The national physician shortage in the United States, exacerbated by the COVID-19 pandemic, presents an ongoing challenge for radiology practices in meeting growing imaging volumes and expectations for patient access to care. These factors in turn contribute to radiologists' stress level and adversely impact radiologists' well-being [
1]. A study by Metter et al. [
2] showed that, although the total number of radiologists in the United States increased from 2007 to 2017, the rate of increase slowed between 2015 and 2017, raising concerns about whether the growth would be sufficient to address the imaging needs of the aging population. According to a 2019 report by the Association of American Medical Colleges, 53% of active radiologists are 55 years old or older, with only seven of 46 other specialties having a higher percentage of active physicians in this age group [
3]. As radiologists age along with the general population, the need for strategies to fill gaps in the physician workforce becomes more relevant.
International medical graduates (IMGs) represent a source of physicians who could play a role in alleviating the radiologist shortage in the United States. IMGs are medical students or medical school graduates who are receiving, or who received, their medical school training outside of the United States or Canada. The National Resident Matching Program (NRMP) classifies IMGs as either U.S. IMGs, defined as IMGs who are U.S. citizens, or non-U.S. IMGs, defined as IMGs who are not U.S. citizens [
4]. IMGs play a critical role in providing health care to many Americans, especially in underserved and lower-income areas of the United States [
5]. Indeed, approximately 21 million people live in areas of the United States where IMG physicians constitute a minimum of 50% of the physician workforce [
1].
IMGs currently comprise 11% of all U.S. graduate medical education (GME) diagnostic radiology trainees and 12% of the active practicing radiologist workforce [
6]. While completing U.S.-based residency training, these individuals are usually supported by a J-1 visa, as endorsed by the Educational Commission for Foreign Medical Graduates (ECFMG); less commonly, these individuals are supported by an H-1B visa during U.S. residency training.
After completion of a U.S. radiology residency program, IMGs may pursue job opportunities at radiology practices in the United States [
7,
8]. However, IMGs seeking faculty positions in the United States face unique challenges associated with the visa options for independent practice [
9]. In general, to remain in the United States and begin clinical practice as radiologists, IMGs are required to switch from the J-1 visa that is typically in place during training to a different visa status. Specifically, H-1B and O-1 visas are the visas most conducive to physicians being hired as faculty within U.S. academic departments.
Academic radiology departments vary in their policies and practices for the hiring of IMG faculty candidates, in part because departments are subject to larger institutional norms for addressing visa options. These differing practices can lead to potential barriers in the recruitment of IMGs. Such obstacles in turn have broad implications for the radiologist job market in the United States in terms of the ability of this potential source of faculty candidates to help address the workforce shortage. We therefore conducted this study to assess the policies and experiences of U.S. academic radiology departments in offering visas to IMGs applying for faculty positions.
Methods
The institutional review board of Johns Hopkins University approved this study under exempt status. Completion of the survey indicated consent to participate, and additional written informed consent was not required.
Description of Survey
The survey questions were written by the study investigators, one of whom (D.M.Y.) has training and experience in creating academic surveys. The survey comprised a total of 14 questions, some with multiple subparts. The survey questions related to the visa policies and practices of U.S. academic diagnostic radiology departments for hiring IMG candidates for faculty and trainee positions. For example, questions asked how frequently the department sponsors particular visas for eligible individuals, the department's baseline visa policy for non-U.S. citizens applying for faculty positions (hereafter described as the department's default policy), reasons why certain visas are not sponsored, and whether the department establishes its own visa policy. Additional questions related to department characteristics (e.g., department type, faculty size, and U.S. state). Certain questions or response options were explicitly related to J-1, H-1B, or O-1 visas. Information about these three visas is provided in
Table 1 and the
Supplemental Methods.
Survey questions comprised primarily multiple-choice questions, which asked the respondent to select the single best answer or to select all answers that applied. Some of these questions included a free-text field that accompanied particular response options and provided an opportunity for further commentary or clarification. Numerous survey questions included the response option “I don't know.” Three survey questions had separate subparts for faculty and trainees. Two survey questions, both of which related to H-1B visas, collectively referred to faculty and trainees. One survey question was purely a free-text response question asking for any comments on the department's visa policy or visa issues at the respondent's institution. One survey question requested that if the individual who initially received the survey was not knowledgeable about the department's visa policy, that individual should provide the name and e-mail address of the person at their institution who would be most knowledgeable about their department's visa policy, so that person could be contacted as an alternate respondent from the institution.
Survey Administration
In September 2022, a directory from the NRMP was reviewed for chairs of U.S. radiology departments with a diagnostic radiology residency training program, with 193 individuals identified. The chairs' e-mail addresses were then obtained through publicly available resources, including institutions' websites and the online membership directories of the Radiological Society of North America (RSNA) [
10] and the Society of Chairs of Academic Radiology Departments (SCARD) [
11]. An e-mail address was identified for all 193 chairs in the initial sample.
The survey was distributed to participants via e-mail by use of web-based survey software (Qualtrics Research Suite, version June 2023, Qualtrics). The survey was initially distributed on October 7, 2022. Six reminder e-mails, released 3–7 days apart, were sent to nonrespondents, with the last reminder sent on November 7, 2022. The survey's introductory e-mail indicated that participation was voluntary. All survey items were optional. When the initial survey respondent indicated that an alternate individual was most knowledgeable about the department's visa policies, then the survey was sent to the indicated alternate individual, who acted as a surrogate respondent from the given department.
Statistical Analysis
The survey response rate was determined as the ratio between the number of surveys in which the recipient responded to at least one item and the number of survey recipients after exclusion of those who were found to have invalid e-mail addresses at the time of initial survey distribution. When the initial respondent provided information for an alternate individual who was most knowledgeable about the department's visa policies, the initial individual's responses were discarded from the analysis, and only the responses of the surrogate (if they responded) were included.
The survey responses were summarized descriptively using counts and percentages. When percentages were computed for response options to individual questions or when other statistical comparisons based on response frequencies were performed, “I don't know” responses were excluded. The frequencies with which H-1B and O-1 visas were available to at least some extent were computed as the sums of the percentages of departments that offered these visas frequently, sometimes, or rarely. Themes associated with free-response portions of the survey questions were qualitatively identified by the study investigators.
On the basis of the survey responses, departments were classified as university-based departments or non–university-based departments (with the latter encompassing community-based departments, university-community hybrid departments, government, Veterans Affairs [VA] or military departments, or departments designated “other”), and they were also classified as small (50 or fewer faculty members) or large (more than 50 faculty members). Department location was classified in terms of U.S. Census Bureau regions [
12].
The frequency with which H-1B visas and O-1 visas were offered to at least some extent (either frequently, sometimes, or rarely) was compared between university-based and non–university-based departments as well as between small and large departments, by use of chi-square tests.
Given the present investigation's focus on the U.S. radiologist workforce shortage, the responses to survey item subparts that related specifically to trainees are not presented among the results of the current study.
p values less than .05 were considered statistically significant. Statistical analysis was performed using SPSS version 26 (IBM).
Results
Survey Respondents
Of the 193 initial survey recipients, 16 were found to have an invalid e-mail address, resulting in a maximum of 177 possible participants. Six of the initial survey respondents provided contact information for a surrogate respondent; all six of these surrogates were contacted and participated in the survey. The overall survey response rate (i.e., completion of at least one survey item) was 81% (143/177) when surrogate survey recipients were included to replace corresponding initial survey recipients. Other than questions that received an entirely free-text response, the response rate for individual questions (inclusive of “I don't know” responses) ranged from 62% (110/177) to 67% (118/177).
Department Characteristics
Table 2 summarizes the characteristics of the departments represented by survey respondents. A total of 68% (75/110) of respondents were from university-based departments, and 32% (35/110) were from non–university-based departments. A total of 55% (61/110) of respondents were from large departments (i.e., more than 50 faculty members), and 45% (49/110) were from small departments (i.e., 50 or fewer faculty members). Respondents were from departments in 40 different U.S. states as well as Washington, DC, with 25% (27/110) from the Midwest, 27% (30/110) from the Northeast, 28% (31/110) from the South, and 20% (22/110) from the West. The largest number of respondents were from departments in New York (
n = 12) or California (
n = 11).
Visa Policies and Experiences
Table 3 summarizes the frequency with which departments offered H-1B and O-1 visas for eligible faculty. A total of 24% (28/115), 38% (44/115), 17% (20/115), and 20% (23/115) of departments offered H-1B visas frequently, sometimes, rarely, and never, respectively. A total of 3% (3/113), 27% (31/113), 22% (25/113), and 48% (54/113) of departments offered O-1 visas frequently, sometimes, rarely, and never, respectively. Thus, 80% (92/115) and 52% (59/113) of departments offered H-1B and O-1 visas to at least some extent.
Table 4 summarizes the responses to additional survey questions relating to departments' visa policies and experiences. The default policy of departments for offering visas to IMG faculty applicants was to “offer the positions preferentially to American citizens” in 28% (32/113), to offer H-1B visas in 41% (46/113), to offer O-1 visas in 5% (6/113), to not offer visas in 11% (12/113), and to follow some other policy (selecting “other” as a response) in 22% (25/113). Fourteen respondents provided a comment in the question's free-text portion that indicated that their default policy was based on recruiting the most qualified candidates, regardless of a candidate's visa status.
A total of 71% (72/102) of respondents indicated that their department's policy did not vary depending on whether the faculty candidate was a research or clinical applicant. Of the 30 respondents who indicated that their policy varied depending on this factor, eight provided a comment in the question's free-text portion that indicated that they preferred to sponsor visas for research rather than clinical faculty because of the perceived greater feasibility of obtaining H-1B visas for research faculty.
The most common reasons given for why departments do not offer H-1B visas were as follows: time-consuming (25% [29/117]), the lack of reliability of the candidate's starting time because of embassy or immigration issues (22% [26/117]), and the expense of the visa application (22% [26/117]). The most common reasons given for why departments do not offer O-1 visas were as follows: the time-consuming process (20% [23/114]), the lack of reliability of the candidate's starting time because of embassy or immigration issues (16% [18/114]), the expense of the visa application (11% [13/114]), and lack of expertise (11% [13/114]). Additional reasons provided by at least three departments for both H-1B and O-1 visas included lack of applicants for the visa and a preference to employ U.S. citizens; for the O-1 visa, additional reasons included the need to pay IMGs more than U.S. citizens to match prevailing wages. In the free-text portion of this question for H-1B visas, 20 respondents indicated that the policies were established by the institution or GME office in a manner that was beyond the department's control, and two respondents indicated that state regulations were dominant in their environment and governed the ability to offer visas for employment. In the free-text portion of this question for O-1 visas, 14 respondents expressed concern about the high bar set for qualifying for this visa. For example, one respondent stated the following: “Our university interprets the rules as though O-1 visa candidates should be eligible to win the Nobel prize, etc.”
A total of 15% (16/108) of respondents indicated that their department sets its own policy for offering visas; 75% (81/108) specified that their institution sets the policy, and 10% (11/108) indicated that the policy is set by an entity other than the institution or department. Responses to this question's free-text portion indicated that departments' policies were set by the U.S. Department of State–Bureau of Consular Affairs (identified in the survey as the “U.S. Visa Office”), medical school dean, state government, or an international office.
A total of 41% (36/87) of respondents stated that their department's H-1B visa policy resulted in the department losing applicants to another institution either frequently or rarely. A total of 48% (42/87) of respondents stated that the department has had IMG applicants but the applicants declined for reasons not related to the visa policy. Five respondents to this question stated that the faculty candidates accepted the department's employment offer using J-1 or O-1 visa options; four respondents to this question stated that their department had not had IMG faculty applicants.
Table 5 summarizes responses to survey questions assessing departments' experiences in offering visas to IMG faculty applicants in terms of counts of individuals. Over the past 10 years, 16% (15/96) of departments had not considered any faculty applicants that required visas, 66% (63/96) considered one to 12 such faculty, and 19% (18/96) considered more than 12 such faculty. A total of 42% (42/99)ad been receiving) visas, 18% (18/99) had five to eight such faculty, and 18% (18/99) had nine or more such faculty.
Additional Free-Text Comments
Representative responses to the survey question asking for additional free-text responses regarding the departments' visa policy or the issue of visas at the institution can be found in
Appendix 2.
Stratification by Department Characteristics
University-based departments, in comparison with non–university-based departments, had significantly higher rates of offering H-1B visas (89% [65/73]) vs 68% [23/34]; p = .01) and of offering O-1 visas (63% [45/72] vs 38% [13/34]; p = .02) to at least some extent. Large departments, in comparison with small departments, had significantly higher rates of offering H-1B visas (93% [56/60] vs 67% [31/46]; p < .001) and of offering O-1 visas (68% [41/60] vs 37% [17/46]; p = .001) to at least some extent.
Discussion
IMGs could potentially help address growing radiologist work-force shortages. However, this national survey found that, although most U.S. radiology departments offered H-1B and O-1 visas to IMG faculty applicants, such applicants face a variety of potentially substantial barriers in successfully obtaining positions. For example, only 41% and 5% of departments offered H-1B and O-1 visas, respectively, to IMG faculty applicants as their default policy. Departments alternatively had default policies of preferring U.S. citizens or of not offering visas at all. Reported obstacles to offering visas included monetary and time expenses, uncertain timelines for the process, regulatory issues, and the need for expertise.
Only 15% of departments set their own visa policies, whereas 75% of departments followed a policy set by the institution. Respondents also indicated the potential impact of factors originating outside the institution on the institution's visa policies. For example, states vary in the ease with which IMGs in the state obtain a medical license; an inability to obtain a medical license would essentially preclude an applicant from obtaining a clinical faculty position at a given institution.
The presence of IMGs has been shown to increase the racial, ethnic, and cultural diversity of the physician workforce [
13,
14]. Thus, the challenges faced by IMGs in obtaining faculty positions are expected to hinder opportunities to increase the diversity of the radiologist workforce in the United States. The benefits of a diverse physician workforce, paralleling the diversity of the U.S. population, have been described [
15–
17].
One of the most common reasons given for not offering H-1B visas was the lack of reliability of the candidate's start date. This issue may reflect the perceived or real challenges of meeting regulatory requirements. For example, current U.S. immigration laws provide J-1 visa holders (most often trainees including residents and fellows) two options for becoming eligible for an H-1B visa or permanent U.S. residency: fulfilling a minimum of a 2-year home return or waiving this obligation through the receipt of a J-1 waiver [
18]. An interested U.S. federal government agency must submit an application requesting sponsorship for the non-U.S. IMG to waive the home return requirement. After waiver applications are reviewed by the U.S. Department of State, the physician is required to practice for at least 3 years in a VA hospital or Health Professional Shortage Area [
7]. Alternatively, IMGs may apply for a J-1 waiver by providing evidence either that the applicant has family members who are U.S. citizens and will be exposed to exceptional hardship or that the applicant may be subject to persecution after returning to their home country [
19].
The radiology community should attempt to address the reasons why departments are reluctant to offer visas to non-U.S. IMG candidates through education and training. For example, informational seminars could be delivered at meetings of professional societies, providing instructions on visa policies. Standardized visa application templates could also be made available for radiology departments. With increasing experience in sponsoring visas, departments will gain knowledge of the typical time-lines for visa procurement, which may help reduce the ambiguity regarding start dates.
U.S. health care systems have been reported to be reluctant to hire IMGs for numerous reasons other than visa issues [
18]. These reasons include potential biases and misperceptions regarding IMGs as well as discrimination against IMGs. The present survey identified a small number of departments that did not offer visas due to a stated preference to hire U.S. citizens. It will be important for future studies to systematically evaluate such non–visa-related barriers for the hiring of IMGs by academic radiology departments and to seek strategies to address these issues within the specialty.
This study had limitations. First, the survey did not consistently differentiate visa policies for IMGs who completed radiology residency in the United States or Canada versus those who did not. IMGs who did not complete radiology residency in the United States or Canada would need to apply to the alternate pathway program offered by the American Board of Radiology (ABR), whereby the individual must (among other requirements) spend 4 years at a single institution to qualify for ABR Diagnostic Radiology Certification [
8]. When applying for a faculty position, candidates applying for the ABR alternate pathway program experience visa issues similar to those experienced by candidates who completed radiology residency in the United States or Canada; however, they may nonetheless face greater challenges overall in seeking successful employment related to the additional need to fulfill the ABR requirements. Second, a fraction of surveyed individuals did not respond, introducing potential bias. Nonetheless, the overall response rate was high, exceeding target survey response rates provided in various sources [
19,
20]. Third, although the overall survey response rate was high, the rates of response to individual questions were variable, and respondents at times indicated that they did not know the response to a particular question. Fourth, although department chairs represented the initial surveyed individuals, practices and experiences may have been shaped by other individuals with perspectives different from those of the chair (e.g., a division chief). Fifth, exclusion of departments that were not listed by the NRMP as having an affiliated diagnostic radiology residency program affects the generalizability of the study. Finally, IMG faculty applicants themselves were not surveyed regarding their experiences in applying for U.S. faculty positions and associated challenges.
In conclusion, although most U.S. academic radiology departments offer, at least to some extent, H-1B and O-1 visas for IMG candidates seeking faculty positions, the use of such visas typically is not the default policy of departments. Barriers to offering visas included financial constraints, time commitments, and regulatory burden. Radiology departments' policies for offering visas for IMG faculty applicants were primarily determined at the institutional level and potentially were also impacted by state-level policies. Initiatives by radiology departments to increase the availability and utilization of visas for IMG faculty applicants, to the extent permitted by institutional and state policies, could help expand the nation's radiologist supply and thereby address the developing U.S. radiologist shortage.