The Appalachian region of the United States is historically associated with rugged geography, scenic beauty, a unique value system, serious economic challenges,1 – 4 and inadequate health care.5,6 More than 24 million people live in Appalachia, which encompasses parts of Alabama, Georgia, Kentucky, Maryland, Mississippi, New York, North Carolina, Ohio, Pennsylvania, South Carolina, Tennessee, and Virginia, and all of West Virginia.7
Although health conditions have improved in some parts of Appalachia in recent years,8 health disparities relative to other regions of the United States continue.9 – 14 Physician supply in Appalachia has increased, but the region remains subject to physician shortages, and many economically distressed counties lack adequate medical care.8,12,15 Because 42% of Appalachian residents live in rural areas,7 the expected continuing or worsening shortage of rural physicians in the United States16,17 will disproportionately affect the people of Appalachia.
One of medical schools' basic missions is to graduate physicians who will meet the health care needs of local and regional citizens.18,19 Many state-supported medical schools have institutional missions involving service to their states, and all medical schools that accept public support have some obligation toward meeting their states' workforce needs.17 Although physicians choose their own specialty and practice location, medical schools have significant influence on their graduates' decisions.20,21
In a study published in 1991, Roberts and colleagues22 determined which medical schools had produced the most physicians practicing primary care and non-primary-care specialties in rural and urban areas of Appalachia. Their findings supported existing concerns regarding attracting and retaining physicians to serve residents of rural Appalachia. To update Roberts and colleagues' findings, we identified the schools that, more recently, have been turning out the most graduates who practice in Appalachia. We analyzed graduates of three time periods (1991–1995, 1996–2000, 2001–2005) to identify recent trends and to categorize graduates as practicing primary care or non-primary-care specialties. We also identified the schools with the most recent graduates (1991–2005 period) practicing in Appalachia's economically distressed and at-risk counties and those that have contributed the most physicians to the Appalachian counties of each state.
Physician cohorts and data
For this study, conducted in fall 2010 to summer 2011, we obtained June 2009 physician data from the American Medical Association's Physician Masterfile.23 Physician Masterfile data are used for many purposes, including national health workforce analysis and planning. The Physician Masterfile has many data strengths and is widely regarded as the most complete, uniform, and accurate source of information on physicians training and working in the United States,24 – 26 although it also has known limitations.26 – 28
The Physician Masterfile includes information on all graduates of U.S. MD-granting medical schools, all graduates of U.S. DO-granting medical schools since the late 1980s, and all international medical graduates who have trained in U.S. residency programs and/or have practiced in the United States. Data on physicians' practice specialty and practice location are continually updated using information from a variety of sources,23,29 including the physicians themselves.
To identify physicians practicing in the Appalachian region, we used data from the Physician Masterfile's “MaxOffice” address field, which reflects the practice location for 86% of office-based physicians and 76% of physicians overall.30 There were 64,798 physicians with MaxOffice addresses in Appalachian counties as of June 2009. Of these, 45,415 worked in clinical practice (i.e., they were not in training, in nonclinical positions, retired, or otherwise not working). We then excluded 521 physicians whose specialty was unknown and, for the purposes of our analyses, classified the remaining 44,894 as practicing in primary care specialties (family medicine, general internal medicine, and general pediatrics) or non-primary-care specialties (all other specialties).
To analyze recent U.S. medical school graduates' Appalachian practice locations and specialties, we first sought to identify the most recent five-year cohort that would have had time to complete their first residencies and for whom Physician Masterfile data would adequately capture specialties and postresidency practice locations.27,28 Accordingly, we assessed the type of practice as of June 2009 for all 2000–2007 U.S. graduates nationally. Among physicians in primary care specialties, we found that 94.7% of 2006 graduates were still listed in residency positions; however, the proportion dropped to 17.4% for 2005 graduates, 4.4% for 2004 graduates, and 0.3% for 2001 graduates. File address and specialty information on transitions out of residency positions were more delayed for non-primary-care physicians, presumably because of longer residency training periods: We found that 88.6% of 2005 graduates were listed in residency positions, as were 33.9% of 2003 graduates, 7.0% of 2001 graduates, and 2.5% of 2000 graduates.
We selected 2001–2005 as the most recent five-year period for which we could track graduates into practice as of June 2009, recognizing that about 4% of this period's graduates who selected primary care specialties and 40% who selected non-primary-care specialties were still listed in residency positions. We did not include physicians listed as in training or those in other nonpracticing roles in our counts of physicians practicing in Appalachia or in the denominators of all graduates from each school when calculating percentages of schools' graduates practicing in Appalachia. Basing our analyses only on the physicians who had completed their residencies and whose posttraining practice locations and specialties were captured in the Physician Masterfile as of June 2009 should not substantially bias analyses of which medical schools produced the most physicians practicing in Appalachia, because longer residency periods and data delays for some graduates should affect graduates of all schools. We also identified physicians practicing in Appalachia who graduated during two prior five-year periods—1991–1995 and 1996–2000—to allow us to assess changes in schools' numbers of physicians produced for Appalachia.
Appalachian county data
The Appalachian Regional Commission (ARC) identifies 420 counties in 13 states as being within Appalachia (Figure 1).31 The ARC uses an index of three indicators—three-year average unemployment rate, per capita market income, and poverty rate—to compare these 420 counties with counties nationally in order to classify Appalachian counties into five economic levels: distressed (in the lowest 10% of U.S. counties), at risk (bottom 10%–25% of U.S. counties), transitional (25%–75% of U.S. counties), competitive (top 75%–90% of U.S. counties), and attainment (in the top 10% of U.S. counties).
In fiscal year 2011, the ARC classified 82 of the 420 Appalachian counties as distressed and 86 as at risk.32 For our analyses, we grouped these two categories into a single “distressed/at-risk” group of 168 counties (Figure 1). We also classified each of the 420 Appalachian counties as rural or urban according to 1999 federal metropolitan and nonmetropolitan designations.33
We determined which physicians among the 44,894 with known specialties practicing in Appalachian areas in June 2009 were graduates of U.S. versus international medical schools. We then identified those U.S. graduates who were practicing primary care and non-primary-care specialties in the region's rural and urban counties; next, we generated rank-ordered lists of the medical schools that had graduated the highest numbers of these physicians during 2001–2005. We also identified the 1996–2000 and 1991–1995 graduates of these schools who were practicing primary care and non-primary-care specialties in rural and urban Appalachian counties.
In addition, we calculated the number of 1991–2005 graduates of U.S. and international medical schools who were practicing in Appalachia's economically distressed and at-risk counties in June 2009. We used these data to identify the top 10 U.S. medical schools in terms of the number of their graduates working in these counties as primary care and non-primary-care physicians. Finally, we identified the U.S. medical schools with the most 1991–2005 graduates practicing in each state with Appalachian counties as of June 2009.
We performed analyses using SAS version 9.1 (SAS Institute, Inc.; Cary, North Carolina). Our findings are purely descriptive; we do not use inferential statistics. This study was exempted from human subjects review by the West Virginia School of Osteopathic Medicine's institutional review board.
Of the 44,894 physicians with known specialties whom we determined to be in clinical practice in Appalachia's 420 counties as of June 2009, 34,784 (77.5%) were graduates of U.S. medical schools, and 10,110 (22.5%) were international medical graduates. Among the U.S. medical school graduates, we identified 13,277 (38.2%) practicing in primary care specialties. The practice locations we found for 13,258 (99.8%) of these primary care physicians placed 8,655 (65.3%) of them in rural counties and 4,603 (34.7%) of them in urban counties. Among the 21,507 graduates of U.S. medical schools practicing in non-primary-care specialties, we found practice locations for 21,491 (99.9%). In contrast to primary care physicians, 15,759 (73.3%) of the non-primary-care physicians worked in urban counties, whereas 5,732 (26.7%) practiced in rural counties.
Medical schools with the most recent graduates practicing in primary care specialties in Appalachia
The U.S. medical schools with the most 2001–2005 graduates practicing in primary care specialties in rural or urban Appalachian counties as of June 2009 were all located in the 13 states that the region encompasses (Table 1). Ten schools produced 50.5% (197/390) of all U.S. medical school graduates from these years who practiced primary care in rural Appalachian counties. The West Virginia School of Osteopathic Medicine ranked first, followed by the University of Pikeville Kentucky College of Osteopathic Medicine and the University of Alabama School of Medicine. Six of these 10 schools are located not just in Appalachian states but actually in Appalachian counties.
Ten schools produced 44.1% (366/829) of all the U.S. medical school graduates from 2001 to 2005 who practiced primary care in urban Appalachian counties. The University of Alabama School of Medicine ranked first, followed by the Lake Erie College of Osteopathic Medicine and the University of Pittsburgh School of Medicine. Seven of these 10 schools are located in Appalachian counties. Seven schools appear on both top 10 lists.
The rank-ordering of the top schools in terms of numbers of graduates practicing primary care in rural and in urban Appalachian counties held fairly constant over each of the three periods. Two schools that did not exist in 1991 and produced their first graduates during one of the later cohorts were among the top contributors of primary care physicians in rural counties (University of Pikeville Kentucky College of Osteopathic Medicine) or urban counties (Lake Erie College of Osteopathic Medicine) in 2009.
Medical schools with the most recent graduates practicing in non-primary-care specialties in Appalachia
The U.S. medical schools that produced the most 2001–2005 graduates practicing as non-primary-care physicians in rural and urban Appalachian counties are all located in states that contain Appalachian counties (Table 2). Ten schools accounted for 42.1% (128/304) of 2001–2005 U.S. medical graduates practicing in non-primary-care specialties in rural areas of Appalachia in June 2009. The top-producing schools among these 10 were the West Virginia University School of Medicine, Georgia Health Sciences University, and the University of Mississippi School of Medicine. Five of the 10 schools are located in the Appalachian region.
Eleven medical schools (there was a tie for 10th place) produced 38% (323/840) of 2001–2005 U.S. medical graduates working as non-primary-care physicians in urban areas of Appalachia in June 2009. The top-producing schools among these 11 were the University of Alabama School of Medicine, the West Virginia University School of Medicine, and the University of Pittsburgh School of Medicine. Six of the top 11 schools are located within the Appalachian region.
Six schools appear on both of these top 10 lists. As with schools producing graduates practicing primary care, the rank-order of schools over the three study period cohorts is fairly constant. Lake Erie College of Osteopathic Medicine, which did not exist in 1991 but produced graduates in the 1996–2000 and 2001–2005 cohorts, was among the top contributors of non-primary-care physicians to both rural and urban areas.
Medical schools with the most recent graduates practicing in economically distressed and at-risk Appalachian counties
We identified 747 primary care physicians who graduated from medical school from 1991 to 2005 and were practicing in Appalachia's distressed/at-risk counties33 in July 2009 (Table 3). A total of 592 (79.3%) of these primary care physicians graduated from U.S. medical schools. Slightly more than half (305; 51.5%) of the U.S. medical school graduates attended just 10 schools. The top producers were the University of Kentucky College of Medicine (52 graduates), the West Virginia School of Osteopathic Medicine (48 graduates), and the University of Louisville School of Medicine (39 graduates).
We also identified 505 non-primary-care physicians who graduated from medical school during 1991–2005 and were practicing in distressed/at-risk Appalachian counties in July 2009 (Table 3). Of these, 398 (78.8%) graduated from U.S. medical schools; half of them (199; 50%) graduated from just 10 schools. The University of Kentucky College of Medicine contributed the highest number (34 graduates), followed by the University of Louisville School of Medicine (27 graduates) and the University of Mississippi School of Medicine (24 graduates).
Medical schools with the most recent graduates practicing within each state's Appalachian counties
As physicians often practice near where they trained,34 we were not surprised to find that the top-producing school for each of the 13 states in the Appalachian region was located within the state. Further, in 11 of these states, the school ranked second was also located in the state. Supplemental Digital Table 1 (see http://links.lww.com/ACADMED/A80) highlights for each Appalachian-region state the U.S. medical schools that produced the most 1991–2005 graduates practicing in that state's Appalachian counties.
One of the important challenges medical schools face is fulfilling their social mission of providing physicians to meet the needs of their regions.18 Although schools' institutional missions vary,35 many schools, including most of those in or near Appalachia, have in their missions service to their state, their geographic area, and often underserved areas or populations. The majority of the schools we identified as producing the highest numbers of physicians practicing in Appalachia are located within the region's 13 states: These schools therefore appear to be meeting this aspect of their respective missions. Schools varied in their ranking on lists of top producers of primary care versus non-primary-care physicians, and rural versus urban physicians practicing in Appalachia, which likely reflects other aspects of their respective missions.
The schools that produced the highest numbers of graduates practicing in Appalachia were remarkably stable across the three time periods we studied (1991–1995, 1996–2000, 2001–2005). Furthermore, our results were similar to those of Roberts and colleagues'22 1991 study of graduates serving the Appalachian region in 1989 or 1990: Both studies identified the same top school for each category (i.e., those producing the most rural and urban primary care physicians and rural and urban non-primary-care physicians). The major difference between our findings and theirs is that our lists of top schools include new schools located in or near Appalachia that did not exist in the late 1980s: University of Pikeville Kentucky College of Osteopathic Medicine, Lake Erie College of Osteopathic Medicine, and East Tennessee State University James H. Quillen College of Medicine.
Because of methodological differences between the two studies, direct comparisons between the two are not possible: however, our results may indicate a slight decline in the number of graduates practicing in Appalachia from institutions that were identified as top producers in the 1991 study. This decline appears to be more than made up for by the numbers of graduates produced by new medical schools, including the new schools appearing in our top 10 lists. The expansions in class size36,37 taking place at several schools on the top 10 lists should help these schools place more physicians in Appalachia in the future. Further, contributions are expected from other new schools in the Appalachian region that do not appear in the top 10 lists, including but not limited to the University of South Carolina School of Medicine in Greenville, which has received provisional accreditation from the Liaison Committee on Medical Education,38 as well as Lincoln Memorial University–DeBusk College of Osteopathic Medicine in Harrogate, Tennessee, and Edward Via College of Osteopathic Medicine in Blacksburg, Virginia, which are both accredited by the American Osteopathic Association Commission on Osteopathic College Accreditation.39
Although we focused our study on graduates of U.S. medical schools, 10,110 (22.5%) of the physicians practicing in Appalachia as of June 2009 had graduated from international medical schools during 1991–2005. Further, international medical graduates accounted for more than one-fifth (21%) of the physicians practicing in economically distressed and at-risk Appalachian counties (Table 3). Although there are important policy questions regarding recruitment and retention of international medical graduates, particularly in primary care specialties,40,41 these physicians are important to the Appalachian region. Policy changes that make recruitment of international graduates more difficult41,42 could have a significant impact on the health care in the region, particularly in its underserved areas.
The AMA's Physician Masterfile is widely used in health workforce analyses, but it is not free of errors.27,28 The Physician Masterfile's data limitations should not, however, favor one school over another and, therefore, should not have influenced the order of schools in our top 10 lists. This assumption is confirmed by the general consistency in the schools identified as producing the highest numbers of primary care and non-primary-care physicians in urban and rural Appalachia across the three time periods studied.
This report does not consider medical students' origins (e.g., whether graduates entering practice in Appalachia were originally from the region) or where graduates completed their residency training. Further research is warranted regarding both of those factors. It is possible that students from the region may be more likely to enter schools in or near Appalachia, to complete residencies in the region, and/or to return to Appalachia to practice. In addition, some of the schools on the top 10 lists have educational programs designed to encourage students to practice in their state or in the Appalachian region.43 – 45
The physician workforce in Appalachia consists largely of graduates of MD- and DO-granting schools in or near the region—this holds true for physicians practicing in urban and rural counties and in primary care and non-primary-care specialties. The medical schools that produce the highest numbers of these physicians are meeting, at least in part, medical schools' shared mission to educate physicians who will serve local populations with particular needs. The new medical schools now being developed within the region will likely make important additional contributions to the region's physician workforce.38,39
The authors would like to thank their colleagues, particularly C. Ken Shannon, MD, PhD, for critical review of earlier versions of this article. They would also like to thank Katie Gaul of the Cecil G. Sheps Center for Health Services Research at UNC–Chapel Hill for creating the map.
Supplemental digital content for this article is available at http://links.lww.com/ACADMED/A80.
1. Williams JA. Appalachia: A History. Chapel Hill, NC: University of North Carolina Press; 2002.
2. Lengerich EJ, Bohland JR, Brown PK, et al.. Images of Appalachia. Prev Chronic Dis. 2006;3:A112.
3. Coyne CA, Mian-Popescu C, Friend D. Social and cultural factors influencing health in southern West Virginia: A qualitative study. Prev Chronic Dis. 2006;3:A124.
4. Jones L. Appalachian Values. Ashland, Ky: Jesse Stuart Foundation; 1994.
5. Barney SL. Authorized to Heal: Gender, Class and the Transformation of Medicine in Appalachia, 1880–1930. Chapel Hill, NC: University of North Carolina Press; 2000.
6. Elnicki DM, Morris DK, Shockcor WT. Patient-perceived barriers to preventive health care among indigent, rural Appalachian patients. Arch Intern Med. 1995;155:421–424.
9. Hendryx M, Zullig KJ. Higher coronary heart disease and heart attack morbidity in Appalachian coal mining regions. Prev Med. 2009;49:355–359.
10. Zullig KJ, Hendryx M. A comparative analysis of health-related quality of life for residents of U.S. counties with and without coal mining. Public Health Rep. 2010;125:548–555.
11. McGarvey EL, Leon-Verdin M, Killos LF, Guterbock T, Cohn WF. Health disparities between Appalachian and non-Appalachian counties in Virginia USA. J Community Health. 2010;36:348–356.
12. Denham SA, Wood LE, Remsberg K. Diabetes care: Provider disparities in the US Appalachian region. Rural Remote Health. 2010;10:1320.
13. Bailey BA, Cole LK. Rurality and birth outcomes: Findings from southern Appalachia and the potential role of pregnancy smoking. J Rural Health. 2009;25:141–149.
14. Barker L, Crespo R, Gerzoff RB, Denham S, Shrewsberry M, Cornelius-Averhart D. Residence in a distressed county in Appalachia as a risk factor for diabetes, Behavioral Risk Factor Surveillance System, 2006–2007. Prev Chronic Dis. 2010;7:A104.
15. Khanna R, Bhanegaonkar A, Colsher P, Madhavan SS, Halverson J. Breast cancer screening, incidence, and mortality in West Virginia. W V Med J. 2009;105 spec no:24–32.
16. Whitcomb ME. The challenge of providing doctors for rural America. Acad Med. 2005;80:715–716.
17. Rosenblatt RA. Commentary: Do medical schools have a responsibility to train physicians to meet the needs of the public? The case of persistent rural physician shortages. Acad Med. 2010;85:572–574.
18. Mullan F, Chen C, Petterson S, Kolsky G, Spagnola M. The social mission of medical education: Ranking the schools. Ann Intern Med. 2010;152:804–811.
19. McCurdy L, Goode LD, Inui TS, et al.. Fulfilling the social contract between medical schools and the public. Acad Med. 1997;72:1063–1070.
20. Colwill JM, Cultice JM, Kruse RL. Will generalist physician supply meet demands of an increasing and aging population? Health Aff (Millwood). 2008;27:232–241.
21. Rabinowitz HK, Diamond JJ, Markham FW, Wortman JR. Medical school programs to increase the rural physician supply: A systematic review and projected impact of widespread replication. Acad Med. 2008;83:235–243.
22. Roberts A, Davis L, Wells J. Where physicians practicing in Appalachia in 1978 to 1990 were trained and how they were distributed in urban and rural Appalachia. Acad Med. 1991;66:682–686.
25. Cherkin D, Lawrence D. An evaluation of the American Medical Association's Physician Masterfile as a data source—One state's experience. Med Care. 1977;15:767–769.
26. Williams PT, Whitcomb M, Kessler J. Quality of the family physician component of AMA Masterfile. J Am Board Fam Pract. 1996;9:94–99.
27. Staiger DO, Auerbach DI, Buerhaus PI. Comparison of physician workforce estimates and supply projections. JAMA. 2009;302:1674–1680.
28. Rittenhouse DR, Mertz E, Keane D, Grumbach K. No exit: An evaluation of measures of physician attrition. Health Serv Res. 2004;39:1571–1588.
30. Fleming R. Medical Marketing Service. Personal communication with DE Pathman, 2004.
34. Dorner FH, Burr RM, Tucker SL. The geographic relationships between physicians' residency sites and the locations of their first practices. Acad Med. 1991;66:540–544.
35. Hirsch J. Social responsibility of medical schools. Arch Intern Med. 1969;124:113–114.
38. Liaison Committee on Medical Education. Institutions with developing medical education programs that have applied to the LCME for preliminary accreditation. http://www.lcme.org/newschoolprocess.htm
. Accessed December 10, 2011.
40. Fink KS, Phillips RL Jr, Fryer GE, Koehn N. International medical graduates and the primary care workforce for rural underserved areas. Health Aff (Millwood). 2003;22:255–262.
41. Thompson MJ, Hagopian A, Fordyce M, Hart LG. Do international medical graduates (IMGs) “fill the gap” in rural primary care in the United States? A national study. J Rural Health. 2009;25:124–134.
42. Terhune KP, Zaydfudim V, Abumrad NN. International medical graduates in general surgery: Increasing needs, decreasing numbers. J Am Coll Surg. 2010;210:990–996.
43. Stookey JR, Baker HH, Nemitz JW. How West Virginia School of Osteopathic Medicine achieves its mission of providing rural primary care physicians. J Am Osteopath Assoc. 2000;100:723–726.
44. Florence JA, Goodrow B, Wachs J, Grover S, Olive KE. Rural health professions education at East Tennessee State University: Survey of graduates from the first decade of the community partnership program. J Rural Health. 2007;23:77–83.
45. Wheat JR, Brandon JE, Leeper JD, Jackson JR, Boulware DW. Rural health leaders pipeline, 1990–2005: Case study of a second-generation rural medical education program. J Agromedicine. 2007;12:51–61.
References in Figure 1 only
This study was exempted from IRB approval by the West Virginia School of Osteopathic Medicine institutional review board.
An abstract emphasizing the rural findings in this study was presented at the 2011 annual meeting of the National Rural Health Association; Austin, Texas; May 4–6, 2011.