13
ORIGINAL ARTICLE MAYO CLINIC COJ Longitudinal Study Evaluating the Association Between Physician Burnout and Changes in Professional Work Effort ®crOSSMark Tait D. Shanafelt, MD; Michelle Mungo, MBA; Jaime Schmitgen, MPA; Kristin A. Storz, MA; David Reeves, PhD; Sharonne N. Hayes, MD; Jeff A. Sloan, PhD; Stephen J. Swensen, MD; and Steven J. Buskirk, MD Abstract Objective: To longitudinally evaluate the relationship between burnout and professional satisfaction with changes in physicians' professional effort. Participants and Methods: Administrative/payroll records were used to longitudinally evaluate the profes- sional work effort of faculty physicians working for Mayo Clinic from October 1, 2008, to October 1, 2014. Professional effort was measured in full-time equivalent (FTE) units. Physicians were longitudinally surveyed in October 2011 and October 2013 with standardized tools to assess burnout and satisfaction. Results: Between 2008 and 2014, the proportion of physicians working less than full-time at our organization increased from 13.5% to 16.0% (P=.05). Of the 2663 physicians surveyed in 2011 and 2776 physicians surveyed in 2013, 1856 (69.7%) and 2132 (76.9%), respectively, returned surveys. Burnout and satisfaction scores in 2011 correlated with actual reductions in FTE over the following 24 months as independently measured by administrative/payroll records. After controlling for age, sex, site, and specialty, each I-point increase in the 7-point emotional exhaustion scale was associated with a greater likelihood of reducing HE (odds ratio [OR], 1.43; 95% CI, 1.23-1.67; P<.OOl) over the following 24 months, and each I-potnr decrease in the 5-point satisfaction score was associated with greater likelihood of reducing HE (OR, 1.34; 95% CI, 1.03-1.74; P=.03). On longitudinal analysis at the individual physician level, each l-point increase in emotional exhaustion (OR, 1.28; 95% CI, 1.05-1.55; P=.OI) or l-point decrease in satisfaction (OR, 1.67; 95% CI, 1.19-2.35; P=.003) between 2011 and 2013 was associated with a greater likelihood of reducing FTE over the following 12 months. Conclusion: Among physicians in a large health care organization, burnout and declining satisfaction were strongly associated with actual reductions in professional work effort over the following 24 months. © 2016 Mayo Foundation for Medical Education and Research ,. Mayo Clin Proc. 2016:91(4}:422-431 he United States faces a large projected physician workforce shortage over the next decade, with several agencies, including the US Department of Health and Hu- man Services, projecting a shortfall of 45,000 to 90,000 physicians by 2025 1 ,2 The magnitude of this problem varies by specialty. Although some question whether there will be a shortage in the absolute number of physicians or simply a rela- tive shortage due to geographic and specialty distribution.' all agree there will be too few pri- h .. d I 134 mary care p YS1Clans an genera surgeons. ' , A variety of factors contribute to the physician workforce shortage, including population growth, an aging US population, more For editorial comment. see page 408 From the Department of In- temal Medicine (T.D.5., S.N.H.), Human Resources (l.S), Department of Health Sciences Research UAS.), Department of Radiology (S.j.S.),and Office of Organi- zation and Leadership Devel- opment (S.j.S.), Mayo Clinic, Rochester, MN; Sirota Survey Affiliations continued at the end o( this article. widespread access to medical care because of the Affordable Care Act, a rapid increase in the complexity of medical care and need for certain specialists, an inadequate supply of physicians in medical school training, limits on the number of residency training positions, and mismatch between the residency positions available and the type of specialists needed. 1,2 Physician burnout is another factor that may influence the adequacy of the physician work- force. National studies suggest that approxi- mately half of US physicians are experiencing burnout and that this proportion continues to increase. 5 ,6 Cross-sectional studies evaluating self-reported career plans suggest that burned 422 Mayo Clin Proc. ,. April2016:91(4h422-431 ,. http'//dx.doi.org/l0.1016/j.mayocp.2016.02.001 www.mayoclinicproceedings.org ,. © 2016 Mayo Foundation for Medical Education and Research

ORIGINAL ARTICLE MAYO COJ - mfprac.com

  • Upload
    others

  • View
    4

  • Download
    0

Embed Size (px)

Citation preview

Professional Work Effort
®crOSSMark
Tait D. Shanafelt, MD; Michelle Mungo, MBA; Jaime Schmitgen, MPA; Kristin A. Storz, MA; David Reeves, PhD; Sharonne N. Hayes, MD;
Jeff A. Sloan, PhD; Stephen J. Swensen, MD; and Steven J. Buskirk, MD
Abstract
Objective: To longitudinally evaluate the relationship between burnout and professional satisfaction with changes in physicians' professional effort. Participants and Methods: Administrative/payroll records were used to longitudinally evaluate the profes- sional work effort of faculty physicians working for Mayo Clinic from October 1, 2008, to October 1, 2014. Professional effort was measured in full-time equivalent (FTE) units. Physicians were longitudinally surveyed in October 2011 and October 2013 with standardized tools to assess burnout and satisfaction. Results: Between 2008 and 2014, the proportion of physicians working less than full-time at our organization increased from 13.5% to 16.0% (P=.05). Of the 2663 physicians surveyed in 2011 and 2776 physicians surveyed in 2013, 1856 (69.7%) and 2132 (76.9%), respectively, returned surveys. Burnout and satisfaction scores in 2011 correlated with actual reductions in FTE over the following 24 months as independently measured by administrative/payroll records. After controlling for age, sex, site, and specialty, each I-point increase in the 7-point emotional exhaustion scale was associated with a greater likelihood of reducing HE (odds ratio [OR], 1.43; 95% CI, 1.23-1.67; P<.OOl) over the following 24 months, and each I-potnr decrease in the 5-point satisfaction score was associated with greater likelihood of reducing HE (OR, 1.34; 95% CI, 1.03-1.74; P=.03). On longitudinal analysis at the individual physician level, each l-point increase in emotional exhaustion (OR, 1.28; 95% CI, 1.05-1.55; P=.OI) or l-point decrease in satisfaction (OR, 1.67; 95% CI, 1.19-2.35; P=.003) between 2011 and 2013 was associated with a greater likelihood of reducing FTE over the following 12 months. Conclusion: Among physicians in a large health care organization, burnout and declining satisfaction were strongly associated with actual reductions in professional work effort over the following 24 months.
© 2016 Mayo Foundation for Medical Education and Research ,. Mayo Clin Proc. 2016:91(4}:422-431
he United States faces a large projected physician workforce shortage over the next decade, with several agencies,
including the US Department of Health and Hu- man Services, projecting a shortfall of 45,000 to 90,000 physicians by 20251
,2 The magnitude of this problem varies by specialty. Although some question whether there will be a shortage in the absolute number of physicians or simply a rela- tive shortage due to geographic and specialty distribution.' all agree there will be too few pri-
h .. d I 134mary care p YS1Clansan genera surgeons. ' , A variety of factors contribute to the physician workforce shortage, including population growth, an aging US population, more
For editorial comment. see page 408
From the Department of In- temal Medicine (T.D.5., S.N.H.), Human Resources (l.S), Department of Health Sciences Research UAS.), Department of Radiology (S.j.S.),and Office of Organi- zation and Leadership Devel- opment (S.j.S.), Mayo Clinic, Rochester, MN; Sirota Survey
Affiliations continued at the end o( this article.
widespread access to medical care because of the Affordable Care Act, a rapid increase in the complexity of medical care and need for certain specialists, an inadequate supply of physicians in medical school training, limits on the number of residency training positions, and mismatch between the residency positions available and the type of specialists needed. 1,2
Physician burnout is another factor that may influence the adequacy of the physician work- force. National studies suggest that approxi- mately half of US physicians are experiencing burnout and that this proportion continues to increase.5,6 Cross-sectional studies evaluating self-reported career plans suggest that burned
422 Mayo Clin Proc. ,. April2016:91(4h422-431 ,. http'//dx.doi.org/l0.1016/j.mayocp.2016.02.001 www.mayoclinicproceedings.org ,. © 2016 Mayo Foundation for Medical Education and Research
PHYSICIAN BURNOUT AND REDUCTIONS IN PROFESSIONAL WORK EFFORT
out physicians are more likely to be planning to reduce the amount of time they devote to clin- ical work over the next 12 to 24 monrhs.r!' Other studies have found a link between decreased physician satisfaction and intent to reduce work hours or leave clinical care.ll-13
Conceptual models of the relationship between physician stress and intent to quit or reduce work hours have been proposed, with cross- sectional data providing validation of this framework. 9
Despite these associations, it is unknown how well self-reported intentions to reduce work hours relate to actual future behavior. If burned out physicians are truly more likely to reduce clinical work hours, it could pose a large threat to the adequacy of the US physi- cian workforce as well as a large financial cost to society.!" The potential impact of burnout on the physician workforce may be compounded by the fact that many of the spe- cialties facing the largest projected shortage of physicians are the disciplines with the highest rates of burnout5 To explore the relationship between burnout and actual changes in work hours, we conducted a prospective, longitudi- nal study of physicians working at a large US health care organization.
PARTICIPANts AND METHODS
Demographic Characteristics and Professional Effort Mayo Clinic is a nonprofit, physician-led health care organization with 3 large academic campuses (Rochester, Minnesota; Scottsdale, Arizona; and Jacksonville, Florida) as well as an integrated group of community-based hospi- tals and health care facilities. The professional effort of physicians at Mayo Clinic is measured as a fraction of full-time equivalent (FTE) units. For example, an FTE of 1.0 indicates a physician working full-time, and an FTE of 0.5 indicates a physician working half-time. Administrative! payroll records were used to describe the FTE level of all Mayo Clinic physicians at the Rochester, Scottsdale, and Jacksonville cam- puses who were members of the permanent faculty. Available demographic information included age, sex, and specialty area. Physi- cians were categorized into 1 of 8 specialty areas: primary care (general internal medi- cine, family medicine, general pediatrics),
internal medicine subspecialty, surgical disci- pline, radiology, anesthesiology, pathology/ laboratory medicine, other medical specialty area (eg, dermatology, neurology, physical medicine/rehabilitation, psychiatry, radiation oncology), or other.
Survey Participants As part of its efforts to foster the health of the or- ganization, Mayo Clinic surveys its physicians, scientists, and allied health staff approximately every 24 months. This all-staff survey is admin- istered by an independent consulting organiza- tion (Sirota Survey Intelligence) and covers a broad array of topics including employees' perception of quality and safety, measures of professional burnout, and satisfaction with the organization.
The 2 most recent surveys for which data are available were administered in October 2011 and October 2013. Participation in these surveys was voluntary and all data were confidential. Although the external survey consulting firm tracks responses by employee identification number, identifying information is not available to any Mayo Clinic employee or the Mayo- employed researchers. To facilitate analysis of the relationship between responses to survey questions and changes in FTE, information on the FTE level of permanent faculty physicians at each survey time point was determined using administrative!payroll records and sent from Mayo Clinic to the external survey consulting firm so they could pair this information with sur- vey responses. The use of data collected from the survey for the research analysis reported in this article was reviewed by the Mayo Clinic Institu- tional Review Board and determined to be exempt.
Burnout and Satisfaction Burnout is a syndrome characterized by emotional exhaustion (losing your enthusiasm for work) and depersonalization (viewing/treat- ing people as if they were objects) that results in decreased effectiveness at work.15 Although the Maslach Burnout Inventory (MBI)15 is the criterion standard for assessing burnout, its length (22 items) limits feasibility for use in an organization-wide survey covering a wide range of topics such as the one reported here. Thus, to evaluate the emotional exhaustion and deper- sonalization domains of burnout in physicians,
Mayo Clin Proc. - April 2016;91(4),422-431 - hltp:!/dx.doi.orgI10.10161j.mayocp.2016.02.001 www.mayoclinicproceedings.org
TABLE 1. Proportion of Physicians at Mayo Clinic Working Less Than Full-time. 2008-~014a.b
Working less than full-time 2008 (N=2518) 2009 (N=2555) 20 I0 (N=2569) 2011 (N=2641) 2012 (N=2759) 2013 (N=2836) 2014 (N=29 IS) P value for trend
All physicians 339 (13.5) 362 (14.2) 362 (14.1) 392 (14.8) 427 (155) 455 (16.0) 466 (16.0) .045
Menc 115 (6.0) 131 (6.8) 128 (6.6) 143 (7.3) 167 (8.3) 184 (9.0) 191 (9.1) <.001
:::;35 Y 3 (2.6) 4 (4.0) 2 (2.3) 2 (2.2) 3 (3.1) I (1.2) o (0.0) .61 36-45 y 20 (3.7) 18 (35) 14 (2.8) IS (3.0) 15 (3.0) 15 (2.8) 16 (3.0) .98 46-55 Y 29 (3.8) 34 (4.6) 31 (4.4) 35 (5.0) 36 (5.2) 32 (4.8) 33 (5.2) .88 >55 y 63 (12.6) 75 (13.2) 79 (12.8) 91 (13.6) 113 (15.7) 134 (17.6) 142 (17.7) .QI
Womend 224 (375) 231 (37.0) 234 (36.7) 249 (36.9) 260 (35.3) 271 (34.6) 275 (33.4) .62 :::;35 Y 21 (29.2) 19 (25.3) 15 (22.7) 15 (19.2) 15 (16.9) 10.(10.9) II (11.8) .02 36-45 Y 102 (395) 99 (39.3) 92 (37.4) 92 (35.8) 103 (35.8) 103 (33.6) 95 (30.4) .26 46-55 Y 73 (38.2) 78 (37.3) 82 (385) 92 (41.3) 89 (38.9) 96(41.2) 105 (41.2) .96 >55 y 25 (36.8) 35 (40.7) 43 (41.3) 48 (43.2) 53 (41.7) 60 (42.6) .60 (39.2) .98
Specialty Primary care" 49 (25.8) 49 (28.7) 50 (30.1) 52 (29.7) 53 (30.3) 51 (30.0) 52 (30.4) .96 Other medical specialty' 92 (17.7) 101 (18.4) 101 (185) 101 (17.8) 106 (17.7) 112 (17.8) 115 (17.6) >.99 Anesthesiology 23 (15.6) 22 (145) 22 (14.4) 24 (15.0) 23 (13.9) 28 (165) 28 (15.8) >.99 Intemal medicine subspecialty 125 (15.5) 136 (16.7) 140 (17.0) 156 (18.3) 167 (18.9) 175 (19.4) 184 (20.0) .16 Radiology 13 (11.1) II (9.5) 9 (7.7) 9 (7.6) 19 (14.5) 27 (20.0) 27 (19.7) .004 Pathology and laboratory medicine 12 (8.5) 13 (8.8) 15 (9.8) 20 (12.7) 19 (11.2) 19 (11.3) 17 (9.7) .89 Other 9 (4.5) 12 (5.6). 9 (4.2) 7 (34) II (5.0) 10 (4.2) 13 (55) .93 Surgical specialty 16 (4.1) 18 (4.6) 16 (4.0) 23 (5.6) 29 (6.9) 33 (7.8) 29 (6.6) .12....•.. -- ---
'Data are presented as No. (percentage) of participants.
"Based on the full-time equivalent of each physician on October I of each year for 2008 to 2014 according to Mayo Clinic Human Resources employment records. 'Values indicate the % of men physicians with an FrE < 1.0 at the indicated time-point See Supplemental Table I (available online at http://www.mayoclinicproceedings.org) for the total number of male physicians
for each year. dValues indicate the % of women physicians with an FrE < 1.0 at the indicated time-point See Supplemental Table I for the total number of male physicians for each year.
eFamily medicine, general pediatrics. general intemal medicine. 'Neurology. demnatology. physical medicine/rehab. radiation oncology. subspecialty pediatrics, psychiatry, etc.
'.
40 g 35 ~ V 30
~ 25 :§ 20 s ~ 15 .~ 10
1 5 Q.
O~--,---.---,---,---,---.---. • • -+---"--'- .
2008 2009 20 I0 20 I I 20 12 20 I3 20 14 Year
A
•....$35 Y (P=.61) -- 46-55 Y (P=.88) ___ 36-45 Y (P=.98) >55 Y (P=.O I)
20 C- 18 ~ 16v 14LU
ti: 12 -= 10 'j:
8Ul c:: 6ra 'u
0 2008 2009 20 I 0 2011 2012 2013 2014
Year
B
__ 46-55 Y (P=.96)
>55 y (P=.98) 50-L----------------------~
g 45 ~ 40 ~ 35 ti: 30 :§ 25 ~ 20 ~ 15 :~ 10 >-'&. 5
04---,---,--.---,--,---,--, 2008 2009 2010 2011 2012 2013 2014
Year
C
FIGURE 1. Changes in physicians working less than full-time between 2008 and 2014. A, Changes in full-time equivalent (FTE) in men and women. B, Changes in FTE among men stratified by age group. C, Changes in FTE among women stratified by age group.
we used 2 single-item measures adapted from the full MBI. These 2 items have been used in previ- ous studies involving more than 30,000 physi- cians16
- 18 and have been reported to have a
high correlation with burnout as measured by the full MBI in samples of over 10,000 physi- cians.19
,20 The area under the receiver operating characteristic curve for emotional exhaustion for the single emotional exhaustion item relative to the full MBl is 0.94,19 and the area under the receiver operating characteristic curve for the single-item depersonalization measure is 0.9319
The concurrent validity of this approach for assessing burnout has also been established. 20
These 2 items remain the property of Mind Gar- den Inc (which holds the copyright on the MBI) and were used with appropriate license .
Overall satisfaction with the organization was also evaluated by asking, "Considering everything, how would you rate your overall satisfaction with Mayo Clinic as a whole at the present time?" Physicians responded to this question using a 5-point Likert scale (5 = very satisfied, 4 = satisfied, 3 = neither satisfied nor dissatisfied, 2 = dissatisfied, and 1 = very dissatisfied).
Statistical Analyses Continuous variables were summarized using mean ± SD, and categorical variables were summarized using frequency. Multivariate logistic regression analysis was used to eval- uate the relationship between changes in FTE and both burnout and satisfaction after adjusting for age, sex, and specialty. Changes in emotional exhaustion score between 2011 and 2013 were also calculated for individuals who provided data at both time points. The relationship between changes in emotional exhaustion and changes in FTE were subse- quently evaluated. Two-tailed P<.05 was considered statistically Significant. All analyses were performed using IBM SPSS Statistics version 20 (IBM Corporation).
RESULTS
Changes in Professional Effort of Physicians Over Time Complete data on the percentage of physicians working less than full-time were available for all physicians working at Mayo Clinic from October 1,2008, to October 1, 2014. The proportion of
Mayo Clin Proc .• April 2016:91 (4),422-431 • http,lldx.doi.orgI10.1016/j.mayocp.2016.02.001 www.mayoclinicproceedings.org
2011 2013
Variable (n= I856) (n=807) (n=2132) (n=644)
Sex Men 1359 (73.2) 61 I (75.7) 1527 (71.6) 477 (74.1) Women 497 (26.8) 196 (24.3) 602 (28.2) 167 (25.9) Missing 0(0) 0(0) 3 (0.1) 0
Age <25 y 0(0) 0(0) 0(0) 0(0) 25-34 Y 106 (5.7) 49 (6.1) I 15 (5.4) 32 (5.0) 35-44 Y 506 (27.3) 241 (29.8) 624 (29.3) 169 (26.2) 45-54 Y 654 (35.2) 258 (32.0) 676 (31.7) 216 (33.5) 55-64 Y 504 (27.2) 211 (26.2) 590 (27.7) 176 (273) 2:65 y 86 (4.6) 48 (5.9) 124 (5.8) 51 (7.9) Missing 0(0) 0(0) 3 (0.1) 0(0)
FTE <1.0 269 (14.5) 128 (15.9) 334 (15.7) 94 (14.6) Mean 0.97 0.96 0.96 0.97
aFTE = full-time equivalent bData are presented as No. (percentage) of participants. Percentages may not total 100 because of rounding.
'The number of physicianssurveyed in 20 II and 2013 (2663 and 2776. respectively) differsslightly from the number of physicians in the 'FTE analysis in 20 I I and 20 I3 (Table I) because these values came from independent data sources (ie, data for Table I from Mayo Clinic Human Resources employment records; the physicians in the survey sample for Table 2). The roster for physicians to be included in the 20 I I and 2013 staff surveys were created the first week of September in 20 I I and the first week of August in 2013, whereas the data in Table I are based on the FTE of each physician on October I of each year for 2008 to 2014.
physicians working less than full-time on October 1 of each year according to Mayo Clinic Human Resources employment records is pro- vided in Table J. (see Supplemental Tables 1 and 2 for additional details, available online at http://www.mayoclinicproceedings.org). The proportion of physicians working less than full-time increased from 13.5% (339 of 2518) to 16.0% (466 of 2915; P=.05) over the 7-year interval from 2008 to 2014.
A strong relationship between sex and changes in FTE was observed. Women physi- cians were more likely to work less than full-time in all years evaluated (Figure 1, A), The proportion of male physicians working less than full-time increased from 6.0% (1l5 of 1921) in 2008 to 9.1 % (191 of 2092) in 2014 (P<. 001). Among men, a statistically significant increase in the proportion of physicians working less than full-time was observed among men older than 55 years (from 12.6% [63 of 499] to 17.7% [142 of 802]; P=Ol) between 2008 and 2014 (Figure 1, B). No statistically
significant change was observed among men in other age categories.
Although not statistically Significant, the proportion of women physicians working less than full-time decreased from 37.5% (224 of 597) in 2008 to 33.4% (275 of 823) in 2014. Among women, the proportion of physicians working less than full-time decreased for those aged 35 years or younger (from 29.2% [21 of 72] to 11.8% [ll of 931; P=.02) between 2008 and 2014 (Figure 1, C). A decrease in the proportion of women physicians working less than full-time was also observed among those aged 36 to 45 years (from 39.5% [102 of258] to 30.4% [95 of312]; P=.26); how- ever, this difference was not statistically signifi- cant. No statistically Significant changes were observed among women aged 46 to 55 or older than 55 years.
Specialty-specific differences in the per- centage of physicians working less than full-time were also observed (Table 1). Pri- mary care physicians were more likely to work less than full-time in all years evalu- ated. The only specialty area with a statis- tically Significant change' in the proportion of physicians working less than full-time between 2008 and 2014 was radiology (from 1(1% [13 of 117] to 19.7% [27 of 1371; P=.004).
Relationship Between Professional Burnout and Satisfaction and Changes in Professional Effort Of the 2663 and 2776 physicians employed at the time of the 2011 and 2013 staff surveys (rosters for physicians to be included in the 20ll and 2013 staff surveys were created the first week of September in 20ll and the first week of August in 2013), 1856 (69.7%) and 2132 (76.9%), respectively, returned surveys. Demographic characteristics of responding and nonresponding physicians were similar at both survey time points with respect to age and sex (Table 2).
We evaluated whether physicians with higher emotional exhaustion scores at the time of the 2011 survey were more likely to reduce their FTE over the following 24 months (ie, before the 2013 survey). After controlling for the effects of age, sex, site (Arizona, Florida, Minnesota), and specialty, each I-point increase in the 7-point emotional
426 Mayo Clin Proc. _ April 2016:91(4),422-431 _. httpj/dx.doi.org/10.10161j.mayocp.2016.02.001 www.mayoclinicproceedings.org
---------- -- --
PHYSICIAN BURNOUT AND REDUCTIONS IN PROFESSIONAL WORK EFFORT
exhaustion scale was associated with a greater likelihood of reducing FTE over the 24 months following the 2011 survey (OR, 1.43; 95% CI, 1.23-1.67; P<.OOl). Individuals with lower satisfaction at the time of the 2011 survey were also more likely to reduce their FTE over the following 24 months. After adjusting for the effects of age, sex, site, and specialty, each l-point decrease in the 5-point satisfac- tion scale was associated with a greater likeli- hood of reducing FTE over the 24 .months following the 2011 survey (OR, 1.34; 95% CI, 1.03-1.74; P=.03). The results of X2 tests regarding model fit indicated acceptable fit of the model to the data. No association between depersonalization score in 2011 and changes in FTE over the following 24 months was observed.
We next evaluated how changes in emotional, exhaustion at the individual physician level be- tween 2011 and 2013 related to the likelihood of reducing FTE in the 12 months following the 2013 survey. Of the 1856 physicians who completed surveys in 2011, 1483 (79.9%) also completed a survey in 2013. Changes in the emotional exhaustion level of physicians who completed surveys at both time points are pro- vided in Table 3. After controlling for age, sex, site, and specialty area, each 1-point increase in a physician's emotional exhaustion score be- tween 2011 and 2013 was associated with a higher likelihood of reducing FTE in the 12 months following the 2013 survey (OR, l.28; 95% CI, 1.05-1.55; P=.Ol). Changes in satisfaction score were also related to changes in FTE. After controlling for age, sex, site, and specialty area, each I-point decrease in a physician's satisfaction score be- tween 2011 and 2013 was associated with a higher likelihood of reducing FTE in the 12 months following the 2013 survey (OR, 1.67; 95% CI, l.19-2.35; P=.003). The re- sults of X2 tests regarding model fit indicated acceptable fit of the model to the data. No as- sociation between changes in depersonaliza- tion score and the likelihood of reducing FTE were observed on similar analysis.
DISCUSSION In this prospective study of physicians, burnout and satisfaction were related to future reductions in professional work hours as assessed by administrative/payroll records.
TABLE 3. Changes in Burnout Among Physicians Who Completed Surveys in Both 2011 and 2013"
Variable All physicians Men Women
Changes in overall satisfaction' ~1463b n=I072 n=39 I ~2-point improvement 33 (2.3) 26 (2.4) 7 (1.8) I-point improvement 228 (15.6) 175 (16.3) 53 (13.6)
No change 882 (60.3) 654 (61.0) 228 (58.3)
I-point worsening 262 (17.9) 180 (16.8) 82 (21.0)
~2-point worsening 58 (4.0) 37 (3.5) 21 (5.4)
Bumout Changes in emotional exhaustion scored N= I4552 n=1065 n=390
~3-point improvement 25 (1.7) 18 (1.7) 7 (1.8) 2-point improvement 63 (4.3) 47 (4.4) 16 (4.1) I-point improvement 158 (10.9) 108 (10.1) 50 (12.8) No change 473 (32.5) 367 (34.5) 106 (27.2) I-point worsening 340 (23.4) 250 (23.5) .90 (23.1) 2-point worsening 248 (170) 173 (16.2) 75 (19.2) ~3-point worsening 148 (10.2) 102 (9.6) 46 (I 1.8)
Changes in depersonalization score" N=1438b n=1051 n=387 ~3-point improvement 32 (2) 21 (2) I I (3) 2-point improvement 54 (4) 40 (4) 14 (4) I-point improvement 176 (12) 126 (12) 50 (13) No change 658 (48) 513 (49) 172 (44) I-point worsening 256 (18) 184 (18) 72 (19) 2-point worsening III (8) 86 (8) 25 (6) ~3-point worsening 124 (9) 81 (8) 43 (II)-
'Data are presented as No. (percentage) of participants. Percentages may not total 100 because of
rounding. bA total of 1483 physicians completed surveys at both the 20 I I and 20 I3 survey time points, of
whom 1463 provided data for overall satisfaction at both time points, 1455 provided data for.
emotional exhaustion at both time points, and 1438 provided data for depersonalization at both
time points. 'Change in satisfaction between 20 I I and 20 I3 surveys; 5-point scale. dChange in emotional exhaustion score between 20 I I and 2013 surveys; 7-point Likert scale.
"Change in depersonalization score between 20 I I and 2013 surveys; 7-point Likert scale.
Specifically, physicians with greater degrees of emotional exhaustion in 2011 were more likely to reduce their FTE over the following 24 months. Physicians with lower degrees of satisfaction were also more likely to reduce their FTE over the following 24 months. The relationship between emotional exhaustion and satisfaction with reductions in FTE exhibited a "dose-response" relationship. Inde- pendent of age, sex, site, and specialty, each 1- point change in emotional exhaustion (7-point scale) or satisfaction (5-point scale) was associ- ated with a 43% and 34% higher likelihood of reducing FTE over the ensuing 24 months, respectively. We were also able to longitudi- nally evaluate the effects of changes in burnout and satisfaction at the level of individual phy- sicians over a 2-year interval and prospectively assess the subsequent association with changes
Mayo Clin Proc. - April 2016;91(4),422-431 - httpd/dx.doi.org/10.1016/j.mayocp.2016.02.001 www.mayoclinicproceedinqs.orq
MAYO CLINIC PROCEEDINGS
in professional effort. Physicians with wors- ening burnout or declining satisfaction be- tween 2011 and 2013 were more likely to reduce their FTE between 2013 and 2014 (ie, in the 12 months following the 2013 assessment). A dose-response relationship was once again observed between changes in emotional exhaustion and satisfaction and actual reductions in FTE. No relationship between the depersonalization domain of burnout and reductions in FTE were observed in either of these analyses, suggesting speci- ficity to these findings.
Independent of burnout, we observed an increase in the proportion of physicians work- ing less than full-time over the 7 years of our study. Notably, this trend was primarily due to an increased proportion of men, particularly ,men over age 55 years, reducing their FTE. Although women were more likely to work less than full-time than men throughout all years of the study, the proportion of women working less than full-time did not increase during the 7-year study interval, and the pro- portion of women aged 35 years or younger working less than full-time actually decreased. These findings suggest that the trends in FTE we observed were not attributable to millen- nials oriented toward better work-life integra- tion but rather reductions in work hours by their older colleagues. This observation is consistent with findings from other studies that indicate the increase in physicians work- ing less than full-time is not limited to women or younger physicians21,22 Although the rea- sons for this trend are unknown, several studies have suggested that the interval of 10 to 20 years after entering practice is a particu- larly stressful time for physicians.v' and it is possible that many physicians reduce their workload in response to the distress they experience during this period of their career.
Reducing professional work hours may be a. helpful strategy for individual physicians experiencing burnout.21,24-26 A part-time physician is certainly preferable to a physi- cian leaving practice altogether. Organiza- tions that provide physicians the flexibility to adjust their FTE as a "safety valve" to pre- serve meaning and satisfaction may have a competitive advantage in recruiting and retaining physicians. Although working less than full-time creates its own set of
challenges,27,28 the available cross-sectional data suggest that reducing work from full- time to part-time can be an effective strategy for physicians to improve satisfaction21,24-26 and increase their sense of control over work24,25 These studies also indicate that physicians working part-time does not adversely impact patient satisfaction?5,29-32 quality of care,30,31or effictency'"
Although it may be a beneficial approach to improve personal well-being, physicians reducing their professional effort has the potential to exacerbate the already substantial US physician workforce shortage and could also impact continuity of care for patients.v" Physicians reducing their professional effort may also compound the challenge that many medical centers have in preserving adequate patient access to physicians-particularly primary care physicians. It is notable that primary care physicians, who experience some of the highest rates of professional bumout.i" were more likely to work part- time across all years of our study.
There is a societal imperative to provide physicians a better option than choosing between reducing their professional effort or burning out14,33 Efforts to address this chal- lenge must first recognize that physician burnout is, in large part, a system issue. Sincere efforts to alleviate the problem must address the drivers of burnout, including excessive work- loads, inefficiencies in the practice environment, loss of flexibility and control over work, barriers to healthy work-life integration, and erosion of the meaning in work34-37 Individual, work unit, organization, and national factors also contribute to each of these dimensions (Figure 2). A number of important studies have provided insights in this regard. Health care organizations must improve the efficiency fh . . ~~~ 1"or t e practice environment, ' , set rea 1StlC
productivity expectations.'? cultivate physician engagement.I" help physicians optimize "career fit,,,37 nurture teamwork and community.T' foster flexibility and control over work,35,36 and help physicians cultivate meaning in their work37,42
Our study is subject to several important limitations. Although it is a prospective and longitudinal study, we cannot definitively determine cause and effect. There are many reasons to reduce work hours other than
428 Mayo Clin Proc. - April 2016;91(4):422-431 _ hltp'//dx.doi.org/10.1016/j.mayocp.2016.02.001 www.mayodinicproceedings.org
PHYSICIAN BURNOUT AND REDUCTIONS IN PROFESSIONAL WORK EFFORT
• 9ii;{~({1\ ' l ,- I-- -\l'if.fl Primary drivers \ )- ,- of burnout and "'-D Q satisfaction '""\(~~\} in physicians Individual Work unit Organization National factors
Workload • Specialty • Productivity expectations • Productivity targets • Structure reimbursement • Practice location • Team structure • Method of Medicare/Medicaid • Decision to increase work to • Efficiency compensation Bundled payments
Increase Income • Use of allied health Salary Documentation professionals Productivity based requirements
Efficiency • Experience • Availability of support • Integration of care • Integration of care • Ability to prioritize staff & their experience • Use of patient portal • Requirements for: • Personal efficiency • Patient check-in • Institutional efficiency Electronic prescribing • Organization skills efficiency/process EHR Medication reconciliation • Willingness to delegate • Use of Scribes Appointment system Meaningful use of EHR • Ability to say "no" • Team huddles Ordering systems • Certification agency facility
• Use of allied health regulations professionals
Control over • Personality • Degree of flexibility: • Scheduling system • Precertifications for work/autonomy/ • Assertiveness Control of physician • Policies testsitreatments flexibility • Intentionality calendars • Affiliations that • Insurance networks that
Clinic start/end times restrict refenrals restrict referrals Vacation scheduling • Rigid application • Practice guidelines Call schedule practice guidelines
Work-life • Priorities and values • Call schedule • Vacation policies • Requirements for: integration • Personal characteristics • Structure night/weekend • Sick/medical leave Maintenance certification
Spouse/partner coverage • Policies Licensing Children/dependents • Cross-coverage for time Part-time work Health issues away Flexible scheduling
Meaning in work • Self-awareness most • Match work to talents & • Organizational culture • Evolving supervisory role of personally meaningful aspect interests of individuals • Organizational physicians (potentially less of work • Collegiality in practice leadership direct patient contact)
• Ability to shape career to environment • Organization values • Reduced funding focus on interests • Work unit leadership • Practice environment Research
• Doctor-patient relationships • Opportunities for • Opportunities for Education • Personal recognition of involvement professional
positive events at work Education development Research • Organization's mission Leadership Service/quality vs profit
• Collegiality across the organization
FIGURE 2. Drivers of burnout and examples of factors that influence each driver.
experiencing problems with burnout and satisfaction, including personal or family health problems, child rearing, need to spend more time with family, and a desire to pursue other interests.T Providing flexi- bility that allows physicians to adjust their professional effort at different points in their career may be a useful approach to promote diversity and help physicians meet both per- sonal and professional priorities29,43-45
Because we were unable to measure and control for all these potential confounders,
however, it could be argued that our find- ings on the relationship between burnout! satisfaction and reductions in FIE are all the more striking. It should be noted that although the physicians in our study worked on separate academic campuses in 3 different states, they all worked for a single health care organization, a factor that may influence the generalizability of our findings.
Our study also has several important strengths. First, the study was prospective, included physicians from all specialties,
Mayo Clin Proe .• April 2016:91(4):422-431 • http://dx.doi.orgi10.1016/j.mayoep.2016.02.001 www.mayoclinicproeeedings.org
430
and utilized actual human resources/payroll department data to measure changes in pro- fessional effort. Participants were unaware of the specific aspects of this analysis corre- lating survey response to changes in profes- sional effort, which were independently assessed using administrative payroll records. Second, participation in the survey component of our study was exceptionally high, particularly for a study of physicians.I? Third, burnout was assessed using a well-validated metric. 19,20
Fourth, we were able to longitudinally evaluate changes in burnout and satisfaction at the level of individual physicians and assess the relation- ship with changes in FTE.
CONCLUSION Among the physicians in a large health care orga- nization, measures of burnout and satisfaction were strongly associated with subsequent reduc- tions in work effort as measured by independent employment records over the following 24 months. Further research is needed to under- stand whether these factors are causal and whether modifications to the practice environ- ment aimed at reducing burnout and improving satisfaction would modify this relationship.
ACKNOWLEDGMENTS Dr Reeves had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
SUPPLEMENTAL ONLINE MATERIAL Supplemental material can be found online at http://www.mayoclinicproceedings.org. Sup- plemental material attached to journal articles has not been edited, and the authors take re- sponsibility for the accuracy of all data.
Abbreviations and Acronyms: FTE = full-time equivalent; MBI = Maslach Burnout Inventory; OR = odds ratio
Affiliations (Continued from the first page of this article.), Intelligence, Purchase, NY (K.A.S., DR); and
Human Resources (M.M.) and Department of Radiology (S.j.B.), Mayo Clinic, Jacksonville, FL.
Grant Support: Funding for this study was provided by the
Mayo Clinic Office of Organization and Leadership Devel- opment and the Mayo Clinic Department of Internal Med-
icine Program on Physician Well-being.
Correspondence: Address to Tait D. Shanafelt, MD, Department of Internal Medicine, Mayo Clinic, 200 First St SW, Rochester, MN 55905 ([email protected]).
REFERENCES 1. Association of American Medical Colleges. The Complexities
of Physician Supply and Demand: Projections from 20 I3 to 2025: Final Report. Association of American Medical Colleges website. https:llwww.aamc.org/download/426248/datalthe complexitiesofphysiciansupplyanddemandprojectionsfrom20 13 toz.pdt, Published March 2015. Accessed September 2, 2015.
2. US Department of Health and Human Services. The Physician Workforce: Projections and Research into Current Issues Affecting Supply and Demand. US Dept of Health and Human Services website. http://bhpr.hrsa,gov/healthworkforce/reports/ physwfssues.pdf. Published December 2008. Accessed April 15,2015.
3. Eden j, Berwick 0, Wilensky G, eds. Graduate Medical Education That Meets the Notion's Health Needs. Washington, DC: National Academies Press; 2014.
4. Bressi C, Manenti S, Porcellana M, et al Haemato-oncolcgy and burnout: an Italian survey. BrJ Cancer. 2008;98(6): I046-1 052.
5. Shanafelt TO, Dyrbye LN, West CPo Physician burnout: an urgent call for early intervention-reply. JAMA Intern Med. 2013;173(8):710-711.
6. Shanafelt TO, Hasan 0, Dyrbye LN, et al. Changes in burnout and satisfaction with work-life balance in physicians and the general US worl<ing population between 20 I I and 2014. Mayo Clin Proc. 20 15;90( 12):1600-1613.
7. Shanafelt T, Sloan J, Satele 0, Balch C. Why do surgeons consider leaving practice [letter]? ] Am Call Surg 20 I 1;212(3): 421-422.
8. Shanafelt TO, Raymond M, Kosty M, et ai. Satisfaction with work-life balance and the career and retirement plans of US oncologists. J Clin Oneal. 20 14;32( I I): I 127-1 135.
9. Williams ES, Konrad TR, Scheckler WE, et al. Understanding physicians' intentions to withdraw from practice: the role of job satisfaction, job stress, mental and physical health. Health Core Manage Rev. 20 I0;35(2): I05-1 15.
10. Dewa CS, Loong 0, Bonato S, Thanh NX, jacobs P. How does bumout affect physician productivity? a systematic literature re- view. BMC Health SeN Res. 2014; 14:325.
11. Zhang Y, Feng X. The relationship between Job satisfaction, burnout, and turnover intention among physicians frorn urban state-owned medical institutions in Hubei, China: a cross- sectional study. BMC Health SeN Res. 20 I I; I I:235.
12. Landon BE, Reschovsky JD, Pharn HH, Blumenthal D. Leaving medicine: the consequences of physician dissatisfaction. Med Core. 2006;44(3):234-242.
13. Sibbald B, Bojke C, Gravelle H. National survey of job satisfac- tion and retirement intentions among general practitioners in England. BM). 2003;326(7379):22.
14. Dewa CS, jacobs P, Thanh NX, Loong D. An estirnate of the cost of burnout on early retirement and reduction in clinical hours of practicing physicians in Canada. BMC Health SeN Res. 2014; 14:254.
15. Maslach C, jackson SE, Leiter MP. Maslach Burnout Inventory Manual. 3rd ed. Palo Alto, CA: Consulting Psychologists Press; 1996.
16. Shanafelt TO, Boone S, Tan L, et al, Burnout and satisfaction with work-life balance arnong US physicians relative to the general US population. Arch Intern Med. 2012; 172( 18): 1377-1385.
17. Shanafelt TO, Kaups KL, Nelson H, et al. An interactive indi- vidualized intervention to promote behavioral change to in- crease personal well-being in US surgeons. Ann Surg. 2014; 259( I):82-88.
Mayo Clin Prae. II April 2016:91(4):422-431 II http://dx.doi.org/l0.1016Jj.mayocp.2016.02.001 www.mayoclinicproceedings.arg
PHYSICIAN BURNOUT AND REDUCTIONS IN PROFESSIONAL WORK EFFORT
18. West CP, Shanafelt TD, Kolars [C, Quality of life, burnout, educational debt, and medical knowledge among internal medicine residents. )AMA. 20 I 1;306(9):952-960.
19. West CP, Dyrbye LN, Sloan jA. Shanafelt TD. Single item mea- sures of emotional exhaustion and depersonalization are useful for assessing burnout in medical professionals.) Gen Intern Med. 2009;24( 12): 1318-1321.
20. West CP, Dyrbye LN, Satele DV, Sloan jA, Shanafelt TD. Con- current validity of single-item measures of emotional exhaus- tion and depersonalization in burnout assessment. ) Gen Intern Med. 20 12;27( II): 1445-1452.
21. Cull WL, O'Connor KG, Olson LM. Part-time work among pediatricians expands. Pediatrics. 20 I0; 125( I): 152-157.
22. Orkin FK, McGinnis SL, Forte Gj, et al. United States anesthesi- ologists over 50: retirement decision making and workforce im- plications. Anesthesiology. 2012; I 17(5):953-963.
23. Dyrbye LN. Varkey P, Boone SL, Satele DV, Sloan jA, Shanafelt TD. Physician satisfaction and burnout at different career stages. Mayo Clin Proc. 20 13;88( 12): 1358-1367.
24. Murray A, Safran DG, Rogers WH, Inui T. Chang H, Montgomery jE. Part-time physicians: physician workoad and patient-based assessments of primary care performance. Arch Fom Med. 2000;9(4):327-332.
25. Mechaber HF, Levine RB. Manwell LB, et al. Part-time physi- cians ... prevalent, connected, and satisfied.) Gen Intern Med. 2008;23(3):300- 303.
26. McMulTaY jE, Heiligers Pj, Shugerman RP, et al; Society of Gen- eral Internal Medicine Career Satisfaction Study Group (CSSG). Part-time medical practice: where is it headed? Am J Med. 2005; . 118(1):87-92.
27. Ofri D. Part-time medicine. JAMA. 2009;30 I (3):254-255. 28. Harrison RA, Gregg jL. A time for change: an exploration of at-
titudes toward part-time work in academia among women in- ternists and their division chiefs. Acod Med. 2009;84( I):80-86.
29. Levine RB, Harrison RA, Mechaber HF, Phillips C. Gallagher TH. Professional characteristics and job satisfaction among SGIM rnembers: a comparison of part-time and full-time physician members. J Gen Intern Med. 2008;23(8): 1218-1221.
30. Fein OT, Garfeld R. Impact of physicians' part-time status on inpatients' use of medical care and their satisfaction with physi- cians in an academic group practice. Acod Med. 199 I ;66( I I): 694-698.
31. Parkerton PH, Wagner EH, Smith DG. Straley HL. Effect of part-time practice on patient outcomes. J Gen Intern Med. 2003; 18(9):717-724.
32. Panattoni L, Stone A, Chung S, Tai-Seale M. Patients report bet- ter satisfaction with part-time primary care physicians, despite less continuity of care and access. J Gen Intern Med. 2015; 30(3):327-333.
33. Dyrbye LN, Shanafelt TD. Physician burnout: a potential threat to successful health care reform. JAMA. 20 I I; 305( 19):2009-20 I O.
34. Shanafelt TD, Sloan jA, Habermann TM. The well-being of phy- sicians. Am) Med. 2003; I 14(6):513-519.
35. Sinsky CA, Willard-Grace R, Schutzbank AM, Sinsky T A, Margolius D, Bodenheimer T. In search of JOY in practice: a report of 23 high-functioning primary care practices. Ann Farn Med. 2013; I 1(3):272-278.
36. Linzer M, Levine R, Meltzer D, Poplau S, Warde C. West CPo 10 Bold steps to prevent burnout in general internal medicine [editorial]. J Gen Intern Med. 20 14;29( I): 18-20.
37. Shanafelt TD, West CP, Sloan jA, et al. Career fit and burnout among academic faculty. Arch Intern Med. 2009; 169( 10):990-995.
38. Linzer M, Baier Manwell L, Mundt M, et al Organizational climate, stress, and error in primary care: the MEMO Study. In: Henriksen K Battes jB, Marks ES, Lewin DI, eds. Advances in Patient Safety: From Research to Implementation, Vol I Rockville, MD: Agency for Healthcare Research and Quality; 2005:65-77.
39. Balch CM, Shanafelt TD, Dyrbye L, et al Surgeon distress as calibrated by hours worked and nights on call. J Am Call Surg. 20 I0;21 1(5):609-619.
40. Shanafelt TD, Gorringe G, Menaker R, et al, Impact of organi- zational leadership on physician burnout and satisfaction. Moyo c/in Proc. 2015;90(4):432-440.
41. West CP, Dyrbye LN, Rabatin jT, et al. Intervention to pro- mote physician well-being, job satisfaction, and professional- ism: a randomized clinical trial. JAMA Intern Med. 2014; 174(4):527 -533.
42. Horowitz CR, Such man AL, Branch WT Jr, Frankel RM. What do doctors find meaningful about their work? Ann Intern Med. 2003; 138(9):772-775.
43. Carr PL, Ash AS, Friedman RH. et al. Relation of family re- sponsibilities and gender to the productivity and career satisfaction of medical faculty. Ann Intern Med. 1998; 129(7):532-538.
44. Linzer M, Warde C, Alexander RW, et al. Part-time careers in academic internal medicine: a report from the Association of Specialty Professors Part-Time Careers Task Force on behalf of the Alliance for Academic Intemal Medicine. Acod Med. 2009;84( I0): 1395-1400.
45. Kahn jA. Degen S], Mansour ME, et al Pediatric faculty mem- bers' attitudes about part-time f~culty positions and policies to support part-time faculty: a study at one rnedical center. Acod Med. 2005;80( I0):931-939.
46. Asch DA, jedrziewski MK, Christakis NA. Response rates to mail surveys published in medical joumals. ) Clin Epidemiol. I997;50( I0): I 129-1 136.
Mayo Clin Proc. III April2016;91(4h422-431 III httpJldx.doi.org/l0.1016/j.mayocp.2016.02.001 www.mayoclinicproceedings.org
431
http://www.mayoclinicproceedings.org
EDITORIAL
®crOSSMark
408
COJ
The Secret of the Care of the Doctor Is in Caring for the Doctor
y first clinic appointment of the day runs late. After an initial greeting, my patient opens up about feeling
overwhelmed. He is struggling to manage home and work responsibilities. In his pro- fessional sphere, he is experiencing pressure to be more productive and notes less energy and joy related to his work. At home, he is often too tired to spend quality time with his family. He worries that he is not supporting his spouse enough, who is at home with their young children. We discuss strategies to create better balance in his life. We never get a chance to talk about his medical issues. The next day I see my patient again. This time he is walking the halls of the hospital. He looks harried, but smiles and pauses to say to me in passing, "Thanks for letting me vent yesterday." Today he is wearing his white coat; he is an attending physician overseeing the care of his own patients.
In the current issue of Mayo Clinic Pro- ceedings, Shanafelt et all use a prospective, longitudinal research design to explore the impact of physician burnout and low work satisfaction on reductions in professional effort in a single large health care organization. As the authors state, previous studies have linked physician burnout and low satisfaction with the intent to decrease work hours or to leave clinical practice altogether, suggesting a sig- nificant but unknown potential effect on the physician workforce.
Using administrative and payroll statistics, Shanafelt et al discovered an overall increase in the number of physicians working less than full time during the study period (2008-2014). Women and primary care physicians were more likely to work less than full time compared with other physician groups during the study, and the authors noted an increase in the percentage of men older than 55 years working less than full time. Burnout and satisfaction were measured in 2011 and 2013 as part of an organization-wide physician survey. These results were then linked to
full-time equivalent employment data to assess reductions in work effort prospectively. Physi- cian burnout and low satisfaction predicted reductions in full-time equivalent over time. This was true for data analyzed in aggregate and at the individual physician level.
The authors appropriately acknowledge that their study does not provide a complete picture; there are many reasons why physicians may choose to decrease their work effort, and although the Shanafelt et al research found an association between burnout and low satisfac- tion vs decreased effort, causality cannot be fully established. However, their findings provide further compelling evidence that now is the time to address physician burnout and dimin- ished work satisfaction. We are currently facing projected workforce shortages, especially among primary care specialties, at the same time that our population is aging 'and there is increased demand to care for newly insured patients.
Burnout and low work satisfaction are common among all types of physicians, although those in primary care and other frontline roles report the highest levels of bumout' In addition to reducing work effort, physician burnout has been associated with a variety of undesired physician behaviors and outcomes, including medical errors, decreased empathy, substance abuse, divorce, and depres- sion3
,4 Concerns over physician burnout are widely held, as indicated elsewhere in this issue of the Proceedings, in the correspondence of Khoo' and Eiser6 Physician burnout has a tremendous potential effect on our health care system in overall costs and the quality of care that we provide to our patients, not to mention the personal cost to physicians and the people and relationships in their lives.
Although Shanafeltet all demonstrate that burned out physicians are more likely to reduce their work effort, the impact of work- ing less than full time in the study group is unknown. Will these physicians eventually reenter the full-time workforce refreshed, not burned out, and more satisfied? Part-time
Mayo Clin Proc. II April 2016:91(4),408-410 II http'//dx.doi.org/l0.l016/j.mayocp.2016.02.008 www.mayodinicproceedihgs.org II © 2016 Mayo Foundation for Medical Education and Research
physicians must still manage the challenges of the current health care system, such as requirements for increased documentation, lack of control over schedules, and working in low resource settings (common in primary care and emergency departments). Of note, in previous cross-sectional studies, control over schedule and fewer total work hours were associated with decreased bumout.I'"
Following this narrative, the authors' pro- pose that reducing work effort may serve as a "safety valve" for physicians who are experi- encing burnout and low work satisfaction. The safety valve option provides an alternative for physicians to leaving clinical practice altogether. Extending the safety valve metaphor may help to highlight the implications of physician burnout and low work satisfaction on the health care system and offer strategies to address them.
Safety valves are designed to prevent catastrophic events by automatically releasing a substance, such as steam from a closed sys- tem, when certain limits are reached. Gener- ally, safety valves work well in this type of setting. However, the more complex the sys- tem, such as in politics, labor relations, envi- ronmental protection concerns, and health care, the less effective a safety valve becomes. A safety valve works by averting an impending disaster, often at the last possible moment, not by addressing root problems or by changing the way a system functions. Safety valves rarely lead to sweeping innovation.
Medicine as a profession has traditionally embraced the "ideal worker" role, in which physicians are available to work full time and without Significant family obligations.i" Typi- cally, US physicians work, on average, 50 hours per week, with 30% working 60 hours or more per week.r ' Reducing work effort provides a very real option for many individ- ual physicians experiencing burnout or other types of distress, and in some situations has led to local novel approaches to traditional medical careers and clinical care.12 For some physicians, the part-time option represents a proactive, conscious decision to "work differ- ently," for example by creating time to practice the self-care and healthy behaviors they encourage among their patients and to spend more time with family.13 In fact, quality and satisfaction are not diminished (and are often improved) among patients of part-time
h .. 1415 H . d bP ysicians." " owever, as pomte out y the authors, physician reductions in work effort may increase pressures on our health care system to maintain timely access to care and continuous clinical coverage. If up to 50% of our physicians are burned out and burnout predicts reduction in work effort, we may be at risk for overwhelming the part-time safety valve option.
To avert overwhelming our health care system, we must look beyond safety valves and address larger system issues-but where to start and on what foundation? In 1927, Dr Francis Peabody famously addressed what many at the time saw as an impending crisis in medicine: the triumph of scientific advancement to the detri- ment of humanism in caring for patients. Pea- body wisely and simply reminded medical students, physicians, and the public that "the secret of the care of the patient is in caring for the patient.v'" Any effort to address burnout and low work satisfaction among physicians should begin with understanding the context within which we currently live and practice, and from a place of also caring about the physician. My patient (the physician described at the begin- ning of this article) is unlikely to choose part- time work for a variety of reasons, and so I fear he is at risk for burnout. How shall we care for him and others like him?
An initial step to adequately address physician burnout is to measure it and to identify the factors that are associated with it. Mayo Clinic and the study authors should be commended for collecting and sharing these data at regular intervals. It is hoped that organizational leaders are using this informa- tion to develop and track efforts to address physician burnout and satisfaction at the sys- temwide level and in a way that values and cares for the physician. Other academic in-. stitutions and health care organizations are taking a similar path and have created wellness committees and programs to address physi- cian burnout using a systemwide approach.V
ACKNOWLEDGMENTS The title of this editorial is a reimagining of Francis Peabody's famous statement, "for the secret of the care of the patient is in caring for the patient."
Details of the patient described in this editorial have been altered in compliance with
Mayo Clin Proc. II April 2016;91(4);408-410 II http;//dx.doi.org/10.1016/j.mayocp,2016.02.00S www.mayoclinicproceedings.org
Health. Insurance Portability and Account- ability Act of 1996 regulations.
Rachel B. Levine. MD. MPH Division of General Intemal Medicine
Department of Medicine Johns Hopkins School of Medicine
Baltimore, MD
Correspondence: Address to RachelB. Levine,MD, MPH, Division of General Intemal Medicine, Department of Medicine, Johns Hopkins School of Medicine, Baltimore, MD ([email protected]).
REFERENCES 1. Shanafelt TD, Mungo M, Schmitgen j, et al. Longitudinal study
evaluating the association between physician burnout and changes in professional work effort. Moyo Clin Proc 20 I6; 91 (4):422-431.
2. Shanafelt TD, Boone S, Tan L, et al. Burnout and satisfaction with work-life balance among US physicians relative to the general US population. Arch Intern Med. 2012; 172( 18): 1377-1385.
3. Wallace jE, Lemaire jB, Ghali WA Physician well ness: a missing quality indicator. Lancet 2009;374(9702): 1714-1721.
4. Shanafelt TD, Sloan jA Haberman TM. The well-being of physicians. Am} Med. 2003; I 14(6):513-519.
5. Khoo TK. Factors affecting burnout in physicians. Mayo ClinProc 2016;91 (4):536-537.
6. Eiser AR. Burnout and the ethos of medical practice. Mayo Clin Proc 2016;91 (4):537.
7. Keeton K, Fenner DE, johnson TR, Hayward RA Predictors of physician career satisfaction, work-life balance, and burnout Obstet Gyneco/. 2007; I09(4):949-955.
8. McMunray jE, Linzer M, Konrad TR, Douglas j, Shugerman R, Nelson K The SGIM Career Satisfaction Study Group. The work lives of women physicians results from the physician work life study.) Gen Intern Med. 2000;15(6):372-380.
9. Mechaber HF, Levine RB, Manwell LB, et al; MEMO Investiga- tors. Part-time physicians ... prevalent. connected, and satisfied. } Gen Intern Med. 2008;23(3):300-303.
10. Williams j. Unbending Gender Why Family and Work Conflict and Whatto Do About It New York NY: Oxford University Press; 2001.
11. Ly DP, Seabury SA jena AB. Divorce among physicians and other healthcare professionals in the United States: analysis of census survey data. BM). 20 15;350:h706.
12. Levine RB, Mechaber HF. Opting in: part-time careers in academic medicine. Am} Med 2006; I 19(5):450-453.
13. Harrison RA, Gregg u, A time for change: an exploration of attitudes toward part-time work in academia among women internists and their division chiefs. Acod Med. 2009;84( I):80-86.
14. Parkerton PH, Wagner EH, Smith DG, Straley HL Effect of part-time practice on patient outcomes. J Gen Intern Med. 2003; 18(9):717-724.
15. Panattoni L, Stone A Chung S, Tai-Seale M. Patients report better satisfaction with part-time primary care physicians, despite less continuity of care and access.} Gen Intern Med. 2015;30(3):327-333.
16. Peabody FW. The care of the patient JAMA I927;88( 12): 877-882.
17. Linzer M, Levine R, Meltzer D, Poplau S, Warde C, West CPo 10 bold steps to prevent burnout in general internal rnedicine.
} Gen Intern Med. 20 14;29( I): 18-20.
Mayo Clin Proc, - April 2016:91 (4):408-41 0 _ httpd/dx.doi.orgI10.1016/j.mayocp.2016.02.008 www.mayoclinicproceedings.org
Titles
TABLE 1. Proportion of Physicians at Mayo Clinic Working Less Than Full-time. 2008-~014a.b
3:
Statistical Analyses
429
Images
Mayo Clin Proc. III April2016;91(4h422-431 III httpJldx.doi.org/l0.1016/j.mayocp.2016.02.001
431
The Secret of the Care of the Doctor Is in
408
Images
REFERENCES