Surveying Fourth-Year Medical Students Regarding the Choice of Diagnostic Radiology as a Specialty

PurposeThe aim of this study was to survey fourth-year medical students, both those choosing and those not choosing diagnostic radiology as their specialty, regarding factors influencing their choice of specialty and their perceptions of radiology.MethodsA voluntary anonymous online survey hyperlink was sent to 141 US medical schools for distribution to fourth-year students. Topics included demographics, radiology education, specialty choice and influencing factors, and opinions of radiology.ResultsA representative sampling (7%) of 2015 fourth-year medical students (n = 1,219; 51% men, 49% women) participated: 7% were applying in radiology and 93% were not. For respondents applying in radiology, the most important factor was intellectual challenge. For respondents applying in nonradiology specialties, degree of patient contact was the most important factor in the decision not to choose radiology; job market was not listed as a top-three factor. Women were less likely than men to apply in radiology (P < .001), with radiology selected by 11.8% of men (56 of 476) and only 2.8% of women (13 of 459). Respondents self-identifying as Asian had a significantly higher (P = .015) likelihood of selecting radiology (19 of 156 [12.2%]) than all other races combined (44 of 723 [6.1%]). Respondents at medical schools with required dedicated medical imaging rotations were more likely to choose radiology as a specialty, but most schools still do not require the clerkship (82%).ConclusionsThe reasons fourth-year medical students choose, or do not choose, diagnostic radiology as a specialty are multifactorial, but noncontrollable factors, such as the job market, proved less compelling than controllable factors, such as taking a radiology rotation.     

The Stability of Factors Influencing the Choice of Medical Specialty Among Medical Students and Postgraduate Radiology Trainees

ObjectiveTo investigate whether general psychological motivating factors that guide career selection of a medical specialty differ over the course of medical school and to compare differences in motivating factors among students choosing “controllable” lifestyle specialties, students choosing “uncontrollable” lifestyle specialties, and a cohort of radiology residents.Materials and MethodsAn anonymous survey was distributed to first- through fourth-year medical students and radiology residents at a single institution. Participants were asked to select their top three of seven factors that most influenced their choice of medical specialty. Fourth-year students were asked to designate the specialty to which they had applied.ResultsThe survey was distributed to 259 students and 47 radiology residents with a response rate of 93.8% (243 of 259) and 95.7% (45 of 47), respectively. The top three factors indicated by medical students were finding the daily work fulfilling, work–life balance, and interest in the subject. These top three factors were common to all medical student classes and did not differ between students choosing “controllable” versus “uncontrollable” fields. The factors uncommonly selected were similar personality to others in the field, attending income, competitiveness or prestige, and job market conditions. For radiology residents, the top three motivating factors were the same as for medical students.ConclusionThree out of seven motivating factors were universally important to trainees, regardless of their stage of medical training or their selection of a controllable versus uncontrollable lifestyle specialty. These data suggest the variety of career choices made by students may not derive from differing underlying values.     

The Short and the Long of It: Transitioning to a Blended Longitudinal Curriculum in Radiology

PurposeThe aim of this study was to demonstrate that the transition from a stand-alone radiology clerkship block to a longitudinally integrated radiology curriculum leverages newer teaching tools favored by today’s learners.MethodsIn 2013 and 2014, medical students attended a dedicated 1-week radiology clerkship course. In 2015, the block clerkship model for radiology transitioned to a vertically integrated curriculum. By 2019, radiology content was integrated into many of the health illness and disease course blocks. Pre- and postcourse multiple-choice question tests as well as anonymous surveys were administered for both clerkship and integrated curriculum blocks. The student survey questions assessed perceptions regarding interpretation skills, imaging modality knowledge, and radiologists’ roles.ResultsAmong 197 total students in the clerkship block, surveys were completed by 170 respondents, yielding a response rate of 86.3%. Among 106 students in the longitudinal course, surveys were completed by 71 respondents, yielding a response rate of 67%. For both clerkship and longitudinally integrated courses, the average number of correct responses after completion of the courses was significantly greater than the average number of correct precourse responses. Compared with students in the clerkship block curriculum, students in the longitudinal curriculum demonstrated a significantly greater frequency of agreement in response to survey questions regarding significant exposure to radiology, feeling comfortable interpreting CT images, and being familiar with how to use the ACR Appropriateness Criteria.ConclusionsTransitioning from a single clerkship block to a more integrated format allows a more effective patient-centered clinical approach to medical imaging.     

Factors leading to radiology career selection: results of the 2004 National Physician Survey.

Data from the 2004 National Physician Survey were used to determine when current Canadian radiologists made their decision to pursue a radiology career and to determine which factors were most influential in their decision. Most respondents reported having made this decision during their clerkship years of medical school (32.7%) or after a period of time in practice (25.4%). The most influential factors involved in this decision were perceived intellectual stimulation and workload flexibility or predictability. These results provide insight into the trainee's decision to pursue radiology and may be useful to those recruiting for radiology as well to medical students considering the field.     

Factors influencing selection of an interventional radiology training program

PurposeTo understand factors influencing the choice and ranking of Interventional Radiology (IR) training programs among a cohort of medical students and diagnostic radiology residents pursuing careers in IR.Materials and methodsAn IRB approved, 34 question online survey (surveymonkey.com) evaluated the impact of twenty-two different factors and demographics on IR training program selection for medical students and residents. The factors analyzed included programmatic features, location characteristics, academic reputation, program size, benefits/financial incentives, emphasis on clinical care, and future job opportunities. Comparison of Likert scale responses between medical students and residents were performed by using unpaired two-sample t-tests.Results181 (145 male, 35 female) individuals responded to the survey, 74 medical students (40.9%) and 107 residents (59.1%). Medical students and residents both selected variety of IR cases as the most important and highest rated factor when choosing an IR program. Medical students ranked availability of a mentor (p = .03), inpatient consultation service (p = .003), outpatient clinic experience (p = .003), and ICU rotation experience (p < .001) significantly higher. Residents rated job placement/accomplishments of prior fellows (p = .03) and opinion of spouse/significant others (p = .002) significantly higher than medical students.ConclusionsThe top rated factors are similar among medical students and residents however medical students value the clinical aspects of the program (ICU experience, inpatient consultation service, outpatient clinic) more than residents. Residents placed more value on job placement opportunities in selecting an IR program.     

Medical Student Performance After a Vertically Integrated Radiology Clerkship

PurposeProper selection of imaging examinations and basic image interpretation skills are essential for all physicians, yet only approximately 25% of US medical schools require clerkships in radiology. Although there is limited time in most medical school curricula to allow the addition of a required radiology clerkship, the authors developed one that is vertically integrated over a two-year period. This clerkship includes one week of contact with radiologists distributed over the M2 and M3 years, podcasts, online modules, required readings, and presentations. A standard national examination is administered at the end of the clerkship period. This clerkship was designed to address the educational needs of students while occupying minimal time in the curriculum. The purpose of this study was to determine if students completing this clerkship perform as well on a national radiology examination as students from other medical schools, regardless of their curricula.MethodsAt the end of the M3 year, these students take a computer-based radiology examination developed by the Alliance of Medical Student Educators in Radiology and used by students at multiple medical schools nationally. The mean and median scores of these students were compared with those of students at these other institutions.ResultsThe mean and median scores of the students were 74% and 74% (standard deviation, 7.5%) compared with 74% and 50% (standard deviation, 8.4%) at other institutions.ConclusionsStudents completing this vertically integrated radiology clerkship had test scores comparable with those of students at other medical schools.     

Effectiveness of resident-prepared conferences in teaching imaging utilization guidelines to radiology residents

RATIONALE AND OBJECTIVES: The authors evaluated the effectiveness of a resident-prepared conference series for teaching imaging utilization guidelines to radiology residents. MATERIALS AND METHODS: Brown University radiology residents (n = 17) gave 61 presentations on imaging utilization to their colleagues during 16 1-hour conferences. The residents were later examined on the topics presented and surveyed about their familiarity with the American College of Radiology appropriateness criteria, their exposure to issues of cost-effectiveness, and their degree of confidence in providing imaging consultation. The same examination and survey were administered to control residents from the University of Wisconsin (n = 14) and the Oregon Health Sciences University (n = 14). Scores were compared by using linear regression and Wilcoxon rank sum tests. RESULTS: Controlling for years in radiology residency, residents at Brown scored on average 16.0% (standard error = 2.2%) higher than residents at the other universities (P < .001). Controlling for institution, 3rd- and 4th-year residents scored on average 7.4% (standard error = 2.1%) higher than 1st- and 2nd-year residents (P = .001). Brown residents expressed more familiarity with American College of Radiology appropriateness criteria and appeared to have more exposure to cost-effectiveness issues in conferences than residents at Wisconsin or Oregon Health Sciences University (P < .005). Residents from the three universities did not differ in their level of confidence in providing imaging consultation. CONCLUSION: Resident-prepared conferences are an effective means of teaching imaging utilization guidelines to residents, but they do not affect the residents' perception of their ability to provide imaging consultation.     

Assessing Medical Students’ Knowledge of IR at Two American Medical Schools

PurposeTo determine if there was a difference in the level of knowledge about interventional radiology (IR) between medical students in preclinical years of training compared with medical students in clinical years of training at two medical schools and to compare awareness of IR based on the curriculum at each school: one with required radiology education and one without such a requirement.Materials and MethodsAn anonymous survey was distributed to students at two medical schools; the survey assessed knowledge of IR, knowledge of training pathways, and preferred methods to increase exposure. Responses of the preclinical and clinical groups were compared, and responses from the clinical groups at each school were compared.Results“Poor” or “fair” knowledge of IR was reported by 84% (n = 217 of 259) of preclinical students compared with 62% of clinical students (n = 110 of 177; P < .001). IR was being considered as a career by 11% of all students (15%, n = 40 of 259 preclinical; 5%, n = 9 of 177 clinical). The main reason respondents were not considering IR was “lack of knowledge” (65%, n = 136 of 210 preclinical; 20%, n = 32 of 162 clinical). Students in the clinical group at the institution with a required radiology rotation reported significantly better knowledge of IR than clinical students from the institution without a required clerkship (P = .017).ConclusionsThere are significant differences in knowledge of IR between preclinical and clinical students. Required radiology education in the clinical years does increase awareness of IR.     

Analysis of Gender Disparity in US and Canadian Radiology Residency Programs

BackgroundEquity, diversity, and inclusion in academic radiology are a work in progress and although the gender gap has decreased, there remains a paucity of studies examining female representation among radiology trainees over the past decade.ObjectiveThe aim of our undertaking was to evaluate gender parity in United States (US) and Canadian radiology residency programs and to suggest future directions to improve female representationMaterials and MethodsRetrospective analysis of publicly available data on radiology residents from the US and Canada was performed from 2007to 2019. Data on diagnostic radiology residents was collected from the Accreditation Council for Graduate Medical Education for the US and the Canadian Post M.D. Education Registry for Canada. Statistical tests including regression and ANOVA were used to study the gender proportions from 2007to 2019.ResultsThere has been little progress in bridging the gender gap in the last 12 years. The proportion of female residents pursuing radiology has remained at an average of 26.74% (n = 1,238of 4,629) in US programs and 31.78% (n = 28 of88) in Canadian programs. The average change in the percentage of female residents was 0.0% per year (P = 0.0) for US programs and -2.9% per year (P = 0.3) for Canadian programs.DiscussionDespite a higher proportion of females in North American medical schools, gender disparity persists among radiology residents. More research is needed to identify barriers limiting female representation and improve gender parity across North American radiology programs.     

The Radiology Resident as Teaching Consultant: An Innovative Peer-to-Peer Teaching Consultation Service to Strengthen Relationships With Referring Colleagues in the Era of Imaging 3.0

BackgroundThe American College of Radiology Imaging 3.0 paradigm emphasizes the need for radiologists to serve as imaging consultants to their referring colleagues. However, outside the reading room, teaching interactions between radiology and nonradiology residents are limited. Internal Medicine Morning Report (IMMR) is a resident-run educational program widely employed by internal medicine (IM) residencies. Although medical imaging is regularly discussed in IMMR, radiology residents are not typically involved in case preparation. We aimed to develop a peer-to-peer imaging teaching consultation service (TCS) incorporated into the well-established structure of IMMR. By creating illustrative, “dynamic” teaching slides for use at these conferences, we sought to provide salient radiology teaching material, demystify jargon, discuss appropriate imaging use, and review relevant anatomy. We hypothesized that TCS could improve the quality of IMMR as perceived by the IM presenter.MethodsTCS was piloted over a 7-month period. Each referred case was reviewed by a senior radiology resident who produced a set of “dynamic” teaching slides for each case. These included patient imaging overlayed with extensive annotations and animations highlighting teaching points, with particular attention to radiologic terminology. Slides were shared with the IM presenter, who could use them for preparation and include the animations in the talk if desired. TCS effectiveness was evaluated with a survey distributed to participating IM residents.ResultsIn the pilot period, 12 TCS requests were received and 10 were performed in collaboration with 6 IM residents. Survey results indicated that most IM residents did not consult radiologists prior to TCS (5/6, 83%). IM residents used the “dynamic” teaching slides to both prepare for and present at IMMR (5/6, 83%). TCS improved IM residents’ perceived ability to engage their audience (6/6, 100%), confidence in teaching radiology material (4/6, 67%), ability to understand radiology reports (4/6, 67%) and appreciation for what radiologists do (6/6, 100%).ConclusionsThe TCS pilot resulted in successful radiology-IM collaboration and improved knowledge and confidence in teaching imaging concepts. Continuous program evaluation will be performed and future work will assess the effect of TCS on radiologist confidence in real-world clinical consultations.     

Radiology Clerkship Requirements in Canada and the United States: Current State and Impact on Residency Application

PurposeAn unmet need for radiology education exists even in this era of medical school curricular renewal. The authors examined the radiology clerkship requirements in Canadian and US medical schools to interpret radiology residency applicant trends.MethodsThe curricula of Canadian and US medical schools were reviewed for radiology rotation requirements. The radiology residency applicant trends for 2010 to 2019 were analyzed using linear regression. The number of radiology electives taken by matched radiology applicants was examined. Regression analysis was performed to assess the impact of radiology rotation requirements on residency application.ResultsOnly 1 of 17 Canadian medical schools required a radiology rotation despite major curricular renewal at the majority of medical schools. Approximately 20% of US medical schools required radiology rotations, without a significant change from 2011 to 2018, whereas the duration of required radiology rotations increased significantly. The numbers of total and first-choice radiology applicants showed significant decreases from 2010 to 2019 in Canada but not in the United States. Nearly all matched radiology applicants took electives in radiology, the majority of whom took three or more electives. Both the presence and duration of radiology rotation requirements showed significant, positive relationships with the number of radiology applicants.ConclusionsOnly a minority of medical schools in North America have radiology clerkship requirements, both the presence and duration of which significantly affect students’ choice of radiology as a career. Radiology clerkship requirements can be a solution to meet the expanding demand for diagnostic imaging in modern medicine.     

Invasive Procedural Versus Diagnostic Imaging and Clinical Services Rendered by Radiology Trainees Over Two Decades

PurposeTo characterize evolving radiology trainee exposure to invasive procedures.MethodsUsing Physician/Supplier Procedure Summary Master Files from 1997 to 2016, we identified Medicare services performed by radiology trainees in approved programs by extracting information on services billed by diagnostic and interventional radiologists reported with “GC” modifiers. Services were categorized as (1) invasive procedures, (2) noninvasive diagnostic imaging services, or (3) clinical evaluation and management (E&M) services. Relative category trainee work effort was estimated using service-level work relative value units.ResultsNationally from 1997 to 2016, invasive procedures declined from 34.2% to 14.3% of relative work effort for all Medicare-billed radiology trainee services. Radiology trainees’ noninvasive diagnostic imaging services increased from 65.1% to 85.4%. Clinical E&M services remained uniformly low (0.7%-0.3%). Diagnostic radiology (DR) and interventional radiology (IR) faculty supervised 81.0% and 19.0%, respectively, of all trainee invasive procedures in 1997, versus 68.3% and 31.7%, respectively, in 2016. Despite declining relative procedural work, trainees were exposed to a wide range of both basic and complex invasive procedures in both 1997 and 2016. Over this period, trainee noninvasive diagnostic imaging services shifted away from radiography to CT and MRI.ConclusionRadiology trainees’ relative invasive procedural work effort has declined over time as their work increasingly focuses on CT and MRI. As DR and IR-DR residency curricula begin to diverge, it is critical that both DR and IR residents receive robust training in basic image-guided procedures to ensure broad patient access to these services.     

Design,Implementation, and Evaluation of a Diversity Program for Radiology

Rationale and ObjectivesDiagnostic radiology training programs are less diverse than graduating US medical school classes and the patient populations they serve. Inclusion of physicians who are underrepresented minorities in medicine (URMM) can strengthen the profession and help to meet the needs of an increasingly diverse population. Our Department of Radiology developed and implemented a plan to increase the number of URMMs in our residency applicant pool and residency training program.Materials and MethodsWe designed a recruitment strategy to diversify the radiology residency applicant pool. This included website development, advertising, early exposure opportunities, travel to predominantly minority institutions and national meetings, and mentoring URMM medical students. We implemented parallel activities to increase the number of URMMs in our residency program. These included holistic screening tools for residency application review, a diverse residency recruitment committee, a welcome environment for visiting candidates, and “Second Look Weekend” visits for talented candidates. Primary outcomes measured include change in percentages of URMM applicants in our applicant pool and URMM residents in our residency program.ResultsThe percentage of URMM radiology residency applicants increased from 7.5% (42 of 556) of the total applicant pool in the 2012 to 2013 recruitment year to 12.6% (98 of 777) in the 2017 to 2018 recruitment year (P = .001). URMM radiology residency representation increased from 0% (0 of 32) in the 2013 to 2014 academic year to 20% (6 of 30) in the 2018 to 2019 academic year (P = .01).ConclusionAn intentional, strategic diversity program can diversify an institution’s residency applicant pool and increase representation of URMMs in a diagnostic radiology residency program.     

Developing the Evidence Base for M-Learning in Undergraduate Radiology Education: Identifying Learner Preferences for Mobile Apps

PurposeThere is a lack of evidence for developing radiology mobile apps for medical students. This study identifies the characteristics which students perceive as most valuable to teaching radiology with mobile apps (m-learning).MethodsAn online anonymous survey was administered to second- to fourth-year medical students at a single institution. The survey, which was based on established theoretical framework, collected students' preferred content organization, content presentation, and delivery strategies. The Copeland method was used to rank student preferences and a 2-tailed t test was used to determine if student responses were related to their clinical experience, with statistical significance at P < .05.ResultsThe response rate was 25.6% (163/635). For content organization, image interpretation (66.9%), imaging anatomy (61.3%), and common pathological conditions (50.3%) were selected as the most important. For content presentation, quizzes (49.1%) and case presentations (46.0%) were selected as the most useful. Students with clinical experience rated algorithms as more important (P < .01) and quizzes as less important (P = .03). For delivery strategies, ease of use (92.6%), navigation (90.8%), and gestural design (74.8%) were deemed the most applicable.ConclusionThis study documents medical students' preferences for m-learning in radiology. Although learner preferences are not the only feature to consider in the development of educational technology, these provide the initial framework for radiologists wishing to develop and incorporate mobile apps into their teaching.     

Evolution of the Preliminary Clinical Year and the Case for a Categorical Diagnostic Radiology Residency

PurposeWhile other specialties traditionally utilizing a segregated clinical internship year have slowly progressed toward integrated training curricula, diagnostic radiology has been slow to adopt this path. The aim of this study was to analyze the trends in stand-alone preliminary clinical years as well as the shift toward categorical residencies currently being undertaken in other specialties. Advantages of mimicking the trends of other specialties and current integrated radiology programs are discussed. The perception of diagnostic radiology as a competitive specialty is explored, and the prospect of change as a recruiting tool is examined.MethodsData assimilated by the NRMP from 1994 through 2016 were processed and analyzed.ResultsThe total number of postgraduate year (PGY) 1 preliminary year programs has remained relatively constant over the past 10 years despite a gradual increase in overall NRMP applicants. The proportion of these programs offered as a transitional year declined from 31% in 1994 to 20% in 2016. The proportion of categorical anesthesiology positions gradually rose from 43% in 2007 to 70% in 2016. The fraction of categorical neurology positions increased from 30% in 2007 to 59% in 2016. The percentage of diagnostic radiology programs beginning at the PGY 1 level has been relatively constant at 12% to 14% since 2007. Dermatology has increased advanced (PGY 2) positions while decreasing categorical (PGY 1) positions. Those matching in diagnostic radiology have performed at a high level compared with the composite NRMP average since 2007. In the 2015 match, there were 65 diagnostic radiology programs that did not fill all of their offered positions. Of the institutions housing these programs, only 22% of them had preliminary internal medicine or transitional year positions available after the match.ConclusionsIn response to the evolving nature of health care and graduate medical education, other specialties are gradually shifting toward curricular structures that begin at the PGY 1 level. By considering such a transition, diagnostic radiology would be well served to position itself as a valuable clinical specialty while maintaining a lesser dependence on other specialties to train its physicians.     

Introducing First-Year Radiology Residents to the ACR at the AMCLC From 2009 to 2013: Summary of Experiences and Five-Year First-Cohort Follow-Up

PurposeThe aim of this report is to provide a five-year summary of the Minnesota Radiological Society’s initiative to send first-year radiology residents to the ACR at the AMCLC. The authors provide an update of the survey data for the first five years (2009-2013) and a report of the ACR membership status of the original 2009 cohort (class of 2012) five years after their conference experience.MethodsParticipating residents from 2009 to 2013 completed pre- and postconference surveys assessing their knowledge of ACR-related topics, conference satisfaction, and intention to join the ACR. ACR membership status of the first cohort was determined using the ACR membership database and compared with both the previous five graduating classes and the national average for practicing radiologists.ResultsSeventy first-year Minnesota radiology residents attended the conference from 2009 to 2013. Knowledge of the ACR significantly increased after the conference. Most residents were highly satisfied or satisfied with their conference experience and highly likely or likely to join the ACR in the future. Two years after residency, 87% of the first cohort (13 of 15) were ACR members, compared with an average membership rate of 57% (63 of 110) for the previous five graduating classes.ConclusionsExposing radiology residents early to the ACR at the AMCLC leads to a significant increase in knowledge pertaining to the professional organization. This exposure likely leads to increased ACR membership when residents enter practice. This early engagement in radiology affairs can lead to a higher rate of ACR membership and to a better informed membership.     

Scientific Publications on Thyroid Ultrasound between 2001 and 2020: Differences in Research Characteristics by Disciplines

ObjectiveTo analyze the characteristics and trends of scientific publications on thyroid ultrasound (US) from 2001 to 2020, specifically examining the differences among disciplines.Materials and MethodsThe MEDLINE database was searched for scientific articles on thyroid US published between 2001 and 2020 using the PubMed online service. The evaluated parameters included year of publication, type of document, topic, funding, first author''s specialty, journal name, subject category, impact factor, and quartile ranking of the publishing journal, country, and language. Relationships between the first author''s specialty (radiology, internal medicine, surgery, otorhinolaryngology, and miscellaneous) and other parameters were analyzed.ResultsA total of 2917 thyroid US publications were published between 2001 and 2020, which followed an exponential growth pattern, with an annual growth rate of 11.6%. Radiology produced the most publications (n = 1290, 44.2%), followed by internal medicine (n = 716, 24.5%), surgery (n = 409, 14.0%), and otorhinolaryngology (n = 171, 5.9%). Otorhinolaryngology and internal medicine published significantly more case reports than radiology (p < 0.001, each). Radiology published a significantly higher proportion of publications on imaging diagnosis (p < 0.001 for all) and a significantly lower proportion of publications on biopsy (p < 0.001 for all) than the other disciplines. Publications produced by radiology authors were less frequently published in Q1 journals than those from other disciplines (p < 0.005 for internal medicine and miscellaneous disciplines and < 0.01 for surgery and otorhinolaryngology). China contributed the greatest number of publications (n = 622, 21.3%), followed by South Korea (n = 478, 16.4%) and the United States (n = 468, 16.0%).ConclusionRadiology produced the most publications for thyroid US than any other discipline. Radiology authors published more notably on imaging diagnosis compared to other topics and in journals with lower impact factors compared to authors in other disciplines.     

Women in Radiology: Exploring the Gender Disparity

PurposeIn 2015, only 1.5% of female Canadian medical students pursued radiology as a specialty, versus 5.6% of men. The aim of this study was to determine what factors attract and deter Canadian medical students from pursuing a career in radiology, and why fewer women than men pursue radiology as a specialty.MethodsAn anonymous online survey was e-mailed to English-speaking Canadian medical schools, and 12 of 14 schools participated. Subgroup analyses for gender and radiology interest were performed using the Fisher exact test (P < .05).ResultsIn total, 917 students (514 women; 403 men) responded. Direct patient contact was valued by significantly more women who were not considering specialization in radiology (87%), compared with women who were (70%; P < .0001). Physics deterred more women (47%) than it did men (21%), despite similar educational backgrounds for the two gender groups in physical sciences (P < .0001). More women who were considering radiology as a specialty rated intellectual stimulation as being important to their career choice (93%), compared with women who were not (80%; P = .002). Fewer women who were not interested in radiology had done preclinical observerships in radiology (20%), compared with men who were not interested in radiology (28%; P = .04).ConclusionsA perceived lack of direct patient contact dissuades medical students from pursuing radiology as a career. Women have less preclinical radiology exposure than do men. Programs that increase preclinical exposure to radiology subspecialties that have greater patient contact should be initiated, and an effort to actively recruit women to such programs should be made.