The increasing migration of international medical graduates (IMGs) to affluent countries has been addressed in the literature by medical educationalists and policymakers.1–5 The fact that 60% of IMGs in the United States and 75% in the United Kingdom originate from poor countries6 has fueled an economic and ethical debate about the problems that less developed nations face related to physician needs; often the situation is viewed from a perspective of controlling, compensating for, or managing migration.7–9 Because migratory flows may increase in the future,10,11 suggestions have been made to counter the effects on the global health care labor market through policy initiatives.12,13 Migration of physicians from Africa3,14 has been addressed in the literature, but relatively little has been published about such migration from Pakistan, which is the third leading source of IMGs in affluent countries.6
Migration is but one of the forces affecting adequacy of physician supply. The need or demand for physicians is an additional factor in rapidly evolving countries such as Pakistan. In less developed countries, population growth, evaluated health needs of the population, and expectations are all dynamic factors that affect this need or demand but have not received adequate attention.
This article presents the findings of a national case study on Pakistan. In this study, we reviewed and analyzed physician production and considered the causes of migration and other losses from the physician workforce, and we considered the impact given the projected needs and demands of the nation. The limited human resource pool for health care is global. Current and anticipated physician shortages are also therefore global issues. The implications are discussed in relation to policy options.
Information on physician production and losses, educational capacity, health indicators, and effectiveness of physician substitutes was obtained from multiple sources, including the Pakistan Medical and Dental Council (PMDC) and College of Physicians and Surgeons (CPSP); publications indexed in PubMed; migration surveys; news items; annual publications of the Mehboob-ul Haq Human Development Center; the Web sites of the United States Agency for International Development, the World Health Organization, and the Government of Pakistan; and working papers for conferences. Where data on annual physician migration were not available, the literature was searched for information on which reasonable estimates could be based. We deliberately erred on the side of underestimating physician loss, to avoid overestimating physician shortages.
Physician availability in January 2006 was calculated from the listing of registered physicians, on the basis of Khan’s15 assumptions. No adjustment was made for practicing physicians who had not registered with the PMDC, an unknown number. Projections of physician supply were based on the current average medical school output, adjusted to take into account increased output by maturing medical colleges and increasing success of IMGs on certifying examinations. In projecting physician losses, we assumed that at least a third of the graduates would be women, of whom 10% would not practice on graduation; that migration would continue at current rates; that decertification of physicians would not be significant by 2020; and that annual loss of physicians due to causes other than migration and women not practicing would remain a constant figure at 0.5% of physicians on the register in December 2005. All losses were deliberately underestimated to avoid overestimating shortages. Migration estimates were based on information from papers by E. Salsberg (personal communication, February 2006), Mullan,6 and H. Owen (personal communication, March 2006), and best estimates of numbers of physicians in Arabic-speaking nations were obtained from Web sites and news items. Emigration to Africa and Europe was ignored as comparatively insignificant.
Future demands for physicians were calculated for four scenarios: (1) current physician to patient ratios, (2) 1 physician per 1,000 population (P/TP), (3) 1 P/TP until 2010 and 1.5 P/TP thereafter, and (4) the current ratio in the United States (2.93 P/TP).6 Physician shortages were derived by subtracting the number of physicians available from the expected demand as per each scenario.
Population estimates were based on medium growth rate projections of 2.1% (in the period 2006–2010), 1.7% (2011–2015), and 1.4% (2016–2020) (M. S. Karim, personal communication, February 2006). Current thinking on reasons for migration was obtained by interviewing or administering questionnaires to limited convenience samples and through personal communications.16 Information that one of us (JT) gained by interviewing the faculty of the refugee-physician–based Afghan University in Peshawar, Pakistan, in 2000 provided insights on Afghan refugee doctors in Pakistan.
Most physicians practicing in Pakistan are graduates of Pakistani medical schools or reentering migrant citizens. Combined, the 39 medical colleges in Pakistan produced an average of 5,038 physicians per year (range, 4,422 and 5,520) from 2000 to 2004.17 In 2000, 50 physicians migrated into Pakistan with the MBBS or an equivalent degree and no postgraduate qualification; this number reached 395 in 2004.17 These international medical graduates (IMGs) were trained in China, the Philippines, the Caribbean, and former Russian Republics.15 The PMDC has recognized five Chinese medical schools.
Combined, the number of graduates from Pakistan’s medical school and number of reentering IMGs can be reasonably expected to reach 6,800 per year as schools mature. Afghan-refugee doctors, who have limited access to modern knowledge and interventions, do not contribute significantly to the pool of practicing physicians. Since 1960, Pakistani-trained physicians have assumed teaching posts in medical colleges after their postgraduate medical education abroad. At the Aga Khan University Medical College (AKUMC), 63% of faculty in professorial ranks have foreign qualifications. The returning IMGs, although small in number, play critical roles in boosting research and education in key institutions in the country.
The PMDC is responsible for the quality of practicing physicians and requires that all physicians register themselves with the council. As of February 2006, Pakistan had 18,029 registered specialists.18 Many returning specialists may not have registered themselves as the CPSP alone has certified 6,418 members (MCPS) and 7,329 fellows (FCPS) since its inception in 1962.
The number of medical colleges doubled from 1997 to 2005 despite a dearth of teachers, facilities, and teacher training institutions for medical colleges. The number of students entering medical colleges increased from 400 (in four medical colleges) in the 1960s to 4,239 (of whom 3,552 qualified) in 16 colleges in 1981.19,20 Since the first private medical college, AKUMC, was established in 1983, 18 other private sector colleges have been established. Class size, at times exceeding 300 in some public institutions, impedes efforts to improve learning and assessment. There is an appropriate move to establish uniformity across institutions, but the “new” curriculum specifies a number of hours for lectures and insists on having physicians teach basic sciences. A handful of institutions, including AKUMC, are moving rapidly to newer ways of improving learning. Research capacity is severely limited at medical colleges in Pakistan.
Fortunately, access to medical education is not limited by poverty or educational or geographical disadvantage. Although the total cost of providing the five years of medical education is U.S. $100,000,15 students are charged tuition of U.S. $833 at public colleges and U.S. $10,000–35,000 at private colleges. In the Punjab province, 1.2% of seats are reserved for students from underdeveloped districts (not all who so enter return to rural or tribal areas); 4% of students, including seven Afghans and two Palestinians, receive financial assistance. The low fees are partly offset by fees from self-financing foreign students, who are charged U.S. $20,800 at public colleges and up to U.S. $97,000 at private colleges.
Causes of physician loss include emigration and cessation (or diminution) of practice for a variety of reasons, including migration.
We estimate that 1,700 physicians per year are lost from the pool of practicing physicians. This figure is based on review of data from a variety of sources. The Bureau of Emigration and Overseas Employment (Khaskheli, personal communication, March 2006) estimates that annually about 1,000 to 1,500 physicians leave the country, of whom 10–15% return, for a net migration of 900 to 1,275 physicians. We can account for 1,150 emigrating physicians per year. Of this number, we estimate that 418 (the number starting graduate medical education in 2004 according to the Association of American Medical Colleges’ GMEtrack database; E. Salsberg, personal communication, February 2006) to 500 emigrate to the United States. As 4.4%6 of IMGs in the United Kingdom are from Pakistan, and there were 4,185, 4,325, and 5,904 IMGs (H. Owen, personal communication, March 2006) in the United Kingdom (2000–2002), we have calculated expected annual emigration to the United Kingdom as 260 (4.4% of 5,904). Using ratios of the numbers of Pakistani physicians emigrating to the country each year to the total pool of Pakistani physicians in those countries where both figures were available, we extrapolated that 16 emigrate to Canada annually and 6 to Australia. However, we know that 1,061 and 367 candidates have attempted the Medical College of Canada MCCEE and MCCQE qualifying Part 2 examinations, respectively, indicating a possibly higher rate of migration to Canada (D. Dauphinee, personal communication, March 2006). The flows to Arabic-speaking nations (ASNs) are probably about 350 annually; 300 a year have emigrated to Saudi Arabia alone.21 Although a large number of fellows of the CPSP are in ASNs, in the absence of a tracking system, the only information available officially is about those who maintain e-mail contact with the college. So recorded, there are only 31 fellows in the ASN and 38 in other countries (I. Waheed, personal communication, March 2006).
Although as reported in 2005 by Mullan, 12,813 physicians from Pakistan were in the United States, the United Kingdom, Australia, and Canada,6 we cannot determine the total number in ASNs. Khan15 assumed that there were 25,000 graduates outside Pakistan, implying that there were 12,200 emigrants in countries other than the above four, chiefly in ASNs. At least 3,000 are accounted for: 2,000 in Saudi Arabia21 and 1,000 in the Persian Gulf states. We can assume that this outflow will continue. An additional 20,000 Pakistani doctors are required in Saudi Arabia, and Pakistan has been asked to send “as many physicians as it can spare.”22 The estimate that there are 1,000 physicians from Pakistan in the Gulf is based on the report that of the 20,000 IMGs in the Persian Gulf, 25% are South Asian (G. Wajid, personal communication, February 2006), of whom 80% probably are Indian.
On adding 3,000 Pakistani physicians working abroad in ASNs to Mullan’s6 estimate of 12,813 (in four advanced countries), the emigration ratio (the number of physicians working abroad as a percent of total pool of physicians) becomes as high as 13.5% (if all 101,090 registered physicians are practicing) or 17.6% (if we use the more likely estimate of 73,890 practicing physicians).
Gender-related factors, of increasing importance as the number of women in medical school has increased, are a major reason that physicians stop or slow down their practice. Because of marriage, childbearing, and family, a sizeable number of women graduates are not practicing; anecdotal estimates of the percentage range from 5% to 50%. On average, 50% of those admitted to medical school are women; however, despite a higher pass rate for women than for men, as of December 2005 only 38% of registered physicians are women.18 We assumed that at least a third (i.e., 2,000) of the 6,000 graduating (non-IMG) physicians will be women and that 10% will not be practicing, and estimated that at least 200 women will have to be considered as an annual continuing loss from practice. Physicians also are lost to practice because they are practicing part time, teaching in basic science departments, working in government offices, not practicing for other reasons, or retired or dead. Altogether, these causes result in loss of 370 physicians from practice.
In this section, we start with the number of physicians available in Pakistan in December 2005 and make projections about future supply of physicians based on a series of assumptions. We then state the numbers of physicians needed or demanded in various scenarios. The difference between projected supply and demand is the projected surplus or shortfall.
On the basis of Khan’s15 assumptions, we estimated that 73,890 physicians were practicing in Pakistan in December 2005, for a P/TP ratio of 0.473. This number was calculated by subtracting from the 101,090 registered physicians the estimated 27,200 physicians who had emigrated or who were dead, retired, or not practicing.
The net annual physician gain is 5,100, as 6,800 physicians are produced and 1,702 lost. Although the number of medical students in Pakistan has grown over the past years, it is unlikely that the increase in graduates from accredited medical colleges will continue. We therefore assume a static rate of production of about 6,000 MBBS graduates annually, with a gradual accumulation of physicians over time. The number of IMGs (mostly Pakistani IMGs) coming to Pakistan, is projected to be static at 800 per year. We therefore predict that Pakistan will have approximately 150,500 physicians in 2020.
Whereas the supply of physicians in Pakistan is somewhat predictable, the level of demand is less certain. Although needs for physicians in the country might be set arbitrarily, the demand or perceived need reflects a complex series of factors:
- Pakistan’s population is growing rapidly, and at medium growth rates, it will reach 173.6 million in 2010, 191.4 million in 2015, and 208.4 million in 2020.
- Life expectancy had increased to 62 years by 2004, from 45 years in the 1950s.
- A total of 6.7 % of the population was older than 60 years (in 2005).
- The economy is expanding rapidly.
The incidences of accidents (unintentional injuries), cancer, and degenerative disease are increasing. In addition, there is easier access to technology and therefore a greater recognition of diseases. Education and access to television have led to demands for advanced health care that are visible every day: Patients surviving stage four testicular cancer ask why blood for transfusion was not screened for hepatitis C; patients with end-stage renal disease (of which incidence is increasing23) demand dialysis; the epidemics of heart disease and unintentional injuries lead to opening of tertiary centers.
These issues regarding need or demand led us to conceptualize a series of scenarios to better understand possible future trends. These scenarios give us a view of future trends as a set of a range of requirements of physicians for different levels of demands. We have calculated the shortfall, or gap, between projected supply and demand for the four scenarios in List 1.
The first scenario assumes that the demand will be remain constant at the current ratio of physicians to population (0.473 P/TP). Under this scenario, the accumulation of physicians in Pakistan will provide a modest surplus of physicians by 2010. Despite the rapid growth of population, the accumulation of physicians over the next decade will increase the current very low ratio of physicians to population. This scenario obviously does not take into account the unsatisfactory state of the population’s health, the increasing burden of disease in an aging population, and a population with increasing expectations.
The second scenario uses the ratio of 1 P/TP. In this scenario, Pakistan will fall grossly short of its needs for physicians in 2010 and 2020. Indeed, according to this standard, Pakistan already has a serious dearth of physicians. The shortfall will be even greater than in Scenario 2 if needs increase as in Scenario 3, which reflects the genuine possibility that physician demand from 2010 onward will rise to at least 1.5 P/TP. The fourth scenario assumes high needs and high expectations: We have applied the P/TP ratio of the United States (2.93/1,000) to Pakistan. The gap or shortfall is even greater.
Physician supply problems are compounded by the urbanization-induced demoralization in villages, which makes them unattractive for physicians and villagers alike. Pakistan is rapidly becoming an urban nation, with as much as half of the population (officially 40%) living in cities and towns. Although Pakistan may not soon see levels of care resembling those in the United States, the escalating health care demands fueled by education and urban living make scenario three a useful benchmark.
Views of young physicians on migration
The reasons that recent graduates state for emigrating suggest that key factors are educational quality, research opportunities, lifestyle, opportunities for professional and career progression, and international recognition. Migrants are highly motivated, well-educated high performers, deeply interested in improving health parameters and systems and capable of it because of academic strengths and well-developed peer, social, and political linkages. Until the events of 2001, almost every graduate of AKUMC, which selects 2–4% of students who apply for admission, desired to emigrate. Frossard16 notes the poor self-image of the native country and admiration of things foreign as factors in migration. Other factors include peer pressure and role modeling, as two thirds of the faculty at AKUMC have been educated abroad. Owen’s survey of 1,021 non-European IMGs in the United Kingdom shows that the reputation of British training, similarities in education (from commonalities in language of instruction, curricular content, and nature of clinical clerkships), promotions and prestige obtained on attaining a membership or fellowship degree of the Royal Colleges of the United Kingdom, and experience of a different sociocultural environment are some of the factors in the decision (H. Owen, personal communication, March 2006).
The decision to migrate appears to be individual and not based on family decisions. Repatriation of foreign exchange by Pakistani physicians is insignificant in comparison with the large flows (billions of U.S. dollars) from low-skill émigrés.
Four insights emerge from this overview: (1) high-quality data on physician migration are scarce, (2) physician loss from migration is numerically insignificant yet has a critical impact, (3) the physician workforce shortage (PWS) is gargantuan, and (4) it is impossible to educate enough physicians worldwide, and therefore a framework is needed for educating other health workers and designing educational programs to deliver the required skills.
Our predictions are being played out on a geographically, politically, and economically highly charged and progressing territory. Pakistan is already emerging from poverty, and this transition in itself will improve education and lower death rates.24
Problems with data
Across the world, data on migration are fragmentary and incomplete.25,26 New methods of tracking physicians are required, as suggested by Eckhert.27 Data should be freely available and easily retrievable.
We have displayed our collected data without the sophisticated econometric analysis used by Brown and Connell.29 Some of our assumptions could well be challenged: Physicians on the register in 2005 might not include a significant proportion of the 25,000 we have deleted as émigrés, on Khan’s assumptions; the number of physicians not practicing might be underestimated; 50% of the 39,000 women doctors might not be practicing. Migration is known to be unpredictable.29 The number of émigrés might easily shift suddenly: Saudi Arabia’s call for 20,000 physicians might be acceded to; the proportion emigrating to the United States might increase to ease the predicted PWS of 100,000.30 Economic revival may reverse migration by improving postgraduate programs and retaining physicians. Our calculations of the PWS are based on medium population growth projections, and the population could well be larger if fertility is not controlled. Economic union with India might provide a new stream of IMGs.
Small but critical losses: the migration paradox
Migratory exodus as an escape from targeted sectarian killings and kidnappings has given way to a more rational flow in two streams: to advanced technology-inventing nations for postgraduate education and citizenship, and to ASNs for service and lifestyle.
Migration significantly depletes poor countries,3,6 but for Pakistan, with an anticipated shortfall in the year 2020 of 58,000 to 451,000 physicians, depending upon population demands, a total ban on migration, even stemming the loss of 1,500 emigrants per year, would still leave a deficit of 35,400–428,600 physicians in 2020. Migration causes only a small proportion of the anticipated shortage. The main cause is the inability to educate an adequate number of physicians for a socially emancipated, demanding population with a complex burden of infectious and degenerative disease, cancer, and injuries.
However, even though numerically insignificant, physician migration is a serious problem in Pakistan for the following reasons. First, émigrés are academically powerful individuals with existing social and political skills and acquired international polish. Thus, when they leave, Pakistan loses people who can reform, organize, stabilize, and expand the health system and who can drive change in academics and research. (Some say, though, that physical presence is not necessary for the diaspora to contribute to progress in the home country. 31) Second, Pakistan has a limited stock of physicians, many of whom are inadequately educated, and there is a threat of an ever-increasing draw from that stock (so that the emigration factor rises beyond the current 17%) as other countries improve their health care. As example, the United States had an IMG force of 10% in the 1960s, 18% in 1970, and 25% in 2002.32 Granted, some of these IMGs are U.S. citizens educated elsewhere—and here lies a benefit for both parties: The cost of education is low for the citizen from the affluent country; and the training country benefits from the higher fees charged.
Nevertheless, migration should not be controlled: health care requires a common knowledge base, many undergraduate programs in Pakistan are of high quality and quite similar to those in technologically developing countries, the Internet has opened access to modern information, and the top 10% of Pakistani physicians are high achievers (some with United States Medical Licensing Examination scores in the 99th percentile). In these circumstances, it seems inappropriate to restrict bright and able individuals.
Affluent countries have to address the demands of their populations, and except as in the case of Japan, their patients cannot be moved to the service point across the globe.33 Government-to-government agreements on recruitment have been evaded (albeit in the recruitment of nurses).
Additionally, migration is a very powerful inbuilt instinct. It stems from an urge to improve lifestyle, education, and international visibility. Migration is a move for betterment that mirrors our earlier migrations: It is a freedom and a right, and it benefits human society—witness the Nobel Prizes that pour out of the Swiss Federal Institute of Technology,34 with its 35% migrant faculty. In today’s fractured world, migrants have the potential for changing one-way acculturation into a mutually beneficial understanding of societies and cultures. Migration for postgraduate study (or for an elective opportunity or twinned joint program) opens up the mind and provides opportunities for people to learn from each other newer perspectives in their professional field.35
Implications for health care human resource management: the global challenge
There is clear evidence that the current ratio of 0.473 P/TP is associated with poor National Health parameters.36 Pakistan has a double burden of disease: a growing epidemic of cancer, cardiovascular diseases, and disabilities, along with a continuing burden of infectious diseases and nutritional deficiencies. More than a third of adults in the country are hypertensive; of these, very few are aware of their condition. In the mainly rural province Baluchistan, 800 mothers die for every 100,000 live births, a figure far worse than the national average of 340 per 100,000.37 Pakistan ranks 129th of 174 nations on the Human Development Index,38 a measure of “the achievement of a country in terms of health, longevity, education and standards of life”39 However, Pakistan has improved access to health services,38 reduced infant mortality from 139 per 1,000 live births in 1961 to 84 in 2004, curbed population growth, increased contraceptive use (from 5.5% in 1968 to 23.9% in 1998), decreased the total fertility rate (from 6.4% in 199 to 4.8% in 2000), increased life expectancy to 62 years, and increased the national immunization rate to more than 80%.
Many improvements have been wrought by trained Lady Health Workers (LHWs) and others who are not physicians or nurses. Initiated in 1994, basic refresher and continuous training programs educate resident LHWs in common community-based maternal and child health issues. LHWs have knowledge of basic medicines, supplies, and preventive care, and their work is supervised by more advanced workers. Tuberculosis, leprosy, trachoma, and malaria control program workers and immunization workers also have contributed. In addition to 74,000 physicians, there are some 32,619 nurses (Pakistan Nursing Council database, 1996–2005). A total of 675 midwives, Lady Health Visitors (LHVs, a different category from LHWs), and Dais (traditional birth attendants) registered with the nursing council in 2005; here too, not all who qualify register. In 2004 alone, 2,270 nurses, 1,384 midwives, and 1,013 LHVs completed training. The 70,000 LHWs who had been trained as of 2003–2004 served 63 million people. This workforce is being increased to 100,000 workers. The value of LHWs can be judged from their involvement after the October 2005 earthquake in Northern Pakistan. A total of 3,311 LHWs assisted in relief operations and were taught grief counseling.40
These success stories have their limitations. Although maternal mortality has declined, postpartum hemorrhage or obstructed labor still kills. But evidence is accumulating that the program of basic, refresher, and continuous training in areas of specific need (for example, grief counseling, or indications for use of antibiotics in respiratory infections)41 can reduce mortality in infant and childhood pneumonias and that decentralization, as well as delegation of task responsibility to nonphysicians, works.42–44 It is worth pursuing the education of these trained workers—the literate, motivated youth of the local region who can “speak” the culture and gain compliance. They are married to the region and unlikely to leave.
The total health expenditure of Pakistan (2004) was 2.4% of the gross domestic product.45 Thus, health expenditures clearly are inadequate to the needs of the country.
Options to meet overwhelming physician shortages include borrowing IMGs, producing more physicians, instituting efficiencies in postgraduate education, upgrading the skills of (rural) health workers, and establishing mechanisms to reduce burden on physicians. Karle46 stated that there were 2000 medical schools in the world (in 2003). Pakistan has 1.95% of the world’s medical schools. It has about 2.55% of the world population, and 39 schools, for a ratio of four million people per medical school, whereas Nigeria has 8.3 million per medical school.3 Pakistan should not expand its schools without first addressing standards. Norcini and Mazmanian2 emphasize the value of raising educational standards and Harden47 the value of utilizing concepts of virtual medical schools to improve education.
Across the world, anticipated shortages are dealt with by borrowing IMGs. Czechs borrow from East Europe to replace losses to Germany; the French lose their family physicians to the United Kingdom48; the United Kingdom loses physicians to Canada, and Canada to the United States.6 Pakistan—which pays poorly and remains unattractive and unsafe—cannot freely recruit foreign IMGs. The quality of Pakistani IMGs taught in non–English-speaking countries needs to be evaluated.
Although Anand and Barnighausen49 correlate health outcomes to physician density, we believe that achieving outcomes requires specific competences, many of which can be imparted by training any of a wide range of individuals, not necessarily physicians. For example, control of maternal mortality is related to the availability of a skilled birth attendant, who need not be a physician. Physicians are reluctant to move to rural areas, for valid reasons—thwarted growth of self and children; insecurity from threats, robberies, and kidnappings; and interference from community leaders. In contrast, the rural health worker is married to the land. Therefore, investment in development of skills and decision-making powers of nonphysician health workers will have great value.
Because Pakistan is also battling a heavy disease burden, there is a pressing need for specialists and therefore for improvement in residency programs. The rapidly developing world could assist in buttressing these programs, and as the need for specialists is growing across the world, we need to explore ways of rapidly developing specialists. What educational technologies and skills laboratories will help? Should specialization begin after a core three-year undergraduate education? Should twinning of institutions, with free movement of residents between countries, be encouraged? Will twinning mandate an initial year of core courses that build general universal competences in cognito-psychomotor skills, ethics, culture, behavior, communication, and learning skills, preparing the individual for further development in any country with a high standard? Malaysia and Ireland have successfully initiated a twinned MBBS program.50
Why do so many brilliant young people leave their home country? The idea of serving the population that supported their early development is not a strong motivator. Is this a lack of altruism, or evidence of intolerable home conditions, or a lack of nationalistic spirit evolving from globalized roles? Altruism cannot be “delivered” or inculcated, packaged neatly in a curricular box. Universities must ensure that during premedical general education students develop an ability to work in communities and take what is international and convert it into an application for the local people. Universities need to capture their students’ imaginations and convert universal theory into acceptable practice. The consequent benefit to others will gain for the students gratitude and a prestige that generates self-esteem, confidence, and self-efficacy—the ability to do more, and better, beyond one’s current capacity; it will also provide the assurance needed to move away from an admiration of and dependence on what is foreign. One could include service in health care teams as a prerequisite for eligibility to apply for admission to medical college. Viewpoints and insights so gained might alter future physicians’ perceptions of need to serve rural areas and poorer communities and might generate new ideas for health care from the grassroots level.
Looming PWSs cannot be easily met by increasing the number of medical schools or by enacting legislation on migration. The shortage of physicians is a problem that Pakistan shares with the developed world. The need for physicians is rapidly outstripping production globally and is forcing us to rethink medical education in both resource-rich and resource-poor countries. At current levels of production, Pakistan and other developing countries cannot meet their current needs and will not be able to meet the needs of the developed world.
A global strategy includes reconsideration of the need for categories of health professionals other than physicians. Demand for physicians can escalate without limits unless health needs are stratified, education of middle-level workers is enhanced, and physicians’ roles and the degree to which they can be delegated are redefined. Physicians may need to be retrained to increasingly work in teams with other professionals and to supervise these workers, who may assume increasingly independent roles.
Residency programs could be improved through better assessment of inductees, a basic core program common across nations, and free movement of residents between programs, especially those in different countries, for part of their education. Migration causes critical impact even when numerically insignificant, but it is inevitable. It is unreasonable to expect the diaspora in high-pressure advanced environments to leave their demanding environments to work in Pakistan, unless work environments in their adopted homes provide time for engagement in their parent country.
For effective planning of health services, reliable databases on sources of adequately educated physicians and their movements are required. Equally important is the better assessment of candidates for residency programs and of their progression through the programs. Twinning and research on how best to provide for a nation’s health also are needed. Global professional bodies, foundations, and especially the Educational Commission for Foreign Medical Graduates are aptly poised and have the capacity and wherewithal to address two foundational tasks: the recording of accurate data on schools and migration, and the enhancement of standards through refined assessment.
Above all, an engagement of society is essential, as technologies radically disruptive51 of conventional approaches and current ethics have arrived and will consume physicians’ time. Universities need to develop graduates’ sense of altruism and the need to serve resource-poor settings.
We hope that the current case study of migration, medical education, and health care in Pakistan has yielded insights that will assist global efforts to address physician workforce shortages and allow new perspectives to come to bear on the problem.
This article has been made possible because of thoughtful contributions of a large number of Aga Khan University faculty including Camer Vellani, FRCP, Distinguished University Professor; Murad Moosa Khan, MRCPsych, professor and chair, Department of Psychiatry; Mehtab S. Karim, PhD, professor and head of Section of Reproductive Health, Community Health Sciences; and James Hallock, MD, president, Educational Commission for Foreign Medical Graduates; John Norcini, PhD, president, Foundation for Advancement of International Medical Education and Research; Janet Grant, director, Director Open University Centre for Education in Medicine Milton Keynes; Barbara Stillwell, PhD, RN, coordinator, World Health Organization, Geneva; Jordan Cohen, MD, president, AAMC; Irshad Waheed, FRCS, secretary, College of Physicians and Surgeons; Dale Dauphinee, MD, FRCPC, executive director, Medical Council of Canada; and Amy Hagopian, PhD, faculty, University of Washington, School of Public Health.
1 Hallock JA, Seeling SS, Norcini JJ. The international medical graduate pipeline: recent trends in certification and residency training. Health Aff (Millwood). 2006;22:94–96.
2 Norcini JJ, Mazmanian PE. Physician migration, education and health care. J Contin Educ Health Prof. 2005;25:4–7.
3 Hagopian A, Ofusus A, Fatusi A, et al. The flight of physicians from West Africa: views of African physicians and implications for policy. Soc Sci Med. 2005;61:1750–1760.
4 Aluwihare AP. Physician migration: Donor country impact. J Contin Educ Health Prof. 2005;25:15–21.
5 Bundred PE, Levitt C. Medical migration: who are the real losers? Lancet. 2000;356:245–246.
6 Mullan F. The metrics of the physician brain drain. N Engl J Med. 2005;353:1810–1818.
7 Khan MM. The NHS International Fellowship Scheme in Psychiatry: robbing the poor to pay the rich? Psychiatr Bull. 2004;28:435–437.
8 Shafqat S, Zaidi A. Unwanted foreign doctors: what is not being said about the brain drain. J R Soc Med. 2005;98:492–493.
9 Salafsky B, Glasser M, Ha J. Addressing issues of maldistribution of health care workers. Ann Acad Med Singapore. 2005;34:520–525.
10 Zurn P. Understanding and influencing the international migration of health workers. WHO Geneva Briefing of HRH experts in WHO Africa Region; April 8–11, 2003; Ouidale, Benin.
11 Cooper RA. Physician migration: a challenge for America, a challenge for the world. J Contin Educ Health Prof. 2005;25:8–14.
13 Forcier MB, Simoens S, Giuffrida A. Impact, regulation and health policy implications of physician migration to OECD countries. Human Resources for Health, BioMed Central. Available at: (http://www.human-resources-health.com/content/2/1/12
). Accessed July 28, 2006.
14 Chen CC, Boufford JI. Fatal flows—doctors on the move. N Engl J Med. 2005;353:1850–1852.
15 Khan AJ. Scope of medical colleges in the private sector. J Ayub Med Coll Abbotabad. 2004;16:1–3.
16 Frossard PM. Reversing the brain drain: a matter of unity, faith and discipline. In: Talati J, ed. Developing the Complete Physician: Challenge for the University in the 21st
Century. Available at: (forthcoming)
17 Fresh Registration of Medical/Dental Practitioners from 200-2004. Pakistan Medical and Dental Council Gazetteer. Oct-Dec 2004;volume 21:issue IV.
23 Jafar T. The growing burden of chronic kidney disease in Pakistan. N Engl J Med. 2006;354:995–997.
24 Pappas G, Queen S, Hadden W, Fisher G. The increasing disparity in mortality between socioeconomic groups in the United States, 1960 and 1986. N Engl J Med. 1993;329:103–109.
25 Carrington WJ, Detragiache E. How extensive is the brain drain? Finance Dev. 1999;36(2):46–49.
26 Alkire S, Chen L. Medical exceptionalism in international migration: should doctors and nurses be treated differently? Joint Learning Initiative, Human Resources for Health and Development, The Global Equity Initiative, Harvard University Asia Center Working Paper 7-3, 2004. Available at: (http://www.globalhealthtrust.org/doc/abstracts/WG7/Alkirepaper.pdf
). Accessed September 14, 2006.
27 Eckhert NL. The global pipeline: too narrow, too wide or just right? Med Educ. 2002;36:600–601.
28 Brown RPC, Connell J. The migration of doctors and nurses from South Pacific Island Nations. Soc Sci Med. 2004;48:2193–2210.
31 Hyder A. Third World brain drain [letter]. BMJ. 2006;327:929–930.
33 Martineau T, Decker K, Bundred P. Briefing note on international migration of health professionals: levelling the paying field for developing country health systems. Available at: (http://www.eldis.org/static/DOC12324.htm
). Accessed July 28, 2006.
35 Eckhert NL. International health electives. In Talati J, ed. Developing the Complete Physician, Challenge for the University in the 21st Century. (forthcoming)
36 Pappas G, Akhtar T, Gergen PJ, Hadden WC, Khan AQ. Health status of the Pakistani population: a health profile and comparison with the United States. Am J Public Health. 2001;91:93–98.
39 Haq M. Human Development in South Asia. Karachi: Oxford University Press; 2000.
41 Darmstadt GL, Bhutta ZA, Cousens S, et al. Evidence-based, cost-effective interventions: how many newborn babies can we save? Lancet. 2005;365:977–988.
42 Sibley L, Sipe TA. What can a meta-analysis tell us about traditional birth attendant training and pregnancy outcomes? Midwifery. 2004;20:51–60.
43 Sazawal S, Black RE, Pneumonia Case Management Trials Group. Effect of pneumonia case management on mortality in neonates, infants, and preschool children: a meta-analysis of community-based trials. Lancet Infect Dis. 2003;3:547–556.
44 Bang AT, Bang RA, Baitule SB, Reddy MH, Deshmukh MD. Effect of home-based neonatal care and management of sepsis on neonatal mortality: field trials in India. Lancet. 1999;354:1955–1961.
47 Harden RM. A new vision for distance learning and continuing medical education. J Contin Educ Health Prof. 2005;25:43–45.
48 Ballard KD, Robinson SI, Laurence PB. Why do general practitioners from France choose to work in London practices? A qualitative study Br J Gen Pract. 2004;54:747–742.
49 Anand S, Barnighausen T. Human resources and health outcomes: cross country econometric study. Lancet. 2004;364:1603–1609.
50 Bury G, Cullen W, Beng Khoo S. Educational bridges between Ireland and Malaysia: Initial student responses to the Penang Medical College project. Asia Pacific Family Medicine. 2003;2:200–205. Available at: (http://www.apfmj.com/afm2.4/afm_90.pdf
). Accessed July 28, 2006.