Five Periods of Health Policy and Physician Career Choice

 Robert C. Bowman, M.D.   October 30, 2006

The nation has had two excellent natural experiments to test the impact of health policy upon medical student career choice. This work compares these two periods and the remaining three over the past 41 class years.

Background:

General practice levels have consistently declined throughout the past century in a close parallel with rural population levels and rural physician levels. The deteriorations in general practice left the nation without a suitable physician for rural and underserved areas. Only in recent decades has the pattern changed. National debates highlighted the particular problems of the poor and elderly and the need for a return of general practice. The nation made sweeping changes in a number of areas during the late 1960s and 1970s. This involved a time period of primary care emphasis, increased state and federal funding for medical schools, and the creations of Medicare, Medicaid, and family medicine. In many ways this was a perfect storm for family medicine and primary care choice. Medical student choice of family medicine soared. During the 1980s the nation had a different focus but maintained some level of support. It was a neutral period. In the 1990s the nation entered a new level of support for rural and underserved populations, primary care, and areas outside of major medical centers. The nation again changed health policy in the form of managed care and health reforms. Prominent factors were a "forced" choice of family medicine and primary care, increases in primary care reimbursement and decreases in subspecialty funding. In more recent years the nation has returned to a “market forces” plan for physician support with career choices much like 40 years ago. Losses in primary care include more than just new medical student decisions as seen in losses of nurse practitioners and physician assistants. Primary Care Retention

At current levels of deterioration, family medicine will increase share to 40 - 60% of total primary care, rural primary care, and underserved primary care. Changing Primary Care Contributions 1970 - 2015 Sadly this has to do more with losses of other forms than gains in medical student choice of family medicine.

Differences and Definitions provides more background on the types of students who distribute, the types of schools, the various differences in primary care types, definitions, and limitations of physician distribution.

Rural Coding - about the RUCA coding and distributions

Despite regular medical student career choice changes with health policy changes, few studies have explored this relationship. To understand health policy impacts it is important to explore in a number of dimensions.

The following themes are supported:

• Policy and practices that facilitate choice of family medicine, improve funding for primary care, and increase support outside of major medical centers will improve physician distribution.
• Policy and practices that impair choice of family medicine, decrease primary care support (decreased reimbursement or increased costs or impaired primary care delivery), and increase support to major medical centers will concentrate physicians in major medical centers.

In the following graphic the family medicine and general practice career choice is shown by class year of medical school graduation. A comparison is made with the rural born graduate percentage, a relative measure of the potential for distribution in medical students. This relationship is best seen in the neutral health policy impact period from 1979 to 1990 when FM choice % paralleled rural born medical student %.

When considering medical student choices there have been 5 health policy periods.

1. Little or no health policy support as in the time period before 1965.
2. Medicare and Medicaid Creation - Health policy support for lower income, older, inner city, and rural populations increasing from 1965 to 1978
3. Neutral health policy impacts from 1979 to 1990
4. Managed care and health reform impacts from 1991 – 1997
5. Rapid return to market forces from 1998 to the present and likely to 2010 or longer given the years it has taken to change health policy (selections 2 and 4)

Note that the rural born "line" is a proxy for distributional student admissions which include all who have origins outside of major medical centers: rural born, lower income, older, children not of professionals, middle income. This is a group of about 30% of admitted medical students but represents 70% of the population. Those with origins outside have the most potential for outside location, most fully expressed by choice of family medicine and during times when the nation maximizes health policy for distribution.

The First Period: Below the Line

The first period prior to significant intervention was characterized by severe problems for lower and middle income elderly, those in poverty, and particularly for young people in poverty and rural areas. Access to physicians and hospital care was limited to the most urban areas and those with higher income or health care insurance. It was difficult for physicians to locate in rural and underserved locations without the support of state and federal government or a booming local economy with health care coverage. Rural and underserved locations also did not enjoy the economics generated by federal and state health care. (Few understand how health care and education dollars from federal and state sources are important for economics and other parts of rural and underserved areas, how efficient education and health care can be in such areas, and how cuts in education and health devastate these areas.) Major national reports illustrated the problems stemming from declining primary care and general practice choices. State and federal support for medical education was poor.

In the graphic, the levels of family medicine and rural choice were “below the line” or lower than the percentage of rural born students admitted. The nation’s health policy did not support distribution.

Second Period: Massive New Investment

The second period was influenced by the nation’s War on Poverty and involved efforts in education, social programs, child development, and economics targeting low income and underserved areas. Despite the long term and determined opposition of organized medicine, the nation created Medicare and Medicaid. These programs grew rapidly relieving pent up demand for health care. The nation also created family medicine and funded new medical schools with state and federal funding. The nation also supported expansion of medical schools during this period of primary care emphasis.

The result was a doubling of medical students and a doubling of primary care percentage, with a quadrupling of total primary care and family medicine numbers in just ten years. The new allopathic and the only 6 osteopathic public schools that were created continue to lead the nation in primary care, family medicine, rural, and underserved distributions.

By the graduating class of 1975, the new family physician graduates were choosing rural practice at 30%. Family physicians have supplied 30% of the rural physicians for each subsequent graduating class as still found in practice in 2005 and despite major health policy changes. In the newer schools the contributions of family medicine were 50 – 80% of all found in rural locations.

The nation divided into medical schools with higher levels of distribution and those without much distribution. The distributional medical schools provided concentrations of graduates for rural and for underserved areas that made up much ground for the medical schools failing to distribute. Expansions in these non-distributional schools provided lower contributions to distribution and a dilution effect that captured the attention of workforce planners in the nation and resulted in a slowing of medical education growth.

The earlier period of time also involved a different United States. Education was emphasized 1960s and 1970s at local, state, and national levels. A much broader range of candidates was presented to medical school admissions committees. The lower income, middle income, and rural born physicians had much higher levels of admission in this time period. Sadly these humble origin types did not gain ground with the opportunity of medical school expansion. There was some increase in the numbers of rural born, Black, and Hispanic medical students admitted, but the percentages remained much the same or continued to decline as in rural born admissions.

Expansion did help the nation discover the value of older medical students (over 29 years at graduation) who increased from 5% to 15% of admissions and continued to increase to the present 22%. Older physicians are an important source of primary care, family medicine, psychiatry, rural, and underserved physicians. Divisions in Physician Career and Location Choice Related to Age at Graduation

Asian and foreign born medical students were not a major part of medical education during this time period. The top income quintile of predominantly white medical students continued to dominate admissions with 60% of graduates; however the differences between the highest and lowest income students and education differences were much lower. The nation has separated richest from middle and lower income groups in all ethnicities and races since this time with great impacts on the composition of the medical students. Changes in Admissions in Allopathic Medical Schools

In the graphic, the levels of family medicine and rural choice increased to a level above the percentage of rural born students admitted, indicating some measure of support for physician distribution. Reimbursement support such as Medicare is also a consideration.

Third Period: Neutral Health Policy or Admissions-Determined Distribution

The third period was a neutral period of health policy influence with a slow decline in choice of primary care and family medicine at about the same rate of decline as the rural born student admissions (again a reflection of all distributional type admissions not just rural born). This period involved limited growth of state and federal programs, energy problems, massive increases in foreign policy and military spending, and lower levels of support for rural and underserved areas. The nation did not ignore health policy, but it also did not improve health policy support. New wrinkles began to appear. The reimbursement formulas in health care were based on percentage increases. Each year left behind the lowest paid physicians and facilities especially in rural areas, primary care, and locations outside of major medical centers. Young physicians in the early parts of the period were treated to even lower levels of funding. Rural hospitals were closing from declines in funding and as major medical centers discarded their chains of rural hospitals that they had acquired during cost-plus reimbursement.

Family medicine became increasingly focused on academic settings as the first generation “from the field” leadership gave way to those connected to academics throughout life, education, training, and practice. Not surprisingly the focus has remained medical school departments, programs in every state, match rates, and Title VII funding.

The focus of medical schools also changed. A passing interest in the millions provided in medical education support regarding primary care was soon surpassed by graduate medical education funding that grew into the billions. Subspecialty, graduate medical education, and major medical center emphasis grew while primary care, teaching, rural, and underserved areas lagged.

Multiple steady changes were difficult to comprehend. Questions were raised about primary care and family medicine. Late in the period the steady declines in family medicine choice became the subject of conversations and publications. These may have had little impact other than becoming a major topic of conversation at medical schools and among medical students, potentially resulting in lower choices of family medicine. This was minor compared to the swings in opinion to come during managed care and the managed care rebound. National changes in health policy, education, and medical school admissions were not considered as factors in the steady declines.

The levels of family medicine choice remained “above the line” but were decreasing steadily along with rural born admissions.

Fourth Period: Managed Care and Health Reform, Redirecting Resources

The fourth period was the managed care era. This was also an era of health reform although managed care conflicts overshadowed many of the changes. This is also the period of the most abrupt changes in medical student career choice toward and then away from primary care and family medicine. Not since the creation of Medicare and Medicaid had health policy changes generated so much controversy. Within a short period of time all medical students and all residents were more than aware of coming problems that would impact all not choosing primary care. The abrupt increase and decrease provided a specific impact on a limited number of class years, providing a natural experiment regarding health policy impact. The impact was greatest upon family medicine choice for those in medical school at the time (medical school graduates of 1995 – 1997 at peak) and on internal medicine retention in primary care for those in internal medicine residency.

Managed care dominated the agenda along with workforce reports projecting massive decreased need for subspecialists and hospital beds. Almost lost in the controversy were the health reform components. The nation doubled Medicaid from 1990 – 1995. Reimbursement support such as Medicare is also a consideration. No longer did the nation allow new massive investments as in the 1960s. Increases in one area had to come from a different and related area. Reimbursement reforms shifted funding away from subspecialty care and major medical center care and into primary care and outpatient settings with a series of increases. The older reimbursement systems that guaranteed steady declines in primary care and rural payments relative to other areas were replaced by better payment systems, although still far from equitable. Organized medicine and medical education protested the managed care changes vigorously. Increasing costs of delivering care (liability, supplies, personnel), decreasing state and federal support for medical education, exodus of patients from training centers with the new health plans emphasizing patient choice, declining local support (closures of county hospitals) and retention of the most subspecialized faculty with losses of the more general subspecialty faculty that did more teaching were just a few of the changes facing medical schools. The changes moved significant funding out of major medical centers to primary care and to locations outside of major medical centers. The controversies and protests added to the impact on medical student choices.

New sources of primary care were supported and physician assistant and nurse practitioner programs grew. The initial programs often involved an emphasis on students with origins outside of major medical centers and training outside of major medical centers, but these patterns would change with health policy leading the way.

Residents in internal medicine during the time period turned back to office based primary care. After declining steadily to 40% for the 1992 residency graduates, retention in primary care increased back to 53.5% for the 1996 graduates and then decreased back to 40% for the 2001 residency graduates. The correlations between Masterfile office based choices and reported primary care levels of internal medicine graduates are within a few percentage points. Now first year internal medicine residents indicate that they plan to remain in primary care at less than 20% levels. (Garibaldi 2005)³

Internal medicine retention in generalist office base care had declined to 40% and increased back to 53%. These are comparisons in the 2005 Masterfile for the previous graduating classes. The family medicine choices are stable. The internal medicine residency graduates in office based primary care deteriorate steadily with time after graduation and this may influence observed patterns at a later date (see medicine pediatrics figures). Overall across this managed care period this was a fairly level period of time with similar career choices.

Family medicine choice went through the roof and somehow programs managed to expand from 2200 positions to 3400 by 1996 although the levels are half of this total now. (Pugno 2006)¹² Family medicine choice increased above 60% for the peak levels compared to the trough levels of 1988 – 1990 graduates. At maximum over 2200 US MD Grads chose family medicine in the match but actual peak levels of US MD Grads choosing family physicians were over 2600. Rural location in family medicine remained at 25% choice but numbers increased 68% from an average of 405 for the 1988 – 1990 classes to 638 for the 1995 – 1997 graduates as found in 2005 locations. Family physicians found in underserved areas also increased from 12% to 14% of family physicians but increasing from 200 to 360 or an 80% increase before declining again.

Current 2005 Careers and Locations, 3 year Averages at Trough and Peak Levels

Medicine pediatrics numbers did double during the time period, but the apparent increases above using Masterfile data are not accurate. They appear to be higher due to losses outside of medicine pediatrics, most likely to subspecialization. These substantial losses do not allow calculations of changes. The steady losses of past class years mean that more recent graduates and the hardy office based crew that remained in medicine pediatrics skew the observed contributions as in 17% serving in rural areas. Those remaining in medicine pediatrics are older and tended to graduate from schools with lower subspecialization rates and higher choice of family medicine. When the denominator becomes all who matched into medicine pediatrics and these are adjusted for rural location rates for subspecialists (7 – 8%), the medicine pediatrics contribution decreases to less than 10%.

Increases in internal medicine are also likely to be impacted by the same factors as subspecialization does occur over time. Pediatrics and obstetrics changes are smaller.

Graduates in primary care that subspecialized later would not be counted in the 1988 – 1990 graduates as primary care physicians.

The table above uses 2005 careers and locations. For stable specialties it is accurate. For specialties with changes over time, it is a reflection of current and recent contributions more than past contributions. Retrospective career and location indicators do not capture earlier primary care careers, obligations fulfilled, or brief first time rural and underserved careers. However the 2005 snapshot does capture a workforce in equilibrium conditions and those making long term contributions to primary care, rural, and underserved areas. For a specialty such as family medicine where 99% remain active, 98% remain in family medicine, and over 90% remain in office based primary care; there is a consistency in the contributions. Primary Care Retention⁴ In family medicine, first time practice choices and distributions within states, types of states, and geographic and socioeconomic classifications within the nation tend to remain the same over time. 

The greatest levels of change in family medicine choice involved instate born students and those born in the same county where they attended medical school. A switch to family medicine may well have preserved options to remain in the same location as birth, education, training, family, and other contacts. The career changes in this massive group of medical students greatly added to total primary care numbers and with family medicine choice, these primary care levels were preserved over time. Those at either end of the distributional spectrum were least impacted by health policy changes with 30 – 40% increased choice, the humble origin types and the elite students in elite schools. The elite choices involved up to 50% increase, but this impacted an actual change of 1 – 3 medical students a year more choosing family medicine. The most humble and most elite students may represent the students that have decided on medical careers earlier than more typical medical students. The typical students in the broad middle group were impacted at greater levels. Family medicine choice in the historically black schools and the medical schools with more Black and Hispanic students were also impacted at greater than average rates for the peak 1995 – 1997 graduates compared to choices before managed care or current choices. (Bowman 2006) 2  Primary Care Retention 4 

Also missed during the controversy was a glimpse of what sufficient numbers of medical students could do for family medicine curricula. Usually family medicine programs appear to be all things to all types of students in the hope of attracting enough candidates. For the first time the programs had more than enough students interested. Family medicine programs no longer faced the same pressure to be much like other programs. Some were able to pursue a form of subspecialization quite different than the past 50 years. This movement largely appeared to be guided by the influence of program directors and their contacts, interests, and past experiences. Family medicine programs subspecialized into people and population needs and into available clinic populations, not organs, body parts, or technologies. The programs diversified into inner city training, liaisons with Community Health Centers, smaller programs, programs in rural areas, rural training tracks, fellowships (in geriatrics, obstetrics, procedures, rural, or sports medicine), and combinations of family medicine training with psychiatry, emergency care, other forms of primary care, and medical schools (Accelerated Family Medicine Training Programs).

Distribution levels were at a peak, those trained were trained more specifically for rural and underserved areas, and health policy increased the capacity of rural and underserved areas to support the increased numbers of primary care physicians, physician assistants, and nurse practitioners.

Also missed during the controversy was a natural experiment regarding limitations of graduate medical education positions. There were no official policies that cut subspecialty positions or any other first year resident positions, but the nation’s workforce reports forced student career changes that greatly diminished the choice of anesthesia, radiology, and pathology. Mental health and subspecialty choices also suffered during the era as health policy dictated different career choices. The students were forced into family medicine in subtle way that escaped much notice. Of course a more direct approach would have added to the controversy, but would also be likely to generate much the same results.

Also unnoticed was the rapid rise of Asian and foreign born medical students during this period. Asians in Education and Medical Education These are medical students most closely connected to medical schools by proximity, professional parents, most urban origins, and highest income levels. (Bowman 2006) 2  The full impact of increasing admissions of those least likely to distribute outside of major medical centers would not be discovered for some time yet, certainly not at a time when the nation was experiencing peak distribution and proving that the nation could distribute physicians, with the proper health policy support.

The nation’s choices of family medicine during the managed care period greatly exceeded the rural born percentages which continued to decline steadily. The departure was abrupt and unprecedented and in some ways beyond the health policy changes, indicating some level of panic generated from the controversy. The rebound from this would be just as abrupt.

Fifth Period: The Return of the Major Medical Center

The fifth and current period appears to be a complete reversal of the gains in health policy in periods 2 and 4. Some new problems have been added as well. Diminishing support for Medicare and Medicaid, internal shifts of Medicaid away from primary care (to chronic care and hospital care), academic primary care investments gone sour when increasing government and insurance support vanished, increases in a number of regulations, constant changes in regulations making practice more complicated (especially for the smaller groups), increases in the costs of delivering health care, massive increases in the cost of living and working in most medical school areas, massive increases in medical student tuition and cost of living, state budget crunches, the return of energy costs, deteriorations in education and social programs complicating the work of all professionals in underserved areas, active movements of private and so-called not for profit health care resources away from underserved populations, reversals in affirmative action and other indications of the use of the legal system to improve education and health opportunities for those with the highest income levels (and highest standardized test scores), massive declines in federal support of programs for lower and middle income areas, diversions of funding to defense and foreign policy, massive deficits, tax cuts, and resulting poor economics in rural and underserved areas have all played a role.

Exposing Changes in Admissions  This massive changes in this period did not result entirely from health policy changes alone. This period exposed the long term cumulative losses of lower income and rural born students that occurred before the period. The period did include more massive losses of the lower and middle income medical students from admissions. The changes were greatest for the 1997 – 2001 matriculants that were under the direct assault of medical school interpretations of reversals of affirmative action and US News and World Report rankings and changes in education. For the 16,000 entering medical students each year about 1500 lowest income and 1500 middle income types were replaced by medical students from parents making over $100,000 that increased by 3000 (3765 to 6828) during the period from 1997 to 2004. (Minority Students in Medical Education: Facts and Figures XI and XIII) ^{13, 14} This was an increase from 23% to 42% of medical students with continued increases likely for some time. The highest income and most urban replaced those most different in socioeconomics and distant in miles. Those with the greatest probability of distribution were replaced by those most connected to major medical centers. Students of the most urban and highest income origins, Asian students, and foreign born students had the highest probability of admission with probabilities 2 to 10 times greater. These groups are found in family medicine, primary care, rural, and underserved areas at the lowest levels in 2005. They were all born and raised in the closest proximity to the nation’s medical schools and they were the least likely to depart these practice locations, especially with loss of family medicine choice. There is little chance that current medical students with even more connections to major medical centers will leave such locations for practice at any higher level. Subspecialty needs in rural areas are particularly dependent upon rural born admissions. (Bowman 2006) 2

With the loss of health policy support for primary care and distribution and with the loss of the lowest income and rural born students, family medicine choice is becoming the only way to enhance distribution for the remaining students most connected to major medical centers during their first 30 years. Family medicine doubles distributional choices for all types of medical students. Those least likely to distribute are the most numerous medical students and increase from below national averages for distribution to twice the nation’s level of rural (8% to 19%) and twice the level of underserved (6% to 13%) physician location in the urban and highest income types. Family Medicine Physician Distribution 11

Impacts upon the highest income urban origin types are critical as 90% of US MD Grads are urban origin and 70% of the nation’s rural physicians are urban origin. Without rural origins or choice of family medicine, rural workforce concentrations decline below the 11% rural average. Without lower income origin students or choice of family medicine, underserved workforce concentrations decline below 7%. More graduates only dilute the concentrations further. NP and PA contributions also appear to have declined below the line. This is also a likely impact of health policy that supports subspecialty and major medical center employment at higher levels.

In the current health policy, the distribution of health professionals is reduced to admissions of those most likely to choose family medicine. This involves humble origin types of students. This in turn means changes in education and child development, most prominently in rural areas and in the major urban areas with medical schools where 50% fail to finish high school. (Jay Greene High School Graduation Rates in the United States)¹⁵ Teachers, nurses, public servants, and family physicians are difficult to graduate without the support of local, state, and national efforts in child development and early education. What the nation fails to observe is the tens of thousands of school teachers, nurses, and other serving young professionals that are born and raised and educated in states that distribute income and education who are recruited by states that fail to invest enough in children to support their own needs. They take what they need from every other state and nation. This results in losses of education and health resources from other nations, many of them allies in need of the stabilizing influences of health, education, economics, and leadership that stem from their best and brightest.

Admissions of distributional types of students is no easy task. The nation’s top 146 colleges have 74% from the top income quartile (Carnevale in Kahlenburg http://www.tcf.org/Publications/Education/leftbehindrc.pdf) ¹⁶ and the major socioeconomic advantages involved can be tracked back to child development. (Hart, B., and Risley, T. 1995)¹⁷ Medical students are much the same with 60% from the top quintile and another 20% from the second quintile. (AAMC Debt) ¹⁸ Declines in the admissions of distributional types of students and in choice of family medicine in US MD Grads are a disaster for distribution. The nation’s physician assistants and nurse practitioners have also followed the subspecialty and major medical center location pattern laid out by the nation. The primary care contributions of internal medicine and medicine pediatrics have declined below 30%. Only family medicine at 90% and pediatrics at 66% remain solidly behind primary care and only family medicine distributes outside of major medical centers with 57% compared to 31% of pediatrics. About 69% of office based internal medicine and pediatrics remains within major medical centers where multiple revenue lines supplement primary care and multiple duties dilute actual primary care contributions. (Bowman 2006)²

Other than family medicine outside of major medical centers with higher patient care volume, longer hours, higher levels of call, and some measure of procedures and hospital care; the nation has reduced primary care to a loss leader necessary for referrals or teaching purposes or cost savings.

The declines of NP and PA contributions in primary care also indicate health policy changes. PAs working with family physicians have declined from 40% to 28.5% in the past ten years and this trend is likely to continue at about 1 percentage point a year for some time. PAs in rural areas have declined from 25% to 17% and also decrease at a similar rate. http://www.aapa.org/research/index.html ⁵ The reason is simple. PAs and NPs are changing in origins, training, and practice locations. PAs and NPs were “designed” to help in primary care, rural, and underserved areas. However the design also allows (and in some way facilitates) movements away from areas that receive less support from health policy. PAs and NPs can generate more revenue in emergency rooms, procedural care, and major medical centers where health policy rewards hospital and subspecialty care. PAs get the higher salaries in major medical centers and those switching from primary care can gain 10 – 12% income boosts.  http://www.aapa.org/research/index.html   Primary Care Retention ^{4, 5} Primary care physicians, particularly family physicians and pediatricians with incomes fixed by Medicaid and Medicare and serving lower and middle income populations outside of major medical centers cannot meet these salary increases. About 60% of PAs have no call and this allows them to do shifts in emergency rooms, urgent care centers, corporate clinics, or chain stores. The massive support of major medical centers fueling the growth of PA and NP programs is also not a surprise. PAs and NPs allow major medical centers to save on their number one budget item: personnel costs. One of their most expensive items is physicians. PAs and NPs are much the same as family physicians in limited productivity studies comparing the best of the PAs and NPs to the most limited forms of family medicine in managed care and major medical center settings. PAs and NPs replacing emergency physicians, residents in training, and various hospital physicians and their benefits and shared profit plans can save millions.

The nation also failed to observe that the rebound from managed care and health reform basically returned choice in family medicine below the line of rural born percentage, indicating that the nation has little support for distributional careers. Fortunately the nation has other sources of family physicians in osteopathic and Caribbean and distant international schools. Also missed is the fact that family medicine continues at levels of graduates not much different than peak levels. The distorted allopathic focus has missed this.

Missing and Disturbing Information in the Age of Information

The rapid increases in the higher income and most urban medical students are changes ignored or impacts yet to be identified. Changes in NPs and PAs are evident only with national data that is often not available. It is no longer accurate to note that NPs and PAs will be the source of primary care in the future. The best that can be expected without changes in policy is for PA program growth to be able to maintain current numbers in the face of declining percentages in primary care and rural areas. Further declines in primary care from physician sources outside of family medicine have been mentioned, but have been underestimated in total impact in a number of areas. Continued family medicine contributions have been minimized, again largely a part of the allopathic US school focus.

Family medicine retention in primary care is a critical piece of evidence regarding health policy impacts. With declines or with compromises in family medicine, the differences would become invisible and the nation’s medical leaders could state that distribution is impossible. No medical educator should ever state this again. The nation has more than proven the ability to distribute physicians, even in difficult health policy periods and despite changes in education and income distribution.

Also missing throughout all the major health policy changes and in the workforce literature is the continued dedication of family physicians to their initial purpose. Family medicine continues to serve half of Community Health Center physician positions and 61% in rural locations. (Rosenblatt 2006)  ⁷  The RN plus FP model remains the nation’s choice in efficient, low cost, effective health care. Family medicine rural choices remain at 24% or 2 times the national average of 11% and underserved distribution at 16% or over 2 times the national average of 7%. Graduating more family physicians distributes physicians away from major medical centers. Graduating any other type of physician means more for major medical centers and the most served locations. ¹¹

Also disturbing in recent decades has been increased use of federal programs designed for rural and underserved areas by major medical centers. In some cases the programs were abused by a few to the point of near termination as in Rural Health Clinics, the termination of other forms of cost-based funding, and major reforms and regulations. Abuses in billing in a few medical schools have forced increasing regulations which have made it most difficult for the training programs in primary care or those with smaller numbers. Areas outside of major medical centers often do not have the expertise to apply for the more and more complicated forms of funding. Major medical centers have more than enough expertise to adapt to all funding methods, including rapid adaptations to prospective payment and managed care. Major medical centers have mastered federal and state shortage designations, use of National Health Service Corps scholarship and loan repayment physicians, bonus programs designed for shortage areas, and J-1 Visa positions (to meet increasing faculty shortages in medical schools).

What is perhaps most disturbing is that the nation has made the same “trickle down” mistake twice, fails to support the primary care and distributional areas that have been successful, allows health policy to distort primary care career choices and training investments, and continues with the same workforce research that fails to provide leadership with the information needed to make good decisions, especially decisions for those in most need of health care. This is unlikely to change until the nation has a solid primary care foundation.  

Graduating more total physicians failed to result in improvements in rural and underserved areas. Graduating more NPs and PAs has also failed. Both failed for the same health policy reasons.

NP and PA programs were created to increase primary care, rural, and underserved workforce. Health policy and major medical center focus (academics, graduate degrees, program location, jobs, funding for training) have resulted in NPs and PAs moving to secondary and tertiary care and major medical centers and away from primary care.

Admitting more students with origins outside of major medical centers that have the greatest choice of family medicine and graduating more family physicians by specific programs, schools, statewide efforts, and national health policy has worked consistently.

NP and PA programs remodeled back to distributional admissions, training locations, and missions could return to primary care contributions, but this is unlikely given current health policy and domination of PA and NP programs by major medical centers, even if policy temporarily favors primary care choice. Only PA program growth keeps the level of primary care and rural PAs at a stable annual number but fails to grow with increasing population, increasing underserved population, or increasing primary care need.

Any support of workforce research for objective research would help. Workforce research that moved beyond promotion of any agenda or funding mechanism or particular discipline or type of primary care practitioner would be helpful. Workforce research that integrated the needs of the broad range of America with child development, education, other infrastructure areas, and that focused on making the nation more efficient and competitive from basic primary care and public health delivery to top level research developments would be ideal.

Primary Care Workforce Outside of Major Medical Centers

Pediatrics

Current indications are that generalist pediatrics will remain stable, but will also remain 69% in major medical centers. Pediatric demand is in balance with pediatric workforce but maldistributions are significant. (Committee on Pediatric Workforce⁾¹⁹ Pediatric perspectives are greatly needed to guide the nation in child development. The voices of these child advocates are lost in the current health and education decisions of the nation. The tragedy of pediatrics is that many enter medical school specifically for generalist pediatric careers, but endure much training that is wasted or less appropriate. A six year program with a generalist pediatric focus would provide the same or better training at less cost. Accelerated family medicine graduates had early commitment to family medicine, older age, specific focus on family medicine training (starting in the third year), and top levels of distribution with 6 years of training. Pediatrics training could also target training outside of major medical centers with any flexibility from accrediting bodies. Pediatricians could also provide significant collaborative support for complicated pediatric patients for those outside of major medical centers. Complicated patients could be focused on certain days in certain small towns for collaborative care involving local physicians, pediatricians, nurses, and others involved in the care of such patients. Some pediatric partnership demonstration models in Nebraska and in other states have combined a variety of dental, mental, and health needs of children. Healthy Tomorrows Partnership for Children Program ²⁰ Geriatric physicians are 75 – 85% locked inside of major medical centers. Team geriatric efforts are heavily supplemented by donated residents, pharmacists, social workers, and other personnel. Others rely on temporary grant programs. There are geriatric outreach programs that continue to serve rural areas and practitioners and also help to educate medical schools about real world health care needs.

Adult Primary Care

For primary care beyond major medical centers and involving adult primary care, the situation does not look promising. Other than family medicine, fewer physicians escape major medical centers and limited physician distribution is likely for urban areas outside of major medical centers, urban underserved areas, and all rural areas.

For family physicians in many ways, the real training begins after departure from major medical centers where they will spend most of their careers. They learn to support themselves and their patients and communities. Those that manage to do so will also stay outside major medical centers. Those not able to do so or with connections to major medical centers for a variety of reasons will return. Others chafing at the confines of major medical centers will leave for more welcome family medicine territory. A constant equilibrium is maintained.

For those in other specialties that most often do not leave major medical centers, there is little stimulus to shape person, place, or the practice of medicine. The revered ground lies within major medical centers and more are connected to such centers. They will return consistently over time.

Surveys indicate that internal medicine primary care retention will begin with only 20% in generalist primary care at graduation (Garibaldi 2005) 3  with further declines likely over time. Returns for specialty training are likely in the youngest, Asian, foreign born, and those graduating from the top MCAT medical schools. Office Based Proportions in IM, FP, Pediatrics²¹ The nation is also admitting more Asian, foreign born, and higher income types and the MCAT score for admitted medical students continues to increase yearly despite scoring held to 1993 standards (1993 standards according to Ellen Julian VP for MCAT). Health policy, admissions, and major medical center training will continue to drive subspecialization. Older graduates and those from schools graduating more family physicians retain more in primary care.

Medicine pediatrics residents and new practitioners rapidly fall away from primary care and within 10 years the levels are below 10%. Only 14 graduating in 1987 can be found with a primary specialty as medicine pediatrics compared to the 156 in the 1987 match. Medicine pediatric levels are 20% for 1987 – 1990 graduates and 50% for all 1987 – 1999 graduates. About 1580 remain out of 3100 who matched into medicine pediatrics and most of these are recent residency graduates. These figures do not consider subspecialty training common in medicine pediatric graduates and the primary care workforce losses for time spent in the care of subspecialty patients and losses due to major medical center duties. Any other primary care choice would have resulted in more primary care workforce. The loss of any rural born, lower income, older, Black, and Mexican American medical students to medicine pediatrics instead of family medicine or pediatrics is particularly tragic given their enhanced distribution rates (over 27%) to rural and to underserved areas.

Few have explored why and how primary care can survive outside of major medical centers. Family medicine involves a higher volume of patients and often a greater number of hours, especially compared to physician assistants. Family medicine outside of major medical centers often includes procedures, emergency room, nursing homes, and hospital care. Some get public health funds or pay for on call time. Others have cost-based salaries that can be higher as in Rural Health Clinics. The broad scope of all types of patients of all ages, socioeconomic levels, acuity levels, and locations adapts family medicine to more locations. Locations outside of major medical centers also have lower cost of living and this also means lower cost of office space, personnel, transportation, and office supplies. Problems are likely to continue and increase, especially for some areas. These include increased liability insurance and other costs, decreased ability to do obstetrics and endoscopy and other procedures, loss of the more routine and less complex patients, increasing costs of living, and decreased ability to negotiate the best reimbursements or lowest costs as a smaller entity means real problems.

What is more and more apparent is that primary care is not able to survive outside of supported positions in major medical centers without significant adaptations to increase reimbursement and minimize costs.

The Dependency Status of Primary Care

Primary care pediatrics and internal medicine components are highly dependent upon the support of subspecialists sharing the same department and a chair who attempts to retain some balance. This may also make it difficult to assert issues important to primary care within departments or medical schools. Almost 100% are trained in major medical centers and few know any other form of existence. This is reflected in the comments of medical education leaders (Why Doctor's Don't Go Where They Are Needed) ²² and the composition of workforce conferences where those attending graduated from elite schools or currently work at elite schools.

Family physicians are trained in major medical centers. Family medicine programs and departments are dependent upon support from medical schools and major medical centers. Family medicine leaders and leadership training efforts depend upon government programs funneled through major medical centers and other major medical center support. Program directors, department chairs, administrators, association leaders, and those involved with accreditation all depend upon sources outside of primary care. With the passing decades few remain that have had any experience other than dependent positions. Family medicine residents face the constant challenge of trying to balance hospital rotations, ambulatory clinics, and obstetrical care. Family medicine leaders and faculty that assert family medicine needs in areas that might conflict (privileges in obstetrics or endoscopy, primary care training) may risk major medical center support. Family medicine has had to constantly fight to be sure that federal and state funds earmarked for family medicine training actually are used for family medicine training. 

No “firewalls” exist between the primary care functions outside of major medical centers and the major medical center priorities. As more and more health care support goes to major medical centers steadily with each passing year, support for primary care outside of major medical centers declines. Medicaid is a primary example. Largely created for poor young people who are nearly 75% of those served, the major expenditures of 75% now go to the elderly, those in chronic care, and those near the end of life. This has been a slow steady shift away from primary care and areas outside of major medical centers to secondary and tertiary care and major medical center locations. Medicare and the various state and federal support programs are all fair game for major medical centers.

In some ways it is difficult to blame MMCs. Federal and state support is inadequate to address all of the various areas: indigent, uninsured, underinsured, the costs of training, and the unique missions. It is also difficult to do well when you are training more and more of your own competition in the form of subspecialists and those converting from primary care to subspecialties.

Primary care problems will continue to exist until primary care achieves some element of separation and identity and priority in the nation’s health care scheme including reimbursement priority, adequate support personnel, primary care control of accreditation, a focus on preparation for primary care practice, and separate primary care training funding. Medical educators outside of the United States recognize this and even a few who have studied rural medical education, (Robert Talley, former Dean in South Dakota) but those inside the United States have never seen anything different than the current system.

Choices for the United States

Nations have choices. When the short term decisions of a nation appear to fail, there are methods to make longer term decisions. Defense and public welfare are such areas. Social security, Medicare, Medicaid, and other areas have also been set aside as long term priorities. Child development, primary care, public health, and education have suffered, squeezed by other areas. The nation needs to move these areas beyond year to year control.

Best Choice

The best choice for the nation (and one that would help increase physician distribution) is improvements in child development and education. Later efforts in college or high school are inefficient (Hart and Risley 1995) ¹⁷ and there is evidence that the nation has inadequate education to prepare enough medical students. (Cooper)²³ Improving child development and improving education tend to impact mainly lower and middle classes, but these changes also impact the careers chosen by such groups and their distribution. When the most geographically and socioeconomically different do well, the nation distributes professionals. Inadequate levels of lower income, Black, Hispanic, rural, Native, and middle income students make it difficult to admit the students most likely to distribute where needed as professionals. Perhaps a more pressing matter is that our complex most urban areas demand a cohesive, informed, educated group. Not having hopelessness breed terrorists inside our own borders may be as important as those outside. Hopelessness also results in more costs in all facets of society.

Those who are on the lower income and middle income site that avoid hopelessness and that continue to be motivated by education and future income potential include those who will become service oriented professionals such as school teachers, nurses, counselors, those serving as public servants of all types and the military, and family physicians.

This is also an approach likely to improve the general efficiency of American society, reduce divisions, improve health care, and decrease future costs of education, health care, unemployment, prisons, and social programs.

The major problem with such an early focus is that the impact is far later than any political career. Everyone wants quick fixes. It is an investment in the next generations of children and the benefits are likely to be attributed to the next generation of politicians. It is also a focus that does not appeal to the highest income groups who provide their own child development, private schools, and top college education who have forgotten how important building and revising and rebuilding infrastructure is. The lack of public investment in medical schools in some ways is responsible for reversals of affirmative action. A public with enough investment to admit rich and poor medical students or law students would not fight about who got in. Lack of public investment makes it difficult to admit all income groups and the nation has spoken regarding who is likely to win the court, political, economic, and voting battles involved.

Only a nation that understands that it is only as strong as its weakest is able to overcome its own limitations.

Taking the development of infrastructure out of the hands of year to year and month to month issues should be a top priority. Guidelines that would move federal and state funds into the areas of most need would kick in with lack of improvement in outcomes or with worsening outcomes. Investment in Age 0 – 6 child development now at 0.5% of GDP would move first move incrementally closer to the 2% spent by Denmark. Further changes would be determined by outcomes in education, college completion, prison rates, and unemployment.

The nation’s current choices reveal priorities. A focus on science and math at high school levels is an investment in the top 30% who still remain interested in education and higher education at this point in time. Earlier investments such as universal preschool and enhanced Headstart (not the current sunsetted programs) move spending away from the 30% highest income and toward the 70% lower and middle income groups that need a better start. The 30% highest income can continue to spend the tens of thousands that they already spend on child development, but the 70% would have some level of matching investment.

Critics often point to failures of increased expenditures to make a difference. They often point to areas such as Washington DC where massive funding increases have not improved outcomes. There is a reason for such extreme examples to make the case of no hope. Washington DC may well be the most inequitable place in the United States. The highest income quintile income is over $210,000 and the lowest makes $7000. This is a 30 to 1 ratio that has increased each decade from 12 to 19 to 28. These are the highest and lowest income quintiles in the nation. Not surprisingly those form the Washington area have the highest and lowest education levels, infant mortality rates, medical school admission rates, longevity, abortion rates, and more. The areas that divide the nation the most are all represented just zip codes apart. The situation is particularly tragic for the lowest income chronically poor. They are surrounded by the symbols of freedom, independence, and opportunity and watch as those born in other nations move in and move up and move out within a few short years while they and their children remain behind with little hope of moving up. The major vote regarding Washington DC has been the departure of population. Washington DC is one of the leaders regarding shrinking major county populations with losses of 5 – 15% each decade for 40 years. Fifty-four medical schools are in counties that have lost population at a time when similar areas over 1 million have gained over 30%. Those not connected by family or ownership of housing or highest income jobs have been leaving for decades. Bright Future Rankings

Comprehensive efforts addressing multiple facets of education, public safety, health, social, and housing needs simultaneously, at the earliest ages, and for the long term are the only solutions for areas of chronic poverty, in rural or in inner city poverty areas.

Even the most subspecialized physician is familiar with the course of recovery of a seriously ill intensive care patient. The recovery of such a patient requires an intensivist general coordinator with subspecialists working on all the major systems that are damaged. Organ systems are stressed to the point of failure and reserves are tested. Progress is slow and complicated by setbacks in a variety of areas, sometimes unpredictable. For those without health insurance, the recovery of the body may be impaired or delayed by the lack of resources and support to complete the process of repair and restoration.

For areas of chronic poverty, overall coordination and leadership must oversee the recovery of the neighborhood (and family) and its organ systems of education, public safety, housing, economics, etc. The efforts require months or years and great patience and diligence. Setbacks are frequent and sometimes unpredictable. The decades of neglected infrastructure that have resulted in the failures require decades of infrastructure for a return to productive function.

The lessons of our major cities, Katrina, and Iraq all demand that we learn the ultimate importance of infrastructure in preventing disasters, improving the response, and hastening recovery.

The Choice for Primary Care

Another important choice must be made to develop a real primary care component in the health care system. Subspecialty care must take a secondary and tertiary role. Separation from the current situation will require a number of years and much cooperation and this will take much leadership. The nation will need to steadily increase funding for primary care and for patient populations outside of major medical centers. This could involve new funds as in the 1970s but this is unlikely with current deficits. Steady decreases in subspecialty reimbursement or reimbursement held at steady levels while increasing primary care funding focused outside of major medical centers would improve distribution. This is also an area that needs to be moved beyond political or medical association or advocacy manipulation.

There is no need for the abrupt changes of the managed care era. This natural experiment confirmed earlier health policy regarding emphasis on primary care. It also demonstrated that the nation could control graduate medical education positions with effective movement of physicians to needed career choices. With the current medical school expansion, a limitation of GME positions except in family medicine and psychiatry would begin the process. The limitations would be in the hands of the medical students, medical schools, and admissions committees. With decreasing choices in family medicine and psychiatry, more pressure would be brought to bear on changing admissions and training. With increasing choice of family medicine and psychiatry, other graduate medical education positions could be allowed to expand. The increases required should continue until the nation has clear evidence of improved health care access.

This is not a difficult task. The lesson of Critical Access Hospitals demonstrated that a few percent redistributed to rural hospitals could mean the difference between closure and continuation, between access and limited access.

Given the low percentages of rural physicians needed (increase from 11% to 13 or 14%) and of underserved physicians needed (increase of 7% to 10%), only a few percent more are all that is needed. This must be a reliable steady contribution. Since family physicians stay in primary care and distribute and remain in distributional locations, this is a preferred method with the least disruption, the greatest probability of reaching the desired goals, and the greatest probability of staying at target distributional goals for the long run. More temporary rural and underserved practitioners staying a short time in primary care and needed areas only dilutes the concentrations further and defeat earlier efforts.

Finally the nation has a choice in medical education. The nation does not need huge numbers of family physicians, but it does need enough to meet current needs and enough to grow with anticipated population changes. The nation needs a stable base for primary care and family medicine fits this better than the rest. There was some evidence that past primary care physicians plus the managed care and health reform impacts plus the PA and NP numbers dating from programs with different emphasis may have saturated primary care needs. The health policy-related declines in primary care practitioners in all but family medicine have also created additional positions to fill and the losses are likely to be in the least supported, most rural, least organized, and most underserved locations.  

Distribution involves selection of those outside of major medical centers (especially those who have an interest in returning outside), training focused on outside, and health policy focused on support of practice outside of major medical centers. This following suggestion involves a typical physician focus on health care, but with emphasis on what is working to distribute physicians.

• The nation should halt the current undisciplined expansion which will benefit major medical centers far in excess of those in need of health care and in need of economic impact in the nation. These are peoples and populations largely outside of major medical centers. The nation needs more physicians, but it should not expand in areas guaranteed to graduate subspecialists for major medical centers at great excess compared to those needed for primary care and areas outside of major medical centers.
• The nation should consider expansion only in the schools or involving the types of programs that are distributing physicians. Any medical school can graduate subspecialists. However the schools graduating the most subspecialists have great difficulty distributing physicians. Those that have figured out how to meet state and national needs should be allowed to expand and should be encouraged and supported to do so.
• The best models for expansion or for replication are osteopathic public medical schools, a model ignored since 1978. All six make sustained top contributions to rural areas, underserved locations, primary care, and family medicine.
• Allopathic public schools with a focus on rural and underserved admissions, training location, and distribution have also done well. Historically black schools and those admitting increased percentages of Black, Hispanic, and Native students combined with increased choice of family medicine are important. Some medical schools in California, Texas, North Carolina, New Mexico, and Washington State have earned a reputation as being able to distribute physicians where needed. 
• Statewide efforts involving coordination at multiple levels have played a key role in states such as West Virginia, Arkansas, Iowa, North Carolina, Minnesota, Nebraska, and North Dakota. Distributions of education may also play a key role. Even so the inner city populations and the 7 million in lower income rural minority areas (border, black, and tribal) are underrepresented in admissions and in physician distribution.
• Standardized tests designed for upper income students fail to predict performance in those most different in multiple complex and significant ways (rural plus low income, low income plus language, culture, background, parent education, geographic distance, and more). Each of these characteristics contributes to enhanced probability of physician distribution, each difference contributes to a lower standardized test score impacting all levels of placement, and each difference contributes to lower probability of admission. Each also has a higher probability of medical school failure, but no admissions evaluation can identify the individual who will fail. Anticipating failure is one of the major failures of our time regarding lower and middle income peoples.
• A focus on standardized tests is recognized as a significant narrowing influence at more colleges across the nation. Medical schools with lower MCAT scores (placing a focus on other characteristics) admit the most different types of students and distribute the most physicians. The problem does not lie with the standardized tests. The problem involves failures in interpretation. To assess the more and more diverse students in the United States takes significant expertise in understanding the limitations of testing. The limitations in these areas have been known since the beginning of the MCAT test. (McGaghie, Thorndike) or local review on this site
• That medical schools, colleges, and professional schools fail to take the time and effort and resources to evaluate beyond the limited contributions of standardized tests is not a matter of courts. It is a matter of ethics, morality, social responsibility, and public trust. Higher education and professional education that truly met the needs of society would not be at the lowest levels of funding in 25 years.
• Australia has reformed higher education, medical education, admissions, and health policy with a focus on distribution. Rural born student declines from 25% to 10% have reversed to 19%. New general practice schools and even rural medical schools complement the efforts earlier in the pipeline and support for rural practice. (Australian Workforce Reports)
• Other nations are passing the United States in education, higher education, health care costs, health care quality, and efficiency. No longer does the nation enjoy the same advantages it once had. Rather than looking later and at higher expenses, the nation needs to vote early and vote often, for children.

New models to consider: 

• Admission tracks of older students given a chance to prepare and perform and then enter medical school.
• Six year schools specifically for generalist pediatrics, psychiatry, and/or family medicine that require over 10 years of service with rigorous requirements – Why have 20 – 40% family medicine when 100% is possible? Aligning career interest to training is possible in careers where medical students have commitments before admissions. This tends to be at either end of a academics and socioeconomic point of view, for the most service oriented as well as those most interested in research.
• New schools, branch schools, or conversions of existing schools to admit specifically lower income origin and older students, similar to Historically Black schools. This may include schools that train in languages other than English that would facilitate more physicians of other languages, but would also prepare native English speakers to meet important needs. Medical schools with long histories of meeting diverse needs and reaching out significantly to education and career preparation and primary care emphasis should receive priority for such considerations (UCLA, UTMB Galveston, UT San Antonio, the University of New Mexico, Brody-East Carolina, Morehouse, Howard, Meharry)   
• Working toward the end of all private medical school designations and requiring all medical schools to be responsible for a certain workforce, region, city, or state in health outcomes impact. This would also come with the requirement of significant support from states and cities matched by federal dollars and should have the potential enforcement with requirements and penalties, including reduced class sizes or even termination. The expertise of all our professional schools should be brought to bear on our most difficult societal problems, not allowed to fester in the neighborhoods right across the street. Academic Medicine's Season of Accountability and Social Responsibility
• Required commitments for all medical school graduates of 10 years service from research to academics to underserved to military to public health. Those accepting admissions should all be prepared to serve the nation and need to make sure before graduation that this is what they desire to do. All will have obligations to serve the nation in times of crisis and this could include disasters, pandemics, or meeting the needs of populations. Ideally improvements in physician distribution will decrease the need for obligations.

Primary care should not be a loss leader subject to the changing tides of health policy. Primary care, public health, and public education should focus on driving an efficient economic and innovation engine that has no equal, a cornerstone of a national model where all have a chance to be treated equitably. Any thing less was not in the minds of the founding fathers or the framers of the Constitution. We have also managed to descend to their worst fears. That it has taken 200 years for us to learn how to convert public coffers to the use of the privileged is a testament to the design, and the need for serious and widespread reforms.

The world no longer allows leaders to hide the worst and present the best. The world actually focuses on the worst inequities in any nation. Those who profess to be the best, better become the best, and better encourage nations to become better.

Bibliography for Five Periods and for Differences and Definitions

1.         Xu G, Hojat M, Veloski JJ, Brose J. A national study of factors influencing primary care career choices among underrepresented-minority, white, and Asian American physicians. Acad Med. Oct 1996;71(10 Suppl):S10-12.

2.         Bowman RC. The Distributional Student, the Distributional School, and the Distributional Specialty. pending acceptance, from Rural WONCA Plenary and Birth Origins Research, Rural and Remote Health. 2006.

3.         Garibaldi RA, Popkave C, Bylsma W. Career plans for trainees in internal medicine residency programs. Academic Medicine. May 2005;80(5):507-512.

4.         Bowman RC. Primary Care Retention.  http://www.unmc.edu/Community/ruralmeded/primary_care_retention.htm. Accessed October 26, 2006.

5.         American Academy of Physician Assistants. Data and Statistics.  http://www.aapa.org/research/index.html. Accessed October 26, 2006, 2006.

6.         Bowman RC. Retention Within the Specialty of Family Practice.  http://www.unmc.edu/Community/ruralmeded/retention_within_fp.htm. Accessed October 26, 2006.

7.         Rosenblatt RA, Andrilla CH, Curtin T, Hart LG. Shortages of medical personnel at community health centers: implications for planned expansion. Jama. Mar 1 2006;295(9):1042-1049.

8.         Bowman RC. Birth Origins Database from American Medical Association Masterfile Database Using OfficeMax Practice Locations; 2005.  Birth Origins Articles

9.         Association of American Medical Colleges. Minority Students in Medical Education: Facts and Figures XII Available at https://services.aamc.org/Publications/showfile.cfm?file=version12.pdf&prd_id=89&prvid=87  Accessed April, 2005. Washington DC 2002.

10.       Hart G. WWAMI Rural-Urban Commuting Areas (For Residential & Commercial/Geographic Zip Codes: Version 1.1). Available at www.fammed.washington.edu/wwamirhrc/rucas/descript.html. 2005;Accessed April.

11.       Bowman RC. Family Medicine: the Distributional Specialty.  http://www.unmc.edu/Community/ruralmeded/FM_distribution.htm.

12.       Pugno PA, McGaha AL, Schmittling GT, Fetter GT, Jr., Kahn NB, Jr. Results of the 2006 national resident matching program: family medicine. Fam Med. Oct 2006;38(9):637-646.

13.       Association of American Medical Colleges. Minority Students in Medical Education: Facts and Figures XI Available at https://services.aamc.org/Publications/showfile.cfm?file=version12.pdf&prd_id=89&prvid=87  Accessed April, 2003. Washington DC 1998.

14.       Association of American Medical Colleges. Minority Students in Medical Education: Facts and Figures XIII Available at https://services.aamc.org/Publications/showfile.cfm?file=version53.pdf&prd_id=133&prv_id=154&pdf_id=53 , Accessed July 2006. Washington DC 2005.

15.       Greene JP. High School Graduation Rates in the United States.  http://www.manhattan-institute.org/html/cr_baeo.htm#11 <http://www.manhattan-institute.org/html/cr_baeo.htm>, Accessed August 2005 at http://www.manhattan-institute.org/html/cr_baeo.htm#11 <http://www.manhattan-institute.org/html/cr_baeo.htm>.

16.       Carnevale A, Rose SJ. Left Behind: Unequal Opportunity in Higher Education, Reality Check Series. In: Kahlenburg R, ed. New York: The Century Foundation Press; 2004:p. 106.

17.       Hart B, and Risley, T. Meaningful Differences in the Everyday Experience of Young Children. Baltimore: Paul H. Brookes; 1995.

18.       Association of American Medical Colleges. Medical Student Education Costs and Student Debt Available at https://services.aamc.org/Publications/showfile.cfm?file=version35.pdf&prd_id=121&prv_id=137&pdf_id=35   Accessed July 2006. Washington DC 2005.

19.       Committee on Pediatric Workforce. Pediatrician workforce statement. Pediatrics. Jul 2005;116(1):263-269.

20.       American Academy of Pediatrics. Healthy Tomorrows Partnership for Children Program.  http://www.aap.org/commpeds/htpcp/descriptionsbystate.html. Accessed October 26, 2006, 2006.

21.       Bowman RC. Office Based Proportions in Internal Medicine, Family Medicine, and Pediatrics.  http://www.unmc.edu/Community/ruralmeded/office_based_proportions.htm.

22.       Cohen JJ. Why doctors don't always go where they're needed. Acad Med. Dec 1998;73(12):1277. Copy on site with critique.

23.       Cooper RA. Impact of trends in primary, secondary, and postsecondary education on applications to medical school. I: gender considerations. Acad Med. Sep 2003;78(9):855-863.

24.       McGaghie, Assessing Readiness for Medical Education: Evolution of the Medical College Admission...
JAMA.2002; 288: 1085-1090. Local review here

25.       Australian Workforce Reports

Rural WONCA Plenary 2006 Disturbing News from the United States - powerpoint presentation

Managed Care Comparison Table

Distributional Analysis Policy Center

Distributional Choices and Health Policy

Legislative and Health Policy

Physician Workforce Studies

www.ruralmedicaleducation.org