A Consensus Study: Generational Differences in Neurosurgeons’ Perception of Retirement and Practice Limits

Abstract

Background

Concerns about the changing demographics in the US and the aging of the neurosurgical workforce exist. Both the importance and inherent risk of surgical responsibilities suggest that thought be given to whether workloads should change later in surgeons’ careers.

Objective

To assess current neurosurgeons’ expectations concerning their late-stage careers.

Methods

A survey was sent to 3,317 US board-certified neurosurgeons. It was designed to assess surgeons’ perceptions of call and operative responsibilities in the later stages of their careers. Statistical analyses were completed in R version 3.6.1, with an alpha set to 0.05.

Results

659 neurosurgeons completed the questionnaire. Seventy-seven percent believed that the call burden should decrease later in practice and 66% planned to decrease their own call burden later in their career. The most common age range for planned retirement was 65 to 69 years (36%), followed by 70+ years (33%). Most (67%) believed that there should not be a mandatory age to stop operating. More recent year of residency completion was negatively associated with the belief that call burden should decrease at older age groups, and positively associated with support for a mandatory age to stop operating as well as an earlier retirement age.

Conclusion

This study suggests that neurosurgeons have differing views on how workloads should change later in their careers. Younger neurosurgeons support an earlier decrease in workload or even a policy-mandated stop to operating after a certain age. These results may give insight into future trends and turnover in neurosurgery and provide a valuable tool to help practices anticipate workforce changes.

Introduction

There are few papers published on the topic of retirement in surgery and even fewer about retirement in neurosurgery.^{1,2,3,4,5,6,7,8} The ones that exist begin with Sir William Osler’s famous and controversial 1905 address in which he discusses the “comparative uselessness of men above 40 years of age” and the even more “uselessness of men above 60 years of age”.^9,10,11,12

The discussion of this question often focuses on the negative effect of aging on cognitive and motor function versus the positive gains from experience. Cognitive and motor function are essential for the practice of neurosurgery, however, there is no substitute for experience. Satkunisivam et al. found that increasing surgeon age was associated with decreasing rates of postoperative death, readmission and complication.¹³ There are multiple surveys and studies that demonstrate that most surgeons, regardless of specialty, do not agree with a mandated retirement age.^9,14,15

The attitudes of surgeons toward retirement are of more pressing concern now because of projected shortfalls in the neurosurgical workforce and increasing needs of an aging population.^16,17 If we can understand when surgeons plan to curtail their clinical responsibilities, we can potentially better predict future workforce needs.

The importance of and inherent risk in surgical responsibilities suggest that thought be given to how surgical workloads should change in late career. This study seeks to understand neurosurgeons’ opinions on these issues, as well as their own plans for the later stages of their career. We sought to assess whether these beliefs differ by demographics, type of surgical practice, surgical sub-specialty, and/or how recently residency training was completed. We aim to give insight into the modern neurosurgeon’s attitude toward retirement and infer the effect this may have on future workforce trends and turnover.

Methods

The survey was designed by a team of three neurosurgeons in the Research Electronic Data Capture (REDCap) survey system. It consisted of 10 questions, and is presented in Appendix A. It was pre-tested within the Virginia Commonwealth University Department of Neurosurgery by 12 attending neurosurgeons.

The multi-item questionnaire was sent to 3,317 board-certified neurosurgeons across the United States. All responses were confidential. The questionnaire included items for demographics, and questions about the perceptions of call and operative responsibilities in the later stages of the respondents’ careers. The questions were asked from both a general perspective (e.g., “Do you believe that call burden should decrease later in practice?”), as well as a personal perspective (e.g., “Do you plan on decreasing your personal call burden later in practice?”). The survey was sent out in October 2020, and responses were collected until the end of December 2020.

Summary statistics were calculated using count and frequency for categorical variables and median and interquartile range for numeric variables. Univariate logistic regression models were fit for each of the covariates for the survey questions with binary responses (yes or no). Then, ordinal logistic regression models were fit to the related question (e.g., “At what age do you believe that call burden should decrease later in practice?”) for those who answered “yes” to the binary question, using any covariates significant in univariate analyses for the binary question. Given the strong correlation between surgeon age and when they completed residency training, only one of these variables was included in the ordinal regression models if they were both significant in univariate analyses. Summary statistics for the ordinal regression models were reported, including model coefficients, standard errors, t-values, and p-values. Odds ratios and confidence intervals were calculated and presented for significant regression coefficients.

Two secondary analyses were performed regarding the relationships between multiple survey responses. The first examined the association between surgeons’ belief that a mandatory retirement age should exist in their practice and their own planned retirement age. The second examined the association between surgeons’ plan to decrease their own call burden later in their career and the age at which they believed other surgeons should decrease their call burdens.

All analyses were completed in R, version 3.6.1, with an alpha set to 0.05 to designate statistical significance.

Results

A total of 659 neurosurgeons completed the questionnaire—a 20% response rate (659/3317). Relevant demographics are summarized in Table 1. The most common age group was 50 to 59 years (36%). 85% of the surgeons were male. The most common practice types were private (44%) and academic (30%). The most common sub-specialty was general neurosurgery (41%) followed by spine (20%). The median year of completed residency training was 1999 with interquartile range [1992, 2007].

Table 1.

Respondent Demographics

Variable	Summary
Age a
30 to 39 years	33 (5%)
40 to 49 years	157 (24%)
50 to 59 years	234 (36%)
60 to 69 years	170 (26%)
70 to 79 years	28 (4%)
80 to 89 years	1 (0%)
Decline to answer	36 (5%)
Gender a
Female	60 (9%)
Male	561 (85%)
Decline to answer	38 (6%)
Practice Type a
Academic	197 (30%)
Private	288 (44%)
Combination	94 (14%)
Other	55 (8%)
Decline to answer	25 (4%)
Sub-specialty a
Cerebrovascular	49 (7%)
Functional/Epilepsy	36 (6%)
General Neurosurgery	269 (41%)
Neuro-Oncology	51 (8%)
Other	13 (2%)
Pediatrics	60 (9%)
Peripheral Nerve	2 (0%)
Spine	134 (20%)
Decline to answer	45 (7%)
Year Completed Residency b	1999 [1992, 2007]

^aCategorical variables summarized with count and frequency

^bNumeric variables summarized with median [inter-quartile range]

Responses to survey questions are summarized in Table 2. Most surgeons (77%) believed that call burden should decrease later in practice and 66% planned to decrease their own call burden later in their career. Fewer surgeons (58%) planned to decrease their operative burden later in their career. Only 23% planned to stop operating but continue to maintain a clinical practice later in their career. The most common age range for planned retirement was 65 to 69 years (36%) followed by after 70 years (33%). Most surgeons (67%) believed that there should not be a mandatory age to stop operating in their field.

Table 2.

Summary of responses to retirement-related questions.

Question	Count (Frequency)
Do you believe that call burden should decrease later in practice?
Yes	509 (77%)
No	121 (19%)
Decline to answer	29 (4%)
Do you think neurosurgeons should have a mandatory age to stop operating?
Yes	180 (27%)
No	443 (67%)
Decline to answer	36 (5%)
Do you plan to decrease your call burden later in your career?
Yes	438 (66%)
No	191 (29%)
Decline to answer	30 (5%)
Do you plan to decrease your operative burden later in your career?
Yes	381 (58%)
No	245 (37%)
Decline to answer	33 (5%)
Do you plan to stop operating but maintain clinical practice later in your career?
Yes	150 (23%)
No	477 (72%)
Decline to answer	32 (5%)
When do you plan to retire?
Before 55 years	11 (2%)
55 to 59 years	38 (6%)
60 to 64 years	123 (19 %)
65 to 69 years	234 (36%)
After 70 years	217 (33%)
Decline to answer	36 (5%)

Surgeons did not differ significantly with respect to practice type (private vs. academic) in their responses to any of the main questions about retirement and practice limits.

Surgeons differed by age group with respect to their support for a mandatory age to stop operating (p = 0.01). Significantly more surgeons in the 40- to 49-year age group supported a mandatory age than expected by chance, supporting the notion that younger surgeons tend to support this concept more than older surgeons. There was also a significant association between age group and planned retirement age (p < 0.001). Significantly more surgeons in the 30- to 39-year age group planned to retire between 60 and 64 years, and more surgeons in the 40- to 49-year age group planned to retire between 55 and 59 years than expected by chance. Additionally, fewer surgeons in the 40- to 49-year age group planned to retire after 70 years than expected.

A summary of the univariate analyses performed for dichotomous survey responses is given in Table 3. The year of completed residency was negatively associated with belief that call burden should decrease later in practice, and positively associated with support for a mandatory age to stop operating and with an earlier planned retirement age. Younger surgeons (aged 40 to 59) were more likely to support a mandatory age to stop operating, and older surgeons (aged 60+) were less likely to support such a requirement. Practice sub-specialty was not significantly associated with belief that call burden should decrease later in practice, support of a mandatory age to stop operating, and personal plan to decrease call burden later in career.

Table 3.

P-values for covariates in univariate analyses.

Covariate	Age	Gender	Practice	Subspecialty	Year Completed Residency
Do you believe that call burden should decrease later in practice?	0.431	0.982	0.253	<0.001*	0.016*
Do you think neurosurgeons should have a mandatory age to stop operating?	0.013*	0.086	0.259	0.032*	0.003*
Do you plan to decrease your call burden later in your career?	0.234	0.976	0.465	<0.001*	0.631
Do you plan to decrease your operative burden later in your career?	0.506	0.300	0.559	0.307	0.653
Do you plan to stop operating but maintain clinical practice later in your career?	0.155	0.730	0.127	0.064	0.081
When do you plan to retire?	<0.001*	<0.001*	0.085	0.112	<0.001*

Tables 4, 5, and 6 display output from ordinal regression models, which include the significant covariates from the univariate analysis. Table 4 includes those who believed that call burden should decrease later in practice. More recent graduates were significantly less likely (OR = 0.910, 95% CI (0.892, 0.928)) to believe in an older age group for planned call burden decrease.

Table 4.

Ordinal logistic regression model for “At what age should call burden decrease in practice?”

Covariate	Value	SE	T-Value	P-Value
Subspecialty: Functional/Epilepsy	−0.235	0.475	−0.495	0.621
Subspecialty: General Neurosurgery	−0.163	0.370	−0.440	0.660
Subspecialty: Neuro-Oncology	0.491	0.445	1.102	0.270
Subspecialty: Other	0.664	0.724	0.917	0.359
Subspecialty: Pediatrics	0.470	0.449	1.046	0.296
Subspecialty: Peripheral Nerve	0.037	1.218	0.030	0.976
Subspecialty: Spine	−0.353	0.388	−0.910	0.363
Year Completed Residency	−0.094	0.010	−9.327	<0.001*

Table 5.

Ordinal logistic regression model for “Do you think neurosurgeons should have a mandatory age to stop operating?”

Covariate	Value	SE	T-Value	P-Value
Subspecialty: Functional/Epilepsy	1.123	1.099	1.022	0.307
Subspecialty: General Neurosurgery	0.669	0.631	1.060	0.289
Subspecialty: Neuro-Oncology	0.656	0.789	0.831	0.406
Subspecialty: Other	4.107	1.484	2.767	0.006*
Subspecialty: Pediatrics	0.539	0.711	0.758	0.448
Subspecialty: Spine	0.682	0.635	1.075	0.282
Year Completed Residency	−0.043	0.018	−2.466	0.014*

Table 6.

Ordinal logistic regression model for “When do you plan to retire?”

Covariate	Value	SE	T-Value	P-Value
Gender: Male	0.547	0.265	2.063	0.039*
Year Completed Residency	−0.094	0.009	−10.401	<0.001*

Table 5 includes the covariates to the question answer of whether a mandatory age to stop operating should exist. More recent graduates were significantly less likely (OR = 0.958, 95% CI (0.925, 0.991)) to support an older age group for a mandatory stop to operating.

Table 6 shows more recent graduates were significantly less likely (OR = 0.911, 95% CI (0.894, 0.927)) to plan to retire in an older age group. Broadly, those who more recently completed their training were more likely to support decreases to responsibilities later in neurosurgeons’ careers, and in some cases, an age after which operating was prohibited. Other ordinal regression models fit did not show any significant associations between covariates and outcomes and are not presented.

Plots of predicted probabilities of responses for each of these questions and models are shown in Figures 1-3. Figure 1 demonstrates that a neurosurgeon who completed residency in 1990 was approximately 30 percent likely to support a decrease to call burden by age group 55-59 or before, however a neurosurgeon who completed residency in 2010 was approximately 80 percent likely to support such a decrease by the same age range.

Figure 1. Predicted probability plot for “At what age should call burden decrease in practice?”.

Note: In Figure 1, the model fixes sub-specialty as General Neurosurgeon in the model predicted probabilities, since that is the most common response in the sample.

Figure 3. Predicted probability plot for “When do you plan to retire?”.

Note: In Figure 3, the model fixes gender as Male in the model predicted probabilities, since that is the most common gender in the sample

In Figure 2, respondents who completed their residency in 1990 and in 2010 were approximately 50 and 75 percent likely, respectively, to support an age of 70 to be one after which neurosurgeons should not be allowed to operate.

Figure 2. Predicted probability plot for “Do you think neurosurgeons should have a mandatory age to stop operating?”.

Note: In Figure 2, the model fixes sub-specialty as General Neurosurgeon in the model predicted probabilities, since that is the most common response in the sample.

In Figure 3, respondents who completed their residency in these years were approximately 10% and 40% percent likely, respectively, to state that they planned to retire between 60-64 years of age.

Secondary analyses examined the association between responses to multiple survey questions. First, support for a mandatory age to stop operating was associated with a significantly lower planned age group for retirement (OR = 0.415, 95% CI (0.301, 0.572), p < 0.0001). Second, those surgeons with plans to decrease their own call burden later in their career were significantly less likely (OR = 0.568, 95% CI (0.370, 0.877), p = 0.0107) to believe in later age ranges to decrease call burden. These results suggest consistency between surgeons’ beliefs for others and themselves.

Discussion

This study is unique in its focus and scale. No other report has assessed neurosurgeons’ perceptions and beliefs regarding limited practice and retirement and compared these to their own intentions, nor has any done so with such a large number of respondents. The most clear conclusion that emerges from our analysis is that those who completed training more recently were more likely to support decreases to operative and call responsibilities in late career, and more likely to support an age after which operating is prohibited. Younger respondents were more likely to support reducing call burden late into their career. They were more likely to support an earlier mandatory age cutoff to stop operating and more likely to plan to retire earlier in their career.

It is difficult to determine whether these findings are due to a true generational effect or if they stem from distance from retirement, in which case this attitude may change as these surgeons age. If it is due to the former, it may reflect that work-life balance concerns are influencing surgeons’ responses. There are significant population generational differences and young adults value lifestyle and quality of life over volunteerism out of a sense of duty. ¹⁸ The responses of the neurosurgeons who have been in practice longer may suggest they are not concerned with the aging effects on cognitive and motor function skills or that experience outweighs these concerns, and therefore they may not entertain positions or policies that truncate their careers. In addition, the majority of the respondents were male, which reflects the current state of the field; however, as more women continue to join the neurosurgical workforce, there may be an evolution of these results.

Our survey also found that neurosurgeons have internal consistency between their general beliefs about retirement and limiting practice, and their own personal plans in the later phases of their careers. Generally, if they believed in an earlier age for a decrease in some workplace responsibility, then they also planned to decrease their responsibility at an earlier age. There were few significant or systemic patterns between other covariates such as practice type and subspecialty training suggesting survey responses did not vary significantly between neurosurgeons in academic and private practice or between specific subspecialties.

The only other publication which deals with this subject was written by Babu et al. Their survey found broad support for assessing the medical knowledge and case logs of older neurosurgeons, but two-thirds rejected a cutoff mandatory retirement age.¹⁴ This finding is exactly mirrored in our study, and we build upon these earlier findings by assessing actual plans regarding late career practice and retirement. A separate consideration for senior partners is that there are physical and cognitive tests, like the Aging Surgeon Program, that exist to assess aging physician competence.⁹ These may be used to judge senior physicians in the future.

What do these findings mean for the neurosurgical workforce? The Transamerica Center for Retirement Studies in a 2019 survey showed that the average retirement age for the general population was 61, but more than half (54%) of workers plan to keep working past the age of 65.¹⁹ If younger neurosurgeons who more recently finished residency are planning on practicing for fewer years than their predecessors, this could exacerbate the already identified shortfall in the neurosurgical manpower.¹⁶ There have been calls for expanding residency spots to accommodate this perceived need and our findings may lend urgency to this.

There were several limitations to this study. First, though the survey was sent to a large group of board-certified surgeons practicing nationally, results could have been affected by selection bias. Our methodology did not allow us to assess whether non-respondents would have answered differently than those who chose to participate.

Further, only a few demographic questions were included in the survey, meaning that other factors potentially influencing neurosurgeons’ beliefs about end-of-career responsibilities were not included. The response rate is a further indicator of possible selection bias. Additionally, it is possible that neurosurgeons later in their career responded at a higher rate because of more interest in the topic. Also, the survey did not include an evaluation of differences in incentives for call. In the past, early-career call was a way of building a practice and junior partners desired more call. With more recent changes in payments and regulations, perspectives may have changed. Whether or not a physician is paid for call and how much the corresponding compensation is may play a role in their opinion. Overall career satisfaction may also have influenced the responses and this was not evaluated in the survey.

This topic warrants future investigation. Opinions can and do change throughout the course of a career, particularly as one accrues more experience and understanding of the health care system. It would be interesting to collect follow-up data five to ten years later to assess changes in opinion.

This study offers evidence to suggest that younger and older neurosurgeons have differing views on how their workloads should change as they age. Younger neurosurgeons endorse an earlier decrease, or even a potentially policy-mandated stop to operating after a certain age. Attention should be paid to these generational differences when assembling teams of neurosurgeons and projecting workforce needs for the future. Given the increase in the aging population and the increased neurosurgical burden this entails, the results of our survey suggest junior faculty and younger partners will face an increasingly burdensome clinical load that they may find undesirable later in their careers.

Conclusions

This is a unique and large survey of the attitudes of practicing neurosurgeons toward limiting practice as they age and retirement. Our results show that most neurosurgeons plan on decreasing their call burden in their late 50s and then plan on retiring in their late 60s. In concordance with previous studies, most do not agree with a mandatory retirement age. This study offers evidence to suggest that younger and older neurosurgeons have differing views on how their workloads should change later in their careers, with younger neurosurgeons suggesting an earlier decrease, or even a potentially policy-mandated stop to operating after a certain age.

Table 2a.

Summary of responses to retirement-related questions, stratified by practice type.

Question	Academic	Private	P-Value
Do you believe that call burden should decrease later in practice?			0.69
Yes	162 (83%)	229 (81%)
No	33 (17%)	53 (19%)
Do you think neurosurgeons should have a mandatory age to stop operating?			0.13
Yes	65 (33%)	74 (27%)
No	130 (67%)	205 (73%)
Do you plan to decrease your call burden later in your career?			0.63
Yes	141 (72%)	197 (70%)
No	54 (28%)	85 (30%)
Do you plan to decrease your operative burden later in your career?			0.44
Yes	122 (63%)	167 (59%)
No	71 (37%)	115 (41%)
Do you plan to stop operating but maintain clinical practice later in your career?			0.07
Yes	53 (27%)	56 (20%)
No	141 (73%)	225 (80%)
When do you plan to retire?			0.27
Before 55 years	2 (1%)	6 (2%)
55 to 59 years	10 (5%)	20 (7%)
60 to 64 years	32 (17%)	62 (22%)
65 to 69 years	78 (40%)	109 (39%)
After 70 years	72 (37%)	84 (30%)

Quantities summarized with count and frequency. P-value is from chi-squared test of independence between covariate (practice type) and response to question.

Table 2b.

Summary of responses to retirement-related questions, stratified by age group.

Question	30-39	40-49	50-59	60-69	70-79	P-Value
Do you believe that call burden should decrease later in practice?						0.35
Yes	25 (76%)	119 (77%)	185 (81%)	142 (85%)	23 (85%)
No	8 (24%)	36 (23%)	44 (19%)	25 (15%)	4 (15%)
Do you think neurosurgeons should have a mandatory age to stop operating?						0.01*
Yes	9 (27%)	60 (39%)	63 (28%)	35 (21%)	5 (18%)
No	24 (73%)	95 (61%)	165 (72%)	130 (79%)	23 (82%)
Do you plan to decrease your call burden later in your career?						0.36
Yes	26 (79%)	114 (74%)	151 (66%)	115 (69%)	21 (75%)
No	7 (21%)	41 (26%)	78 (34%)	52 (31%)	7 (25%)
Do you plan to decrease your operative burden later in your career?						0.66
Yes	19 (58%)	94 (61%)	132 (58%)	107 (64%)	18 (67%)
No	14 (42%)	60 (39%)	97 (42%)	59 (36%)	9 (33%)
Do you plan to stop operating but maintain clinical practice later in your career?						0.15
Yes	8 (24%)	29 (19%)	48 (21%)	50 (30%)	7 (25%)
No	25 (76%)	125 (81%)	181 (79%)	116 (70%)	21 (75%)
When do you plan to retire?						< 0.001*
Before 55 years	1 (3%)	7 (5%)	1 (1%)	0 (0%)	0 (0%)
55 to 59 years	2 (6%)	23 (15%)	13 (6%)	0 (0%)	0 (0%)
60 to 64 years	12 (36%)	35 (23%)	57 (24%)	17 (10%)	0 (0%)
65 to 69 years	13 (39%)	66 (42%)	90 (40%)	59 (35%)	0 (0%)
After 70 years	5 (15%)	24 (15%)	66 (29%)	91 (55%)	28 (100%)

Quantities summarized with count and frequency. P-value is from chi-squared test of independence between covariate (age group) and response to question.

^*Significant on p < 0.05 level.

Appendix

Appendix A –. Full survey

Neurosurgeons’ Perception of Call and Operative Responsibilities

Thank you for participating in our survey regarding Neurosurgeons’ Perception of Call and Operative Responsibilities. Our goal is to accumulate data on Neurosurgeons’ perception about how much call and operative responsibilities should remain a significant role for late career neurosurgeons.

Please complete the 10 question survey below. It will take 5 minutes or less to complete Thank you!

1. What is your current age?

   30 to 39

   40 to 49

   50 to 59

   60 to 69

   70 to 79

   80 to 89

2. Gender

   Male

   Female

   Decline to answer

3. How would you describe your practice?

   Private

   Academic

   Combination

   Other

4. What year did you complete residency training? Fill in

5. Do you believe that call burden should decrease later in practice?

   Yes

   No

   If yes was chosen then a drop down question as follows:

   • At what age should it decrease?

   • Less than 55

   • 55 to 59

   • 60 to 64

   • 65 to 69

   • Greater than 70

6. Do you plan on decreasing your personal call burden later in practice?

   Yes

   No

   If yes was chosen then a drop down question as follows:

   • At what age will you decrease call burden?

   • Less than 55

   • 55 to 59

   • 60 to 64

   • 65 to 69

   • Greater than 70

7. Do you plan on decreasing your operative case load later in practice?

   Yes

   No

   If yes was chosen then a drop down question as follows:

   • At what age do you plan on decreasing your operative case load?

   • Less than 55

   • 55 to 59

   • 60 to 64

   • 65 to 69

   • Greater than 70

8. Do you plan on stopping operative cases but still continuing clinical practice later in your career?

   Yes

   No

   If yes was chosen then a drop down question as follows:

   • At what age do you think you will stop operating?

   • Less than 55

   • 55 to 59

   • 60 to 64

   • 65 to 69

   • Greater than 70

9. At what age do you plan on retiring from practice?

   Less than 55

   55 to 59

   60 to 64

   65 to 69

   Greater than 70

10. Do you think neurosurgeons, like airline pilots, should have a mandatory ago to stop operating?

    Yes

    No

    If yes was chosen then a drop down question as follows:

    • What age of mandatory retirement from operating would you propose?

    • 65

    • 70

    • 75

    • Greater than 75