Changes in the US healthcare system, including alterations in reimbursement schema, limitations on autonomy, and escalating costs of malpractice, have placed ever-increasing strains on physicians. Studies show that 10% to 35% of physicians are dissatisfied with their work, varying by specialty and geography.1–3 Professional isolation and work-life stress mar the well-being of many physicians.4,5
Physician job dissatisfaction is costly. Physicians incur direct personal costs, in the forms of impaired health6 and lost income.4,5 Dissatisfied physicians are more likely to leave their practices, cut down on their clinical hours, and retire early,6–8 leading to substantial economic costs.9,10 Most importantly, patients are affected.11–15 Dissatisfied physicians have less satisfied and less adherent patients,11,12 and dissatisfied physicians seem to be more likely to engage in risky prescribing practices.13,14
Given the prevalence of physicians’ professional dissatisfaction and its associated costs, it is essential to characterize the correlates of dissatisfaction. With increasing emphasis on quality of care and patient safety, it is important to elucidate the relationships between efforts to improve quality and safety and physicians’ professional lives. Given that autonomy and time-stress are important mediators of job satisfaction, it is possible that quality improvement (QI) activities, especially those that occur without physician input or those that increase physician workload, may worsen well-being.16–18 We therefore undertook this study to examine how physician and practice factors, including self-reported practice involvement in QI, are related to physician work-life dissatisfaction.
Survey development, administration, and sampling methods have been described elsewhere19 and are summarized here.
We identified all physicians practicing in Massachusetts in the spring of 2005 and excluded those who were retired, in training, or not engaged in direct patient care. This population included 20,227 physicians practicing in 6174 unique sites. We drew a stratified random sample of 1921 practice sites and randomly selected one physician to represent each practice site. After excluding closed practices, the final sample size was 1887 physicians.
An 8-page survey including questions about physician and practice characteristics was mailed to participants. Physician characteristics included race, gender, age, number of visits per week, primary care versus specialty care, and whether the physician was an owner of the practice. Practice characteristics included the type of practice (ie, solo or group practice), practice location (rural vs. nonrural), presence of electronic health record, perceived availability of resources for practice expansion or improvement, and whether teaching of medical students or residents occurred at the site. Physicians also reported their perceptions about their practice's quality and engagement in QI activities (Box 1).
Main Outcome Measures
Survey items pertaining to professional well-being are shown in Box 2. In multivariate models predicting each outcome, a consistent pattern of relationship was observed between all significant covariates and the survey items regarding stress, long hours, and demoralization. Exploratory factor analysis indicated a high degree of communality among these items (Cronbach's alpha = 0.76). We therefore created a composite outcome measure representing “work-life stress” that was calculated as the average of each participant's responses to these 3 items. Work-life stress, isolation, and job dissatisfaction were considered as 3 separate outcome measures.
Data were analyzed using SAS, version 9.1 (SAS Institute, Inc, Cary, NC). Bivariate relationships between predictor and outcome measures were assessed using χ2 tests for categorical variables. We used linear regression to identify the independent correlates of professional isolation, work-life stress, and job dissatisfaction with each outcome considered in its original ordinal form. All available physician and practice characteristics were included in each model. As the outcome measures were not normally distributed, we relied on the large sample size and the Central Limit Theorem to justify the linear regression approach. We also dichotomized each measure and carried out logistic regression analyses to confirm the results of the linear regression; because the results did not materially differ, only the linear regression results are presented. Standardized regression coefficients are presented to allow for comparison of the strength of associations. Correlation between the 3 outcome measures and between the availability of practice resources and QI activities was also examined using the Pearson correlation coefficient.
A total of 1345 physicians returned questionnaires (response rate: 71.4%). We excluded physicians who did not see outpatients (n = 162) and those who lacked responses for one or more outcomes variables (n = 12), leaving 1171 for analysis. Respondents and nonrespondents did not differ with respect to practice size, specialty, and practice location. Table 1 shows the characteristics of the respondents.
Figure 1 illustrates the proportion of physicians who reported that their practices engaged in QI activities and had systems in place to prevent errors, as well as the proportion of physicians perceiving quality problems within their practice. Most physicians (85%) reported QI activities in their practices; about one-third (33%) reported that their practices had quality problems.
Reporting increasing levels of resources available for practice improvement was associated with self-reported QI activities. (Pearson coefficient = 0.13, P < 0.001).
A total of 17% of respondents rated isolation from colleagues as a moderate or serious problem (Supplemental Table 1, https://links.lww.com/A1452). Physicians whose practices engaged in QI activities were less likely to report isolation (15% vs. 27%, P < 0.001). Similarly, physicians reporting that their practices evaluated QI interventions after implementing them also reported lower levels of isolation (14% vs. 21%, P < 0.001). Physicians who reported that their office did not have quality problems reported lower levels of isolation (14% vs. 19%, P = 0.04).
In linear regression analysis, after adjusting for all available potential confounders, evaluation of practice QI efforts and presence of practice quality problems remained independent correlates of the degree of professional isolation (Table 2).
The median score on the work-life stress scale was 2.3, indicating that work-life stress was at least a slight problem for more than half of respondents. Nearly one-third (31%) had a score of 3 or greater, suggesting that work-life stress was a moderate or serious problem (Supplemental Table 1 https://links.lww.com/A1452). All significant relationships between the covariates and each of the outcomes that contribute to the work-life stress scale (work long hours, demoralization, personal/professional stress) were of the same magnitude and direction as those with the composite scale (Supplemental Table 2, https://links.lww.com/A1452).
Self-reported practice quality and QI activities were associated with physician work-life stress. Physicians who reported that their practices did not have quality problems were less likely to experience work-life stress (24% vs. 34%, P = 0.002). Physicians who reported that their practices evaluated the effectiveness of their QI interventions were less likely to have difficulties with work-life stress (27% vs. 37%, P < 0.001) as were those who reported that their practices had systems in place to prevent errors (25% vs. 35%, P < 0.001).
After controlling for all covariates, reporting quality problems in the practice was associated with greater levels of work-life stress, whereas evaluation of practice QI efforts was associated with lower levels of work-life stress (Table 2).
Dissatisfaction With Practice
Overall, 26.6% of respondents reported being somewhat dissatisfied (21.2%) or very dissatisfied (5.4%) with their practice situation (Supplemental Table 1, https://links.lww.com/A1452).
Physicians who reported higher levels of quality problems in their practice were more likely to be dissatisfied than were physicians who reported fewer quality problems (31% vs. 19%, P < 0.001). Practice QI activities, the evaluation of these activities, and having procedures and systems in place to prevent errors, were all associated (P < 0.001) with lower levels of dissatisfaction (Supplemental Table 1, https://links.lww.com/A1452).
In the fully adjusted model, perceiving that their practice had fewer quality problems and reporting greater participation in QI activities and evaluation of these activities, as well as having systems to prevent errors, all remained independently associated with lower levels of dissatisfaction (Table 2).
Correlation Between Outcome Measures
There was weak to moderate correlation between the 3 outcomes. Work-life stress and practice dissatisfaction were moderately associated (r = 0.58, P < 0.001). Isolation was less correlated with the other 2 outcomes (r = 0.28 for each pairwise correlation, both with P < 0.001).
This statewide study assessed the relationships between physician and practice characteristics and global professional isolation, work-life stress, and practice dissatisfaction. Isolation, work-life stress, and practice dissatisfaction affected a substantial proportion of providers. Physicians whose practices were engaged in QI efforts were less likely to report isolation, work-life stress, and dissatisfaction. Moreover, physicians who indicated that their practices had quality problems reported more professional isolation, work-life stress, and practice dissatisfaction. These relationships, persistent in multivariate analysis, suggest that physicians’ perceptions of quality and QI are directly related to factors influencing their well-being and professional satisfaction.
Because these data are cross-sectional, we cannot assert causal relationships between quality and QI activities and satisfaction. However, because participation in QI activities is modifiable by practices or individual physicians, these results may suggest that practices might effect positive change in physicians’ professional well-being by engaging physicians in QI efforts. The availability of resources for practice improvement seems to be an important correlate of both job satisfaction and reduced work-life stress and correlates with practice QI activities. This finding may have broad implications as payers, foundations, or medical societies may focus on creating rewards for practices making changes to improve quality and safety; the medical-home movement is an example of such efforts attempting to improve not only the quality of health care but also the professional lives of physicians.20,21
What does it mean that QI activities are correlated with physicians’ positive feelings about their professional lives? Intuitively, perceiving increased quality problems within the practice is associated with all of the negative outcomes: isolation, work-life stress, and job dissatisfaction. Challenges to autonomy and feeling stressed for time adversely affect physician well-being.16–18 It may be surprising, then, that QI activities were associated with positive trends in all outcome measures. Perceiving that QI activities are rigorously assessed within the practice was associated with lower levels of isolation, work-life stress, and dissatisfaction. This observation might suggest that practices that truly assess their QI activities undertake an iterative process that solicits physician input and collaboration, thereby decreasing stress they may feel with interventions imposed by others, and creating a feeling of belonging and believing in shared commitment to providing good care. That these processes would contribute to improvement in physicians’ work-lives is consistent with demand-control models of stress that postulate that having control counteracts the stressful effects of practice demands.22
The strengths of this study include the statewide sampling frame with physicians from many specialties, the high response rate, and the large sample size, which enabled the assessment of many factors related to professional well-being.
There are several study limitations. These results represent practices in Massachusetts, potentially limiting their generalizability. The cross-sectional nature of this study prohibits us from drawing conclusions regarding the directionality of observed relationships. We speculate that engaging in QI activities leads physicians to report lower levels of dissatisfaction, isolation, and work-life stress. However, it is equally plausible that being less dissatisfied, less isolated, and less burdened by stress leads physicians to engage in QI activities and to perceive higher levels of quality in their practices. Although response rate did not vary by practice size, specialty, or location, nonresponse bias may have influenced the study results and unmeasured confounders. Future studies should incorporate longitudinal assessments to disentangle the temporal relationships between QI activities and measures of physician well-being and should furthermore attempt to obtain measures of practice systems and quality indicators that do not rely on physicians’ self-report.
This study builds on a strong foundation of literature that identifies physician work-life issues as important not only for the well-being of the physicians themselves but also for the quality of care they deliver to patients.12–14,20 Given the apparent positive relationship between being involved in QI activities and physician work-life measures, future efforts attempting to engage physicians in QI programs should assess the impact of these programs on physicians themselves and on the care that they deliver.
1. Landon BE, Reschovsky J, Blumenthal D. Changes in career satisfaction among primary care and specialist physicians, 1997–2001. JAMA
2. Landon BE, Reschovsky JD, Pham HH, et al. Leaving medicine: the consequences of physician dissatisfaction. Med Care
3. Skolnik NS, Smith DR, Diamond J. Professional satisfaction and dissatisfaction of family physicians. J Fam Pract
4. Linn LS, Yager J, Cope D, et al. Health status, job satisfaction, job stress, and life satisfaction among academic and clinical faculty. JAMA
5. Lavanchy M, Connelly I, Grzybowski S, et al. Determinants of rural physicians’ life and job satisfaction. Soc Indic Res
6. Williams ES, Skinner AC. Outcomes of physician job satisfaction: a narrative review, implications, and directions for future research. Health Care Manage Rev
7. Buchbinder SB, Wilson M, Melick CF, et al. Primary care physician job satisfaction and turnover. Am J Manag Care
8. Linzer M, Konrad TR, Douglas J, et al. Managed care, time pressure, and physician job satisfaction: results from the physician worklife study. J Gen Intern Med
9. Buchbinder SB, Wilson M, Melick CF, et al. Estimates of costs of primary care physician turnover. Am J Manag Care
10. Atkinson W, Misra-Hebert A, Stoller JK. The impact on revenue of physician turnover: an assessment model and experience in a large healthcare center. J Med Pract Manage
11. Haas JS, Cook EF, Puopolo AL, et al. Is the professional satisfaction of general internists associated with patient satisfaction? J Gen Intern Med
12. DiMatteo MR, Sherbourne CD, Hays RD, et al. Physicians’ characteristics influence patients’ adherence to medical treatment: results from the Medical Outcomes Study. Health Psychol
13. Melville A. Job satisfaction in general practice: implications for prescribing. Soc Sci Med [Med Psychol Med Sociol]
14. Grol R, Mokkink H, Smits A, et al. Work satisfaction of general practitioners and the quality of patient care. Fam Pract
15. Halbesleben JR, Rathert C. Linking physician burnout and patient outcomes: exploring the dyadic relationship between physicians and patients. Health Care Manage Rev
16. Ahluwalia S, Offredy M. A qualitative study of the impact of the implementation of advanced access in primary healthcare on the working lives of general practice staff. BMC Fam Pract
17. Chehab EL, Panicker N, Alper PR, et al. The impact of practice setting on physician perceptions of the quality of practice and patient care in the managed care era. Arch Intern Med
18. Solberg LI, Crain AL, Sperl-Hillen JM, et al. Effect of improved primary care access on quality of depression care. Ann Fam Med
19. Simon SR, Kaushal R, Cleary PD, et al. Correlates of electronic health record adoption in office practices: a statewide survey. J Am Med Inform Assoc
20. Goroll AH, Berenson RA, Schoenbaum SC, et al. Fundamental reform of payment for adult primary care: comprehensive payment for comprehensive care. J Gen Intern Med
21. Blue Ribbon Panel of the Society of General Internal Medicine. Redesigning the practice model for general internal medicine. A proposal for coordinated care: a policy monograph of the Society of General Internal Medicine. J Gen Intern Med.
22. Karasek RA, Theorell T, Schwartz JE, et al. Job characteristics in relation to the prevalence of myocardial infarction in the US Health Examination Survey (HES) and the Health and Nutrition Examination Survey (HANES). Am J Public Health