Author + information
- Received February 20, 2019
- Revision received April 10, 2019
- Accepted April 16, 2019
- Published online June 24, 2019.
- Laxmi S. Mehta, MDa,∗ (, )@DrLaxmiMehta,
- Sandra J. Lewis, MDb,
- Claire S. Duvernoy, MDc,
- Anne K. Rzeszut, MAd,
- Mary Norine Walsh, MDe,
- Robert A. Harrington, MDf,
- Athena Poppas, MDg,
- Mark Linzer, MDh,
- Philip F. Binkley, MDa,
- Pamela S. Douglas, MDi,
- on behalf of the American College of Cardiology Women in Cardiology Leadership Council
- aDepartment of Medicine, Division of Cardiology, The Ohio State University, Columbus, OhioDepartment of Medicine, The Ohio State University, Columbus, Ohio
- bLegacy Health, Portland, OregonLegacy Health, Portland, Oregon
- cDepartment of Medicine, Cardiology Section, VA Ann Arbor Healthcare System, University of Michigan Health System, Ann Arbor, MichiganVA Ann Arbor Healthcare System, University of Michigan Health System, Ann Arbor, Michigan
- dMarket Strategy Division, American College of Cardiology, Washington, DCMarket Strategy Division, American College of Cardiology, Washington, DC
- eSt. Vincent Heart Center, Indianapolis, IndianaSt. Vincent Heart Center, Indianapolis, Indiana
- fDepartment of Medicine, Stanford University, Stanford, CaliforniaDepartment of Medicine, Stanford University, Stanford, California
- gDepartment of Medicine, Division of Cardiology, Lifespan Cardiovascular Institute, Brown University, Providence, Rhode IslandLifespan Cardiovascular Institute/Brown University, Providence, Rhode Island
- hDepartment of Medicine, Hennepin Healthcare Systems, University of Minnesota, Minneapolis, MinnesotaHennepin Healthcare Systems, University of Minnesota, Minneapolis, Minnesota
- iDepartment of Medicine, Division of Cardiology, Duke University School of Medicine, Duke Clinical Research Institute, Durham, North CarolinaiDuke University School of Medicine, Duke Clinical Research Institute, Durham, North Carolina
- ↵∗Address for correspondence:
Dr. Laxmi S. Mehta, The Ohio State University, 473 West 12th Avenue, DHLRI Suite 200, Columbus, Ohio 43210.
Physician burnout has a negative impact on patient care, productivity and job retention, whereas the cost of recruiting and replacing burned-out physicians can be steep (1). The current health care environment places strong emphasis on accomplishing the triple aim: improving population health and enhancing patient experience while reducing overall costs; however, absence of clinician well-being has a negative impact on achievement of these aims. Accordingly, many have called for modification of health care goals to include clinician well-being, thereby expanding to a “quadruple aim” (2). To date, there has been highly variable uptake of the quadruple aim's call to action with inconsistent efforts to address well-being across institutions.
The American College of Cardiology recognizes the seriousness of clinician burnout and has incorporated the quadruple aim into its new strategic plan. There is a dearth of burnout data specific to cardiologists. The American College of Cardiology included questions specific to burnout and job satisfaction in its third Professional Life Survey (PLS) (3) in order to address this gap and to better characterize burnout among cardiologists.
The third PLS included questions on career and personal issues; details regarding the survey have been previously published (3). The Mini-Z survey, which measures the emotional exhaustion component of burnout, consisted of 10 questions regarding professional wellness (4). The PLS was e-mailed to 10,798 physicians; 2,313 (21%) completed the survey. The additional Mini-Z survey questions were completed by 2,274 cardiologists and fellows in training, including 1,321 men (58%) and 953 women (42%).
Survey respondents were categorized into 2 groups on the basis of self-reporting: no burnout group (no burnout or feeling stressed, but not burned out) versus burnout group (≥1 symptoms of burnout, constant feelings of burnout, or complete burnout feelings). For analyses, all responses from the Mini Z 10 questions were dichotomized using the top 2 options on a 5-point Likert scale; categorical variables were recoded into binary variables for analysis. PLS questions pertaining to demographic data, career satisfaction, and discrimination along with Mini-Z questions pertaining to work environment (control, chaos, culture/values, teamwork, and electronic medical record [EMR]) were correlated with presence or absence of burnout.
The p values were calculated using z-tests with Bonferroni-Holmes multiple comparison adjustments, chi-square and 1-way analysis of variance methods. Thresholds for statistical significance were set at p ≤ 0.05, p ≤ 0.01, and p ≤ 0.001 levels. Bivariate relationships with p < 0.05 were included into the multivariable forward stepwise likelihood ratio logistic regression model, with the explanatory variables entered simultaneously. IBM SPSS Statistics for Windows version 24.0 (IBM, Armonk, New York) was used for analyses.
The majority of survey respondents (73.2%) did not report having burnout symptoms. Of this group, 23.7% reported enjoying their work and 49.5% reported being under stress with less energy. Approximately one-quarter of respondents (26.8%) reported being burned out; 19.2% experienced at least 1 symptom of burnout, 6.4% reported chronic burnout symptoms that led to frequently thinking of work frustrations, and 1.2% reported feeling completely burned out to the point of possibly needing outside intervention. Mid-career cardiologists (8 to 21 years of practice) had the highest prevalence of experiencing burnout compared with fellows in training, early career and late career cardiologists (39% vs. 10%, 23% and 28%, respectively; p ≤ 0.01). Overall, women reported burnout more frequently than men (31% vs. 24%; p ≤ 0.001). Burnout respondents were less likely to report being married (79% vs. 85%; p ≤ 0.01), and more likely to report dissatisfaction with family life (30% vs. 10%; p ≤ 0.001) and that family responsibilities hinder their ability to do professional work (46% vs. 29%; p ≤ 0.001).
Neither cardiovascular subspecialty nor practice setting had an impact on burnout. Burnout respondents self-reported a greater percentage of their weekly time in direct clinical practice (73% vs. 69%; p ≤ 0.01) compared with those without burnout. Overall all cardiologists were satisfied with their career, but burnout respondents were significantly more likely to report feeling less satisfied with achieving professional goals (43% vs. 18%), financial compensation (49% vs. 32%) and level of advancement (75% vs. 65%), and less likely to recommend cardiology as a career (44% vs. 20%) compared with no burnout respondents, all p ≤ 0.001.
Physicians reporting burnout were less likely to report being treated fairly at work (61% vs. 86%), feeling valued (63% vs. 87%) and feeling that their contributions matter (65% vs. 88%) compared with their peers, all p ≤ 0.001. Burnout respondents were more likely to report experiencing discrimination during career or training (50% vs. 37%; p ≤ 0.001), and it adversely affected professional activities with colleagues and patients (64% vs. 48%, 29% vs. 18%, respectively; p ≤ 0.05) compared with the no burnout group.
For each of the Mini-Z survey responses, an adverse work environment was least common in no burnout/no stress physicians with increasingly higher rates in those stressed without burnout and highest among the burnout physicians (Figure 1). Multivariable analyses were performed to determine which factors independently correlated with burnout. This model had good accuracy (81.4%) and high correlation with a C-statistic of 0.86 (95% confidence interval [CI]: 0.84 to 0.87). Factors independently associated with a higher rate of burnout include no control over workload (odds ratio [OR]: 2.03; 95% CI: 1.57 to 2.62; p ≤ 0.001), hectic work environment (OR: 1.90; 95% CI: 1.45 to 2.50; p ≤ 0.001), misalignment of values (OR: 1.65; 95% CI: 1.21 to 2.25; p ≤ 0.001), family responsibilities hindering professional work (OR: 1.47; 95% CI: 1.15 to 1.88; p = 0.002), intent to renegotiate for support staff (OR: 1.68; 95% CI: 1.14 to 2.48; p = 0.009) and insufficient documentation time (OR: 1.40; 95% CI: 1.08 to 1.82; p = 0.011). Factors independently associated with lower rates of burnout include family life satisfaction (OR: 0.46; 95% CI: 0.34 to 0.63; p ≤ 0.001), encouraging cardiology as a career (OR: 0.61; 95% CI: 0.47 to 0.89; p ≤ 0.001), having a mentor who served as a career role model (OR: 0.70; 95% CI: 0.54 to 0.89; p = 0.004), and feeling of being treated fairly at work (OR: 0.683; 95% CI: 0.509 to 0.918; p ≤ 0.011).
Burnout among physicians is increasingly recognized as an important barrier to health care quality and efficiency. We believe our study to be the first to report burnout data specific to cardiologists and the first to explore aspects of their professional lives associated with burnout or its absence. More than one-quarter of U.S. cardiologists reported feeling burnout, and nearly one-half of the survey respondents reported feeling stressed, which may be a precursor to burnout. These data may inform efforts to enhance cardiologists’ well-being, which in turn is thought to improve patient care and reduce costs.
Compared with a Medscape survey that recently reported that 43% of U.S. cardiologists are burned out and/or depressed (5), our data show a much lower rate of burnout. Furthermore, nearly 50% in our survey reported experiencing a great deal of job stress, which was associated with adverse work conditions. However, the Medscape survey does not use a validated measure of burnout and is a convenience sample with no defined response rates. Others have reported that ∼44% of physicians are experiencing burnout (6), which is also much higher than the ∼27% of cardiologists in our study. It is unclear why the prevalence of burnout in the current study of cardiologists is lower than prior reports of other specialties but may be attributable to the burnout assessment tool used, cardiologists’ resilience, and/or unique elements of the field/practice of cardiology that may be more protective than those found in other disciplines. The Maslach Burnout Inventory (MBI) is a widely accepted tool to assess the emotional exhaustion, depersonalization, and low sense of personal accomplishment domains of burnout; however, the cost can significantly limit its use. The Mini-Z survey, which measures the emotional exhaustion domain of burnout, is freely available in the public domain and has questions specific to the health care work environment; hence it has been selected by the American Medical Association for use in its Steps Forward program. The Mini-Z’s single item measure of burnout has been shown to have good correlation with the MBI’s emotional exhaustion scale (7). Moreover, the discrepancy seen between our survey and previously reported rates of burnout is not unexpected, as the prevalence of burnout in studies using the Mini-Z survey instrument are lower than other studies using the full MBI among general internal medicine physicians (4,6).
Clinicians in all fields of medicine are experiencing an increase in burnout; however, the highest rates are apparent in those on the front lines of medicine who spend the most time on direct patient care (1). Among the many causes of burnout, the transition of medical records from paper charts to EMRs has been singled out as among the most important factors because it places a clerical burden on physicians that deprofessionalizes their work and has a negative impact on work-life balance (8,9). For every 1 h of time spent in direct face-to-face care with patients, approximately 2 additional hours are spent on desk work and electronic medical records; furthermore, physicians spent 1 to 2 h of personal time to complete additional computer work each night (9). In our study, 72% of those experiencing burnout reported documentation time pressures, and 57% reported increased EMR used at home, yet self-reported poor EMR efficiency did not correlate with burnout. This has been seen in other studies and may reflect inability to adequately or accurately judge EMR proficiency. Comparable to data on general internists (4), burned-out cardiologists in our study reported significant rates of job stress, lack of control over workload, misalignment of professional values, poor care team efficiency, and hectic work atmosphere.
The U.S. physician workforce is already projected to be short-staffed by 2030, and this may profoundly worsen with potentially reduced clinical work hours/effort and early retirement among physicians experiencing burnout. Therefore, more efforts are needed to understand and reduce physician burnout. There are many drivers of burnout, including, but not limited to, unrealistic efficiency/productivity targets, reduced time allocated per patient encounter, increased administrative and professional burdens, and exponential rise in documentation time with EMR. The intense workloads along with loss of autonomy, financial strain, and escalation in negative views toward doctors have resulted in a rise in burnout nationally among physicians (1,10). A great deal of attention is currently being directed toward finding solutions (e.g., medical scribes, workflow redesign) to burnout and improving physician well-being.
There were several limitations, including low response rate to the PLS, which were previously published (3). Demographic data regarding survey nonresponders were lacking and therefore limits comparability to survey responders. Another potential limitation was the lack of questions pertaining to the consequences of burnout (such as depression or intent to decrease productivity), however this could be considered for future study.
More than one-quarter of U.S. cardiologists reported burnout, which is increasingly recognized as a barrier to improving health care, and one that needs to be systematically addressed. Interestingly, compared with other studies, burnout rates among cardiologists were lower relative to other specialties in medicine. This warrants further investigation into potential protective factors of cardiology. By identifying modifiable drivers of burnout, our data may inform efforts to understand the causes of burnout, and to design solutions at an individual and organizational level.
All authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- confidence interval
- electronic medical record
- Maslach Burnout Inventory
- odds ratio
- Professional Life Survey
- Received February 20, 2019.
- Revision received April 10, 2019.
- Accepted April 16, 2019.
- 2019 American College of Cardiology Foundation
- Shanafelt T.D.,
- Noseworthy J.H.
- Bodenheimer T.,
- Sinsky C.
- Lewis S.J.,
- Mehta L.S.,
- Douglas P.S.,
- et al.
- Linzer M.,
- Poplau S.,
- Babbott S.,
- et al.
- Kane L.
- Shanafelt T.D.,
- West C.P.,
- Sinsky C.,
- et al.
- Rohland B.M.,
- Kruse G.R.,
- Rohrer J.E.
- Shanafelt T.D.,
- Dyrbye L.N.,
- Sinsky C.,
- et al.
- Sinsky C.,
- Colligan L.,
- Li L.,
- et al.