Table of Contents Author Guidelines Submit a Manuscript
Multiple Sclerosis International
Volume 2014, Article ID 624790, 14 pages
http://dx.doi.org/10.1155/2014/624790
Research Article

Factors Associated with Neurologists’ Provision of MS Patient Care

1RTI International, 701 13th Street NW No. 750, Washington, DC 20005, USA
2RTI International, 3040 East Cornwallis Road,Research Triangle Park, NC 27709-2194, USA

Received 21 December 2013; Accepted 24 March 2014; Published 24 April 2014

Academic Editor: Wolfgang Bruck

Copyright © 2014 Michael T. Halpern et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

Neurologists are central to providing quality care for individuals with MS. However, neurologist shortages may restrict access to care for MS patients. To examine factors influencing neurologists’ provision of MS care, we surveyed 1,700 US neurologists to assess demographic/practice characteristics, training, and attitudes toward MS care. The study population consisted of 573 respondents: 87 (15.2%) MS subspecialists and 486 (84.8%) “other neurologists,” including subspecialists in other neurology areas (i.e., non-MS) and general neurologists. MS subspecialists indicating they “enjoy interacting with MS patients” had a significantly greater rate of MS patients seen per week. In separate analyses of the “other neurologists” group, the rate of MS patients seen was lower among neurologists in university-based groups or those practicing in major cities; female neurologists; and neurologists who indicated lack of sufficient knowledge regarding MS patient care. Rates of MS patients seen were significantly greater for other neurologists who agreed that MS care involved “ability to improve patient outcomes and quality of life”; “dynamic area with evolving treatment options”; and “enjoy interacting with MS patients.” Understanding factors influencing MS patient care by neurologists and developing policies for appropriate access to care is critical for optimal outcomes among this population.

1. Introduction

Neurologists are central to the provision of quality, patient-centered care for individuals with MS. The substantial majority of MS patients depend on neurologists for high quality treatment and education about their condition [1, 2]; for example, MS patients receiving care from neurologists are more likely to receive and use disease-modifying therapies (DMT), participate in rehabilitation clinics, and receive care from rehabilitation specialists (e.g., occupational and physical therapists) and urologists compared with those MS patients receiving care from primary care providers [3]. Neurologists are also critical to quality MS care because of their detailed knowledge of treatment options for MS patients. MS treatment options have become increasingly complicated as the available DMTs increase and their benefits and safety profiles remain unclear [1, 4].

Among neurologists, MS subspecialists have the greatest familiarity with MS patient care. Little information is available regarding difference in treatment patterns or patient perceptions of care among individuals with MS receiving care from MS subspecialists versus from other neurologists. Although MS patients receiving care from neurologists and from MS subspecialists have been reported to experience similar use of healthcare services, those receiving care from MS subspecialists reported greater knowledge about and current use of a particular DMTs (i.e., interferon beta-1b), expressed more confidence in their physicians, and indicated greater participation in nondrug research [5].

Although neurologists are critical to providing MS care, shortages among the neurologist workforce (including both MS subspecialists and other neurologists) may restrict access to neurologists among individuals with MS [6]. Anecdotal evidence from patient groups suggests that MS patients experience difficulty in gaining access to appropriate care, especially from MS subspecialists; new patients may experience substantial waits and difficulties in finding available neurologists. Understanding the factors that influence the number of MS patients seen by neurologists can aid in identifying challenges that MS subspecialists and other neurologists encounter in seeing MS patients and can suggest potential solutions for resolving these challenges. This study examines results from a recently completed study assessing factors associated with the number of MS patients seen by MS subspecialists and other neurologists.

2. Materials and Methods

2.1. Instrument Development

We developed the MS Physician Workforce Neurologist Survey initially by reviewing previously developed physician career surveys to identify items relevant to assessing neurologists’ care for MS patients. We selected and adapted existing questions and crafted new items to create a draft survey, which included questions on respondents’ demographic characteristics, medical training, attitudes toward providing MS patient care, MS subspecialist status, practice characteristics, and numbers of MS patients seen.

To ensure comprehensiveness and usability of the survey, we sought input from an advisory panel of experts identified for this study and the American Academy of Neurology (AAN) MS Section Executive Committee. The advisory panel, including neurologists, reviewed and provided comments on the survey, which was then revised based on their feedback. Members of the AAN MS Section Executive Committee then reviewed the survey to ensure that it was feasible for AAN members to complete. This final version of survey was approved by the RTI International Institutional Review Board prior to distribution. The survey is available from the authors on request.

2.2. Study Population and Survey Administration

The sampling frame consisted of neurologists practicing in the United States who are members of the AAN. We collaborated with the AAN, which administered the survey and collected responses. The AAN contacted a random sample of 1,700 neurologists who resided in the United States and were members of the AAN, excluding retired members; those in medical school, residency, or fellowship; and AAN committee members who participated in survey review. Seven respondents were later removed from the sample because they were not in practice or did not see adult patients.

The AAN administered the survey via email as well as postal mail or fax to the sample of neurologists in January 2012. The AAN sent neurologists an email describing the study and including a hyperlink to the online version of the survey. The paper version also included a cover letter signed by the Chair of the AAN MS Section. AAN staff tracked responses to determine whether invited neurologists submitted surveys and, for nonresponders, distributed two email reminders to participate. Of the 1,693 neurologists invited to participate, 662 submitted responses (response rate of 39.1%).

2.3. Study Data

We used identification numbers to merge (deidentified) neurologists’ survey responses with demographic information and practice characteristics from the 2008 AAN Member Census. From the Census, practice arrangement were categorized as solo practice, neurology group, multispecialty group, university based group, other, and unknown. The “other” category included staff-model HMO; government hospital or clinic; and other public or private hospital or clinic setting.

2.4. Study Variables

The dependent variable for this analysis was self-reported total number of MS patients seen in an average week. Independent variables included demographic characteristics, practice characteristics, and attitudes toward MS patient care. Self-reported demographic characteristics included age group (categorized in quartiles), sex, race (White, Asian, and other race including Black/African American, American Indian/Alaska Native, and Native Hawaiian/other Pacific Islander groups combined because of small numbers), ethnicity (Hispanic/Latino or not), and year began medical practice (categorized in quartiles). Urban/rural designation of practice area was coded into three categories: within a major city, suburban or moderate-sized city, and rural area or small city.

Survey respondents were categorized as MS subspecialists using two sets of criteria. First, based on fellowship training those who had completed a MS fellowship and reported seeing, on average, at least one MS patient per week were classified as MS subspecialists. As neurologists may also subspecialize in MS patient care without completing a MS fellowship, respondents who reported seeing, on average, more than 12 MS patients per week (the median number of MS patient seen per week among neurologists who did not complete a MS fellowship but considered themselves to be MS subspecialists) were also classified as MS subspecialists.

The survey captured information on 17 attitudes toward MS patient care. Ten of these attitudes reflected factors that could limit the number of MS patients seen by respondents, while seven corresponded to positive aspects of providing MS care (Table 2). Attitudes were coded dichotomously (i.e., zero or one), based on whether a respondent indicated that each attitude was applicable to his or her practice.

2.5. Analyses

We computed bivariate analyses to examine factors associated with number of MS patients seen. This dependent variable (number of MS patients seen) was examined separately for MS subspecialists versus other neurologists (i.e., general neurologists and subspecialists in areas other than MS, referred to as “non-MS subspecialists”). Since this dependent variable was not normally distributed (Shapiro-Wilk tests for normality, ), we computed nonparametric bivariate analyses to examine factors associated with MS patient seen using the Wilcoxon-Mann-Whitney tests for independent variables with two categories and the Kruskal-Wallis tests for independent variables with three or more categories.

We also conducted bivariate analyses to compare characteristics and attitudes of other neurologists (non-MS subspecialists and general neurologists) who did not see any MS patients with those who saw MS patients, using tests or, because of small cell sizes, Fisher’s exact tests to examine relationships between discrete variables (e.g., sex) and the proportion seeing any MS patients.

We performed multivariate regression analyses to examine the association of neurologist characteristics and attitudes with numbers of MS patients seen while controlling for other potentially associated factors. Since residuals for the dependent variables were not normally distributed, we used negative binomial regression models with robust standard errors to identify predictors of number of MS patients seen. As factors associated with numbers of MS patients seen may differ between MS subspecialists and other neurologists, regression analyses were performed separately for these two populations. Results from the negative binomial regressions are presented as incidence rate ratios (IRRs), that is, the “rate” of seeing individuals with MS associated with a predictor variable relative to the reference case for that variable. IRRs less than 1.0 indicate decreased rates of seeing MS patients compared with the reference case, while IRRs greater than 1.0 indicate increased rates. Regression models included all independent variables (i.e., survey respondents’ characteristics) that were significantly or marginally significantly associated with the number of MS patients seen in bivariate analyses (Tables 1 and 2). All variables included in regression analyses were examined for multicollinearity (VIF > 5) prior to inclusion in the final model. Because of multicollinearity with age, the variable “year began medical practice” was removed from the regression model predicting MS patients seen per week among general neurologists and non-MS subspecialists (as age was not significantly associated with number of MS patients seen among MS subspecialists, age was not included in that regression model). Missing data generally comprised less than 3% of responses; individuals with missing data for descriptive, bivariate, or regression analyses were excluded from those analyses.

tab1
Table 1: Descriptive statistics for study populations.
tab2
Table 2: MS patients seen per week for MS subspecialists versus Non-MS subspecialists/general neurologists.

3. Results

3.1. Population Characteristics and MS Care Attitudes

After excluding neurologists who did not provide information about their subspecialty status, the study population consisted of 573 neurologists. Of these neurologists, 87 (15.2%) were categorized as MS subspecialists and 486 (84.8%) were “other neurologists,” including subspecialists in other areas of neurology (i.e., not MS) and general neurologists (Table 1). Among the 87 MS subspecialists, 26 (approximately 30%) were classified as MS subspecialists based on having completed an MS fellowship, while the remaining 61 were classified in this group based on seeing, on average, more than 12 MS patients per week.

After excluding neurologists who did not respond to the questions for other specified dependent variables, MS subspecialists saw approximately 25 MS patients per week on average (median = 20). Other neurologists saw approximately three MS patients per week on average (median = 3). MS subspecialists and other neurologists indicated substantial differences in attitudes regarding MS patient care (Table 2).

3.2. Characteristics Associated with Number of MS Patients Seen per Week

MS subspecialists saw, on average, more MS patients per week if they indicated the following attitudes: “enjoy interacting with MS patients” , “care involving a multidisciplinary team” (marginally significant), and “personal connection to individuals with MS who are not your patients” (marginally significant) (Table 2). MS subspecialists who began medical practice between 1972 and 1985 were also marginally more likely to see more MS patients per week than MS subspecialists who began practice later. Practice arrangement, physician age, sex, race, ethnicity, urban/rural practice area, and the remaining attitudes toward providing MS patient care were not significantly associated with number of MS patients seen (Table 2).

Among survey respondents in the “other neurologist” group, general neurologists saw significantly more MS patients per week than did non-MS subspecialists (4.4 versus 3.2, resp., ) (Table 2). Physicians in a neurology or multispecialty groups also saw more MS patients that did those in solo practice. Younger neurologists, female neurologists, those practicing for fewer years, and those based within a major city saw fewer MS patients.

General neurologist and non-MS subspecialist respondents who indicated the following attitudes saw significantly fewer MS patients per week: “lack of sufficient knowledge to feel comfortable caring for this patient population”; “lack of sufficient knowledge regarding newer disease-modifying drugs”; and “seldom encounter MS patients” (Table 2). In contrast, neurologists who indicated “insufficient reimbursement for time involved” surprisingly saw more MS patients. Other neurologist respondents saw more MS patients if they agreed that MS patient care involved “ability to improve patient outcomes and quality of life,” “dynamic area with evolving treatment options,” “care involving a multidisciplinary approach,” “enjoy interacting with MS patients,” and “research opportunities” (marginally significant). As with MS subspecialists, general neurologist and non-MS subspecialist respondents based within a major city saw fewer MS patients. Race, ethnicity, and the remaining attitudes toward providing MS patient care were not significantly associated with number of MS patients seen.

3.3. Multivariate Regression Analyses of Number of MS Patients Seen

We used negative binomial regressions (as presented in Section 2) to examine the association of neurologist characteristics and attitudes toward MS care with the number of MS patients seen per week while controlling for other potentially associated factors. Analyses were performed separately for MS subspecialists (Table 3) and general neurologists/non-MS subspecialists (Table 4). All survey items that had significant or marginally significant associations with number of MS patients seen in bivariate analyses (Table 2) were included as independent variables in these regression models.

tab3
Table 3: Negative binomial regression predicting MS patients seen per week among MS subspecialists.
tab4
Table 4: Negative binomial regression predicting MS patients seen per week among Non-MS subspecialists/general neurologists.

Among MS subspecialists, the only factor significantly associated with number of MS patients seen per week in regression analysis was the attitude “enjoy interacting with MS patients”; MS subspecialists who indicated this attitude had a rate of MS patients seen per week that was 1.6 times greater than did those not indicating this attitude. MS subspecialists who began medical practice between 1986 and 1993 saw fewer MS patients than did those who had been practicing longer, although this difference was only marginally significant .

In regressions analyses of the “other neurologists” group, the rate of MS patients seen per week among neurologists in a university-based group was approximately 60% (0.59 times) the rate for those in a solo practice. The rate of MS patient seen among female other neurologists was approximately 80% the rate among males. The rates of MS patients seen were significantly lower for non-MS subspecialists/general neurologists who indicated “lack of sufficient knowledge to feel comfortable caring for this patient population” and “seldom encounter MS patients.” The rates of MS patients seen were significantly greater for other neurologists who agreed with the following attitudes toward MS patient care: “ability to improve patient outcomes and quality of life,” “dynamic area with evolving treatment options,” and “enjoy interacting with MS patients.” Among other neurologists practicing in rural areas or small cities, the rate of MS patients seen per week was 1.30 times greater than among those practicing within major cities; practicing in suburban areas or moderate-sized cities was marginally associated with seeing more MS patients.

Figure 1 graphically presents results from Table 4, illustrating the incidence rate ratios (IRR) and 95% confidence intervals for the association of non-MS subspecialist/general neurologist characteristics and attitudes with the number of MS patients seen per week. The solid horizontal line indicates IRR of 1.0, corresponding to no significant difference from the reference group for that characteristics or attitude. Results are presented in three groups. Among neurology practice characteristics, those in university-based practices had a significantly decreased rate of seeing MS patients, while those in small city/rural practices had increased rates. Examining physician characteristics, female neurologist had significantly decreased rates of seeing MS patients. Finally, for attitudes towards MS patient care, “lack of sufficient knowledge for MS care” and “seldom encounter MS patients” were associated with decreased rates of MS patients seen while “ability to improve outcomes/quality of life,” “dynamic area with evolving treatment options,” and “enjoy interacting with MS patients” were associated with increased rates.

624790.fig.001
Figure 1: Incidence rate ratios for MS patients seen per week among non-MS subspecialists/general neurologists.

4. Discussion

To our knowledge, this is the first published study examining factors influencing the numbers of MS patients seen among a broad sample of US neurologists. Given the importance of neurologist care for MS [2, 3, 7], barriers to access neurologists could have substantial impacts on symptom control, disease progression, and quality of life among individuals with MS. Several studies have described barriers that individuals with MS may experience in receiving needed care. For example, MS patients in rural areas have been reported to have more limited access to MS subspecialists and related care services and had longer travel times than did patients in urban areas to receive specialized MS care [3, 8, 9]. Racial and ethnic disparities are also barriers to MS patient care: African Americans are less likely to have been treated by an MS subspecialist; Latinos are less likely to have received rehabilitation care compared to Caucasians [10]. These barriers to care may have direct consequences on patient outcomes; for example, individuals with MS residing in rural areas had significantly greater reductions in their physical components of health-related quality of life [11].

We found that MS subspecialists responding to this study’s survey saw approximately 25 MS patients per week, while other neurologists saw approximately 3.4 patients per week. It is not surprising that general neurologists and neurologists who subspecialize in clinical areas other than MS see, on average, fewer MS patients than do MS subspecialists. However, as the population of individuals with MS in the US continues to increase and only limited numbers of MS subspecialists (particularly outside of large urban/suburban locations) are available to serve MS patients, more MS patients may seek care from these other neurologists. The results presented in this study, providing information on how much care these other neurologists are currently providing for MS patients, may be useful for estimating the future capacity of the neurologist workforce for care for individuals with MS.

Among MS subspecialists, we identified only one factor that was significantly associated with number of MS patients seen per week in multivariate regressions (Table 3). Those who indicated that they enjoyed interacting with MS patients saw more individuals with MS. This may relate to “burnout” among MS subspecialists; that is, subspecialists who do not enjoy interacting with MS patients likely experience less enjoyment from providing medical care and instead may focus on other activities (e.g., teaching or administration) and see fewer individuals with MS. Shanafelt et al. [12] reported that neurologists had the third highest rate of burnout among all physician specialties.

More factors were found to be significantly associated with numbers of MS patients seen among other neurologists (i.e., general neurologists and non-MS subspecialists) (Table 4). Neurologists in university-based practices had significantly lower rates of seeing MS patients compared with the reference group, neurologists in solo practice. This may reflect the nonpatient care activities that are required of university neurologists (including teaching, research, and administration) and the likelihood that many of the “other neurologists” at universities specialized in care for patients with conditions other than MS.

Several attitudes toward providing MS patient care were also associated with numbers of MS patients seen among the “other neurologists.” Respondents who indicated that they lack sufficient MS knowledge or seldom encounter MS patients saw fewer MS patients. These attitudes may be linked; that is, physicians who feel they lack sufficient knowledge may be less likely to have MS patients referred to them. Among physicians who feel they lack sufficient knowledge regarding MS care, increasing knowledge may increase their interest and willingness to see MS patients. In contrast, those who indicated the ability to improve MS patient outcomes and quality of life, considered MS a dynamic area with evolving treatment options, or indicated that they enjoy interacting with MS patients were more likely to see more MS patients. These attitudes likely reflect better knowledge regarding MS care. For example, given the recent (and expected future) development of new MS disease modifying therapies [1, 4], this is clearly a dynamic area with the ability to improve patient outcomes. Knowledge of these newer treatment options as well as familiarity with the benefits of multidisciplinary approaches to MS care will likely increase physician interest and enjoyment in providing care to MS patients. Educational programs to increase neurologist familiarity with MS care may strengthen these positive attitudes and increase access to care for MS patients. Improving general neurologists’ knowledge of new treatment options may also enhance MS patient care, as general neurologists may be less likely to discuss newer treatment options than are MS subspecialists [5].

Surprisingly, other neurologists who practice in small cities or rural areas were marginally significantly more likely to see more MS patients than were those practicing in major cities. This appears to contradict previous research findings that MS patients in rural areas have decreased access to care and are more likely to choose other providers (e.g., primary care providers) to direct their care [3, 8]. However, these findings may not be in conflict. MS patients in rural areas likely have few neurologists available; they may therefore be more likely to see neurologists who practice in their local area rather than going to farther away MS subspecialists. Thus, non-MS subspecialist neurologists outside of large urban areas may be more likely to see MS patients as there are no MS centers or MS subspecialists easily available to provide this care.

There are a number of limitations associated with this study. Although the population of neurologists invited to participate in the survey was randomly selected from the relevant members of the AAN, only 39% responded to the survey; this group may not be representative of U.S. neurologists in general. All information provided by survey participants was by self-report; we did not attend to validate any responses. In addition, as with all surveys, we limited the number of items asked to minimize respondent burden. There are likely additional factors that influence the number of MS patients seen by both MS subspecialists and other neurologists that were not captured in our survey. Finally, as our survey data are cross-sectional (i.e., collected at one time point), we are unable to assess causality in the analysis results. For example, we can only determine that an attitude of enjoy interacting of MS patients is associated with seeing more MS patients; we cannot determine whether this attitude preceded initiation of providing MS patient care or whether the attitude was generated based on providing MS patient care (or both). Future studies could involve longitudinal data collection from a cohort of neurology residents, to examine changes in attitudes and subspecialization decisions over time to assess causal relationships in this physician population.

Despite these limitations, this study provides important information on factors affecting numbers of MS patients seen among the neurologist workforce and identifies potential actions to increase access to care for this population.

5. Conclusions

Prior studies have suggested current or impending shortages among neurologists [6]. Among individuals with MS, who may require frequent interactions with neurologists, barriers to access care may affect symptom control, quality of life, and potentially even survival. Our results indicate that several factors are associated with decreased provision of MS patient care among “other neurologists” (i.e., general neurologists and neurologists subspecializing in areas other than MS). These factors include demographic and practice characteristics (female neurologists, those in university-based practices, and those practice in major cities were less likely to provide MS patient care) and attitudes towards MS care. The significant associations of attitudes towards MS care with number of MS patients seen by neurologists suggest that increasing neurologist knowledge about treatments for individuals with MS may decrease barriers to seeing more patients. To ensure optimal health outcomes among individuals with MS, it is critical both to gain better insights regarding factors influencing provision of MS patient care by neurologists and to develop policies that ensure appropriate access to high-quality MS care.

Conflict of Interests

The authors declare that there is no conflict of interests regarding the publication of this paper.

Acknowledgments

The authors would like to thank Dr. Nicholas LaRocca and the National Multiple Sclerosis Society for their funding of this study and providing guidance during the study. The authors would also like to thank Oksana Drogan and the American Academy of Neurology for review, dissemination, and collection of data from the neurologist survey and members of the MS Physician Workforce Study Advisory Panel for their thoughtful comments on earlier versions of this paper. Advisory Panel members include Dr. Corey Ford, Dr. Barbara Giesser, Ms. June Halper, Dr. Shana Johnson, Dr. Aaron Miller, and Dr. Steven Ringel.

References

  1. D. S. Bandari, D. Sternaman, T. Chan, C. R. Prostko, and T. Sapir, “Evaluating risks, costs, and benefits of new and emerging therapies to optimize outcomes in multiple sclerosis,” Journal of Managed Care Pharmacy, vol. 18, no. 9, pp. S3–S15, 2012. View at Google Scholar
  2. R. J. Buchanan, M. Kaufman, L. Zhu, and W. James, “Patient perceptions of multiple sclerosis-related care: comparisons by practice specialty of principal care physician,” NeuroRehabilitation, vol. 23, no. 3, pp. 267–272, 2008. View at Google Scholar · View at Scopus
  3. S. L. Minden, D. C. Hoaglin, L. Hadden, D. Frankel, T. Robbins, and J. Perloff, “Access to and utilization of neurologists by people with multiple sclerosis,” Neurology, vol. 70, no. 13, part 2, pp. 1141–1149, 2008. View at Publisher · View at Google Scholar · View at Scopus
  4. G. Giovannoni and R. W. Rhoades, “Individualizing treatment goals and interventions for people with MS,” Current Opinion in Neurology, vol. 25, pp. S20–S27, 2012. View at Publisher · View at Google Scholar
  5. B. G. Vickrey, Z. V. Edmonds, D. Shatin et al., “General neurologist and subspecialist care for multiple sclerosis: patients' perceptions,” Neurology, vol. 53, no. 6, pp. 1190–1197, 1999. View at Google Scholar · View at Scopus
  6. T. M. Dall, M. V. Storm, R. Chakrabarti et al., “Supply and demand analysis of the current and future US neurology workforce,” Neurology, vol. 81, no. 5, pp. 470–478, 2013. View at Publisher · View at Google Scholar
  7. Multiple Sclerosis Trend Report: Perspectives from Managed Care, Providers, and Patients, Kikaku America International, 2007.
  8. R. J. Buchanan, S. Wang, A. Stuifbergen, B. J. Chakravorty, L. Zhu, and M. Kim, “Urban/rural differences in the use of physician services by people with multiple sclerosis,” NeuroRehabilitation, vol. 21, no. 3, pp. 177–187, 2006. View at Google Scholar · View at Scopus
  9. M. Finlayson, J. D. Garcia, and C. Cho, “Occupational therapy service use among people aging with multiple sclerosis,” American Journal of Occupational Therapy, vol. 62, no. 3, pp. 320–328, 2008. View at Google Scholar · View at Scopus
  10. R. J. Buchanan, M. A. Zuniga, G. Carrillo-Zuniga et al., “Comparisons of latinos, African Americans, and Caucasians with multiple sclerosis,” Ethnicity and Disease, vol. 20, no. 4, pp. 451–457, 2010. View at Google Scholar · View at Scopus
  11. R. J. Buchanan, L. Zhu, R. Schiffer, D. Radin, and W. James, “Rural-urban analyses of health-related quality of life among people with multiple sclerosis,” Journal of Rural Health, vol. 24, no. 3, pp. 244–252, 2008. View at Publisher · View at Google Scholar · View at Scopus
  12. T. D. Shanafelt, S. Boone, L. Tan et al., “Burnout and satisfaction with work-life balance among US physicians relative to the general US population,” JAMA Internal Medicine, vol. 172, no. 18, pp. 1377–1385, 2012. View at Publisher · View at Google Scholar