Koch, Paul E. MD; Simpson, Deborah PhD; Toth, Heather MD; Marcdante, Karen MD; Densmore, Emily MD; Young, Staci PhD; Weisgerber, Michael MD; Morzinski, Jeffrey A. PhD; Havas, Nancy MD
Dr. Koch is assistant professor of family and community medicine, Medical College of Wisconsin, Milwaukee, Wisconsin.
Dr. Simpson is adjunct professor of family and community medicine, Medical College of Wisconsin, and medical education program director, Aurora Health Care, Milwaukee, Wisconsin.
Dr. Toth is associate professor of medicine and pediatrics, Medical College of Wisconsin, Milwaukee, Wisconsin.
Dr. Marcdante is professor of pediatrics, Medical College of Wisconsin, Milwaukee, Wisconsin.
Dr. Densmore is assistant professor of pediatrics, Medical College of Wisconsin, Milwaukee, Wisconsin.
Dr. Young is assistant professor of family and community medicine, Medical College of Wisconsin, Milwaukee, Wisconsin.
Dr. Weisgerber is associate professor of pediatrics, Medical College of Wisconsin, Milwaukee, Wisconsin.
Dr. Morzinski is associate professor of family and community medicine, Medical College of Wisconsin, Milwaukee, Wisconsin.
Dr. Havas is associate professor of family and community medicine, Medical College of Wisconsin, Milwaukee, Wisconsin.
Funding/Support: This project was partially supported by a U.S. Department of Health and Human Resources Faculty Development in Primary Care award #D55HP00093.
Other disclosures: None reported.
Ethical approval: The project was granted exempt status by the Medical College of Wisconsin institutional review board.
Previous presentations: Results of this qualitative analysis were presented at the Association of American Medical Colleges Annual Meeting, November 4–9, 2011, Denver, Colorado.
Correspondence should be addressed to Dr. Koch, Medical College of Wisconsin / Department of Family and Community Medicine, 1121 E. North Ave., Milwaukee, WI 53212; telephone: (414) 267-6541; e-mail: email@example.com.
The Institute of Medicine has estimated that anywhere from 44,000 to 98,000 patient deaths are attributable to preventable medical errors annually,1 and the Joint Commission estimates that 80% of serious medical errors involve miscommunication between caregivers when patients are transferred or handed off.2 Today’s health care system, which involves increased multidisciplinary care, adds to communication complexity. Patient transitions occur between care locations (e.g., acute care hospitals, long-term care facilities, rehabilitation centers, home) and among providers (e.g., primary care/specialty physicians, nurses, other providers). Each care transition has the potential for error, especially when different care providers are involved in multiple settings. Further, resident duty hours restrictions have resulted in an increased number of transitions among providers, raising resident and program director concern about the higher potential for error.3 The new care model known as the patient-centered medical home (PCMH) is intended to provide a central location for coordinating patient care and follow-up on care plans; however, information does not always flow easily to and from the PCMH.4–7
Medical students observe and participate in a variety of care transitions in both the hospital and ambulatory settings. In 2011, the authors of a systematic review of transitional care curricula for physicians-in-training and independently practicing physicians found 25 reports describing educational interventions published between January 1973 and June 2010.8 The majority of the reports (56%) were published in 2009 or later, and 63% of the interventions focused on students in their clinical years of medical school (typically their third or fourth year).8 However, our literature review revealed a paucity of publications which specifically detail the methods for successfully teaching medical students about effective care transitions.9–11 Although 62% of U.S. Liaison Committee on Medical Education–accredited medical schools report teaching about quality and safety,12 we could find no published reports describing how medical students’ experiences of care transitions as learners and/or as health care providers have informed their instruction.
Students’ perceptions and experiences influence the degree to which the instruction they receive transfers to clinical practice.13 Effective instructional design includes needs assessment, a step through which educators identify content gaps, skill deficiencies, and the key learner characteristics (e.g., self-efficacy, anxiety/negative affectivity, perceived utility) that influence transfer of learning to clinical settings. An effective needs assessment results in data-driven objectives and instruction appropriate to the targeted learner audience (e.g., medical students).14
In this study, we sought an enriched understanding of third-year medical students’ experiences with care tran sitions, which would, in turn, inform care-transition instructional design to maximize transfer of learning to clinical practice.
We selected a qualitative methodological approach15 to analyze information regarding how medical students experience care transitions. Researchers across health care professions and in medical education have used a constant comparative approach to explore teaching effectiveness, professionalism,16 and student perspectives on patient-centered care as conveyed in critical incident (CI) descriptions.17 We selected student-authored descriptions of CIs as our data collection approach to garner unfiltered student perspectives on care transitions.
In April 2010, 193 third-year medical students at the Medical College of Wisconsin completed a mandatory two-hour intersession on care transitions and patient handoffs.18 In the intersession’s didactic portion, we gave an overview of care-transition definitions and methods, framed in the context of patient safety and professional communication. Then we asked each student to complete a CI form describing a care transition he or she had witnessed that evoked a strong emotional response. The deidentified narratives included the setting of the care transition, the type and level of participating care providers (e.g., nurses, residents, attending physicians), and nonidentifying patient information including diagnosis, context, and outcome. We then introduced the PRIMARY care-transitions mnemonic (People, Reason, Input, Medical course, Assessment, Recommendations and responsibilities, Your turn; see also Table 1) and asked the students to practice communicating their CIs with one another using this mnemonic. At the end of the intersession, we collected the incident descriptions, which included no information identifying students or patients.
Next, we (clinical and social science faculty with MDs and/or PhDs) met and reviewed the constant comparative coding methodology.15 We then developed criteria for determining CI care-transition status: failed, successful, or indeterminate. The two criteria for failed transitions were (1) the description included a patient care transition with reported communication between sender and receiver, and (2) per the medical student, a key element was missing or poorly done, resulting in an unsuccessful transition (e.g., potential adverse clinical outcome, time delay, confusion). We designated successful care transitions as those in which a care transition was completed between sender and receiver, and there were no student-reported gaps in patient care. The reasons we deemed CI transitions to be indeterminate were as follows: The descriptions conveyed a CI in which a transition between a sender and receiver did not occur (e.g., the sender or receiver was unable to contact the other for the transition), the descriptions were composed in indecipherable handwriting, and/or the descriptions lacked sufficient information about the incident to make a sound judgment. We used the “PRIMARY” care-transitions mnemonic18 to code which care-transition stage the student identified as critical. As the PRIMARY care-transitions mnemonic outlines sequential stages, we used the first stage at which the transition went awry to characterize the transition stage.
Consistent with constant comparative methodology,15 we met again to review and collaboratively code 10 randomly selected CI descriptions, which resulted in an initial descriptive codebook (e.g., categories and associated data codes). Then, in pairs, we independently coded 10 to 20 randomly assigned CI descriptions. We used memos to document data, and we coded each CI description with multiple memos to record data codes (e.g., emotion evoked, situation, participants) and the appropriate PRIMARY stage. Coding pairs met to reconcile differences. Collectively, we resolved any remaining differences in CI judgment (successful, failed, or indeterminate), in data codes, and/or in PRIMARY transition stage designation.
After coding, we collated the CI descriptions by transition criterion (i.e., successful/failed/indeterminate) and by the depicted PRIMARY transition stage. Then, we organized the memos into broad categories, including care-transition setting and patient description, providers involved, and students’ affective responses (see also Results). Finally, we summed the CI descriptions in these categories and converted them to percentages.
The project was granted exempt status by the Medical College of Wisconsin institutional review board.
Transition status and PRIMARY care-transition stage
Transition status was either unavailable or inadequately described in 42% (82/193) of student CI narratives and thus not coded. Of the remaining 111 CI descriptions, 64% (n = 71) conveyed failures, 13% (n = 14) described successes, and 23% (n = 26) were deemed indeterminate because of incomplete transitions.
We were able to determine the transition stage, coded using the PRIMARY mnemonic, depicted by students in 73% of the CI descriptions (141/193). Care-transition stage coding and analysis revealed that over 50% (n = 78) of the 141 care-transition communication issues focused on two PRIMARY stages: 35% (n = 50) lacked clarity regarding sender/receiver roles and responsibilities (second R of mnemonic), and 20% (n = 28) lacked details about the patient’s medical course (M of the mnemonic). The five remaining care-transition stages accounted for the remainder of the incidents (see Table 1): identification of the people (P) involved in the transition (15% [n = 21]), assessments (A) of what needed to be done for the patient (13% [n = 18]), reasons (first R) for transition (8% [n = 11]), input (I) from the receiver regarding his or her knowledge of and caregiving relationship with the patient (6% [n = 8]), and addressing the receiver’s or your (Y) understanding of the care plan (3% [n = 5])
Students’ experiences with care transitions
At the time of the intersession, students had completed both inpatient and outpatient rotations, and their CI descriptions reflected that. About two-thirds of the students (69%; 134/193) identified the setting where the care transition was initiated. The majority of transitions (92%; 123/134) were initiated in the inpatient setting; only 8% (11/134) originated in the outpatient setting. Students usually identified the receiver in the transition: 70% of incidents (135/193) specified the receiver. Often—88% of the time (119/135)—the receiver was described as another inpatient caregiver. Only 12% of the CI descriptions (16/135) involved transitions to an outpatient setting.
We could clearly identify a sender in 39% (75/193) of the CI descriptions, and coincidentally, we could also identify a clear receiver in 39% (n = 75) of the descriptions; however, notably, not all descriptions mentioned both a sender and a receiver. Students reported residents and fellows as the individuals most commonly engaged in transitions (59% [44/75] as senders; 63% as receivers [47/75]), followed by attending physicians (15% [11/75] as senders; 21% as receivers [16/75]). Others involved in transitions, according to student descriptions of CIs, were as follows: medical students (16% [12/75] as senders; 7% [5/75] as receivers), nurses (8% [6/75] as senders; 8% [6/75] as receivers), and community physicians (3% [2/75] as senders; 1% [1/75] as receivers). The sum of the percentiles exceeds 100%, as multiple individuals may have been reported in some transitions as either senders or receivers.
Most of the CI descriptions (63%; 121/193) conveyed strong emotional responses, and of these 121, 92% (n = 111) were negative. The negative emotions students wrote about most frequently were frustration (mentioned in 41% [39/94] of the descriptions), anger (16% [15/94]), and annoyance (9% [8/94]). Some of the descriptions (14% [17/121]) also conveyed compound emotions, such as frustration and irritation, shock and anger; and some (14% [n = 13/94]) depicted disturbing emotions, such as horror, fear, or regret. For example, one student described his/her emotions on receiving an inadequate handoff from another student on inpatient service thusly: “I was disappointed and, at the time, frustrated.” Another student described his/her emotions at observing a “rushed” evening resident-to-resident inpatient handoff as “a game of telephone.” He or she described feeling “Helpless to intervene, but annoyed at the ineffectiveness of the process.” Importantly, a few of the students did express positive emotions; pride and appreciation were among the most common of these. We have provided additional descriptions of the transition situation and emotions evoked (organized by their PRIMARY stage) in Table 2 to illustrate the range and depth of students’ emotional reactions.
Discussion and Conclusions
CI descriptions authored by third-year medical students reveal these students’ experiences with and reactions to care transitions. Many of the CI descriptions focused largely on miscommunication or omitted communication between senders and receivers. Students perceived that the majority of communication issues centered on establishing clear roles and responsibilities (i.e., who will do what) between transition sender and receiver, and on providing appropriate medical information so that the receiver is clear about the patient’s condition. According to these CI descriptions, the providers most actively represented in the care-transition process are residents and fellows—a finding that indicates an opportunity to influence medical student transition competence and attitudes.
Although collectively, the third-year students wrote about all stages of the PRIMARY transition process, two stages accounted for the majority of the incidents. Confirming roles and responsibilities (the second R) for ongoing care and follow-up was the most frequent problem, followed by failing to provide all of the pertinent clinical details (M / Medical course). Although the medical community accepts that clear and accurate communication of the patient’s condition and the explicit transfer of responsibility and authority are key components of any transition,19 students related that these important stages in transitions are sometimes omitted. Their frustration, anger, fear, and/or annoyance, expressed in the CI descriptions, may indicate that they do not feel empowered to act—even when they recognize the need for improved communication.20
These negative emotions may also highlight the fact that best-remembered events are often those that fail. Recent research has confirmed that stimuli evoking strong emotions are more accurately remembered21; additionally, some research may indicate that negative events may be remembered more often than positive events.22 Thus, the use of student-authored CI descriptions may be a powerful instructional strategy to motivate students to learn and to engage them in enhancing transfer. Indeed, reflective learning in a patient safety course for senior medical students that has used incident report cards revealed that students attended more closely to details that could and did go wrong, which, in turn, positively influenced their intention to behave more safely.20 Student-authored CI reports may be incorporated into clinical instructor training to better prepare students to plan for and manage care transitions in student teaching settings.
Although the results of this study provide a robust description of the emotional impact of care transitions experienced by 193 medical students, we acknowledge certain limitations. The incidents were gathered at one point in time, from one medical student class at a single medical school. Although students at our institution participated in both inpatient and ambulatory rotations, 92% of the incidents were inpatient focused. Students may not witness care transitions in the ambulatory setting as frequently, nor view them as care transitions per se (e.g., referring a patient from the office to a subspecialist, a physical therapist, or a mental health counselor). The paucity of outpatient events in our data set may also be due to the lack of feedback students receive about transitions once the patient is discharged from the acute care hospital or the student has left the inpatient service. Scarce knowledge of follow-through would limit students’ reactions to and recall of such events, decreasing the likelihood that they would report them in their CI descriptions. Given the current nature of medical care, specifically the growing ambulatory component, this study provides little information about whether students perceive inpatient and outpatient transitions in the same light. Further study may help to answer this question.
As the medical education community moves to a competency- and performance-based medical student educational system, we must design instruction to optimize our trainees’ ability to transfer what they learn to clinical practice. Our findings reveal that experiences of care transitions evoke strong negative emotions in our third-year students. Students recognize transition problems but perceive that they are unable to influence care-transition success. Thus, training in patient-transfer skills within clinical settings will be influenced by the degree to which instruction explicitly addresses these perceptions.13 For example, training students to use a structured transition mnemonic could serve a valuable role in identifying and filling care-transition stage gaps. However, to maximize the chances of effective patient transfer, the training must also address how to overcome subjective norms (e.g., relationships with supervisors, team culture),20 and it may need to occur in conjunction with residents/fellows, who, as the most frequent transition communicators in the CI descriptions our students wrote, play a crucial role in team culture.
This study’s findings allow educators to take the next step to ensure safe patient care transitions in their institutions. By incorporating learner needs and experiences during the design of instruction, transfer of care-transition skills from classroom to clinic can be enhanced.
Acknowledgments: The authors wish to thank Tess Chandler for coordinating this project and the Safe Transitions for Every Patient (STEP) Collaborative for working on the STEP curriculum.
1. Institute of Medicine. To Err Is Human: Building a Safer Health System. 2000 Washington, DC National Academies Press
3. Wohlauer MV, Arora VM, Horwitz LI, Bass EJ, Mahar SE, Philibert IHandoff Education and Assessment for Residents (HEAR) Computer Supported Cooperative Workgroup. . The patient handoff: A comprehensive curricular blueprint for resident education to improve continuity of care. Acad Med. 2012;87:411–418
4. Kripalani S, LeFevre F, Phillips CO. Deficits in communication and information transfer between hospital-based and primary care physicians implications for patient safety and continuity of care. JAMA. 2007;297:831–841
5. Forrest CB, Glade GB, Baker AE, Bocian A, von Schrader S, Starfield B. Coordination of specialty referrals and physician satisfaction with referral care. Arch Pediatr Adolesc Med. 2000;154:499–506
6. Gandhi TK, Sittig DF, Franklin M, Sussman AJ, Fairchild DG, Bates DW. Communication breakdown in the outpatient referral process. J Gen Intern Med. 2000;15:626–631
7. Schoen C, Osborn R, Huynh PT, et al. Primary care and health system performance: Adults’ experiences in five countries. Health Aff (Millwood). 2004(suppl Web exclusives):W4–W487
8. Buchanan IM, Besdine RW. A systematic review of curricular interventions teaching transitional care to physicians-in-training and physicians. Acad Med. 2011;86:628–639
9. Ouchida K, LoFaso VM, Capello CF, Ramsaroop S, Reid MC. Fast forward rounds: An effective method for teaching medical students to transition patients safely across care settings. J Am Geriatr Soc. 2009;57:910–917
11. Chu ES, Reid M, Burden M, et al. Effectiveness of a course designed to teach handoffs to medical students. J Hosp Med. 2010;5:344–348
12. Anderson MB, Kanter SL. Medical education in the United States and Canada, 2010. Acad Med. 2010;85(9 suppl):S2–S18
13. Burke LA, Hutchins HM. Training transfer: An integrative literature review. Hum Resour Dev Rev. 2007;6:263–296
14. Kern DE, Thomas PA, Hughes MT Curriculum Development for Medical Education—A 6-Step Approach. 2009 Baltimore, Md Johns Hopkins University Press
15. Harris IBNorman G, van der Vleuten C, Newble D. Qualitative methods. International Handbook of Research in Medical Education. 2002 Part I. Dordrecht, The Netherlands Kluwer Academic Publishers:45–96
16. Rademacher R, Simpson D, Marcdante K. Critical incidents as a technique for teaching professionalism. Med Teach. 2010;32:244–249
17. Bower DJ, Young S, Larson G, et al. Characteristics of patient encounters that challenge medical students’ provision of patient-centered care. Acad Med. 2009;84(10 suppl):S74–S78
19. Patterson ES, Wears RL. Patient handoffs: Standardized and reliable measurement tools remain elusive. Jt Comm J Qual Patient Saf. 2010;36:52–61
20. deFeijter JM, deGrave W, Hopmans EM, Koopmans RP, Scherpbier AJ. Reflective learning in a patient safety course for final-year medical students. Med Teach. 2012;34:946–954
21. Ochsner KN. Are affective events richly “remembered” or simply familiar? The experience and process of recognizing feelings past. J Exp Psych Gen. 2000;129:242–261
22. Kensinger EA, Corkin S. Effect of negative emotional content on working memory and long-term memory. Emotion. 2003;3:378–393