Choice of Family Medicine: Past, Present, and Future

Robert C. Bowman, M.D.

The new education must be less concerned with sophistication than compassion…it must teach man the most difficult lesson of all - to look at someone anywhere in the world and be able to see the image of himself.  Norman Cousins, Anatomy of an Illness   via John West MD Nebraska Academy of Family Physicians president

Being able to share that image is a critical area for teachers, doctors, and all who interface with people. The situation for physicians has become much more complicated. For all physicians other than family medicine, seeing the image is a difficult process. No other specialty has the wide variety of differences covered so that patients can find a physician most like them. That is because family physicians have origins most similar to patients, in geography, diversity, income level, and other key areas that improve patient interactions and can enhance health care or the perception of health care.

To understand how family medicine choice varies over birth origins, age, medical school type, and other factors, try

Multidimensional Choice of Family Medicine

Choice of Family Medicine Regression

Birth Origins and Distribution Tables

Birth Origins and FP Choice

Driving Difficulty or Distinction

MCAT and FP 2005

Summary see Short and Sweet on FP Choice or   Choice of FP Update, Progress beyond the Arizona Study

Introduction

Physicians need to be intellectual and practical, sophisticated and personal, but what happens if there are not enough with all of the qualities needed? Can physicians retain humanistic qualities in a medical world increasingly dominated by intellectual testing to get admitted, to survive basic sciences, to graduate, and to get the specialty of choice. If someone is intellectual, can they be practical?

If we have a system that selects for the intellectual and sophisticated, is this assisting us with the problems that we have in medicine, or bringing additional complications? Regardless, it is a system that we all have created, and can change. In this nation there is far too much blame and not enough understanding. Setting priorities is important rather than allowing "the world" to make the changes. Important areas such as education and health care are far too critical to allow them to move in directions not helpful for our nation, or the world. Understanding Higher Education and Income

We need strong leaders for a challenging time. There are no more important weapons in the fight against hopelessness and terrorism than education and health care, particularly for those in most need of it today.

Family and General Practice in the Past Century

Family medicine makes important health care contributions in the United States. No other type of physician contributes as much to improve access to care 1 and to serve rural and underserved areas. 2 3 New studies confirm the long standing association between family physicians and reductions in health care costs and increases in the quality of health care. 5, 6  Decreases in family physicians would compromise areas of national interest. 7

Family medicine, as measured by the choices of US medical students, has had a consistent 30 year decline. Only a brief period of popularity stimulated by managed care efforts interrupted this decline from 1994 – 1998, with continued decline since. The consistent nature of this decline does not suggest a recent problem. Nor does it indicate the swing of a pendulum. Without significant numbers of physicians imported from international medical schools, the situation would be even worse. The managed care panic years and the contributions of foreign medical schools have hidden the magnitude of the changes in US medical schools.

A precipitous drop in the number of US medical students choosing general practice in the 1950s and 1960s resulted in a series of reports and the creation of "family medicine" as a specialty in 1969. Initially family medicine gained state support and medical schools grudgingly accepted the new specialty. However in more recent years the new specialty has been eclipsed by massive increases in total numbers of US physicians and the more growth of the US population. Medical school priorities have also shifted away from teaching and primary care in favor of research and specialty care. Looking back a few more decades, family and general practice physicians have been in a steady decline for the past century in the US. 8

See changes in US physicians, FP

http://www.unmc.edu/Community/ruralmeded/admissions_and_origin.htm

Adapted from COGME and Colwill using total physicians and population instead of ratios

There have been repeated efforts to increase the choice of family medicine by US medical students. The most recent effort is the Future of Family Medicine project.9  The one area that is indeed critical to the future of family medicine is the selections process. Selections has yet to become a significant focus of family medicine or national efforts. This is puzzling since selections has been the cornerstone of every successful effort to graduate more family physicians at the medical school, state, or program level. 

Selections: Do US Medical Schools Admit Students Who Will Choose Family Medicine?

Birth Origin

A consistent element in the choice of family medicine has been admissions of students from small towns, particularly those who have interest in family medicine at matriculation. Medical schools, special admission tracks, and statewide efforts concentrating on students from rural background have all graduated more family physicians with success rates as high as 50 %. 10, 11, 12, 13. These programs remain models that are largely unreplicated. Medical schools have not adopted changes that would increase the numbers of family physicians.

There is a general impression that medical schools are indeed attempting to admit students who  "want to become family physicians and go to a small town to practice." The importance of rural selections has been highlighted. Without special admissions significant numbers of rural background students would not have even reached the interview stage  (Basco) Some 47 medical schools profess a policy of rural emphasis in rural selections. These have been documented in annual reports regarding medical education (JAMA Barzansky).

These rural admissions preferences are not confirmed by reality. Data from the Association of American Medical Colleges (AAMC) data reveals major decreases in rural background students admitted to US allopathic medical schools, from 27 % in 1983 to 16% in 1999.14 Yearly reporting has hidden longer term trends. The magnitude of the change using AAMC data was a 55% decrease in white rural background students in all medical schools from 1983 - 1999.

Such an important finding needs a confirmation by a different method, if possible. AMA Masterfile data includes birth city and state on over 600000 US physicians. Studies using the birth origin of physicians confirm this decline, revealing a consistent and steady decline in those admitted to medical school who were born in rural areas. Using the birth origin data, the average decrease in each medical school regarding admissions of students born in rural areas was 43.4 % from 1976 - 1980 as compared to 1996 - 2000 (Bowman birth origin).

Did any group of medical schools demonstrate rural admissions preference. Review of the data for individual medical schools notes that the decline was apparent even in schools with a rural mission, osteopathic schools, and even schools in states with a higher percentage of rural population. Only 2 medical schools managed to show even a small gain in rural birth admissions from 1976 – 1980 as compared to 1996 - 2000. The average decline for all US medical schools was 43.4%. (Bowman Birth Origin) This is comparable to the rural background declines calculated from AAMC data. The consistency in birth origin and the choice of family medicine is even more remarkable.

Family Physicians Are Born, Not Made

Studies demonstrate that family physicians are different, even from other primary care physicians. "Family physicians were more likely to have made their career decision before medical school, and were more likely to have come from inner-city or rural areas. Personal values and early role models play a very important role in influencing their career choice." (Comparisons Among Three Types of Generalist Physicians: Personal Characteristics, Medical School Experiences, Financial Aid, and Other Factors Influencing Career Choice  XU G.[1]; VELOSKI J.J.[1]; BARZANSKY B.[1]; HOJAT M.[1]; DIAMOND J.[1]; SILENZIO V.M.B.[1]  ).

The choice of family medicine by students born in various urban and rural locations has been remarkably steady from 1976 – 2000, with the exception of the 1994 - 1998 managed care impact years. This is data using RUCA coding applied to the birth city and state of physicians.

The consistency is also demonstrated in various urban categories as well. Students from the most urban locations in the nation choose family medicine at lower rates as compared to students from slightly less urban locations. 15 (Bowman Birth Origin One pager).

Birth Location of US Physicians Graduating After 1975

Updated Data

AMA Masterfile and Robert Graham Center, Birth Coding by RCB

Birth origins can divide students into those likely to distribute or not.

The choice of family medicine seems to be more related to student origins and experiences before medical school than what happens after admissions. The greater numbers of urban students results in more family physicians from urban origins. The rural contribution has diminished greatly over time, but rural born students contribute a higher percentage.

Changes in Matriculants

The rural origin group has been replaced by urban students who are much less likely to choose family medicine.

See graphic on matriculant changes at web site

http://www.unmc.edu/Community/ruralmeded/admissions_and_origin.htm

Allopathic US Medical Student Admissions, FP Choice, Income Levels

*Income level of Mexican American parents used. Other Hispanic incomes are higher and FP choices lower.

Black, rural (mostly white), and Native males admitted at even lower ratios.

Only white, Mexican American, and Native groups have any appreciable percentage born in rural areas and this is reflected in their choice of rural practice locations.

Those with the highest income levels by urban, ethnicity, and population density considerations are the most likely to be admitted and the least likely to distribute to rural and poverty primary care locations. In blacks, Natives, and Mexican Americans there are more females that choose rural locations compared to males. This is a reflection of who gains access to college and medical school. This female dominance also includes some residency programs in states with high poverty levels. The effect of poverty seems to penalize male changes of gaining college and medical school admission. Black males and rural males are admitted in the lowest levels, share some of the best distribution when admitted, and share other important characteristics such as income and first time college attendance.

New studies define the differences further in the Asian subgroups (Medicine, Education, and Social Status)

Asian US medical school graduates choose FP at 7.0 %. Other US Medical student choices (most recent US graduates and last 3 years of FP graduate data from 2001 - 2003.)

Birth Origin and Ethnicity

Includes only those with birth city and ethnicity in AMA Masterfile, basically US citizens graduating from US and International Medical Schools and choosing FP

When reviewing the following it is important to remember that urban origins are far more common in non-white as compared to white. Also there is great variation in socioeconomic status that seems to have more to do with FP choice   Medicine, Education, and Social Status, although rural locations are a concern as well.

US Medical School Expansions: No Help for Family Medicine Workforce

Throughout the early 1900s the US greatly reduced the number and size of medical schools. Concerns regarding the supply of family physicians and total physicians resulted in increased numbers of medical schools and medical students. There was little attention to the types of students admitted or their origins. The expansions of medical schools and class sizes in the past 40 years have involved medical students born in urban areas.

Cost, Quality, Access, and Physician Workforce Expansion

Probability of rural born admissions decreased here too, even with expansions.

Some might attempt to attribute this change to demographics or to fewer rural students applying. Non-metropolitan population grew 10% while metro population grew at 16% from 1990 – 2000. The changes in admissions are much greater than demographics. There was also no major change in applications of students from rural background. The percentage of rural background students taking the Medical College Admission Test remained 16 % in both 1991 and 1999 with 30 % not giving their rural or urban background in both years (AAMC data).

Alternate or Matriculant: Points and Polish

In more recent years the MCAT and GPA of applicants has been rising. The mean scores for the national applicant pool now exceeds minimal requirements for prediction of USMLE Step 1 success. This means that most candidates could become physicians if admitted. Those who are more likely to choose family medicine are available and have demonstrated the academics necessary. Health professional advisors note that their greatest challenge involves assisting prospective medical students. Admissions to medical school are much less predictable than other health professions areas. (NAAHP meeting, consensus of discussions).

The Influence of the Medical College Admissions Test

The most influential component of the admissions process is the Medical College Admission Test score. This test significantly reduced the attrition rate many decades ago. The primary design of the test has been prediction of graduation. Currently the focus is on USMLE 1 scores. The primary component still remains multiple choice testing involving sciences and verbal ability. A multimillion dollar industry has arisen to assist potential medical students with their MCAT scores. This makes the design and interpretation of the MCAT a more difficult task (personal communication ellan julian) 

The admissions process has also become more complex. Medical school admissions committees face increasing pressure to admit students with higher MCAT scores. The pressures involve legal issues, accreditation, the cost of education, remedial resources, student cost considerations, school reputation, and competition among schools for research dollars. Perhaps the most important concern regarding students with lower MCAT scores is the possibility that they might fail or have delayed graduation. Admissions committees are most sensitive to the possibility that a student may incur debt and have little means to repay it. Although students with lower scores are often the focus of remedial efforts, MCAT scores of 8 with a total of 25 are sufficient to predict academic success without excluding too many students from a variety of diverse backgrounds (Albanese)

Legal pressures have increased greatly in recent years. Students (and parents of students) with higher MCAT scores have pursued a variety of legal and other efforts to gain admission. It is harder and harder to turn such students away even though they may have characteristics that may make them less desirable. The virtual lawyer present at the birth of every child may now also be present in the admission committee chambers at the birth of every doctor.

Admissions committee work already demands a great deal of time from a large number of medical faculty. Students with more diverse educational backgrounds and scores also take more time and demand more expertise for a better admissions evaluation. This means more time and effort from medical faculty already giving large amounts of time with little chance to benefit from such efforts. Madison demonstrated that admissions committees could greatly enhance the selections of students that choose primary care (madison), but the process of determining the service-orientation of students added significantly to the evaluation effort (owens). Meanwhile deans and chairs are demanding more time from faculty in areas that do count, such as research and clinical revenues. With increasing pressure in all phases of academic life, admissions committees may depend more upon the MCAT than ever before.  Medical schools with fewer failing students are also less likely to face accreditation challenges and also may need to spend less time and resources in remedial efforts.

The competition for top students in many schools is intense. Deans urge admissions committees to recruit students with higher MCAT scores. Students with higher MCAT scores are felt to have great potential as researchers. Schools are also expanding M.D./Ph.D. programs and revising curricula (Duke, Stanford) to emphasize research and scholarly work. The deans all hope this will help the medical school to a bigger share of the multibillion-dollar pot at the National Institutes of Health. The declines in state funding, graduate medical education funding, and reimbursement for clinical revenues, Medicaid, and indigent care leave few options. Long forgotten are studies raising concern regarding use of the MCAT, especially those noting that schools with higher MCAT scores are less likely to graduate family physicians. 17

State Education Opportunity

The nation's medical schools also depend upon education at the K-12 and college levels. Recent decades have presented significant challenges to state government. Economic declines and taxpayer concerns have resulted in fewer revenues. Health care, prisons, and social programs all demand more funding and increasingly compete with education for fewer remaining state tax dollars. Health care costs have risen at a higher rate and states soon are likely to have less federal assistance. One view of this situation is that increasing health care costs are eroding education funding in many states. At the same time education itself faces challenges regarding maintaining and expanding technology and competing with other careers for workforce.

Understanding Higher Education and Income

Rural and inner city schools face the most difficult challenges regarding higher costs and  maintaining quality. Rural schools attempt to provide a wide variety of education for a smaller number of students. Inner city schools face additional challenges regarding language, culture, poverty, fatherlessness, and security 18 (rural education). Both rural and inner city share in teacher shortages and both have problems with education quality in the college prep courses. Both also have larger numbers of teachers who are near retirement as well as those new to teaching. Both have lower property values and less ability to compensate for state and federal cutbacks. Some rural and inner city areas also have large tracts of land that cannot be taxed at all. All schools have had increasing costs just to satisfy federal and state requirements with more to come. Rural and inner city school districts in several states have filed lawsuits to improve the distribution of state education funding. It is unknown whether the improvement in funding distribution is worth the cost of the lawsuits, since both sides deplete limited resources that could best be used on education.

In today's highly competitive admissions environment, it may not take much to keep students from being admitted to college or to professional school. Potential young professionals from rural and inner city areas may be the ones most impacted. Parents in urban areas with more income can move to better school districts or access private schools. Rural areas have fewer private schools and parents are often tied to sources of income that are not as mobile. Lower income students compete for federal grants that cover less total college costs. Lower income students often have to work during college, making high marks in GPA and MCAT more difficult.

A decreased flow of students from lower income origins, from rural areas, from non-professional parents, and from students who are somewhat older or married would have also have a greater impact on the numbers choosing family medicine. 19   ( need other reference regarding fp choice)    Given that rural physicians often come from rural areas and physicians that serve the underserved often come from inner city locations, declines in rural and inner city education would greatly limit access to care for areas already most in need of care.

The “Natural Experiment” of Managed Care

There are many that cling to the hope that medical school curricular reform could improve the flow of family physicians. There is little hope for this line of thought. Medical students graduating in the years 1994 through 1998 had a maximal incentive to choose family medicine and primary care. During this time there was far less prospect of a job outside of primary care. Despite this maximal "intervention," the numbers of US medical students choosing family medicine increased by only 15 – 25% over this time period before declining to a level lower than pre-managed care levels. This was only 600 additional FP graduates and 1000 increase in all primary care disciplines from US medical schools at peak levels. Even if these choices has been maintained, the US would have still imported nearly 50 % international medical school graduates.

Family Physician Workforce: How Many?

In order to replace retiring primary care and family physicians and to make progress toward providing a continuity medical home for most Americans, the US will need to graduate additional physicians above attrition and at least enough to keep up with US population growth. Current growth is about 800 above attrition. To increase at a pace with US population, the growth should be 1100. At 1500 a year FP physicians contribute

The nation's medical schools are graduating fewer family physicians instead of more. The obstacles to more family physicians appear to be at the admissions level or before. Studies regarding admissions and state education and other barriers should be a top priority in a nation that desires to have better health care, lower health costs, and increased access to health care.  

Methods

Hypothesis:

Medical schools admitting more students born in rural areas (or fewer from urban origin), schools admitting more instate students, medical schools that have a rural mission or person, schools with a family medicine department, schools with a lower MCAT average scores, those located further west, and medical schools located in states with better education opportunity are likely to graduate more family and general practice physicians.

Dependent Variable

The dependent variable was the percentage of family medicine and general practice physicians graduating from each medical school that had a final year of residency training between 1997 and 2003. This included 21000 graduates who completed a family medicine residency program as noted in the AAFP database and 4000 additional physicians from the AMA Masterfile who designated their specialty as FP and had at least one year of primary care graduate training.

Independent Variables

Medical School Variables

Medical school type: public, private, osteopathic, and allopathic.

Establishment of a family medicine department by 1992 20 (Kahn)

Existence of a rural mission 21 (Rosenblatt) or rural medical education person at the school (surveys by the STFM Group on Rural Health survey).

Percentage of students born in rural areas using the birth city and state from the AMA Masterfile. Over 98% of US cities were coded rural or urban (Hart reference).

Percentage of instate medical students for each school

Average MCAT for each school in 2000

Grade Point Average for students at the school

State Education Variables

Average of all Achievement Test scores for the state’s 8^th graders

Product of high school graduation rate and college continuation rate for the state. 24 (from Education Weekly).

State data was compiled from the US Census regarding population. 23 The percentage of non-metro population was used to weight the regression as several of the variables were sensitive to population distribution.

Footnote: Commercial sources and internet sources that compile MCAT scores for schools had to be utilized as AAMC would not release individual MCAT scores for schools.22 

These scores were verified by accessing medical school admissions web sites (    of  130 accessed). Use of 5 different sets of MCAT scores from 2000 - 2003 revealed no differences in the regressions.

Some medical schools were excluded due to significant variation in mission, ethnicity, and comparability of state education data. These atypical schools included the 2 year, branch, or regional campuses (University of Minnesota Duluth, Charles Drew); the 6 year medical school (U of Missouri Kansas City); those with large variation in ethnicity and mission (Howard, Meharry, Charles Drew, Hawaii, Puerto Rico schools); and those with too few graduates (Arizona College of Osteopathic Medicine, Florida State, Touro, Edward Via, and Pikeville).

The data was analyzed with SPSS v 11.5 using linear regression weighted for the percentage of the population of the state in non-metropolitan areas. 25

See Birth Origins and FP Choice for update

The average state achievement test score also conflicted with the educational opportunity variable. Both could not be included in the same regression.

Longitude and education opportunity variables had less interactions with other variables in the regression.

Regressions using only board certified FP physicians did not vary from those using a FP and GP physicians.

Predicting the Family Practice Match

Since the choice of family medicine is steady regarding background and ethnicity, it is possible to predict the nationwide the US contribution to family practice workforce 7 years into the future. The following graphic notes the actual and predicted family practice match for recent decades. The only variation is the managed care years impacting the graduates of 1994 – 1998 as shown, with a return to baseline decline after this period.

s

Predicted estimates of FP vs Actual AAFP data on US seniors choosing FP

Graphic on web site

http://www.unmc.edu/Community/ruralmeded/admissions_and_origin.htm

Conclusions

Medical schools are admitting fewer of the students that are most likely to choose family medicine. Changes in education may be limiting college and professional school options for certain students. Admissions decisions related to the MCAT appear to be excluding students that are more likely to choose family medicine.

Linear regressions involving medical schools are not the same as those involving individual students. The observations involve policies and procedures rather than characteristics of individual students. Also the associations involved with MCAT, ethnicity, or background do not imply problems with any test or group. These are only associations regarding how these variables are related to family medicine choice.

Beyond the regressions and returning to a more global viewpoint, it would be hard to design a system with more barriers to the graduation of family physicians. These include a broad range of dimensions involving education, admissions, curriculum, support, appreciation, and health policy at the state and national level. Without comprehensive statewide and national efforts, the nation should expect continued decreases in family physicians and rural physicians with decreasing access to health care, increasing cost, and decreasing quality.

Changes in admissions have been a key part of the improvements in medical education in the past 100 years. However, the centralization and infrastructure associated with medical education may have gone too far. The MCAT has played a key role in decreasing medical school failure rates to nil, however, there is increasing evidence that the MCAT, as currently utilized, is becoming a barrier to certain groups that may be important in several areas in medicine and medical education. Each medical school should carefully consider decisions that might limit the opportunities for certain students, especially those greatly needed by local, state, or national priorities. This is not limited to family physicians.

• MCAT also loads negatively with OB-GYN, office based physicians, office based pediatricians, and primary care physicians.
• Rural birth is associated with ob-gyn, office based physicians, and primary care
• Longitude factor only in primary care and FP
• Older students and state education factor only in FP
• Ethnicity impacts FP more, with whites tending toward FP and Asians away from FP

Longitude favors IM and Peds in the east and FP in the west. Medical schools with FP departments or rural mission/person graduate fewer pediatricians and emergency physicians

The weaknesses of this study involve the data. The use of a variety of data sources is always a concern. However in key areas such as education opportunity, all of the variables (higher education expenditure per capita, median income, poverty level) illustrate the importance of greater education opportunity.

This data includes the classes graduating from 1994 - 2000. During 5 of these 7 years, the students had a 25% higher choice of family medicine compared to other groups over the past 30 years. This "managed care panic" influenced urban students to choose FP at a 30% higher rate as compared to 15% more for rural origin students. Schools with more urban students had a higher FP/GP choice during this period. These schools with more urban students are schools that tend to have lower FP/GP percentages. This means less differentials between the higher and lower producing medical schools. The impact of the MCAT, birth, longitude, and other variables might be more pronounced in medical school students graduating before 1994 and after 1998, including the most recent graduates. The true impact may actually be higher than the regression equation notes.

The MCAT scores did not correspond directly to the classes involved in the cohort. However MCAT scores across all medical schools do not tend to vary from year to year. Regressions using 5 different yearly sets of MCAT data did not differ from the reported results. Accessing numerous medical school admissions sites, some with multiple years of data, noted good compatibility with the actual scores used in the regression.

The rural population percentage of the state could have been used as an independent variable. Instead it was used to weight the regression. Rural states tend to have higher state education opportunity, lower MCAT scores, more rural birth origin students, more older students, more instate students, etc. Using this weighted regression compensates for these known interactions.

The study highlights a continuing controversy regarding the utilization of the MCAT test as a part of the admissions process. Medical leaders such as AAMC President Jordan Cohen have widely promoted a more diverse admissions process emphasizing personal characteristics. Several authors have pleaded with medical schools to use the test more as a baseline rather than as a ranking determination. A very few schools have stopped using the test. Use of the MCAT can predict 90% of students likely to face academic difficulty, but 80% of those identified will still graduate (alabama web site). Multistate studies such as those by Albanese note that MCAT scores of 8, a total MCAT of 25, and a GPA of 3.0 are sufficient to predict passage of USMLE Step 1 and promotion to the clinical years without excluding students unnecessarily. 26 

The MCAT remains a “speeded” test where time pressure is a part of the performance. This is due to concerns regarding student ability to process the large body of information over the basic science years.  Recent studies indicate that additional MCAT testing time can improve scores. 27 28  In such tests it is indeed possible that those who are quicker test takers can outperform others that would be equal in quality as physicians and perhaps better for national interests. It is not known whether primary care or family physicians or more diverse students would perform better if given more time, nor are we likely to find this out. The MCAT designers are careful not to be influenced by such areas so as to avoid bias in the implementation. (Ellen Julian)    In theory they may have a point, but the reality of national needs is such that further investigations should continue.

If speededness is a problem for national workforce needs, then alternative routes might satisfy USMLE Step one. These include additional preparation before admission, more than 2 years of basic sciences, or even a basic science curriculum over several years based on a combination of traditional and internet-based medical education. Those who have the greatest potential for rural practice location are those already in rural locations. If medical education can better accommodate their needs, this might mean more physicians for rural areas, reservations, and population groups in the US that have cultural, language, or income barriers. 

In the current pathway to medicine, lower income students and those from rural and inner city school districts may particularly be punished their origins, by lack of exposure to various resources and technology and advice, and by their own inability to afford thousands of dollars in expenses for areas such as MCAT preparation and colleges that have a better admissions potential. Such students may be unwilling to take the risks of education and finance without supportive health advisors and the assistance of family, older siblings, or medical students that reach back to help those facing the same barriers that they faced. It is often easier to identify the intellectual student during high school and college and they are often told that they should apply for medical school .Those a bit lower on the scoring scales may not get the same early identification and encouragement.

With a speeded intellect test, only certain types of students may be able to escape certain educational settings. In states with better education opportunity, those with a broader scope, the ones more likely to become generalists, may survive and become viable candidates. In states with less opportunity, only the most intellectually gifted may be able to overcome barriers of education, language, or culture. Inner city and rural youth with a greater tendency toward family medicine may be the ones excluded. Could it be that only the intellectual are escaping reservations, rural areas, and the inner city? If so, students with these characteristics are unlikely to choose primary care and underserved areas. Indeed native and minority students have had decreasing rates of choice of family medicine in recent years. This may indeed be an indication of changing admissions.

One approach taken by some schools is to minimize the quantitative MCAT impact by blinding the actual scores or minimizing the use of MCAT as a ranking system. This could be a national decision or it could be a local decision by the school. Schools would set a minimum and would only consider students that had scores above this level. This would be enough to insure a reasonable probability of academic success without the potential of overemphasis. This would also take the pressure off of medical school committees. It has become difficult to turn down students with high scores, even with less than stellar qualifications in other areas. With a blinded MCAT score, they could honestly note that they were only doing their best to choose from the pool of candidates presented to them in order to graduate the best physicians. This may also help address the legal and social pressures that can be exerted by higher income and professional parents. Such parents can present a much more significant challenge than those of students from more humble or distant origins.

The impact of environment is a more difficult task. The longitude variable suggests that eastern medical schools are unwilling to change to help the nation meet workforce goals, perhaps at the expense of other school priorities. Also the competition for National Institutes of Health dollars may be driving medical schools to admit more and more academic-intensive students who are less and less likely to be able to choose careers most in need in the nation.

Family medicine needs to reconsider its options.

§         FP Department efforts - Since 1992 every medical school that is likely to ever have an FP department has one. Continued efforts to set up other FP departments are unlikely to result in more FP physicians.

§         Curricular changes - Changes in medical school curricula are unlikely to increase medical school graduation rates beyond that stimulated by the managed care panic years. The small increase in FP choice during these years is not enough to provide gains of family physicians in relation to expansions in total physicians or to even keep up with the growth of the US population.

§         Increase reimbursement is also unlikely to influence the choice of FP, again as demonstrated by the managed care era. Mainstream membership expects improved working conditions and income, but the current situation demands leadership that focuses the efforts of all FP physicians upon a future for family medicine. Real change in reimbursement, respect, funding for training, and other areas may not be possible unless 40 - 50 % of Americans, including the most influential ones, have a continuity medical home and learn to appreciate the value of primary care and family medicine. Without massive growth in FP physicians, at least above the population growth, each passing generation grows less aware of the value of family practice.

§         Family medicine should focus efforts on admitting students who have the potential to become family physicians. It appears currently that those most likely to do so are being screened out. Special considerations for rural background students by medical schools were either a temporary approach, good rhetoric to appease certain interests (like those in more rural states), or they were overcome by other considerations. 

§         New tracks involving older students offer some real advantages. These give rural students a second chance. Later admissions also allow students without educational advantages to compensate and enter medicine. The maturity of older students can contribute to the character of a class as well. Older students have also contributed to accelerated programs and other special tracks in significant numbers.

Rural medical educators have an important role in facilitating the choice of family medicine. Admissions efforts are one area, but the preprofessional area is perhaps a more important goal. Direct efforts are important, but rural faculty are limited in numbers and availability. They can multiply efforts by working with students from small towns. Inner city students traditionally have provided a great boost in admissions, reaching out to those who face similar barriers. This attempt to bridge between two cultures is often stressful, but is extremely important to younger students. Rural medical students often assist family members, friends, and direct contacts, but there is not the same recognition of the need as compared to inner city students. More work needs to target high school and middle school students to motivate them to more challenging education and interest in health careers.

There is no stronger influence on any student facing significant barriers than to see someone like them who has made it to medical school. This should also be apparent to any medical students who participate. There are other reasons that students should make "home visits." After all, when they reach out to other rural students like them, they are actually recruiting future colleagues and replacements. There are even formal teaching activities that students can take to rural schools that are educational, fun, and helpful for the schools involved. Only 5% of rural high school students in Nebraska knew about a long-standing special admissions program for rural students. (John Klein and Prime). A half day of effort can target 1 - 2 schools, some key teachers and counselors and could get a few students moving down the medical school pipeline that otherwise would not.

Rural communities may again face the greatest consequences regarding the current education and admissions policies. Currently the boost in total primary care numbers in recent years coupled with the outputs of specific rural programming has resulted in some of the best levels of rural family physicians in decades. This tends to confirm the work of Newhouse 30 and COGME and Jack Colwill 8 noting the need for enough primary care and family physicians to be able to meet rural workforce needs. However this is not likely to last. Family practice programs have found it difficult to recruit medical students. This has already resulted in fewer choosing smaller and rural programs and rural training tracks. Many smaller programs are closing after years of difficulty with recruitment and with new graduate medical education funding restrictions. Small towns once had hope of attaining 30 physicians per 100,000. Now it looks like the level will decline to 20 - 25 physicians per 100000.  8

It is also hard not to notice what is happening in international medical schools. These schools have become the major source of US family physicians. Ross University in Dominica now provides the most family and general practice physicians for the US with 93 entering family practice in 2003. Two other international schools, three osteopathic schools, and 4 allopathic medical schools join Ross in the top ten. Some of the Caribbean schools have classes made of 70 - 90% US citizens. Family practice graduates from these schools have rural graduation rates of 25%. This suggests that at least some of the rural background students in the US have had to go to other countries to pursue their dreams. This may only be a small percentage of those who could become physicians. Others in more difficult circumstances in this nation may not be able to afford or risk this effort.

It would be tragic to find that many potential medical students are giving up at an early stage because of obstacles beyond their control. This would include state education decisions, parent income, teacher expectation, and access to career advisors and health professionals (shadowing, advice, encouragement) in the high school and college years. It is entirely possible that these limitations are impacting all young professionals, not just physicians. If this applies to teaching professionals, then an accelerating cycle is a possibility, where lack of quality teaching leads to fewer professionals from rural and inner city areas which leads to poorer quality education, with impacts on a number of different types of professions.

It may not be a surprise that two of the most successful models for graduating family physicians are located in Minnesota and Pennsylvania, states with more resources, strong small colleges, and better emphasis on education at all levels. In such states it may be possible to find enough who are 1) academically qualified, and 2) interested in family medicine, and 3) interested in returning to rural locations. In states with less education capacity for young professionals, it may not be possible to obtain more than one or two of these three criteria. This may be a reason for inconsistency in the connection between rural background and choice of family medicine or underserved areas.

Screening out those likely to choose family medicine is likely to worsen the health care quality in the nation and to increase health care costs. Studies by Baicker and Chandra just published in Health Affairs noted that an increase of general/family practice physician sufficient to raise the ratio of such physicians by only 1 per 10,000 people improved the state's quality ranking by more than 10 places (out of 50 states). This also reduced health care costs an average of $684 per individual per year. The states with the highest health care quality had the most FP/GP docs and also tended to be states with a higher percentage of rural people. By contrast, states with more specialists had lower quality and higher costs. 5 (Baicker and Chandra Health Affairs)

One of the great mysteries of our time is why our state and federal governments continue to tolerate policies that are 

1.      Not contributing to national workforce goals,

2.      Likely to increase health care costs, and

3.      Likely to decrease health care quality

Medical schools would do well to listen to any number of medical leaders from Abraham Flexner to Jordan Cohen, who have all pointed to the need to have better physicians through a better admissions process.

“Use MCAT scores and GPAs only as threshold measures. Rather than giving more weight to higher scores, why doesn't each school decide for itself, from data available from its previous students, what level of GPA and MCAT performance is sufficient for predicting success in clearing the high academic hurdles of medical school -- and leave it at that. We would send a powerful signal to those intelligent idealists who are currently eschewing medicine if they knew that, once having met the academic achievement threshold, they would be evaluated solely on the basis of their humanistic qualities, their penchant for serving others, their leadership abilities, and so on.” 31

Don Madison has given sage advice in his article regarding generalist outcomes of the class of 1985 at the University of North Carolina:  32

In the best of all possible worlds the majority of all entering medical students would end up as generalists, but all physicians - family practitioners, physiatrists, urologists, allergists - would be well-educated, highly intelligent, well-rounded, personable, honest, altruistic, highly motivated individuals who had tested themselves prior to medical school in some tangible way against their goal of a medical career and a service profession. 

If an admission committee informs itself of "what finally happens" to those it admits, its decisions can contribute to achieving whatever policy its medical school adopts with respect to the mix of physicians it wishes to produce. 

 

Future studies will focus on the origins of those likely to choose family medicine. This includes rural location but studies will include suburban and inner city origin students to examine the impact of income, being raised in a continuity medical home, service orientation, and education. There are also indications that general surgeons and other specialties may be impacted by changing admissions in recent decades.

Medical school admissions are based on the MCAT, grades, the application, and the interview. Studies have highlighted the problems regarding the less objective and non-cognitive areas. In the absence of better measures, better training of committee members, better understanding of the impact of admissions beyond USMLE 1, and more time spent in admissions deliberations, there will be increasing reliance on MCAT, GPA, and college information. Those who do not do as well on speeded tests, those who do not have the highest grades for a variety of reasons, and those who cannot afford colleges with the best reputations and track records will continue to have less opportunity to become physicians.

The relationship between rural and GP/FP goes back over 100 years. Over this time period the two have remained basically one slowly descending line. The US population and the US doctor population have increased, with more rapid increases in physicians in recent years. I have a graph of this at http://www.unmc.edu/Community/ruralmeded/fpgrad/decreasing_rural_fp.htm

The FP and GP numbers are important to access in underserved areas, and also to health care quality (Baicker and Chandra, Medicare Spending, The Physician Workforce, And Beneficiaries Quality of Care,  Health Affairs April 2004 http://content.healthaffairs.org/cgi/reprint/hlthaff.w4.184v1.pdf) Studies by Baicker and Chandra on Medicare patients noted that additional health care expense adds little if any additional quality and may be unhealthy in some areas. The addition of 1 general practice physician per 10000 improved the state's quality ranking by more than 10 places (out of 50 states) and a reduction in spending of $684 per individual. Increasing specialists by 1 per 10000 resulted in a quality drop of 9 places and an additional $526 cost per individual

MCAT Correlations

Probability of admission tables

Cost, Quality, Access, and Physician Workforce Expansion

Origin, Admissions, Family Medicine