of workload and two weekends off per month. Increasing demands in other areas must not be allowed to undo this benefit. Faculty satisfaction is important not only for faculty well-being but also for the role modeling provided to trainees.11,18,19 Previous in-depth interviews report that residents do appreciate that preferentially protecting residents may have a deleterious effect on attendings.24
There are several limitations to our study. Although we reached theoretical saturation with our data collection, participation was voluntary, and therefore it is possible that those who did not participate had somewhat different views about the system change. In addition, this is a single-institution study, and our findings may not apply to different educational settings. Particularly important is that this study occurred in a Canadian province in which DHRs still permit 24 plus 2 consecutive hours of call; this is different from regulations in some other parts of Canada, such as Quebec,32 and the United States.16 In these systems, at least some residents are restricted to 16 hours of work, so the system described in our study may be less feasible in other settings. In the future, our own province may also adopt similar restrictions and, therefore, challenge the implementation of the system described here.
In summary, the experience at our institution provides an in-depth examination of a call structure change within restricted duty hours. The change was perceived to improve continuity of care and accountability, in part by enhancing team responsibility for patient care. However, the educational effects of the new structure were significant. Despite this, residents’ views were refreshingly flexible and altruistic because they felt that the new system was overall better for patient care. Program leaders must ensure that residents’ sacrifices are not too great and that their education remains a priority.
Funding/Support: This research was supported by the Department of Medicine Professors Research and Education Fund, University of Toronto, and the Education Development Fund, Faculty of Medicine, University of Toronto.
Other disclosures: None.
Ethical approval: This study was approved by the University of Toronto research ethics board.
Previous presentations: Parts of this research were presented at the Association for Medical Education in Europe Conference, Glasgow, Scotland, September 6, 2010 (poster); at the International Conference on Residency Education, Ottawa, Ontario, Canada, September 24, 2010 (oral); and at the Research in Medical Education Conference, Association of American Medical Colleges annual meeting, Washington, DC, November 10, 2010 (oral).
1. Ryan J. Unintended consequences: The Accreditation Council for Graduate Medical Education work-hour rules in practice. Ann Intern Med. 2005;143:82–83
2. Fletcher KE, Underwood W 3rd, Davis SQ, Mangrulkar RS, McMahon LF Jr, Saint S. Effects of work hour reduction on residents’ lives: A systematic review. JAMA. 2005;294:1088–1100
3. Goitein L, Shanafelt TD, Wipf JE, Slatore CG, Back AL. The effects of work-hour limitations on resident well-being, patient care, and education in an internal medicine residency program. Arch Intern Med. 2005;165:2601–2606
4. Myers JS, Bellini LM, Morris JB, et al. Internal medicine and general surgery residents’ attitudes about the ACGME duty hours regulations: A multicenter study. Acad Med. 2006;81:1052–1058
5. Fletcher KE, Davis SQ, Underwood W, Mangrulkar RS, McMahon LF Jr, Saint S. Systematic review: Effects of resident work hours on patient safety. Ann Intern Med. 2004;141:851–857
6. Logio LS, Djuricich AM. Handoffs in teaching hospitals: Situation, background, assessment, and recommendation. Am J Med. 2010;123:563–567
7. Fischer JE. Continuity of care: A casualty of the 80-hour work week. Acad Med. 2004;79:381–383
8. Horwitz LI, Krumholz HM, Huot SJ, Green ML. Internal medicine residents’ clinical and didactic experiences after work hour regulation: A survey of chief residents. J Gen Intern Med. 2006;21:961–965
9. Peterson LE, Johnson H, Pugno PA, Bazemore A, Phillips RL Jr.. Training on the clock: Family medicine residency directors’ responses to resident duty hours reform. Acad Med. 2006;81:1032–1037
10. Reed DA, Levine RB, Miller RG, et al. Effect of residency duty-hour limits: Views of key clinical faculty. Arch Intern Med. 2007;167:1487–1492
11. Arora V, Meltzer D. Effect of ACGME duty hours on attending physician teaching and satisfaction. Arch Intern Med. 2008;168:1226–1228
12. Coverdill JE, Finlay W, Adrales GL, et al. Duty-hour restrictions and the work of surgical faculty: Results of a multi-institutional study. Acad Med. 2006;81:50–56
13. Coverdill JE, Adrales GL, Finlay W, et al. How surgical faculty and residents assess the first year of the Accreditation Council for Graduate Medical Education duty-hour restrictions: Results of a multi-institutional study. Am J Surg. 2006;191:11–16
14. Mathis BR, Diers T, Hornung R, Ho M, Rouan GW. Implementing duty-hour restrictions without diminishing patient care or education: Can it be done? Acad Med. 2006;81:68–75
15. Professional Association of Internes and Residents of Ontario (PAIRO).. Agreement Between: The Professional Association of Internes and Residents of Ontario (PAIRO) and the Council of Academic Hospitals of Ontario (CAHO). Re: Terms and conditions of employment of residents in Ontario teaching hospital. July 1, 2008 to June 30, 2011.. http://www.pairo.org/Content/Default.aspx?pg=1003
. Accessed June 12, 2012.
16. . Accreditation Council for Graduate Medical Education. ACGME Common Program Requirements. 2012 http://www.acgme.org/acWebsite/home/Common_Program_Requirements_07012011.pdf
Accessed June 12
17. Arora V, Wetterneck TB, Schnipper JL, et al. Effect of the inpatient general medicine rotation on student pursuit of a generalist career. J Gen Intern Med. 2006;21:471–475
18. Garibaldi RA, Popkave C, Bylsma W. Career plans for trainees in internal medicine residency programs. Acad Med. 2005;80:507–512
19. Hauer KE, Durning SJ, Kernan WN, et al. Factors associated with medical students’ career choices regarding internal medicine. JAMA. 2008;300:1154–1164
20. Strauss AL, Corbin J Basics of Qualitative Research: Techniques and Procedures for Developing Grounded Theory. 1998 Thousand Oaks, Calif Sage Publications
21. Creswell JW. Data collection. In: Qualitative Inquiry and Research Design. 1998 Thousand Oaks, Calif Sage Publications:109–135
22. Creswell JW. Data Analysis and Representation. In: Qualitative Inquiry and Research Design. 1998 Thousand Oaks, Calif Sage Publications:139–165
23. Coverdill JE, Carbonell AM, Fryer J, et al. A new professionalism? Surgical residents, duty hours restrictions, and shift transitions. Acad Med. 2010;85(10 suppl):S72–S75
24. Szymczak JE, Brooks JV, Volpp KG, Bosk CL. To leave or to lie? Are concerns about a shift-work mentality and eroding professionalism as a result of duty-hour rules justified? Milbank Q. 2010;88:350–381
25. Lopez L, Katz JT. Perspective: Creating an ethical workplace: Reverberations of resident work hours reform. Acad Med. 2009;84:315–319
26. Park J, Woodrow SI, Reznick RK, Beales J, MacRae HM. Patient care is a collective responsibility: Perceptions of professional responsibility in surgery. Surgery. 2007;142:111–118
27. Van Eaton EG, Horvath KD, Pellegrini CA. Professionalism and the shift mentality: How to reconcile patient ownership with limited work hours. Arch Surg. 2005;140:230–235
28. Rybock JD. Residents’ duty hours and professionalism. N Engl J Med. 2009;361:930–931
29. Frost P, Wise M. Residency duty-hour limits: Time to change how key clinical faculty work? Arch Intern Med. 2008;168:244
30. Fields RC, Bowman MC, Freeman BD, Klingensmith ME. Implementation of an “after hours” resident educational program in a general surgery residency: A paradigm for increasing formal didactic training outside of the hospital setting in the era of the 80-hour workweek. J Surg Educ. 2009;66:340–343
31. Temple J Time for Training: A Review of the Impact of the European Working Time Directive on the Quality of Training. May 2010 London, United Kingdom National Health Service http://www.mee.nhs.uk/PDF/14274%20Bookmark%20Web%20Version.pdf
. Accessed June 21, 2012