A 15-Year Profile of U.S. Radiation Oncology Residency Growth by Geographic Region, Metropolitan Size, and Program Size

Applicant and Program Director Perceptions of Second Look Events During the 2023 Radiation Oncology Residency Match

Purpose

Virtual radiation oncology (RO) residency interviews may impair applicant and program evaluation. Second look events (SLEs) exist; however, the frequency, nature, and implications are unknown. We surveyed applicants and program directors (PDs) to characterize the 2023 RO Match SLEs and assess perspectives.

Method and Materials

An online, anonymous survey was distributed to 2023 RO Match applicants and American College of Graduate Medical Education-accredited RO PDs post-Match. Number and percentage are reported as response per question. Likert-type scores (1, strongly agree; 5, strongly disagree) are reported as median, IQR.

Results

Responses were received from 51 of 246 applicants (21%) and 52 of 88 PDs (59%). Forty applicants (87%) were offered in-person and virtual SLEs; 20 (51%) and 17 (44%) applicants were invited to 1 to 3 and 4 to 6 events, respectively. Most invited applicants attended none (21, 54%). Applicants reported that all (21, 54%) or some (16, 41%) programs communicated intentions to finalize rank order lists (ROLs) before SLEs. Most applicants (29, 74%) agreed that SLEs were optional without ROL consequences (median, 2, IQR 1-3). Applicants declined in-person SLEs due to city/facility indifference (10, 43%), finances (10, 43%), and logistics (9, 39%). Most (12, 86%) in-person SLE attendees agreed that SLEs influenced their ROL (median, 2, IQR 1-2). Nineteen PDs (40%) reported offering SLEs, with 18 of 19 being in-person. PDs who did not offer SLEs cited ethical concerns (13, 45%) and institutional policies (11, 38%). All PDs reported that SLEs were optional, and 18 of 19 explained that the SLE would be without ROL consequences. SLEs mostly occurred in February before (11, 58%) and after (15, 79%) ROL submission.

Conclusions

In-person SLEs occurred during Match 2023. All PDs considered SLEs optional which was trusted by most applicants. Attendance at in-person SLEs influenced applicants' ROLs; however, finances and logistics impaired applicant attendance. Further work is needed to appreciate SLE implications and ensure equitable residency recruitment.

Radiation and Cancer Biology Educators of Radiation Oncology Residents and the Courses They Teach1

The purpose of this study was to characterize today's radiation and cancer biology educators of radiation oncology residents, and the biology courses they teach. An e-mail list of 133 presumptive resident biology educators was compiled, and they were invited to participate in a 46-item survey. Survey questions were designed to collect information about the educational and academic backgrounds of the educators, how they self-identify, characteristics of the courses they teach, the value that they assign to their teaching activities, their level of satisfaction with their courses and how they see these courses being taught in the future. Findings of this survey were compared and contrasted with prior surveys of biology educators (conducted 12 and 20 years ago, respectively), and with more recent surveys of radiation oncology residents and radiation oncology residency program directors conducted in 2018 and 2019. A total of 67 survey responses were received. Biology educators range in age, academic rank and years of teaching experience from junior (18%) to quite senior (45%). Only about 40% self-identify as radiation biologists, biophysicists or chemists, compared to 56% in 2001. The majority of the others consist of cancer biologists (15%), radiation oncologists (15%) and radiation oncology physician-scientists (16%). Educators prioritize their resident teaching as important or very important. Biology courses are widely variable in contact hours between programs and have not changed significantly over the past 20 years. About 75% of the courses are team-taught, including 15% involving multiple training programs. An average biology course consists of about 42% foundational ("classical") radiobiology, 28% clinical radiobiology and 28% cancer biology. While biology educators and radiation oncology program directors are highly satisfied with their biology courses, approximately a third of residents report being not very, or not at all, satisfied. That fewer biology educators are radiobiologists by training and their courses have remained quite variable in length and content over long periods point to the need for a consensus core curriculum for resident education in radiation and cancer biology. Both current educators and program directors also support making online teaching resources available, diversifying course instructors and consolidating biology teaching across multiple training programs.

United States Radiation Oncology Fellowship Growth From 2010 to 2020

PURPOSE

To provide a comprehensive analysis of radiation oncology (RO) fellowship growth from 2010 to 2020.

METHODS AND MATERIALS

A collated database of RO fellowship programs and matriculants was created using (1) RO residency program (n = 92) and graduate (n = 2082) web searches, (2) prospective American Society for Radiation Oncology (ASTRO) Career Center postings database, (3) Association of Residents in Radiation Oncology Fellowship Directory, (4) RO fellowship survey data, (5) ASTRO Membership Directory, and (6) direct e-mail contact with fellowship program directors. Linear regression was used to assess the statistical significance of RO fellowship program, position, and matriculant growth over time.

RESULTS

From 2010 to 2020, the number of RO fellowship programs and annual positions significantly increased from 20 to 37 (1.60 increase per year; 95% confidence interval, 1.32-1.89; P < .001) and 20 to 39 (1.81 increase per year; 95% confidence interval, 1.52-2.10; P < .001), respectively. The most commonly offered fellowship disciplines were proton therapy (n = 10), brachytherapy (n = 7), stereotactic radiosurgery/stereotactic body radiation therapy (n = 6), general RO (n = 5), and customizable to trainees' interests (n = 3). Only 10 (27%) fellowships had a formal curriculum. All fellowships were unaccredited. Four (10.8%) programs were offered at institutions without an Accreditation Council for Graduate Medical Education-accredited RO residency training program, all established within the past 2 years. In addition, 54.8% (171 of 312) of available fellowship positions were filled between 2010 to 2020. Of these, 94 (55.0%) were graduates of US RO residency programs. The mean number of total fellows and US-residency trained fellows per year was 15 (range, 5-23) and 8 (range, 2-20), respectively. There was no significant increase in the number of annual matriculated fellows over time (P = .077). Among US-residency trained fellows, 27 (28.7%), 37 (39.4%), and 29 (30.9%) were from small (≤6), medium (7-12), and large (>12) residency programs, respectively. Twenty-eight (29.8%), 13 (13.8%), 25 (26.6%), and 27 (28.7%) trained in the Northeast, Midwest, South, and West, respectively.

CONCLUSIONS

There has been significant growth in unaccredited RO fellowship programs and annual positions during the past decade, although the number of matriculants has remained stable. We report for the first time the recent establishment of fellowships at institutions without an Accreditation Council for Graduate Medical Education-accredited RO residency program. The impact of fellowship programs on the training of RO residents should be studied.

Quantitative Analysis of Practice Size Consolidation in Radiation Oncology: A Trend Toward Bigger and Fewer Practices

PURPOSE

There is evidence of practice consolidation in US health care in recent years. To our knowledge, a detailed quantitative study of recent changes in radiation oncology practice size has not been performed. We aim to evaluate radiation oncology practice size changes between 2012 and 2020 in the United States.

MATERIALS AND METHODS

Using the Medicare Physician Compare Database, we identified practices employing radiation oncologists using their taxpayer identification number and individual radiation oncologists using their national provider identifier. We grouped individual radiation oncologists into categories by practice size (which includes the number of physicians of all specialties) and compared the number of radiation oncologists in each category between 2012 and 2020. Further analyses by US geographic census region, single-specialty practice, academic practice, and high- and low-population density areas were performed.

RESULTS

Between 2012 and 2020, the total number of practicing radiation oncologists increased by 9%, and the number of practices employing radiation oncologists decreased by 11.5%. The number of radiation oncologists in practices of size 1 to 2, 3 to 9, 10 to 24, and 25 to 49 decreased by 3.7%, 4.7%, 4.9%, and 2%, respectively, and the number of radiation oncologists in practices of size 50 to 99, 100 to 499, and 500+ increased by 1.4%, 2.1%, and 11.8%, respectively (all 500+ practices are multispecialty groups). The increase in practice size was significant in all regions, for single-specialty and multispecialty practices, academic and nonacademic practices, and for practices in high-, middle-, and low-population density areas (P < .05 for all comparisons). The proportion of single-specialty practices has decreased significantly (P < .001), and the proportion of academic practices increased significantly (P = .004). Additionally, the proportion of practices and physicians in high- and low-population density regions remained stable during this period (P > .05).

CONCLUSIONS

Our analysis suggests that practice size consolidation has occurred within the US radiation oncology workforce from 2012 to 2020. The impact of this consolidation on quality, cost, and patient access deserves further attention.

The Geography of Employment Outcomes for Radiation Oncology Graduates in 2019

PURPOSE

The geographic distribution of practicing radiation oncologists is of concern for multiple stakeholders within the field. Employment outcomes of graduating residents can affect that distribution, and they are of major concern to current residents. Data investigating employment outcomes of recent graduates are sparse. We aimed to analyze the employment outcomes of the radiation oncology residency class of 2019.

METHODS AND MATERIALS

Using publicly available information, we identified the employment of 179 of 183 graduating residents in the class of 2019. For each, the place of employment, residency program, and medical school were geocoded using Google Maps. We used the rural-urban continuum code (RUCC) published by the United States Department of Agriculture to determine the rurality of each location and compared employment outcomes by RUCC and program size.

RESULTS

Two thirds of graduates (66%) took a position in a county within a metropolitan area with a population greater than 1,000,000 people; only 3.4% took a position in a county outside of a metropolitan area. Graduates of smaller programs (≤6 residents) and those in smaller metropolitan areas were more likely to take positions in smaller metropolitan areas or nonmetropolitan areas. The geographic distance between location of employment and residency program did not significantly vary by program size or size of metropolitan area where a residency program was located.

CONCLUSIONS

Among the class of 2019, a small proportion took positions in nonmetropolitan areas. Smaller programs and those in smaller metropolitan areas may be more likely to produce graduates that practice in similar settings, but those graduates might not do so locally. We advocate for a centralized, prospective data collection of employment outcomes for graduating residents to refine these analyses and to reduce employment prospect information asymmetry for trainees.

Residency Match Trends, Racial Disparity, and Matching Amid a Pandemic

PURPOSE

Radiation oncology has been facing an evolving crisis in recruitment for several years, and the events of 2020 to 2021 will certainly add to that crisis with the urgency of addressing systemic racial injustice amid a global pandemic. The purpose of this study is to examine applicant data to gain insight on residency match trends and evaluate these findings within the backdrop of a novel match year.

METHODS AND MATERIALS

National Residency Matching Program (NRMP) data between 2009 and 2020 were assessed for the number of applicants, programs, and positions available, number of ranked applicants needed to fill positions, and successfully matched applicant data. Additionally, Electronic Residency Application Service data were evaluated for race/ethnicity identification among applicants.

RESULTS

The number of applicants who ranked radiation oncology as their preferred specialty has declined for 3 consecutive years from 223 in 2017 to 155 in 2020. In 2020 the applicant-to-position ratio was at an all-time low at 0.82, and the unmatched position rate increased to 18.5%. The percentage of Black or African American applicants applying to radiation oncology has also declined to 4.9%, and this population represents 7.2% of all applicants. The number of ranked applicants needed to fill the available radiation oncology PGY2 positions increased from 4.0 in 2010 to 6.0 in 2020.

CONCLUSION

Declining interest in radiation oncology among applicants, and an even further decline of black applicants, along with the challenges of interview and travel restrictions during the pandemic provide heightened concern for this year's match. Innovative efforts to expand the reach of radiation oncology to prospective applicants is needed to engage diverse, bright, and committed students for the continued progress of radiation oncology and most importantly, our patients.

Radiation Oncology Resident Quality by National Resident Matching Program Metrics From 2007 to 2018

PURPOSE

To quantify how the quality of US medical students accepted to radiation oncology (RO) training programs, as defined by National Resident Matching Program (NRMP) metrics, has changed over time.

METHODS AND MATERIALS

We examined NRMP data of senior US medical students matched into RO training programs from 2007 to 2018. Metrics include United States Medical Licensing Exam (USMLE) Step 1 and 2-Clinical Knowledge scores, research output, percentage with PhD, and percentage in Alpha Omega Alpha (AOA), among others. Linear regression analysis assessed the statistical significance of changes in available metrics of matched RO residents over time. The Student t test and χ² test compared quality metrics between matched students in RO versus all other specialties.

RESULTS

From 2007 to 2018, the mean USMLE Step 1 and 2-Clinical Knowledge for RO residents significantly increased from 235 to 247 (1.0 point/year; 95% confidence interval [CI], 0.70-1.52; P = .002) and from 237 to 253 (1.3 points/year; 95% CI, 1.27-1.62; P <.001), respectively. The mean number of research experiences and abstracts/presentations/publications increased from 3.7 to 6.1 (0.2/year; 95% CI, 0.12-0.29; P = .003) and from 6.3 to 15.6 (0.78/year; 95% CI, 0.60-1.04; P <.001), respectively. The percentage of RO residents inducted into AOA increased from 24.2% to 35.2%, whereas those with a PhD remained stable (∼21%). Matched RO residents had statistically superior metrics versus all other specialties for USMLE Step 1 scores (mean +13.5 points; 95% CI, 7.26-19.67; P <.001), research experience (mean +2.04; 95% CI, 1.11-2.97; P <.001), abstracts/presentations/publications (mean +6.80; 95% CI, 3.38-10.22; P = .001), percentage with a PhD (22.2% vs 4.1%; P <.001), and percentage in AOA (29.5% vs 15.8%; P <.001).

CONCLUSIONS

RO resident quality, defined by routinely reported NRMP metrics, increased from 2007 to 2018. Furthermore, RO resident quality is significantly higher than in all other specialties combined for most metrics. Whether the recent decline in medical student interest in RO will correlate with reduced NRMP quality metrics is unknown.