Abstract
As the proportion of women in the physician workforce increases, burnout in this population warrants further investigation. Exercise is an often-proposed strategy to combat burnout. Evaluating physical activity across a cohort of women physicians can assess associations of health behaviors with burnout. Cross-sectional study of women attending physicians in the United States who are actively engaged in a social media group for runners. An electronic survey comprised of 60 questions covering demographics, health behaviors, and burnout was administered. A healthy lifestyle subgroup (HLS) was defined based on American Heart Association physical activity and nutrition recommendations. We determine the prevalence of burnout and investigate associations between health behavior factors and burnout.
Of the 369 included surveys, most respondents were at least six years out from medical training (85.9%) and White (74.5%). Forty-two percent experienced burnout symptoms. Time exercising was significantly associated with fruit/vegetable consumption (P=.00002). There was no significant difference in burnout between the HLS compared to others (P = .37).
This group of self-reported physically active women physicians was found to have a lower prevalence of burnout when compared to other women physicians. Exercise and nutrition may be protective against burnout in women physicians but deserve further investigation.
Keywords: burnout, physician burnout, job stress, health behavior, running, aerobic exercise, exercise, mental health, occupational health, women’s health
“...exercise and nutrition may be protective against burnout in women physicians...”
Introduction
An estimated 50% of physicians have burnout, the symptoms of which have been found at a higher incidence in women than men.1-3 To combat burnout in the health professions, wellness promotion strategies have included exercise.4-7 Recent studies have shown improved burnout symptoms in those performing regular physical activity; however, most of these participants were men.6,8,9 Nutrition has been implicated as a modifiable risk factor for mental illness, particularly depression. 10 With the number of women entering the physician workforce increasing, burnout and wellness practices of physically active women physicians warranted further investigation.
Our objective was to determine the prevalence of burnout in this cohort of physically active women physicians in the United States. We further aimed to investigate which demographic and health behavior factors were associated with burnout. We hypothesized that exercise and healthy dietary practices are associated with reduced symptoms of burnout. To our knowledge, this is the first study to survey a cohort of self-identified physically active women physicians and analyze how burnout symptoms are affected by exercise and nutrition practices.
Methods
This cross-sectional study surveyed women physicians actively engaged in a running-related social media group over a 3-week period in February and March 2020. An online posting containing the survey hyperlink was posted to the page, inviting all group members to anonymously participate. In an effort to reduce non-response bias, only actively engaged group members, as determined by the platform’s engagement insight metrics, were considered when assessing response rate.
The survey comprised 60 questions assessing demographics, finances, domestic responsibility, and self-reported health behavior as it relates to exercise and nutrition practices. Questions from the American Medical Association’s Mini-Z Burnout Survey (MZBS) were included to evaluate burnout and work satisfaction.
Surveys belonging to women attending physicians practicing in the United States were included in the analysis. All respondents provided written informed consent. Summary tables account for missing data that were otherwise excluded from analysis. Prevalence of burnout was calculated using MZBS responses. Chi-square testing was performed to compare burnout in this group to available data in similar populations. One-way analysis of variance was performed to evaluate the relationship between the mean MZBS score, demographic factors, and health behavior responses. Chi-square tests were performed to assess the relationship between hours exercised weekly and hours worked weekly, number of fruits/vegetables consumed, and percent of domestic responsibilities.
A healthy lifestyle subgroup (HLS) was created to include those meeting healthy lifestyle criteria, defined as exercising at least 2–3 hours per week and consuming an average of at least five servings of fruits and vegetables per day, with those who did not. These criteria are based on the American Heart Association recommendations for physical activity and nutrition.11,12 Wilcoxon rank-sum tests were used to compare MZBS scores between the subgroups. P-values are reported where applicable with a significance threshold of <.05.
Analysis was conducted in R. 13 The University of Texas Health Science Center at Houston Institutional Review Board classified this study as exempt.
Results
Three hundred and eighty-two (38.2%) of the 1000 actively engaged group members completed the survey. Three hundred and sixty-nine met inclusion criteria. Racial/ethnic identity included 74.5% White, 13.8% Asian, 6.8% Hispanic, 2.4% multiple races, and .8% Black. The majority were at least six years out from medical training (85.9%). Demographics and health behavior data are shown in Tables 1 and 2.
Table 1.
Demographic Factors.
| Category | n (%) |
|---|---|
| Race/ethnicity | |
| White | 275 (74.5) |
| Asian | 51 (13.8) |
| Hispanic | 25 (6.8) |
| Multiple races | 9 (2.4) |
| Other | 4 (1.1) |
| Black or African American | 3 (.8) |
| Native Hawaiian or Pacific Islander | 1 (.3) |
| American Indian or Alaska native | 1 (.3) |
| Years out of training | |
| 5 years or less | 48 (13.0) |
| 6-10 years | 137 (37.1) |
| 11-20 years | 151 (40.9) |
| More than 20 years | 29 (7.9) |
| Other | 4 (1.1) |
| Specialty | |
| Obstetrics and gynecology | 49 (13.3) |
| Family medicine | 43 (11.6) |
| Internal medicine subspecialty | 42 (11.4) |
| Pediatrics | 39 (10.6) |
| Pediatric subspecialty | 35 (9.5) |
| Internal medicine | 28 (7.6) |
| Emergency medicine | 20 (5.4) |
| Surgical subspecialty | 18 (4.9) |
| Psychiatry | 14 (3.8) |
| Other | 14 (3.8) |
| Anesthesiology | 13 (3.5) |
| Dermatology | 12 (3.3) |
| Physical medicine and rehabilitation | 10 (2.7) |
| Radiology | 8 (2.2) |
| Neurology | 8 (2.2) |
| Pathology | 6 (1.6) |
| Ophthalmology | 3 (.8) |
| Allergy and immunology | 3 (.8) |
| General surgery | 2 (.5) |
| Preventive medicine | 2 (.5) |
| Number of children | |
| 0 | 0 (0) |
| 1 | 39 (10.6) |
| 2 | 208 (56.4) |
| 3 | 86 (23.3) |
| 4 | 27 (7.3) |
| 5 or more | 8 (2.2) |
| No response | 1 (.3) |
Table 2.
Health Behaviors.
| Category | n (%) |
|---|---|
| Hours spent exercising per week | |
| None | 2 (.5) |
| <1 | 3 (.8) |
| 1–2 | 12 (3.3) |
| 2–3 | 26 (7.0) |
| 3–4 | 53 (14.4) |
| 4–5 | 70 (19.0) |
| 5–6 | 65 (17.6) |
| 6–7 | 61 (16.5) |
| >7 | 77 (20.9) |
| Step count | |
| 1000 < 2500 | 2 (.7) |
| 2500 < 5000 | 13 (4.3) |
| 5000 < 7500 | 36 (12.0) |
| 7500 < 10 000 | 69 (22.9) |
| 10 000 < 12 500 | 86 (28.6) |
| 12 500–15 000 | 56 (19.6) |
| 15 000 < 17 500 | 20 (6.6) |
| 17 500 < 20 000 | 9 (3.0) |
| >20 000 | 10 (3.3) |
| No response | 68 (18.4) |
| Percentage of time spent in each type of exercise | |
| Cardiovascular/aerobic | |
| 0–25% | 57 (15.4) |
| 26–50% | 33 (8.9) |
| 51–75% | 86 (23.3) |
| 76–100% | 193 (52.3) |
| Resistance training | |
| 0–25% | 318 (86.2) |
| 26–50% | 45 (12.2) |
| 51–75% | 5 (1.4) |
| 76–100% | 1 (.3) |
| Yoga/pilates | |
| 0–25% | 356 (96.5) |
| 26–50% | 11 (3.0) |
| 51–75% | 1 (.3) |
| 76–100% | 1 (.3) |
| Calisthenics/gymnastics | |
| 0–25% | 369 (100) |
| 26–50% | 0 (0) |
| 51–75% | 0 (0) |
| 76–100% | 0 (0) |
| Recreational sports | |
| 0–25% | 368 (99.7) |
| 26–50% | 1 (.3) |
| 51–75% | 0 (0) |
| 76–100% | 0 (0) |
| Percentage of time spent exercising alone | |
| 0–25% | 39 (10.6) |
| 26–50% | 42 (11.4) |
| 51–75% | 81 (22.0) |
| 76–100% | 207 (56.1) |
| Servings of fruits and vegetables per day | |
| None | 3 (.8) |
| 1 < 3 | 56 (15.2) |
| 3 < 5 | 180 (48.8) |
| 5 < 7 | 100 (27.1) |
| 7–9 | 24 (6.5) |
| >9 | 6 (1.6) |
One hundred and fifty-five respondents (42.1%) indicated they were experiencing burnout symptoms on the MZBS. Our group had significantly less burnout (P < .0001) compared to a study of women physicians, the majority of whom were mothers, that found burnout in 64.9% of those surveyed using a similar question and 5-point response scale to the MZBS (Larson, et al, 2020). One hundred and twenty-two respondents (33.1%) fit the HLS criteria, but no difference was found in burnout scores between this subgroup and others (P = .37). One hundred and thirty respondents (35.2%) reported consuming at least 5 servings of fruit and vegetables daily. Increased fruit and vegetable consumption was associated with increased time spent exercising per week (P = .00002). All but 1 respondent (99.7%) reported that exercise improved their overall sense of well-being. A median of 5–6 hours per week was spent on exercise with the majority (69% on average) of that time spent performing aerobic exercise. There was not a significant relationship between time exercising and hours worked (P = .95) or percent of domestic responsibilities performed (P = .30). While the MZBS trended toward higher scores (indicating lower burnout) with greater hours of exercise, there was not a significant association (P = .078; see Figure 1).
Figure 1.
Trend of increased MZBS score (lower burnout) with increased hours of exercise per week.
Discussion
In this group of women attending physicians who self-identify as runners, we found a reduced prevalence of burnout compared to a similar cohort of attending women physicians at a leadership conference. 14 Increased fruit and vegetable consumption was associated with increased time exercising per week. We observed a trend of lower burnout with increased exercise. There was no significant difference in burnout scores between those in the HLS and not.
We observed a trend of increased burnout in those reporting less exercise per week although this association did not meet statistical significance. Importantly, very few respondents (4.6%) were inactive, failing to meet the physical activity recommendations. The majority of the cohort reported at least 2–3 hours of exercise weekly with a median of 5–6 hours, mostly aerobic. Consequently, this study consists primarily of active women physicians. Though these healthy lifestyle factors were not assessed in the comparison study that consisted of women physicians, the majority of whom (76%) were also mothers, one would expect the physical activity level to be more heterogeneous since they were not sampled from a fitness-related group (Larson, et al., 2020). As such, it is possible that the significantly reduced burnout in our sample is attributable to healthy lifestyle practices, like exercise. The major limitation in considering this comparison is the different phrasing in the five response options for the question in each survey that asked participants to self-assess their degree of burnout. The assessment of exercise and burnout should be further explored with a consistent burnout measurement tool in a more fitness-diverse population, investigating both volume and type of physical activity, to better elucidate this relationship.
Our study adds to the limited existing literature on nutrition and physician well-being. 15 Dietary patterns rich in plant-based foods, such as the Mediterranean Diet, are associated with a lower likelihood of experiencing mental disorders such as depression and anxiety.10,16 Our findings support combined nutrition and exercise strategies reduce likelihood of physician burnout.
A statistically significant difference was not found between burnout scores of the HLS compared to the others. However, the sampling bias introduced by the group of women surveyed deserves consideration as does the definition of our subgroup. Since this was a predominantly active group, the HLS criteria were largely influenced by the inclusion of healthy dietary practices, the assessment of which has several limitations. Self-reported dietary information was obtained via an unvalidated brief screening question subject to recall bias and only able to provide a gross estimation of intake. 17 By diversifying the group of women physicians assessed to include a more varied array of physical activity practices and by obtaining a validated, comprehensive measure of health behavior, future research will be better equipped to analyze the relationship between HLS criteria and burnout.
This self-reported data are subject to biases such as recall, social desirability, and non-response bias. While we only considered those actively engaged in the social media group as possible participants, non-response bias is still possible. Those choosing to complete the survey may have had more time or energy to do so. Respondents may have had greater motivation to participate because they were experiencing fewer symptoms of burnout or, conversely, were experiencing burnout so found the survey of personal interest. Despite these limitations, this study adds to the limited existing literature on burnout and health behavior in women physicians.
Conclusion
By investigating burnout in this population, we can begin to better understand its associations, thereby guiding the development of targeted efforts to combat burnout. Health behaviors like exercise and nutrition may be protective against burnout in women physicians but deserve further investigation to elucidate this relationship.
Acknowledgments
This study did not receive any funding. An abstract of these findings was presented at the Association of Academic Physiatrists Annual Meeting, Physiatry ’21, held virtually in February 2021.
Footnotes
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding: The author(s) received no financial support for the research, authorship, and/or publication of this article.
ORCID iD
Hannah Uhlig-Reche https://orcid.org/0000-0002-8417-394X
References
- 1.del Carmen MG, Herman J, Rao S, et al. Trends and factors associated with physician burnout at a multispecialty academic faculty practice organization. JAMA Netw Open. 2019;2(3):e190554. doi: 10.1001/jamanetworkopen.2019.0554. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Dyrbye LN, West CP, Satele D, et al. Burnout among U.S. medical students, residents, and early career physicians relative to the general U.S. population. Acad Med. 2014;89(3):443-451. doi: 10.1097/ACM.0000000000000134. [DOI] [PubMed] [Google Scholar]
- 3.West CP, Shanafelt TD, Kolars JC. Quality of life, burnout, educational debt, and medical knowledge among internal medicine residents. J Am Med Assoc. 2001;306(9):952–960. [DOI] [PubMed] [Google Scholar]
- 4.Aggarwal R, Deutsch JK, Medina J, et al. Resident wellness: an intervention to decrease burnout and increase resiliency and happiness. MedEdPORTAL . 2017;13:10651. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Shanafelt TD, Novotny P, Johnson ME, et al. The well-being and personal wellness promotion strategies of medical oncologists in the North Central cancer treatment group. Oncology. 2005;68(1):23-32. [DOI] [PubMed] [Google Scholar]
- 6.Shanafelt TD, Oreskovich MR, Dyrby LN, et al. Avoiding burnout: the personal health habits and wellness practices of US surgeons. Ann Surg . 2012;255(4):625-633. [DOI] [PubMed] [Google Scholar]
- 7.Weight CJ, Sellon JL, Lessard-Anderson CR, et al. Physical activity, quality of life, and burnout among physician trainees: the effect of a team-based, incentivized exercise program. Mayo Clin Proc . 2013;88(12):1435-1442. [DOI] [PubMed] [Google Scholar]
- 8.Brand S, Ebner K, Mikoteit T, et al. Influence of regular physical activity on mitochondrial activity and symptoms of burnout-an interventional pilot study. J Clin Med. 2020;9(3):667. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Naczenski LM, Vries JD, van Hooff MLM, et al. Systematic review of the association between physical activity and burnout. J Occup Health . 2017;59(6):477-494. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Marx W, Moseley G, Berk M, Jacka F. Nutritional psychiatry: the present state of the evidence. Proc Nutr Soc . 2017;76(4):427-436. [DOI] [PubMed] [Google Scholar]
- 11.American Heart Association. American heart association recommendations for physical activity in adults and kids [Internet]. (2018) Available from: https://www.heart.org/en/healthy-living/fitness/fitness-basics/aha-recs-for-physical-activity-in-adults.
- 12.American Heart Association. Fruits and vegetables serving sizes infographic [Internet]. (2017). Available from: https://www.heart.org/en/healthy-living/healthy-eating/add-color/fruits-and-vegetables-serving-sizes.
- 13.R Core Team. R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing; 2017. Available from: https://www.R-project.org/. [Google Scholar]
- 14.Larson AR, Jagsi R, Moeschler SM, Silver JK. Association of compensation and educational debt with burnout and perceived impact of debt on women physicians’ career and lifestyle choices. Health Equity . 2020;4(1):565-570. [Google Scholar]
- 15.Hamidi MS, Boggild MK, Cheung AM. Running on empty: a review of nutrition and physicians’ well-being. Postgrad Med. 2016;92(1090):478-481. [DOI] [PubMed] [Google Scholar]
- 16.Owen L, Corfe B. The role of diet and nutrition on mental health and wellbeing. Proc Nutr Soc . 2017;76(4):425-426. [DOI] [PubMed] [Google Scholar]
- 17.Yaroch AL, Tooze J, Thompson FE, et al. Evaluation of three short dietary instruments to assess fruit and vegetable intake: the National cancer institute’s food attitudes and behaviors survey. J Acad Nutr Diet. 2012;112(10):1570-1577. [DOI] [PMC free article] [PubMed] [Google Scholar]