It is well established that many medical students experience burnout and other forms of psychological distress.1–7 A multisite longitudinal survey of 4,287 U.S. students at seven medical schools (52.4% response rate) found that 49.6% of medical students reported burnout and 11.2% reported suicidal ideations.2 Studies also suggest that distress can have serious professional and personal ramifications. For example, in a recent multi-institutional study burnout was associated with self-reported unprofessional conduct and less altruistic professional values among medical students at seven U.S. schools.1 In other studies burnout was inversely correlated with empathy.8,9 Burnout has a strong dose–response relationship with suicidal ideation and serious thoughts of dropping out of medical school that persists on multivariable analysis controlling for other factors including symptoms of depression.2,3 Factors contributing to the high prevalence of distress among medical students include stress from academic demands, experiences of major stressful personal life events, harassment or discrimination, low social support, and suboptimal learning environment.4,10–13 A recent national study reported that burnout was more prevalent among medical students than similarly aged U.S. population controls.14 In comparison with U.S. college graduates aged 22 to 32, medical students were also more likely to exhibit symptoms of depression and had higher levels of fatigue.
It is less clear whether students start medical school more distressed than age-similar individuals in the general population. Previous studies of matriculating medical students (MMSs) have been small and findings have not been uniform,12 with some investigators reporting a higher prevalence of depression at matriculation than age-matched controls15,16 and others finding the reverse.17
The competitive application process, demands for academic excellence, and extracurricular activities prior to medical school could contribute to distress in medical school applicants. Medical students may begin medical school already compromised, which could contribute to subsequent burnout during training years. On the other hand, if students begin medical school with similar or better mental health than age-similar controls, this further supports existing concerns about the negative impact of training on student well-being and its capacity to overwhelm even highly capable individuals. This finding would provide additional support for modifying medical school curricula to improve student well-being, funding student wellness programs, and implementing screening programs to facilitate identification of medical students in need of help.
To address the question of whether MMSs are more distressed than age-similar individuals in the general population, we conducted a multisite survey of MMSs and compared these data with those from a probability-based sample of the general U.S. population.
Medical student sample.
In the fall of 2012, we invited all MMSs attending six diverse U.S. medical schools (Mayo Medical School; University of Washington School of Medicine; University of Alabama School of Medicine; University of California, San Diego School of Medicine; Rutgers New Jersey Medical School; and the Uniformed Services University of the Health Sciences) to participate via e-mail with a link to a Web-based survey or by filling out the survey during orientation activities. Participation was elective, and responses were anonymized. The institutional review board at each institution approved the study prior to participation of their students.
Population control sample.
The details of the methods used to obtain the population control sample have been previously reported.18 In collaboration with Knowledge Panel (additional technical information available at http://www.knowledgenetworks.com/knpanel/index.html), we surveyed a probability-based sample of U.S. individuals aged 22 to 65 years, with modest oversampling of those younger than 34 years to allow a larger sample for comparison with medical students in 2011. The Mayo Clinic institutional review board reviewed and approved this component of the study.
Surveys for both students and population control included two items from the Maslach Burnout Inventory (MBI),19,20 two-item Primary Care Evaluation of Mental Disorders (PRIME-MD),21 and Linear Analogue Scales Assessment22 to measure burnout, symptoms of depression, and multiple domains (mental, emotional, physical, overall) of quality of life (QOL), respectively, as well as demographics (age, sex, race/ethnicity, and relationship status). Population control subjects also provided information on their current occupation and highest level of education completed.
Medical students completing the survey answered the 22-item MBI, which contains three subscales (emotional exhaustion, depersonalization, and sense of personal accomplishment).23
Given the length of the full MBI, the population control group was asked to answer two single-item measures adapted from the full MBI, as previously described.19,20 Those scoring high on the emotional exhaustion and depersonalization single items (defined as occurring at least weekly in accord with thresholds previously reported)19,20 were considered to have at least one manifestation of professional burnout. In multiple independent samples of over 10,000 physicians and medical students, these single items stratify the risk of burnout.19,20 In previous studies, the areas under the receiver operating characteristic curve for the emotional exhaustion and depersonalization single items in comparison with the full MBI domain scores were 0.94 and 0.93, respectively, and the positive predictive values of these single-item thresholds for high levels of emotional exhaustion and depersonalization were 88.2% and 89.6%, respectively.19,20 We used data from the two-item measures when comparisons were made between the burnout prevalence rates of medical student and control samples.
Symptoms of depression and QOL.
Both surveys included the two-item PRIME-MD, a screening tool which performs as well as longer instruments.24 The PRIME-MD has a sensitivity of 86% to 96% and a specificity of 57% to 75% for major depression disorder.21,24 With a reported positive likelihood ratio of up to 3.42 for the diagnosis of major depression,24 and an estimated 25% prevalence of depression among medical students,12 screening positive implies a 50% probability of current major depression.
All respondents were asked to rate their overall, mental, physical, and emotional QOL over the past week on a standardized linear analogue scale (0 = “As bad as it can be”; 10 = “As good as it can be”). The validity of this scale has been established in a variety of medical conditions and populations.22
From the population sample, we selected four-year college graduates and used a simple randomization procedure in SAS statistical software to equally weight the age strata (< 25, 25–30, 31–35, 36–40, > 40) in the two samples. To age-match the matriculants with the normal population data, we needed a population sample with a similar age stratification (72.6%, < 25; 22.2%, 25–30; 3.8%, 31–35; 1.2%, 36–40; and 0.2%, > 40) as the medical student sample. Our limiting factor was the first group (age < 25) because, in the normal data (population data), we only had 121 people age < 25 who are college graduates. Therefore, the maximum size that age group could be was 121. Because we needed for that group of 121 to account for 72.6% of the normal population (to age-match the matriculates), we were limited on the overall size of the normal group and the proportions or sizes of the other age groups. By combining the < 25 and the 25–30 age groups, we had a larger population of college graduates in the 25–30 group, and we could inflate the normal group to about 546 people while keeping the similar age stratification (94.8%, < 30; 3.8%, 31–35; 1.2%, 36–40; and 0.2%, > 40).
We used standard descriptive statistics and Kruskal-Wallis test or chi-square test for univariate comparisons to characterize and make comparisons between the MMSs and controls. All tests were two-sided with type I error rates of 0.05. Pooled multivariate logistic and mixed-effects linear regression analyses of MMSs and population controls were performed to identify demographic and professional characteristics associated with the dependent outcomes. Age, sex, relationship status, race/ethnicity, and parental status were included in the models as these covariates have previously been found associated with burnout or other forms of psychological distress among medical students or physicians in practice.3,10,12,18
We performed all analysis using SAS version 9 (SAS Institute, Cary, North Carolina).
Demographic characteristics of the 582/938 (62%) responding MMSs are shown in Table 1. The sex distribution of responders mirrored that of matriculants at the six U.S. medical schools.25 Fifty-four percent of respondents were male compared with 52.8% of MMSs nationally.25 MMS participants were more likely to be married and reported higher debt burden than students matriculating nationally.26
Relative to 546 age-similar college graduates (controls), MMSs were more likely to be male and single and less likely to be white non-Hispanic and have children (all P < .0001; Table 1). MMSs had lower rates of burnout (27.3% versus 37.3%, P < .001) and depression symptoms (26.2% versus 42.4%, P < .0001) and had higher QOL scores across all domains relative to age-similar college graduates (all P < .0001; Table 2). Results were similar when we repeated the analysis using the control group with matched proportions of age strata in all age categories (< 25; 25–30; 31–35; 36–40; > 40) in the two samples.
In the pooled multivariate analysis of all college graduates (e.g., both MMSs and college graduates from the population sample) adjusting for age, sex, relationship status, race/ethnicity, and parental status, MMSs had lower rates of burnout and depression symptoms and higher QOL scores (Table 3).
Implications of findings
In this multisite study, we found that MMSs had better mental health across a variety of specific domains relative to age-similar college graduates from the general population pursuing other careers. This finding persisted after controlling for a host of demographic factors, suggesting that MMSs enter medical school in relatively good mental health. These findings about medical students at the time they begin training are a distinct contrast to the findings from national studies indicating that burnout in enrolled medical students, residents, early career physicians,14 and physicians in general18 are higher than the general population. Previous studies also found depression to be more prevalent in enrolled medical students and residents relative to similarly aged college graduates.14 This apparent shift from students being at lower risk of distress at the time they begin medical school to higher risk during medical school and all subsequent stages of their career supports the contention that being a physician is a high-risk profession. Importantly, this finding suggests that the training process may contribute to the deterioration of mental health in developing physicians. Our findings support the concept that medical student distress appears to be a “nurture” rather than a “nature” problem, indicating that changes in the learning environment are needed.
Recently, the Association of American Medical Colleges’ Innovation Lab Working Group and Admissions Initiative reported the results of an extensive process and literature review to identify important personal competencies in medical school applicants and ways to measure them during the admissions process.27 One core competency considered to be important was resilience. Resilience has been defined as the ability to recuperate and remain positive after experiencing adversity.4 One tempting approach could be to develop admissions criteria that evaluate applicants’ resilience to screen out individuals potentially at high risk for burnout. We would caution against a strategy that overemphasizes selection of “resilient” individuals capable of navigating a dysfunctional training process. Instead, we should address the problems within the learning environment that precipitate burnout in highly talented and dedicated individuals with the traits desired in future physicians. Our findings that MMSs possess similar or better mental health than peers even after going through a demanding application and selection process suggest that the students currently entering medical school may, on average, have better-developed coping mechanisms than those in the general population at similar stages of personal development. Although we should unquestionably seek to help medical students further develop their coping skills once enrolled,18,28 changes to the process of training and the learning environment will also likely be necessary.
Modifications in curriculum and examination schedule, pass–fail grading systems, and duty hours limitations for medical students may all be useful approaches to improving the training experience for medical students and improve their vulnerability to distress. In addition, adequate financial reimbursement and fair academic advancement to support educational endeavors of faculty members so that they in turn can better mentor and support students would also likely help. Furthermore, modulating faculty and resident on-call duty, as well as enhancing faculty and resident development programs, could help reduce cynicism and student belittlement, resulting also in improvement in the environment in which our students are educated.4,29,30 One group has reported on a new preclinical curriculum designed to support medical student wellness that included changes in course content, contact hours, scheduling, and grading, as well as institution of small learning communities. The authors noted lower levels of depression and anxiety during preclinical years in students exposed to this new curriculum compared with students who experienced the curriculum prior to those changes.31 Longitudinal studies are needed to evaluate the impact of specific curriculum changes and modification to the learning environment that attempt to improve all stages of the medical student education experience. Although designing and implementing such studies have numerous logistic and financial challenges, our results suggest that changing the way we educate students and the environment in which they learn should be an integral part of interventions to reduce or prevent burnout and distress in medical students.
One limitation of our study is the cross-sectional design. A longitudinal study that showed a decline in mental health indicators of MMSs over time would be even more compelling. Another limitation is the inclusion of a limited array of personal variables. Other variables on which the students and population controls may differ might create more stress in the controls than in the MMSs. For example, looking for a job as a college graduate in the current challenging economy or lower socioeconomic status among the age-matched college graduates than the MMSs might be confounding factors. MMSs may be excited about starting medical school, which may improve their mental health indicators very early in medical school. Strengths include that the study was multisite and conducted with students from diverse medical schools. Other strengths include the use of validated instruments, an MMS sample similar to U.S. MMSs overall, and a probability-based sample of age-similar U.S. college graduates.
Medical students begin medical school with better mental health indicators than age-similar college graduates in the general population. These findings, coupled with other studies that demonstrate high rates of distress among medical students, support existing concerns that the learning environment and training process contribute to the deterioration of mental health in medical students. Although longitudinal studies are needed, our findings suggest that efforts to improve the health and resilience of our future physician workforce should seek to improve the medical school and physician training process.
1. Dyrbye LN, Massie FS Jr, Eacker A, et al. Relationship between burnout and professional conduct and attitudes among U.S. medical students. JAMA. 2010;304:1173–1180
2. Dyrbye LN, Thomas MR, Massie FS, et al. Burnout and suicidal ideation among US medical students. Ann Intern Med. 2008;149:334
3. Dyrbye LN, Thomas MR, Power DV, et al. Burnout and serious thoughts of dropping out of medical school: A multi-institutional study. Acad Med. 2010;85:94–102
4. Dyrbye LN, Power DV, Massie FS, et al. Factors associated with resiliency and recovery from burnout: A prospective, multi-institutional study of U.S. medical students. Med Educ. 2010;44:1016–1026
5. Schwenk TL, Davis L, Wimsatt LA. Depression, stigma, and suicidal ideation in medical students. JAMA. 2010;304:1181–1190
6. Goebert D, Thompson D, Takeshita J, et al. Depression symptoms in medical students and residents: A multischool study. Acad Med. 2009;84:236–241
7. Wallace JE, Lemaire JB, Ghali WA. Physician wellness: A missing quality indicator. Lancet. 2009;374:1714–1721
8. Brazeau C, Schroeder R, Rovi S, Boyd L. Relationships between burnout, empathy, and professionalism climate. Acad Med. 2010;85(10 suppl):S33–S36
9. 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
10. Dyrbye LN, Thomas MR, Eacker A, et al. Race, ethnicity, and medical student well-being in the United States. Arch Intern Med. 2007;167:2103–2109
11. Dyrbye LN, Thomas MR, Huntington JL, et al. Personal life events and medical student well-being: A multicenter study. Acad Med. 2006;81:374–384
12. Dyrbye LN, Thomas MR, Shanafelt TD. Systematic review of depression, anxiety, and other indicators of psychological distress among U.S. and Canadian medical students. Acad Med. 2006;81:354–373
13. Frank E, Carrera JS, Stratton T, Bickel J, Nora LM. Experiences of belittlement and harassment and their correlates among medical students in the United States: Longitudinal survey. BMJ. 2006;333:682
14. Dyrbye LN, West CP, Satele D, et al. Burnout among US medical students, residents, and early career physicians relative to the general US population. Acad Med. 2014;89:443–451
15. Sherry S, Notman MT, Nadelson CC, Kanter F, Salt P. Anxiety, depression, and menstrual symptoms among freshman medical students. J Clin Psychiatry. 1988;49:490–493
16. Buchman BP, Sallis JF, Criqui MH, Dimsdale JE, Kaplan RM. Physical activity, physical fitness, and psychological characteristics of medical students. J Psychosom Res. 1991;35:197–208
17. Lloyd C, Miller PM. The relationship of parental style to depression and self-esteem in adulthood. J Nerv Ment Dis. 1997;185:655–663
18. Shanafelt TD, Boone S, Tan L, et al. Burnout and satisfaction with work–life balance among U.S. physicians relative to the general U.S. population. Arch Intern Med. 2012;172:1377–1385
19. West CP, Dyrbye LN, Sloan JA, Shanafelt TD. Single item measures of emotional exhaustion and depersonalization are useful for assessing burnout in medical professionals. J Gen Intern Med. 2009;24:1318–1321
20. West CP, Dyrbye LN, Satele DV, Sloan JA, Shanafelt TD. Concurrent validity of single-item measures of emotional exhaustion and depersonalization in burnout assessment. J Gen Intern Med. 2012;27:1445–1452
21. Spitzer RL, Williams JB, Kroenke K, et al. Utility of a new procedure for diagnosing mental disorders in primary care. The PRIME-MD 1000 study. JAMA. 1994;272:1749–1756
22. Gudex C, Dolan P, Kind P, Williams A. Health state valuations from the general public using the visual analogue scale. Qual Life Res. 1996;5:521–531
23. Maslach C, Jackson SE, et al. Maslach Burnout Inventory Manual. 1996 Palo Alto, Calif Consulting Psychologists Press
24. Whooley MA, Avins AL, Miranda J, Browner WS. Case-finding instruments for depression. Two questions are as good as many. J Gen Intern Med. 1997;12:439–445
27. Koenig TW, Parrish SK, Terregino CA, Williams JP, Dunleavy DM, Volsch JM. Core personal competencies important to entering students’ success in medical school: What are they and how could they be assessed early in the admission process? Acad Med. 2013;88:603–613
28. Brazeau CM. The surest way to prevent burnout in medical school and beyond. Acad Med. 2010;85:569–570
29. Reed DA, Shanafelt TD, Satele DW, et al. Relationship of pass/fail grading and curriculum structure with well-being among preclinical medical students: A multi-institutional study. Acad Med. 2011;86:1367–1373
30. Dyrbye LN, Thomas MR, Harper W, et al. The learning environment and medical student burnout: A multicentre study. Med Educ. 2009;43:274–282
31. Slavin SJ, Schindler DL, Chibnall JT. Medical student mental health 3.0: Improving student wellness through curricular changes. Acad Med. 2014;89:573–577