Definitions and Statistical Analysis

Descriptive summary statistics were used to describe the overall group of respondents. Select survey responses were combined into categories for the purposes of statistical analysis based on previous literature and number of responses within the original selections. All categories were determined prior to completing any statistical analysis. For example, responses to specific questions regarding professional perceptions that were believed to have a large impact on the desired outcome were converted from a Likert-type scale to dichotomous variables. The number of age categories was collapsed from 7 to 3 to approximate the early (up to and including age 45 years), middle (46–55 years), and later (> 55 years) stages of a person's career, with each stage selected to represent an approximate 10-year period.

We decided a priori to identify and quantify predictors of 2 survey outcomes: 1) burnout and 2) career satisfaction. Burnout was measured using the validated Maslach Burnout Inventory (MBI) that assesses emotional exhaustion (9 questions), depersonalization (5 questions), and personal accomplishment (8 questions). Using the same definition as Shanafelt et al. and the cutoff values specific for those who work in medicine as set forth in the MBI manual, burnout was defined as having high scores for emotional exhaustion or depersonalization on the MBI.^30,37 Respondents were classified as being satisfied with their career if they selected “very satisfied” or “somewhat satisfied” on Question 1 of the professional satisfaction section of the survey. Subgroup analysis evaluating the same outcomes in respondents who described their job as “academic” versus those who did not was also performed.

After conducting univariate analysis, forward multivariable logistic regression was undertaken to assess independent associations related to the dichotomous variables of burnout and career satisfaction. Variables with p < 0.2 in the univariate analysis were placed in the logistic regression model initially. The same methodology was used for the subgroup analysis of academic and nonacademic neurosurgeons. Two-tailed statistical tests were performed, and p ≤ 0.05 was determined to represent statistical significance in the univariate and multivariable analysis. The level of statistical significance for multiple comparisons was not adjusted. Results were reported as adjusted odds ratio with corresponding 95% confidence intervals. All data were analyzed using SPSS software (IBM SPSS Statistics for Windows, Version 21.0, Released 2012, IBM Corp.).

Survey Results

Of the 3247 US neurosurgeons invited to complete the survey, 783 responded (24%). Responses from 33 physicians were excluded due to incompleteness on the 22-question MBI, yielding 750 completed surveys.

The majority of respondents were male, between the ages of 40 and 60 years, in a stable relationship with children, and working in a group or university practice with subspecialty training. Most worked more than 50 hours per week, took call 6 or more days per month, and completed over 200 cases annually. Just over half indicated that they had not been involved in any malpractice suits over the last 5 years (2008–2012). A detailed summary of personal and practice characteristics of the 750 survey participants is provided in Table 1.

A breakdown of responses for the survey sections on professional stressors, satisfaction, and quality of life is shown in Figs. 1–3. Too much call (59.2%), uncertainty about future earnings/health care reform (75.2%), inadequate administration time (46.9%), low salary (48.8%), and too little vacation time (43.7%) had at least a moderate impact on neurosurgeons' professional careers (Fig. 1). Participants were satisfied with the location of their practice (75.1%) and their neurosurgical colleagues (65.1%), but were dissatisfied with the time allowed for personal growth/development (44.6%) and believed that reimbursement should be improved (69.6%; Figs. 2 and 3).

Response to questions for Professional Stressors (Appendix, Section 2). Figure is available in color online only.

Response to questions for Professional Satisfaction (Appendix, Section 3). Figure is available in color online only.

Response to questions for Quality of Professional Life (Appendix, Section 4). Figure is available in color online only.

Selected questions from each of the 3 sections—professional stressors, satisfaction, and quality of life—and from the general questions section of the survey are shown in Table 2. More than 80% cited that they were satisfied with their career as a neurosurgeon. Strong responses by participants included being intellectually challenged by their work (77.3%), having a spouse that understood their demanding work schedule (73.5%), and being concerned about future earnings and health care reform (53.5%). Only 36.5% believed that they had adequate time for personal growth/development, and 48.1% had achieved a balance between work and life. Although almost 70% of respondents would choose a career in neurosurgery again, only 35.5% would recommend neurosurgery as a career to their children.

When asked if they felt their professional life would improve, only 26% of neurosurgeons agreed. Moreover, when asked if they felt their professional life would worsen, 52% agreed, whereas only 24.8% did not. The direction neurosurgeons believe their careers will take in the future is depicted graphically in Fig. 4.

Perceptions among neurosurgeons on the future direction of their career. Figure is available in color online only.

The burnout rate overall was 56.7% (Table 3). High emotional exhaustion was reported in 35.1% of neurosurgeons, high depersonalization in 31.3%, and a low sense of personal accomplishment in 28.4%.

Predictors of Satisfaction and Burnout

The results of univariate analysis are shown in Table 4. Factors that were independently associated with career satisfaction and burnout are listed in Table 5. Six factors were predictive of burnout and 7 for career satisfaction. Being accused of malpractice (OR 1.6) and having uncertainty regarding future earnings/health care reform (OR 1.96) significantly increased the odds of being burned out, while 4 other practice characteristics or perceptions were protective. Four of the 7 factors independently associated with career satisfaction were positive correlations, including having children (OR 2.4) and a good work/life balance (OR 10.0). Achieving a balance between work and life and anxiety over future earnings/health care reform were factors shown to significantly impact the odds of both outcomes. The presence of burnout reduced the odds of being satisfied with one's career almost 6-fold (OR 0.17).

Academic Versus Nonacademic Neurosurgeons

Characteristics of academic and nonacademic neurosurgeons are shown in Table 6. Variables that were independently associated with career satisfaction and burnout among academic and nonacademic neurosurgeons are shown in Tables 7 and 8, respectively.

Study Findings

Neurosurgery has been—and will remain—a stressful yet rewarding profession. This survey study represents the most comprehensive attempt to identify and quantify sources of satisfaction and burnout in neurosurgeons. Among all survey participants, factors shown to independently increase the odds of being satisfied with a career in neurosurgery included being surgically productive, having children, being intellectually stimulated, and achieving balance between career and life outside of work. The burnout rate overall was 56.7% (35.1% emotional exhaustion and 31.3% depersonalization), which is one of the highest reported rates among surgical^{8,25,33,37,44,45} and nonsurgical specialties,^{10,13,14,26,27,31,32,40} including those in previously published reports for neurosurgeons (Table 9).^7,23,37 Burnout was associated with malpractice litigation and uncertainty regarding future earnings and health care reform, which was also shown to decrease the odds of being satisfied with neurosurgery.

Burnout

The study of burnout and career satisfaction is immensely important in today's health care system with effects that permeate many different aspects of health care delivery. Physician burnout has been shown to lead to medical errors,^{19,36,39,47,48} early retirement,^11,22,43 diminished patient satisfaction with care,^20,28 and increased malpractice suits.²¹ Surgeons from every specialty are at very high risk for burnout.^7,37 The current study and prior pilot study are the only studies, to our knowledge, that have been designed specifically to garner feedback from neurosurgeons. Despite a high burnout rate among the neurosurgeons who participated in the survey, the overall career satisfaction response rate was 81.2%. Similar studies have found this seemingly paradoxical finding,^7,25 which supports the innate subjectivity of burnout assessment. Burnout can occur in transient or periodic episodes, thus causing physicians to suffer from burnout at some point in their career while still being satisfied overall. Interestingly, a physician's perception of his or her own life seems to play a significant role in the relative impact of burnout on career satisfaction.

Sources of Satisfaction and Dissatisfaction

In almost all models for satisfaction and burnout, neurosurgeons have indicated that they enjoy being challenged at work. The data presented suggest that neurosurgeons are content when they are in the early-to-middle portion of their career, are consistently challenged, and spend considerable time in the operating room. In a previous cross-sectional study, Dyrbye and colleagues examined how the stage in a physician's career affects burnout and satisfaction.¹⁷ Physicians from all specialties in the US were analyzed and showed that the lowest satisfaction occurs at 2 times during their careers. First, during the early stages of their careers, physicians were dissatisfied with their overall career choice (that is, being a physician) because of work/home conflicts and feelings of depersonalization. Second, during the midstages of their careers, they were dissatisfied with their specialty career choice because of excessive work demands (that is, working long hours and taking overnight call).¹⁷ This finding is contrary to data in the current study. We hypothesize that the typical neurosurgeon experiences excitement and energy in practicing independently during the early part of their career and then may achieve a “professional peak” in terms of skill level, experience, or impact during the middle portion of their career. It is reasonable to infer that the latter stage of their career may be negatively affected by greater administrative duties, less operating room time, and increased physical fatigue or personal health issues. Such challenges may contribute to the pessimistic and unfavorable outlook held by some neurosurgeons regarding their future career (Fig. 4). Age, by itself, was not predictive of burnout or achieving career satisfaction.

It is equally clear that neurosurgeons desire a good work/life balance. The professional perception of a good work/life balance was significant in all multivariable analyses. The long hours and hard work required of a neurosurgeon make it difficult to achieve a satisfactory balance. Numerous studies have addressed the impact of work/home conflicts on medical professionals and have shown that physicians with career and family imbalance, recent work/home conflicts, and recent conflicts resolved in favor of work typically have higher rates of burnout.^11,16,18 However, if a neurosurgeon perceives a good balance between work and personal life, then career satisfaction may be higher and risk of burnout may be lower.

Neurosurgeons, as with all health care workers, have been impacted in recent years by health care reform. As new measures of the Patient Protection and Affordable Care Act (PPACA) continue to be unveiled, uncertainty about the impact of health care reform may have adverse effects on the well-being of physicians. It has been suggested that health care reform poses an increased risk for burnout by augmenting physician workload, decreasing autonomy, and potentially disrupting physician work/life balance.¹⁵ The economic implications of health care reform on a neurosurgeon's practice should also be considered as a source of anxiety. To our knowledge, no previous study has assessed this issue despite the potential for changes in the US health care system to exacerbate burnout. Our data revealed that career satisfaction among US neurosurgeons is strongly impacted by the uncertainty of future earnings and health care reform. More importantly, those who feel this uncertainty in their practice have an increased risk for burnout.

Medical malpractice was another important characteristic associated with burnout. Malpractice, which was not addressed by the PPACA, has been shown to strongly influence physician burnout, depression, and suicidal ideation among surgeons.³ Physicians who had experienced medical malpractice were more likely to score a higher burnout level in all 3 domains assessed by the MBI.¹² Even though less than half of the polled neurosurgeons were involved in malpractice suits over a 5-year period (2008–2012), malpractice was still found to be associated with an increased risk of burnout and can affect patient care.

Academic Versus Nonacademic Neurosurgeons

Another intriguing aspect of the present study was our comparison of academic and nonacademic neurosurgeons. Many of the predictors of satisfaction and burnout were shared between the 2 groups. Both were more likely to be satisfied if they felt challenged at work and perceived themselves as having a good work/life balance. They were less likely to be satisfied if they were considered burned out. Academic neurosurgery had a lower rate of burnout than nonacademic neurosurgery (47.7% vs 62.9%). These results were similar to those in previous studies in which private-practice general surgeons and oncologists had a greater burnout risk and lower career satisfaction rate.^7,40 Previous research has suggested that nonacademic surgeons may be at greater risk because of the higher average number of call days, the longer hours worked per week, or earnings based entirely on billing.³⁷ In contrast to their academic colleagues, who are more likely to be salaried, nonacademic neurosurgeons in our study also had lower levels of career satisfaction when they reported uncertainty about future earnings and health care reform.

Interestingly, academic neurosurgeons were less likely to be satisfied with their careers if they had practiced as a neurosurgeon for more than 20 years. This finding suggests that while burnout may be an overall problem for nonacademic neurosurgeons, it may be a bigger problem for more senior academicians. Academia presents unique stressors, such as administrative and teaching responsibilities, tenure applications, and research requirements. These additional obligations may promote burnout as the number of years that such stressors are experienced increases and by leaving inadequate time for personal development, which, as found in our study, decreases satisfaction. It is also possible that more specialized academic neurosurgeons may be at an increased risk of burnout given the impact of repetition and decreased operative variability.⁴⁰

Study Limitations

Our response rate of 24% was reasonable but on the lower end of the spectrum compared with the rates in other studies listed in Table 9. We did not offer any incentive(s) to participate as others have, although doing so may have improved our response rate. This study may suffer from selection bias whereby individuals who completed the survey may be different as a group compared with those who did not complete the survey. For example, those who currently suffer from burnout or career dissatisfaction may be less likely to take the time to complete a voluntary survey, which would underestimate the prevalence of both outcomes, or vice versa.²³ Another example of selection bias relates to the results of our earlier pilot study. That study had 85 respondents, most of whom were in academics or had an academic affiliation (73%). The senior author (P.K.) personally asked many of them to participate in the pilot study, resulting in a higher response rate (50%). On virtually all measures, the pilot study respondents seemed to be a “happier” group than the 750 neurosurgeons in the present study (which includes the 85 in the pilot study). For example, the neurosurgeons in the pilot study had a lower burnout rate (27% vs 57%), higher overall satisfaction (96% vs 81.2%), greater work/life balance (74% vs 48.1), and greater likelihood of choosing neurosurgery again as a career (88% vs 68.3%). While results of the present study are internally valid, the results may not be completely generalizable to surgeons who did not complete the survey. Nonetheless, we think that the 750 survey respondents in the current study are probably a more representative sample of US neurosurgeons than the 85 participants in the pilot study, and thus the full sample provides a more accurate assessment of burnout and job satisfaction.

Neurosurgeons are constantly asked to participate in various surveys, which leads to apathy or “survey fatigue,” making it difficult to conduct such studies. The results may not be applicable to neurosurgeons outside the US, although it is reasonable to assume that there is some degree of universality in terms of sources of stress and satisfaction among all neurosurgeons, regardless of practice location. As with any survey study, some of the questions may not have been clearly understood by all participants. For example, the divorce rate of the neurosurgeons sampled was only 9.1%. This is significantly lower than previously published rates, suggesting that miscommunication probably occurred at some level of survey administration.^16,37 Shortening the survey was necessary to improve our response rate; however, doing so resulted in less information in comparison with that in the pilot study. Nonetheless, we believe that the shortened survey was able to capture the important aspects of a career that have direct relevance on satisfaction, stress, and burnout.

Future Studies

To safely and reliably provide for the needs of the increasing patient population in the US, it is essential that individual and organizational interventions be created to diminish the potency of potential stressors in all medical specialties that can lead to unhealthy levels of stress and burnout.^29,41,42 Such programs could be tailored to practice types (for example, academic vs private vs hospital based) based on research that would further clarify predictors of job satisfaction and burnout. Some researchers have already begun work on programs to reduce stress and burnout.^24,46 An intervention involving facilitated physician discussion groups around mindfulness, reflection, shared experience, and small-group learning has shown increased empowerment and engagement at work, decreased rates of depersonalization, and decreased overall burnout among participating physicians.⁴⁶ Another intervention that offered an educational course on mindfulness meditation, self-awareness, and meaningful clinical narratives was shown to improve scores for mindfulness, burnout, depersonalization, personal accomplishment, and empathy.²⁴ Finally, it would be interesting to repeat this study in the future to ascertain the full impact of health care reform.