Gender, Parenting Status, and the Academic Productivity of Pediatricians During the COVID-19 Pandemic

We distributed an anonymous, internet-based survey to faculty, residents, and fellows using departmental e-mail lists at a quaternary-care children’s hospital in August 2021 regarding the impact of the COVID-19 pandemic on work environment, household obligations, and academic productivity (Supplemental Information). The survey featured mixed-format questions including Likert-type scales, multiple choice, and free-response questions. All questions were voluntary. Before distribution, a multidisciplinary group of experts in pediatrics, medical education, and qualitative research assessed the survey for face validity. We conducted cognitive interviewing of the survey with 1 pediatric fellow, 1 pediatric surgeon, 1 pediatric generalist, and 3 pediatric subspecialists. Two residents pilot-tested the survey to assess clarity and readability and determine the estimated time burden (~10 minutes). The institutional review board deemed this study exempt (STUDY20040076).

We used descriptive statistics for demographics and dichotomized and analyzed Likert-scale responses with χ² or Fisher’s exact tests. We performed multivariable logistic regression to determine factors associated with self-reported academic productivity. We adjusted models for age, trainee versus faculty status, race, ethnicity, gender, and parenting status. We excluded respondents who did not disclose their gender from our multivariable analysis. We used RStudio (Boston, MA) and R version 4.1.1 (R Foundation, Vienna, Austria) for statistical analysis.

We analyzed comments from the free-response questions using inductive thematic analysis. The senior author (TK) developed an initial codebook, and 2 coders (ES, EF) then applied the codebook to each comment and resolved discrepancies through consensus. The PI, senior author, and a coinvestigator (ES, TMK, BM) adjudicated any differences in interpretation. We identified central themes and representative quotations.

We gathered 366 responses, representing a 46.4% response rate (65.1% women; Table 1). Nearly one-half of respondents (48.9%; 179/366) were parents and 79.9% (143/179) reported their child(ren) required childcare. Of these, 60.8% (87/143) experienced unreliable childcare during the pandemic.

One-half of respondents (50.3%, 184/366) reported decreased academic productivity during the pandemic. Women were significantly more likely to report decreased academic productivity than men (65.8% [146/222] vs 29.5% [38/129], P <.001; Fig 1). Parents with unreliable childcare reported significantly decreased academic productivity compared with those with reliable childcare (76.2% [64/84] vs 36.7% [18/49], P <.001; Fig 2). Among physician parents with unreliable childcare, women disproportionally reported decreased academic productivity compared with men (88.5% [54/61] vs 43.5% [10/23], P <.001; Fig 3). In multivariable logistic regression, women physicians with children were significantly more likely to report decreased academic productivity than men with children (adjusted odds ratio: 10.19, 95% confidence interval: 4.68–22.23; P <.001; Supplemental Tables 2 and 3).

All resident (n = 10), 96.4% (n = 27) of fellow, and 75.7% (n = 106) of faculty parents reported that their child(ren) required childcare. Of these, 60.8% (n = 87) reported experiencing unreliable childcare during the pandemic (80.0% of residents [n = 8], 81.5% of fellows [n = 22], 53.8% of faculty [n = 57]). The most common reasons for unreliability were caregiver illness and/or quarantines (n = 23), daycare closures (n = 23), school closures (n = 13), and concern about potential exposures (n = 15).

Twenty percent of respondents elected to share a free-response comment. After thematic analysis, 4 major themes emerged: (1) respondents struggled to adapt to increased household obligations and childcare challenges, (2) respondents perceived decreased academic productivity from their baseline, (3) work–life conflicts left respondents feeling inadequate both at work and at home, and (4) respondents experienced various negative emotions throughout the pandemic, the most ubiquitous of which were stress, fear, and uncertainty. Figure 4 highlights the key themes/subthemes and provides representative quotations.

Women physician parents were significantly more likely to report decreased academic productivity during the COVID-19 pandemic compared with men physician parents, and unreliable childcare was a significant contributor to this disparity.

In our study, 60% of physician parents reported unreliable childcare and/or had difficulty finding childcare during the pandemic. We found that childcare reliability had no significant effect on the productivity of men. On the contrary, women physician parents with unreliable childcare reported significantly decreased academic productivity. Women were significantly more likely to report decreased academic productivity compared with men, echoing studies revealing gender disparities around objective productivity measures such as number of publications, research funding, and promotion rates.^5,9,10  Our findings are in line with previous studies suggesting that women physicians are more likely than men to take on the extra needs at home, even at the expense of academic productivity.^11,16,17  Women may also experience bias and be passed over for opportunities after they have children, (ie, the “maternal wall”), resulting in fewer prospects for leadership, publication, and career advancement.¹⁸  However, as the consequences of inadequate childcare may impact women physicians disproportionately, institutional prioritization of childcare supports can help women physicians overcome systemic barriers to gender equity in academic medicine.¹⁹ 

Institutions should acknowledge and accommodate caregiving duties and household responsibilities. The transition to allow virtual participation in activities such as conferences, division meetings, Grand Rounds, and various lectures afforded increased flexibility for many parents.¹⁹  However, parents in our study reported significant stress and conflict with virtual meetings extending into the early morning, late evening, or weekend hours. Possible institutional responses to these concerns could be limiting virtual meetings to business hours and considering virtual participation as equivalent to onsite participation when new opportunities arise and/or evaluations for promotion occur.

Many faculty respondents in our survey voiced concerns regarding promotion. Physician parents reported difficulty meeting deadlines, ongoing interruptions to research, and anxieties surrounding grant funding. Respondents with and without children discussed the trauma of the pandemic and the difficulty of achieving baseline productivity given ongoing uncertainty and stress. As blanket extensions of the tenure clock can inadvertently exacerbate gender inequities in academics,^20,21  the tenure evaluation process should instead consider the challenges faced during the pandemic and account for dedication to patient care and education under challenging circumstances.²²  Although institutions play a critical role, journals, national societies, and grant-funding agencies are additional sentries to gender equity.²³  Thus, change should be comprehensive and multifaceted.

Our study has several limitations. First, it was conducted at a single quaternary-care pediatric hospital, which may limit generalizability. Although our response rate was 46.4%, this is considerably higher than typical response rates for health care provider surveys (~20%),^24–26  suggesting that participants have high perceived importance of this topic. The subanalysis of men with unreliable childcare had a low number of respondents, which may have limited the ability to make gender comparisons. However, the gender breakdown of our respondents is in line with the gender breakdown in the field of pediatrics.¹  In addition, these low numbers may be reflective of how men may be less likely to perceive childcare as unreliable because they may be less involved in household responsibilities. To prioritize anonymity, we did not ask questions about faculty roles (ie, assistant versus associate professor) or breakdown of clinical time. Also, we limited our investigation to questions regarding childcare and did not explore the impact of eldercare. Participants who felt more strongly about the topic may have been more likely to complete the survey, leading to sampling bias if these respondents differ from nonrespondents. We relied on respondent self-report of academic productivity, which may be subject to social desirability and recall bias. Additionally, participants may underreport their accomplishments, and this may be more prominent among women participants, leading to gender bias.

Although gender disparities in academic productivity have been previously observed,^9–12  the sudden and pervasive challenges to securing reliable childcare were relatively unique to the COVID-19 pandemic. Thus, although the baseline difference in productivity between men and women surveyed is unknown, women were more likely to report decreased academic productivity and unreliable childcare was a significant contributor to this disparity. Because the perception of decreased productivity may have been amplified by the multiple other stressors physicians experienced during the pandemic, future investigations should examine objective measures of productivity, such as number of publications and presentations, grant funding procurement, and academic promotion, to determine if this self-reported decreased productivity aligns with lack of advancement in academic medicine. Finally, we acknowledge the intersectional aspects of identity (such as race/ethnicity and sexual orientation) and the detrimental impact of gender stereotypes, recognizing that our respondents have multiple identities and experiences beyond those of their gender and parenting status.

In this survey of pediatricians and pediatric subspecialists during the COVID-19 pandemic, women with children were significantly more likely than men with children to report decreased academic productivity. Unreliable childcare was a significant contributor to this disparity. Institutions must take action to mitigate the long-lasting effects this pandemic will have on academic pediatricians and gender equity in medicine.

Thank you to Drs Stephanie Dewar, Andrew Nowalk, Arvind Srinath, and Emily Trauernicht for their contributions to development of the survey tool and Drs Brian Martin and Terence Dermody and Ms Karen Anderson for their support and role with survey distribution.