Disease Site Specialization in the Academic Radiation Oncology Workforce: Evidence of Gender Differences

Abstract

Purpose:

Because some stakeholders within medicine seek to diversify and attain greater workforce equity, it is critical to understand gender-based divisions within specialization. Radiation oncology (RO) has one of the smallest proportions of women representation of all specialties, and to our knowledge, no prior studies have investigated gender differences in all the disease site specializations within RO. Thus, we analyzed the relationship between gender and disease site(s) treated in academic RO (ARO).

Methods and Materials:

Faculty gender and disease site(s) treated by faculty from ARO departments were collected via publicly available department websites in January 2020. X² analyses were conducted to assess differences between the proportions of women faculty treating each disease site.

Results:

Of 1337 ARO faculty, 408 (30.5%) were identified as women. Breast, gynecology, and pediatrics had the largest proportions of women faculty (all >40%; P < .001). A majority (53%; P < .001) of women ARO faculty treated breast. Genitourinary, thoracic, and head and neck had the smallest proportions of women faculty (all <25%; P < .001). Women ARO faculty were twice as likely to treat breast and gynecologic malignancies compared with men faculty (risk ratio [RR] with 95% CI, 2.01 [1.75–2.50]; P < .001 and RR [95% CI], 2.06 [1.72–2.79]; P <.001, respectively). Men ARO faculty were 3 times more likely to treat genitourinary cancer compared with women faculty (RR [95% CI], 0.40 [0.34–0.48]; P < .001). There was no difference in the mean number of disease sites treated between women and men ARO faculty (2.63 vs 2.53; P = .29).

Conclusions:

Gender differences in disease site specialization were observed in ARO. Future research into the drivers of disease site selection should be explored. © 2024 Elsevier Inc. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

Introduction

Among nonsurgical specialties, radiation oncology has one of the smallest proportions of women representation (27.4% in 2020).¹ Previous studies have shown an increase in the number of women in radiation oncology over time; however, the representation of women treating each disease site within radiation oncology is not well known.² A prior study explored the role of gender in differential specialization in radiology and found a higher rate of women in mammography, pediatric radiology, and sonography subspecialties.³ Similar gender differences in subspecialization have also been found within medical oncology, with more women medical oncologists treating breast cancer and fewer treating genitourinary (GU) cancer.⁴ Subsequently, one study focused on radiation oncology and found that nearly half of radiation oncologists treating breast cancer were women while less than one-sixth of those treating GU cancer were women; however, additional disease sites were not addressed.⁵ Thus, we aimed to examine gender differences within all disease site specializations in radiation oncology by analyzing the relationship between gender and disease site(s) treated in academic radiation oncology (ARO) faculty.

Methods and Materials

In January 2020, an internet-based search of information from Accreditation Council for Graduate Medical Education (ACGME)-accredited ARO departmental websites was performed. All faculty physicians at the institution’s main campus or satellite clinics were included for analysis. The authors imputed binary gender (defined as man or woman) for faculty as self-reported gender was not available. A multifaceted approach was used to determine faculty and department chair gender based on the assessment of name, the use of pronouns in the website verbiage, images available on departmental websites, and the Gender Balance Assessment Tool. The institutional websites did not state any individuals who identified as transgender or nonbinary. Discrepancies were resolved by investigator review of all available resources. Disease site specialization was determined by review of departmental or faculty websites, and the number and type of disease sites treated by each faculty member were recorded; faculty treating multiple disease sites were included for analysis. Faculty members without disease site specialization listed on departmental or faculty websites were excluded from the analysis. Pearson’s X² testing was used to assess differences in the proportion of women ARO faculty by disease site. P values <.05 were considered statistically significant. Statistical analyses were performed using SPSS software (version 24.0, IBM SPSS Inc). The study was deemed exempt, and consent was waived by the institutional review board, given the use of publicly available data.

Results

Ninety-one departments with 1337 full-time ARO faculty were included in the initial cohort. One hundred fifty-three (11.4%) faculty did not have disease site specialization listed on departmental or faculty websites and were excluded, leaving 1182 ARO faculty in 88 departments in the analytic cohort. Four-hundred-eight (30.5%) ARO faculty were identified as women.

The distribution of each disease site by gender is displayed in Figure 1. Breast, gynecology, head and neck (H&N), thoracic, pediatrics, and GU cancer disease sites had the strongest association with representation by faculty gender. Breast (47.1%), gynecology (46.7%), and pediatrics (40.7%) cancers had the largest proportion of women faculty (P < .01). GU (14.6%), thoracic (23.7%), and H&N (23.2%) cancers had the smallest proportion of women faculty (P < .01). Relative risk analysis showed that women ARO faculty were twice as likely to treat breast and gynecologic malignancies compared with men (risk ratio [RR] with 95% CI, 2.01 [1.75–2.50]; P < .001 and RR [95% CI], 2.06 [1.72–2.79]; P < .001, respectively; Table 1). Conversely, male faculty were nearly 3 times more likely to treat GU cancer compared with female faculty (RR [95% CI], 0.40 [0.34–0.48]; P < .001; Table 1).

Fig. 1.

Distribution of disease site specialization by gender in academic radiation oncology. Percentages by gender reflect the proportion of faculty treating that disease site. Overall percentages reflect the proportion of academic radiation oncologists that treat a given disease site. Asterisks denote a significant X² test of the proportion of men vs women treating each disease site compared with not treating that disease site (all P < .01). Note that the summed percentages for each gender (as well as pooled among all physicians in the “All” category) exceed 100% as many individuals treat multiple disease sites.

Table 1.

Relative risk of women versus men gender treating each disease site in academic radiation oncology

Disease site	Risk ratio (95% CI)	P value
Breast	2.01 (1.75–2.50)	<.001*
Genitourinary	0.40 (0.34–0.48)	<.001*
Gastrointestinal	0.84 (0.71–1.01)	0.08
Thoracic	0.73 (0.61–0.87)	0.003*
Hematologic	1.04 (0.87–1.25)	0.37
Gynecological	2.06 (1.72–2.79)	<.001*
Sarcoma	0.94 (0.79–1.13)	0.47
Skin	0.91 (0.76–1.08)	0.29
Head and neck	0.71 (0.59–0.85)	0.002*
Central nervous system	0.91 (0.76–1.09)	0.212
Pediatrics	1.61 (1.35–1.93)	0.005*

A risk ratio > 1.0 represents greater representation by women, and a risk ratio < 1.0 represents greater representation by men.

^*Asteriks denote a significant p-value from relative risk analysis of the liklihood of women vs men treating each disease site.

There was no significant difference in the proportion of women versus men treating one disease site compared with more than one disease site (8.3% vs 18.4%; P = .14) or in the mean number of disease sites treated between women versus men ARO faculty (2.63 vs 2.53; P = .29).

Discussion

This study investigated gender representation in ARO and found that gender differences exist in disease site specialization. Specifically, we identified that a higher proportion of women AROs treat breast, gynecologic, and pediatric cancers (ie, all specialties with a socially perceived association with the woman gender), and a lower proportion of women AROs treat GU, thoracic, and H&N cancers; this is consistent with prior research showing gender differences in subspecialization in other medical fields.⁶ Finally, our findings corroborate previous findings of underrepresentation of women in ARO but demonstrate positive trends with increasing numbers of women AROs from 23.8% in 2010 and 27.7% in 2015 to the 30.5% observed at the time of our data collection in 2020.^7,8

Although we identified that gender differences in disease site specialization exist in ARO, this study did not investigate the etiologies of this finding. Future research is needed to examine this disparity and explore driving factors of gender differences in ARO. One potential driving factor is the comfort level and affinity of physicians and patients to work with others of the same gender, especially regarding sensitive topics and body areas. Another possible driving factor of this disparity may be differences in disease site reimbursement, given known gendered disparities in radiation oncology salaries.⁹ Additionally, there may be an effect of existing gender dominance in a disease site that attracts faculty of the same gender to that disease site to experience kinship, mentorship rooted in shared experiences, and potentially increased advocacy for career growth.^10–13 Finally, another factor to be considered as potentially driving gender differences in ARO is the role of disease site leadership and institutional support for gender equity as differences in support, such as parental leave policies, leadership term limits, support for national conference attendance, and equity in research funding, may impact genders differently when choosing their disease site specialization.^14,15 Overall, many factors may be at play in driving this disparity, and a meaningful investigation of the causative structural, sociological, and societal factors of gender differences in disease site specialization in ARO is needed.

This study has some limitations. First, as a cross-sectional analysis examining the pattern of ARO disease site specialization by gender in 2020, the current landscape of disease site specialization and potential changes in disease site subspecialization over time is not known. Additionally, this analysis was restricted to academic centers and, thus, is not generalizable to the entire radiation oncology community. Moreover, our results are limited due to the exclusion of ARO faculty without disease site specialization listed, although this was a small proportion of the total sample size. Overall, this data serve as a benchmark for future efforts. Because these data were based on the availability and accuracy of information on departmental websites, it is conceivable that website-based information was inaccurate or not updated. Finally, this method of data acquisition is rooted in the gender binary, which does not take into account the role of other gender identities. In a prior study of self-reported gender in the American Society for Radiation Oncology Annual Meeting, speakers reflected at least one respondent who identified as nonbinary.¹⁶ Future studies should consider collecting self-reported gender to accurately reflect gender identity. Finally, because this analysis was based on the assumption of gender and gender concordance, it could underestimate the role of gender identity in disease site preference and woman representation in ARO.

Conclusions

Despite the overall underrepresentation of women faculty members in ARO, there is a higher proportion of women treating disease sites that are typically associated with female gender, including breast, gynecologic, and pediatric cancers. Conversely, there is a lower proportion of women AROs treating GU, thoracic, and H&N cancers. These gender differences in disease site specialization warrant further investigation into the drivers of disease site selection by ARO faculty.

Data Sharing Statement:

Research data are stored in an institutional repository and will be shared upon request to the corresponding author.