Accelerated Family Medicine Training Programs

Robert C. Bowman, M.D.     2009 update

Typical physicians have 9% rural location and 7% are located in underserved areas outside of Major Medical Centers. Family physicians have had a consistent 20% rural location and 14 - 15% found in underserved locations. Older age and lower and middle income background facilitate the highest levels of rural or of underserved distribution using complete populations in representative career and location choices (not just first time practices). 

Accelerated graduates have maintained 40% rural location and 18% underserved location and one of the highest rates of underserved plus rural location. Age is a key factor. About 22% of typical allopathic graduates are older than 29 years at graduation, 28% of family physicians, and 56% of accelerated graduates are older. FPs who are older are more likely to be found in rural and underserved areas. Age, experience, rural and lower income origins, and the commitment known as family medicine maximizes distribution.

Short and Sweet on Accelerated Family Medicine Training Programs

Top Family Medicine Residency Programs Ranked by Rural Percentage of Graduates

An Analysis of the Distribution of Accelerated Graduates

While exploring variation in residency training length in the Masterfile, accelerated graduates were noted to have a 2 year gap between medical school and residency graduation dates. Data was collected regarding 12 programs and 132 accelerated residents graduating between 1997 and 2003 as identified in the 2005 Masterfile.

Accelerated Family Medicine Advantages

•           40% of accelerated graduates in rural locations or four times the national average for all physicians or twice the level of rural distribution (20%) of all family physicians.

•           18% in underserved locations outside of major medical centers compared to 14% underserved for family medicine and a 7% national average for all physicians.

•           Inservice scores, chief resident positions, and increased choice of faculty careers1,2

•           Distribution at the same high levels for female, urban origin, younger, higher income origin, and foreign born students that are increasingly admitted and traditionally have lower levels of distribution

•           Adaptable to specific distributional needs such as frontier rural practice3

•           Savings of a year of medical school costs and costs of living. Many programs actually provided some salary support for the transitional year with or without incurring an obligation for distribution.

•           Savings of the costs of moving as trainees have 3 years of medical school and 3 years of residency in the same academic location, an advantage for older grads, those with children, or those with a spouse finishing education.

•           Back to back placement of the major clinical years, the first clinical year (M-3) and the M-4/PGY-1 year

The challenges in accelerated programs involve

•           Insufficient numbers of US MD Grads choosing family medicine such that community-based programs successfully pressed forward their argument that accelerated programs had an unfair advantage in recruitment

•           Admissions of candidates ready to commit a year early to family medicine and a challenging curricula

•           Sufficient adaptation of the M-4/PGY-1 year to facilitate a difficult transition; including an optimal balance between resident development and supervision to “accelerate” development; including the active participation of the accelerated residents in program development and recruitment; and including an experienced, secure, committed, versatile program director willing to devote significant time to the project

•           Superior faculty and volunteer faculty to provide a top quality training experience, especially in the first years of the program, enough for students to take the risk of not going through the match

•           A superior group of initial accelerated residents to reassure medical school leadership and ensure the best future recruitment, a critical factor since all residents remain local, vocal, accessible and influential

•           Sufficient support from the medical school, residency, hospitals, faculty, and federal facilities involving licensure, privileges, and other adaptations

•           Loss of potential family medicine residents when accelerated applicants are turned down

•           Lack of understanding of distributional outcomes and poor coordination across family medicine leadership, across medical association leadership, and across both accrediting bodies, the ACGME and LCME

Accelerated programs have the potential

•           To specifically train primary care graduates for all of the most complex family medicine careers: inner city,  international, military, frontier rural, faculty positions, public health, behavioral emphasis

•           To assimilate older student admissions tracks, family medicine-oriented basic sciences, and community-based programs to form a new type of medical education with the lowest costs, the best quality, and the best distributional outcomes in the nation

•           To attract new funding for permanent primary care that remains within the state and targets greatest need

References

1.         Petrany SM Crespo R, The Accelerated Residency Program: The Marshall University Family Practice 9-year Experience, Family Medicine 2002; 3 4(9):669-72

2.         Galazka SS, Zweig S, Young P. A progress report on accelerated residency programs in family practice. Acad Med 1996;71:1253-5 

3.         Stageman JH, Bowman RC, Harrison JD. An accelerated rural training program. J Am Board Fam Pract. 2003 Mar-Apr;16(2):124-30.

Accelerated Family Medicine Training Programs:

Fulfilling the Promise of the Discipline

Family medicine is about helping others to fulfill their potential. This article is dedicated to James Stageman MD, now retiring, who has continued to develop the full potential of students, residents, patients, residency programs, and family medicine itself over a lifetime of service.

Abstract: This study involves 142 accelerated family medicine residents graduating from 1997 – 2003 from 12 different programs across the nation. These residents attended medical school and residency at the same location, 3 years at each level of training (3-3). Nearly 80 % of accelerated family medicine residency graduates have chosen rural, poverty, or teaching locations at rates 50 – 100 % higher than traditional graduates. The model potentiates the distributional characteristics of rural origin and older students. The model facilitates the distribution of those least likely to choose rural practice. This includes greater rural location percentages in students born in the most urban counties and female physicians. This is important because these are the two groups increasing the most in US allopathic admissions and both are less likely than their counterparts to choose family medicine and rural careers of all physician types. Female accelerated graduates chose smaller towns in the same distribution pattern as males and even had an advantage in urban poverty location levels that was even more dramatic for those from public medical schools. Accelerated graduates also had a much higher rate of retention compared to traditional graduates with 80 % of graduates retained within the state of training. Selections based on early readiness and desire for a family medicine career combined with early and specific family medicine training appears to fulfill medical student potential for service in underserved areas.

Introduction

Accelerated family medicine training programs linked 3 years of medical school and 3 years of family medicine residency training at the same location. At least 15 allopathic medical schools have had accelerated programs in the past 10 years, although only a few departments developed mature programs (Tolleson).¹ Students were selected based on maturity and academic standing (Bratton).² Not surprisingly, accelerated graduates did very well academically and have contributed in teaching and leadership areas during and after residency (Petrany and Crespo)³. (Galazka, Zweig, Young)⁴. Over 80 % of  Marshall accelerated graduates chose West Virginia after graduation (Petrany and Crespo)².

The University of Nebraska Medical Center (UNMC) accelerated program focused on the needs of counties with less than 10,000 people by adding a rural-oriented procedural fellowship year (3-3-1 program). The UNMC program graduated over 70% into instate rural practices (stageman bowman Harrison).⁵  Without comparisons with other programs, it was not known whether such rates were a result of the specific Nebraska approach, the rural nature of the state, or a function of the accelerated model itself in selections, age of graduates, design advantage, or training. With the addition of other reports and comparisons the UNMC model did not have the highest rural graduation percentage, nor was it atypical compared to other accelerated programs with shorter duration. ⁶

Accelerated programs have not been popular with residency programs not connected to Family Medicine departments. Much of this has been the result of increased competition for students during times of decreased family medicine career choice. Since accelerated students make decisions a year earlier and avoid the "match," the community-based programs have long been concerned about loss of some of the best candidates for a residency position. From the perspective of departments, they have been able to retain some of the best students, stabilizing recruitment and enhancing the quality of the residency training program. When the Accreditation Council for Graduate Medical Education (ACGME) purged a number of hybrid training models there was not a solid base of support. Without adequate data and advocates, accelerated programs were not granted a stay of execution. The unique design of the program allowed a national database to be assembled using secondary data collections.

Beyond local questions and issues within disciplines are some key questions about the value of 3-3 accelerated training programs:  to the trainees, to residency programs, to those who fund residency training, to those who expect improved distribution of graduates, to those who feel that physician distribution is not possible with medical education efforts, and to the communities and populations in most need of family physicians in the United States and perhaps in other nations.

The initial intent of accelerated programs was to offer older students a facilitated training for those with potentially less time remaining to practice. Accelerated training was not designed to address specific workforce needs, however it is unlikely that any graduate medical education intervention will be able to surpass the improved access for those in most need of care.

Methods

The data sources include the American Medical Association Masterfile for 2000 and 2004 ⁷ and Robert Graham Center and American Academy of Family Physician medicine databases. This database was used to identify all graduates with a 2 year gap between residency graduation year and medical school graduation. When such graduates came from a school with a known accelerated program, they were included in the accelerated database. Other data comparisons involve physicians listed in the 2004 Masterfile who graduated from a medical school anywhere in the world from 1987 – 2000.

Birth origins and practice location data were available for over 95 % of physicians involved in allopathic family medicine training. Birth origins were converted to zip code locations. Birth location and practice location zip codes were converted to rural-urban commuting area code (RUCA)⁸. After RUCA coding, the location data was divided into urban-focused, large rural, medium rural, and isolated rural categories for birth origins and practice locations using the method suggested by Hart (Hart)⁹. Where zip codes were unique (medical school, large corporation), the closest zip code with a population was matched to the graduate. When there was no practice zip code the home or other zip code was used. This has provided over 95 % accuracy in other graduate location studies. ¹⁰

Zip code locations were screened to insure that the graduates had left their residency zip location. This tends to exclude the graduates (at least 8) that have chosen medical teaching careers. Teaching careers may not rate as highly as poverty practice locations or rural areas, but they do have shortage designations in many states and family medicine residency zip codes involve populations in poverty service areas often in zip codes with high poverty levels.

Zip code determinations were also made the percentage of the population at or below the poverty level. Zip codes with over 20 % in poverty were coded as poverty locations. This coding may under represent group that tend to have more rural locations, such as accelerated or rural training track programs. Rural areas have greater numbers of working poor, higher rates of underinsurance, and greater use of high deductible insurance that are less supportive of primary care (Mueller).¹¹ Decreasing the poverty cut point to 14 % instead of 20 % increased the differentials between accelerated and traditional graduates.

The determinations regarding use of zip codes for teaching and rural locations therefore tend to make the numbers more conservative than the actual impact considering rural, poverty, and teaching impacts. In an analysis of primary data it is entirely possible that there would be few accelerated graduates remaining that were not involved in some sort of career of service to the underserved.

The results can also be compared using secondary data files with rural and poverty choices for all US physicians by age and other characteristics, family medicine residency programs, medical schools, types of schools, or categories of residency programs.

Proper interpretation of the distribution characteristics of any physician intervention must involve consideration of some important areas. These include gender, age of graduates, and the timing of graduation. Female family medicine graduates usually lag 3 - 5 percentage points behind males in choice of rural location with variation depending upon the type of medical school attended.¹² Older graduates have 1 - 5 percentage points of advantage over those less than 30 years of age at medical school graduation in rural and poverty location.

The timing of the graduates is also most important. Those graduating from medical school from 1995 - 1999 graduating under health policy conditions greatly favorable to primary care. Choices of allopathic medical students were altered in a way greatly different than the past 25 years including increased family medicine choice (+30 %), rural family medicine choice (25%), and office-based primary care choice (+25%) when compared to previous years of graduates. The impact was greater on those born in counties of over 1 million and those born in other countries who had 50 % increase in choice of family medicine and over 30 % increase in rural choice. The overall impact of this time period was to saturate urban and large rural practice locations, facilitating the distribution of all primary care programs to rural and poverty locations. The proper interpretation of accelerated program outcomes, rural training track outcomes, Title VII outcomes, and the recent lifestyle career choice outcomes requires consideration of this aberrant period in US physician workforce history. ¹³

It appears that primary care got to bat, the fully developed primary care training models got hits, and the older and rural born candidates brought in the runs by distributing even better than usual. In the current time period with less and less total primary care and family medicine graduation and fewer distributional students admitted, any distributional intervention is not likely to have the same degree of impact. Indeed the rural location rates have been falling across the nation and across physician characteristic groups for the past 3 years.

Results

There were over 140 graduates that appear to have completed an accelerated training program from 1997 - 2003. The gender distribution was 55 female and 85 male.

Accelerated programs have consistent and superior rural graduation rates. The Marshall program was atypical in rural graduation rates compared to other accelerated programs and also compared to the usual Marshall medical school graduates. Marshall ranks 20^th among all medical schools in the world regarding percentage of total graduates in the US in rural areas (ranking). Marshall had the youngest accelerated graduates of any accelerated program and also may have graduated the most teaching physicians. Marshall accelerated graduates had an 81 % retention in the state ³ which was no different than all accelerated graduates from allopathic public schools.    (note in 2009 - teaching positions have been a transition pattern for those leaving rural practices also, possibly an impact of 6 or more years of training in such locations. Also this still represents rural workforce and also future rural workforce due to well trained practice experienced accelerated grads as teachers. If accelerated models expanded to training in rural areas, the teaching would match up with the rural location of the teacher/clinician and the needs to train medical students and residents and the needs of rural communities for workforce).

The UNMC accelerated program ranked 7^th in the nation for all family medicine residency programs (RTT included) in rural graduation rates at 75 %,⁶ but was still behind the original Kentucky accelerated program (78%) in this category. Interspersed with accelerated programs in the top 50 FP programs in % rural choice are rural training tracks and residency programs in smaller locations. The rural location rates for accelerated graduate in public medical schools in rural states (NC, KY, NE, AL, TN) are comparable to rural training tracks location rates nationwide (Stageman, Bowman, Harrison).⁵ Accelerated programs also contributed a similar number of rural family practice physicians compared to published rural training track numbers ¹⁴ with 50 % of 150 total graduates from 1997 – 2003. The programs had consistency in rural location pattern with only 3 programs below 50 %. Graduates of South Carolina were not identified by the secondary methods used by the author. Data from UNMC for additional years reveals 39 graduates with 74% choosing rural practice. The few graduates leaving rural practice have tended to move to other rural practices or teaching positions.

The following data includes birth location and practice location on 120 accelerated graduates.

Accelerated graduates born in urban areas have 3 times the rural practice location rate compared to traditional graduates of urban birth. Accelerated graduates born in the most rural areas have enhanced rural practice location rates. Urban born students involve over 71 % of US allopathic medical students and in some schools as high as 90 %. The urban  component is growing yearly with the rise of foreign born US medical students. This group has grown from a few percent up to 14 % of allopathic students in recent years. This group is 97 % urban in origin.

Any training model that hopes to succeed in the future must be able to impact the groups increasing them most in medical school admissions, females and urban origin students. The accelerated model has been able to do both.

The overall rural location rate nears 60 % when excluding those still at residency zip codes.

The distribution of accelerated graduates is also important to understand compared to other graduating allopathic physicians. It is important to note that the urban and large rural areas are supplied by family medicine and a number of other primary care disciplines. Few aside from family medicine venture into the medium rural and isolated rural practices, particularly those with only a few physicians or in poverty areas.

Accelerated graduates choose rural locations at 2 - 4 times the rate of usual family medicine graduates at every category of rural location. Only 41.5% chose urban locations despite training for 6 or 7 years at some of the most urban academic locations in the nation.

Note the tendency for those born in a location to return to a similar location (diagonal bold) This is more pronounced in other FP studies.

The total rural graduates from accelerated programs are equivalent to the contributions of the rural training tracks (RTT). RTT programs have an edge in rural location rate with 70 - 80 %, but also have specific rural location and selection bias toward rural born and those interested in rural practice. Accelerated students involved only 29 % rural born, had no rural training locations, and with the exception of the Nebraska program 5, had no selection bias.

Accelerated graduates also had increased location in areas of higher poverty in the nation. For this comparison the zip codes with over 20 % of the population at or below poverty levels were selected. The accelerated graduates had higher choice of urban poverty locations (20 % vs 14 %, female advantage noted) and rural poverty locations (27 % vs 12.7 %). These levels are equivalent to the poverty choices of black (27 %), Hispanic (26 %), and foreign born Asian (22 %) family physicians in the past 3 years. White and Asian FP graduates born in this nation choose poverty locations 14 % of the time. The accelerated graduates were predominantly white in the past 3 years of graduates. Accelerated programs nearly doubled the poverty location rate of white graduates.

Conclusions

The accelerated model as implemented nationwide clearly makes a contribution to efforts to resolve the maldistribution of physicians. Not since the creation of family medicine has the nation had the opportunity to do so much to resolve this chronic problem. More rural training tracks, more programs in rural locations, and more accelerated programs would likely improve physician distribution.

Admissions of more rural, older, and lower income origin students and national health policy toward primary care are also major contributors to physician distribution. Without the recent "perfect storm" of primary care health policy, now long gone, the effectiveness of almost every primary care distributional intervention would have been reduced.

The accelerated model is more and more important with each passing year.

• The model works even as the nation graduates more females and more of urban origins.
• The model offers an end to debates about 3 versus 4 years of graduate training. Six focused years appear to be enough and a 7 year effort can be targeted to specific locations such as urban or rural underserved.
• The model highlights what true primary care can do if it has more and more control of the medical school process from admissions to graduation.

The major limitation of this study is the use of secondary data. Even using multiple sources, not all of the accelerated graduates are likely to be included. Smaller numbers from programs with intermittent activity are missing as well as the entire South Carolina delegation. The data is, however, consistent for coding and timing of practice location and not biased by individual program determinations.

State legislatures may be more than willing to assist programs with a high rate of retention and proven results in graduation of more physicians for rural and poverty areas. The rural graduates of the two private schools, Creighton and Case Western, did not have the same rural location rates or instate retention. Some may have been fulfilling obligations that often require a change of state. The utility of such a dedicated primary care preparation program for National Health Service Corps scholars or others committed to rural and underserved practices is obvious. The accelerated model can also been adapted to the needs of the inner city poverty locations by linking to an urban Community Health Center.⁵

Although 12 accelerated programs endured, another 5 or 6 did not survive.¹ It is not know why these programs closed. Even among the 12, there were some who had few total graduates. Some experts feel that only 3 accelerated programs really worked to develop their model. ¹. There were no specific funding efforts that involved any of the programs. The efforts of a previous faculty development program, the Minifellowship in Rural Family Medicine at East Tennessee State University, aided in the dissemination and development of this model in later replications. RTT programs also had consultants and specific faculty development efforts. It is likely that supportive funding would impact accelerated programs in significant ways. Perhaps the Nebraska model was the most challenging since it involved the additional development of the fellowship year, with additional faculty, locations, and rotations. Without faculty of a variety of specialties who had rural connections, this recruitment would have been difficult. The success of this fellowship year was important to future recruitment efforts. The Nebraska model has adapted to a senior year track. It is not known whether other accelerated programs will survive. The new Edward Via osteopathic medical school in Virginia has developed a 6 year combined model. The flexibility of osteopathic medical education plus increased numbers of rural and older students may provide even greater results.

The accelerated model managed to negate the influences of age and gender on rural practice location. The specifics of selection and training should be a major focus of investigation for those truly interested in distributing physicians where needed. The rising awareness of socioeconomic characteristics of students and choice of family medicine and rural practice may also be a worthwhile theme to explore. Those of more humble origins are less likely to be admitted to medical school but are more likely to distribute where needed as physicians (Birth Origins).¹³ Accelerated programs involve a more specific focus on primary care and family medicine. Trainees have active "hands-on" patient care responsibilities earlier compared to their fellow medical students. The success of the programs brings larger questions about selection and training methods that would result in more office-based primary care physicians for rural, underserved, and poverty locations.

It is always difficult to discern the contributions of selections or training. Those ready for a career decision a year earlier are likely the ones who have always known what they want to do. Studies show that those interested in rural practice as senior medical students were twice as likely to have known that they were interested in family medicine as a career (30%) as compared to their peers (15%) prior to medical school. In this group of rural-interested and rural background students (RBI), 60% were interested in serving the underserved as compared to 10% of their 13000 senior medical student peers (Bowman AAMC GQ 1995)¹⁵. This prior commitment to rural and family medicine is also noted in studies at Duluth (Boulger)¹⁶ and in the graduates of the Jefferson Physician Shortage Area program (Rabinowitz) ¹⁷. The match of RBI students and specific rural medical education (RME) training is powerful. With supportive national health policy in primary care and admissions of a broader range of candidates to medical school, maldistribution can be defeated or at least taken into extra innings.

Even though it appears that background and selection bias may be a large part of the rural outcomes, there is also the potential that the design of the accelerated program has influence. The back-to-back placement of the two most critical learning years (M-3 and PGY-1) could "accelerate" medical education such that graduates may be more comfortable with the choice of challenging locations and practices. The enhanced outcomes of the Nebraska program regarding isolated rural location may be a good indicator of what might improve the outcomes further. The flexibility of the fellowship year also allows graduates to "tailor their training" to the needs of their location and work with consultants who are already in place near such locations. During the fourth graduate year of training, a designation as a "fellow" has also opened doors for advanced training that are usually shut for even senior family medicine residents. Having 7 years of training in a 3 - 4 configuration may also end the current debates about 4 - 4 or having all family medicine residents train for 4 years.

Comparisons of rural and urban background and measures of competence and confidence may reveal other benefits of a curricula that is focused on family medicine earlier and longer than most. The Rural Physician Associate Program (RPAP) at the University of Minnesota involves 9 months of training in a rural family practice during the third year of medical school. It is much "purer" in terms of family medicine orientation and has far more family medicine content when compared to almost any other form of medical education. RPAP also has superior evaluations regarding behavioral, cognitive, and procedural areas (Verby ) ^18,19. It also has high rural graduation rates and has graduated a number of family medicine leaders and teachers. It has also distributed physicians well  and improved rural economics across Greater Minnesota to the tune of 2.2 billion dollars in economic impact so far.²⁰  The accelerated programs are showing such promise, even in the early years of the program.

The development of family medicine as a discipline may be greatly hampered by curricula that is grossly deficient in primary care, family medicine, and "hands-on" training. A more aggressive approach may students to fulfill their maximal potential. Adjusting the amount of training, the timing, and the priority of primary care training may result in graduates with the great workforce flexibility demonstrated by accelerated graduates. Such graduates have adapted to the widest range of primary care practice in geographic and socioeconomic dimensions. Such a program can allow graduates to adapt their training to fit a wide range of population-based, procedural, and behavioral needs. This might also improve the potential for retention.

Eliminating an accelerated model with superior outcomes, one that is perhaps the "most family medicine" of all curricula, one that has achieved much of what family medicine was designed and funded to do, has been a bad decision for the country.

The model should return from the ashes even stronger and more able to serve the country. The model could move forward a year as a 2-4 model by integrating a long term preceptorship such as RPAP for an enhanced rural effort. Accelerated graduates could finish 3 years of medical school and do an initial graduate year (3 - 1 - 2) before moving out to rural training track programs or community based program, a move that might be politically correct and could facilitate training as well.

References

1. Tolleson, Joe   Personal Communication

2. Bratton RL, David AK The University of Kentucky's Accelerated Family Practice Residency Program   Family Medicine 1993 Feb; 25(2):107-10

3. Petrany SM Crespo R, The Accelerated Residency Program: The Marshall University Family Practice 9-year Experience, Family Medicine 2002; 3 4(9):669-72

4. Galazka SS, Zweig S, Young P. A progress report on accelerated residency programs in family practice. Acad Med 1996;71:1253-5

5. Stageman JH, Bowman RC, Harrison JD. An accelerated rural training program. J Am Board Fam Pract. 2003 Mar-Apr;16(2):124-30.

6. Ranking Medical Schools and Residency Programs http://www.unmc.edu/Community/ruralmeded/ranking_rural.htm

7. American Medical Association Masterfile Medical Marketing Service, Inc; 2000 and 2004,

8. Rural urban commuting area coding http://www.fammed.washington.edu/wwamirhrc/rucas/descript.html

9. Hart's method 1 used, described at  http://www.fammed.washington.edu/wwamirhrc/rucas/00C8994E-005B90E7.-1/use_healthcare.html

10. Krist AH, Johnson RE, Callahan D, Woolf SH, Marsland D   Title VII Funding and Physician Practice in Rural or Low-Income Areas, Journal of Rural Health, Winter 2005 21 1 3-11

11. Mueller K Comer, J., & Mueller, K. (1995). Access to health care: Urban-rural comparisons from a midwestern agricultural state, 11(2), 128-136

12. Bowman RC Ethnicity, Gender, and Choice of Rural Practice http://www.unmc.edu/Community/ruralmeded/ethnicity_gender_and_rural_pract.htm

13. Birth origins of allopathic physicians and choice of family medicine http://www.unmc.edu/Community/ruralmeded/birth_origins_and_fp_choice.htm

14. Rosenthal T Outcomes of Rural Training Tracks: A Review--Tom Rosenthal, M.D. The Journal of Rural Health Volume 16, No.3, Summer 2000

15. Bowman RC  Characteristics of Rural Interested Senior Medical Students in 1995 at http://www.unmc.edu/Community/ruralmeded/model/medsch/aamcdat.htm

16. Boulger JG. Family medicine education and rural health: a response to present and future needs. J Rural Health. 1991;7:105-115.

17. Rabinowitz HK, Diamond JJ, Markham FW, Paynter NP, Critical factors for designing programs to increase the supply and retention of rural primary care physicians, JAMA 2001; 286: 1041-1048

18.  Verby JE, Schaefer MT, Voeks RS   The impact of a long term preceptorship on clinical confidence of senior medical students    Minnesota Medicine May 1982  p 297- 300

19.  Garrard J, Verby JE Comparison of medical student experiences in rural and university settings Journal of Medical Education 52:802 1977

20.  Bowman RC, Crump WJ   Impact of the Rural Physician Associate Program: External Consultation for the University of Minnesota, 1998

Other References related to subject

Weintraub W, Balis G, Mackie J. The combined accelerated program in psychiatry: a progress report. Am J Psychiatry 1974;131:1213-6.4.

Thompson JS, Haist SA, DeSimone PA, et al. The accelerated internal medicine program at the University of Kentucky. Ann Intern Med1992;116:1084-7.5. Coleman C. Traveling the nontraditional path. AAMC Reporter 1999Aug;8(11).6.

State of West Virginia. West Virginia Health Care Authority. The West Virginia State Health Plan, volume 1, 2000– 2002. Charleston, WV:West Virginia Health Care Authority, 1999.7.

Denton DR, Cobb JH, Webb WA. Practice locations of Texas family practice residency graduates, 1979–1987. Acad Med 1989;64:400-5.8.

Dorner FH, Burr RM, Tucker SL. The geographic relationships between physicians’ residency sites and the locations of their first practices. Acad Med 1991;66:540-4.

Decreasing Rural FP Physicians

Rural Background

Gender and Ethnicity in FP Graduates 1997-2003

www.physicianworkforcestudies.org

www.basichealthaccess.org

www.ruralmedicaleducation.org