A change in the balance between personal and professional life among female gynecologic oncologists

Background: Female representation in gynecologic oncology has increased over the last two decades. Our objective was to compare work-life balance issues faced by female gynecologic oncologists between 1998 and 2015. Material/Methods: We conducted a cross-sectional survey of physician members of the Society of Gynecologic Oncology. A survey sent to female gynecologic oncologists in 1998 was expanded, piloted with 10 volunteers, and administered in electronic format (DatStat Illume) in February 2015. There were 75 fixed response questions regarding 4 domains: demographics, mentoring issues, work-life balance, and caregiving responsibilities. We compared 2015 responses to the 1998 aggregate survey data. Data were analyzed using Stata 10 (Statacorp, College Station TX) with Chi-square/Fisher’s exact tests using aggregate data functions. Results: 172 of 643 female gynecologic oncologists (26.7% response rate) completed the 2015 survey. The historical comparison group included 82 females (56.2% response rate). While more women in 2015 versus 1998 reported starting a family during residency or fellowship (57.7% vs. 36.0%, p<0.009), 42% still waited until after training. More than half (55.9%) of respondents in 2015 said the timing of becoming a parent led to some or a great deal of relationship stress compared to only 20% in 1998 (p<0.0001). The majority of divorces were in fellowship for both groups with 8 (50.0%) in 2015 compared to 5 (45.5%) in 1998 (p=0.8). In 2015, 5 (83.4%) women divorced after fellowship and felt their career had a moderate to great influence on their divorce. Conclusions: Despite changes in work-life balance and caregiving responsibilities in female gynecologic oncologists between 1998 and 2015, challenges still exist today. Correspondence to: Ashley R Stuckey, Women and Infants Hospital, 101 Dudley St, Providence, RI 02905, USA, Tel: (401) 453-7520; Fax: (401) 453-7529; E-mail: astuckey@wihri.org


Introduction
Work-life balance is defined as a concept of proper prioritization between "work" (career and ambition) and "personal life" (health, pleasure, leisure, family and spiritual development/meditation) [1].
At the time of its founding in 1969, the Society of Gynecologic Oncology (SGO) included one female member. Female representation in gynecologic oncology has steadily increased since that time, and as of 2016, 43% of attending gynecologic oncologists and 76% of fellows were women [2]. Data from other surgical subspecialties have shown concerns amongst females regarding job flexibility and mentorship [3]. In addition, female surgeons are more likely to experience burnout compared to their male counterparts [4]. Finally, female academic physicians with children spend five hours per week less time at work and four hours per week more time providing childcare; it is thus no surprise that female gynecologic oncologists have a lower publication rate in the first two years of practice compared to male colleagues (2.7 vs 5.3 publications) [5][6][7].
In an effort to improve understanding of factors that determine personal and professional satisfaction among female gynecologic oncologists, Gordinier et al. surveyed all female members of the SGO as well as female fellows in gynecologic oncology in 1998 [8]. At the time, women comprised 16% of all SGO members and 33% of American Board of Obstetrics and Gynecology accredited gynecologic oncology fellows. The survey was mailed to the women gynecologic oncologists listed in the 1998 SGO directory as well as to female fellows identified from the American Board of Obstetrics and Gynecology. Fifty eight percent of eligible participants responded and identified mentorship, childbearing and childcare concerns as the most salient issues at the time. Suggested areas for improvement within the field included "recognition of the need for flexible schedules and decreased overall workload for working mothers, without major academic penalties" as well as increased access to quality childcare and a need for more female mentoring. More substantial maternity leave policies and enhanced discrimination awareness were also identified as areas for improvement.
With the changing demographics of the specialty in mind, our objective was to compare the work-life issues faced by female gynecologic oncologists between the years 1998 and 2015.

Materials and methods
An updated electronic questionnaire was developed based on the original questionnaire that was sent to female gynecologic oncologists in 1998. The expanded instrument was more comprehensive and contained 75 fixed response questions regarding 4 domains including: demographics, mentoring issues, work-life balance, and caregiving responsibilities. Mentoring data will be presented in a future manuscript. Content validity was assessed through a review by 10 gynecologic oncologists who were selected based on informative sampling to represent diverse sociodemographic characteristics (gender, partner status, children, age, sexual orientation). Each reviewer assessed the questions and response options to ensure inclusivity with regards to diverse life experiences as well as all social variations of a family (i.e. step parents, adoption, divorce, single parents, surrogate gestational carriers). The survey was administered anonymously in electronic format (DatStat Illume) in February 2015 to all female and male gynecologic oncology physician members of the SGO. A link to the survey was sent via email with 3 reminder emails regardless of completion status. This study was approved by the Women & Infants' Hospital Institutional Review Board. We compared the 2015 survey responses to the previous 1998 aggregate survey data. Data was analyzed using Stata 10 (Statacorp, College Station TX) with Chisquare or Fisher's exact test for aggregate contingency table data (Stata tabi command). All p-values presented were two-sided with p<0.05 considered statistically significant.

Results
Of the 1246 SGO members or fellows invited to participate, 268 (21.5%) responded. Of the respondents, 172 women out of 643 total female SGO members were included in these analyses (response rate 26.7%). There were no differences in position or type of practice between 1998 and 2015. Most respondents in both 1998 and 2015 were between 36 and 45 years of age. In 2015, 148 women (86.1%) were married or living with a partner, compared to 65 (79.3%) in 1998 (p=0.02). In 2015, 14 (9.1%) women reported a female partner compared to 3 (4.6%) in 1998 (p=0.4). Fewer women were currently divorced or separated in 2015, 3 (1.7%) compared to 8 (9.8%) in 1998 (p=0.02) although there were no differences in respondents who were ever divorced ( Table 1). The majority of divorces were in fellowship for both groups with 8 (50.0%) in 2015 compared to 5 (45.5%) in 1998 (p=0.8). In 2015, for those 6 women who had a post-fellowship divorce, 5 (83.4%) felt their career had a moderate to great influence on their divorce. These questions were not asked in 1998. The number of women who reported having a physician spouse or partner was similar between 2015 and 1998, 55 (42.6%) and 38 (54.3%) respectively (p= 0.1).
There were no differences in the number of respondents with children or the number of children per respondent in 2015 compared to 1998 (Table 2) In response to how career plans affected decision-making regarding parenting, women in both 2015 and 1998 felt career affected the timing to become a parent somewhat or very much, 85.6% and 88.0% respectively (p=0.8). More than half (55.9%) of respondents in 2015 said the timing of becoming a parent led to some or a great deal of stress in their relationship compared to only 20% in 1998 (p<0.0001) ( Table 2). In 2015, 50.9% of women reported having taken off at least 7 weeks after becoming a parent compared to only 28% women in 1998 (p=0.01). Most women in both 2015 and 1998 (57.3% and 62.0%, respectively) felt that if allowed, they would have taken more time off after becoming a parent.
In 2015, when asked to select all sources of childcare assistance, 43 (38.7%) women reported having a nanny, 65 (58.6%) relied on their spouse, 33 (29.7%) used daycare, and 12 (10.8%) women reported having no needs. In 2015, women were much more likely to rely on a spouse/partner for childcare, 65 (58.6%) compared to 6 (12%) in 1998 (p<0.0001) ( Table 2). On the other hand, women in 2015 were much less likely to have live-in childcare, 10 (17.0%) compared to women in 1998 where 21 (50.0%) reported having live-in childcare (p=0.001). As there have been more women entering the field since 1998, we expected to have more women in 2015 with younger children. To attempt to adjust for this we removed those women >55 as well those with no childcare needs and/or the youngest child was greater than 18 and we did not note any changes from the initial calculation. While not asked in 1998, in 2015, 42 (42.4%) women reported being primarily responsible for arranging childcare, 33 (33.3%) women shared responsibility with a spouse/partner, and 24 (24.2%) women had a spouse/partner who arranged childcare.

Discussion
There have been substantial changes since the late 1990s in the practice of medicine and cultural norms, prompting us to reassess how female gynecologic oncologists are balancing the demands of the workplace and the home. We found that female gynecologic oncologists in 2015 still faced many of the same challenges they faced outcomes, and should not impose punitive repercussions on those surgeons who chose to have children [13]. Since the majority of female gynecologic oncologists in both the 1998 and 2015 surveys thought their career affected their decision-making regarding parenthood, we feel this topic may merit more consideration by our professional organizations.
The United States as a whole lags behind its counterparts with respect to family leave, as only 2 out of 185 countries do not mandate paid maternity leave: the United States and Papua New Guinea [14]. While in the 2015 survey, 50.9% of respondents who had ever been a new mother reported having taken off at least 7 weeks compared to 28% of women in 1998 (p=0.01), over half of female gynecologic oncologists (57.3%) still felt that if allowed, they would have taken more time off after becoming a parent. This highlights an area we can continue to improve on in the work-life balance of gynecologic oncologists.
Childcare issues among female gynecologic oncologists have changed somewhat since 1998. Women in 2015 tended to rely on their spouse or partner more often for assistance with childcare compared to 1998. While the literature has traditionally supported gender differences in domestic activities, with more responsibility falling on women, our findings could reflect a shift to an increase in shared responsibility [15,16]. Additionally, the use of a nanny decreased significantly with only 38.7% of participants in 2015 relying on a nanny compared to 78.0% in 1998. Reasons for this are unclear but could include increased joint spousal responsibility, more flexible alternative childcare options, or an increase in the cost of childcare. Finally, similar to other reports, 42% of gynecologic oncologists in 2015 reported having a spouse or partner who is also a physician [17][18][19][20]. Given responsibilities at work and home, these dual physician relationships may be challenging, especially when 42% of the female gynecologic oncologists in 2015 were solely responsible for arranging childcare.
Our study supports the importance of balancing work-life responsibilities given that an imbalance leads to burn out, decreased job satisfaction, and potentially deleterious effects on health, interpersonal relationships, sexual relationships and increased stress levels [21]. Shanafelt et al. [22] examined burnout and work life balance among physicians in the United States comparing 2011 to 2014. They noted that when compared with 2011, rates of burnout among physicians were higher (54.4% vs. 45.5%; p<.001) in 2014 and satisfaction with work life balance was lower (40.9% vs 48.5%; p<.001). Interestingly, the same study noted a substantially decreased satisfaction with work life balance among physicians between 2011 and 2014, though there was no increase in the median number of hours worked per week. While our study did not address burnout, we did see that career sometimes changed the choices women made in their personal lives as evidenced by delayed childbearing after training, less time for maternity leave, and potentially electing to not have children.
The 1998 study to which we compared 2015 results was limited to female gynecologic oncologists, so our current comparison included only women. However, we know that men are affected by work life balance issues as well, and while gender differences in 2015 will be addressed in a separate manuscript, we are not able to compare the work life challenges male gynecologic oncologists face now with those in 1998. An additional limitation of our study is the low response rate of 27% which contrasts with the 58% response rate in 1998. One reason for this difference may be the novelty of inquiry in 1998 regarding the female gynecologic oncologist experience. It was also the first chance many of these women had to express their concerns on subjects that seventeen years ago, including the effect of career on the decision to divorce, bear children, and choices of childcare. However, we found they were more likely to have a spouse/partner that they can rely on for childcare responsibilities in 2015 compared to 1998.
In Western countries, there is a growing trend for women to delay childbearing for purposes of career attainment. Factors contributing to this include changes in societal roles and values, increasing gender equity, accessible and effective contraception, and economic pressures [9][10][11][12]. Consequently, women are often entering fellowship or medical practice at the peak of their reproductive years. While 64% of female gynecologic oncologists delayed childbearing until after training in 1998, only 42% did so in 2015. This partial shift in the timing of childbearing suggests that more recently trained fellows may deem it more feasible to have children prior to or during training. However, despite the fact that the majority of respondents with children chose to have them prior to or during training in 2015, when asked their opinion regarding when the best time was to have children, only 24% felt during training was the best time and 51% would ideally choose post-fellowship. Likewise, most of these women (55.9%) said the timing of becoming a parent led to some or a great deal of stress in their relationship. Interestingly, only a small percentage of the gynecologic oncologists surveyed in 1998 (20%) felt the timing of becoming a parent led to some or a great deal of stress. Therefore, although the field has more women having children during training, there is still a perception that this is not an ideal time for childbearing, which may be leading to increased stress for female gynecologic oncologists and trainees.
The importance of parental leave has recently been brought to the forefront of multiple medical specialties. In 2016, the American College of Surgeons put forth a statement recognizing that a successful surgical career should not preclude a surgeon's choice to be a parent.
As a profession, surgery should be supportive of healthy pregnancy  were somewhat taboo to discuss at that time. While we acknowledge the 2015 response rate is low, it is typical of electronic national survey studies of the members of physician societies and we did have a similar distribution of respondents to that seen in 1998 [4,23,24]. Additionally, there may be a small number of women that responded in both 1998 and 2015 which could skew the results. However, when excluding those who might have been included in the 1998 survey when analyzing the stress of the timing to parent, the results remained the same. Finally, only 1998 aggregate data was available to use for comparisons, which limited our ability to further categorize the data and make comparisons between questions. Over the past decade there has been a stronger focus on work-life integration. In 2016 and 2017, a main focus at the SGO Annual Meeting was on wellness and burnout prevention. While this study demonstrates increased numbers of trainees having children and some improvements in shared childcare responsibilities, we continue to see challenges associated with balancing our professional and personal lives. As a sub-specialty of obstetrics and gynecology, it is important for our specialty to be a leader in creating environments that promote work-life balance. One of our respondents summed it up in the following statement "I like the term work life integration. Balance implies they are weighed against each other, and the reality is that all the aspects of a life need attention and bring satisfaction. How they integrate to make that possible is the real challenge."