The number of international medical graduates (IMGs) practicing in the United States has increased steadily over the past four decades, going from 10% in 1963 to 18% in 1970 (1), and currently comprising approximately 25% of the physician workforce and 27% of residents (2, 3). In 2005–2006, patient visits to IMGs comprised 24.6% of all visits to office-based physicians (4). Compared with graduates of U.S. and Canadian medical schools, a greater proportion of IMGs specialize in the primary-care disciplines, locate in areas of need, and take care of poorer patients (5, 6). Likewise, on average, IMGs’ patients are more likely than U.S. medical graduates’ patients to live in areas with lower median household incomes and a higher percentage of households with incomes below the federal poverty level (4). As part of the overall U.S. healthcare delivery system, IMGs have both gap-filling and safety-net roles (7–9). Most important, they provide healthcare to underserved populations by entering specialties and geographic areas that U.S. medical graduates (USMGs) tend to avoid. Even with increases in the number of U.S. medical schools and associated enrollment, the projected shortfall in physicians in the U.S. (10, 11) will likely necessitate that IMGs, across all specialties, continue to play a significant patient-care role, especially for underserved populations, for some time to come.
There have been numerous studies, both specialty- and non–specialty-specific, that have looked at the demographics, distribution, and practice patterns of IMGs in the U.S. healthcare system (12–14). Studies have specifically noted a shortage of specialists (15), including psychiatrists (16). Maldistribution of psychiatrist physicians has also been documented, with workforce needs not likely to be met in some health services areas, particularly those that are rural (17). The maldistribution problem could be exacerbated by a number of factors, including the composition of the workforce (e.g., IMG proportion, gender), the availability of residency positions, effects of managed care, expansion of the healthcare system, interest in the field of psychiatry, and even the burden of psychiatric illness. Historically, IMGs have played a critical role in some psychiatric disciplines (e.g., child psychiatry), but recent trends reveal that the numbers selected for training in this discipline have decreased (18). Nevertheless, over one-third of current psychiatric residents are IMGs (3), and their contribution to the diversity of the workforce has been documented in numerous studies (19, 20). It thus seems likely that, even with some variability in IMG entrance to various specialties, IMGs will continue to play a significant role in the U.S. psychiatric workforce for some time.
Given the need for competent practitioners, regardless of medical school of origin (21), more detailed investigations of quality of care for various provider groups are needed. Unfortunately, although the role of IMGs in the U.S. healthcare system has been extensively documented (22, 23), there have been relatively few studies that have specifically focused on quality of care (12). Given the difficulty of attributing patient-care outcomes to individual practitioners, other markers of competence, including specialty board certification and maintenance of certification, have been forwarded. As an indicator of quality (i.e., demonstrated competence to provide quality care), there is considerable evidence indicating a relationship between specialty certification and later performance in practice (24). As such, studies that document board-certification, especially for distinct cohorts of physicians (including psychiatrists), can be valuable, at least in aggregate, for making general inferences regarding quality of care.
International medical graduates are a heterogeneous group, having immigrated to the United States from over 130 countries and territories. Since IMGs will continue to play a role in U.S. healthcare and, specifically, in U.S. mental health care for some time to come, knowing more about the supply, distribution, and characteristics of IMG psychiatrists can help inform specialty-specific workforce policies. The purpose of this study was to document the characteristics and qualities of IMG psychiatrists in the United States. Where appropriate, we make comparisons with U.S. graduates.
We used the 2010 American Medical Association Physician Masterfile (AMA Masterfile) to obtain information on physicians who currently practice medicine in the United States. The AMA Masterfile includes data on all physicians who have met the requirements for recognition as a physician. For IMGs, a record is started when the individual enters a graduate medical education (GME) program. Additional information is added from primary sources and surveys of the physicians.
For this study, AMA Masterfile data were merged, via a common identifier, with data from the Educational Commission for Foreign Medical Graduates (ECFMG). ECFMG is the certifying body for all international medical graduates who seek GME positions in the United States. ECFMG certification requirements are described elsewhere (25). To be eligible for accredited GME training, IMGs must have achieved ECFMG certification; to be eligible for licensure in any of the U.S. jurisdictions, some GME training is also required.
The American Board of Medical Specialties (ABMS) maintains current and historical data on member board-certification, including certification by the American Board of Psychiatry and Neurology (ABPN). Board certification is achieved through satisfactory completion of an accredited GME program and successful examination performance. Certification by an ABMS-member board is not, however, required to practice in a specialty. Physician board-certification data were obtained as part of an ECFMG–AMA–ABMS data exchange and were merged with the combined AMA/ECFMG database.
This study was exempt from IRB review. ECFMG registrants have acknowledged that their data can be used for research purposes. Personal identification information has been removed, and only group-level results are reported.
Both the ECFMG database and the AMA Masterfile contain demographic information. Major professional activity (resident, office-based practice, etc.), type of employment (solo practice, state hospital, etc.), primary self-designated practice specialty, location (state), and type of physician (IMG, U.S. M.D., or U.S. D.O.), were derived from the AMA Masterfile. For IMGs, additional demographic data were drawn from ECFMG records. This included citizenship at the time of entry into medical school and country of medical school training. An IMG was defined as an individual who graduated from a medical school located outside of Canada or the U.S., regardless of citizenship. Citizenship at medical school was used to classify IMGs as either U.S.-citizen IMGs (USIMGs) or non-U.S. citizen IMGs (non-USIMGs).
Of the over 200 self-designated practice specialties represented in the AMA Masterfile, only those physicians who indicated a psychiatric specialty were selected. Psychiatrists were accordingly limited to those individuals who indicated Addiction Psychiatry, Child and Adolescent Psychiatry, Forensic Psychiatry, Geriatric Psychiatry, Psychiatry, Psychoanalysis, or Psychosomatic Medicine as a practice specialty. Interdisciplinary subspecialties (e.g., Clinical Neurophysiology, Pain Medicine, Sleep Medicine, Hospice and Palliative Medicine) were excluded.
For most parts of this investigation, the study group included all physicians involved in patient care activities (residents, full-time hospital staff, or those in office-based practice) whose primary self-designated practice specialty was one of the psychiatric specialties listed above. For the analyses of specialty board certification rates, the resident psychiatrists were excluded.
Descriptive statistics were used to determine the distribution and characteristics of psychiatrists in the United States. Although IMG and non-IMG comparisons were made, no inferential statistics were applied, since the study group includes the whole population of practicing psychiatrists in the U.S.
The 2010 AMA Physician Masterfile includes 804,445 physicians involved in patient-care activities (full-time hospital staff, those in office-based practice, and residents). For the cohort of all practicing physicians for whom self-designated practice specialty was available (N=786,272; 97.7%), including IMGs, D.O.s and M.D.s, there were 46,322 psychiatrists (self-designated practice specialty of Addiction Psychiatry, N=330; Child and Adolescent Psychiatry, N=7,254; Forensic Psychiatry, N=388; Geriatric Psychiatry, N=752; Psychiatry, N=37,239; Psychoanalysis, N=398; Psychosomatic Medicine, N=61). Overall, psychiatrists make up 5.9% of the patient-care workforce. The total number of psychiatrists, including those not directly involved in patient-care activities, was 63,134 (N=43,883 D.O.s or M.D.s; N=19,251 IMGs). Although most of the M.D. and D.O. psychiatrists were directly involved in patient-care activities (N=31,943, 72.8%), a greater percentage of IMGs were classified as such (N=14,379; 74.7%).
A breakdown of the psychiatric workforce (those involved in patient-care activities), by medical school type, is provided in Table 1.
TABLE 1.Psychiatrists in Patient-Care Activities
| Add to My POL
|Type||N||% of All Psychiatrists in Patient Care|
|IMG: U.S. citizen (USIMG)||2,415||5.2|
|IMG: non-U.S. citizen (non-USIMG)||11,509||24.9|
|Allopathic physicians (M.D.)||30,178||65.2|
|Osteopathic physicians (D.O.)||1,765||3.8|
Counting all psychiatric disciplines, IMGs represent almost one-third of the workforce (31.0%). For all other self-designated specialties (non-Psychiatry, but involved in patient-care activities), IMGs (N=173,759) represent only 23.5% of the workforce. Eliminating the IMGs for whom citizenship information at medical school was not available, U.S. citizen-IMGs (USIMGs) make up 17.3% of the internationally-educated (IMG) psychiatrist workforce.
The majority of IMG psychiatrists in patient care are male (N=8,684; 60.5%), which is similar to the gender breakdown for USMG psychiatrists (N male=19,899; 62.3%). Compared with the age of USMG psychiatrists (mean=53.2 years; standard deviation [SD]=13.5), practicing IMG psychiatrists, on average, are a little older (mean=54.7 years; SD=13.1). Native language was available for most of the IMG cohort. Over 82% (N=11,156) indicated that English was not their native language, although at least 70% received their medical degrees from schools where English was a language of instruction.
On the basis of present employment data from the Masterfile, IMG psychiatrists were less likely to be self-employed in solo practice (N=2,669; 18.6%) than USMG psychiatrists (N=9,373; 29.3%). IMG psychiatrists were more likely to be employed in a city/county/state government hospital (N=1,738; 12.1%) than USMG psychiatrists (N=1,673; 5.2%).
International Medical Graduate (IMG) Psychiatrists
Of the 14,379 IMG psychiatrists in patient-care activities, 1,756 were residents (32.1% of all 5,463 psychiatry residents); 9,347 were in office-based practice (28.4% of all 32,966 psychiatrists in office-based practice); and 3,276 were full-time hospital staff (41.0% of all 7,893 psychiatrists who were full-time hospital staff).
Country of medical school training could be determined for 14,161 (98.5%) of the IMG psychiatrist cohort. The top 10 (of 131) countries of medical school education are shown in Table 2. Nearly half of the IMG psychiatrists attended medical school in South-Central Asia (N=4,942; 34.90%) or the Caribbean (N=1,704; 12.0%). Citizenship at medical school was available for 13,924 (96.8%) of the IMG psychiatry cohort. The top 10 (of 139) countries of citizenship of IMG psychiatrists are also shown in Table 2. Contrasting the countries of medical school training and citizenship, it is clear that many individuals attended medical schools outside of their home country. For those who were U.S. citizens at entry to medical school, the most common country of medical school training was Mexico (N=470; 19.5%) followed by the Dominican Republic (N=375; 15.5%). On the basis of the top 10 countries of medical school training listed in Table 2, many students, even excluding those educated in the Caribbean and Mexico, were still American citizens. For example, of all the IMG psychiatrists who attended medical school in Spain (N=282,104 [41.8%]) were American citizens. A proportionately large number of Cuban citizens (N=58 [23.3%]) also attended medical school in Spain. In contrast, most of the psychiatrists who attended medical school in India (N=3,374) were Indian citizens (N=3,227; 95.6%). Of all practicing psychiatrists who attended medical school in the Caribbean (N=1,704), almost two-thirds (N=1,085, 63.7%) were U.S. citizens at entry into medical school.
TABLE 2.Medical School Country for IMG Psychiatrists in Patient-Care Activities
| Add to My POL
|Medical School Country||N||%||Citizenship at Medical School||N||%|
|Dominican Republic||617||4.4||South Korea||333||2.4|
|All others||5,262||37.2||All others||4,123||29.6|
Although nearly one-quarter of the practicing IMG psychiatrists attended medical school in India, the top two provider medical schools are located in the Philippines (University of Santo Tomas: N=409; 2.9%) and Mexico (Universidad Autónoma de Guadalajara: N=352; 2.5%). Two of the top five schools for educating IMG psychiatrists are located in the Caribbean: Ross University (N=245; 1.8%) and St. George’s University (N=244; 1.8%).
IMG psychiatrists practice in all 50 states. The number of IMG psychiatrists by state (top 10) and total number of practicing psychiatrists are shown in Table 3. Although many of these states have large proportions of IMGs across all specialties, in some states (e.g., New Jersey), IMG psychiatrists are present in even greater proportions, representing more than 50% of the workforce involved in patient-care activities.
TABLE 3.Number of Psychiatrists in Patient-Care, by State (top 10)
| Add to My POL
|State||IMG Psychiatrists (Patient Care)||Total Psychiatrists (Patient Care)||% IMG||% IMG, All Specialties|
Excluding residents (N=5,463), there were 40,859 psychiatrists in patient-care activities. The majority of these practitioners (N=28,825; 70.6%) were board-certified in a psychiatric specialty or subspecialty. Specialty board certification data in the primary board or subspecialty, by provider type (USIMG, non-USIMG MD) are presented in Table 4. Compared with U.S. graduates, the board-certification rates for IMGs, particularly USIMGs, are much lower.
TABLE 4.ABPN Certification of Psychiatrists in Psychiatric Specialties, by Physician Type
| Add to My POL
|Physician Type||Total Patient Care (excluding residents)||Total Board-Certifieda||% Board-Certified|
|M.D. (U.S. allopathic)||26,799||20,796||77.6|
International medical graduates play a key role in the provision of healthcare in the United States. They have diverse educational and ethnic backgrounds, and, therefore, are especially valuable in caring for an increasingly diverse U.S. population, with variable English language proficiency. Their personal experience with cross-cultural issues, non-English language proficiency, and sensitivity to other ethnic groups are certainly beneficial for psychiatric care (7). Although U.S. medical schools continue to expand to meet current physician shortfalls, recently enacted healthcare reform will likely result in an even greater demand for physician services. If the number of available residency positions remains fixed, future U.S. GME opportunities for IMGs may wane as more U.S. graduates fill those slots. However, given the working lifespan of a physician, there still is, and will continue to be, a significant role for IMGs in today’s healthcare system, particularly for underserved areas and populations. From a workforce policy perspective, whether specific to psychiatry or to other disciplines, it is important to know who they are, where they came from, how they practice, and, where measurable, their competence to provide quality care.
On the basis of the 1996 National Survey of Psychiatric Practice data, Blanco estimated that 22.3% of active psychiatrists in the U.S. were IMGs (6). The current analysis of the 2010 AMA Masterfile shows that over 30% of practicing psychiatrists in the U.S. are IMGs, a notable increase. With respect to physicians at the beginning of their careers, IMGs constitute over one-third of the psychiatric residents. Whether IMG specialization is a matter of choice, or simply a reflection of gap-filling after U.S. graduates have chosen their career paths, it remains that psychiatric care is, and will continue to be, heavily dependent on IMGs. However, looking to the longer-term future, if there is no broad expansion of residencies to accommodate the larger output from U.S. medical schools, the option for IMGs to fill certain specialty gaps could eventually be eliminated. To the extent that U.S. graduates are motivated—or not—to practice in the psychiatric specialties, shortages in some disciplines (e.g., Child Psychiatry) may not improve (18).
Over 10 years ago, Blanco et al. found that IMG and USMG psychiatrists had different practice patterns (6). Although the current study does not include contrasts of work-hours or patient characteristics, IMG psychiatrists were found to be older and more likely to be employed in hospitals, especially government hospitals. With respect to practice location, it is reasonable to assume that IMGs are more likely to work in government hospitals because that is where they received their residency training (26). Likewise, because of acculturation and language issues, they may, at least initially in their career, resist entering solo or even group practice and choose to live and practice in areas where there are greater concentrations of foreign-born individuals. As noted in previous studies (4, 5, 17), it is logical to conclude that IMGs and USMGs still play somewhat different roles in psychiatric practice. Therefore, if policies serve to restrict the availability of IMGs, patient care in some areas, especially for more vulnerable groups, may eventually suffer (4).
The geographical results, presented at the state level, suggest that some states heavily rely on IMGs to provide psychiatric care to their populations. Although this result may be a function of individuals staying in states where they did their residency training, and may not directly point to an overall maldistribution issue, it does suggest that the overall ebb and flow of IMGs into psychiatric residency training programs could have an impact on the availability of psychiatric care in some areas. As documented for the U.S. and other countries that rely on IMGs, a large percentage of IMGs, including those who become psychiatrists, originated from one of three countries: India, the Philippines, or Pakistan (27, 28). To the extent that emigration from these countries changes, and different groups choose different specialties or locations for specialty training, the state-based patterns of IMG psychiatric dependency may also change. Unfortunately, this cross-sectional analysis of the 2010 AMA Masterfile does not allow for discussion of any specific trends. Nevertheless, given changing demographics of individuals achieving ECFMG certification in the last 10 years (29) and the propensity for USIMGs to secure residency training positions ahead of their non-U.S. citizen counterparts (23), the future psychiatric workforce, at least in terms of IMG characteristics, may be quite different.
Although a sizable proportion of IMG psychiatrists have achieved board-certification in their practice specialty, U.S. M.D.s, on average, have much higher board-certification rates. This finding, noted elsewhere, is generally consistent across all specialties (4, 30). Still, compared with overall board-certification rates across specialties, proportionately fewer psychiatrists achieve this status. This may be due to a number of factors, including the historically challenging nature of the written and oral examinations, less perceived need for certification as a marker of personal and professional recognition, and absence of service delivery system and insurance panel requirements for subspecialty certification in psychiatry (31). Additional longitudinal research on board-certification rates among IMG psychiatrists and their potential relationship to patient outcomes, is certainly warranted.
Historical data from ECFMG indicate that the number of U.S.-citizen IMGs achieving ECFMG certification is increasing (29). These individuals, upon achieving certification, are more likely to secure residency positions (23). This trend, combined with the facts that specialty board certification has a demonstrated relation to physician competence (24), and USIMGs have the lowest board certification rates, has some implications for the future psychiatric workforce. The lower board-certification rates of USIMGs in psychiatric (or any other) disciplines are unlikely to be associated with language difficulties. Instead, they may be due, at least in part, to admission criteria of the schools they attended, the variability in educational experiences the schools provide, selection factors in residency training programs (32), other experiential factors, or, simply, lower overall ability. Since more USIMGs are coming into the system, and many would certainly end up involved in direct patient care in the psychiatric disciplines, possibly without board certification, this could have an impact on clinical outcomes. Going forward, and, given that changes in the healthcare system have placed greater emphasis on board certification, policymakers should be aware of the fact that the credentials of IMGs may be increasingly associated with their ability to fill the workforce gap and maintain the quality of care (24).
Although this analysis provides a detailed cross-sectional overview of the role of IMG psychiatrists in the United States, it is not without limitations. First, some of the data, such as practice specialty, are self-reported. It is therefore possible that some estimates are biased. Second, even though the analyses of board-certification were restricted to active physicians in patient care (i.e., full-time hospital staff, office-based practice), there would still be some individuals, perhaps just having finished a residency program, who have not had the opportunity to achieve board certification. This may explain the small differences between the rates reported in this study and those reported elsewhere (30). Also, board certification rates were based on all physicians, regardless of age; historical data on past certification and recertification were not available. To the extent that more recent graduates (younger physicians) are more likely to obtain board certification, the potential negative impact of not having board certification on patient-care quality may be waning.
International medical graduates make up a substantial proportion of the psychiatric workforce, especially among residents in training. As with USMGs, their distribution within the United States is uneven, but their practice patterns are somewhat different from USMGs. In light of the recently observed changing nature of psychiatry internationally as a medical discipline (33), the continued strong representation of IMGs in the specialty is important to consider when envisioning the future of the psychiatric workforce in the U.S. The strength of IMGs’ presence in the specialty also has implications for GME training programs, where increased attention to acculturation and social support could ease IMGs’ transition to American medicine and even improve their mental health (34). Additional longitudinal studies are needed, especially those concentrating on trends in distribution and entry into psychiatric specialties and subspecialties.