Assessing Burnout and Associated Risk Factors

in Medical Students
Misha Armstrong, M.D., Kimberly Reynolds, M.D.

academic demands and results in cynicism and feelings of

Acknowledgements: The authors acknowledge the support of Dr. Jeffrey Brosco, incompetence.5,6 The learning environment, in addition to
MD, PhD and Dr. Stephen Symes, MD.

Abstract: Purpose: Our study aims to build on existing literature by assessing

specific character traits such as being self-driven, perfec-
factors that may be associated with an increased risk of burnout amongst tionists and competitive, has led to higher rates of psy-
medical students, particularly students of color.
chological stress in medical students compared to their
Methods: Our cross-sectional survey included the Copenhagen Burnout
Inventory (CBI) and additional de novo questions. Surveys were administered peers in other professional programs.5,7,8
electronically in June 2017 using a convenience sampling method.
Santen et al. (2010) found that low support and high
Results: A total of 162 survey results were recorded. Of those, 159 completed
demographic information with 43% of respondents being non-White, 64%
stress were significantly associated with burnout.2
women, 50% reported not having a mentor in medicine, 30% having an Stressors identified amongst medical students include
immediate family member in medicine, and 71% being concerned about the
financial burden associated with medical school. Black students were more delayed income, lack of leisure time, and educational
likely to be the first in their family to attend college, not have a physician family
member, and have financial concerns. The average CBI burnout scores
content volume.8 These are similar to those cited by
(n ¼ 138) indicated that overall students are not experiencing burnout. medical students in Pakistan, with high parental expecta-
However, nearly 50% of students experience personal, 42% work, and 12% client
related burnout based on their individual scores. Women were significantly more tions, high frequency of tests, and lack of acknowledge-
likely to experience work related burnout (p ¼ 0. 028) and had significantly ment also reported as significant stressors.1 The impact of
higher personal burnout scores (p ¼ 0.017). Additionally, Black students have
significantly higher personal burnout scores (p ¼ 0.013) compared to all other this stress is depression symptoms and anxiety in 14.3% of
reported races.
medical students compared to 6.7% in the general popu-
Conclusion: Although factors assessed during this study showed no significant
effect, the data trends suggest that both women and Black students
lation.3 Furthermore, burnout has been associated with
experienced higher rates of burnout. Further discussion regarding solutions to unprofessional behavior and a decreased willingness to
burnout is required in order to intervene early on in training for those at highest
risk. help others, resulting in a detrimental impact on personal
Keywords: Burnout-Medical student burnout-Undergraduate medical
and professional life.1,9,10 With the potential impact on
education-Minority students-Women in medicine patient outcomes and decrease in efficiency, stress could
be considered an occupational hazard.6,11,12
Author affiliations: Misha Armstrong, University of Miami Miller School of Medicine, Miami,
The unique experiences of minorities in medicine
FL, USA; Kimberly Reynolds, Department of Pediatrics, University of Miami Miller, School of makes understanding the implications of burnout and so-
Medicine, Miami, FL, USA
cial stressors an important part of addressing attrition rates
Correspondence: Department of Pediatrics, University of Miami Miller, School of Medicine,
USA., email: [email protected]
and the success of diverse students. AAMC data from
ª 2020 by the National Medical Association. Published by Elsevier Inc. All rights reserved.
2010 illustrated that Black and Hispanic students
https://doi.org/10.1016/j.jnma.2020.05.019 accounted for 15% of all matriculants. American Indians
and Native Hawaiians represented less than 1% of
matriculants.13 As such, underrepresented minorities
INTRODUCTION (URMs), defined as African Americans, Hispanics, and
Native Americans, are more likely to report difficulty

U
ndergraduate medical education exposes students
to demanding coursework that can be emotionally developing support networks and finding same race men-
difficult.1 Burnout commonly occurs when an tors as well as feeling their race affects their educational
individual experiences chronic stress without the necessary experience.14 With social support being a necessity to
social support and resources to handle the resulting offset stress and prevent emotional exhaustion, previous
depersonalization and emotional exhaustion.2,3 Further, findings indicate that URMs may have a higher risk of
individuals may experience feelings of inadequacy and burnout.
loss of motivation in response to high expectations.1 Ul- Previous research has examined the levels of burnout
timately, burnout is a reactive response to a negative work experienced by residents, physicians, and other healthcare
environment.4 In students, burnout is largely due to providers suggesting that burnout is a problem across the

JOURNAL OF THE NATIONAL MEDICAL ASSOCIATION VOL -, NO -, - 2020 1

BURNOUT IN MEDICAL STUDENTS

field of medicine. Additional studies, mostly from single and fourth year medical students. Reminder emails were
institutions, have evaluated medical students for burnout sent ten days later, with surveys open for a total of four
with the purpose of determining when burnout begins.2 weeks. These students were asked to reflect on their pre-
There has been limited research on the experiences of vious academic year when completing the CBI questions.
minority students in medical school. Our study aims to Prior to data collection, approval was obtained from the
build on existing literature and assess medical student Institutional Review Board of University of Miami.
burnout in minority and non-minority students. In addi- Study data was collected and managed using REDCap
tion, the objective of this study was to move beyond hosted at the University of Miami. REDCap (Research
quantifying levels of burnout amongst medical students Electronic Data Capture) is a secure, web-based applica-
and determine social factors that may be associated with an tion. Analysis was completed using IBM SPSS Statistics
increased risk of burnout as a means to identify at risk (24.0) involving descriptive statistics to obtain the preva-
students and develop interventions to decrease burnout. lence of burnout and stratify the data by demographic
information. Additional analysis included chi square,
T-tests, and logistic regression analysis to assess for
MATERIALS AND METHODS potential associations between burnout and social factors.
The study utilized a cross sectional survey method to
administer surveys to medical students electronically dur-
ing the beginning of the 2017 academic year. The survey RESULTS
consisted of the Copenhagen Burnout Inventory (CBI) and A total of 162 survey results were recorded; of those, 160
additional questions developed by the authors to identify completed demographic information with one student not
demographic information and potential factors associated responding to specific demographic questions, but
with an increased risk of burnout. The survey included a answering all other survey questions. The majority of
total of 35 questions, four of which were free response. participants were from the University of Miami with other
The rest of the questions were multiple choice using a institutions including University of Central Florida and
Likert Scale or simple binary responses. Florida International University. The results include 69 s
CBI is a validated survey tool that utilizes three sub- year students, 43 third year students, 46 fourth year stu-
dimensions: personal, work-related, and client-related dents, and 1 MD/PhD student in their seventh year. The
burnout.15 The authors instructed second year students to average age was 25; 64% of the respondents were women
replace “client” with “classmates” and third- and fourth- and 43% were non-White, with 66% of those being URMs
year students to replace “client” with “patients”, based based on the AAMC definition. Additionally, 50% re-
on the exposure that is most common during the preclin- ported not having a mentor in medicine, 30% having an
ical and clinical curriculum. CBI emphasizes fatigue and immediate family member in medicine, and 71% being
exhaustion as the core issues leading to burnout. This is in concerned about the financial burden associated with
concordance with a definition of burnout as a form of medical school. Black students were more likely to be the
exhaustion due to long term exposure to emotionally first in their family to attend college, not have a physician
demanding work.6,15 Personal burnout is related to phys- family member, and have financial concerns. Demographic
ical and psychological fatigue; work and client related data and survey responses are summarized in Table 1.
burnout is the fatigue perceived to be associated specif- The average CBI burnout scores (n ¼ 138) indicated
ically with work or clients, respectively.16 As such CBI that overall students are not experiencing burnout. The
seeks to quantify how much of the burnout experienced by average personal burnout score was 45.1, work related 44,
an individual is perceived or attributed to specific domains and client related 26. However, nearly 50% of students met
in a person’s life.15 Each domain uses a Likert scale the criteria for personal, 42% work, and 12% client related
correlating to a score of 100,75, 50,25, or 0. The total burnout based on their survey results. The average per-
score is based on the average of the score for each item. A sonal burnout score for minority students was 47.9
total score of 50 or higher is indicative of burnout.1 compared to 43.1 in non-minority students. The average
All medical students at any allopathic or osteopathic work-related burnout score for minority students was 46.2
medical school in the United States were eligible. and 42.4 for non-minority students. There was no statis-
Recruitment for participation relied on convenience sam- tically significant difference between the burnout scores
pling. The surveys were created in REDCap and made for minority and non-minority students. However, 67%
available via public links sent out via email and social and 57% of Black students met the criteria for personal and
media pages for medical students. Surveys were available work-related burnout, respectively. Additionally, Black
for completion starting June 19, 2017 for second, third, students, specifically, had significantly higher personal

2 VOL. -, NO -, - 2020 JOURNAL OF THE NATIONAL MEDICAL ASSOCIATION

 BURNOUT IN MEDICAL STUDENTS

financial concerns, physician family members, or first-

Table 1. Characteristics of survey respondents and key responses
summarized. The majority of participants in the study identified as generation college status on burnout.
female, attended a Florida program, and were in their second year of A total of 132 students completed a third section
medical school. Financial concerns were prominent. Only 138 students
completed the burnout inventory completely. Nearly half of all
regarding the resources necessary to improve their
participants experienced personal burnout as medical students. academic performance and personal well-being. 26% of
students reported feeling there is not enough resources or
Demographics (n [ 159) N (%) support from administration. Students indicated that lack
Gender Identity of time, barriers to accessing mental health care, and dif-
ficulties finding mentorship contributed to their levels of
Female 101 63.5
burnout. Importantly, 83% of students reported being
Male 58 36.5
happy with their medical school choice.
Race
Asian 32 20.1
Black/African American 26 16.4 DISCUSSION
Multi race 7 4.4 In a cross-sectional survey of medical students, our results
Other 3 1.9 showed that the majority of medical students are con-
cerned about their financial burden (71%) and reported not
White 91 57.2
having a mentor in medicine (50%). Additionally, these
Hispanic/Latino 19 11.9
barriers appear to be larger for Black medical students.
Underepresented in Medicine 45 28.3 Black students are less likely to come from a wealthy
Year in Medical School background, experience increased debt burden, and
First 0 0.0 perceive more stress due to racial discrimination.17,18 In
Second 69 43.4 addition, as URM students enter medical training at rates
Third 43 27.0 less than expected based on US representation, the lack of
a critical mass of minority students leads to social
Fourth or higher 47 29.6
isolation.18
Location of Medical School (n ¼ 158)
Amongst our population of medical students, nearly 50%
Florida 144 91.1 of students met the criteria for personal, 42% work, and 12%
Other States 14 8.9 client-related burnout. Black students had higher CBI scores
Social factors (n[160) N (%) and women were more likely to experience burnout than
First generation 13 8.1
men. Overall, there was no significant difference between
college student the rates of burnout or burnout scores between minority and
Immediate family 48 30.0
non-minority students. This is in concordance with a survey
member is a physician of students in Minnesota where rates of burnout were similar
Has a mentor in medicine 80 50.0 between minority and non-minority students at 43% and
45% respectively.19 However, minority students had a lower
Concerned about 113 70.6
financial burden sense of accomplishment, mental quality of life scores, and
more perceived difficulties regarding support from medical
Copenhagen Burnout administration.19
Inventory (n[138) N (%)
Student burnout begins during the first year of medical
Personal Burnout 65 47.1 school, necessitating the development of programs to
Work Related Burnout 55 39.9 address mental health early on.2 Causes of burnout are
Client Related Burnout 16 11.6 multifaceted with increased stress, debt, and academic
pressure being factors to consider.20 Additional stressors
have been found to have significant associations with
burnout scores (p ¼ 0.01) compared to all other reported burnout including grades, uncertainty, lack of time off,
races. Although not statistically significant, Black students gender, and physician parents.1,2 Our study did not find a
also had higher work-related burnout scores (p ¼ 0.06). significant effect of mentorship, financial concerns,
Women were also significantly more likely to experience physician family members, or first-generation college
work related burnout (p ¼ 0. 03) and had significantly status on burnout. However, students reported that time
higher personal burnout scores (p ¼ 0.012). Logistic constraints, lack of career advising, and barriers to mental
regression models showed no effects of mentorship, health counseling contribute to their experience of

JOURNAL OF THE NATIONAL MEDICAL ASSOCIATION VOL -, NO -, - 2020 3

BURNOUT IN MEDICAL STUDENTS

burnout. Prevention of burnout requires recognition of minority students, particularly Black students and women
symptoms with the inclusion of burnout awareness in may experience burnout at higher rates than their coun-
coursework being a potential solution.3 Strategies for terparts. Further discussion regarding solutions to burnout
addressing these identified stressors include pass-fail is required in order to intervene early on in medical
grading systems and interventions to improve stress training for those at highest risk. Additionally, as the di-
management, enhance resiliency, and increase feelings of versity of medical schools increases it is important for
personal achievement.1,2 institutions to evaluate the resources available to support
Limitations of this study include the cross-sectional minority students.
survey design with a self-administered, online survey
design based on convenience sampling. Convenience sam- FUNDING/SUPPORT
pling was chosen in order to increase the racial diversity in This research did not receive any specific grant from
the study; however as a result our study population may not funding agencies in the public, commercial, or not-for-
be representative of the medical student population. The profit sectors.
participants in this study represent a small sample of the
entire population of medical students. Associations may
OTHER DISCLOSURES
change with a larger sample size. The range in CBI scores
suggests that a larger sample size may have identified more The authors have no competing interests or disclosures.
significant findings. Also, response rates tend to be low
amongst medical students and minority students tend to be ETHICAL APPROVAL
less likely to respond to surveys.19 This impacts the ability to This study was approved by the University of Miami
obtain a large, diverse enough sample size to create gener- Miller School of Medicine Institutional Review Board.
alizability. Cultural and regional differences may exist in
regards to contributing factors to burnout. As such, pooling DISCLAIMERS
of all minority groups into a single category for analysis Not applicable.
ignores cultural differences regarding dealing with burnout,
acceptance of depression, and coping mechanisms. Selec-
PREVIOUS PRESENTATIONS
tion bias could impact the reported prevalence of burnout,
with students experiencing a large number of stressors being Results of the study were presented at the American
more likely to respond. Additionally, social desirability bias Medical Women’s Association 103rd meeting; March 24,
could lead to students altering their response to survey items. 2018; Philadelphia, PA. It was also presented at the As-
Of note, the majority of the study included validated survey sociation of American Medical Colleges Continuum
tools allowing for adequate comparison of data. Connections Joint Meeting; April 28, 2018; Orlando, FL.
To our knowledge, this is the first study to use CBI to
compare rates of burnout amongst minority and non- CONFLICT OF INTEREST STATEMENT
minority medical students with a large proportion of mi- The authors have no competing interests or disclosures.
norities included in the study. Our results are similar to the
high rates of burnout amongst medical students seen in the
literature. Further, we identified that women and Black REFERENCES
students in particular experience more personal burnout 1. Muzafar, Y., Khan, H., Ashraf, H., et al. (2015). Burnout and its
than their peers. More research is necessary to evaluate associated factors in medical students of Lahore, Pakistan.
Cureus(11), 7. https://doi.org/10.7759/cureus.390.
potential associations between social factors and burnout
risk amongst medical students. Additionally, nationwide 2. Santen, S., Holt, D., Kemp, J., & Hemphill, R. (2010). Burnout in
surveys and focus groups are necessary to identify existing medical students: examining the prevalence and associated
stressors in the learning environment as well as assess the factors. South Med J, 103(8), 758e763. https://doi.org/10.1097/
impact of current medical school curricula at modifying smj.0b013e3181e6d6d4.

existing stressors. 3. Chigerwe, M., Boudreaux, K., & Ilkiw, J. (2014). Assessment of
burnout in veterinary medical students using the Maslach
Burnout Inventory-Educational Survey: a survey during two se-
CONCLUSION mesters. BMC Med Educ(1), 14. https://doi.org/10.1186/s12909-
Burnout is a significant problem amongst medical students 014-0255-4.

and has widespread implications for future practice and 4. Dahlin, M., & Runeson, B. (2007). Burnout and psychiatric morbidity
patient outcomes. Specifically, our research shows that among medical students entering clinical training: a three year

4 VOL. -, NO -, - 2020 JOURNAL OF THE NATIONAL MEDICAL ASSOCIATION

 BURNOUT IN MEDICAL STUDENTS

prospective questionnaire and interview-based study. BMC Med newly graduated dentists. J Int Soc Prev Community Dent, 6(6),
Educ(1), 7. https://doi.org/10.1186/1472-6920-7-6. 535. https://doi.org/10.4103/2231-0762.195509.

5. Barbosa, J., Silva, Á., Ferreira, M., & Severo, M. (2016). Transition 13. Castillo-Page, L. (January 4, 2017). Diversity in medical edu-
from secondary school to medical school: the role of self-study cation facts and figures 2012. AAMC. https://www.aamc.org/
and self-regulated learning skills in freshman burnout. Acta initiatives/diversity/179816/facts_and_figures.html.
Med Port, 29(12), 803. https://doi.org/10.20344/amp.8350.
14. Bright, C., Duefield, C., & Stone, V. (1998). Perceived barriers
6. Maslach, C., & Leiter, M. (2016). Understanding the burnout and biases in the medical education experience by gender
experience: recent research and its implications for psychiatry. and race. J Natl Med Assoc, 90(11), 681e688.
World Psychiatr, 15(2), 103e111. https://doi.org/10.1002/wps.
15. Kristensen, T., Borritz, M., Villadsen, E., & Christensen, K. (2005).
20311.
The Copenhagen Burnout Inventory: a new tool for the
7. Villwock, J., Sobin, L., Koester, L., & Harris, T. (2016). Impostor assessment of burnout. Work Stress, 19(3), 192e207. https://doi.
syndrome and burnout among American medical students: a org/10.1080/02678370500297720.
pilot study. Int J Med Educ, 7, 364e369. https://doi.org/10.5116/
16. Molinero-Ruiz, E., Basart-Gómez Quintero, H., & Moncada-Lluis,
ijme.5801.eac4.
S. (2013). Validation of the Spanish version of the copenhagen
8. Oliva Costa, E., Santos, A., Abreu Santos, A., Melo, E., & burnout inventory questionnaire. Rev Esp Salud Pública, 87,
Andrade, T. (2012). Burnout Syndrome and associated factors 165e179.
among medical students: a cross-sectional study. Clinics, 67(6),
17. Liebschutz, J. M., Darko, G. O., Finley, E. P., Cawse, J., Bharel, M. M.,
573e579. https://doi.org/10.6061/clinics/2012(06)05.
& Orlander, J. D. (2006). In the minority: black physicians in resi-
9. Houpy, J., Lee, W., Woodruff, J., & Pincavage, A. (2017). dency and their experiences. J Natl Med Assoc, 98(9), 1441e1448.
Medical student resilience and stressful clinical events during
18. Dugger, R. A., El-Sayed, A. M., Dogra, A., Messina, C.,
clinical training. Med Educ Online(1), 22. https://doi.org/10.
Bronson, R., & Galea, S. (2013). The color of debt: racial
1080/10872981.2017.1320187.
disparities in anticipated medical student debt in the
10. Rothenberger, D. (2017). Physician burnout and well-being. Dis United States. PloS One(9), 8. https://doi.org/10.1371/jour-
Colon Rectum, 60(6), 567e576. https://doi.org/10.1097/dcr. nal.pone.0074693.
0000000000000844.
19. Dyrbye, L., Thomas, M., Huschka, M., et al. (2006). A multicenter
11. Dyrbye, L., West, C., Satele, D., et al. (2014). Burnout among U.S. study of burnout, depression, and quality of life in minority and
Medical students, residents, and early career physicians rela- nonminority US medical students. Mayo Clin Proc, 81(11),
tive to the general U.S. Population. Acad Med, 89(3), 443e451. 1435e1442. https://doi.org/10.4065/81.11.1435.
https://doi.org/10.1097/acm.0000000000000134.
20. Cecil, J., Mchale, C., Hart, J., & Laidlaw, A. (2014). Behaviour
12. Dagli, N., Kulkarni, S., Duraiswamy, P., Desai, H., Vyas, H., & and burnout in medical students. Med Educ Online, 19(1),
Baroudi, K. (2016). Stress and professional burnout among 25209. https://doi.org/10.3402/meo.v19.25209.

JOURNAL OF THE NATIONAL MEDICAL ASSOCIATION VOL -, NO -, - 2020 5