- Email: [email protected]
Emotional Wellness of Current Musculoskeletal Radiology Fellows
Original Investigation
Emotional Wellness of Current Musculoskeletal Radiology Fellows Jack Porrino, MD, Michael J. Mulcahy, PhD, Hyojeong Mulcahy, MD, Annemarie Relyea-Chew, JD, MS, Felix S. Chew, MD Rationale and Objectives: Burnout is a psychological syndrome composed of emotional exhaustion, depersonalization, and sense of lack of personal accomplishment, as a result of prolonged occupational stress. The purpose of our study was to determine the prevalence of burnout among current musculoskeletal radiology fellows and to explore causes of emotional stress. Materials and Methods: A 24-item survey was constructed on SurveyMonkey using the Maslach Burnout Inventory. We identified 82 musculoskeletal radiology fellowship programs. We recruited subjects indirectly through the program director or equivalent. Results: Fifty-eight respondents (48 male, 10 female) identified themselves as current musculoskeletal radiology fellows and completed the survey. Comparison of the weighted subscale means in our data to the Maslach normative subscale thresholds for medical occupations indicates that musculoskeletal radiology fellows report relatively high levels of burnout with regard to lack of personal accomplishment and depersonalization, whereas emotional exhaustion levels in our sample are within the average range reported by Maslach. Although male musculoskeletal radiology fellows experience relatively high levels in two of the three dimensions of burnout (depersonalization and personal accomplishment), female musculoskeletal radiology fellows experience relatively high burnout across all three dimensions. Job market-related stress and the effort required providing care for dependents significantly affect personal accomplishment. Conversely, imbalances in the work-life relationship and feelings of powerlessness are significantly associated with depersonalization and emotional exhaustion. Conclusions: Musculoskeletal radiology fellows report relatively high levels of burnout. Because the consequences of burnout can be severe, early identification and appropriate intervention should be a priority. Key Words: Burnout; emotional wellness; occupational stress. © 2017 The Association of University Radiologists. Published by Elsevier Inc. All rights reserved.
INTRODUCTION
B
urnout is a psychological syndrome occurring in human service workers and defined by Maslach along three dimensions: emotional exhaustion, depersonalization, and sense of lack of personal accomplishment (1). The syndrome is a result of prolonged occupational stress, in which an individual may become increasingly cynical, may become more callous toward those whom they are serving, and may grow increasingly dissatisfied with their accomplishments within the workplace (1,2). The field of radiology, as in every area of medicine in the United States, has undergone rapid changes in the past decade. Radiologists in both the private and academic sectors have been compelled to adapt to the implementation of the Affordable Care Act and accompanying legislation
Acad Radiol 2017; 24:682–693 From the Department of Radiology, University of Washington, 1959 NE Pacific St, Seattle, WA 98195 (J.P., H.M., A.R.-C., F.S.C.); Central Washington University, Ellensburg, Washington (M.J.M.). Received October 5, 2016; revised December 12, 2016; accepted December 14, 2016. Address correspondence to: J.P. e-mail: [email protected] © 2017 The Association of University Radiologists. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.acra.2016.12.024
682
directed at healthcare reform (3). The resulting effects, such as consolidation in the provider and insurance markets and reduced reimbursement rates, have led to negative personal and work satisfaction for individual physicians (4). In the ecology of radiologists in the United States, subspecialty fellows occupy a unique niche between residents and practicing academic physicians. We surmised that this intermediate position, subjecting them to the clinical responsibilities of academic practice but the transience, lower pay, and limited autonomy that come with a training position, coupled with the uncertainty of a job search, might result in high levels of burnout. If burnout is prevalent in high proportions, then preventing or ameliorating it in trainees should become an important focus of fellowship program directors. The purpose of our study was to determine the current prevalence of burnout among musculoskeletal (MSK) radiology fellows, and to explore causes of stress that may contribute to burnout. We were also able to compare the prevalence of burnout and contributing stresses that MSK radiology fellows experience to that of MSK radiology faculty members. MSK radiology fellows were chosen as the specific study population because the authors are affiliated with an MSK radiology section and had access to this population; we did not have access to other populations of subspecialty fellows.
Academic Radiology, Vol 24, No 6, June 2017
MATERIALS AND METHODS Human Subjects
This investigation of human subjects was granted exempt status from review by the University of Washington institutional review board. Population and Sampling
We conducted a survey of current MSK radiology fellows. The survey instrument was created on SurveyMonkey (Professional) (5). MSK fellows are defined as individuals who have successfully completed a residency in diagnostic radiology and are currently engaged in a post-residency structured training program in MSK radiology. We identified 82 different MSK radiology fellowship programs through the Society of Skeletal Radiology web site: www.skeletalrad.org (6). The Society of Skeletal Radiology web site is believed to be the most complete and accurate listing of MSK radiology fellowships. In February 2016, 180 MSK radiology fellowship positions were listed as available for the academic year 2018. Each listing in-
Figure 1.
EMOTIONAL WELLNESS OF MSK RADIOLOGY FELLOWS
cluded contact information for the program director (an MSK radiologist), for the program coordinator (administrative person), or for both. We recruited subjects indirectly through the MSK program directors or other program contacts by e-mail, and requested that MSK fellows be made aware of our survey. Participation in the survey by MSK fellows was voluntary. An internet link to the survey was included in the message. Two requests were made to the MSK program directors or other contacts, separated by 1 week. The survey was closed 1 week after the second request. The responses were collected by SurveyMonkey and remained anonymous to the authors. No identifiable information about the respondents was solicited or obtained during the survey process, and there was no contact between the authors and the MSK fellows.
Survey
A 24-item survey was constructed on SurveyMonkey (Fig 1—sample item). The survey includes demographic questions, questions from the Maslach Burnout Inventory (MBI),
Survey question. Example of survey question as presented to respondents on surveymonkey.com.
683
PORRINO ET AL
questions about stress, and an open-ended comment question. The full survey is available as an Appendix. Demographics The demographic questions were placed at the start of the survey, and the responses were used to confirm the status of the respondent as an MSK fellow and to establish gender. Respondents who indicated that they were not currently fellows were excluded. Maslach Burnout Inventory In 1981, Maslach and Jackson defined burnout as a syndrome of emotional exhaustion and cynicism that occurs frequently among individuals who perform services and interact with other human beings. The key aspects of burnout are emotional exhaustion, the development of a negative and cynical attitude toward the client (depersonalization), and the tendency to evaluate oneself negatively (personal accomplishment) (1). Based on their initial burnout research, Maslach and Jackson developed the MBI designed to measure hypothesized aspects of the burnout syndrome (1). The inventory evaluates the frequency and intensity of three “subscales” of the burnout syndrome: emotional exhaustion (nine items), depersonalization (five items), and (the lack of a sense of) personal accomplishment (eight items) (1). This 22-item self-administered tool, later revised, asks respondents to rate the frequency with which they experience various feelings or emotions on a 7-point Likert scale, with response options ranging from “never” to “daily” (7,8). The MBI has been used in the past in an effort to assess burnout, and is a useful instrument in detecting the pattern of burnout among physicians. High levels of emotional exhaustion and depersonalization have come to be viewed as the most salient dimensions of burnout among medical professionals, a specific occupational group in which low levels of personal accomplishment may be less likely (2,8,9). The length of the MBI reflects a practical limitation of its routine use, prompting the use of condensed, single-item measures of emotional exhaustion and depersonalization adapted from the original tool as a more efficient means of assessing burnout in healthcare professionals (8,10). Emotional exhaustion has been assessed with a single item, “I feel burned out from my work,” whereas depersonalization was addressed with the item, “I have become more callous toward people since I took this job,” as these two particular measures have demonstrated the highest factor loading among their respective burnout domains across numerous studies (8,10). Utilization of single-item measures has proven to stratify the risk of burnout validly when compared to the use of the full MBI (8,10). In our survey, we used seven questions adapted from the MBI. As validated in earlier work, we used single items to address emotional exhaustion and depersonalization. We also incorporated five of the eight items from the MBI to assess personal accomplishment, as described by McNeeley et al. (2). 684
Academic Radiology, Vol 24, No 6, June 2017
Stressors Based on our review of the literature regarding the burnout syndrome in medicine, we included 13 questions measuring factors that may contribute to the development of burnout. Radiology as a career.—Substantive differences in burnout have been observed by specialty, with particularly high rates noted among emergency medicine, general internal medicine, neurology, and family medicine physicians (11). Specifically, two questions regarding satisfaction with radiology as an occupation were included. Financial stress.—High levels of perceived financial strain have been linked to worse overall mental health (12). Quality of life, satisfaction with work-life balance, emotional exhaustion, feelings of depersonalization, and burnout have all been demonstrated to be negatively affected by increasing levels of education debt among internal medicine residents (13). Conversely, quality of life and satisfaction with work-life balance were higher among internal medicine residents who participated in moonlighting (work at a second job outside of regular residency training duty hours) in the same study (13). Similar results have been found in radiology residents (2). Studies examining individuals with high debt typically employ a singleitem measure for assessing financial strain or debt burden (12). Selenko and Batinic developed a more comprehensive sixitem survey, on a 7-point scale, to assess the relationship between economic stress and individual well-being (12). In our study, we used three questions devised by Selenko and Batinic that address a general feeling of financial strain (12). Balancing work and life.—Physicians work longer hours and have greater difficulties with work-life balance when compared to other workers in the United States. These difficulties include the feeling that a commitment to their children has slowed their career advancement, as well as work-home conflicts. This imbalance, and resultant dissatisfaction, has been linked to physician burnout (11,14). To assess the effects of work-life imbalance and time needed for the appropriate care for dependents among radiology fellows, we utilized two items focused on the integration of work and personal and family life. Healthcare evolution and job market.—In a study conducted comparing US physicians to the general population, there was a 10% rise in the prevalence of physician burnout over a 3-year period, 2011–2014 (4). During this time frame, there were rapid changes in the practice and reimbursement of medicine, including implementation of the Affordable Care Act, reduced reimbursement rates, and the inadvertent effects on the job market. These issues continue. We therefore included four survey items to evaluate the possible association between the evolution of healthcare and job market constraints and physician burnout. Feelings of powerlessness and isolation.—Shanafelt et al. demonstrated that physicians who spent at least 20% of their time
Academic Radiology, Vol 24, No 6, June 2017
EMOTIONAL WELLNESS OF MSK RADIOLOGY FELLOWS
TABLE 1. Descriptive Statistics for Reduced MBI Subscales for MSK Radiology Fellows, and Comparison to Normative Values for Medical Occupations† Panel A: Descriptive Statistics for Reduced MBI Subscales and Independent Variables Variable Personal accomplishment subscale Emotional exhaustion subscale Depersonalization subscale Gender (unweighted) Weighted Stress measures Financial Work-life balance Dependent care Job market Healthcare regulations Isolation at work Powerlessness at work Satisfaction with MSK subfield
Obs.
Number of Items
Mean
Linearized Std. Error
58 58 58 58 58
5 1 1 1 1
17.73 2.79 2.22 .17 .21
.49 .22 .21 .38* .06
16.74–18.72 2.35–3.23 1.81–2.63
58 58 58 58 58 58 58 58
3 1 1 2 2 1 1 2
6.14 2.59 2.08 4.98 4.08 1.34 2.08 3.52
.27 .10 .12 .18 .18 .08 .11 .28
5.60–6.69 2.40–2.79 1.84–2.32 4.63–5.34 3.72–4.45 1.17–1.51 1.87–2.30 2.95–4.1
95% CI
.09–.33
Panel B: Comparison of Reduced MBI Subscale Means for MSK Radiology Fellows to Normative Subscale Means and High Burnout Thresholds for Medical Occupations
Variable Personal accomplishment subscale Emotional exhaustion subscale Depersonalization subscale
Number of Items in Full MBI
Number of Items in Reduced MBI items
8 9 5
5 1 1
Item Weight
Item-weighted Subscale Means (MSK Fellows)
Normative Subscale Means (Medicine)
Normative Subscale Thresholds for High Burnout (Medicine)
1.6 9 5
28.37 25.11 11.10
36.53 22.19 7.12
≤33 ≥27 ≥10
CI, confidence interval; MBI, Maslach Burnout Inventory; MSK, musculoskeletal. * Standard deviation, rather than linearized standard error, for unweighted gender variable. † Post-stratification gender weights applied, based on estimated female population proportion of .205.
in the aspect of work they found most meaningful had a burnout rate roughly half that of those physicians who spent less than 20% effort in the activity that was meaningful to them (15). The parameters evaluated included patient care, research, education, administration, and non-visit patient care activities (15). These data indicate that “career fit” is an important variable in determining physician burnout. To assess satisfaction with personal practice trends, our survey included two items that address the feeling of powerlessness regarding decision-making and isolation in the workplace (Appendix). Open-ended comments.—One open-ended question that solicited free-text comments concluded the survey. Statistical Analysis
Statistical analysis was performed using Stata (StataCorp, 2015, Stata Statistical Software: Release 14, College Station, TX) and Microsoft Excel (Microsoft, Redmond, WA).
RESULTS The mean, standard deviation, and minimum and maximum scores for the variables used in our analyses are presented in Table 1, Panel A.
Demographics
There were 58 respondents who identified themselves as current MSK radiology fellows and who completed the survey. A subsequent independent census of current MSK radiology fellowships with 100% response rate found that there were 190 physicians currently in these fellowships during the July 1, 2015, to June 30, 2016, academic year, 39 of whom were women (16). Of the 58 respondents in our study, 48 (82.8%) identified themselves as male, and 10 (17.2%) as female. Because the proportion of women in our sample is smaller than the actual proportion of women in MSK radiology fellowships, we have sampling bias. As earlier research finds that women differ from men in their susceptibility to burnout, as well as 685
PORRINO ET AL
Academic Radiology, Vol 24, No 6, June 2017
TABLE 2. Comparison of MSK Radiology Fellows Mean Weighted rMBI Scales, Disaggregated by Sex, to Maslach’s Normative MBI Subscale Thresholds and Sex-disaggregated MBI Subscale Means
Normative thresholds for high burnout (physicians and nurses) Male Maslach 2010 MSK radiology Female Maslach 2010 MSK radiology Adjusted Wald tests of subscale means (male-female comparisons) F (1, 57) Prob > F
Emotional Exhaustion
Depersonalization
Personal Accomplishment
≥27 19.86 24.37 20.99 27.90*
≥10 7.43 10.73* 7.02 12.5*
≤33 36.29 28.3* 36.50 28.64*
0.71 0.40
0.52 0.47
0.05 0.83
MBI, Maslach Burnout Inventory; MSK, musculoskeletal; rMBI, reduced Maslach Burnout Inventory. * Indicates value in “high burnout” range of Maslach et al. (2010) normative distributions.
exposure and responses to stressors, our analyses used poststratification weights based on the female proportion of 20.5% in the population of MSK radiology fellows (Chew FS, unpublished data, 2016). Burnout
As described, we utilized a modified, reduced MBI in which single items measured the emotional exhaustion and depersonalization subscales, whereas five items were used to assess personal accomplishment. In our data, the raw average scores for the single-item emotional exhaustion and depersonalization subscales were 2.2 and 2.8, respectively. The raw average aggregate score on the fiveitem personal accomplishment subscale was 17.73, with an item average of 3.5. Maslach et al reports normative ranges for low, medium, and high levels of burnout for various occupational groups (7). In comparing our subscales to those normative ranges, we used Maslach’s “medicine” occupational group (based on MBI scales for 1104 physicians and nurses, not otherwise described) as our reference cohort. Maslach et al states that for this occupational group, depersonalization subscale scores above 9.0, emotional exhaustion subscale scores above 26.0, and personal accomplishment subscale scores below 34.0 indicate high levels of burnout in the respective dimensions (7). To render the subscale scores in our data directly comparable to the normative data published in Maslach et al, we created item weights for the items used for our reduced MBI subscales (7). For each subscale, we applied an item weight using the ratio of the number of items in the full MBI subscale to the number of items in our modified, reduced MBI subscale (Table 1). Panel B of Table 1 includes the number of items in the full MBI and in the reduced MBI used in our research, the resulting weights, and our weighted subscale means. The final two columns in Panel B of Table 1 show Maslach’s normative subscale means and thresholds for “high burnout” for the occupation of “medical workers” (7). Maslach et al 686
used physicians and nurses for the “medical workers” occupation, but did not further identify them. Tables 1–3 show the distributions of item-weighted depersonalization, emotional exhaustion, and personal accomplishment subscale totals for MSK radiology fellows. In our data, the item-weighted mean depersonalization subscale score (Table 1, Panel B) is 11.1, above the 9-point threshold for high burnout. On average, then, MSK radiology fellows experience relatively high levels of “depersonalization” burnout, in comparison to the data on physicians and nurses that were used to form the normative ranges. In contrast, the itemweighted mean emotional exhaustion subscale score in our data (Table 1, Panel B) is 25.1. Although higher than the mean score of 22.19 in Maslach’s general population normative data, it remains below their high emotional exhaustion threshold of 27 for physicians and nurses (7). This indicates that MSK radiology fellows do not experience particularly high levels of emotional exhaustion, compared to other medical workers. Our respondents do, however, experience relatively high levels of burnout in association with the experience of lack of personal accomplishment (Table 1, Panel B), with an itemweighted average subscale score of 28.4, that is, well below Maslach’s threshold of 34 for physicians and nurses (7). In summary, comparison of the weighted subscale means in our data to the normative subscale thresholds for medical occupations indicates that MSK radiology fellows report relatively high levels of burnout with regard to lack of personal accomplishment and depersonalization, whereas emotional exhaustion levels in our sample, although relatively high, are still within the average range reported by Maslach et al (7) (Table 1, Panel B). The average levels of personal accomplishment and depersonalization burnout among the MSK radiology fellows in our sample are higher than those in the medical worker samples analyzed using the full MBI (7). Table 3 shows that the pattern of correlations among the (gender and itemweighted) subscales in our data is remarkably similar to the correlations among the full MBI subscales based on large samples, for a range of occupational groups, as reported by
Academic Radiology, Vol 24, No 6, June 2017
TABLE 3. Correlations Among Reduced MBI Subscales for MSK Fellows, Compared to Correlations Among Full MBI Subscales, Various Occupational Groups†
Depersonalization Personal accomplishment
Emotional Exhaustion
Depersonalization
.54* (.52) −.11 (−.22)
−.17 (−.26)
MBI, Maslach Burnout Inventory; MSK, musculoskeletal. * Significant at P < 0.000. † Numbers in parentheses are intercorrelations between full MBI subscales, reported in Maslach et al. (2010), appendix B. Poststratification gender weights applied to reduced MBI subscales, based on estimated female population proportion of .205.
Maslach et al. (7). The statistically significant 0.54 correlation coefficient for the single-item measures of depersonalization and emotional exhaustion mirrors the findings of Maslach et al for full MBI depersonalization and emotional exhaustion almost exactly. In our data, the negative and not statistically significant coefficients for the personal accomplishmentdepersonalization and personal accomplishment-emotional exhaustion correlations, although slightly smaller than those found in the validation studies by Maslach and collaborators, are otherwise entirely consistent with that research. Gender and Burnout
Maslach et al also reported demographic norms for the MBI subscales, although these are not disaggregated by occupational grouping (7). Table 2 shows the distributions of the depersonalization, emotional exhaustion, and personal accomplishment subscales by respondents’ gender, and compares the subscale means in our data, disaggregated by gender, to the normative demographic means in Maslach et al (7). Table 2 also includes the results of mean comparison tests. In both our data and in the comparison data, females report higher levels of emotional exhaustion; the differences between females and males are greater among MSK radiology fellows than in the Maslach et al data, although the difference is not statistically significant (7). It is noteworthy, however, that in our gender-disaggregated data, the average score for females is above the threshold for “high emotional exhaustion burnout,” but the average score for males remains below this threshold. Whereas Maslach et al found little difference between males and females with regard to depersonalization, female MSK radiology fellows report higher levels of depersonalization burnout than males. Again, these differences do not reach statistical significance in this small sample. In comparison to Maslach et al, we find no differences between males and females with regard to personal accomplishment scores (7). In summary, Table 2 shows that, although male MSK radiology fellows experience relatively high levels in two of the three dimensions of burnout (depersonalization and personal accomplishment), compared to the samples of medical
EMOTIONAL WELLNESS OF MSK RADIOLOGY FELLOWS
professionals analyzed in Maslach et al, female MSK radiology fellows experience comparatively higher burnout across all three dimensions (7). The gender differences in the levels of burnout are not, however, statistically significant in our relatively small sample. Prevalence of Burnout
Among MSK radiology fellows, 49 (84%) had weighted individual totals for the personal accomplishment burnout items that were at or below the threshold of 33 identified by Maslach et al for high personal accomplishment burnout among medical professionals (7). With regard to the depersonalization dimension of burnout, 39 (67%) of our respondents had weighted items scores at or above the threshold of 10 for high depersonalization burnout among medical workers. Even in the burnout dimension of emotional exhaustion—the only burnout dimension in which our sample means were within Maslach’s normal range for medical professionals—the weighted scores for over half of the MSK radiologists in our sample (33 respondents, or 57%) were at or above the normative threshold of 27 for high emotional exhaustion burnout (7). Overall, 21 respondents, or 36.2% of the sample, had weighted subscale scores that placed them in the high burnout range in all three dimensions of burnout measured in the study. An equal number of MSK radiology fellows (21, or 36.2%) exhibited high burnout in two of the three measured burnout dimensions, and another nine respondents (15.5%) reported a high level burnout in only one of the three dimensions. Only seven respondents (12.1%) did not meet the threshold for burnout in any of the dimensions. Thus, the overall prevalence of burnout among MSK radiology fellows was 88%. We can compare the prevalence of burnout among MSK radiology fellows to the prevalence of burnout among MSK radiology faculty members by looking at the results from an independently performed survey of practicing MSK radiologists (16). Radiologists in academic practice are likely to be the faculty members who are training the MSK radiology fellows, although not every academic radiology department has an MSK fellowship. It would seem appropriate to compare the prevalence of burnout in this subgroup of practicing MSK radiologists to that of MSK radiology fellows to gain some sense of whether the situation of the trainees is better or worse than that of the faculty. Burnout was present among many MSK faculty members, but was less prevalent and less severe than that of MSK fellows. The proportion of MSK faculty with no manifestations of burnout was approximately double that of MSK fellows (25% vs 12%), whereas the proportion of faculty with three manifestations was approximately half that of fellows (19% vs 36%) (Fig 2). Sources of Burnout
We examined the relationship between stressors and burnout. The survey included 13 questions that measured stress arising from financial concerns (three items), job market concerns 687
PORRINO ET AL
Figure 2. Burnout affecting MSK faculty and fellows. Column chart comparing the percentage of MSK faculty and fellows with the number of burnout manifestations. MSK, musculoskeletal).
(two items), regulatory changes in medicine including reference to the Affordable Care Act (two items), work-life balance concerns (one item), care for dependents (one item), workplace experiences of isolation (one item) and powerlessness (one item), and satisfaction with the MSK radiology subfield (two items). In the survey instrument, the stress items used a 5-point Likert scale, but in our analyses, we collapsed the extremes on both ends of the distribution to create three-level response categories, with 1 = “strongly disagree/disagree,” 2 = “neutral,” and 3 = “agree/strongly agree.” The average scores for these measures are reported below the subscale averages in Table 1. We also examined gender differences in reported stress levels. Compared to females, males report higher levels of financial stress (6.39 vs 5.2, P < 0.11), slightly lower levels of stress from job market concerns (4.87 vs 5.4, P < 0.23), significantly lower levels of stress associated with the regulatory changes of the Affordable Care Act (3.92 vs 4.7, P < 0.032), slightly higher levels of stress due to poor work-life balance (2.62 vs 2.5, P < 0.65), and substantially and significantly higher levels of stress related to care for dependents (2.21 vs 1.6, P < 0.037). Males did not differ from females with regard to stress associated with feelings of isolation at work (1.35 vs 1.3 for females, P < 0.81) or feelings of powerlessness at work (2.1 vs 2.0 for females, P < 0.71), whereas males reported significantly higher levels of dissatisfaction with the MSK radiology subfield than females (3.71 vs 2.8, P < 0.05). Although we found some significant differences between males and females with regard to sources of stress, the aggregate stress scores are quite similar, with males’ stress reports aggregating only slightly higher, to an average of 27.19, compared to females’ average aggregate stress score of 25.5 (P < 0.34). Regression Analyses
To understand the effects of specific stressors on burnout, we regressed our emotional exhaustion, depersonalization, and per688
Academic Radiology, Vol 24, No 6, June 2017
sonal accomplishment subscales on the stress measures described. The regression models also included a variable measuring the respondents’ satisfaction with the MSK radiology subfield, and a gender control. To facilitate interpretation of the effects of these stressors on various dimensions of burnout, we used reverse-coded emotional exhaustion and depersonalization subscales in the regressions. All of the stressor and subfield satisfaction variables were also reverse-coded, so that for the burnout subscales, the stressor measures, and the subfield satisfaction measures, higher values represent higher levels of burnout, stress, and dissatisfaction, respectively. The results of the regression analyses are shown in Table 4. We first note that, reading across the three models in Table 4, we find no effects of financial stressors, changes in healthcare regulations, or isolation at work on any of the three burnout subscales. These variables appear unrelated to burnout among MSK radiology fellows. Furthermore, stress related to concerns about the job market only significantly affects the personal accomplishment component of burnout (model 1), and is unrelated to the depersonalization and emotional exhaustion burnout subscales. Stress resulting from poor work-life balance, in contrast, does not significantly increase the personal accomplishment component of burnout, but is positively and significantly associated with higher levels of the other two burnout components—depersonalization and emotional exhaustion— with particularly strong effects on the latter component. Stress among MSK radiology fellows resulting from dependent care responsibilities, on the other hand, is significantly related to higher levels of the personal accomplishment component of burnout, and has no significant effects on the depersonalization component of burnout. Model 3 shows an unexpected, statistically significant negative effect of stress from care for dependents on the emotional exhaustion subscale. This indicates that those MSK radiology fellows who report higher levels of stress resulting from care for dependents report, on average, lower levels of emotional exhaustion. This apparently counterintuitive finding makes more sense when we consider the gender differences in levels of stress associated with care for dependents in our data, on the one hand, and emotional exhaustion, on the other. Males—who, even among dual-career couples, typically bear much less responsibility for caring for dependents—report significantly higher average levels of stress from this source (2.2 for males vs 1.6 for females, P < 0.05) (17–19). The modal response to our question about stress associated with care for dependents is “agree/strongly” agree for males and “disagree/strongly disagree” for females. The result, as shown in model 3, is that the demographic group (males) reporting higher levels of stress from caregiving responsibilities is also the demographic group reporting lower levels of emotional exhaustion. To clarify the relationship between gender, stress from care for dependents, and emotional exhaustion, we estimated a fourth regression model (not shown in Table 4) that included an interaction term to test for statistical interaction effects between gender and stress from care for dependents. When the inter-
Academic Radiology, Vol 24, No 6, June 2017
EMOTIONAL WELLNESS OF MSK RADIOLOGY FELLOWS
TABLE 4. Regression of Reduced MBI Personal Accomplishment, Depersonalization, and Emotional Exhaustion Subscales on Stressors and Subfield Satisfaction† Model 1
Independent Variables Gender Stress measures Financial stress Job market stress Healthcare regulations Work-life balance Care for dependents Isolation at work Powerlessness at work Satisfaction with MSK radiology subfield Constant Observations R-squared
Model 2
Model 3
Personal Accomplishment Subscale
Depersonalization Subscale
Emotional Exhaustion Subscale
0.149 (0.211)
0.576 (0.675)
0.486 (0.352)
−0.125 (0.176) 0.348** (0.135) −0.167 (0.173) 0.054 (0.105) 0.323*** (0.105) 0.063 (0.140) 0.171 (0.128) 0.094 (0.096) 0.133 (0.408) 58 0.279
0.017 (0.347) 0.104 (0.258) −0.051 (0.347) 0.386* (0.224) −0.180 (0.288) 0.403 (0.265) 0.473** (0.236) 0.469*** (0.137) −1.073 (0.938) 58 0.318
−0.062 (0.239) 0.140 (0.291) 0.031 (0.320) 0.690*** (0.204) −0.481** (0.204) 0.340 (0.236) 0.689*** (0.199) 0.627*** (0.126) −1.381* (0.822) 58 0.532
MBI, Maslach Burnout Inventory; MSK, musculoskeletal. *** P < 0.01, **P < 0.05, *P < 0.10. † Standard errors in parentheses; post-stratification gender weights applied, based on estimated female population proportion of .205.
action of gender with stress from care for dependents is included in the model, the main effect of high levels of stress from care for dependents represents the effect of this stressor for males (ie, when gender = “0”). This coefficient is negative and statistically significant (P < 0.05). The interaction coefficient indicates the effects of stress from care for dependents for women (ie, gender = “1”). The interaction coefficient of 1.17 is, in magnitude, nearly identical to the main effect of stress from care for dependents (ie, the effect for males, at −1.16), and also statistically significant (P < 0.05), but in the opposite direction, indicating, among female MSK fellows, a positive effect of stress from care for dependents on emotional exhaustion burnout. Stress resulting from feelings of powerlessness at work is not significantly related to the personal accomplishment component of burnout, but it is a significant and relatively powerful predictor of the burnout experiences of depersonalization and emotional exhaustion. Finally, we find a similar pattern of effects of MSK radiology fellows’ level of (dis)satisfaction with their medical subfield: as with stress resulting from feelings of powerlessness at work, dissatisfaction with the subfield does not
significantly affect the personal accomplishment component of burnout, but those MSK radiology fellows who report dissatisfaction with their subfield also tend to report higher levels of depersonalization and emotional exhaustion.
Free-text
Only six MSK fellows provided an open-ended response. In summary, these responses cited pressures or stress related to an apparent need, whether real or perceived, to read more studies, balancing work and private life demands, the cost of higher education/medical school, and a competitive job market.
DISCUSSION Why Does Burnout Matter?
According to Maslach et al, the burnout syndrome can lead to deterioration in the quality of care or service that is provided by staff, and can contribute to job turnover, absenteeism, and low morale (1). Additionally, burnout may lead to physical 689
PORRINO ET AL
exhaustion, insomnia, increased use of alcohol and drugs, and marital and family problems (1). Burnout among medical professionals has been linked to a lower degree of medical knowledge, the deterioration of professionalism, suboptimal patient care, medical errors, early retirement, and reduced empathy (8,11,13,15). Burnout also appears to be associated with personal consequences among physicians, including contributions to broken relationships, problematic alcohol use, and suicidal ideation (11). As a societal issue, burnout among physicians has received enough attention in the lay press that the trope of the burned-out physician in popular culture has become commonplace on television and on social media. The protagonist of the show “House,” as portrayed by the actor Hugh Laurie, is an example of the burnedout and marginally functional but brilliant physician whose manifestations of emotional exhaustion, depersonalization, and low personal achievement inform the plot lines of the popular show. Maslach’s initial research into the phenomenon of burnout included a variety of human service professions besides medicine, such as social services, mental health, criminal justice, and education, so burnout is not exclusive to physicians. Recent attention to burnout in doctors in the lay press does not appear to be unique to that profession; there have been recent New York Times articles on burnout in disparate occupations and situations, including athletes, coaches, athletic administrators, business executives, priests, professional dancers, unpaid caregivers for family members, teachers, employees of various types of businesses, and even military drone operators, a job that would not initially appear to fall under the human services professions rubric. Addressing this apparent rising tide of burnout that broadly affects our society is beyond the scope of our research, but it is important that we acknowledge the problem in our own profession and help those for whom we have responsibility. Burnout in MSK Radiology Fellows
Unfortunately, MSK radiology fellows report relatively high levels of burnout with respect to lack of personal accomplishment and depersonalization. Emotional exhaustion levels from our sample are within the average range reported for medical occupations (7). Why such high levels of burnout in the MSK radiology field? Is there something about the nature of MSK that promotes such a result? We think perhaps the opposite may be true. Compared to many other subspecialties of radiology and other fields of medicine in general, diagnosis and management of MSK patients seem in many respects as if it should be less stressful: fractures will heal, joints can be replaced, severe morbidity and death are uncommon, and bone cancer is rare. If this is the case, then one might expect burnout to be even more severe in other subspecialty fellows. A comparison between MSK fellows and other radiology subspecialty fellows is not possible at this time because our own sample was restricted to MSK fellows and we have found no published studies of burnout in other radiology subspecialty fellows. We suggest, however, that our results are generalizable to fellows in other radiology subspecialties. Al690
Academic Radiology, Vol 24, No 6, June 2017
though the domain-specific subspecialty knowledge base differs from one radiology subspecialty to another, the hierarchical niche and function of a subspecialty fellow are very similar, so the consequent stresses should also be similar. We believe that it would be useful to confirm this notion by studying other subspecialty fellows. Job market-related stress and the effort required in providing care for dependents significantly affect the personal accomplishment component of burnout in MSK radiology fellows. Conversely, imbalances in the work-life relationship and feelings of powerlessness are significantly associated with the depersonalization and emotional exhaustion components of burnout. Those MSK radiology fellows who were dissatisfied with their area of subspecialty tended to report higher levels of depersonalization and emotional exhaustion. How Do We Put Out the Fire?
Because burnout is defined as a work-related psychological syndrome, Nicola et al. report a series of strategies that may serve as remedies for burnout in the context of radiology (20). These authors advocate physical fitness, a minimization of external disruptions and distractions while immersed in work, and attention to “latent benefits” endorsed by the late social psychologist Marie Jahoda. Specifically, these latent benefits include structured time commitments, routine workplace activity, an enhanced social network, a sense of purpose, and augmented social status (20). The American College of Radiology’s Commission on Human Resources published a report in 2016 addressing burnout in radiologists, but there was no comment about radiology residents or subspecialty radiology fellows. They recommended several workplace modifications, including reducing workloads and call schedules, as well as generically improved lifestyle balance and counseling (21). The American Medical Association’s STEPS Forward program has a module on “preventing physician burnout” with suggestions for improving wellness at the group or system level (22). Implementation and evaluation of the strategies to reduce burnout are beyond the scope of our current study, but may be appropriate topics for future research. Limitations
The major limitation to our study is the nature of the MBI itself: it is essentially a self-administered, self-report of the psychological manifestations of burnout. The definition of burnout does not have any other criteria for determining its presence or absence other than responses to the MBI, and burnout is not included in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) as a psychiatric diagnosis. Although the response rate to our survey was 30.5% (58 of 190), the population we studied was relatively small. CONCLUSIONS Radiology fellows have a high prevalence of burnout (88%). To prevent significant adverse outcomes of burnout, more
Academic Radiology, Vol 24, No 6, June 2017
time and effort must be dedicated to the early recognition and intervention of the condition. To give optimal care to patients, a work environment that supports the well-being and mental health of its healthcare providers is essential.
REFERENCES 1. Maslach C, Jackson SE. The measurement of experienced burnout. J Occup Behav 1981; 2:99–113. Available at: http://www.jstor.org/stable/ 3000281. Accessed October 3, 2016. 2. McNeeley MF, Perez FA, Chew FS. The emotional wellness of radiology trainees: prevalence and predictors of burnout. Acad Radiol 2013; 20:647–655. doi:10.1016/j.acra.2012.12.018. 3. Patient Protection and Affordable Care Act (“ACA”), Public Law 110– 148. Available at: https://www.gpo.gov/fdsys/granule/PLAW111publ148/PLAW-111publ148/content-detail.html. Accessed October 3, 2016. 4. Shanafelt TD, Hasan O, Dyrbye LN, et al. Changes in burnout and satisfaction with work-life balance in physicians and the general US working population between 2011 and 2014. Mayo Clin Proc 2015; 90:1600– 1613. doi:10.1016/j.mayocp.2015.08.023. 5. Society of Skeletal Radiology. Available at: https://skeletalrad.org/mskfellowships. Accessed October 3, 2016. 6. SurveyMonkey Inc. (Professional version). Palo Alto, CA. Available at: https://www.surveymonkey.com. Accessed October 3, 2016. 7. Maslach C, Jackson SE, Leiter MP. Maslach Burnout Inventory manual and non-reproducible instrument and scoring guides. Mountain View, CA: CPP, Inc., 2010. 8. West CP, Dyrbye LN, Sloan JA, et al. Single item measures of emotional exhaustion and depersonalization are useful for assessing burnout in medical professionals. J Gen Intern Med 2009; 24:1318–1321. doi:10.1007/s11606-009-1129-z. 9. Rafferty JP, Lemkau JP, Purdy RR, et al. Validity of the Maslach Burnout Inventory for family practice physicians. J Clin Psychol 1986; 42:488– 492.
EMOTIONAL WELLNESS OF MSK RADIOLOGY FELLOWS
10. West CP, Dyrbye LN, Satele DV, et al. Concurrent validity of singleitem measures of emotional exhaustion and depersonalization in burnout assessment. J Gen Intern Med 2012; 27:1445–1452. doi:10.1007/s11606012-2015-7. 11. Shanafelt TD, Boone S, Tan L, et al. Burnout and satisfaction with worklife balance among US physicians relative to the general US population. Arch Intern Med 2012; 172:1377–1385. 12. Selenko E, Batinic B. Beyond debt. A moderator analysis of the relationship between perceived financial strain and mental health. Soc Sci Med 2011; 73:1725–1732. doi:10.1016/j.socscimed.2011.09.022. 13. West CP, Shanafelt TD, Kolars JC. Quality of life, burnout, educational debt, and medical knowledge among internal medicine residents. JAMA 2011; 306:952–960. doi:10.1001/jama.2011.1247. 14. Dyrbye LN, Shanafelt TD, Balch CM, et al. Relationship between workhome conflicts and burnout among American surgeons: a comparison by sex. Arch Surg 2011; 146:211–217. doi:10.1001/archsurg.2010.310. 15. Shanafelt TD, West CP, Sloan JA, et al. Career fit and burnout among academic faculty. Arch Intern Med 2009; 169:990–995. doi:10.1001/ archinternmed.2009.70. 16. Chew FS, Mulcahy MJ, Porrino JA, et al. Prevalence of burnout among practitioners of a radiology subspecialty. RSNA 2016 Annual Meeting Program Book, 2016. 17. Bianchi SM, Sayer LC, Milkie MA, et al. Housework: who did, does or will do it, and how much does it matter? Soc Forces 2012; 91:55–63. 18. Henz U. Parent care as unpaid family labor: how do spouses share? J Marriage Fam 2010; 72:148–164. 19. Shelton BA, John D. The division of household labor. Annu Rev Sociol 1996; 299–322. 20. Nicola R, McNeeley MF, Bhargava P. Burnout in radiology. Curr Probl Diagn Radiol 2015; 44:389–390. doi:10.1067/j.cpradiol.2015.04.007. 21. Harolds JA, Parikh JR, Bluth EI, et al. Burnout of radiologists: frequency, risk factors, and remedies: a report of the ACR Commission on Human Resources. J Am Coll Radiol 2016; 13:411–416. doi:10.1016/j.jacr.2015 .11.003. 22. American Medical Association. Preventing physician burnout. AMA, 2015. Available at: https://www.stepsforward.org/Static/images/modules/15/ downloadable/Preventing_Physician_Burnout.pdf. Accessed December 7, 2016.
691
PORRINO ET AL
APPENDIX. SURVEY QUESTIONS DEMOGRAPHIC ITEMS 1. Which best describes your current practice setting? o Academic practice o Private practice o Teleradiology practice 2. How long has it been since you completed your postgraduate residency and fellowship training? o Not yet completed o 5 years or less o 6–10 years o 11–20 years o 21 years or more 3. What is your gender? o Male o Female MASLACH BURNOUT INVENTORY ITEMS 4. How often does this statement describe your feelings about practicing radiology: “I can effectively solve the problems that arise in my work.” o Many times a day o Every day o Every week o Every month o A few times per year o Every year or less o Never 5. How often does this statement describe your feelings about practicing radiology: “I feel I am making an effective contribution to my radiology department or practice.” o Many times a day o Every day o Every week o Every month o A few times per year o Every year or less o Never 6. How often does this statement describe your feelings about practicing radiology: “In my opinion, I am good at my job.” o Many times a day o Every day o Every week o Every month o A few times per year o Every year or less o Never 7. How often does this statement describe your feelings about practicing radiology: “I have accomplished many worthwhile things in this job.” o Many times a day o Every day 692
Academic Radiology, Vol 24, No 6, June 2017
o Every week o Every month o A few times per year o Every year or less o Never 8. How often does this statement describe your feelings about practicing radiology: “I feel I am making an effective contribution to patient care.” o Many times a day o Every day o Every week o Every month o A few times per year o Every year or less o Never 9. How often does this statement describe your feelings about practicing radiology: “I feel burned out from my work.” o Many times a day o Every day o Every week o Every month o A few times per year o Every year or less o Never 10. How often does this statement describe your feelings about practicing radiology: “I have become more callous toward people since I began my current job.” o Many times a day o Every day o Every week o Every month o A few times per year o Every year or less o Never POTENTIAL SOURCES OF STRESS 11. My current financial situation is a serious strain. o Agree strongly o Agree o Neutral o Disagree o Disagree strongly 12. I often think about my current financial situation. o Agree strongly o Agree o Neutral o Disagree o Disagree strongly 13. I have difficulties meeting my financial obligations. o Agree strongly o Agree o Neutral o Disagree o Disagree strongly
Academic Radiology, Vol 24, No 6, June 2017
14. Decreased reimbursement rates for radiology have generated a serious strain on my life. o Agree strongly o Agree o Neutral o Disagree o Disagree strongly 15. Trying to balance the demands of my work with the time needed for my personal and family life is a major source of stress. o Agree strongly o Agree o Neutral o Disagree o Disagree strongly 16. Providing, organizing, or paying for care for dependents (e.g. children, elderly parents) is a major source of stress in my life. o Agree strongly o Agree o Neutral o Disagree o Disagree strongly 17. The current competition in the musculoskeletal radiology job market serves as a major source of stress in my life. o Agree strongly o Agree o Neutral o Disagree o Disagree strongly 18. The prospect of losing my job or not finding a suitable one, as a reflection of the competitive job market, serves as a major source of stress in my life. o Agree strongly o Agree o Neutral o Disagree o Disagree strongly 19. The lack of input into decisions that affect my medical practice is a major source of stress in my life. o Agree strongly
EMOTIONAL WELLNESS OF MSK RADIOLOGY FELLOWS
o Agree o Neutral o Disagree o Disagree strongly 20. The new regulations resulting from the Affordable Care Act (Obamacare) are a major source of stress in my life. o Agree strongly o Agree o Neutral o Disagree o Disagree strongly 21. Professional isolation (working alone) serves as a major source of stress in my life. o Agree strongly o Agree o Neutral o Disagree o Disagree strongly CAREER CHOICE 22. How often does this statement describe your feelings about practicing radiology: “I wish I were in a different specialty of medicine.” o Many times a day o Every day o Every week o Every month o A few times per year o Every year or less o Never 23. I wish I worked in a different field instead of diagnostic radiology. o Agree strongly o Agree o Neutral o Disagree o Disagree strongly OPEN-ENDED COMMENT 24. Do you have any other comments, questions, or concerns?
693
Emotional Wellness of Current Musculoskeletal Radiology Fellows Jack Porrino, MD, Michael J. Mulcahy, PhD, Hyojeong Mulcahy, MD, Annemarie Relyea-Chew, JD, MS, Felix S. Chew, MD Rationale and Objectives: Burnout is a psychological syndrome composed of emotional exhaustion, depersonalization, and sense of lack of personal accomplishment, as a result of prolonged occupational stress. The purpose of our study was to determine the prevalence of burnout among current musculoskeletal radiology fellows and to explore causes of emotional stress. Materials and Methods: A 24-item survey was constructed on SurveyMonkey using the Maslach Burnout Inventory. We identified 82 musculoskeletal radiology fellowship programs. We recruited subjects indirectly through the program director or equivalent. Results: Fifty-eight respondents (48 male, 10 female) identified themselves as current musculoskeletal radiology fellows and completed the survey. Comparison of the weighted subscale means in our data to the Maslach normative subscale thresholds for medical occupations indicates that musculoskeletal radiology fellows report relatively high levels of burnout with regard to lack of personal accomplishment and depersonalization, whereas emotional exhaustion levels in our sample are within the average range reported by Maslach. Although male musculoskeletal radiology fellows experience relatively high levels in two of the three dimensions of burnout (depersonalization and personal accomplishment), female musculoskeletal radiology fellows experience relatively high burnout across all three dimensions. Job market-related stress and the effort required providing care for dependents significantly affect personal accomplishment. Conversely, imbalances in the work-life relationship and feelings of powerlessness are significantly associated with depersonalization and emotional exhaustion. Conclusions: Musculoskeletal radiology fellows report relatively high levels of burnout. Because the consequences of burnout can be severe, early identification and appropriate intervention should be a priority. Key Words: Burnout; emotional wellness; occupational stress. © 2017 The Association of University Radiologists. Published by Elsevier Inc. All rights reserved.
INTRODUCTION
B
urnout is a psychological syndrome occurring in human service workers and defined by Maslach along three dimensions: emotional exhaustion, depersonalization, and sense of lack of personal accomplishment (1). The syndrome is a result of prolonged occupational stress, in which an individual may become increasingly cynical, may become more callous toward those whom they are serving, and may grow increasingly dissatisfied with their accomplishments within the workplace (1,2). The field of radiology, as in every area of medicine in the United States, has undergone rapid changes in the past decade. Radiologists in both the private and academic sectors have been compelled to adapt to the implementation of the Affordable Care Act and accompanying legislation
Acad Radiol 2017; 24:682–693 From the Department of Radiology, University of Washington, 1959 NE Pacific St, Seattle, WA 98195 (J.P., H.M., A.R.-C., F.S.C.); Central Washington University, Ellensburg, Washington (M.J.M.). Received October 5, 2016; revised December 12, 2016; accepted December 14, 2016. Address correspondence to: J.P. e-mail: [email protected] © 2017 The Association of University Radiologists. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.acra.2016.12.024
682
directed at healthcare reform (3). The resulting effects, such as consolidation in the provider and insurance markets and reduced reimbursement rates, have led to negative personal and work satisfaction for individual physicians (4). In the ecology of radiologists in the United States, subspecialty fellows occupy a unique niche between residents and practicing academic physicians. We surmised that this intermediate position, subjecting them to the clinical responsibilities of academic practice but the transience, lower pay, and limited autonomy that come with a training position, coupled with the uncertainty of a job search, might result in high levels of burnout. If burnout is prevalent in high proportions, then preventing or ameliorating it in trainees should become an important focus of fellowship program directors. The purpose of our study was to determine the current prevalence of burnout among musculoskeletal (MSK) radiology fellows, and to explore causes of stress that may contribute to burnout. We were also able to compare the prevalence of burnout and contributing stresses that MSK radiology fellows experience to that of MSK radiology faculty members. MSK radiology fellows were chosen as the specific study population because the authors are affiliated with an MSK radiology section and had access to this population; we did not have access to other populations of subspecialty fellows.
Academic Radiology, Vol 24, No 6, June 2017
MATERIALS AND METHODS Human Subjects
This investigation of human subjects was granted exempt status from review by the University of Washington institutional review board. Population and Sampling
We conducted a survey of current MSK radiology fellows. The survey instrument was created on SurveyMonkey (Professional) (5). MSK fellows are defined as individuals who have successfully completed a residency in diagnostic radiology and are currently engaged in a post-residency structured training program in MSK radiology. We identified 82 different MSK radiology fellowship programs through the Society of Skeletal Radiology web site: www.skeletalrad.org (6). The Society of Skeletal Radiology web site is believed to be the most complete and accurate listing of MSK radiology fellowships. In February 2016, 180 MSK radiology fellowship positions were listed as available for the academic year 2018. Each listing in-
Figure 1.
EMOTIONAL WELLNESS OF MSK RADIOLOGY FELLOWS
cluded contact information for the program director (an MSK radiologist), for the program coordinator (administrative person), or for both. We recruited subjects indirectly through the MSK program directors or other program contacts by e-mail, and requested that MSK fellows be made aware of our survey. Participation in the survey by MSK fellows was voluntary. An internet link to the survey was included in the message. Two requests were made to the MSK program directors or other contacts, separated by 1 week. The survey was closed 1 week after the second request. The responses were collected by SurveyMonkey and remained anonymous to the authors. No identifiable information about the respondents was solicited or obtained during the survey process, and there was no contact between the authors and the MSK fellows.
Survey
A 24-item survey was constructed on SurveyMonkey (Fig 1—sample item). The survey includes demographic questions, questions from the Maslach Burnout Inventory (MBI),
Survey question. Example of survey question as presented to respondents on surveymonkey.com.
683
PORRINO ET AL
questions about stress, and an open-ended comment question. The full survey is available as an Appendix. Demographics The demographic questions were placed at the start of the survey, and the responses were used to confirm the status of the respondent as an MSK fellow and to establish gender. Respondents who indicated that they were not currently fellows were excluded. Maslach Burnout Inventory In 1981, Maslach and Jackson defined burnout as a syndrome of emotional exhaustion and cynicism that occurs frequently among individuals who perform services and interact with other human beings. The key aspects of burnout are emotional exhaustion, the development of a negative and cynical attitude toward the client (depersonalization), and the tendency to evaluate oneself negatively (personal accomplishment) (1). Based on their initial burnout research, Maslach and Jackson developed the MBI designed to measure hypothesized aspects of the burnout syndrome (1). The inventory evaluates the frequency and intensity of three “subscales” of the burnout syndrome: emotional exhaustion (nine items), depersonalization (five items), and (the lack of a sense of) personal accomplishment (eight items) (1). This 22-item self-administered tool, later revised, asks respondents to rate the frequency with which they experience various feelings or emotions on a 7-point Likert scale, with response options ranging from “never” to “daily” (7,8). The MBI has been used in the past in an effort to assess burnout, and is a useful instrument in detecting the pattern of burnout among physicians. High levels of emotional exhaustion and depersonalization have come to be viewed as the most salient dimensions of burnout among medical professionals, a specific occupational group in which low levels of personal accomplishment may be less likely (2,8,9). The length of the MBI reflects a practical limitation of its routine use, prompting the use of condensed, single-item measures of emotional exhaustion and depersonalization adapted from the original tool as a more efficient means of assessing burnout in healthcare professionals (8,10). Emotional exhaustion has been assessed with a single item, “I feel burned out from my work,” whereas depersonalization was addressed with the item, “I have become more callous toward people since I took this job,” as these two particular measures have demonstrated the highest factor loading among their respective burnout domains across numerous studies (8,10). Utilization of single-item measures has proven to stratify the risk of burnout validly when compared to the use of the full MBI (8,10). In our survey, we used seven questions adapted from the MBI. As validated in earlier work, we used single items to address emotional exhaustion and depersonalization. We also incorporated five of the eight items from the MBI to assess personal accomplishment, as described by McNeeley et al. (2). 684
Academic Radiology, Vol 24, No 6, June 2017
Stressors Based on our review of the literature regarding the burnout syndrome in medicine, we included 13 questions measuring factors that may contribute to the development of burnout. Radiology as a career.—Substantive differences in burnout have been observed by specialty, with particularly high rates noted among emergency medicine, general internal medicine, neurology, and family medicine physicians (11). Specifically, two questions regarding satisfaction with radiology as an occupation were included. Financial stress.—High levels of perceived financial strain have been linked to worse overall mental health (12). Quality of life, satisfaction with work-life balance, emotional exhaustion, feelings of depersonalization, and burnout have all been demonstrated to be negatively affected by increasing levels of education debt among internal medicine residents (13). Conversely, quality of life and satisfaction with work-life balance were higher among internal medicine residents who participated in moonlighting (work at a second job outside of regular residency training duty hours) in the same study (13). Similar results have been found in radiology residents (2). Studies examining individuals with high debt typically employ a singleitem measure for assessing financial strain or debt burden (12). Selenko and Batinic developed a more comprehensive sixitem survey, on a 7-point scale, to assess the relationship between economic stress and individual well-being (12). In our study, we used three questions devised by Selenko and Batinic that address a general feeling of financial strain (12). Balancing work and life.—Physicians work longer hours and have greater difficulties with work-life balance when compared to other workers in the United States. These difficulties include the feeling that a commitment to their children has slowed their career advancement, as well as work-home conflicts. This imbalance, and resultant dissatisfaction, has been linked to physician burnout (11,14). To assess the effects of work-life imbalance and time needed for the appropriate care for dependents among radiology fellows, we utilized two items focused on the integration of work and personal and family life. Healthcare evolution and job market.—In a study conducted comparing US physicians to the general population, there was a 10% rise in the prevalence of physician burnout over a 3-year period, 2011–2014 (4). During this time frame, there were rapid changes in the practice and reimbursement of medicine, including implementation of the Affordable Care Act, reduced reimbursement rates, and the inadvertent effects on the job market. These issues continue. We therefore included four survey items to evaluate the possible association between the evolution of healthcare and job market constraints and physician burnout. Feelings of powerlessness and isolation.—Shanafelt et al. demonstrated that physicians who spent at least 20% of their time
Academic Radiology, Vol 24, No 6, June 2017
EMOTIONAL WELLNESS OF MSK RADIOLOGY FELLOWS
TABLE 1. Descriptive Statistics for Reduced MBI Subscales for MSK Radiology Fellows, and Comparison to Normative Values for Medical Occupations† Panel A: Descriptive Statistics for Reduced MBI Subscales and Independent Variables Variable Personal accomplishment subscale Emotional exhaustion subscale Depersonalization subscale Gender (unweighted) Weighted Stress measures Financial Work-life balance Dependent care Job market Healthcare regulations Isolation at work Powerlessness at work Satisfaction with MSK subfield
Obs.
Number of Items
Mean
Linearized Std. Error
58 58 58 58 58
5 1 1 1 1
17.73 2.79 2.22 .17 .21
.49 .22 .21 .38* .06
16.74–18.72 2.35–3.23 1.81–2.63
58 58 58 58 58 58 58 58
3 1 1 2 2 1 1 2
6.14 2.59 2.08 4.98 4.08 1.34 2.08 3.52
.27 .10 .12 .18 .18 .08 .11 .28
5.60–6.69 2.40–2.79 1.84–2.32 4.63–5.34 3.72–4.45 1.17–1.51 1.87–2.30 2.95–4.1
95% CI
.09–.33
Panel B: Comparison of Reduced MBI Subscale Means for MSK Radiology Fellows to Normative Subscale Means and High Burnout Thresholds for Medical Occupations
Variable Personal accomplishment subscale Emotional exhaustion subscale Depersonalization subscale
Number of Items in Full MBI
Number of Items in Reduced MBI items
8 9 5
5 1 1
Item Weight
Item-weighted Subscale Means (MSK Fellows)
Normative Subscale Means (Medicine)
Normative Subscale Thresholds for High Burnout (Medicine)
1.6 9 5
28.37 25.11 11.10
36.53 22.19 7.12
≤33 ≥27 ≥10
CI, confidence interval; MBI, Maslach Burnout Inventory; MSK, musculoskeletal. * Standard deviation, rather than linearized standard error, for unweighted gender variable. † Post-stratification gender weights applied, based on estimated female population proportion of .205.
in the aspect of work they found most meaningful had a burnout rate roughly half that of those physicians who spent less than 20% effort in the activity that was meaningful to them (15). The parameters evaluated included patient care, research, education, administration, and non-visit patient care activities (15). These data indicate that “career fit” is an important variable in determining physician burnout. To assess satisfaction with personal practice trends, our survey included two items that address the feeling of powerlessness regarding decision-making and isolation in the workplace (Appendix). Open-ended comments.—One open-ended question that solicited free-text comments concluded the survey. Statistical Analysis
Statistical analysis was performed using Stata (StataCorp, 2015, Stata Statistical Software: Release 14, College Station, TX) and Microsoft Excel (Microsoft, Redmond, WA).
RESULTS The mean, standard deviation, and minimum and maximum scores for the variables used in our analyses are presented in Table 1, Panel A.
Demographics
There were 58 respondents who identified themselves as current MSK radiology fellows and who completed the survey. A subsequent independent census of current MSK radiology fellowships with 100% response rate found that there were 190 physicians currently in these fellowships during the July 1, 2015, to June 30, 2016, academic year, 39 of whom were women (16). Of the 58 respondents in our study, 48 (82.8%) identified themselves as male, and 10 (17.2%) as female. Because the proportion of women in our sample is smaller than the actual proportion of women in MSK radiology fellowships, we have sampling bias. As earlier research finds that women differ from men in their susceptibility to burnout, as well as 685
PORRINO ET AL
Academic Radiology, Vol 24, No 6, June 2017
TABLE 2. Comparison of MSK Radiology Fellows Mean Weighted rMBI Scales, Disaggregated by Sex, to Maslach’s Normative MBI Subscale Thresholds and Sex-disaggregated MBI Subscale Means
Normative thresholds for high burnout (physicians and nurses) Male Maslach 2010 MSK radiology Female Maslach 2010 MSK radiology Adjusted Wald tests of subscale means (male-female comparisons) F (1, 57) Prob > F
Emotional Exhaustion
Depersonalization
Personal Accomplishment
≥27 19.86 24.37 20.99 27.90*
≥10 7.43 10.73* 7.02 12.5*
≤33 36.29 28.3* 36.50 28.64*
0.71 0.40
0.52 0.47
0.05 0.83
MBI, Maslach Burnout Inventory; MSK, musculoskeletal; rMBI, reduced Maslach Burnout Inventory. * Indicates value in “high burnout” range of Maslach et al. (2010) normative distributions.
exposure and responses to stressors, our analyses used poststratification weights based on the female proportion of 20.5% in the population of MSK radiology fellows (Chew FS, unpublished data, 2016). Burnout
As described, we utilized a modified, reduced MBI in which single items measured the emotional exhaustion and depersonalization subscales, whereas five items were used to assess personal accomplishment. In our data, the raw average scores for the single-item emotional exhaustion and depersonalization subscales were 2.2 and 2.8, respectively. The raw average aggregate score on the fiveitem personal accomplishment subscale was 17.73, with an item average of 3.5. Maslach et al reports normative ranges for low, medium, and high levels of burnout for various occupational groups (7). In comparing our subscales to those normative ranges, we used Maslach’s “medicine” occupational group (based on MBI scales for 1104 physicians and nurses, not otherwise described) as our reference cohort. Maslach et al states that for this occupational group, depersonalization subscale scores above 9.0, emotional exhaustion subscale scores above 26.0, and personal accomplishment subscale scores below 34.0 indicate high levels of burnout in the respective dimensions (7). To render the subscale scores in our data directly comparable to the normative data published in Maslach et al, we created item weights for the items used for our reduced MBI subscales (7). For each subscale, we applied an item weight using the ratio of the number of items in the full MBI subscale to the number of items in our modified, reduced MBI subscale (Table 1). Panel B of Table 1 includes the number of items in the full MBI and in the reduced MBI used in our research, the resulting weights, and our weighted subscale means. The final two columns in Panel B of Table 1 show Maslach’s normative subscale means and thresholds for “high burnout” for the occupation of “medical workers” (7). Maslach et al 686
used physicians and nurses for the “medical workers” occupation, but did not further identify them. Tables 1–3 show the distributions of item-weighted depersonalization, emotional exhaustion, and personal accomplishment subscale totals for MSK radiology fellows. In our data, the item-weighted mean depersonalization subscale score (Table 1, Panel B) is 11.1, above the 9-point threshold for high burnout. On average, then, MSK radiology fellows experience relatively high levels of “depersonalization” burnout, in comparison to the data on physicians and nurses that were used to form the normative ranges. In contrast, the itemweighted mean emotional exhaustion subscale score in our data (Table 1, Panel B) is 25.1. Although higher than the mean score of 22.19 in Maslach’s general population normative data, it remains below their high emotional exhaustion threshold of 27 for physicians and nurses (7). This indicates that MSK radiology fellows do not experience particularly high levels of emotional exhaustion, compared to other medical workers. Our respondents do, however, experience relatively high levels of burnout in association with the experience of lack of personal accomplishment (Table 1, Panel B), with an itemweighted average subscale score of 28.4, that is, well below Maslach’s threshold of 34 for physicians and nurses (7). In summary, comparison of the weighted subscale means in our data to the normative subscale thresholds for medical occupations indicates that MSK radiology fellows report relatively high levels of burnout with regard to lack of personal accomplishment and depersonalization, whereas emotional exhaustion levels in our sample, although relatively high, are still within the average range reported by Maslach et al (7) (Table 1, Panel B). The average levels of personal accomplishment and depersonalization burnout among the MSK radiology fellows in our sample are higher than those in the medical worker samples analyzed using the full MBI (7). Table 3 shows that the pattern of correlations among the (gender and itemweighted) subscales in our data is remarkably similar to the correlations among the full MBI subscales based on large samples, for a range of occupational groups, as reported by
Academic Radiology, Vol 24, No 6, June 2017
TABLE 3. Correlations Among Reduced MBI Subscales for MSK Fellows, Compared to Correlations Among Full MBI Subscales, Various Occupational Groups†
Depersonalization Personal accomplishment
Emotional Exhaustion
Depersonalization
.54* (.52) −.11 (−.22)
−.17 (−.26)
MBI, Maslach Burnout Inventory; MSK, musculoskeletal. * Significant at P < 0.000. † Numbers in parentheses are intercorrelations between full MBI subscales, reported in Maslach et al. (2010), appendix B. Poststratification gender weights applied to reduced MBI subscales, based on estimated female population proportion of .205.
Maslach et al. (7). The statistically significant 0.54 correlation coefficient for the single-item measures of depersonalization and emotional exhaustion mirrors the findings of Maslach et al for full MBI depersonalization and emotional exhaustion almost exactly. In our data, the negative and not statistically significant coefficients for the personal accomplishmentdepersonalization and personal accomplishment-emotional exhaustion correlations, although slightly smaller than those found in the validation studies by Maslach and collaborators, are otherwise entirely consistent with that research. Gender and Burnout
Maslach et al also reported demographic norms for the MBI subscales, although these are not disaggregated by occupational grouping (7). Table 2 shows the distributions of the depersonalization, emotional exhaustion, and personal accomplishment subscales by respondents’ gender, and compares the subscale means in our data, disaggregated by gender, to the normative demographic means in Maslach et al (7). Table 2 also includes the results of mean comparison tests. In both our data and in the comparison data, females report higher levels of emotional exhaustion; the differences between females and males are greater among MSK radiology fellows than in the Maslach et al data, although the difference is not statistically significant (7). It is noteworthy, however, that in our gender-disaggregated data, the average score for females is above the threshold for “high emotional exhaustion burnout,” but the average score for males remains below this threshold. Whereas Maslach et al found little difference between males and females with regard to depersonalization, female MSK radiology fellows report higher levels of depersonalization burnout than males. Again, these differences do not reach statistical significance in this small sample. In comparison to Maslach et al, we find no differences between males and females with regard to personal accomplishment scores (7). In summary, Table 2 shows that, although male MSK radiology fellows experience relatively high levels in two of the three dimensions of burnout (depersonalization and personal accomplishment), compared to the samples of medical
EMOTIONAL WELLNESS OF MSK RADIOLOGY FELLOWS
professionals analyzed in Maslach et al, female MSK radiology fellows experience comparatively higher burnout across all three dimensions (7). The gender differences in the levels of burnout are not, however, statistically significant in our relatively small sample. Prevalence of Burnout
Among MSK radiology fellows, 49 (84%) had weighted individual totals for the personal accomplishment burnout items that were at or below the threshold of 33 identified by Maslach et al for high personal accomplishment burnout among medical professionals (7). With regard to the depersonalization dimension of burnout, 39 (67%) of our respondents had weighted items scores at or above the threshold of 10 for high depersonalization burnout among medical workers. Even in the burnout dimension of emotional exhaustion—the only burnout dimension in which our sample means were within Maslach’s normal range for medical professionals—the weighted scores for over half of the MSK radiologists in our sample (33 respondents, or 57%) were at or above the normative threshold of 27 for high emotional exhaustion burnout (7). Overall, 21 respondents, or 36.2% of the sample, had weighted subscale scores that placed them in the high burnout range in all three dimensions of burnout measured in the study. An equal number of MSK radiology fellows (21, or 36.2%) exhibited high burnout in two of the three measured burnout dimensions, and another nine respondents (15.5%) reported a high level burnout in only one of the three dimensions. Only seven respondents (12.1%) did not meet the threshold for burnout in any of the dimensions. Thus, the overall prevalence of burnout among MSK radiology fellows was 88%. We can compare the prevalence of burnout among MSK radiology fellows to the prevalence of burnout among MSK radiology faculty members by looking at the results from an independently performed survey of practicing MSK radiologists (16). Radiologists in academic practice are likely to be the faculty members who are training the MSK radiology fellows, although not every academic radiology department has an MSK fellowship. It would seem appropriate to compare the prevalence of burnout in this subgroup of practicing MSK radiologists to that of MSK radiology fellows to gain some sense of whether the situation of the trainees is better or worse than that of the faculty. Burnout was present among many MSK faculty members, but was less prevalent and less severe than that of MSK fellows. The proportion of MSK faculty with no manifestations of burnout was approximately double that of MSK fellows (25% vs 12%), whereas the proportion of faculty with three manifestations was approximately half that of fellows (19% vs 36%) (Fig 2). Sources of Burnout
We examined the relationship between stressors and burnout. The survey included 13 questions that measured stress arising from financial concerns (three items), job market concerns 687
PORRINO ET AL
Figure 2. Burnout affecting MSK faculty and fellows. Column chart comparing the percentage of MSK faculty and fellows with the number of burnout manifestations. MSK, musculoskeletal).
(two items), regulatory changes in medicine including reference to the Affordable Care Act (two items), work-life balance concerns (one item), care for dependents (one item), workplace experiences of isolation (one item) and powerlessness (one item), and satisfaction with the MSK radiology subfield (two items). In the survey instrument, the stress items used a 5-point Likert scale, but in our analyses, we collapsed the extremes on both ends of the distribution to create three-level response categories, with 1 = “strongly disagree/disagree,” 2 = “neutral,” and 3 = “agree/strongly agree.” The average scores for these measures are reported below the subscale averages in Table 1. We also examined gender differences in reported stress levels. Compared to females, males report higher levels of financial stress (6.39 vs 5.2, P < 0.11), slightly lower levels of stress from job market concerns (4.87 vs 5.4, P < 0.23), significantly lower levels of stress associated with the regulatory changes of the Affordable Care Act (3.92 vs 4.7, P < 0.032), slightly higher levels of stress due to poor work-life balance (2.62 vs 2.5, P < 0.65), and substantially and significantly higher levels of stress related to care for dependents (2.21 vs 1.6, P < 0.037). Males did not differ from females with regard to stress associated with feelings of isolation at work (1.35 vs 1.3 for females, P < 0.81) or feelings of powerlessness at work (2.1 vs 2.0 for females, P < 0.71), whereas males reported significantly higher levels of dissatisfaction with the MSK radiology subfield than females (3.71 vs 2.8, P < 0.05). Although we found some significant differences between males and females with regard to sources of stress, the aggregate stress scores are quite similar, with males’ stress reports aggregating only slightly higher, to an average of 27.19, compared to females’ average aggregate stress score of 25.5 (P < 0.34). Regression Analyses
To understand the effects of specific stressors on burnout, we regressed our emotional exhaustion, depersonalization, and per688
Academic Radiology, Vol 24, No 6, June 2017
sonal accomplishment subscales on the stress measures described. The regression models also included a variable measuring the respondents’ satisfaction with the MSK radiology subfield, and a gender control. To facilitate interpretation of the effects of these stressors on various dimensions of burnout, we used reverse-coded emotional exhaustion and depersonalization subscales in the regressions. All of the stressor and subfield satisfaction variables were also reverse-coded, so that for the burnout subscales, the stressor measures, and the subfield satisfaction measures, higher values represent higher levels of burnout, stress, and dissatisfaction, respectively. The results of the regression analyses are shown in Table 4. We first note that, reading across the three models in Table 4, we find no effects of financial stressors, changes in healthcare regulations, or isolation at work on any of the three burnout subscales. These variables appear unrelated to burnout among MSK radiology fellows. Furthermore, stress related to concerns about the job market only significantly affects the personal accomplishment component of burnout (model 1), and is unrelated to the depersonalization and emotional exhaustion burnout subscales. Stress resulting from poor work-life balance, in contrast, does not significantly increase the personal accomplishment component of burnout, but is positively and significantly associated with higher levels of the other two burnout components—depersonalization and emotional exhaustion— with particularly strong effects on the latter component. Stress among MSK radiology fellows resulting from dependent care responsibilities, on the other hand, is significantly related to higher levels of the personal accomplishment component of burnout, and has no significant effects on the depersonalization component of burnout. Model 3 shows an unexpected, statistically significant negative effect of stress from care for dependents on the emotional exhaustion subscale. This indicates that those MSK radiology fellows who report higher levels of stress resulting from care for dependents report, on average, lower levels of emotional exhaustion. This apparently counterintuitive finding makes more sense when we consider the gender differences in levels of stress associated with care for dependents in our data, on the one hand, and emotional exhaustion, on the other. Males—who, even among dual-career couples, typically bear much less responsibility for caring for dependents—report significantly higher average levels of stress from this source (2.2 for males vs 1.6 for females, P < 0.05) (17–19). The modal response to our question about stress associated with care for dependents is “agree/strongly” agree for males and “disagree/strongly disagree” for females. The result, as shown in model 3, is that the demographic group (males) reporting higher levels of stress from caregiving responsibilities is also the demographic group reporting lower levels of emotional exhaustion. To clarify the relationship between gender, stress from care for dependents, and emotional exhaustion, we estimated a fourth regression model (not shown in Table 4) that included an interaction term to test for statistical interaction effects between gender and stress from care for dependents. When the inter-
Academic Radiology, Vol 24, No 6, June 2017
EMOTIONAL WELLNESS OF MSK RADIOLOGY FELLOWS
TABLE 4. Regression of Reduced MBI Personal Accomplishment, Depersonalization, and Emotional Exhaustion Subscales on Stressors and Subfield Satisfaction† Model 1
Independent Variables Gender Stress measures Financial stress Job market stress Healthcare regulations Work-life balance Care for dependents Isolation at work Powerlessness at work Satisfaction with MSK radiology subfield Constant Observations R-squared
Model 2
Model 3
Personal Accomplishment Subscale
Depersonalization Subscale
Emotional Exhaustion Subscale
0.149 (0.211)
0.576 (0.675)
0.486 (0.352)
−0.125 (0.176) 0.348** (0.135) −0.167 (0.173) 0.054 (0.105) 0.323*** (0.105) 0.063 (0.140) 0.171 (0.128) 0.094 (0.096) 0.133 (0.408) 58 0.279
0.017 (0.347) 0.104 (0.258) −0.051 (0.347) 0.386* (0.224) −0.180 (0.288) 0.403 (0.265) 0.473** (0.236) 0.469*** (0.137) −1.073 (0.938) 58 0.318
−0.062 (0.239) 0.140 (0.291) 0.031 (0.320) 0.690*** (0.204) −0.481** (0.204) 0.340 (0.236) 0.689*** (0.199) 0.627*** (0.126) −1.381* (0.822) 58 0.532
MBI, Maslach Burnout Inventory; MSK, musculoskeletal. *** P < 0.01, **P < 0.05, *P < 0.10. † Standard errors in parentheses; post-stratification gender weights applied, based on estimated female population proportion of .205.
action of gender with stress from care for dependents is included in the model, the main effect of high levels of stress from care for dependents represents the effect of this stressor for males (ie, when gender = “0”). This coefficient is negative and statistically significant (P < 0.05). The interaction coefficient indicates the effects of stress from care for dependents for women (ie, gender = “1”). The interaction coefficient of 1.17 is, in magnitude, nearly identical to the main effect of stress from care for dependents (ie, the effect for males, at −1.16), and also statistically significant (P < 0.05), but in the opposite direction, indicating, among female MSK fellows, a positive effect of stress from care for dependents on emotional exhaustion burnout. Stress resulting from feelings of powerlessness at work is not significantly related to the personal accomplishment component of burnout, but it is a significant and relatively powerful predictor of the burnout experiences of depersonalization and emotional exhaustion. Finally, we find a similar pattern of effects of MSK radiology fellows’ level of (dis)satisfaction with their medical subfield: as with stress resulting from feelings of powerlessness at work, dissatisfaction with the subfield does not
significantly affect the personal accomplishment component of burnout, but those MSK radiology fellows who report dissatisfaction with their subfield also tend to report higher levels of depersonalization and emotional exhaustion.
Free-text
Only six MSK fellows provided an open-ended response. In summary, these responses cited pressures or stress related to an apparent need, whether real or perceived, to read more studies, balancing work and private life demands, the cost of higher education/medical school, and a competitive job market.
DISCUSSION Why Does Burnout Matter?
According to Maslach et al, the burnout syndrome can lead to deterioration in the quality of care or service that is provided by staff, and can contribute to job turnover, absenteeism, and low morale (1). Additionally, burnout may lead to physical 689
PORRINO ET AL
exhaustion, insomnia, increased use of alcohol and drugs, and marital and family problems (1). Burnout among medical professionals has been linked to a lower degree of medical knowledge, the deterioration of professionalism, suboptimal patient care, medical errors, early retirement, and reduced empathy (8,11,13,15). Burnout also appears to be associated with personal consequences among physicians, including contributions to broken relationships, problematic alcohol use, and suicidal ideation (11). As a societal issue, burnout among physicians has received enough attention in the lay press that the trope of the burned-out physician in popular culture has become commonplace on television and on social media. The protagonist of the show “House,” as portrayed by the actor Hugh Laurie, is an example of the burnedout and marginally functional but brilliant physician whose manifestations of emotional exhaustion, depersonalization, and low personal achievement inform the plot lines of the popular show. Maslach’s initial research into the phenomenon of burnout included a variety of human service professions besides medicine, such as social services, mental health, criminal justice, and education, so burnout is not exclusive to physicians. Recent attention to burnout in doctors in the lay press does not appear to be unique to that profession; there have been recent New York Times articles on burnout in disparate occupations and situations, including athletes, coaches, athletic administrators, business executives, priests, professional dancers, unpaid caregivers for family members, teachers, employees of various types of businesses, and even military drone operators, a job that would not initially appear to fall under the human services professions rubric. Addressing this apparent rising tide of burnout that broadly affects our society is beyond the scope of our research, but it is important that we acknowledge the problem in our own profession and help those for whom we have responsibility. Burnout in MSK Radiology Fellows
Unfortunately, MSK radiology fellows report relatively high levels of burnout with respect to lack of personal accomplishment and depersonalization. Emotional exhaustion levels from our sample are within the average range reported for medical occupations (7). Why such high levels of burnout in the MSK radiology field? Is there something about the nature of MSK that promotes such a result? We think perhaps the opposite may be true. Compared to many other subspecialties of radiology and other fields of medicine in general, diagnosis and management of MSK patients seem in many respects as if it should be less stressful: fractures will heal, joints can be replaced, severe morbidity and death are uncommon, and bone cancer is rare. If this is the case, then one might expect burnout to be even more severe in other subspecialty fellows. A comparison between MSK fellows and other radiology subspecialty fellows is not possible at this time because our own sample was restricted to MSK fellows and we have found no published studies of burnout in other radiology subspecialty fellows. We suggest, however, that our results are generalizable to fellows in other radiology subspecialties. Al690
Academic Radiology, Vol 24, No 6, June 2017
though the domain-specific subspecialty knowledge base differs from one radiology subspecialty to another, the hierarchical niche and function of a subspecialty fellow are very similar, so the consequent stresses should also be similar. We believe that it would be useful to confirm this notion by studying other subspecialty fellows. Job market-related stress and the effort required in providing care for dependents significantly affect the personal accomplishment component of burnout in MSK radiology fellows. Conversely, imbalances in the work-life relationship and feelings of powerlessness are significantly associated with the depersonalization and emotional exhaustion components of burnout. Those MSK radiology fellows who were dissatisfied with their area of subspecialty tended to report higher levels of depersonalization and emotional exhaustion. How Do We Put Out the Fire?
Because burnout is defined as a work-related psychological syndrome, Nicola et al. report a series of strategies that may serve as remedies for burnout in the context of radiology (20). These authors advocate physical fitness, a minimization of external disruptions and distractions while immersed in work, and attention to “latent benefits” endorsed by the late social psychologist Marie Jahoda. Specifically, these latent benefits include structured time commitments, routine workplace activity, an enhanced social network, a sense of purpose, and augmented social status (20). The American College of Radiology’s Commission on Human Resources published a report in 2016 addressing burnout in radiologists, but there was no comment about radiology residents or subspecialty radiology fellows. They recommended several workplace modifications, including reducing workloads and call schedules, as well as generically improved lifestyle balance and counseling (21). The American Medical Association’s STEPS Forward program has a module on “preventing physician burnout” with suggestions for improving wellness at the group or system level (22). Implementation and evaluation of the strategies to reduce burnout are beyond the scope of our current study, but may be appropriate topics for future research. Limitations
The major limitation to our study is the nature of the MBI itself: it is essentially a self-administered, self-report of the psychological manifestations of burnout. The definition of burnout does not have any other criteria for determining its presence or absence other than responses to the MBI, and burnout is not included in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) as a psychiatric diagnosis. Although the response rate to our survey was 30.5% (58 of 190), the population we studied was relatively small. CONCLUSIONS Radiology fellows have a high prevalence of burnout (88%). To prevent significant adverse outcomes of burnout, more
Academic Radiology, Vol 24, No 6, June 2017
time and effort must be dedicated to the early recognition and intervention of the condition. To give optimal care to patients, a work environment that supports the well-being and mental health of its healthcare providers is essential.
REFERENCES 1. Maslach C, Jackson SE. The measurement of experienced burnout. J Occup Behav 1981; 2:99–113. Available at: http://www.jstor.org/stable/ 3000281. Accessed October 3, 2016. 2. McNeeley MF, Perez FA, Chew FS. The emotional wellness of radiology trainees: prevalence and predictors of burnout. Acad Radiol 2013; 20:647–655. doi:10.1016/j.acra.2012.12.018. 3. Patient Protection and Affordable Care Act (“ACA”), Public Law 110– 148. Available at: https://www.gpo.gov/fdsys/granule/PLAW111publ148/PLAW-111publ148/content-detail.html. Accessed October 3, 2016. 4. Shanafelt TD, Hasan O, Dyrbye LN, et al. Changes in burnout and satisfaction with work-life balance in physicians and the general US working population between 2011 and 2014. Mayo Clin Proc 2015; 90:1600– 1613. doi:10.1016/j.mayocp.2015.08.023. 5. Society of Skeletal Radiology. Available at: https://skeletalrad.org/mskfellowships. Accessed October 3, 2016. 6. SurveyMonkey Inc. (Professional version). Palo Alto, CA. Available at: https://www.surveymonkey.com. Accessed October 3, 2016. 7. Maslach C, Jackson SE, Leiter MP. Maslach Burnout Inventory manual and non-reproducible instrument and scoring guides. Mountain View, CA: CPP, Inc., 2010. 8. West CP, Dyrbye LN, Sloan JA, et al. Single item measures of emotional exhaustion and depersonalization are useful for assessing burnout in medical professionals. J Gen Intern Med 2009; 24:1318–1321. doi:10.1007/s11606-009-1129-z. 9. Rafferty JP, Lemkau JP, Purdy RR, et al. Validity of the Maslach Burnout Inventory for family practice physicians. J Clin Psychol 1986; 42:488– 492.
EMOTIONAL WELLNESS OF MSK RADIOLOGY FELLOWS
10. West CP, Dyrbye LN, Satele DV, et al. Concurrent validity of singleitem measures of emotional exhaustion and depersonalization in burnout assessment. J Gen Intern Med 2012; 27:1445–1452. doi:10.1007/s11606012-2015-7. 11. Shanafelt TD, Boone S, Tan L, et al. Burnout and satisfaction with worklife balance among US physicians relative to the general US population. Arch Intern Med 2012; 172:1377–1385. 12. Selenko E, Batinic B. Beyond debt. A moderator analysis of the relationship between perceived financial strain and mental health. Soc Sci Med 2011; 73:1725–1732. doi:10.1016/j.socscimed.2011.09.022. 13. West CP, Shanafelt TD, Kolars JC. Quality of life, burnout, educational debt, and medical knowledge among internal medicine residents. JAMA 2011; 306:952–960. doi:10.1001/jama.2011.1247. 14. Dyrbye LN, Shanafelt TD, Balch CM, et al. Relationship between workhome conflicts and burnout among American surgeons: a comparison by sex. Arch Surg 2011; 146:211–217. doi:10.1001/archsurg.2010.310. 15. Shanafelt TD, West CP, Sloan JA, et al. Career fit and burnout among academic faculty. Arch Intern Med 2009; 169:990–995. doi:10.1001/ archinternmed.2009.70. 16. Chew FS, Mulcahy MJ, Porrino JA, et al. Prevalence of burnout among practitioners of a radiology subspecialty. RSNA 2016 Annual Meeting Program Book, 2016. 17. Bianchi SM, Sayer LC, Milkie MA, et al. Housework: who did, does or will do it, and how much does it matter? Soc Forces 2012; 91:55–63. 18. Henz U. Parent care as unpaid family labor: how do spouses share? J Marriage Fam 2010; 72:148–164. 19. Shelton BA, John D. The division of household labor. Annu Rev Sociol 1996; 299–322. 20. Nicola R, McNeeley MF, Bhargava P. Burnout in radiology. Curr Probl Diagn Radiol 2015; 44:389–390. doi:10.1067/j.cpradiol.2015.04.007. 21. Harolds JA, Parikh JR, Bluth EI, et al. Burnout of radiologists: frequency, risk factors, and remedies: a report of the ACR Commission on Human Resources. J Am Coll Radiol 2016; 13:411–416. doi:10.1016/j.jacr.2015 .11.003. 22. American Medical Association. Preventing physician burnout. AMA, 2015. Available at: https://www.stepsforward.org/Static/images/modules/15/ downloadable/Preventing_Physician_Burnout.pdf. Accessed December 7, 2016.
691
PORRINO ET AL
APPENDIX. SURVEY QUESTIONS DEMOGRAPHIC ITEMS 1. Which best describes your current practice setting? o Academic practice o Private practice o Teleradiology practice 2. How long has it been since you completed your postgraduate residency and fellowship training? o Not yet completed o 5 years or less o 6–10 years o 11–20 years o 21 years or more 3. What is your gender? o Male o Female MASLACH BURNOUT INVENTORY ITEMS 4. How often does this statement describe your feelings about practicing radiology: “I can effectively solve the problems that arise in my work.” o Many times a day o Every day o Every week o Every month o A few times per year o Every year or less o Never 5. How often does this statement describe your feelings about practicing radiology: “I feel I am making an effective contribution to my radiology department or practice.” o Many times a day o Every day o Every week o Every month o A few times per year o Every year or less o Never 6. How often does this statement describe your feelings about practicing radiology: “In my opinion, I am good at my job.” o Many times a day o Every day o Every week o Every month o A few times per year o Every year or less o Never 7. How often does this statement describe your feelings about practicing radiology: “I have accomplished many worthwhile things in this job.” o Many times a day o Every day 692
Academic Radiology, Vol 24, No 6, June 2017
o Every week o Every month o A few times per year o Every year or less o Never 8. How often does this statement describe your feelings about practicing radiology: “I feel I am making an effective contribution to patient care.” o Many times a day o Every day o Every week o Every month o A few times per year o Every year or less o Never 9. How often does this statement describe your feelings about practicing radiology: “I feel burned out from my work.” o Many times a day o Every day o Every week o Every month o A few times per year o Every year or less o Never 10. How often does this statement describe your feelings about practicing radiology: “I have become more callous toward people since I began my current job.” o Many times a day o Every day o Every week o Every month o A few times per year o Every year or less o Never POTENTIAL SOURCES OF STRESS 11. My current financial situation is a serious strain. o Agree strongly o Agree o Neutral o Disagree o Disagree strongly 12. I often think about my current financial situation. o Agree strongly o Agree o Neutral o Disagree o Disagree strongly 13. I have difficulties meeting my financial obligations. o Agree strongly o Agree o Neutral o Disagree o Disagree strongly
Academic Radiology, Vol 24, No 6, June 2017
14. Decreased reimbursement rates for radiology have generated a serious strain on my life. o Agree strongly o Agree o Neutral o Disagree o Disagree strongly 15. Trying to balance the demands of my work with the time needed for my personal and family life is a major source of stress. o Agree strongly o Agree o Neutral o Disagree o Disagree strongly 16. Providing, organizing, or paying for care for dependents (e.g. children, elderly parents) is a major source of stress in my life. o Agree strongly o Agree o Neutral o Disagree o Disagree strongly 17. The current competition in the musculoskeletal radiology job market serves as a major source of stress in my life. o Agree strongly o Agree o Neutral o Disagree o Disagree strongly 18. The prospect of losing my job or not finding a suitable one, as a reflection of the competitive job market, serves as a major source of stress in my life. o Agree strongly o Agree o Neutral o Disagree o Disagree strongly 19. The lack of input into decisions that affect my medical practice is a major source of stress in my life. o Agree strongly
EMOTIONAL WELLNESS OF MSK RADIOLOGY FELLOWS
o Agree o Neutral o Disagree o Disagree strongly 20. The new regulations resulting from the Affordable Care Act (Obamacare) are a major source of stress in my life. o Agree strongly o Agree o Neutral o Disagree o Disagree strongly 21. Professional isolation (working alone) serves as a major source of stress in my life. o Agree strongly o Agree o Neutral o Disagree o Disagree strongly CAREER CHOICE 22. How often does this statement describe your feelings about practicing radiology: “I wish I were in a different specialty of medicine.” o Many times a day o Every day o Every week o Every month o A few times per year o Every year or less o Never 23. I wish I worked in a different field instead of diagnostic radiology. o Agree strongly o Agree o Neutral o Disagree o Disagree strongly OPEN-ENDED COMMENT 24. Do you have any other comments, questions, or concerns?
693