|
|
| Home | Current Issue | Archives | Topic Collections | Search | Blog | Subscribe | Contact Us | Help |
|
Health Affairs, 22, no. 5 (2003):
241-249
doi: 10.1377/hlthaff.22.5.241 © 2003 by Project HOPE |
STATE REPORTHealth Departments’ Use Of International Medical Graduates In Physician Shortage Areas
The Conrad "State 20" Program places international medical graduates (IMGs) on J-1 visas in health professional shortage areas (HPSAs). The authors surveyed program administrators from health departments in forty-two participating states. Problems reported include unfair working conditions and compensation for physicians. Federal immigration agencies were reported to be unresponsive and difficult. Employers seem to be more satisfied than physicians with the program. After the exit of the U.S. Department of Agriculture as a sponsor for physician J-1 visa waivers, Congress expanded the Conrad Program, signaling a continued reliance on IMGs to serve in shortage areas.
Almost one in four doctors practicing in the United States, or about 180,000 physicians, graduated from medical school in another country. The events of 11 September 2001 led to important changes in U.S. attitudes and policy toward immigration, even for trained professionals. This paper reviews one of the primary mechanisms by which a subset of international medical graduate (IMG) physicians arrange to practice in the United States. The controversy about how many IMGs to attract tends to focus on a key question: Are IMGs more likely than their U.S. medical graduate (USMG) counterparts are to become part of the health care safety net, or do they simply exacerbate physician surpluses?1 In the short term, new IMGs can often be induced to serve in underserved communities while they participate in visa-waiver programs. In the longer term, however, there is conflicting evidence on whether IMGs are more likely than USMGs are to practice in underserved areas.2 We recently studied small, rural "critical access hospitals" and found them to be greatly dependent on IMGs—44 percent have at least one.3 In addition, Leonard Baer and colleagues estimated that if all IMGs in primary care practice were removed, one of every five "adequately served" nonmetropolitan counties would become underserved, and the percentage of rural counties with physician shortages would rise from 30 percent to 44.4 percent.4 The Conrad "State 20" Program was created in 1994 as an amendment to Title III of the Immigration and Nationality Act. Its purpose was to attract new IMGs to vacancies in health professional shortage areas (HPSAs) in both rural and urban settings. Section 220, which was designed to expire in October 2002, allowed each state annually to recommend twenty physicians on J-1 visas to practice in HPSAs.5 As of fall 2002, the new "State 30" Program allows each state to recommend thirty new J-1 visa waivers per year.6 No research has been published specifically analyzing this program. To fill this gap, we surveyed Conrad Program administrators to determine whether the program is meeting its purpose, its strengths and weaknesses, and how the program might be improved. We describe the types of physicians placed in the program, the potential demand if there were no limit on the number of slots available, problems with implementation and design, and the level of satisfaction with the program.
The Conrad Program, named for its sponsor, Sen. Kent Conrad (D-ND), is administered by state health departments. In return for a commitment to practice at least three years in a HPSA, physicians are granted waivers of a visa rule that would require them to return to their country of origin for a minimum of two years before becoming eligible to return. This J-1 visa waiver allows foreign-born, nonimmigrant physicians on exchange visas to remain in the United States during a period of obligated service employment; many subsequently apply for permanent residency status. As of February 2002 forty-one states and the District of Columbia were participating. Each jurisdiction could recommend twenty physicians for J-1 visa waivers each year. Federal law places minimal restrictions on physician placement via the Conrad Program. Employment at a "health facility" must begin with ninety days of waiver approval, the IMG must work full time for at least three years in a federally designated HPSA, and only thirty waivers are granted per state. Physicians who fail to fulfill their commitments must return immediately to their home country and may not apply for an immigrant visa or any other change of nonimmigrant status. The act has a sunset provision, limiting eligible physicians to those who acquire J-1 visa status before June 2004. At least one in five IMG residents are on J-1 student visas, while half are citizens or permanent residents, and one-fourth have unknown immigration status.7 To receive a J-1 visa waiver recommendation through the Conrad Program, an IMG must secure an employment agreement with an eligible employer (located in a HPSA), often with the assistance of an attorney or a recruitment firm. The employer and physician together then ask the state health department to recommend the waiver to both the State Department and the Bureau of Citizenship and Immigration Services (BCIS), formerly the Immigration and Naturalization Service (INS). These federal agencies alone have the ultimate authority to grant approval, although states rarely receive denials. Many physicians and their employers rely on state health departments to process J-1 visa waivers through the Conrad Program, although other federal "interested government agencies" (IGAs) can also recommend J-1 waivers to the BCIS and the State Department. Before the events of September 11 changed U.S. immigration policies, the U.S. Department of Agriculture (USDA) was a primary IGA for waiver applications, along with the Appalachian Regional Commission (ARC).8 With regulations announced 17 December 2002, the U.S. Department of Health and Human Services (HHS) took over the USDA role. In addition, the federal Delta Regional Authority announced that it would begin recommending J-1 visa waivers in its region. At the time this paper was written, proposed HHS program rules were still under review by the federal Office of Management and Budget (OMB). The policy analyst for HHS’s program, to be administered by the Health Resources and Services Administration’s (HRSA’s) Bureau of Health Professions (BHPr), could not predict the number of waivers that might be recommended.9 After September 11, reinterpretations and newly aggressive enforcement of immigration laws have created a less welcoming climate for nonimmigrants on exchange visas generally, but especially those from a list of twenty-five countries, who must engage in "special registration." This list includes Pakistan and other countries that contribute sizable numbers of IMGs. Data are not yet available to show whether this new climate discourages physicians from coming to the United States for medical residency, but universities around the country are reporting difficulty attracting and retaining foreign students generally.10
We contacted Conrad Program managers in each state health department participating in the program as of February 2002. We pre-tested and refined a Web-based questionnaire in two states (Washington and Texas) and subsequently solicited all participating jurisdictions to complete it. We sent paper questionnaires to the handful of program managers who requested them. There was a 100 percent response rate overall, but some questions had a lower response rate. In addition, we reviewed program materials and Web sites for each state’s Conrad Program, allowing us to examine each one’s rules and regulations independent of survey responses. The nine states not participating in the Conrad Program at the time of this study are mostly western.
Administrative factors. Ten states (24 percent of respondents) initiated their Conrad Programs in 1994, the year Congress passed authorizing legislation. Another twenty-two states started their programs in 1995 or 1996. Six were launched in 1997. No additional states entered until 2001. Half of the respondents had been in their Conrad Program management positions at least four years. Conrad Program rules allow each state health department to establish its own application process, deadlines, and selection criteria. Two-thirds of states reported that they do not have formal rules adopted by their state legislatures. One-third require the employer to provide an IMG physician retention plan as part of the application process. More than three-fourths of states require physicians to serve low-income populations; however, only seventeen respondents could say that their rules "effectively ensure that low-income populations are well served" by this program. The rules ranged from simply requiring the physician to accept Medicare and Medicaid patients, to requiring a sliding fee scale, to preferentially selecting health centers or nonprofit agencies for physician placement. More than half of the states did not report having any particular application deadlines: They simply collected and processed applications until all positions were full. Eleven states (27 percent) used a single annual deadline, and six states processed applications in batches. Some states with a single deadline continuously accepted applications if the slots were not full. More than half did not require physicians to have completed their residencies before they apply for a waiver. Just half had specific policies to deal with J-1 waiver physicians wishing to transfer to a new employer—sometimes in another state—before their period of obligation was completed. Goals of participation. Two-thirds of respondents reported that their state’s goal was to fill vacancies in all types of shortage areas through participation in the Conrad Program. Seven states (17 percent) said that they focus exclusively on rural areas, and five (12 percent), on urban areas.11 Only one in four respondents felt that their goals were entirely met, although none reported that their goals were not met at all. Seventeen respondents (41 percent) felt that the goal was "mostly" met, and thirteen (32 percent) said "somewhat."
Demand for positions.
The average number of Conrad Program placements was 13.5 per state per year (Exhibit 1
The number of applicants also could be an artifact of a state’s application cycle structure. If the state accepted continuous applications, it usually refused new applications after its positions were full. This would artificially depress the number of candidates reported to be interested in the positions. Applicants may not apply to more than one state at a time. When asked to speculate how many physicians could be placed in the Conrad Program if there were no limit to the number accepted, respondents reported a range of 0 to 100, with a median of 25 and an average of 27. More than half reported that they could place more than their allotted positions. The largest portion of states (29 percent) indicated, in an open-ended question, that they allocate scarce Conrad Program slots by evaluating the level of shortage in the community from which the employer is applying. The next criterion was "first come, first served" (23 percent); 18 percent said that they considered the type of facility foremost (public or nonprofit facilities first). "Quality" of applicant was not listed as a criterion. Physicians. Only a third of respondents reported perceiving that the IMG physicians in their programs were "highly satisfied" with their J-1 waiver practice sites. Two-thirds of respondents perceived that physicians were somewhat satisfied; two felt that physicians were generally dissatisfied. Conrad Program administrators are in a position to be aware of these levels of satisfaction because they work closely with employers and physicians both before and during placement.
The largest proportion of physicians placed by the states in this program during FY 2000–2001 were internists, followed by family practitioners, psychiatrists, and pediatricians (Exhibit 3
Employers. Although all employers in this program must be in HPSAs, there are no federal restrictions on the type of employers participating, and each state has its own rules about how many underserved patients should be served. Almost half of the 568 Conrad Program IMGs in FY 2000–2001 were employed in private practices or clinics. About one-fifth were placed in public clinics such as health centers, and nearly as many went to rural hospitals (Exhibit 3  ). Three-fourths of states placed 30 percent or more of their Conrad IMGs with private practices or clinics. More than half the states reported that they perceived the employers of Conrad IMGs to be "highly satisfied" with the program, and 43 percent said "somewhat" satisfied. If they had more resources to devote to the program, a third of states said that they would help employers or physicians with recruiting IMGs. Violations and problems. Many respondents reported experiencing problems with how employers treat their Conrad IMGs. Three-fourths of respondents said that their state’s Conrad IMGs had problems with employers changing work practices after the employment agreement was signed. Half said that they had problems with employers’ requiring physicians to work under poor or unfair working conditions, and more than half encountered problems related to compensation disputes. While three-fourths said that discrimination against J-1 physicians was not a problem, the remaining one-fourth said that it was "some problem" for the physicians. Twenty-four respondents (60 percent) reported that they knew of physicians in their Conrad Programs who had violated one or more requirements. The violations were most often related to the physician’s failure to work in the approved practice location, but a few were cited for failure to work forty hours a week or for practicing a specialty when they were supposed to be working as generalists. One state reported that a physician was not serving indigent patients, and another reported a criminal problem. All states reported having at least some problems with employers’ following program rules. When violations by physicians occur, two-thirds of states report the violation to the INS (now the BCIS). At least half of the states require a site visit, ten mandate a correction, and ten provide some kind of technical assistance. Some states use multiple approaches. Federal government relations. Relations between the state health departments and the federal agencies associated with the program are suboptimal. Nearly one-third of respondents said that their relations with the INS or BCIS were "poor" or "very poor," and only 25 percent said that their relationship was any better than "neutral." However, only 19 percent said that their relations with the State Department were "poor" or "very poor," and 57 percent said "good" or "very good." Only four states reported that they had ever had a waiver candidate denied by the federal government. One state reported six denials; the other three states reported only one or two. We asked respondents who their contacts were at the INS or BCIS. Only four were able to provide a contact name and phone number. "Establishing a link has been difficult," one said. Another said: "Are you kidding? INS is the biggest black hole in the universe!" By contrast, twenty states were able to name contacts at the State Department. Tracking. Most states have a three-year period of obligation for each physician (although some states require five years, such as for specialists). Only one-fourth of states track their physicians’ locations after the period of obligation is completed; six respondents said that they track for more than five years. Half of the states require reports from physicians or their employers, or both, every year during the obligation period, and half require a report every six months. Four states limit reporting to once during the entire placement. Only six states (15 percent) conduct exit interviews with their physicians at the end of the obligation period, and only eleven states (28 percent) had conducted a site visit to one or more employers in the past eighteen months. When asked how they would improve their Conrad Programs if they had more resources, 41 percent of respondents said that they would expand their tracking systems. Costs of participation. The federal government provides no financial support for the Conrad Program. Respondents indicated that they allocate a range of zero to one (average of 0.44) full-time-equivalent (FTE) positions for program administration. Respondents’ recommendations. In an open-ended question, respondents were asked to suggest improvements to the program. The improvement suggested most often was to expand the number of J-1 waivers available to each state; the next was federal funding of administrative costs. Suggestions included that the federal government centralize the application process, provide technical assistance to states, and improve communication between the states and federal agencies. In particular, some expressed support for better enforcement of regulations or delegation of authority to enforce regulations themselves.
Limitations. This study is limited in two respects. First, the survey was performed just prior to widespread knowledge of the USDA’s decision to stop sponsoring IMGs for J-1 visa waivers and before HHS assumed the USDA’s role.12 Therefore, state administrators might not have been in a position to accurately estimate current demand for slots. Second, we relied on a single program coordinator in each state health department to provide responses to the survey. We did not verify data with other parties, although we did extensively review program Web sites. Problems identified. We identified several problems that could have a significant bearing on the current and future implementation of this program, and the placement of IMGs generally. These include violations of program rules, staffing costs, the lack of information about retention experience, and the relationship between state health departments and the BCIS. In addition, while physicians must practice in HPSAs, there is no explicit requirement at the federal level that they serve underserved patients per se. More than half of respondents reported problems with how employers treat Conrad Program physicians. Most of these problems related to alterations in work practices after the employment agreements had been signed, but at least half of respondents reported some problem with physician compensation disputes and suboptimal working conditions. Employment contracts are typically drafted by an attorney, who may also represent the employer, which can lead to conflicts of interest. These contracts vary widely and may be vague or misleading. Furthermore, in some circumstances IMG physicians could be reluctant to report violations for fear of adversely affecting their immigration status. Reports of contract violations by employers, many of them private practices, are troubling and apparently widespread. Immigrant physicians who are exploited or mistreated by employers have little access to justice in the absence of any enforcement of rules. Independent interviews we conducted with twenty-one IMGs in one state for a different study substantiated this concern. More effective monitoring and tracking of working conditions during the entire period of physician placement, coupled with clear sanctions for violations, might improve the situation. Is the Conrad Program adequate? Less than two-fifths of states filled all twenty of their available Conrad Program positions in 2000–2001. Some states relied more heavily on alternative programs, such as the former USDA program, to place IMG physicians in underserved locations. The policy of allocating slots by state, of course, disregards the states’ fluctuating populations and needs.13 We can only speculate about reasons for the distribution of IMGs across the country. Migration theory tells us that new immigrants follow previous ones, thus creating geographic regions that are more immigration-friendly. Also, residents tend not to move far from where they trained, and half of all IMGs in residency training programs are in New York, Illinois, Pennsylvania, and New Jersey.14 The fact that almost half the physicians recommended for Conrad waivers went to private practices, coupled with the reports about physician exploitation, remind us that the program benefits employers as well as physicians and communities. Most respondents felt that their state’s rules failed to ensure that low-income clients were well served, further clouding the issue of who benefits most. Two-thirds of Conrad IMGs placed in 2000–2001 were employed as primary care physicians. However, internists or pediatricians might have received fellowship training in a subspecialty area and therefore might not really be practicing as generalists. This could lead to the suboptimal practice of primary care and also to higher rates of attrition when the period of obligation is over. The Conrad Program has no federal funding. Although the program does help states meet demand for physician services in shortage areas, the lack of federal support implementation is clearly frustrating for many states and limits their ability to enforce rules and track placements, especially with current state budget constraints. The program has no legislative provisions to promote the long-term retention of physicians. The lack of tracking of J-1 physicians in three-fourths of states makes it difficult to measure the program’s long-term effects. Since the states agree that tracking after the obligation period ends is a high priority, the federal government should invest in this effort, especially since growth in the program will make the need for this information even more important to policymakers. A shift in policy. Congress has elected to expand the Conrad Program to thirty slots per state, and all states are now reported to participate. Since Conrad is one of the few areas where Congress makes direct and explicit policy affecting IMGs, expansion signals a continued reliance on IMGs to help solve the U.S. physician maldistribution problem. Favorable IMG policy changes seem somewhat at odds with recent stricter overall immigration policies, but U.S. policy has historically been friendlier to highly educated immigrants than to others. HRSA reports the immediate need for 20,000 primary care and psychiatric physicians to serve in HPSAs, with only half that many graduating from the class of 1999, and fewer than 1,000 of those electing to work in a HPSA.15 Further, the declining number of medical students electing to enter generalist residency training is widely documented.16 As a possible alternative to the Conrad Program, the United States could expand the National Health Service Corps, which has been demonstrated to induce USMGs to practice primary care in underserved areas. While that program was expanded by 20 percent in 2003 and an additional expansion is proposed for FY 2004, it is costly to provide scholarships and loan repayments. The J-1 visa waiver programs essentially provide "free physicians," whose education was obtained in locations where U.S. taxpayers have no obligations. Most Conrad physicians are from low- or lower-middle-income countries (as defined by the World Bank).17 The continued reliance on poor countries to supply U.S. physicians represents a decision to transfer wealth from poor countries to rich ones. At some point, the United States should decide either to curtail its appetite for medical care or to pay for the true cost of this decision. Regardless of the ethical implications of relying on IMGs, our findings demonstrate there are opportunities to improve the Conrad Program. Resources should be allocated for administration, tracking of physicians both during and after obligation, investigating and pursuing violations, ensuring that service is provided to the underserved as intended, and improving federal immigration agencies’ responsiveness to state health departments.
Amy Hagopian is a doctoral candidate in the University of Washington’s School of Public Health and Community Medicine and a research associate with the Washington, Wyoming, Alaska, Montana, and Idaho (WWAMI) Center for Health Workforce Studies (CHWS) at the University of Washington (UW) Department of Family Medicine. Matthew Thompson is an investigator with the WWAMI CHWS and an assistant professor at the UW Department of Family Medicine. Emily Kaltenbach is a planner with Presbyterian Medical Services, a New Mexico health and human service agency. L. Gary Hart is director and principal investigator at the WWAMI CHWS and a professor in the UW Department of Family Medicine. The authors are grateful to Jennell Prentice, the Washington State Conrad Program coordinator, and Connie Berry, who coordinates the program in Texas, for their help with questionnaire design and testing and other assistance on this project. They also thank Carolyn Watts, Oscar Gish, and Allen Cheadle, who reviewed this paper. Research was funded by the Health Resources and Services Administration’s National Center for Health Workforce Information and Analysis, Bureau of Health Professions, Cooperative Agreement no. 6 U79 HP 00003-04-01.
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Background
Study Design And Methods
