- Email: [email protected]
Achieving Speaker Gender Equity at the SIR Annual Scientific Meeting: The Effect of Female Session Coordinators
EDUCATION, ECONOMICS, AND WORKFORCE
Achieving Speaker Gender Equity at the SIR Annual Scientific Meeting: The Effect of Female Session Coordinators Christine E. Ghatan, MD, Jonathan Altamirano, MS, Magali Fassiotto, PhD, Marcelina G. Perez, MD, Yvonne Maldonado, MD, Shellie Josephs, MD, Daniel Y. Sze, MD, PhD, and Nishita Kothary, MD ABSTRACT Purpose: To examine the impact of targeted efforts to increase the number of female speakers at the Society of Interventional Radiology (SIR) Annual Scientific Meeting (ASM) by reporting gender trends for invited faculty in 2017/2018 vs 2016. Materials and Methods: Faculty rosters for the 2016, 2017, and 2018 SIR ASMs were stratified by gender to quantify female representation at plenary sessions, categorical courses, symposia, self-assessment modules, and “meet-the-expert” sessions. Keynote events, scientific abstract presentations, and award ceremonies were excluded. In 2017, the SIR Annual Meeting Committee issued requirements for coordinators to invite selected women as speakers. Session coordinators are responsible for issuing speaker invitations, and invited speakers have the option to decline. Results: Years 2017 and 2018 showed increases in female speaker representation, with women delivering 13% (89 of 687) and 14% (85 of 605) of all assigned presentations, compared with 9% in 2016 (46 of 514; P ¼ .03 and P ¼ .01, respectively). Gender diversity correlated with the gender of the session coordinator(s). When averaged over a 3-year period, female speakers constituted 7% of the speaker roster (112 of 1,504 presentations) for sessions led by an all-male coordinator team, compared with 36% (108 of 302) for sessions led by at least 1 female coordinator (P < .0001). Results of the linear regression model confirmed the effect of coordinator team gender composition (P < .0001). Conclusions: Having a woman as a session coordinator increased female speaker participation, which suggests that the inclusion of more women as coordinators is one mechanism for achieving gender balance at scientific meetings.
ABBREVIATIONS AMC ¼ Annual Meeting Committee, ASM ¼ Annual Scientific Meeting
From the Radiology Service (C.E.G.), Veterans Affairs Palo Alto Health Care System, Palo Alto, California; and Office of Faculty Development and Diversity (J.A., M.F., Y.M.), Department of Radiology (M.G.P.), and Divisions of Pediatric Radiology (S.J.) and Interventional Radiology (D.Y.S., N.K.), Stanford University School of Medicine, 300 Pasteur Dr., H3630, Stanford, CA 94305-5642. Received April 30, 2019; final revision received June 28, 2019; accepted July 8, 2019. Address correspondence to N.K.; E-mail: [email protected]; Twitter handle: @Fivenada From the SIR 2019 Annual Scientific Meeting. N.K. receives research grants from Siemens (Siemens Healthcare) and EchoPixel (Santa Clara, California). None of the other authors have identified a conflict of interest. © SIR, 2019 J Vasc Interv Radiol 2019; 30:1870–1875 https://doi.org/10.1016/j.jvir.2019.07.006
Women continue to be underrepresented in the sciences, especially at the higher ranks (1). Lower visibility of female scientists compared with male scientists is one of several reasons for the underrepresentation of women progressing into senior ranks (2). Female speakers are outnumbered by male speakers at national and international meetings in many scientific disciplines (3–8), including fields in which women represent the majority (9,10). The urgency of this issue was recently underscored by the director of the National Institutes of Health, who, in a published statement, called to “end the tradition of all-male speaking panels” so that “scientists of all backgrounds are evaluated fairly for speaking opportunities” (11). Speaking invitations and the resultant visibility at major scientific meetings are an important consideration for rank advancement in academia (5). Further, speaking assignments can provide an opportunity to showcase one’s expertise and research interests,
Volume 30 ▪ Number 11 ▪ November ▪ 2019
thereby providing a path to expand the speaker’s professional network. Visibility of underrepresented groups at the podium may also facilitate more diverse recruitment into the field by signaling to audience members that a diverse group of people can find success in a given discipline (5). Recruitment of women and other underrepresented groups into interventional radiology has been an ongoing challenge (12). Recognizing the impact of visibility on career advancement, retention, and recruitment, efforts to increase the participation of women in major meetings have become part of the effort to address gender disparities (5). In 2017, acknowledging the prevailing gender imbalance within SIR and at the podium, the Annual Meeting Committee (AMC), in conjunction with SIR’s Diversity and Inclusion Committee, provided session coordinators with the names of female interventional radiologists to include as speakers in their sessions. These invitations were required, but they represented only part of the overall roster. Additional speaker selection, session curriculum, and the assignment of topics to individual speakers remained at the discretion of session coordinators. The present study investigates the impact of the AMC’s initiative by reporting the gender trends for invited faculty at the 2017 and 2018 ASMs, examines the factors influencing gains in gender diversity, and provides a framework for future change.
MATERIALS AND METHODS This study did not require institutional review board approval. Archived faculty rosters for the SIR ASMs held in 2016, 2017, and 2018 were reviewed. The analysis was limited to the following sessions: plenary sessions, symposia, categorical courses, self-assessment modules/self-assessment continuing medical education, and “meet-the-expert” sessions (henceforth collectively referred to as “podium sessions”). Keynote events and award ceremonies, as well as nondidactic events (scientific abstract presentations, case-based workshops, and handson workshops), were excluded. Included faculty speakers and session coordinators were stratified by gender (male/female) by using existing SIR member data and institutional websites. Up to and including the 2016 ASM, the AMC invited session coordinators by peer-to-peer recommendations, and these coordinators were responsible for selecting speakers for their sessions, with invitations issued at their discretion. Starting in 2017, even though the means for selecting session coordinators remained unchanged, the AMC provided session coordinators with the names of female interventional radiologists to include as speakers in their sessions. The AMC identified prospective female faculty members based on familiarity with published work and/or peer-to-peer recommendations. Session coordinators were required to invite these individuals (typically 1 or 2 names were provided); selections for the remainder of the session’s speaker roster were left to the discretion of the coordinator. To measure the impact of these recommendations and to examine speaker representation as a function of the gender
1871
composition of the coordinator(s) leading the session, the data were further classified as to whether the session was led by an all-male team or a combination with at least 1 female coordinator.
Statistical Analysis Comparisons of the numbers of female speakers were made across year and by coordinator gender by using c2 tests of independence. Average numbers of female speakers at podium sessions by coordinator gender were analyzed with Wilcoxon– Mann–Whitney tests to account for the nonparametric distribution of data. Hierarchic linear regression models were built, one including only conference year and one including conference year and coordinator gender, to analyze the effects of these predictors on the number of female speakers. A P value < .05 was considered significant.
RESULTS From 2016 to 2018, the SIR ASM had 96, 118, and 105 podium sessions that resulted in totals of 514, 687, and 605 speaker assignments accepted by 319, 413, and 338 unique speakers, respectively (Table 1). In line with SIR’s initiative to promote gender diversity, 2017 and 2018 showed increases in numbers of female speakers: 89 of 687 (13%) and 85 of 605 (14%) podium presentations in 2017 and 2018, respectively, were delivered by female speakers, compared with 46 of 514 presentations (9%) in 2016 (P ¼ .03 and P ¼ .01, respectively, c2 test). Gains in gender diversity were not homogeneously distributed across all sessions, and correlated with the gender of the session coordinator(s) (Table 2). Over the 3year period, female speakers constituted 7% of the speaker roster for sessions led by an all-male coordinator team, delivering a total of 112 of the 1,504 podium presentations. In sessions led by at least 1 female coordinator, women delivered 108 of the 302 podium presentations, constituting 36% of the speaker roster, a statistically significant increase in representation compared with sessions led by an all-male coordinator team (P < .0001). In 2017 and 2018, there were overall increases in the absolute numbers of female speakers irrespective of the gender composition of the coordinator team. However, the proportion of presentations delivered by women in sessions coordinated by an all-male team remained 8% in 2017 and 2018, and was statistically unchanged compared with 2016 (6%; P ¼ .57; Fig a). In contrast, the proportion of presentations delivered by women in sessions with at least 1 female coordinator increased to 38% in 2017 and 37% in 2018 compared with 28% in 2016, even though the increase did not reach statistical significance (P ¼ .40; Fig b). Table 3 shows the results of the hierarchical linear regression model to predict the change in the number of female speakers. Variables included were the year of the annual meeting followed by the gender composition of the session coordinators. The first model demonstrates that the 2017 and 2018 ASMs were significantly and
1872 ▪ Female Session Coordinators at the SIR Annual Meeting
Ghatan et al ▪ JVIR
Table 1. Distribution of Male and Female Speaker Assignments by Year Session Type
2016 Speakers (514 Assignments)
2017 Speakers (687 Assignments)
2018 Speakers (605 Assignments)
Male
Female
Male
Female
Male
8 (2)
0
13 (2)
1 (< 1)
11 (2)
Symposia
53 (10)
9 (2)
115 (17)
Categorical
175 (34)
17 (3)
185 (27)
29 (4)
210 (35)
37 (6)
90 (18)
6 (1)
124 (18)
17 (2)
106 (18)
22 (4)
Meet the expert
142 (28)
14 (3)
161 (23)
14 (2)
105 (17)
10 (2)
Total
468 (91)
46 (9)
598 (87)
89 (13)*
520 (86)
85 (14)†
Plenary
SAM/SA-CME
28 (4)
88 (15)
Female 2 (< 1) 14 (2)
Note–Values in parentheses are percentages. SAM/SA-CME ¼ self-assessment modules/self-assessment continuing medical education. *Significant at P ¼ .03. † Significant at P ¼ .01.
Table 2. Distribution of Male and Female Speaker Assignments for Sessions Based on Gender Composition of Coordinator Team Session
≥ 1 Female Coordinator
All-Male Team Male
Female
Sessions
Male
Female
Sessions -
Plenary 2016
8
0
2
-
-
2017
10
1
1
3
0
1
2018
11
2
3
-
-
-
2016 2017
44 100
4 17
3 12
9 15
5 11
1 6
2018
80
11
11
8
3
1
Symposia
Categorical course 2016
164
13
33
11
4
3
2017
161
11
29
24
18
8
2018
169
11
29
41
26
9
Self-assessment modules 2016 2017
80 104
4 5
16 18
10 20
2 12
2 5
2018
78
7
15
28
15
6
Meet the expert 2016
129
8
32
13
6
4
2017
150
10
34
11
4
4
2018
104
8
30
1
2
1
All sessions combined 2016 2017
425 525
29 (6) 44 (8)
86 94
43 73
17 (28) 45 (38)
10 24
2018
442
39 (8)
88
78
46 (37)
17
Note–Values in parentheses are percentages. *P < .0001 for each year (c2 test).
positively associated with the number of female speakers (P ¼ .027). However, after the addition of gender composition of the coordinator team (all male vs at least 1 woman), the effect of the year disappeared (P ¼ .051), whereas the effect of the gender composition of the coordinator team was highly significant (P < .0001), with
the presence of a female coordinator having a strong positive impact on the number of female speakers.
DISCUSSION The present study highlights the partnership among SIR leadership, the Diversity and Inclusion Committee, and the AMC to advance women in interventional radiology and demonstrates that a small change in the invitation strategy can increase the proportion of female speakers in a short time frame. The results also uncover the effect of female session coordinators, as women were more likely to serve as invited speakers when there was at least 1 female coordinator in a session. This suggests that prudent selection to include more female interventional radiologists in leadership roles will have a downstream impact on creating a more diverse speaker roster. The data fit a larger pattern of recent work in other scientific disciplines (3–10,13) in which invited female speakers are outnumbered by male speakers and the presence of female session coordinators correlates with increased gender parity among speakers. Examination of speaker rosters within diverse fields like neuropharmacology, microbiology, social psychology, anthropology, and evolutionary biology, for example, found that fewer women spoke at major meetings than men (5–9,13). This discrepancy extended into fields with gender parity and fields with a female majority. Similar to the results presented here, these studies (4,5,7–9,13) noted that the inclusion of more women as session coordinators correlated with a greater number of female speakers. Increasing women’s visibility at the podium can impact recruitment and retention, playing a critical role in plugging the “leaky pipeline,” a metaphor that describes the high attrition rate of women in the sciences as they progress along the ranks of academia (2). Although women have constituted more than 40% of medical school matriculants for more than 25 years (14), women continue to be underrepresented in interventional radiology (15), constituting 15.4% of graduating fellows but only 9.5% of practicing
Volume 30 ▪ Number 11 ▪ November ▪ 2019
1873
Figure. Distribution of male versus female speakers by year, comparing all-male coordinator teams versus coordinator teams with at least 1 female member. (a) From 2016 to 2018, all-male teams coordinated a majority of the sessions and showed a zero slope in the number of female speakers over the 3-year period (86, 94, and 88 sessions, respectively). (b) During the same time period, coordinator teams with at least 1 female member exhibited an upward slope for the number of female speakers recruited despite the fewer number of sessions coordinated by them (10, 24, and 17, respectively).
1874 ▪ Female Session Coordinators at the SIR Annual Meeting
Table 3. Linear Regression Model Examining Effect of Gender Composition of Coordinator Team Variable
Model 1
b (95% CI) Year
0.16 (0.02–0.31)
Model 2
P Value .027
b (95% CI)
P Value
0.12 (–0.0004 to 0.23)
.051
< .0001
Coordinator team 1 woman All male (ref. group)
– –
– –
1.68 (1.43–1.94) –
CI ¼ confidence interval.
interventional radiology physicians (12). Although factors such as long hours and exposure to radiation have been implicated, inadequate mentorship and gender bias also have been suggested to be related to the poor recruitment and retention of female interventional radiologists (16–18). Academic promotion has been tied to retention of women physicians (19). By providing evidence of external recognition and scholarly achievement necessary for academic promotion, speaking invitations and visibility at national meetings may indirectly play a role in retention (5). In addition, they provide networking opportunities and allow dissemination of research advances that can lead to new employment and research and clinical collaborations (2). Speaker diversity promotes wider recruitment by providing audience members with conscious and subconscious cues as to who can be successful in the field (5). Diversity initiatives today are no longer aimed solely at fairness for underrepresented groups. Research in management science and organizational behavior shows that diversity leads to adaptability, creativity, and innovation, providing a workforce that often outperforms homogenous organizations (20). Data have also demonstrated that companies with more gender-diverse leadership achieve better financial returns (21). Greater diversity at the podium and a broader exchange of ideas among all participants promote these overarching principles and help maintain interventional radiology as a competitive, cutting-edge, and innovative medical specialty. It is unclear why the gender of the coordinator has such an impact, and assumptions of causation are difficult to prove. Although the correlation between female coordinators and the proportion of female speakers may suggest implicit bias against female speakers by male coordinators (22), it could also indicate that female coordinators have larger networks of female colleagues from which to invite speakers. Alternatively, the “imposter syndrome,” a phenomenon whereby women in science underestimate their ability and performance (4), may prompt female invitees to decline an invitation, especially one from a well-known male colleague (4,8). The present study provides insight into diversity and gender parity at the SIR ASM, but it has limitations. The data were derived from a short period of time, and it is
Ghatan et al ▪ JVIR
therefore unclear if gains were solely a result of interventions in 2016 or part of a larger trend unrelated to the AMC initiative. Sampling was obtained from speaker rosters at the SIR ASM only, and cannot be extrapolated to other international, national, local, or regional interventional radiology conferences. Data regarding which female faculty members were appointed by the AMC and which female faculty members participated in the ASM independent of AMC recommendations are not available. Speaker information captured in the program rosters reflects accepted invitations only and does not take into account declined invitations and last-minute substitutions. Demographic information, including gender assignment, was curated from information volunteered by SIR members, is limited to a binary gender assignment and does not capture the full spectrum of gender identity. Quantitative assessment of equity based on an absolute number of speakers is unidimensional and does not take into account details such as audience size, length of presentations, time of day, and other factors that may reflect the prestige of an invitation. The present study confirms that progress has been made since the implementation of AMC initiatives to increase gender diversity, and that a targeted invitation strategy can improve the visibility of an underrepresented group at a major scientific meeting in a short period of time. These efforts are in accordance with strategies suggested in other scientific fields (23). Female faculty members leading and coordinating sessions resulted in a more gender-diverse speaker roster, which suggests that the inclusion of more female coordinators and moderators is one mechanism for achieving gender balance at scientific meetings.
REFERENCES 1. Ceci SJ, Williams WM, Barnett SM. Women’s underrepresentation in science: Sociocultural and biological considerations. Psychol Bull 2009; 135:218–261. 2. Pell AN. Fixing the leaky pipeline: women scientists in academia. J Anim Sci 1996; 74:2843–2848. 3. Casadevall A. Achieving speaker gender equity at the American Society for Microbiology General Meeting. mBio 2015; 6:e01146-15. 4. Ford HL, Brick C, Blaufuss K, Dekens PS. Gender inequity in speaking opportunities at the American Geophysical Union Fall Meeting. Nat Commun 2018; 9:1358. 5. Johnson CS, Smith PK, Wang C. Sage on the stage: women’s representation at an academic conference. Pers Soc Psychol Bull 2017; 43: 493–507. 6. Jones TM, Fanson KV, Lanfear R, Symonds MRE, Higgie M. Gender differences in conference presentations: a consequence of self-selection? PeerJ 2014; 2:e627. 7. Moghaddam B, Gur RE. Women at the podium: ACNP strives to reach speaker gender equality at the Annual Meeting. Neuropsychopharmacology 2016; 41:929–931. 8. Schroeder J, Dugdale HL, Radersma R, et al. Fewer invited talks by women in evolutionary biology symposia. J Evol Biol 2013; 26: 2063–2069. 9. Isbell LA, Young TP, Harcourt AH. Stag parties linger: continued gender bias in a female-rich scientific discipline. PLoS ONE 2012; 7: e49682. 10. Simon JL, Morris EK, Smith NG. Trends in women’s participation at the Meetings of the Association for Behavior Analysis: 1975-2005. Behav Anal 2007; 30:181–196. 11. Collins FS. Time to end the Manel tradition. National Institutes of Health. 2019. Available at https://www.nih.gov/about-nih/who-we-
Volume 30 ▪ Number 11 ▪ November ▪ 2019
12.
13.
14.
15.
16.
are/nih-director/statements/time-end-manel-tradition. Accessed June 23, 2019. Higgins MCSS, Hwang WT, Richard C, et al. Underrepresentation of women and minorities in the United States IR academic physician workforce. J Vasc Interv Radiol 2016; 27:1837–1844.e2. Casadevall A, Handelsman J. The presence of female conveners correlates with a higher proportion of female speakers at scientific symposia. mBio 2014; 5:e00846-13. Association of American Medical Colleges. Longitudinal applicant, matriculant, enrollment, & graduation tables. Available at http://www. aamcdiversityfactsandfigures2016.org/report-section/applicants-enrollment/ #tablepress-12. Accessed March 28, 2019. Association of American Medical Colleges. Physician Specialty Data Report. 2018. Available at https://www.aamc.org/data/workforce/reports/ 457712/2018-specialty-databook.html. Accessed February 25, 2019. Deipolyi AR, Covey AM, Brody LA, Bryce YCD, Li D, Brown KT. Women’s challenges in IR: #ILookLikeAnIR. J Vasc Interv Radiol 2017; 28: 1195–1196.
1875
17. Perez YV, Kesselman A, Abbey-Mensah G, Walsh J. A glance at genderspecific preferences influencing interventional radiology selection. J Vasc Interv Radiol 2016; 27:142–143.e1. 18. Englander MJ, O’Horo SK. Journal club: women in interventional radiology: how are we doing? AJR Am J Roentgenol 2018; 211:724–729. 19. Carr PL, Gunn CM, Kaplan SA, Raj A, Freund KM. Inadequate progress for women in academic medicine: findings from the National Faculty Study. J Womens Health (Larchmt) 2015; 24:190–199. 20. Diaz-Uda A, Medina C, Schill B. Diversity’s new frontier: diversity of thought and the future of the workforce. Deloitte Insights; July 23, 2013. Available at https://www2.deloitte.com/insights/us/en/topics/talent/ diversitys-new-frontier.html. Accessed March 30, 2018. 21. Herring C. Does diversity pay? Race, gender, and the business case for diversity. Am Sociol Rev 2009; 74:208–224. 22. Chafetz JS, Valian V. Why so slow? The advancement of women. Contemp Sociol 1999; 28:42. 23. Martin JL. Ten simple rules to achieve conference speaker gender balance. PLoS Comput Biol 2014; 10:e1003903.
Achieving Speaker Gender Equity at the SIR Annual Scientific Meeting: The Effect of Female Session Coordinators Christine E. Ghatan, MD, Jonathan Altamirano, MS, Magali Fassiotto, PhD, Marcelina G. Perez, MD, Yvonne Maldonado, MD, Shellie Josephs, MD, Daniel Y. Sze, MD, PhD, and Nishita Kothary, MD ABSTRACT Purpose: To examine the impact of targeted efforts to increase the number of female speakers at the Society of Interventional Radiology (SIR) Annual Scientific Meeting (ASM) by reporting gender trends for invited faculty in 2017/2018 vs 2016. Materials and Methods: Faculty rosters for the 2016, 2017, and 2018 SIR ASMs were stratified by gender to quantify female representation at plenary sessions, categorical courses, symposia, self-assessment modules, and “meet-the-expert” sessions. Keynote events, scientific abstract presentations, and award ceremonies were excluded. In 2017, the SIR Annual Meeting Committee issued requirements for coordinators to invite selected women as speakers. Session coordinators are responsible for issuing speaker invitations, and invited speakers have the option to decline. Results: Years 2017 and 2018 showed increases in female speaker representation, with women delivering 13% (89 of 687) and 14% (85 of 605) of all assigned presentations, compared with 9% in 2016 (46 of 514; P ¼ .03 and P ¼ .01, respectively). Gender diversity correlated with the gender of the session coordinator(s). When averaged over a 3-year period, female speakers constituted 7% of the speaker roster (112 of 1,504 presentations) for sessions led by an all-male coordinator team, compared with 36% (108 of 302) for sessions led by at least 1 female coordinator (P < .0001). Results of the linear regression model confirmed the effect of coordinator team gender composition (P < .0001). Conclusions: Having a woman as a session coordinator increased female speaker participation, which suggests that the inclusion of more women as coordinators is one mechanism for achieving gender balance at scientific meetings.
ABBREVIATIONS AMC ¼ Annual Meeting Committee, ASM ¼ Annual Scientific Meeting
From the Radiology Service (C.E.G.), Veterans Affairs Palo Alto Health Care System, Palo Alto, California; and Office of Faculty Development and Diversity (J.A., M.F., Y.M.), Department of Radiology (M.G.P.), and Divisions of Pediatric Radiology (S.J.) and Interventional Radiology (D.Y.S., N.K.), Stanford University School of Medicine, 300 Pasteur Dr., H3630, Stanford, CA 94305-5642. Received April 30, 2019; final revision received June 28, 2019; accepted July 8, 2019. Address correspondence to N.K.; E-mail: [email protected]; Twitter handle: @Fivenada From the SIR 2019 Annual Scientific Meeting. N.K. receives research grants from Siemens (Siemens Healthcare) and EchoPixel (Santa Clara, California). None of the other authors have identified a conflict of interest. © SIR, 2019 J Vasc Interv Radiol 2019; 30:1870–1875 https://doi.org/10.1016/j.jvir.2019.07.006
Women continue to be underrepresented in the sciences, especially at the higher ranks (1). Lower visibility of female scientists compared with male scientists is one of several reasons for the underrepresentation of women progressing into senior ranks (2). Female speakers are outnumbered by male speakers at national and international meetings in many scientific disciplines (3–8), including fields in which women represent the majority (9,10). The urgency of this issue was recently underscored by the director of the National Institutes of Health, who, in a published statement, called to “end the tradition of all-male speaking panels” so that “scientists of all backgrounds are evaluated fairly for speaking opportunities” (11). Speaking invitations and the resultant visibility at major scientific meetings are an important consideration for rank advancement in academia (5). Further, speaking assignments can provide an opportunity to showcase one’s expertise and research interests,
Volume 30 ▪ Number 11 ▪ November ▪ 2019
thereby providing a path to expand the speaker’s professional network. Visibility of underrepresented groups at the podium may also facilitate more diverse recruitment into the field by signaling to audience members that a diverse group of people can find success in a given discipline (5). Recruitment of women and other underrepresented groups into interventional radiology has been an ongoing challenge (12). Recognizing the impact of visibility on career advancement, retention, and recruitment, efforts to increase the participation of women in major meetings have become part of the effort to address gender disparities (5). In 2017, acknowledging the prevailing gender imbalance within SIR and at the podium, the Annual Meeting Committee (AMC), in conjunction with SIR’s Diversity and Inclusion Committee, provided session coordinators with the names of female interventional radiologists to include as speakers in their sessions. These invitations were required, but they represented only part of the overall roster. Additional speaker selection, session curriculum, and the assignment of topics to individual speakers remained at the discretion of session coordinators. The present study investigates the impact of the AMC’s initiative by reporting the gender trends for invited faculty at the 2017 and 2018 ASMs, examines the factors influencing gains in gender diversity, and provides a framework for future change.
MATERIALS AND METHODS This study did not require institutional review board approval. Archived faculty rosters for the SIR ASMs held in 2016, 2017, and 2018 were reviewed. The analysis was limited to the following sessions: plenary sessions, symposia, categorical courses, self-assessment modules/self-assessment continuing medical education, and “meet-the-expert” sessions (henceforth collectively referred to as “podium sessions”). Keynote events and award ceremonies, as well as nondidactic events (scientific abstract presentations, case-based workshops, and handson workshops), were excluded. Included faculty speakers and session coordinators were stratified by gender (male/female) by using existing SIR member data and institutional websites. Up to and including the 2016 ASM, the AMC invited session coordinators by peer-to-peer recommendations, and these coordinators were responsible for selecting speakers for their sessions, with invitations issued at their discretion. Starting in 2017, even though the means for selecting session coordinators remained unchanged, the AMC provided session coordinators with the names of female interventional radiologists to include as speakers in their sessions. The AMC identified prospective female faculty members based on familiarity with published work and/or peer-to-peer recommendations. Session coordinators were required to invite these individuals (typically 1 or 2 names were provided); selections for the remainder of the session’s speaker roster were left to the discretion of the coordinator. To measure the impact of these recommendations and to examine speaker representation as a function of the gender
1871
composition of the coordinator(s) leading the session, the data were further classified as to whether the session was led by an all-male team or a combination with at least 1 female coordinator.
Statistical Analysis Comparisons of the numbers of female speakers were made across year and by coordinator gender by using c2 tests of independence. Average numbers of female speakers at podium sessions by coordinator gender were analyzed with Wilcoxon– Mann–Whitney tests to account for the nonparametric distribution of data. Hierarchic linear regression models were built, one including only conference year and one including conference year and coordinator gender, to analyze the effects of these predictors on the number of female speakers. A P value < .05 was considered significant.
RESULTS From 2016 to 2018, the SIR ASM had 96, 118, and 105 podium sessions that resulted in totals of 514, 687, and 605 speaker assignments accepted by 319, 413, and 338 unique speakers, respectively (Table 1). In line with SIR’s initiative to promote gender diversity, 2017 and 2018 showed increases in numbers of female speakers: 89 of 687 (13%) and 85 of 605 (14%) podium presentations in 2017 and 2018, respectively, were delivered by female speakers, compared with 46 of 514 presentations (9%) in 2016 (P ¼ .03 and P ¼ .01, respectively, c2 test). Gains in gender diversity were not homogeneously distributed across all sessions, and correlated with the gender of the session coordinator(s) (Table 2). Over the 3year period, female speakers constituted 7% of the speaker roster for sessions led by an all-male coordinator team, delivering a total of 112 of the 1,504 podium presentations. In sessions led by at least 1 female coordinator, women delivered 108 of the 302 podium presentations, constituting 36% of the speaker roster, a statistically significant increase in representation compared with sessions led by an all-male coordinator team (P < .0001). In 2017 and 2018, there were overall increases in the absolute numbers of female speakers irrespective of the gender composition of the coordinator team. However, the proportion of presentations delivered by women in sessions coordinated by an all-male team remained 8% in 2017 and 2018, and was statistically unchanged compared with 2016 (6%; P ¼ .57; Fig a). In contrast, the proportion of presentations delivered by women in sessions with at least 1 female coordinator increased to 38% in 2017 and 37% in 2018 compared with 28% in 2016, even though the increase did not reach statistical significance (P ¼ .40; Fig b). Table 3 shows the results of the hierarchical linear regression model to predict the change in the number of female speakers. Variables included were the year of the annual meeting followed by the gender composition of the session coordinators. The first model demonstrates that the 2017 and 2018 ASMs were significantly and
1872 ▪ Female Session Coordinators at the SIR Annual Meeting
Ghatan et al ▪ JVIR
Table 1. Distribution of Male and Female Speaker Assignments by Year Session Type
2016 Speakers (514 Assignments)
2017 Speakers (687 Assignments)
2018 Speakers (605 Assignments)
Male
Female
Male
Female
Male
8 (2)
0
13 (2)
1 (< 1)
11 (2)
Symposia
53 (10)
9 (2)
115 (17)
Categorical
175 (34)
17 (3)
185 (27)
29 (4)
210 (35)
37 (6)
90 (18)
6 (1)
124 (18)
17 (2)
106 (18)
22 (4)
Meet the expert
142 (28)
14 (3)
161 (23)
14 (2)
105 (17)
10 (2)
Total
468 (91)
46 (9)
598 (87)
89 (13)*
520 (86)
85 (14)†
Plenary
SAM/SA-CME
28 (4)
88 (15)
Female 2 (< 1) 14 (2)
Note–Values in parentheses are percentages. SAM/SA-CME ¼ self-assessment modules/self-assessment continuing medical education. *Significant at P ¼ .03. † Significant at P ¼ .01.
Table 2. Distribution of Male and Female Speaker Assignments for Sessions Based on Gender Composition of Coordinator Team Session
≥ 1 Female Coordinator
All-Male Team Male
Female
Sessions
Male
Female
Sessions -
Plenary 2016
8
0
2
-
-
2017
10
1
1
3
0
1
2018
11
2
3
-
-
-
2016 2017
44 100
4 17
3 12
9 15
5 11
1 6
2018
80
11
11
8
3
1
Symposia
Categorical course 2016
164
13
33
11
4
3
2017
161
11
29
24
18
8
2018
169
11
29
41
26
9
Self-assessment modules 2016 2017
80 104
4 5
16 18
10 20
2 12
2 5
2018
78
7
15
28
15
6
Meet the expert 2016
129
8
32
13
6
4
2017
150
10
34
11
4
4
2018
104
8
30
1
2
1
All sessions combined 2016 2017
425 525
29 (6) 44 (8)
86 94
43 73
17 (28) 45 (38)
10 24
2018
442
39 (8)
88
78
46 (37)
17
Note–Values in parentheses are percentages. *P < .0001 for each year (c2 test).
positively associated with the number of female speakers (P ¼ .027). However, after the addition of gender composition of the coordinator team (all male vs at least 1 woman), the effect of the year disappeared (P ¼ .051), whereas the effect of the gender composition of the coordinator team was highly significant (P < .0001), with
the presence of a female coordinator having a strong positive impact on the number of female speakers.
DISCUSSION The present study highlights the partnership among SIR leadership, the Diversity and Inclusion Committee, and the AMC to advance women in interventional radiology and demonstrates that a small change in the invitation strategy can increase the proportion of female speakers in a short time frame. The results also uncover the effect of female session coordinators, as women were more likely to serve as invited speakers when there was at least 1 female coordinator in a session. This suggests that prudent selection to include more female interventional radiologists in leadership roles will have a downstream impact on creating a more diverse speaker roster. The data fit a larger pattern of recent work in other scientific disciplines (3–10,13) in which invited female speakers are outnumbered by male speakers and the presence of female session coordinators correlates with increased gender parity among speakers. Examination of speaker rosters within diverse fields like neuropharmacology, microbiology, social psychology, anthropology, and evolutionary biology, for example, found that fewer women spoke at major meetings than men (5–9,13). This discrepancy extended into fields with gender parity and fields with a female majority. Similar to the results presented here, these studies (4,5,7–9,13) noted that the inclusion of more women as session coordinators correlated with a greater number of female speakers. Increasing women’s visibility at the podium can impact recruitment and retention, playing a critical role in plugging the “leaky pipeline,” a metaphor that describes the high attrition rate of women in the sciences as they progress along the ranks of academia (2). Although women have constituted more than 40% of medical school matriculants for more than 25 years (14), women continue to be underrepresented in interventional radiology (15), constituting 15.4% of graduating fellows but only 9.5% of practicing
Volume 30 ▪ Number 11 ▪ November ▪ 2019
1873
Figure. Distribution of male versus female speakers by year, comparing all-male coordinator teams versus coordinator teams with at least 1 female member. (a) From 2016 to 2018, all-male teams coordinated a majority of the sessions and showed a zero slope in the number of female speakers over the 3-year period (86, 94, and 88 sessions, respectively). (b) During the same time period, coordinator teams with at least 1 female member exhibited an upward slope for the number of female speakers recruited despite the fewer number of sessions coordinated by them (10, 24, and 17, respectively).
1874 ▪ Female Session Coordinators at the SIR Annual Meeting
Table 3. Linear Regression Model Examining Effect of Gender Composition of Coordinator Team Variable
Model 1
b (95% CI) Year
0.16 (0.02–0.31)
Model 2
P Value .027
b (95% CI)
P Value
0.12 (–0.0004 to 0.23)
.051
< .0001
Coordinator team 1 woman All male (ref. group)
– –
– –
1.68 (1.43–1.94) –
CI ¼ confidence interval.
interventional radiology physicians (12). Although factors such as long hours and exposure to radiation have been implicated, inadequate mentorship and gender bias also have been suggested to be related to the poor recruitment and retention of female interventional radiologists (16–18). Academic promotion has been tied to retention of women physicians (19). By providing evidence of external recognition and scholarly achievement necessary for academic promotion, speaking invitations and visibility at national meetings may indirectly play a role in retention (5). In addition, they provide networking opportunities and allow dissemination of research advances that can lead to new employment and research and clinical collaborations (2). Speaker diversity promotes wider recruitment by providing audience members with conscious and subconscious cues as to who can be successful in the field (5). Diversity initiatives today are no longer aimed solely at fairness for underrepresented groups. Research in management science and organizational behavior shows that diversity leads to adaptability, creativity, and innovation, providing a workforce that often outperforms homogenous organizations (20). Data have also demonstrated that companies with more gender-diverse leadership achieve better financial returns (21). Greater diversity at the podium and a broader exchange of ideas among all participants promote these overarching principles and help maintain interventional radiology as a competitive, cutting-edge, and innovative medical specialty. It is unclear why the gender of the coordinator has such an impact, and assumptions of causation are difficult to prove. Although the correlation between female coordinators and the proportion of female speakers may suggest implicit bias against female speakers by male coordinators (22), it could also indicate that female coordinators have larger networks of female colleagues from which to invite speakers. Alternatively, the “imposter syndrome,” a phenomenon whereby women in science underestimate their ability and performance (4), may prompt female invitees to decline an invitation, especially one from a well-known male colleague (4,8). The present study provides insight into diversity and gender parity at the SIR ASM, but it has limitations. The data were derived from a short period of time, and it is
Ghatan et al ▪ JVIR
therefore unclear if gains were solely a result of interventions in 2016 or part of a larger trend unrelated to the AMC initiative. Sampling was obtained from speaker rosters at the SIR ASM only, and cannot be extrapolated to other international, national, local, or regional interventional radiology conferences. Data regarding which female faculty members were appointed by the AMC and which female faculty members participated in the ASM independent of AMC recommendations are not available. Speaker information captured in the program rosters reflects accepted invitations only and does not take into account declined invitations and last-minute substitutions. Demographic information, including gender assignment, was curated from information volunteered by SIR members, is limited to a binary gender assignment and does not capture the full spectrum of gender identity. Quantitative assessment of equity based on an absolute number of speakers is unidimensional and does not take into account details such as audience size, length of presentations, time of day, and other factors that may reflect the prestige of an invitation. The present study confirms that progress has been made since the implementation of AMC initiatives to increase gender diversity, and that a targeted invitation strategy can improve the visibility of an underrepresented group at a major scientific meeting in a short period of time. These efforts are in accordance with strategies suggested in other scientific fields (23). Female faculty members leading and coordinating sessions resulted in a more gender-diverse speaker roster, which suggests that the inclusion of more female coordinators and moderators is one mechanism for achieving gender balance at scientific meetings.
REFERENCES 1. Ceci SJ, Williams WM, Barnett SM. Women’s underrepresentation in science: Sociocultural and biological considerations. Psychol Bull 2009; 135:218–261. 2. Pell AN. Fixing the leaky pipeline: women scientists in academia. J Anim Sci 1996; 74:2843–2848. 3. Casadevall A. Achieving speaker gender equity at the American Society for Microbiology General Meeting. mBio 2015; 6:e01146-15. 4. Ford HL, Brick C, Blaufuss K, Dekens PS. Gender inequity in speaking opportunities at the American Geophysical Union Fall Meeting. Nat Commun 2018; 9:1358. 5. Johnson CS, Smith PK, Wang C. Sage on the stage: women’s representation at an academic conference. Pers Soc Psychol Bull 2017; 43: 493–507. 6. Jones TM, Fanson KV, Lanfear R, Symonds MRE, Higgie M. Gender differences in conference presentations: a consequence of self-selection? PeerJ 2014; 2:e627. 7. Moghaddam B, Gur RE. Women at the podium: ACNP strives to reach speaker gender equality at the Annual Meeting. Neuropsychopharmacology 2016; 41:929–931. 8. Schroeder J, Dugdale HL, Radersma R, et al. Fewer invited talks by women in evolutionary biology symposia. J Evol Biol 2013; 26: 2063–2069. 9. Isbell LA, Young TP, Harcourt AH. Stag parties linger: continued gender bias in a female-rich scientific discipline. PLoS ONE 2012; 7: e49682. 10. Simon JL, Morris EK, Smith NG. Trends in women’s participation at the Meetings of the Association for Behavior Analysis: 1975-2005. Behav Anal 2007; 30:181–196. 11. Collins FS. Time to end the Manel tradition. National Institutes of Health. 2019. Available at https://www.nih.gov/about-nih/who-we-
Volume 30 ▪ Number 11 ▪ November ▪ 2019
12.
13.
14.
15.
16.
are/nih-director/statements/time-end-manel-tradition. Accessed June 23, 2019. Higgins MCSS, Hwang WT, Richard C, et al. Underrepresentation of women and minorities in the United States IR academic physician workforce. J Vasc Interv Radiol 2016; 27:1837–1844.e2. Casadevall A, Handelsman J. The presence of female conveners correlates with a higher proportion of female speakers at scientific symposia. mBio 2014; 5:e00846-13. Association of American Medical Colleges. Longitudinal applicant, matriculant, enrollment, & graduation tables. Available at http://www. aamcdiversityfactsandfigures2016.org/report-section/applicants-enrollment/ #tablepress-12. Accessed March 28, 2019. Association of American Medical Colleges. Physician Specialty Data Report. 2018. Available at https://www.aamc.org/data/workforce/reports/ 457712/2018-specialty-databook.html. Accessed February 25, 2019. Deipolyi AR, Covey AM, Brody LA, Bryce YCD, Li D, Brown KT. Women’s challenges in IR: #ILookLikeAnIR. J Vasc Interv Radiol 2017; 28: 1195–1196.
1875
17. Perez YV, Kesselman A, Abbey-Mensah G, Walsh J. A glance at genderspecific preferences influencing interventional radiology selection. J Vasc Interv Radiol 2016; 27:142–143.e1. 18. Englander MJ, O’Horo SK. Journal club: women in interventional radiology: how are we doing? AJR Am J Roentgenol 2018; 211:724–729. 19. Carr PL, Gunn CM, Kaplan SA, Raj A, Freund KM. Inadequate progress for women in academic medicine: findings from the National Faculty Study. J Womens Health (Larchmt) 2015; 24:190–199. 20. Diaz-Uda A, Medina C, Schill B. Diversity’s new frontier: diversity of thought and the future of the workforce. Deloitte Insights; July 23, 2013. Available at https://www2.deloitte.com/insights/us/en/topics/talent/ diversitys-new-frontier.html. Accessed March 30, 2018. 21. Herring C. Does diversity pay? Race, gender, and the business case for diversity. Am Sociol Rev 2009; 74:208–224. 22. Chafetz JS, Valian V. Why so slow? The advancement of women. Contemp Sociol 1999; 28:42. 23. Martin JL. Ten simple rules to achieve conference speaker gender balance. PLoS Comput Biol 2014; 10:e1003903.