Radiology stereotypes, application barriers, and hospital integration: A mixed-methods study of medical student perceptions of radiology

Abstract

Rationale and Objectives:

Limited exposure to radiology by medical students can perpetuate negative stereotypes and hamper recruitment efforts. The purpose of this study is to understand medical students’ perceptions of radiology and how they change based on medical education and exposure.

Materials and Methods:

A single-institution mixed-methods study included four groups of medical students with different levels of radiology exposure. All participants completed a 16-item survey regarding demographics, opinions of radiology, and perception of radiology stereotypes. Ten focus groups were administered to probe perceptions of radiology. Focus groups were coded to identify specific themes in conjunction with the survey results.

Results:

Forty-nine participants were included. Forty-two percent of participants had positive opinions of radiology. Multiple radiology stereotypes were identified, and false stereotypes were diminished with increased radiology exposure. Opinions of the impact of artificial intelligence on radiology closely aligned with positive or negative views of the field overall. Multiple barriers to applying for a radiology residency position were identified including board scores and lack of mentorship. COVID-19 did not affect perceptions of radiology. There was broad agreement that students do not enter medical school with many preconceived notions of radiology, but that subsequent exposure was generally positive. Exposure both solidified and eliminated various stereotypes. Finally, there was general agreement that radiology is integral to the health system with broad exposure on all services.

Conclusion:

Medical student perceptions of radiology are notably influenced by exposure and radiology programs should take active steps to engage in medical student education.

Introduction

Although radiology is one of the more populous fields in medicine, it does not receive proportionate representation in medical school curricula. Only 16% of medical schools require radiology as a clerkship during the clinical years (1). Some schools instead rely on radiology exposure through preclinical anatomy labs and lectures followed by indirect exposure during clinical core rotations. Other medical schools are integrated radiology sessions taught by radiologists onto other clinical core rotations, such as abdominal imaging teaching sessions during a general surgery clerkship. In general, this educational framework does not provide an accurate representation of the scope and breadth of radiology practice as it is often colored by the views of non-radiologists. Multiple studies have demonstrated that medical students’ understanding of the field of radiology is very limited, especially for pre-clinical medical students (2–5). As a result, stereotypes of radiology are rampant, especially negative stereotypes which emphasize a lack of patient contact, anti-social behavior, and focus on lifestyle or money (2, 4, 6). Prior studies have demonstrated that medical students are seeking additional information about the field of radiology, but their ability is handicapped by the lack of exposure in the current medical school education framework (7, 8). Limited work has demonstrated that direct early exposure to radiology by practicing radiologists has a net positive impact on medical student perceptions of the field and can help to combat negative stereotypes (3, 5, 9). A comprehensive understanding of how medical student perceptions of radiology change based on exposure and education is currently lacking.

Much of the work to date aimed at understanding medical student perceptions of radiology, and how they influence specialty selection, has been survey based (6, 10–13). Surveys provide an excellent means of collecting data on a large scale and allow for statistical testing of subgroups. However, the published surveys have low response rates and have produced mixed results based on differences in question wording, respondent population (e.g., first year medical students versus fourth year students applying in radiology), and the inability to allow for direct probing and follow up questioning. In contrast, qualitative methodologies are better suited to capture the nuances of complex decision-making, such as career decisions. The downsides of qualitative techniques are the cost, smaller sample sizes, and longer time for data collection and analysis. To date, qualitative approaches have been used in a very limited capacity and primarily among Canadian medical students who are learning under the umbrella of a different healthcare system and who are not subject to the same National Residency Match Program as students in the United States (7, 8). Combining quantitative and qualitative approaches into a mixed-methods study allows the limitations of each approach to be reduced, while capitalizing on the relative strengths of each approach.

The purpose of this study is to understand medical students’ perceptions of radiology and how they may vary based on medical education and experience. By utilizing a mixed-methods approach, both quantitative and qualitative information is obtained to allow for a more nuanced understanding of decision-making regarding residency selection. Although most radiology stereotypes are well known, medical students’ perceptions of both positive and negative stereotypes based on radiology exposure have not been well explored. Understanding how medical students’ perceptions change with radiology exposure could facilitate efforts to provide a more accurate representation of radiology and recruit medical students.

Materials and Methods

Study Population

Email invitations to participate in the study were sent to all current medical student, approximately 500 students, via mailing lists at a single academic medical institution with a curriculum that included a core radiology rotation from December 2020 through February 2021. This time period was chosen to coincide with the middle of the academic year. Students responding in the affirmative were sent an invitation to complete an electronic survey via RedCap (Vanderbilt University; Nashville, TN). All participants provided informed consent, and the study was approved by the Institutional Review Board. Participants were provided $20 for their participation. Four groups of medical students were identified. Group 1 was first year medical students with no direct clinical exposure to clinical radiology. Group 2 was students in their clinical year prior to their radiology core rotation who had indirect radiology exposure via other services. By choosing the middle of the academic year, we ensured that students in Group 2 had indirect exposure to radiology from other rotations. Group 3 was students in their clinical year who had completed their radiology c-ore rotation and thus had direct radiology exposure. Group 4 were senior students who had completed all clinical rotations and had exposure to all fields in the core curriculum.

Setting

The curriculum of (name withheld) provides an environment to study the influence of a more comprehensive radiology educational experience (14). In the first year, medical students are exposed to radiologists during anatomy labs and lectures. Faculty and residents give lectures and students can review CT scans of their cadavers. Medical students then begin their clinical clerkships in their second year. During the clinical second year, all students spend a dedicated 4-weeks on radiology as a core clerkship. Students spend 1–5 days on different subspecialty services, which includes time in the reading room and lectures by faculty, residents, and fellows. Students also get indirect exposure to radiology through their rotations on other clinical services. In the third year, students primarily pursue research. Finally, in the last year, students may pursue radiology electives at home or via an away rotation. There are approximately 125 medical students enrolled in each class at (name withheld).

Survey Development

Study participants completed a 16-item survey prior to the focus groups (Appendix A). Demographic questions included gender, race/ethnicity, age, and the presence of children or dependents. Participants were asked to list their top three specialty choices. Opinions of radiology were queried currently, over time, and based on the impact of COVID-19. Finally, there were seven questions about radiology stereotypes. These questions were based on published radiology stereotypes shown to have both positive and negative influences on medical students (6, 7, 10, 11, 13, 15). All opinion-based questions were asked on five-point Likert scales with a neutral middle option as shown in Appendix A.

Focus Group

We conducted 10 focus groups with 4–8 participants each, all led by an experienced qualitative researcher who was not a radiologist (LJF). Each focus group included only members from the same pre-assigned Group, as defined above. The interview guide included the following topics: perspectives of radiology and radiologists, stereotypes, exposure to radiology, COVID-19, barriers to applying for a radiology residency, future of radiology, and diversity. The full text of the interview guide is shown in Appendix B and was developed based on prior published work, expert opinion, and preliminary discussions with medical students. The focus groups were conducted via Zoom (Zoom Video Communications, Inc; San Jose, CA) due to COVID-19 restrictions on in-person meetings. Focus groups lasted approximately 1 hour and were audio recorded.

Data Analysis

The survey questions were tabulated and broken down based on respondent demographics as appropriate. Analysis was done in Excel (Microsoft Corporation; Redmond, WA) and Jmp Pro (SAS Institute; Cary, NC). A qualitative content analysis was performed using a rapid qualitative analysis (16–18). The reduced timeframe of rapid methods tends to be more deductive and explanatory than inductive and exploratory. Each focus group was summarized using a deductive template to structure the analysis, which was based on the interview guide (Appendix C). Two experienced qualitative researchers (LJF and SF) each summarized the same 2 focus group interviews using the template, then compared and resolved discrepancies to ensure consistency in the use of the template. The remaining 8 focus groups were summarized using the template. Template data was analyzed across all 10 focus groups to identify major themes in each category.

Results

Participant demographics

Email invitations were sent to approximately 500 medical students. There were 75 students that responded in the affirmative to the email invitation (~15% affirmative response). There were 49 participants that completed the study. Of the 49 participants, everyone (100%) completed at least part of the survey (range of questions answered: 3–16) and 48 participants (98%) completed the whole survey. The majority/plurality were female (59%) and white (41%) with a mean age of 25.3 years (Table 1). There were 14 participants in Group 1, 7 participants each in Groups 2 and 3, and 21 participants in Group 4. There was broad specialty interest with diagnostic radiology and interventional radiology receiving 12% and 14% interest respectively as shown in Figure 1. Diagnostic radiology and interventional radiology residents represent only 3.6% and 0.4% respectively of active U.S. and Canadian MD graduates in residency in 2019–2020 (19). The current opinion of radiology as a career option was diverse with a moderate positive shift: very negative (n=1, 2%), negative (n=10, 21%), neutral (n=16, 33%), positive (n=17, 35%), and very positive (n=4, 8%). However, when students were asked, “How has your opinion of radiology as a career option for you changed since you started medical school?” the responses were generally positive: somewhat more negative (n=4, 8%), no change (n=12, 25%), somewhat more positive (n=23, 48%), and much more positive (n=9, 19%).

Table 1.

Demographics of study participants.

Demographics	N (%)
Gender
Female	29 (59)
Male	20 (41)
Race/Ethnicity *
White	20 (41)
Asian	15 (31)
Black	11 (22)
Hispanic	6 (12)
Other/prefer not to answer	2 (4)
Mean age	25.3 years
Children or dependents	9 (18)
Group ǂ
1: 1st year student	14 (29)
2: 2nd year student pre-radiology clerkship	7 (14)
3: 2nd year student post-radiology clerkship	7 (14)
4: 3rd or 4th year student	21 (43)

^*Participants could choose one or more options

^ǂOne missing response

Figure 1.

Specialty interest of study participants. Participants were asked to select their top three specialties.

Stereotypes

From the survey, there was general agreement with the well-established radiology stereotypes that radiologists have flexible work hours (87% agree or strongly agree), high salary (83% agree or strongly agree), good lifestyle (93% agree or strongly agree) and limited patient contact (95% agree or strongly agree) as shown in Table 2. From the focus groups, there was broad agreement that most students had few preconceived ideas about radiology upon entering medical school. Radiology does not have a cultural footprint in media (e.g., television or movies) that would provide the basis for impressions (20). Most impressions were thus based on personal experiences with family or the parents of friends who were radiologists. However, many students noted that radiology was not on their radar of potential specialties as a result.

Table 2.

Responses to survey questions. Do you agree or disagree with the following sentiments about radiology in comparison to other medical specialties?

Questions	Strongly disagree N (%)	Disagree N (%)	Neutral N (%)	Agree N (%)	Strongly agree N (%)
Radiologists have flexible work hours	0 (0)	1 (2)	6 (13)	31 (66)	10 (21)
Radiologists make a high salary	0 (0)	1 (2)	7 (15)	29 (60)	11 (23)
Radiologists have a good lifestyle	0 (0)	0 (0)	3 (6)	28 (58)	17 (35)
Radiologists have limited patient contact	0 (0)	1 (2)	2 (4)	31 (65)	14 (29)
Radiologists get to work with emerging/advanced technology	0 (0)	0 (0)	5 (10)	22 (46)	21 (44)

Stereotypes that emerged from the focus groups can be categorized as negative, neutral, and positive as shown in Table 3. Common negative stereotypes related to a lack of patient interaction and a lack of caring about patient outcomes. Several students in Groups 3 and 4 who noted that radiologists were not as aggressive in pursuing research as colleagues in other medical specialties, especially oncology and surgery. There was also the belief that radiologists are unappreciated by patients, as referring providers get the “credit” from patients for the work that radiologists do. Although students were asked to report stereotypes they had heard, their perceptions of these stereotypes differed by exposure. Senior medical students in years 3 and 4 reported that they did not believe many of the negative stereotypes after being exposed to radiology. Specifically, students who had spent time on breast, pediatric, and interventional radiology pushed back on the notion that there was no or little patient interaction. While other students felt that the stereotype of a lack of patient interaction was solidified during their rotation.

Table 3.

Select quotes from focus group by specific themes.

Stereotypes

(Negative) “You are the one making the diagnosis for all the patients in the hospital, but the doctors are not thanking you, the patients aren’t thanking you”

(Negative) “Living in darkness” or “Basement-dwellers”

(Neutral) “Shadow of other stereotypes … the musculoskeletal radiology reading room, they’re kind of like ortho bros”

(Positive) “They were all very relaxed”

(Positive) “Balance work with family life”

COVID-19

“I found that the technological adeptness of radiologists in particular was superior to people in other specialties”

“I think it really underscores how uniquely radiology relies on technology”

Developments in the field

AI will “supplement but not supplant”

“Will I get replaced by a robot?”

Barriers to applying

“There’s a lot of emphasis placed on like Step 1 and Step 2, Step one is not really a factor anymore”

Length of training “was certainly something I took into consideration, I’ve got two kids”

“Barrier for me was finding a mentor”

Impressions based on exposure

(Initial impression) “Not at the cultural forefront of what you imagine being a doctor when you grow up”

(Initial impression) “I don’t think I would have sought out any experiences, because it wasn’t on my list”

(Over time) “Having more exposure to radiology and radiologists… would be a great recruitment”

(Over time) “Medical students might be more apt to consider it seriously if early exposure in an integrated fashion”

Hospital integration

“They kind of function as a consultant to all the different services”

“I feel like they’re pretty central to a lot of specialties.”

Neutral stereotypes included a belief that radiologists were very laid-back or relaxed, introverted, and very technology oriented. Students posited that the laid-back attitude is a product of the greater control over work hours. This sentiment was felt among respondents from all groups and reflected a belief that radiologists were “accessible,” had good “time management skills,” and were “efficient.” There was also a recognition that personalities differ greatly among subspecialties. Musculoskeletal radiology was seen to closely mirror orthopedics (e.g., “ortho bros”), while breast and pediatrics felt “more feminine.” These stereotypes were more commonly expressed by students in Groups 3 and 4 who had spent dedicated clinical time on radiology. Positive stereotypes included an excellent knowledge of anatomy, good teachers, a belief that radiologists are very smart because of the board score requirements, and that radiologists have a good lifestyle. There was not an appreciable difference in the positive stereotypes based on experience.

COVID-19

Although the study was conducted virtually due to COVID-19 restrictions, 85% (n=41) of participants reported in the survey that COVID-19 did not change their opinion of radiology as a career option. In the focus group discussions, there was consensus that COVID-19 changed the way in which medical students were exposed to the radiology rotation, but not their perception of the field. This was primarily the belief among the 2^nd year students (n=14) on their clinical rotations during 2019–2021 when the radiology rotation was a hybrid between in person and virtual depending on the subspecialty. Medical students generally felt positively that radiologists rely on technology and could easily adapt to work remotely.

Developments in the field

In the survey, 90% of participants agreed or strongly agreed that radiologists get to work with emerging/advanced technology. Artificial intelligence specifically was a major topic of conversation in the focus groups, but there were divergent opinions on the impact of AI on the future of radiology. Students with negative feelings towards radiology expressed concerns about the long-term viability of the field, while students with positive feelings felt optimistic about the role of AI. These beliefs were expressed by students at differing levels of exposure. There was broad agreement that future workflows will be altered by AI but that procedural fields were safe. Many students noted that radiologists did not seem to be concerned about job security from AI which gave them comfort for the future of the field. Some students in Group 2 also reported concerns about mid-level providers assuming more responsibilities thus limiting job opportunities for radiologists, but this was not reported among students with direct exposure.

Barriers to applying to radiology residency

Radiology was seen as one of the more competitive specialties because it is a “lifestyle” specialty. High metrics, especially Step 1 scores, were seen as a major barrier to applying for a radiology residency program. Students generally perceived that failure to achieve a sufficiently high score excluded them from considering radiology as a career option. Several students discussed uncertainty over what will happen now that Step 1 switches to pass-fail, given the past focus on high metrics to be considered for competitive specialties. The length of training was also a reported barrier for many students, especially those with families and children. Finally, students in Group 4 discussed the difficulty connecting with radiology faculty as a significant barrier to applying to radiology residency programs. Several mentioned that limited exposure to radiologists early in their medical school training prevented them from finding mentors for research and faculty willing to write a letter of recommendation for residency. (Name withheld) students have dedicated time during the third year to pursue research, so finding a research mentor early is important. Students reported that the inability to form a close relationship with an attending, such as would be formed during a more tight-knit rotation on another clinical service, created a barrier to apply. Even though (name withheld) has more radiology exposure than most medical schools, many students expressed the belief that even more integrated exposure would be a great recruitment tool.

Hospital integration

There was general agreement that radiology serves as the bedrock of patient care in the health system given the role in disease diagnosis, surveillance, and management. Radiologists were seen to have a major role as a consultant to a wide variety of services. Students noted that they were exposed to radiology on virtually every clinical rotation given how integral radiology was to patient care. Following dedicated radiology exposure during the clinical year, students reported greater interest in reviewing scans, even if they were not pursuing radiology. Participants also noted the strong partnership between radiology and surgery, trauma, as well as emergency medicine.

Discussion

This mixed-methods study provided insights into medical student perceptions of radiology, specifically in a radiology intensive medical school curriculum. The impact of medical education and experience with radiology during medical school had notable effects on medical student perceptions. First, it was clear that students enter medical school with very few preconceived notions of radiology. This tabula rosa is largely because radiology does not fit the media and popular culture image of a doctor. This provides an excellent opportunity for radiologists to craft medical student perceptions and avoid false stereotypes (Table 4). However, this also means that most medical students do not think about radiology initially as a career option and so radiologists must make concerted efforts for early outreach and recruitment. Radiology exposure, both direct and indirect, generally had a positive impact on medical student perceptions. This reinforces the findings of Branstetter et al who demonstrated that early exposure to radiology in the pre-clinical years improved the opinions of radiology later in training (3, 5). At (name withheld), there is integration of radiology into the anatomy labs which now includes point of care ultrasound. Programs should think of innovative ways to provide early exposure outside of the traditional classroom experience as a means of education, engagement, and recruitment. An expanded role for and engagement with radiology interest groups may provide an excellent platform upon which to build out these endeavors (21). Alternatively, the development of an expansive, user friendly website could create an enduring resource for radiology departments (22).

Table 4.

Proposed strategies to improve medical student perceptions of radiology.

Develop innovative approaches to increase exposure during the first year of medical school, such as during anatomy lab or via point of care ultrasound sessions

Demonstrate the breadth of radiology practice, including direct patient care, via departmental websites and rotations on more patient centric subspecialties such as breast, pediatric, and interventional radiology

Directly address false stereotypes in conferences and informal discussions

Provide teaching and educational resources, such as from the Radiological Society of North America, regarding the role of AI in radiology

Organize sessions for medical students to learn about radiology residency selection processes and training programs

Facilitate mentorship programs to encourage research and provide a source for letters of recommendation

Showcase the work-life balance of a career in radiology

Provide direct radiology contacts to fill in the gap left by an absence of radiology in the cultural media footprint

The stereotypes identified during the focus group and confirmed by the survey are not novel. However, it was interesting how many students reported that they did not believe many of the stereotypes they had heard after direct exposure to radiology. The stereotype of no patient exposure or not caring about patient outcomes was described very differently after medical student exposure to radiology. For those students who spent dedicated time on interventional, breast, or pediatric radiology this stereotype was debunked. Radiology as a field is so diverse that it does not neatly conform to the imagined stereotypes, both positive or negative. Programs should seek to show the breadth and depth of practice environments for radiologists in order to provide a more complete picture of radiology for medical students. Medical students may not be aware of the sheer volume of patients that radiologists influence (23). Showing the meaningful interactions between radiologists and patients will progressively erode inaccurate stereotypes regarding absent patient interaction. Furthermore, radiology faculty who teach medical students should consider directly addressing stereotypes during their educational sessions to ensure a consistent accurate message is delivered.

The split opinion regarding AI was a notable finding of the focus groups. AI research and clinical applications have gained a rapid foothold in radiology, and several investigators have reported some recent survey results on medical student perceptions (24–27). A survey of medical students by Park et al in 2017–2018 reported that 44% of respondents indicated that AI made them less enthusiastic about radiology (26). Similarly, a survey of Canadian medical students by Gong et al reported that 48.6% of respondents agreed that AI caused anxiety about radiology as a specialty choice (27). A key finding from our focus group was that opinions of AI paralleled interest in radiology. It is unclear if AI is driving students away from radiology or simply that those not interested in radiology are more susceptible to negative coverage. It is clear that many medical students do not have a clear understanding of the current status or potential future applications of AI and this provides an excellent opportunity for radiologists to educate students (28, 29). Faculty with direct clinical or research experience with AI could lecture medical students or programs could consider some of the many online educational resources available, such as from the Radiological Society of North America (30).

The perceived barriers to applying to a radiology residency provide some important opportunities for radiology programs. It was clear that many medical students feel that an inadequate Step 1 score removes radiology from consideration. However, the announcement of Step 1 switching to a pass or fail result should largely remove this hurdle. It is likely that other metrics will emerge to fill this void, but it is unknown at this time if they will be as restrictive to interested students (31). Given the Step 1 scoring change, departments should be proactive about holding sessions to directly educate medical students on what factors influence radiology residency selection. This could also be an opportunity to educate medical students on the length of radiology training and how it is comparable to popular fields such as medicine or pediatrics when fellowship are included and often shorter than surgical subspecialties. Another common sentiment among participants was the failure to find a radiology mentor, as students do not have the intensive one on one exposure to an attending physician that they might have on other core rotations, such as internal medicine. The failure to connect longitudinally with a radiologist reflects the organization of the medical student rotation, but also the clinical practice patterns of academic radiologists who have intermixed academic and clinical days. Programs could consider matching interested students with faculty to provide a point of contact and allow for a stronger connection to facilitate research and letter writing. Radiology interest group activities can also facilitate student and faculty interactions. Additionally, students could be encouraged to enroll in subspecialty radiology electives to ensure more time with one team.

Much has been written about the impact of COVID-19 on medical student education in radiology and the transition to remote/virtual learning (32–34). While it is very clear that radiology education has been disrupted, a finding confirmed by our focus group participants, this was not viewed as either a net negative or positive. For many focus group participants, it simply highlighted the adaptability of radiology and the ability to work well with new technology platforms. It is worth noting that lack of engagement with radiologists was listed as a barrier to applying to radiology and the dynamics of remote learning will likely exacerbate this further.

There are limitations intrinsic to this study and to mixed-methods work in general. The more intensive radiology exposure, including a core rotation, at (name withheld) provides an environment to study the influence of more direct and prolonged radiology education and clinical encounters. But the findings may not be generalizable to schools with more limited traditional radiology curricula. There is selection bias in who chooses to participate in the study, and 15% of respondents expressed interest in the study. Nonetheless there was a broad range of positive and negative opinions about radiology. There were changes to the radiology core rotation during the study period and so some students were given the opportunity to shift from the previous schedule that rotated in each subspecialty for 2–3 days to a new scheduled that allowed one week each on four subspecialty areas. COVID-19 influenced the learning environment for current second year students. Finally, Zoom offers a focus group experience a little different than a traditional round table discussion.

This mixed-methods study of medical student perceptions of radiology identified many findings that can influence the development of medical student education programs and recruitment efforts. Radiologists should seek out opportunities early in medical school training to engage students when they have few preconceived notions of radiology and before incorrect stereotypes can be established. Medical students would benefit from exposure to the breadth of radiology subspecialties to provide an accurate representation of the field. Radiology exposure also helped to reduce false stereotypes of the field. Finally, radiologists should educate medical students about the role of artificial intelligence and provide opportunities for mentorship and research. Although these proposed changes will require significant time and effort from radiology programs, they will help to ensure that future of our field by facilitating the recruitment of well-informed and motivated medical students.

Abbreviations:

AI

artificial intelligence

APPENDIX

Appendix A: Medical student survey administered via RedCap.

Appendix B: Focus group interview guide

Appendix C: Focus group coding template