Perceptions of Parental Leave Among Ophthalmologists (2024)

Key Points

Question What are the current North American ophthalmologist perceptions of taking parental leave?

Findings In this survey study of 186 North American ophthalmologists, attitudes toward stop-the-clock policies were mixed. There was varied awareness of this policy among academic and private practitioners, and attitudes toward parental leave were statistically significantly associated with sex, work setting, parental status, and years out of training.

Meaning Benchmarking current concerns and attitudes toward parental leave can influence leave policies and better address ophthalmologists’ needs.

Importance Benchmarking attitudes surrounding parental leave among posttraining North American ophthalmologists may reveal possible areas for intervention.

Objective To evaluate perceptions of current parental leave policy and culture among posttraining North American ophthalmologists.

Design, Setting, and Participants This nonvalidated survey used a 19-item online questionnaire. A convenience sample of 186 self-identified North American–based ophthalmologists who had completed training was obtained using listservs, residency program coordinators, and social media. Data collection occurred from May to July 2022.

Main Outcomes and Measures Descriptive analysis demonstrated response frequencies. Pearson χ² comparison of means was performed for categorical variables. Two-tailed t tests were performed for continuous variables.

Results Among the 186 surveys completed, 105 respondents (56.5%) identified as female, 76 (40.9%) worked in academia, 133 (71.5%) were 1 to 20 years out of training, and 156 (83.9%) had children. Attitudes toward stop-the-clock policies, or delays in tenure/promotion/partnership review when taking leave, were mixed. Of 171 respondents, 78 (45.6%) thought that stop-the-clock policies should be optional, 39 (22.8%) thought that they should be required, and 31 (18.1%) thought that they should be removed. Of 76 academicians responding, 56 (73.7%) and 49 (64.5%) were unaware if their institution had a stop-the-clock policy for tenure review or promotion review, respectively. Male and female respondents differed in feeling comfortable with taking leave (36 of 66 [54.5%] vs 67 of 90 [74.4%], respectively; P = .04), as well as in their ratings of stress about peer perception on a scale of 0 to 10 (3.70 vs 4.81, respectively; P = .05). Private practitioners and academicians differed in confirmation of leave options for both parents (52 of 101 [51.5%] vs 49 of 69 [71.0%], respectively; P = .02) and ratings of financial stress (7.10 vs 5.43, respectively; P = .004) and research considerations (1.60 vs 3.85, respectively; P < .001).

Conclusions and Relevance Results of this survey study support the hypothesis that demographic factors affect attitudes toward parental leave among posttraining ophthalmologists; policies could be better publicized. Feelings regarding policies and leave were mixed. These findings should be viewed as hypothesis generating because the survey was not validated and the associations provided could be due to confounding factors.

While the challenges of navigating parental leave in residency are represented in the literature,^1-5 the challenges experienced by posttraining physicians, including ophthalmologists, need to be more fully explored. Posttraining physicians are affected by a unique subset of concerns and policies relating to parental leave.⁶ One such policy is the stop-the-clock policy on tenure, promotion, or partnership review.⁷ This policy, which delays review to compensate for theoretical time and productivity losses, was created to benefit the individual on leave. However, assessment of its true effect has been lacking. Several studies outside of the medical field have demonstrated substantial compensation and promotion deficits following its implementation.^8,9 Practicing physicians must also navigate various concerns when taking leave. Salary¹⁰ and career development/trajectory^11-13 are well-known examples of such concerns. However, evaluation of workplace culture stressors has not been fully explored. Additionally, to the best of our knowledge, the specific concerns and perceptions of both parents have not been well documented. Understanding the current state of parental-leave perceptions may guide future policy for ophthalmologists and other physicians. This survey study aims to better understand ophthalmologist perceptions surrounding parental leave as they pertain to stop-the-clock policies, workplace culture, and career stressors, stratified by sex, workplace setting, years out of training, and parental status.

This study used an anonymous, voluntary, nonvalidated 19-item online questionnaire through Qualtrics (SAP) (Figure 1). Participants were advised that this survey was only for North American ophthalmologists who had completed training. Demographics included self-reported sex, years posttraining, parental status, and primary work setting (private vs academic, with academic related to primary employment by an academic institution).

Stop the clock is defined as tenure/promotion/partnership review extension to compensate for leave. Respondents who worked in academic medicine were asked if their institutions stopped the clock on tenure and nontenure promotion review, while those in private practice were asked if their practice paused the clock on partnership review. Participants whose place of work did stop the clock were asked whether the policy was beneficial. The remainder of the questions were asked of all participants and assessed attitudes toward stop-the-clock policies, parental leave options, desire and comfort with taking leave, leave-related stressors, and perceived workplace culture around leave.

The nonvalidated survey was sent to a convenience sample of ophthalmologists recruited through emails to the listservs of ophthalmologic societies (eTable in the Supplement), a social media group (Moms in Medicine), and academic faculty through residency program coordinators. Data were collected from May to July 2022.

This study was approved by the institutional review board of Columbia University Irving Medical Center, which also determined that individual consent was not needed owing to use of an anonymous survey and data. It followed the American Association for Public Opinion Research guidance for institutional review boards and survey researchers.

Because of the anonymous nature of the survey, as well as the variety of societies and presumed inherent overlap in membership, there is no known denominator for survey distribution. Individuals who failed to answer the 4 demographic questions were excluded from analysis (3 respondents), as were individuals who failed to answer at least 1 question in items 5 through 12 (2 respondents). Analysis was performed using SPSS Statistics, version 28 (IBM). Descriptive analysis summarized overall data and stratification according to sex, workplace setting, children, and years out of training. There were insufficient respondents who indicated “Prefer not to say” when asked about sex; those individuals were excluded from sex-specific analysis but were included in overall analysis and all other subgroup analyses. P values were 2-sided with P ≤ .05 considered statistically significant; however, because statistical significance testing was completed in the context of descriptive and exploratory analyses, they were not adjusted for multiple analyses. Pearson χ² tests were used to assess relationships among the different categories of respondents. Due to small counts, analysis of differences combined the 2 most recent categories of years out of training (1-10 and 11-20 years) and the 2 least recent years out of training (21-30 and ≥31 years). Independent-samples t tests (2-tailed) were used for continuous variables with equal variances not assumed. Descriptive analysis of continuous variables grouped responses into low (0-3), medium (4-7), and high (8-10) stress ratings on an 11-point scale.

A total of 186 respondents were included for analysis (Figure 2): 78 (41.9%) identified as male, 105 (56.5%) identified as female, and 3 (1.6%) preferred not to disclose their sex. Seventy-six respondents (40.9%) worked in an academic setting, and 110 (59.1%) worked in private practice. A reported 156 respondents (83.9%) had children; 89 respondents (47.8%) were 1 to 10 years out of training, 44 (23.7%) were 11 to 20 years out of training, 25 (13.4%) were 21 to 30 years out of training, and 28 (15.1%) were 31 years or more out of training. Male and female respondents varied in number of years out of training, with 48 of 78 male respondents (61.5%) vs 82 of 105 female respondents (78.1%) 20 years or fewer out of training (P = .02).

In private practice, 94 of 110 respondents (85.5%) had children, while 62 of 76 respondents (81.6%) in academia had children. Of 78 male respondents, 73 (93.6%) had children vs 82 of 105 female respondents (78.1%) (P = .004); there was no statistically significant difference in female childbearing in private practice or academia. There were either no statistically significant differences in perceptions of parental leave by parental status or items were not analyzed due to low counts (Tables 1 and 2).

Among the 76 academic practitioners, most were unsure if their institutions stopped the clock for tenure review (n = 56 [73.7%]) or nontenure review (n = 49 [64.5%]). Of the 10 whose institutions did stop the clock for tenure review, 6 found it beneficial and 4 felt neutral about its benefit. Of the 10 individuals whose institutions stopped the clock for nontenure review, 5 found it beneficial, 4 felt it was neutral, and 1 felt negatively. Of the 110 private practitioners, 69 reported that their practice had a partnership track. Of the 69, 34 (49.3%) reported that there was no extension in partnership review, 25 (36.2%) were not sure, and 10 (14.5%) said that there was. Of the 10 whose workplaces did stop the clock on partnership review, 5 thought it was beneficial, 2 were neutral, and 3 felt negatively (Table 3).

Of 170 respondents who answered the question about parental leave options for both parents, 101 respondents (59.4%) reported that parental leave was an option for both parents, though 29 (17.1%) were not sure and 40 (23.5%) reported that it was not. Responses to this item differed according to workplace, whereby 52 of 101 respondents (51.5%) in private practice vs 49 of 69 respondents (71.0%) in academic medicine responded that leave was an option for both parents (P = .02). It also varied by parental status: 89 of 142 respondents (62.7%) with children vs 12 of 142 respondents (42.9%) without reported that they had the option (P = .02).

Of 171 respondents, 78 (45.6%) thought the stop-the-clock policy should be optional, 39 (22.8%) thought it should be required, 31 (18.1%) thought it should not exist, and 23 (13.5%) responded “other.” There were no differences when stratified by sex, workplace setting, or years out of training. Stratification by parental status was not performed due to low counts.

Among 170 respondents, 140 (82.4%) reported that they would want to take leave if they had a child now, and 105 of 158 respondents (66.5%) felt comfortable doing so. Additionally, 33 of 155 respondents (21.3%) were concerned about the negative workplace perception when taking leave, 88 of 155 respondents (56.8%) were not, and the remainder were neutral. All female respondents would either choose to take parental leave or were not sure, compared with 58 of 71 male respondents (81.7%). Of 90 female respondents, 67 (74.4%) would feel comfortable taking leave, compared with 36 of 66 male respondents (54.5%) (P = .04; Table 1).

There was an association between years out of training and whether one would opt to take parental leave. Of 124 respondents 20 years or fewer out of training, 111 (89.5%) reported that they would take leave vs 29 of 46 respondents (63.0%) 21 years or more out of training (P < .001). Of 117 respondents 20 or fewer years out of training, 84 (71.8%) reported that they would feel comfortable taking leave vs 21 of 41 respondents (51.2%) 21 years or more out of training (P = .005; Table 1).

Stressors associated with taking parental leave were assessed using an 11-point scale (0-10), and responses were categorized into low (0-3), medium (4-7), and high (8-10) groupings. Parental leave influence on salary and patient care/coverage had overall mean (SD) stressor ratings of 6.40 (3.56) and 6.22 (2.90), respectively. The mean (SD) ratings for other stressors included 4.27 (3.39) for peer perception, 4.13 (3.48) for effect on referrals, 3.19 (3.33) for effect on promotion, and 2.57 (3.06) for effect on research.

The highest-rated stressors for both male and female respondents were patient care/coverage (5.81 and 6.70, respectively) and salary (6.23 and 6.57, respectively). There was statistically significant overlap between both sexes in mean (SD) stress ratings across all categories except for peer perception, in which male respondents rated it 3.70 (3.06) vs 4.81 (3.55) for female respondents (difference, −1.10; 95% CI, −2.18 to −0.03; P = .05; Table 2).

There was an association between work setting and research stress, in which private practitioners rated it as a mean (SD) of 1.60 (2.48) vs academic practitioners as 3.85 (3.28) (difference, −2.25; 95% CI, −3.24 to −1.27; P < .001). Work setting was also associated with salary stress, in which private practitioners rated it as a mean (SD) of 7.10 (3.62) vs academic practitioners as 5.43 (3.28) (difference, 1.68; 95% CI, 0.55-2.79; P = .004; Table 2).

Perceptions and concerns surrounding parental leave are nuanced and complex. This study attempts to shed light on attitudes toward this complicated topic among North American ophthalmologists. Focus areas included knowledge and attitudes toward existing policies, culture of acceptance, and stressors.

Results of this exploratory study suggest that knowledge and/or transparency of existing policies may be lacking, and feelings toward stop-the-clock policies in general are varied. Most respondents in academia did not know if their workplaces had a stop-the-clock policy on tenure (73.7%) or promotion (64.5%) review, and 36.2% of those in private practice whose workplace had a partnership track were also unsure. This finding may highlight the need for increased communication of current policies, especially within academic medicine. Among those whose institutions did have this policy, attitudes were cautiously optimistic, with most finding it beneficial or neutral. Among all respondents, however, feelings toward this policy seemed to be more mixed, with 45.6% thinking it should be optional and 18.1% thinking it should not exist at all. Numerous respondents commented that the duration and number of parental leaves that one takes were important considerations.

Only 59.4% of respondents reported the presence of an optional parental leave for both parents. This is a particularly notable finding given the Family and Medical Leave Act, which protects parental leave time for both parents. Survey results demonstrated greater tolerance or transparency of leave for both parents in academia, as well as a chicken-and-egg significant association with those who have children. Several respondents noted that it was financially expensive to take parental leave if one worked in private practice due to the difficulty of meeting overhead costs.

Most individuals responded that they would want to take parental leave (82.4%). However, only 66.5% of the overall respondents felt comfortable taking leave, suggesting the possibility of an unwelcoming culture. Male respondents were more likely to be potentially uncomfortable taking leave. Perhaps this difference is associated with perceived stigma or financial concerns, as was found among male residents,¹⁴ though overall younger colleagues seemed to be more likely and more comfortable with taking leave. These findings may suggest a culture shift among younger ophthalmologists, but may also reflect sex as a confounding variable.

Most individuals felt that their workplace would support parental leave (56.8%). Negative perceptions persist, however. In narrative review, several individuals responded that the length of leave, rather than the taking of leave itself, was the issue. Pressure to take shorter leave has been well reported among trainees and practicing physicians.¹⁵ Several individuals mentioned that there was toxicity in their work environment regarding parental leave, that the offered duration of leave was too short, and that leave came hand in hand with an unfair workload afterward. Whether workload referred to increased workload compared with baseline or to difficulty creating a more part-time position is unclear.

When considering parental leave, stressors regarding patient care/coverage and salary were most notable, followed by peer perception, referrals, promotions, and finally research. Average ratings among female respondents were higher than male respondents in every category, though only peer perception was statistically significantly greater. One narrative response from a female respondent indicated that her leave duration was compared with that of other women, leading to increased stress. Another comment revealed stress surrounding increased work burden on colleagues.

Research was a considerable burden for those in academia, and salary seemed to be a greater burden for those in private practice, though both groups noted salary to be a high stressor. Among those in private practice, many respondents emphasized financial burden (such as overhead) as the major consideration when taking leave. It is pertinent to remember the variability of paid leave time and expectation to contribute to financial costs of the practice while on leave.^16,17 Notably, research on paid parental leave has been associated with better maternal-fetal outcomes and decreased stress.^18,19

This study had several limitations. The sample population was potentially biased by the listservs and groups that allowed circulation, itself limited by the societies/groups approached and contact responsiveness. Low response counts for several subgroups did not allow for comparison of means for all nonvalidated survey items between these subgroups. Grouping years out of training into smaller clusters may have helped reveal more granular changes in perception over time. Due to the sample sizes and missing responses for some surveys, we were unable to construct a multivariate model that allowed for assessment of differential contributions such as sex and years out of training, each of which may act as a confounder. Terms such as stop the clock, extension, and academia may have been variably interpreted in the context of personal experience and work settings. The question regarding benefit of pausing the clock for partnership review is somewhat ambiguous in the context of differing practice structures, and greater description may have been helpful. Similarly, adding questions pertaining to respondents’ current leave policies, in addition to their perceptions around these policies, would have been informative. There was a noticeable response rate decrease with each question, demonstrating possible survey fatigue or lack of answer clarity. Duration and standardization of leave policies often vary with workplace setting, evidenced through reported responses, and this likely contributes to varying views on level of comfort and stress associated with leave.²⁰ Some respondents suggested that their answers would be influenced by duration of parental leave, so questions that more specifically addressed the nuances involved with duration of leave would be beneficial for future research.

Results of this survey study provide an overview of some of the major concerns and considerations of taking parental leave, including the role of sex, workplace setting, parental status, and years out of training. The perceptions and considerations involved in taking parental leave are complex, multifaceted, and varied. A deeper understanding of these intricacies is pivotal to addressing them and to implementing beneficial policy changes.

Accepted for Publication: October 10, 2022.

Published Online: December 8, 2022. doi:10.1001/jamaophthalmol.2022.5236

Corresponding Author: Lora R. Dagi Glass, MD, Department of Ophthalmology, Edward S. Harkness Eye Institute, 635 W 165th St, New York, NY 10032 (ld2514@cumc.columbia.edu).

Author Contributions: Ms Kalra had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Kalra, Dagi Glass.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Kalra, Dagi Glass.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: All authors.

Administrative, technical, or material support: Dagi Glass.

Supervision: Dagi Glass.

Conflict of Interest Disclosures: None reported.