In medical school and subsequent clinical training, emotional development is an essential part of professional skill acquisition for physicians.1–3 A broadly inclusive term, emotional development incorporates familiar elements, such as emotional intelligence and empathy. Recently, however, the need for a more comprehensive description of emotional development has been recognized.4 One new, multidimensional model of emotional development incorporates four domains in particular: emotional strength (e.g., empathy), emotional intelligence, emotional resilience, and emotional regulation.5 Further studies are needed to define the essential domains of emotional development, but on the most basic level, little is known about the spectrum of experiences that shape emotional development in the course of regular medical education.
Emotional development can be examined through the lens of “personal growth.”6 In studies of residents and attending physicians, it has been observed that specific events in medicine often trigger personal growth.7,8 These events have been variably described as “sentinel,” “evocative,” and “formative.”9–11 Indisputably, the state of emotional development exhibited by a medical trainee will profoundly impact the extent to which specific experiences have a formative impact and will determine the capacity of the learner to perceive the experiences as formative and to appreciate their impact. Thus, an assessment of formative events becomes an essential step toward improved understanding of the emotional challenges and developmental accomplishments of medical students. Previous studies have indicated that a wide range of experiences, both negative12,13 and positive,6 influence personal and professional growth during internship, residency, and professional practice. The experiences of medical students are less well characterized.
Certainly, the experience of medical school varies from student to student. For some, medical school is a turbulent period, whereas others absorb the demands of work, study, and clinical encounters with little difficulty.12 Nonetheless, for most students, the experiences of medical training present important challenges that may foster or erode positive aspects of emotional development.14,15 The capacity of medical school faculty, housestaff, community physician mentors, and peer–student advisors to attend to the needs of nascent physicians will be determined in part by their capacity to anticipate, recognize, and even influence responses to emotionally meaningful experiences in medical school. To this end, we studied graduating students at our medical school to characterize the extent to which a wide variety of emotionally evocative events shaped their experience of training.
Graduating students from two consecutive years (classes of 2007 and 2008) at the Johns Hopkins University School of Medicine were invited to participate in this cohort survey study.
Instrument development and data collection
The study questionnaire was developed and iteratively revised over a period of several months. In creating an initial list of events that may stimulate professional growth and emotional development of medical students, the medical literature and a focus group of Hopkins' Colleges Advisory Program faculty–advisors were consulted. A preliminary list of over 70 events was assembled. At subsequent meetings, the research team met with the focus group (which included Colleges Advisory Program faculty who were actively serving as longitudinal advisors to medical students) to narrow the list, eliminate redundancy, clarify the language, and minimize ambiguity. Ultimately, the research team reached consensus on a list of 34 items that represented distinctive experiences or described important relationships for medical students. Pilot testing among both resident physicians and nongraduating students was conducted to refine the instrument.
A scale was devised that captured two features simultaneously: (i) whether a student was exposed to the experience or relationship during medical school, and (ii) the magnitude of the impact that it had on the student. The six response options associated with each item were “no,” “yes, but it had no impact,” “yes, and it had little impact,” “yes, and it had some impact,” “yes, and it had a lot of impact,” and “yes, and it had tremendous impact.”
The 34-item list of events was incorporated into an online survey consisting of additional questions asking the graduating students about their experiences in medical school with respect to advising and career choice. The survey was delivered via e-mail in the three weeks prior to graduation.
This study was approved by the Johns Hopkins University School of Medicine institutional review board.
Characteristics of the respondents are presented as medians and proportions. The exposure rate was calculated as the percentage of students choosing impact ratings other than “not experienced.” With respect to evaluating the impact of the 34 events, two approaches were taken. In the first, the six response options were grouped in the following way: (a) not experienced = “no,” (b) no impact = “yes, but no impact,” (c) moderate impact = “a little impact” and “some impact,” and (d) high impact = “a lot of impact” and “tremendous impact.” The descriptive analyses show the data as the proportion of students rating the events as having moderate and high impact. In the second approach, a relative impact score was calculated by assigning numerical values to the verbal descriptors as follows: 0 = “yes, but no impact,” 1 = “a little impact,” 2 = “some impact,” 3 = “a lot of impact,” and 4 = “tremendous impact.” The numerical values associated with the students' responses were then summed for each event and divided by the total number of events experienced. The result was then scaled to 100 by multiplying the result by 25.
Factor analysis was conducted using the varimax rotation algorithm on the full dataset. An assessment of the eigenvalues of the reduced correlation matrix indicated that a maximum of four factors was appropriate for this dataset based on an eigenvalue >1. The factor analysis was performed by assigning numerical values to the six response options: from 0 = “no [did not occur]” and 1 = “yes, but no impact” to 5 = “yes, and it had tremendous impact.” Cronbach alpha was used to characterize the internal reliability of each of the factors that emerged from the factor analysis and to assess the extent to which each item contributed to the overall reliability of the factors. Pearson r was used to assess the strength of association between variables within emergent factors. The factor scores presented are the mean values of the items within each factor.
Characteristics of the study population
Of 216 graduating students, 181 (84%) completed the questionnaire. Demographic characteristics of the study population are shown in Table 1.
Exposure to formative events in medical school
In all, 6,084 event experiences were evaluated for their impact on students; of these, 4,620 (76%) were noted to have occurred, with 628 (14%) event experiences categorized as having a tremendous impact. All of the events were experienced by at least 25% of students. Exposure rates for the events ranged from 99% (“working well with a team”) to 27% (“falling asleep at the wheel”). The exposure rate data are presented in Table 2. Several events were, as expected, widely experienced, as they are compulsory experiences in the medical school, such as “getting to wear your white coat,” “your first clinical experience,” and “encountering a corpse in the anatomy lab.” Other highly frequent events might also be considered nearly universal, formative experiences of medical training, including “working well with a team,” “seeing someone undergo resuscitation or other aggressive intervention,” “encountering a truly exceptional role model in medicine,” “seeing a patient experience severe pain,” “encountering a negative role model,” and “making a mistake.” As a whole, the 34 events included in the list were relatively frequent events for our students.
Impact of formative events in medical school
The events rated by the students had varying degrees of impact. In the primary analysis, two events were distinguished by the number of respondents who assigned them high impact ratings: “encountering an exceptional role model in medicine” (78%) and “discovering an area of medicine that seems perfect for you” (65%). The tremendous impact of exceptional role models on graduating seniors is illustrated in Figure 1A.
A total of 26 events were identified by the majority of students as having either moderate or high impact. Eleven of these events were categorized by over 80% of students as having high-to-moderate impact. These events included important positive medical school experiences such as “being inspired by a special patient-care experience” and “working well with a team.” Other experiences were less positive, including “encountering a corpse in the anatomy lab,” “seeing someone undergo resuscitation or other aggressive intervention,” “seeing a patient experience severe pain,” “encountering a negative role model,” and “making a mistake.” A representative impact profile for an experience with high-to-moderate impact, “seeing a patient experience severe pain,” is shown in Figure 1B.
There were many events that students predominantly rated as having moderate impact. These events included experiences that seem to make up the ordinary parts of medical school training, such as “getting to wear your white coat” (impact profile shown in Figure 1C) and “encountering a corpse in the anatomy lab.” Other experiences rated as having moderate impact may actually have a latent impact—for instance, “feeling threatened by a patient” or “having a patient deceive you.”
Lastly, there were events that were reported by most students to have had “no impact,” meaning either that they did not occur or that they occurred but were not perceived as having an impact. This category includes several events that might be considered undesirable, such as “falling asleep at the wheel” (Figure 1D) and “drinking too much.”
Some events may occur with less frequency but nonetheless have a high impact. To address this, we designed a “relative impact score” to assess the impact of any given experience relative to the others included in the survey. Using this approach, we again found that “encountering an exceptional role model in medicine” and “discovering an area of medicine that seems perfect for you,” with relative impact scores of 82 and 76 (on a scale of 0–100), respectively, are experiences with exceptionally strong impact on students. This analysis identified experiences related to lost idealism and genuinely inappropriate feedback as less frequent experiences that had relatively high impact. In addition, we determined that whereas some undesirable events occur infrequently and are nonetheless rated by students as having low impact when they do occur, such as “one or more episodes of drinking too much” or “having a patient be sexually inappropriate,” other experiences, such as “falling asleep at the wheel,” are recognized by students as having relatively more impact. Implications of these ratings are discussed below. The results of this analysis are shown in Table 2.
Factor analysis of the data indicated that some events showed a strong coassociation. Three factors were found to consist of (i) multiple events with factor loadings over 0.5 and (ii) a high Cronbach alpha (without having any of the individual items decreasing the Cronbach alpha of the factor) (see Table 3). The Cronbach alpha for the factors ranged from 0.68 to 0.78, suggesting that the internal reliability of the factor analysis is acceptable.12 We observed excellent discrimination between events loading onto factors. The average factor loading discrimination for the events listed was 0.40, with a range of 0.25 to 0.59. The percent variance explained by the factors was 3.3, 2.5, and 2.0, respectively. We observed that the three factors demonstrated face validity in that experiences that would be expected to cluster in fact did so. The factors are labeled “inspiring experiences,” “mortality-related experiences,” and “negative experiences in the learning environment.” The relative importance of these experience clusters may be gauged by the associated factor scores reported as mean score ± standard deviation: inspiring events, 3.5 ± 1.0; mortality-related events, 3.0 ± 1.1; and negative experiences relating to the learning environment, 2.8 ± 1.3. The differences in relative impact were significant so that the relative impact was inspiring experiences > mortality-related experiences > negative experiences in the learning environment. Correlation of the factor scores among students was examined, and the scores for inspiring experiences and mortality-related experiences were significantly correlated (0.52); other correlations did not reach significance.
The objective of this study was a broadly inclusive, quantitative assessment of formative experiences and an appraisal of the impact of these experiences on medical school graduates. As such, this study provides unique insights into the scope and scale of formative experiences during medical training as perceived by emerging physicians. The events with the strongest impact on students included fundamentally positive experiences—in particular, finding an exceptional role model and identifying an ideal area of medicine. These observations find validation in prior studies that have examined the importance of role models for professional development16 but go beyond previous observations to highlight the power of role models in comparison with other events. Until now, little has been known about the wider spectrum of formative experiences of medical school and their relative impact. Our findings provide critical support for medical training experiences that maximize encounters with potential role models, exposures to potentially appealing areas of medicine (career matching), and opportunities for formative advising in response to emotionally evocative experiences. Further, the variability in responses between students is a clear signal that medical advising must always be learner-centered. For example, whereas several events with strong perceived impact may have been anticipated, others, such as “seeing a patient die,” are potent in the aggregate but were especially so for some students. The observation that many students experienced lasting impact from “seeing a patient experience severe pain” indicates that additional preparation in rendering compassionate care to those suffering despite medical treatment is a critical need.
There are other approaches to ascertaining information about formative events. These have included narrative analysis of journal entries, semistructured interview approaches, focus groups of students, surveys of student affairs professionals, and surveys of student assistance program staff.17–19 The intensive nature of qualitative data analysis can limit the scale of a study in terms of the number of participants that a study can reasonably incorporate. By contrast, survey instruments are intrinsically better-suited to large-scale studies; in our case, over 200 students were invited to participate. To bridge the gap between these two approaches, we engaged in a systematic process of survey development employing qualitative methods in the first phase of the study as the medical literature was reviewed for candidate formative events and advisors were encouraged to propose, critique, and offer refinements to a broadly inclusive list of candidate formative experiences. The resulting survey instrument was intended to provide new perspectives of the relative impacts resulting from exposures to a wide variety of potentially formative experiences. In the second phase of the study, we asked students to semiquantitatively assess the impact of these events, adopting an approach that is in many respects complementary to more qualitative methodologies. Through this hybridization of qualitative and quantitative methods, we formulated a study of expanded scope and scale that we anticipate, with refinement, may be appropriate for multiinstitutional research questions.
Given the substantial challenges that medical students face, there is an unmet need for future studies of the emotional development of medical students. It is well recognized that there is a steep decline in empathy during medical training.20 This has been attributed to a variety of factors, but it is not clear whether a decline in empathy is a necessary part of emotional maturation or may be an undesirable consequence of insufficient attention to emotional development during training. Interestingly, our students identified the “loss of idealism” as an event with high relative impact, suggesting that most students are keenly self-aware and not entirely happy with the changes in their emotion responses or attitudes. Our study further contributes to the field by identifying a variety of medical school experiences that have the potential to impact and shape the emotional development of emerging physicians.
Currently, a large majority of medical students enter into medical training in their early 20s, a life stage characterized by specific social and emotional processes.21 Arnett22 has described the life stage from age 18 to 25 as a period of “emerging adulthood”; for this reason, we describe our study population as “emerging physicians.” “Emerging adulthood” is observed in cultures where marriage and parenthood are postponed; it is characterized by fluid social interaction, high geographic mobility, and risk-taking behaviors. Emerging adults develop work roles, personal values, and social identity, eventually acquiring a consolidated view of the self as truly “adult.” Little is known about whether young physicians follow standard patterns of maturation; nonetheless, several of the moderate-impact events pertained to personal adjustment to medical school, such as “realizing that you are not as idealistic as you were before,” “missing an important event due to medical school, or “disagreeing with a significant other over work or school.”
The disadvantageous events incorporated into the survey were rated by most students to have had limited impact. One of the events in the query was “falling asleep at the wheel.” A quarter of the students who acknowledged falling asleep at the wheel judged this to be a “no impact” event. Despite this perception, falling asleep at the wheel is potentially life-threatening, and the gravity of this seems not to have been recognized or appreciated by some students. Similarly, students reported little impact associated with “feeling threatened by a patient,” an experience that occurred with surprisingly high frequency (49%). One interpretation of these findings may be that medical students need guidance with respect to processing and managing threats to their safety.23
Several limitations of the study should be considered. First, the study was conducted on graduating students at a single medical school, and thus the results may have limited generalizability. For example, if exceptional role models were more or less prevalent at another medical school, their cumulative perceived impact on students might be different. Second, the data collection relied exclusively on self-report. Perceived impact is an intrapersonal experience, however, making this approach acceptable and feasible. The primary concern then becomes the use of a reliable instrument. Our instrument was created through a formal process of literature review and focus group meetings and was subsequently refined by an expert panel. Third, the “impact scale” used to assess the relative influence of events was newly developed for this study. The scale was designed by individuals with expertise in this area, and the medical literature was consulted (content validity evidence). It captures both frequency of event occurrence as well a relative impact for those events that did occur. The scale performed well in distinguishing the impact of various events from each other, and factor analysis demonstrated related variables grouping together (internal construct validity evidence). Finally, the timing of the survey may be viewed as a limitation. Although it would be interesting to gauge the impact of events contemporaneously with their occurrence, this study did not adopt a longitudinal strategy. On one hand, it is possible to hypothesize that the impact of events early in medical school attenuates with the passage of time. On the other hand, it could be argued that event impact is amplified by additional experience and reflection. Further study of this question is required.
In conclusion, this study is among the first to characterize the relative impact of many common events experienced by medical students. These findings highlight a need to offer individualized advisory support of medical students, provide guidance to mentors seeking to more effectively shape the emotional development of emerging physicians, and indicate opportunities for personal and professional growth across the spectrum of medical education.
The authors thank Drs. David Levine and Thomas Koenig for helpful discussions of this work.
This work was supported in part by grant number R24AT004641 from the National Center for Complementary & Alternative Medicine (B.K. and J.A.H.) and by a grant from the Mayday Fund. Dr. Murinson is the recipient of a NINDS Mentored Career Development Award Grant, number NS048146.
Dr. Wright, Arnold P. Gold Foundation Professor of Medicine, receives support as a Miller-Coulson Family Scholar through the Johns Hopkins Center for Innovative Medicine. Dr. Levine is the Society of General Internal Medicine Horn Scholar.
This study was approved by the Johns Hopkins University School of Medicine institutional review board.
1 Epstein RM, Hundert EM. Defining and assessing professional competence. JAMA. 2002;287:226–235.
2 Morse DS, Edwardsen EA, Gordon HS. Missed opportunities for interval empathy in lung cancer communication. Arch Intern Med. 2008;168:1853–1858.
3 Roter DL, Frankel RM, Hall JA, Sluyter D. The expression of emotion through nonverbal behavior in medical visits. Mechanisms and outcomes. J Gen Intern Med. 2006;21(suppl 1):S28–S34.
4 Satterfield JM, Hughes E. Emotion skills training for medical students: A systematic review. Med Educ. 2007;41:935–941.
5 Murinson BB, Agarwal AK, Haythornthwaite JA. Cognitive expertise, emotional development, and reflective capacity: Clinical skills for improved pain care. J Pain. 2008;9:975–983.
6 Wright SM, Levine RB, Beasley B, et al. Personal growth and its correlates during residency training. Med Educ. 2006;40:737–745.
7 Levine RB, Haidet P, Kern DE, et al. Personal growth during internship: A qualitative analysis of interns' responses to key questions. J Gen Intern Med. 2006;21:564–569.
8 Kern DE, Wright SM, Carrese JA, et al. Personal growth in medical faculty: A qualitative study. West J Med. 2001;175:92–98.
9 Bowers L, Simpson A, Eyres S, et al. Serious untoward incidents and their aftermath in acute inpatient psychiatry: The Tompkins Acute Ward study. Int J Ment Health Nurs. 2006;15:226–234.
10 Lane RD, Quinlan DM, Schwartz GE, Walker PA, Zeitlin SB. The Levels of Emotional Awareness Scale: A cognitive-developmental measure of emotion. J Pers Assess. 1990;55:124–134.
11 Orlander JD, Fincke BG, Hermanns D, Johnson GA. Medical residents' first clearly remembered experiences of giving bad news. J Gen Intern Med. 2002;17:825–831.
12 Gaensbauer TJ, Mizner GL. Developmental stresses in medical education. Psychiatry. 1980;43:60–70.
13 Dyrbye LN, Thomas MR, Harper W, et al. The learning environment and medical student burnout: A multicentre study. Med Educ. 2009;43:274–282.
14 Haglund ME, aan het Rot M, Cooper NS, et al. Resilience in the third year of medical school: A prospective study of the associations between stressful events occurring during clinical rotations and student well-being. Acad Med. 2009;84:258–268.
15 Thomas MR, Dyrbye LN, Huntington JL, et al. How do distress and well-being relate to medical student empathy? A multicenter study. J Gen Intern Med. 2007;22:177–183.
16 Wright S, Wong A, Newill C. The impact of role models on medical students. J Gen Intern Med. 1997;12:53–56.
17 Levine RB, Kern DE, Wright SM. The impact of prompted narrative writing during internship on reflective practice: A qualitative study. Adv Health Sci Educ Theory Pract. 2008;13:723–733.
18 McNamee LS, O'Brien FY, Botha JH. Student perceptions of medico-legal autopsy demonstrations in a student-centred curriculum. Med Educ. 2009;43:66–73.
19 Wear D, Aultman JM, Varley JD, Zarconi J. Making fun of patients: Medical students' perceptions and use of derogatory and cynical humor in clinical settings. Acad Med. 2006;81:454–462.
20 Hojat M, Mangione S, Nasca TJ, et al. An empirical study of decline in empathy in medical school. Med Educ. 2004;38:934–941.
21 Erikson EH. Reflections on the last stage—and the first. Psychoanal Study Child. 1984;39:155–165.
22 Arnett JJ. Emerging adulthood. A theory of development from the late teens through the twenties. Am Psychol. 2000;55:469–480.
23 Sobal J. Health protective behaviors in first year medical students. Soc Sci Med. 1986;22:593–598.