International medical graduates (IMGs) play a significant role in the U.S. health care system. The Educational Commission for Foreign Medical Graduates (ECFMG®) defines an IMG as a physician who received his or her basic medical degree or qualification from a medical school located outside the United States and Canada. To be eligible for ECFMG certification, the physician′s medical school and graduation year must be listed in the International Medical Education Directory of the Foundation for Advancement of International Medical Education and Research. U.S. citizens who have completed their medical education in medical schools outside of the United States and Canada are considered IMGs; non-U.S. citizens who have graduated from medical schools in the United States and Canada are not considered IMGs.
Currently, approximately 25% of all residents and 25% of practicing physicians in the United States obtained their medical degrees outside the United States or Canada.1,2 Translated into patient encounters, the over 150,000 internationally trained physicians who currently actively practice medicine in the United States provide a significant amount of the total patient care for U.S. citizens. Historically, many of these physicians were citizens of India, Pakistan, or the Philippines prior to emigrating from their home country. However, between 1983 and 2002, a sizeable group of U.S. citizens chose to attend medical school outside of Canada or the United States. Because of their relatively large numbers and their unique characteristics within the population of internationally trained physicians, these individuals, henceforth referred to as U.S. international medical graduates (USIMGs), are an important cohort when studying issues such as physician supply, international migration, and health care needs for populations in underserved areas.
Unlike the situation in many other countries, medical school graduates in the United States must successfully complete graduate medical education as one of the requirements for obtaining an unrestricted license to practice medicine in the United States. IMGs, regardless of citizenship, must be certified by the ECFMG to be eligible to enter accredited graduate medical education programs in the United States. U.S. residency programs are accredited by the Accreditation Council on Graduate Medical Education (ACGME).
To be certified, IMGs must meet a number of educational and examination requirements. Most of these requirements are also necessary as part of the licensure process for students who graduate from medical schools in Canada or the United States. ECFMG certification assures directors of ACGME-accredited residency and fellowship programs, and the people of the United States, that IMGs have met minimum standards of eligibility required to enter such programs. ECFMG certification does not, however, guarantee that IMGs will be accepted into programs since the number of applicants frequently exceeds the number of available positions. Additionally, ECFMG certification is one of the eligibility requirements to take Step 3 of the three-step United States Medical Licensing Examination (USMLE™). All state medical licensing authorities in the United States call for ECFMG certification, among other requirements, to issue an unrestricted license to practice medicine.
Assessing the readiness of IMGs to enter graduate medical education in the United States has been a primary mission of ECFMG since its founding in 1956.3 In the process of assessing the readiness of IMGs to enter graduate medical education programs, ECFMG gathers important data on the medical education credentials of all certification applicants, including U.S. citizens. Since the majority of USIMGs return to the United States after graduation to practice medicine,1,4,5 there is a particular interest in knowing more about these individuals, including where they go for schooling and how their choice of schools has changed over time. ECFMG, by virtue of its certification role, is uniquely qualified to provide information on the number of U.S. citizens educated in medical schools abroad and on the countries providing educational opportunities for these individuals. During the certification process, ECFMG also gathers detailed demographic data, performance statistics, and details about medical schools located outside of the United States and Canada. Combined, this information can be used to describe the characteristics and better understand the health care role of U.S.-citizen physicians who obtained their undergraduate medical degrees outside of Canada or the United States
In this report, we provide a general descriptive overview of U.S. citizens who obtained their medical degrees outside of the United States and Canada.
The study group included all IMGs certified by ECFMG in the 20 calendar years 1983–2002 (n = 143,926). We used certification year because the date of certification is the point at which an IMG becomes eligible to enter a graduate medical education program in the United States. Over 18,700 of these ECFMG certified physicians, or 13% of the total, were U.S. citizens at the time of their entry into medical school. We did not include in our study the 2,622 U.S. citizens who successfully completed a Fifth Pathway program.* We recommend a separate study about Fifth Pathway physicians when specific data are available.
The majority of U.S. citizens who graduate from medical schools outside of the United States or Canada ultimately return to the United States for graduate medical education and eventual practice. Since ECFMG certification is a requisite for graduate medical education, the number of certificates issued to U.S. citizens is a reasonable proxy, albeit somewhat lagged, of the total number of U.S. students who graduated from medical school outside of Canada or the United States. Therefore, for the USIMG cohort, certification statistics provide a reasonable gauge of the flow of U.S. citizens who study medicine abroad. Combining these data with demographic information, including a medical school’s location, provides a historical perspective of where IMGs trained.
We calculated descriptive statistics, including historical certification rates for the non-USIMG and USIMG cohorts. For USIMGs, we calculated the number of certificates issued by geographical region, country of medical school, and medical school.
Figure 1 shows summary data on the 143,926 IMGs certified from 1983–2002, stratified by citizenship at the time of entry into medical school. Clearly, non-U.S. citizens (non-USIMGs) make up the greatest proportion of ECFMG certificate holders. A review of Table 1 shows that from 1992–1998, the number of foreign nationals who received certificates was significantly higher than it was at any other time during the period studied. However, starting in 1999, USIMGs, as a percentage of the total certificant population, became a much larger cohort, representing approximately 25% of all certificates issued in this four-year period.
Although USIMGs could potentially attend one of over 1,600 medical schools located outside of Canada or the United States, they tend to concentrate in certain geographic regions. Certificates issued to USIMGs, by geographic region and time (five-year blocks), are shown in Table 2.
Medical schools graduating USIMGs were located in all geographic regions: Africa, Asia, Central America and the Caribbean, Europe, the Middle East, Oceania and the Pacific Islands, and South America. For this study, we considered Mexico to be part of Central America and the Caribbean. From 1983–2002, 164 countries had medical schools listed in either the International Medical Education Directory6 or the World Directory of Medical Schools.7 USIMGs graduated from medical schools in 106 (64.6%) of these countries. From a regional perspective, USIMGs attended medical schools in 14 of 41 countries in Africa, 16 of 30 countries in Asia, four of eight countries in Oceania and the Pacific Islands, ten of 13 countries in the Middle East, 31 of 39 countries in Europe, 20 of 22 countries in Central America and the Caribbean, and 11 of 12 countries in South America.
As Table 2 shows, three geographic regions, Africa, Oceania and the Pacific Islands, and South America, consistently graduated relatively few U.S. citizens. Combined over the 20-year period, only 3.9% (no. = 733) of all USIMGs graduated from medical schools in these regions. In contrast, from 1983–2002, over 12,000 USIMGs (64.9%) graduated from medical schools in Central America and the Caribbean. However, even within Central America and the Caribbean, there was some longitudinal variation. From 1983–1987, just over 64% (no. = 3,655) of the ECFMG certificates issued to U.S. citizens were based on graduation from medical schools in this region. From 1998–2002, almost 73% (no. = 4,837) of the USIMG cohort attended medical schools in this region.
Interestingly, there has been a gradual decline in the number of U.S. citizens graduating from medical schools in Europe. Between 1983 and 1987, over 22% of the USIMGs graduated from medical schools in Europe. This number decreased to 7% in 1998–2002. In contrast, Asia has seen a gradual increase in USIMG certificate holders, going from 391 (6.9%) in 1983–1987 to 668 (10.1%) in 1998–2002.
From 1983–2002, 1,000 or more USIMGs who received certificates graduated from schools in each of the following countries: Grenada, Dominica, the Dominican Republic, the Netherlands Antilles, Israel, Montserrat, Mexico, and the Philippines. Summary data for USIMG certificate recipients by top-ten countries of medical school and five-year blocks are shown in Table 3. Within Central America and the Caribbean, the number of U.S. citizens graduating from medical schools in Grenada increased substantially, rising from 652 (11.4%) in 1983–1987 to 1,521 (22.9%) in 1998–2002. Similar increases, most notably in 1998–2002, were evidenced for Dominica and the Netherlands Antilles. In contrast, the number of certificates issued to USIMGs who graduated from medical schools in the Dominican Republic and Mexico decreased. From 1998–2002, only 239 certificates (3.6% of the total) were issued to U.S. citizens who graduated from medical schools in the Dominican Republic. Likewise, for 1998–2002, only 89 certificates were issued to U.S. citizens who graduated from medical schools in Mexico. It is worth noting that the total number of certificates issued to U.S. citizens who graduated from medical schools in Israel exceeds the totals for several countries, such as Mexico and Montserrat, in the otherwise dominant region of Central America and the Caribbean. U.S. citizens also graduated from medical schools in Italy and India, with more than 1,200 receiving certificates in 1983–2002. In 1998–2002, the number of certificates issued to USIMGs attending medical schools in Italy was negligible, representing less than 1% of the total. In contrast, the number of U.S. citizens attending medical schools in India rose, representing 4.4% of the total for 1998–2002.
The 25 medical schools with the highest total number of U.S. citizens who received certificates from 1983–2002 are shown in Table 4. Again, Central America and the Caribbean dominate with 16 of the top 25 medical schools. Of the 18,762 USIMGs who received certificates, 7,867 (42%) attended one of three medical schools: St. George’s University, Ross University, or American University of the Caribbean.
IMGs play an important role in the U.S. health care system. Although they represent numerous nationalities, a significant number of IMGs are U.S. citizens when they enter medical school overseas. U.S. citizens received over 13% of the ECFMG certificates issued between 1983 and 2002. More importantly, from 1992–2002 the number of certificates issued to USIMGs grew appreciably, representing over 25% of the total number of certificates granted. Since most of these individuals return to the United States for graduate medical education, and eventual practice, it is important to know more about them, including where they go to school and how their choice of schools has changed over time.
As part of an extensive study of ECFMG certification from 1969–1982, ECFMG reported that medical schools in Central America and the Caribbean, specifically Mexico and the Dominican Republic, provided medical education to over 50% of all U.S. citizens who attended overseas medical schools.8 Analysis of the data for our study shows that, between 1983 and 2002, medical schools in Central America and the Caribbean, most notably in Grenada and Dominica, provided medical education to almost 60% of all U.S. citizens who studied abroad.
Our results also show that even though U.S. citizens elected to pursue their medical education in 106 different countries, more than half of all U.S. citizens certified from 1983–2002 graduated from medical schools in four of the 20 Caribbean countries: Grenada, Dominica, the Dominican Republic, and the Netherlands Antilles. Interestingly, the overall number of certificates issued to USIMGs who attended Caribbean schools increased dramatically, especially for schools located in Grenada and Dominica. While this may be a function of students’ choice, it likely reflects an increase in the medical education, training, and practice opportunities for U.S. citizens who do not attend medical schools in the United States or Canada. A notable exception is the Dominican Republic, where the decrease in the number of U.S. students can likely be attributed to the closure of a number of schools. The government of the Dominican Republic closed two medical schools, the Universidad C.E.T.E.C. and the Universidad C.I.F.A.S., in 1984 and three additional schools in the 1990s.
Although from 1998–2002 over 70% of the USIMGs graduated from schools in Central America and the Caribbean, from 1983–2002, U.S. citizens who received ECFMG certificates attended medical schools in 86 countries located outside this region. For medical schools in some countries (e.g., Israel, India) the number of U.S. citizens being educated increased over time; for others (e.g., Italy), the numbers decreased. Given the historically large numbers of IMGs coming from medical schools in India to train and practice in the United States, the increase in the number of U.S. students going to India may reflect second-generation immigration patterns. The U.S. citizens going to India for medical training may be the children of Indian immigrants. This hypothesis could be tested by a more in-depth analysis of the demographic characteristics, including ethnic identity, of all USIMGs. In Israel, the Sackler School of Medicine enrolls approximately 300 U.S. students in a four-year MD program with a curriculum patterned after that of U.S. medical schools.9 This likely explains recent increases in U.S. students graduating from medical schools there.
Why U.S. citizens choose to enroll in medical schools outside the United States and Canada certainly requires further study. However, given that Central America and the Caribbean ranked first in all three categories of the data analysis (certificates issued by region, country, and medical schools), a closer inspection of some of the characteristics of the medical schools in this region is worthwhile and may also shed some light on why students choose to study there. First, it is not unreasonable to think that travel distance is a factor that is considered when selecting a school. The Caribbean is geographically very close to the United States. Second, for the vast majority of the countries in this region, English is the language of instruction in the medical schools. U.S. citizens attending these schools would not have to learn another language. Finally, many of the schools operating in the Caribbean model their programs after U.S. medical schools and have a four-year curriculum. Also, like U.S. medical schools, the title of the medical degree awarded is Doctor of Medicine. These characteristics and similarities may explain why U.S. citizens are attracted in such large numbers to Central America and the Caribbean to complete their medical education.
While this descriptive investigation shows that there is much diversity in the educational pursuits of USIMGs, both geographically and over time, additional studies are certainly warranted. First, a more detailed analysis of the demographic characteristics of USIMGs is necessary. This information, in combination with performance data on certification and licensure examinations, would be useful to better understand factors associated with medical school choice. Second, although detailed information, including curricula, is available for medical schools in the United States and Canada,10 relatively little is known about many international programs. Efforts are being made to obtain this information because additional medical school data (e.g., annual enrollment, faculty, curricula) would be helpful to better understand the educational process for USIMGs. For purposes of research related to physician work force issues, it is important to know that a large number of U.S. citizens study medicine abroad and return to the United States for graduate medical education and medical practice. It would also be valuable to analyze the medical specialties and location of practice of USIMGs because they play a significant role in the U.S. health care system.
1Brotherton SE, Rockey PH, Etzel SI. US graduate medical education, 2002–2003. JAMA. 2003;290:1197–202.
2Whelan GP, Gary NE, Kostis J, Boulet JR, Hallock JA. The changing pool of international medical graduates seeking certification training in US graduate medical education programs. JAMA. 2002;288:1079–84.
3Educational Commission for Foreign Medical Graduates. A History: 1956–2000. Philadelphia: ECFMG, 2000.
4Salsberg E, Nolan J. The posttraining plans of international medical graduates and US medical graduates in New York State. JAMA. 2000;283:1749–50.
5Mullan F, Politzer RM, Davis CH. Medical migration and the physician workforce. International medical graduates and American medicine. JAMA. 1995;273:1521–7.
6Foundation for Advancement of Medical Education and Research (FAIMER). International Medical Education Directory (IMED). Philadelphia: FAIMER, 2004.
7World Health Organization (WHO). World Directory of Medical Schools. Geneva, Switzerland, WHO, 1997.
8A study of candidates for ECFMG certification 1969-1982. Dublin T, Oesterling SF (eds). Philadelphia: Educational Commission for Foreign Medical Graduates, 1987.
9Tel Aviv University, Faculty of Medicine. About Sackler. 2004. Sackler Faculty of Medicine. Tel Aviv, Israel.
10Association of American Medical Colleges. Curriculum Directory, 2004 〈http://services.aamc.org/currdir/start.cfm
〉. Accessed 26 January 2005. Washington, DC, AAMC.
*The American Medical Association defines the Fifth Pathway program as follows: “A pathway for entrance to approved programs of graduate medical education, other than those existing under previous policies, became available as of July 1, 1971, for students who have fulfilled the following conditions:
- Have completed, in an accredited American college or university, undergraduate premedical work of the quality acceptable for matriculation in an accredited United States medical school, evaluated by measures such as college grade point average and scores on the Medical College Admission Test (MCAT).
- Have studied medicine at a medical school located outside the United States, Puerto Rico, and Canada that is listed in the World Directory of Medical Schools, published by the World Health Organization. [Note: This publication has been replaced by the International Medical Education Directory of the Foundation for Advancement of International Medical Education and Research; see 〈www.ecfmg.org〉 for more information.]
- Have completed all of the formal requirements of the foreign medical school except internship and/or social service. Those who have completed all of the requirements, including internship and/or social service, and are, consequently, eligible to apply for ECFMG certification, are not eligible for the Fifth Pathway.”
Source: American Medical Association. GMED Companion: An Insider’s Guide to Selecting a Residency Program, 2004/2005. Chicago: AMA, 2004:334–5. This source also includes additional information on the Fifth Pathway program.