|
|
| |||||||||||||||||||||||||||||||||||||||
Background To help address the geographic and specialty maldistribution of physicians, Jefferson Medical College initiated the Physician Shortage Area Program (PSAP) in 1974. This unique program, which combines a selective medical school admissions policy with a special educational program, has been shown to be successful in increasing the number of family physicians in rural and underserved areas, but it is not known whether they remain in this type of practice.
Methods Graduates of the PSAP were tracked longitudinally and compared with their non-PSAP classmates. Information was obtained about the retention of family physicians in rural areas and areas with a physician shortage over the previous five years, the geographic and specialty choices of more recent graduates, and the recruitment of applicants into the program.
Results Of the 47 PSAP graduates from the classes of 1978 through 1981, reported on earlier, the number who combined a career in family medicine with practice in a rural area or one with a physician shortage remained unchanged, although there was substantial attrition among non-PSAP graduates practicing family medicine in rural (32 percent) and underserved (40 percent) areas. Among the 101 PSAP graduates of the classes of 1982 through 1986, the results were similar to those for the first four classes.
Overall, PSAP graduates from the classes of 1978 through 1986 were approximately four times as likely as non-PSAP graduates to practice family medicine (55 percent vs. 13 percent), to practice in a rural area (39 percent vs. 11 percent), and to practice in underserved areas (33 percent vs. 8 percent). They were approximately 10 times more likely to combine a career in family medicine with practice in a rural (26 percent vs. 3 percent) or underserved (23 percent vs. 2 percent) area. Overall, 85 percent of PSAP graduates were either practicing a primary care specialty or practicing in a rural or small metropolitan area or one with a shortage of physicians.
In parallel with national trends, the number of applicants and matriculants to the program decreased during the past decade, so that the percentage of available places filled decreased from 97 percent to 33 percent. However, there has been a recent increase in the number of applicants and matriculants.
Conclusions The results of this study indicate that the PSAP was successful in increasing the number of family physicians in rural and underserved areas as well as in retaining them. This suggests that medical schools can have a substantial influence on the distribution of physicians according to specialty choice and the geographic location of their practices, principally through admission criteria.
The problem of providing family doctors to rural areas is particularly great for the state of Pennsylvania. Although Pennsylvania is well known for its major metropolitan areas of Philadelphia and Pittsburgh, the U.S. Census Bureau ranks it the most rural state in the nation, on the basis of its having the largest rural population of any state (a rural area is defined as a non-urbanized area with fewer than 2500 people)8. In addition, Pennsylvania has a severe maldistribution of physicians, with more than half of all the physicians in the state located in just 3 counties (Philadelphia County, its suburban Montgomery County, and Pittsburgh's Allegheny County), even though the remaining 64 counties have almost three quarters of the state's population9.
To help correct this geographic and specialty maldistribution of physicians, Jefferson Medical College of Thomas Jefferson University initiated the Physician Shortage Area Program (PSAP) in 1974. This program, which has been described previously,10,11 specially recruits and admits medical school applicants from rural backgrounds who intend to practice family medicine in rural and underserved areas. The program also gives these students more financial aid than is usually awarded to Jefferson students and provides them with a special family medicine program, including a family medicine faculty advisor, a required third-year clerkship in family medicine at one of two nonmetropolitan locations, and a senior outpatient subinternship in family medicine, frequently consisting of a preceptorship with a rural family physician.
Early evaluation showed that PSAP graduates were almost 5 times as likely as their non-PSAP classmates to enter family medicine (60 percent vs. 13 percent), more than 3 times as likely to practice in rural areas (42 percent vs. 12 percent), and 7 to 10 times as likely to combine a career in family medicine with practice in a rural or underserved area11. Despite the success of this program, however, a number of questions remain. First, the previous evaluation of the PSAP followed Jefferson graduates for five to eight years after graduation from medical school, so that most of these physicians were in practice for only two to five years. Whether these PSAP graduates have continued their careers as rural family physicians remains an important question. Second, the original evaluation of the PSAP reflected the specialty and geographic career choices of a relatively small number of physicians (47) who graduated more than a decade ago (the classes of 1978 through 1981). It is important to determine whether more recent graduates of Jefferson's PSAP are continuing to fulfill the goals of the program. Finally, the national applicant pool for medical school has recently shown a disproportionate decline in applicants and matriculants from rural areas and small towns, as well as those interested in family medicine12,13. How these changes have affected the PSAP applicant pool, the number of matriculants, and their academic performance is critical to the future of the program.
To address these issues of recruitment and retention, we undertook a study to obtain follow-up information from all PSAP graduates and their non-PSAP classmates from the 19 classes (1978 through 1996) that have participated in the program.
Methods
Data on the current practice location (county and state) of PSAP graduates and their classmates from the first nine classes (1978 through 1986) were provided by the Alumni Association of Jefferson Medical College in October 1991. As previously described, American Medical Association county group codes (1978), which measure the population density of a county by assigning a value of 1 to the least metropolitan counties and a value of 9 to the most metropolitan, were identified for each county11. The alumni data were then merged with the same county variables from the 1986 Area Resource File of the U.S. Department of Commerce, used in the previous study. Because that study showed similar results with the use of two definitions each for "rural areas" and "physician-shortage areas," this study used only non-Standard Metropolitan Statistical Area counties (those with county codes 1 through 5) to identify rural counties. Likewise, a county was considered to be a physician-shortage area only if the entire county was a shortage area according to Health Manpower Shortage Area Codes for Primary Care (1980) or if a portion of the county was a shortage area and the county was also a rural (non-Standard Metropolitan Statistical Area) county11. For this study, a graduate's current specialty was considered to be that in which he or she had obtained board certification (American Board of Medical Specialties, 1990). For graduates who were board-certified in two or more specialties or who were not board-certified in any specialty, primary self-reported specialty data from the American Medical Association Physician Masterfile (1990) (or if not listed, Jefferson alumni self-reported data on specialties) were used. Graduates with subspecialties were categorized according to their primary specialties; those practicing a subspecialty of internal medicine, pediatrics, or surgery were referred to as subspecialists of their primary specialty.
PSAP and non-PSAP graduates were then compared to determine whether they were practicing in a rural or physician-shortage area and whether they were practicing family medicine or primary care (family medicine, general internal medicine, and general pediatrics). The current specialty of each Jefferson graduate was combined with his or her practice location, and PSAP and non-PSAP graduates were then compared to determine whether they were practicing family medicine in a rural or underserved area, thereby fulfilling the goals of the program. Because the entire population of PSAP and non-PSAP graduates was included in this evaluation, inferential statistics were not needed to compare the groups.
The number of applicants and matriculants to the PSAP from the classes of 1978 through 1996 was obtained from the Office of Admissions of Jefferson Medical College. Data on the sex, credentials before medical school admission, academic performance, and postgraduate specialty choice of PSAP students and their peers in the 14 classes from 1978 through 1991 were retrieved electronically from the data base of the Jefferson Longitudinal Study14. The students' credentials before their admission to medical school and their academic performance during medical school included previously reported variables11. Postgraduate performance was assessed by a rating system described elsewhere,15 in which data-gathering and data-processing skills, interpersonal skills and attitudes, and socioeconomic aspects of patient care were measured. Postgraduate performance was assessed only after consent was obtained from Jefferson graduates. For each of these variables, the effect size (the ratio of the difference in means to the standard deviation of the values for non-PSAP students) was used to describe differences between PSAP graduates and their peers16.
Results
Retention of Graduates from the Classes of 1978 through 1981
The number and percentage of PSAP and non-PSAP graduates from the classes of 1978 through 1981 who were practicing family medicine or who were practicing in a rural area or one with a shortage of physicians remained relatively stable from 1986 to 1991 (Table 1). This resulted in PSAP graduates' remaining approximately four times as likely as non-PSAP graduates to practice family medicine or to practice in a rural or underserved area.
|
Among the graduates of the classes of 1978 through 1981 who were practicing internal medicine or pediatrics in 1986, there was also substantial attrition. Four of the seven PSAP graduates practicing internal medicine or pediatrics in 1986 (none of whom were in a rural or physician-shortage area) were practicing a medical or pediatric subspecialty by 1991. Similarly, of the 302 non-PSAP graduates practicing internal medicine or pediatrics in 1986, only 47 percent were practicing general internal medicine or general pediatrics in 1991.
Overall, 87 percent of the PSAP graduates of these classes (40 of 46) continued either to practice in a rural or small metropolitan county (county code 6) or one with a shortage of physicians or to practice one of the three primary care specialties, a slight decrease from the 93 percent (42 of 45) reported in 1986.
Geographic and Specialty Choices of More Recent Graduates (Classes of 1982 through 1986)
The 101 PSAP graduates of the classes of 1982 through 1986 were approximately four times as likely to practice family medicine (55 percent vs. 14 percent) or to practice in a rural area (38 percent vs. 9 percent) or one with a shortage of physicians (30 percent vs. 7 percent) as non-PSAP graduates. In addition, PSAP graduates were 8 to 10 times as likely as their peers to combine a career in family medicine with practice in a rural area (24 percent vs. 3 percent) or one with a physician shortage (20 percent vs. 2 percent). Although this pattern is similar to that previously reported for graduates of the classes of 1978 through 1981, there has been a slight decrease among both PSAP and non-PSAP graduates in the absolute percentages practicing in rural or underserved areas and practicing family medicine in these areas.
Overall Evaluation of the PSAP
Of the 214 PSAP students admitted into the first 14 graduating classes of 1978 through 1991, 206 graduated, 2 transferred, and 1 was still enrolled as of 1992. This level of attrition (2 percent) was similar to that of the non-PSAP students (3 percent). The proportions of male and female PSAP students (77 and 23 percent) were similar to those of their classmates (75 and 25 percent).
Overall, 81 of the 148 PSAP graduates (55 percent) of the nine graduating classes of 1978 through 1986 were practicing family medicine, 58 of 147 (39 percent) were practicing in rural areas, and 48 of 147 (33 percent) were practicing in areas with a shortage of physicians. In each instance, this is approximately four times the percentages for non-PSAP graduates (Figure 1). PSAP graduates were almost 9 times as likely as their classmates to combine a career in family medicine with practice in a rural area (26 percent vs. 3 percent) and more than 11 times as likely as their peers to practice family medicine in an area of physician shortage (23 percent vs. 2 percent). When analyzed according to sex, these patterns remained unchanged.
|
When the data on Pennsylvania were examined alone, PSAP graduates of the classes of 1978 through 1986 were much more likely to be practicing in the state than were non-PSAP graduates (67 percent vs. 39 percent). As shown in Figure 2, among the Jefferson graduates currently practicing in Pennsylvania, PSAP graduates have preferentially settled in the rural counties in the state. Finally, although almost half of Jefferson's non-PSAP graduates (44 percent) who were practicing in Pennsylvania were located in 1 of the 3 "over-doctored" urban or suburban counties, the vast majority of PSAP graduates (88 percent) were located in the other 64 counties.
|
For the graduating classes of 1978 through 1981, up to 12 places in each class of 223 students were reserved for PSAP students. In 1982, a PSAP Cooperative Family Medicine Education Program was begun with Indiana University of Pennsylvania, and the number of places for PSAP students increased to a maximum of 24 per year (with no change in the overall size of the class). During the initial eight years of the PSAP, the number of applications to the program decreased dramatically, from 305 for the graduating class of 1978 to 121 for the class of 1985. During the next 10 years, the number of applications averaged only 63 per year, although there has been a steady increase in applications over the past 5 years. For the most recent class (1996), there were 94 applications to the PSAP, the highest in the past decade. Similarly, the number of matriculants in the PSAP, which approached the maximal number of available places (97 percent) during the first eight years of the program, has decreased significantly over the past decade, with a mean of 13 matriculants in the graduating classes of 1986 through 1990 (55 percent of available places) and a mean of 8 for the classes of 1991 through 1995 (33 percent of available places). The recent increase in the number of matriculants in the class of 1996 (19) has led to the highest number in the past 10 years and may indicate a reversal of this trend.
Despite the decline in the number of applicants and matriculants to the PSAP, the premedical grade-point averages of PSAP students from the graduating classes of 1986 through 1991 remained similar to those of their classmates, although their mean scores on the Medical College Admission Test were somewhat lower (Table 2), a pattern similar to that previously reported for the classes of 1978 through 198510. Nevertheless, as shown in Table 2, the academic performance of PSAP students from the classes of 1986 through 1991 during medical school and postgraduate training was similar to that of their peers (all effect sizes were 
0.2), which is again similar to results reported for previous classes11.
|
Over the past few decades, a remarkable consensus has emerged among health care experts that increasing the number of primary care physicians in underserved areas is a critical goal for the United States1,2,4,5,6,17,18. However, no similar agreement has emerged on how to accomplish this important task. Eighteen years ago, Jefferson Medical College began the PSAP, a unique program designed to address one major aspect of this problem, increasing the number of family physicians in rural areas, especially in Pennsylvania. The results presented in this paper confirm and extend previous results, indicating that graduates of the PSAP continue to be much more likely than their peers to enter family medicine in rural and underserved areas; their high level of retention in these areas is in sharp contrast to the rates for other programs, such as the National Health Service Corps19. These results indicate that the PSAP represents one successful way of addressing the maldistribution of physicians according to specialty and geographic location. The results are even more dramatic when one considers that there was no formal mechanism to ensure compliance with the goals of the program.
The ability of this or other similar programs to solve the overall lack of primary care physicians in rural and underserved areas of the United States also depends on an adequate pool of candidates. In view of this consideration, the dramatic decline during the 1980s in the number of applicants and matriculants to the PSAP represents a disturbing trend. Although this decline was greater than the overall decrease in applicants to medical schools nationally and to Jefferson specifically, it was similar to the national decline in the number of applicants planning careers in family medicine and those planning to practice in rural areas6,7,12,13. Since there were no changes in the admissions process or the level of institutional support for the PSAP during this time, it is likely that the decline in applicants and matriculants merely paralleled these national trends. To address this problem, Jefferson expanded its PSAP Cooperative Family Medicine Education Program in 1990 to five other undergraduate colleges (Allegheny College, Bucknell University, Franklin and Marshall University, Pennsylvania State University, and the University of Scranton) in addition to Indiana University of Pennsylvania to help recruit students at the high-school and college levels. Although the recent increase in applications to the PSAP has also followed national and local trends, the increase in PSAP matriculants surpassed these trends, partly because of this expanded program.
Because the various components of the PSAP (admission, financial aid, advising, curriculum, practice sites, and role models) are not entirely controlled and have not been systematically studied, it is difficult to assess their specific contributions to the success of the program, although all these factors appear to be important. However, the selective admissions component appears to be the most powerful factor, since national data indicate that students with a rural background or interest in family medicine who enter medical school are many times more likely than their peers to become family physicians or practice in rural areas (and those not interested in family medicine or rural practice are highly unlikely to change their career choice to family medicine)20,21.
Although the issue of selective medical school admissions to improve the geographic distribution of physicians remains controversial, it is not a new idea; in fact, most state schools provide preferential admission for their in-state residents for exactly this purpose. On the other hand, because some applicants from rural areas look forward to leaving these areas, the PSAP chooses only applicants from nonmetropolitan backgrounds who specifically want to return to that type of environment eventually and to practice the specialty of family medicine. Although we have used the term "nonurban" to describe the background, culture, and interests of the applicants most likely to achieve the goals of the program, these students are almost always able to assimilate into the diverse group of their Jefferson classmates. In terms of the choice of specialties, although studies show that medical students frequently change their minds about which specialty they will choose, data also indicate that most students usually see themselves as either generalists or specialists and that a student's career choice of family medicine is more stable and consistent than that of other specialties12,20,22.
In contrast to the admissions and educational portions of the PSAP, the financial-aid component appears to have had the least influence on the success of the program. Although PSAP students obtained more financial support than if they had not been in the program, the total dollar amount was small (averaging only a few thousand dollars per student per year) and was almost entirely in the form of repayable loans. On the other hand, instituting a guaranteed loan-repayment program may have the potential to influence who practices family medicine in rural and underserved areas23.
One limitation of this study is the possibility that some PSAP students might have entered medical school and chosen to practice family medicine in rural and underserved areas even without the program. As was true of the previous study of graduates of the classes of 1978 through 1981, almost two thirds (63 percent) of more recent PSAP students (the classes of 1982 through 1986) were not accepted by any medical school other than Jefferson, according to data from the Association of American Medical Colleges Joint Acceptance Reports. Review of the admission credentials of all Jefferson matriculants during these years suggested that a similar percentage would probably not have been accepted at Jefferson without the PSAP.
In summary, the results presented in this paper indicate that Jefferson Medical College, a large, private, urban medical school located in the northeastern part of the United States (factors traditionally associated with a low output of primary care and rural physicians) has been successful in increasing the number of family physicians practicing in rural and underserved areas as well as retaining physicians in those areas. Jefferson has accomplished this through a strong institutional commitment to these goals, while at the same time balancing its role as a tertiary care medical center in a large urban area with a strong research agenda. Other programs have also achieved similar goals,24,25,26,27 suggesting that medical schools can have a substantial influence on the geographic and specialty distribution of their graduates. Medical schools cannot solve the national problem of the maldistribution of physicians by themselves, however. Changes are also needed in the approach to graduate medical education, the remuneration of primary care physicians, the process of loan repayment, and the practice environment2,12,17,18.
I am indebted to James J. Diamond, Ph.D., for help in statistical evaluation and for assistance in reviewing the manuscript; to Adrienne Martin, B.S., and Carol Rabinowitz, B.A., for help in programming; to Michelle Auciello, B.A., for providing data from the office of admissions; to Paul C. Brucker, M.D., Joseph S. Gonnella, M.D., Benjamin Bacharach, M.D., J. Jon Veloski, M.S., John L. Randall, M.D., and Hyman Menduke, Ph.D., for their support of the PSAP and their helpful comments on the manuscript; and to Rhoda Abrams for assistance in the preparation of the manuscript.
Source Information
From the Department of Family Medicine, Jefferson Medical College of Thomas Jefferson University, Suite 401, 1015 Walnut St., Philadelphia, PA 19107, where reprint requests should be addressed to Dr. Rabinowitz.
References
| |||||||||||||||||||||||||||||||||||||||
Related Letters:
Problems in Primary Care
Shurin S. B., Distenfeld A., Zucker R. S., Zucker J. G., Marmon L. M., Czegledy F., Miele A., Fine A. L., Yealy D. M., Plotzker R. M., Siwek J., Pearson H. A., Stockman J. A., Kieliszek F. X., Petersdorf R. G., Fitzgerald F., Levinsky N. G.
Extract |
Full Text
N Engl J Med 1993;
329:728-732, Sep 2, 1993.
Correspondence
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |
Previous
