The Interplay of Parental Leave,Career Achievement,and Burnout in Academic Neurologists: A Multicenter Study

Study design

This is a sub-analysis from a parent multicenter study titled “Women in Neurology Collaborative Study (WINC).” ²³ This analysis focused on examining the relationship between PL, academic productivity, burnout, and career advancement in neurology. Our sample included neurology faculty with children and excluded those who did not report their gender (n = 3) for a total of 239 out of the 431 respondents to the survey of neurology faculty from 19 U.S. academic centers (Supplementary Appendix). ²³ The survey consisted of 99 questions, including 48 with branching logic. PL was defined in the survey as any leave for a parental event, including maternity leave, paternity leave, and adoption. Use of sick leave, short-term disability, or paid time off is not included in the definition or the duration of PL. Survey questions are available in Supplementary Data(Supplementary Appendix). The survey was piloted at two institutions prior to broader dissemination to ensure clarity, usability, and proper functioning of the Research Electronic Data Capture (REDCap) instrument. Feedback during piloting highlighted concerns about confidentiality and fear of retaliation, particularly when discussing equity, PL, and workplace fairness. These concerns informed survey design decisions, including the exclusion of certain potentially identifying variables.

Data were collected and securely managed using REDCap, hosted at the University of Texas Southwestern Medical Center, between February 19, 2021, and April 6, 2022. Adult and pediatric neurology faculty were eligible to participate in the survey. Trainees and advanced practice providers were omitted because the focus was on academic neurologists’ career advancement.

Respondents’ demographic information, marital status, years in profession, and academic track and rank were collected in the survey (Table 1). PL was defined as time off for maternity leave, paternity leave, or child adoption. This definition specifically excluded sick leave, focusing solely on leave taken in relation to the birth or the adoption of a child. This distinction was made to ensure clarity in the type of leave and to align with the study’s focus on parental roles and responsibilities.

OPEN IN VIEWER

Table 1.

Demographic and Sample Characteristics of Survey Respondents by Parental Leave Status

Demographic and work information	Overall	No leave a	Leave a	p-Value
N	239	84	155
Gender, N female (%)	140 (58.6)	24 (28.6)	116 (74.8)	<0.001
Race/Ethnicity, N (%)				0.667
Asian	55 (23.01)	24 (28.6)	31 (20.0)
White	150 (62.76)	49 (58.3)	101 (65.2)
Underrepresented b	19 (7.95)	6 (7.1)	13 (8.4)
Not reported c	12 (5.02)	4 (4.8)	8 (5.2)
Other d	3 (1.3)	1 (1.2)	2 (1.3)
Marital status, N married (%)	226 (94.6)	78 (92.9)	148 (95.5)	0.583
Years in profession	13.0 (8.2)	14.4 (9.6)	12.3 (7.3)	0.077
Academic track, N tenure-track (%)	34 (14.2)	12 (14.3)	22 (14.2)	1.000
Academic rank, N (%)				0.124
Instructor or senior instructor	2 (0.01)	0 (0.0)	2 (1.3)
Assistant professor	117 (48.9)	35 (41.7)	82 (52.9)
Associate professor	76 (31.8)	27 (32.1)	49 (31.6)
Professor	39 (16.3)	19 (22.6)	20 (12.9)
Other	5 (2.1)	3 (3.6)	2 (1.3)
Full-time employee, N yes (%)	229 (95.8)	81 (96.4)	148 (95.5)	1.000
Hours worked per week				0.212
<20 hours/week	1 (<0.01)	0 (0.0)	1 (0.6)
21–40 hours/week	19 (7.9)	7 (8.3)	12 (7.7)
41–60 hours/week	134 (56.1)	41 (48.8)	93 (60.0)
61–80 hours/week	65 (27.2)	25 (29.8)	40 (25.8)
>80 hours/week	20 (8.4)	11 (13.1)	9 (5.8)
Family–Work Conflict Score e range [5–35], higher score = higher conflict	21.1	18.3	22.6	<0.001

Bold data indicates statistically significant (alpha = 0.05).

Data are presented as mean (standard deviation) or frequency (percentage) as appropriate.

a

Three participants responded with “Prefer not to answer” (N = 3).

b

Data for participants who responded with “American Indian/Alaskan Native” (N = 0), “Black or African American” (N = 3), “Native Hawaiian or other Pacific Islander” (N = 2), and “Hispanic/Latino” (N = 14) were combined under “Underrepresented” due to the small numbers.

c

Data for participants who responded with “Not reported” (N = 2) and “Prefer not to say” (N = 10) were combined under “Not reported” due to the small numbers.

d

Includes two White Asian and one Asian and Hispanic.

e

Score is presented as the sum of five questions related to family–work conflict. Individuals who responded with “N/A” for any of the five questions (N = 7) were excluded from this row.

Outcome measures

Academic achievements were measured with two binary measures (whether or not leadership positions were held and whether or not incentivized leadership positions were held) and six quantitative measures (average number of publications in one year, number of first or last author publications over the last 5 years, number of leadership positions held, number of awards received, number of speaking invitations received, and number of funded projects). The survey asked for the average number of weeks for total PL. Then, we grouped responses into three duration ranges for analysis purposes: less than 6 weeks, 7–12 weeks, and more than 12 weeks. Among the 153 respondents who took PL, 27 individuals reported using sick time or vacation time instead of formal institutional PL, which may have been either exhausted or nonexistent at their respective institutions. Well-being was assessed with questions about contentment with PL (Table 2) and the Maslach Burnout Inventory (MBI), which measures burnout along three domains—personal accomplishment, depersonalization, and emotional exhaustion. The Family–Work Conflict Scale assesses how family-related demands and responsibilities affect work roles and obligations. It measures the intensity of this interaction using Likert-scale items, which are designed to capture the level of conflict between family and work domains. Higher scores on the scale indicate greater perceived conflict. The scale uses a 7-point Likert scale for each item, with total scores ranging from 5 to 35; higher scores indicate a stronger perception of conflict between family and work spheres.

OPEN IN VIEWER

Table 2.

Leave Contentment and Duration in Survey Respondents in Parental Leave Group

Contentment and duration questions	Men a	Women a	p-Value
N	37	116
Content n (%)	29 (78.4)	57 (49.1)	0.003
If not content, why? b
I did not have access to paid parental leave	3 (37.5)	15 (25.4)	0.672
I had to take additional unpaid leave after my paid leave was utilized	0 (0.0)	25 (42.4)	0.021
There were no adjustments in productivity during parental leave	2 (25.0)	23 (39.0)	0.700
Institution was not supportive of parental leave	2 (25.0)	13 (22.0)	1.000
Colleagues were not supportive of parental leave	2 (25.0)	10 (16.9)	0.627
Parental leave instructions were not organized or clear	3 (37.5)	24 (40.7)	1.000
Other	2 (25.0)	24 (40.7)	0.469
How much total paid time did you take for parental leave at current and/or past institution as a faculty member? c			<0.001
≤6 weeks	34 (91.9)	64 (55.2)
7–12 weeks	2 (5.4)	39 (33.6)
>12 weeks	1 (2.7)	13 (11.2)

Bold data indicates statistically significant (alpha = 0.05).

a

Data for two participants who responded with “Prefer not to answer” (N = 2) are not included in the table due to the small number.

b

The proportions listed were computed relative to those who were not content with their parental leave, that is, 8 men and 59 women.

c

Note that the survey had options for 2, 4, 6, 8, and 12 weeks, as well as an “Other” option for write-in responses. There were 27 respondents who reported taking parental leave but logged a total leave time of zero weeks, mostly because they used vacation or sick days in place of official leave.

Statistical analysis

To test for differences in demographic and professional characteristics between PL and NoPL groups, and between men and women subgroups, Welch’s t-test and Pearson’s chi-squared test with Yates’s continuity correction were used for continuous and categorical variables, respectively. Fisher’s exact test was used for the categorical variables in which cell counts were small (i.e., n < 5). General linear regression models were used to test if PL status, gender, and their interaction significantly predicted academic achievement outcome measures, including adjustment for the confounding effect of academic rank. Logistic regression and quasi-Poisson regression models were used for the binary and count variables, respectively, using the same predictors and confounders.

For measurement of burnout, we used four categorical measures: overall burnout, personal accomplishment, depersonalization, and emotional exhaustion. For consistency with the three MBI-9 variables, the responses for overall burnout were pooled to create three levels; responses of “1: Not at all burned out” and “2: Slightly burned out” were considered “low,” responses of “3: Moderately burned out” were considered “moderate,” and responses of “4: Very burned out” and “5: Completely burned out” were considered “high.” Multinomial regression models were used to test if PL status, gender, and their interaction significantly predicted the four categorial measures of overall burnout, personal accomplishment, depersonalization, and emotional exhaustion, adjusting for academic rank.

Institutional review board/ethics committee information

Because the survey was confidential and no identifying information was collected, each site PI obtained neurology chair approval and institutional review board (IRB) approval with waiver of informed consent or IRB exemption, as appropriate.

PL variability

Our findings align with literature that demonstrates variability in PL policies across institutions, which has profound implications for faculty well-being. Slostad et al. found that many institutions provide shorter PL for birth parents compared with nonbirth parents, and some require faculty to use sick or vacation leave due to limited paid PL policies. ¹⁶ This lack of standardized support is evident in our dataset, where some women had no access to paid leave, and many women used sick or vacation days for maternity leave, highlighting variability in institutional support for new parents. Such differences may exacerbate emotional exhaustion among women who take PL. Future studies should explore the impact of institutional policies on burnout and faculty retention, especially for women neurologists.

Academic accomplishments and PL

Interestingly, although faculty members who took PL received fewer awards than their NoPL counterparts, other measures of academic achievement—publications, leadership positions, and funded projects—were not significantly different between the two groups. This suggests that PL itself may not have a direct detrimental effect on academic achievement, but rather the timing of career development stages may play a larger role. Faculty members with children tend to be in earlier stages of their careers, which may require additional time to secure awards and recognition. The pressures to “catch up” upon returning from PL can be compounded by family responsibilities and the emotional exhaustion associated with managing both work and family life. In fact, we noticed that women who did not take a paid PL were more likely to work part-time (work hours 21–40, 25.0%) than women who took PL (10.3%, Supplementary Tables S1 and S2, p = 0.37).

Our study also found that women who took PL had fewer publications as first or last author compared with their male counterparts (Fig. 1). This suggests that while PL may not decrease overall academic output, it could limit women’s ability to secure leadership roles in academic publications, which are critical for career advancement. This finding mirrors existing research by Ngamasana et al., indicating that caregiving and family responsibilities disproportionately burden women and create barriers to career progression in academic medicine. ²⁹ Such disparities may prevent women from advancing at the same rate as their male colleagues, further contributing to the gender disparities in leadership positions. We acknowledge that the higher proportion of professors in the NoPL group may reflect historical policy limitations and gender norms, rather than an effect of PL on advancement.

OPEN IN VIEWER

FIG. 1.

Parental leave (PL) and academic achievements. The figure shows similar findings in PL (Yes) compared with NoPL group (No) in (A) yearly publication numbers, (B) first or last author publication numbers, (C) leadership position, (E) speaking invitations, and (F) funded projects. The number of awards (D) received was higher in the NoPL group compared with the PL group. *This difference is mainly driven by a higher men representation in the NoPL group (p = 0.018). First/last author publication numbers were similar in PL compared with NoPL but were higher in women who took PL compared with women who did not take PL (p = 0.02). Bars with a 95% confidence interval for each subgroup.

Emotional exhaustion and PL

Emotional exhaustion was moderate to high in both the PL and NoPL groups, but higher in the PL group (moderate to high in 73.8% and 51.9% in NoPL, p = 0.017). In addition, women who took PL were significantly more likely to experience emotional exhaustion, being 6.7 times more likely to report moderate emotional exhaustion compared with their male counterparts. This increased emotional exhaustion, coupled with reported discontentment, may be an important factor influencing women’s decisions to leave medicine. Major areas of discontentment, including unclear instructions on PL at their institution, absence of adjusted productively targets, and the need to take an additional unpaid leave specifically in women, were potential drivers of the emotional exhaustion. However, women who took PL did not demonstrate a deterioration in academic achievement, with no significant differences in overall publication rates, leadership positions, or funded projects compared with men. One possible interpretation of our findings is that emotional exhaustion may be associated with increased dissatisfaction but does not necessarily translate into reduced professional output, highlighting broader systemic factors that could influence retention of women in academic medicine. Using the MBI, we captured respondents’ current burnout state. Since the MBI was collected at the time of the PL (or childbirth/adoption), our findings about burnout and exhaustion may not directly reflect PL/NoPL influence on burnout and should be interpreted with caution. Further studies are needed to address challenges in quantifying PL duration and to explore the effects of institutional policies on career outcomes and burnout.

It is also worth noting that the emotional exhaustion observed in our study may be linked to the multiple roles parent neurologists must juggle, particularly upon returning to work after PL. Faculty may face added pressures to “make up” service duties or clinical responsibilities, which could contribute to greater stress, especially for women who often have fewer institutional support systems in place compared with men. Previous research has shown that women, especially those with children, are more likely to experience work–life conflict and consider leaving academia due to the competing demands of family and career. ³⁰ Insufficient resources and institutional infrastructure explain the emotional toll observed in our study. Addressing these concerns through better work–life balance policies could alleviate some of the stress experienced by faculty members with young families.

Our findings suggest that academic achievement could benefit from more flexible institutional policies, such as delaying the tenure clock or adjusting promotion timelines for those who take PL. This approach could provide neurologists in general, and women in particular, the space they need to recover from the emotional exhaustion while juggling work and parenthood. Moreover, the length of PL did not significantly impact academic achievement, further indicating that concerns about extended leave (12 weeks) leading to productivity gaps may be unfounded. Instead, it is the emotional toll of balancing multiple responsibilities that appears to be more detrimental to neurologists’ well-being after PL.

Institutional support to faculty parents

In terms of institutional support, it is clear that better accommodations are needed for parents, including men and women, during their leave. Institutions should consider providing clear and supportive parental PL policies. For instance, we noticed in our survey that faculty with children tend to be at an earlier career stage and need more time to develop their careers to secure awards and receive recognition. Suggested strategies such as limiting or waiving productivity metrics and call pools could reduce the emotional stressors and facilitate career advancement. Such models have been implemented in other countries. For instance, Canada has implemented policies that allow faculty members to be excluded from call pools during PL, potentially providing relief for those with family responsibilities. ³¹ Similar policies could be explored in the United States to reduce burnout and increase retention of faculty, especially women, in academic medicine.

The importance of strong institutional support is further emphasized by findings from other studies. Matulevicius et al. demonstrated that working parent faculty, especially women, were more likely to decline leadership opportunities and consider reducing their employment or leaving academia altogether due to work–life integration stress. ³⁰ This suggests that faculty with children are disproportionately affected by work–life stress, and these challenges have been further exacerbated by the COVID-19 pandemic. ³² Our study also echoes these concerns, as the emotional exhaustion of women who took PL could contribute to higher burnout rates and a desire to leave the field. Unexpectedly, parents who took PL had higher family–work conflict scores. This finding should be interpreted with caution, since this survey did not collect responses immediately after a PL.

Study limitations

Despite the valuable insights provided by this study, there are limitations to consider. Recall and selection bias may affect the accuracy of the data, as only those who completed the survey were able to contribute to the findings. We acknowledge that voluntary surveys are subject to selection bias—particularly among respondents with strong views on equity-related topics—but we believe the broader academic framing of the survey may have mitigated some degree of topic-specific bias. The study relies on retrospective self-report, and some respondents may have taken PL many years prior to survey completion. Our survey collected information about PL but did not differentiate when PL was taken and was not administered directly after PL. So, the current well-being and burnout responses may not reflect a direct link to PL status. To reduce concerns about confidentiality and potential reidentification, particularly in smaller departments, we did not collect age. Instead, we used career stage and time since fellowship (end of training) as less identifying proxies, which imposed additional limitations on this work. Surprisingly, MBI scores (reflecting burnout at the time of survey completion) were higher in the PL group. We acknowledge that this finding may be related to confounding factors such as sex distribution, unequal caregiving responsibilities, and sustained workload expectations. Therefore, this should be interpreted with caution and requires future longitudinal studies or a registry to assess this relationship.

Additionally, our results may not fully represent regional and cultural differences across all U.S. academic neurology centers with heterogeneous PL policies. Future research should investigate how varying family leave policies, home support systems, and institutional accommodations impact faculty burnout, academic success, and retention, particularly for women.

Suggested items and future directions

Accommodation for PL and target adjustment (RVU, service time, etc.) to PL are crucial for faculty success and for supporting parents when they return from PL. If we are to achieve gender equality in academic medicine, we need a cultural paradigm shift that incorporates equitable resources and opportunities for faculty with families. Supporting early career faculty retention will provide more efficient physicians, happier patients, and enrichment of the neurology research field. PL provides faculty the opportunity to experience intimate and precious relationships to build their internal family structure, and this internal stability will promote the workforce and would reduce neurologists’ attrition. Our study shows that PL does not hinder academic achievement, which highlights the importance of parenting support resources during PL and when returning from PL to enhance faculty retention. Future research should investigate the benefits of family supportive interventions among neurology faculty in a longitudinal study to demonstrate better academic and personal achievements. Mental health research can better delineate the impact of major life events on faculty and their offspring and raise the need for improvement.