Accepted Manuscript Title: What's Gender Got to Do with It: Difference in the Proportion of Traditionally Female Cases Performed by General Urologists of Each Gender Author: Katherine Rotker, Sarah Iosifescu, Grayson Baird, Simone Thavaseelan, Kathleen Hwang PII: DOI: Reference:
S0090-4295(18)30195-X https://doi.org/10.1016/j.urology.2017.12.040 URL 20932
To appear in:
Received date: Accepted date:
Please cite this article as: Katherine Rotker, Sarah Iosifescu, Grayson Baird, Simone Thavaseelan, Kathleen Hwang, What's Gender Got to Do with It: Difference in the Proportion of Traditionally Female Cases Performed by General Urologists of Each Gender, Urology (2018), https://doi.org/10.1016/j.urology.2017.12.040. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
What’s Gender Got to Do With It: Difference in the Proportion of Traditionally Female Cases Performed by General Urologists of Each Gender Katherine Rotker, M.D. (Corresponding Author) Assistant Professor of Urology Department of Urology University of Massachusetts Medical School Worcester, MA [email protected]
Sarah Iosifescu, M.D. Resident Physician Mount Sinai Medical Center New York NY
Grayson Baird, Ph.D. Department of Biostatistics Lifespan Medical Center Providence, RI
Simone Thavaseelan, M.D. Assistant Professor of Urology Division of Urology Brown Medical School Providence, RI
Kathleen Hwang, M.D. Assistant Professor of Urology Division of Urology Brown Medical School Providence, RI
*This data was previously presented at the New England Section Meeting of the American Urologic Association (September 30, 2016 in Portland, ME) and the abstract only was published in materials for this meeting. No other presentations, print or online publications. *The authors report no external funding source for this study. *The authors declare no conflict of interest. Abstract Word Count: 242 Manuscript Word Count: 1,967 Keywords: gender differences, career choices, practice patterns ABSTRACT: Objective: To examine surgical case volume characteristics in certifying urologists to evaluate practice patterns, given the long-standing understanding but unproven hypothesis that non-
Page 1 of 19
fellowship trained female general urologists perform more urogynecologic procedures compared to their equally trained male counterparts.
Materials and Methods: Case log data from certifying and recertifying urologists from 2000 to 2015 was obtained from the American Board of Urology (ABU). 37 CPT codes were chosen to represent traditionally urogynecologic cases. Logistic regression analysis models were used to determine the percentage of total CPT codes logged during the certification period made up by traditionally urogynecology cases. Male and female non-fellowship trained, self-described general urologists were compared.
Results: The case logs of 4,032 non-fellowship trained general urologists were reviewed from 2000-2015, 297 of whom were female and 3,735 of whom were male. Urogynecology cases made up 1.27% of the total CPT codes logged by the females and 0.59% of those codes logged by the males (p<0.001); an increase of 2.2 times (p<0.001). This statistically significant difference persisted regardless of certification period, geographic location, population density or full-time versus part-time employment.
Conclusion: Traditional urogynecology cases represented a significantly greater percentage of the total cases logged by non-fellowship trained female general urologists when compared to their non-fellowship trained, generalist male colleagues. The percentage of total cases performed by both is very small. However, it supports a belief that patient populations differ for male and female general urologists, which may impact training/career choices.
Page 2 of 19
INTRODUCTION: Urology has long existed as a predominantly male field. As recently as 1981, only 34 females were practicing nationwide. From 1981 to 2009 the number of females practicing urology increased 1000% to 512. The trend towards increasing numbers of females in urology shows no signs of slowing. The gender distribution amongst residents in training continues to shift toward equality. The percentage of female residents increased from 5% in 1989 to 23% in 2011. The most recent match data in 2017 show that 25.9% of matched residents in urology were female.
Along with the incredible growth and change has been an opportunity to study gender differences in specialty training, practice patterns, operative patterns and pay disparity. A recent article widely publicized in the lay press even suggested a health benefit to being treated by female hospitalists. A perception exists that female urologists more often take care of female urologic issues. Cited reasons for this disparity are patient preferences, referral patterns and subspecialty training. Some research has been performed looking at the increasing number of female urologists performing subspecialty training in urogynecology fellowships. However, to our knowledge, research looking into performance and practice patterns for those female urologists who do not seek out careers in urogynecology are lacking.
Given the ever-changing face of the field for females in urology training, career planning can be difficult. As of 2012, males still made up 92% of the urology workforce making
Page 3 of 19
access to role models difficult. Additionally, training programs are generally located in major academic centers where sub-specialists abound. Therefore, obtaining a realistic picture of a community general urology practice can be difficult for any urologist in training.
We examined surgical case volume characteristics in certifying urologists associated with common female urologic procedures. We sought to evaluate practice patterns, given the long-standing belief but unproven hypothesis that non-fellowship trained female general urologists perform more common female urologic procedures compared to their equally trained male general urology counterparts.
MATERIALS AND METHODS: The American Board of Urology (ABU) serves as the specialty certifying board for the field of urology. As part of the certification process, urologists must complete case logs reflecting all patients seen and operated on over a consecutive six-month period in practice. For urologists certified after 1985, following initial certification, recertification is mandatory at 10-year intervals, which also includes an examination and completion of a 6-month case log. The ABU maintains a record of these certifications and logs that record specific patient demographics (age, gender) and extensive surgeon characteristics including age, gender, certification cycle, practice area, practice type. Surgeons are also asked to self-report as a general urologist or sub-specialist in a listed field.
Page 4 of 19
In this study, we analyzed ABU case logs from 2000-2015. Only case logs for those surgeons who self-reported as general urologists were included in the study. Surgical procedures are logged by CPT code system. For this study, 37 common codes representing urogynecologic cases were chosen. (Appendix A) The total number of CPT codes logged by each provider was also provided. Surgeon data including gender, certification cycle, geographic area, population density, practice type, and part-time versus full-time employment was also evaluated for trends. Case log reports are received from the ABU in annualized format and represent a 1-year representation of individual surgeon practice volume based on a 6-month operative period. This methodology has been used in a number of previous reports of practice patterns.[5-7] Institutional review board approval was obtained from the Lifespan IRB.
Proportion of CPT codes were examined by surgeon gender along with certification cycle, geographic location, and practice type using generalized linear modeling assuming a binomial distribution, where the outcome was defined as total number of CPT codes by total number of codes for each surgeon. Alpha was established a priori at the .05 level and all interval estimates were calculated for 95% confidence. Multiple comparisons were examined using the Holm-Tukey method. All analyses were done using SAS Software 9.4 (SAS Inc., Cary, NC) using PROC GLIMMIX and FREQ.
RESULTS: The case logs of 4,032 non-fellowship trained, general urologists were reviewed from 2000-2015, 297 of whom were female and 3,735 of whom were male. Urogynecology
Page 5 of 19
cases made up 1.27% [1.24-1.30] of the total CPT codes logged by the females and 0.59% [0.59-0.60] of those codes logged by the males (p<0.001), thus making the percentage of total CPT codes logged that were urogynecologic 2.2 times greater for females than males. [Table 1]
Within each certification period, urogynecology cases represented a statistically significantly higher percentage of the total codes logged for females than for males. This held true for those undergoing initial certification, initial re-certification, secondrecertification. It is notable that; no females were in their third recertification period. [Table 1, Figure 2]
When the data were evaluated by geographic region, the statistically significant difference held true regardless of region including New England, Mid-Atlantic, North Central, South Central, South Eastern and Western region. These differences were all statistically significant (p<0.001). [Table 1, Figure 1]
When looking at the population density of the practice area, again the statistically significant difference held true for each grouping except those who practiced in the smallest population area. In the least populated areas, the urogynecology cases logged for females was 1.29% and 0.7% for males. In regions with 100,000-250,000 people, 0.86% of the total cases logged by female general urologists were urogynecology cases compared to 0.63% for male general urologists. In areas with 250,001-500,000 or 500,000-1,000,000 people, 1.02% and 1.39% of total cases logged by females were
Page 6 of 19
urogynecology cases respectively and 0.58% and 0.53% of total cases logged by males were urogynecology cases respectively (both p<0.01). In the most populous areas with >1,000,000 people, the percentage of the total cases logged made up by urogynecology cases was 1.55% for females and 0.54% for males (p<0.01). [Table 1, Figure 1]
When evaluating practice type, the statistically significant difference held true for those in private-practice and hospital employed positions. The trend held but without statistical significance for those in academic positions, military positions, and state/local government positions. The trend reversed for the very few people in veteran’s hospitals or other positions. For those general urologists practicing in an academic setting, the percentage of total cases made up by urogynecology cases was 1.00% for females and 0.65% [0.61-0.69] for males. In private practice, again females had 1.24% of their cases and males had 0.58% made up by urogynecology cases (p<.001). For salaried hospital employee general urologists, females had 1.54% of their cases and males had 0.73% made up by urogynecology cases (p<.01). For general urologists in military practice, the percentage of cases made up by urogynecology cases was 1.47% for females and 0.70% for males. However, in those practice types where the number of general urologists was extremely low, the statistical significance was lost or the trend was reversed. For the single female, general urologist who reported practicing in state/local government, the percentage of cases made up by urogynecology cases was 0.97% compared to 0.80% for the 2 males. For those general urologists practicing in a Veterans Hospital, the percentage of total cases made up by urogynecology cases was 0.81% for 4 females and 0.85% for the 8 males. For those who responded other to the
Page 7 of 19
question, the only female had 0.53% of her cases and the 8 males had 1.04% made up by urogynecology cases. [Table 1, Figure 2]
Lastly, we looked at full-time versus part-time employment. For general urologists in fulltime practice, the percentage of total cases logged made up by urogynecology cases was 1.24% for females and 0.59% for males (p<.001). For those in part-time practice, the percentages were 1.52% and 0.70% for females and males respectively (p<.001). [Table 1, Figure 1]
DISCUSSION: When training in an academic center where sub-specialists abound, it can be difficult for residents to obtain a clear understanding of the practice patterns of general urologists in the communities. A belief existed among the females in our training program that if a woman entered a general practice, she would see a disproportionate percent of female patients and perform a disproportionate percent of urogynecologic cases when compared to her male counterparts. For some, this may be desirable or at least not a deterrent to entering general practice. For others, for whom urogynecology holds no special interest, this may be a deterrent to entering a general practice and may even be a reason to seek additional fellowship training. This fellowship training may afford the female urologist the ability to tailor even a general practice toward cases of greater interest to them.
Page 8 of 19
In the present study, we compared the case logs of non-fellowship trained urologists who self-identified as general urologists of both genders. We measured the percentage of total CPT cases logged that were made up by common urogynecology codes, for each urologist, and found the percentage for females to be higher, on average, than for males. The total percentage performed for both genders was low. However, the statistically significant differences between the two groups remained regardless of certification period, geographic location, practice area population density, practice type and part-time versus full-time employment.
Previous studies have shown a disproportionate percentage of pelvic organ prolapse cases are performed by urologists of female gender and that a growing proportion of urogynecologic cases are performed by female subspecialists, with a significant increase since initiation of Female Pelvic Medicine and Reconstructive Surgery (FPMRS) subspecialty certification. This study differs from these previously published reports in that it sought to determine a practice difference amongst general urologists without a self-reported interest in urogynecology. Therefore, we evaluated the logs of only those urologists who considered themselves generalists by self-report. Notably, these CPT codes only captured surgical procedures performed by the physicians. Cases performed in the operating room make-up a small percentage of the total clinical time spent by urologists, regardless of sub-specialty. The average time spent in office with consultation, testing and non-surgical treatment performed to book one surgical case is different for different conditions and sub-specialties. For instance, the time spent in the office to book a surgical case for a nephrolithiasis or bladder
Page 9 of 19
cancer patient may be much lower when compared to a fertility, BPH or pelvic pain patient. The amount of time spent in the office to book a surgical case in a purely urogynecologic practice probably is on the higher side as much of the testing and treatment utilizes non-surgical options. For this reason, a slight difference in percentages of cases performed may underestimate a much larger difference in number of office patients seen with said conditions. This also may make a difference for compensation. A survey of over 800 urologists showed females earn $76,321 less than males using adjusted mean differences. One contributing factor may be that female urologists in general practice are seeing a larger percentage of patients who are medically managed.
The findings in this study apply more broadly to other fields. As the practice of medicine evolves, and females make up a steadily increasing percentage of the physician workforce, defining the differences in the practices of female physicians and the reasons behind those differences is of great importance. Further studies defining perceptions held and actual gender differences in practice will help program directors and other mentors guide trainees to make informed career choices. This can help guide females in the field to the appropriate training and efforts necessary to build the practices that they desire.
CONCLUSION: Traditional urogynecology cases represented a significantly greater percentage of the total cases logged by certifying non-fellowship trained female general urologists when
Page 10 of 19
compared to their non-fellowship trained, generalist male colleagues. The percentage of total urogynecologic cases performed by both is very small and may not be significant enough to sway career decisions. However, it supports a belief that patient populations differ for male and female general urologists which may inform female residents planning future practice decisions.
REFERENCES: 1. 2. 3.
5. 6. 7. 8. 9. 10.
Pruthi, R.S., et al., Recent trends in the urology workforce in the United States. Urology, 2013. 82(5): p. 987-93. Grimsby, G.M. and C.E. Wolter, The journey of women in urology: the perspective of a female urology resident. Urology, 2013. 81(1): p. 3-6. 2017 Urology Residency Match - Statistics. 2017; Available from: http://www.auanet.org/common/pdf/education/specialty-match/Urology-ResidencyMatch-Statistics-2017.pdf. Tsugawa, Y., et al., Comparison of Hospital Mortality and Readmission Rates for Medicare Patients Treated by Male vs Female Physicians. JAMA Intern Med, 2017. 177(2): p. 206-213. Poon, S.A., et al., Trends in partial and radical nephrectomy: an analysis of case logs from certifying urologists. J Urol, 2013. 190(2): p. 464-9. Silberstein, J.L., et al., Urinary diversion practice patterns among certifying American urologists. J Urol, 2013. 189(3): p. 1042-7. Oberlin, D.T., et al., National practice patterns of treatment of erectile dysfunction with penile prosthesis implantation. J Urol, 2015. 193(6): p. 2040-4. Elterman, D.S., et al., Changes in pelvic organ prolapse surgery in the last decade among United States urologists. J Urol, 2014. 191(4): p. 1022-7. Liu, J.S., et al., Disparities in female urologic case distribution with new subspecialty certification and surgeon gender. Neurourol Urodyn, 2015. Spencer, E.S., et al., Gender Differences in Compensation, Job Satisfaction and Other Practice Patterns in Urology. J Urol, 2016. 195(2): p. 450-5.
Page 11 of 19
FIGURE LEGENDS: Figure 1: The statistically significant differences in percentage of total cases made up by urogynecology cases by gender held regardless of population density, geographic location or employment time. Statistical significance (p<0.01) is denoted by an asterisk.
Figure 2: The differences in percentage of total cases made up by urogynecology cases by gender according to certification period and type of practice. Statistical significance (p<0.01) is denoted by an asterisk.
Page 12 of 19
Gender Female Female Female Male Male Male Male
Gender Female Female Female Female Female Female Male Male Male Male Male Male
Gender Female Female Female Female Female Male Male Male Male Male
Lower CI 1.27% 0.59%
First Certification First Recertification Second Recertification First Certification First Recertification Second Recertification Third Recertification
Population Area Less than 100,000 100,000 - 250,000 250,001 - 500,000 500,001 - 1,000,000 Over 1,000,000 Less than 100,000 100,000 - 250,000 250,001 - 500,000 500,001 - 1,000,000 Over 1,000,000
Lower CI 1.11% 1.57% 1.31% 0.65% 0.63% 0.54% 0.36%
Mid-Atlantic New England North Central South Central Southeastern Western Mid-Atlantic New England North Central South Central Southeastern Western
1.08% 1.51% 1.19% 0.64% 0.62% 0.53% 0.34%
Lower CI 1.21% 0.88% 1.11% 2.10% 1.17% 1.75% 0.49% 0.41% 0.70% 0.75% 0.51% 0.74%
1.08% 0.80% 1.02% 1.99% 1.08% 1.66% 0.47% 0.40% 0.68% 0.74% 0.50% 0.72%
Lower CI 1.29% 0.86% 1.02% 1.39% 1.55% 0.70% 0.63% 0.58% 0.53% 0.54%
0.84% 0.56% 0.70% 1.39% 1.16% 0.63% 0.57% 0.51% 0.47% 0.49%
1.15% 1.63% 1.44% 0.66% 0.64% 0.54% 0.38%
198 80 17 1131 1252 1206 115
1.35% 0.96% 1.22% 2.21% 1.26% 1.85% 0.50% 0.42% 0.71% 0.77% 0.52% 0.76%
Upper CI 1.96% 1.32% 1.48% 1.79% 2.05% 0.77% 0.70% 0.65% 0.60% 0.60%
13 26 23 31 22 38 335 453 560 436 840 489
N 42 38 46 46 84 667 669 533 459 908
Page 13 of 19
Gender Female Female Female Female Female Female Female Male Male Male Male Male Male Male
Gender Female Female Male Male
Academic Private Practice Salaried Hosp Employee Military State/local government Veterans Other Academic Private Practice Salaried Hosp Employee Military State/local government Veterans Other
Employment Time Full-time Part-time Full-time Part-time
Lower CI 1.00% 1.24% 1.54% 1.47% 0.97% 0.81% 0.53% 0.65% 0.58% 0.73% 0.70% 0.80% 0.85% 1.04%
0.85% 1.20% 1.42% 1.25% 0.31% 0.58% 0.22% 0.61% 0.57% 0.71% 0.62% 0.36% 0.61% 0.89%
Lower CI 1.24% 1.52% 0.59% 0.70%
1.21% 1.36% 0.58% 0.66%
1.17% 1.27% 1.67% 1.74% 2.98% 1.14% 1.27% 0.69% 0.58% 0.75% 0.79% 1.76% 1.18% 1.21%
Upper CI 1.28% 1.70% 0.59% 0.73%
20 196 31 16 1 4 1 98 3062 308 61 2 8 8
N 251 18 3433 118
Page 14 of 19
Editorial Comment Damara Kaplan, Ph.D., M.D. Albuquerque Urology Associates 610 Broadway Blvd. NE Albuquerque, NM 87102 [email protected]
As the number of women entering urology increases, it is apparent that the differences in male and female urologists are more complex than just gender of the physician. Rotker et al. demonstrated that traditional urogynecology cases represented a significantly greater percentage of the total cased logged by non-fellowship trained female general urologists compared to male general urologists. These differences held true regardless of the region of the country, the type of employment and the status of employment, full vs. part time. Their findings are of great significance for a number of reasons.
First, the manuscript presents “real world” data about the practice patterns for non-fellowship trained female urologists. It is true that women prefer female urologists and this gender preference radically alters a physician’s patient panel. 1 In an attempt to control or limit the number of female patients, many women will pursue fellowship training outside of female pelvic medicine and reconstructive surgery. This specialized training was thought to be protective from an imbalance of female patients, but a study by Bowen et al. demonstrated this to be untrue. 2 It is also possible that more women opt for medical management of their urologic problems rather
Page 15 of 19
than surgical intervention. This will result in a decreased number of operative cases and a perception of decreased surgical productivity.
Second, their findings may explain some of the causes of gender inequality in the urologic work place. AUA census data from 2016 shows that female urologists work 4 hours more per week than male urologists3, and female urologists, on average, make $76,321 less per year than their male counterparts4. The reasons for this pay discrepancy have not been fully elucidated. The Relative Value Units (RVUs) assigned to common urogynecology cases are substantially lower than the RVUs for the most common “male” urologic procedures. Therefore, if women self select female urologists and the reimbursement for “female” urologic procedures is less, this may begin to explain some of the reasons for the substantial pay inequity.
This is the type of well-developed research that will help generations of female urologists make informed decisions about the unique world of urologic practice.
Damara Kaplan, Ph.D., M.D. Albuquerque Urology Associates, Albuquerque, NM
References 1. Kim SO, Kang TW, Kwon D. Gender Preferences for Urologists: Women
Female Urologists. Urol J. 2017; 14: 3018-3022. 2. Bowen DK, Chi AC, Bachrach L, et al. Practice Expectations Compared to Reality for Women in Urology: A National Survey. Urology Practice. 2014; 2:281-286. 3. 2016 AUA Census Data
Page 16 of 19
4. Spencer ES, et al. Gender Differences in Compensation, Job Satisfaction and Other Practice Patterns, J. Urol. 2016; 195:450-455.
Page 17 of 19
Figure 1 - gold journal revised.png
Page 18 of 19
Figure 2 - gold journal .png
Page 19 of 19