Recent Black Family Physicians: Origins, Age, and Locations

Robert C. Bowman, M.D.

Question from Stephen Wilson on Family L list serve:

When comparing Black US Born and Foreign Born US MD Grad medical students

1.      Do Foreign Origin US MD Grads have lower choice of family medicine?

2.      Do Foreign Origin US MD Grad family physicians distribute at the same levels as US Born graduates?

Answer:

1. There is no difference across US or international birth origins in choice of family medicine for the 1997 – 2003 graduates of family medicine residencies from US allopathic medical schools.
2. When considering all of the various urban underserved contributions, rural contributions, military, and teaching contributions, there is very little difference for Black family physicians by US or foreign birth, by rural or urban birth, or by type of medical school. Each of the various types of Black family physicians contribute in important areas of physician workforce. When considering the sum total of rural, urban underserved, teaching, and military careers as all important workforce outcomes, there is little difference in overall contribution. This may reduce down to the final consideration, which is that those who decide for family medicine are deciding for distribution.
3. US MD Grads younger than 26 at graduation from medical school and those graduating from highest MCAT medical schools have lower levels of distribution but have greater contributions in academic areas.
4. The few rural origin graduates have superior rural distribution. The most urban born are rarely found in rural locations but make greater contributions in urban underserved locations.
5. Actual comparisons by parent income level, by parents who are professionals, or by individual MCAT scores may demonstrate differences but this primary data is not released for study.
6. Older graduates do have advantages in rural and urban underserved distributions and are more likely to be found in the military. The military package is likely to be the best fit for an older graduate who is married with family, however direct studies are needed to confirm this. Military and academic careers in family medicine are more likely to involve rural, lower income, and older graduates who otherwise have consistently the highest distribution levels.

Introduction

Medical student career choices are shaped by factors as early as parent occupation to birth origins to education and college to medical school admissions and training. Medical students from higher income origins, elite schools, urban origins, and professional families are less likely to choose family medicine and are less likely to distribute to rural, underserved, and primary care careers. They also tend to be the youngest at medical school admission.

Foreign born medical students have the lowest choice of family medicine, primary care, and rural locations. Foreign born US MD Grads appear to be combinations of these origins and ages which may explain low levels of family medicine and distributional choice.

Which factors are strongest in career choice and distribution?

Family Medicine Choice, Physician Distribution, and Black Family Physicians: Health Policy

Career choices and locations are shaped by other factors. Health policy is one shaping factor. Career choice in family medicine is particularly sensitive to health policy involving primary care reimbursement and priority on support of practice locations outside of major medical centers. During the period of 1965 to 1978 and during 1992 to 1997, family medicine and primary care choice maximized as did physician distribution to rural and to underserved areas. The primary impact upon family medicine choice is within medical school years. For internal medicine the primary impact is during residency training. When medical student increases in family medicine choice are common, retention of internal medicine graduates within generalist office-based careers is also more common. When categorizing by medical school class year, internal medicine retention peaks are in class years 3 or 4 years earlier compared to family medicine.  Five Periods of Health Policy and Physician Career Choice

The differences are obvious to medical students, but not to workforce researchers. Family medicine is different as it is a permanent choice of primary care. All other physician and non-physician forms must continue to make a choice to remain in primary care with each passing month of practice. Those who desire to hedge their bets will choose a transitional form of primary care. In distribution neutral or supportive health policy, family medicine will do well. In the absence of primary care support or support for those outside of major medical centers, family medicine choice will decline. There is also evidence that declines in choice in family medicine are more likely to involve Black and Hispanic medical students. These are students that have made some of the earliest decisions for a medical career and are admitted to medical school later. After spending additional years to gain admission, it may be that these students are looking for health policies that will support their permanent primary care career choice. Historically black medical school students and students in the medical schools with the most Black and Hispanic students have had the greatest decline from 1997 graduates to 2005 graduates. One hypothesis worth testing for all lower and middle income students: after fighting against barriers of income and education for decades, it may well be that some reassurance on the part of government is needed such that more decades of systemic battles are not likely with choice of career. At the current time period and especially for urban family medicine, the reimbursements are low and the costs of practice and living are great and growing. Declines of family medicine choice below the level of distributional type students may indicate that the nation has entered another collapse of health policy eras similar to the time period before 1965. Five Periods of Health Policy and Physician Career Choice

Family Medicine Choice, Physician Distribution, and Black Family Physicians: Matching Ethnicity, Race, Gender and Pratice Location

Another factor involves considerations of the locations where particular graduates will practice. The best matches of patients and physicians clearly improve quality or at least the perception of quality. Better communication may well prevent unnecessary costs as well. Access issues also are important in a complex career such as primary care. Rookie family physicians in rural or in urban underserved areas can make contributions, but the contribution is far less than an experienced family physician who has had a few years to learn the patients, the system, the support personnel, and the referral sources. Estimates of $200,000 in losses for primary care turnover (Buchbinder) may be vast underestimates in underserved locations where losses involve teamwork, institutional memory, experiential learning, and other areas more than just recruitment, lost productivity, and orientation.

Matching race, gender, and ethnicity is a challenge for Black and Hispanic populations since population growth is much greater than the growth of Black and Hispanic physicians. Increases in Black and Hispanic foreign born physicians may help to address certain needs, but do foreign born physicians choose family medicine, choose primary care, and stay in primary care?

Rural populations are growing also but rural born admissions are declining, especially for the males that are found at 15 – 25% greater rural location. Foreign born physicians are rarely rural born, with the exception of nations such as Canada or others that distribute income and education at higher levels and speak English. Even in America declining health and education policies disadvantage rural populations to the same degree as urban underserved populations. Rural born students had the same 40 – 50% probability of medical school admissions compared to Black medical students and the same greater admission rate for females compared to males. Admissions for lower income rural areas was similar to Hispanic students.

Matching up rural and urban backgrounds with patients may also be important, but quality studies have failed to progress beyond gender and race to class and geographic origin. There is little doubt regarding the findings however. Matchups have already played a key role in workforce distributions. Female primary care physicians have been a recent development and still receive a top priority in major medical centers and urban settings. There may well be greater demand than current supply. Major medical centers train 100% of physicians and get the first chance to recruit graduates. Urban served areas share training duties and offer attractive suburban locations. Females who marry physicians or professionals may also prefer the urban locations.Lack of training in rural areas makes it difficult to find rural spouses. Black females may face some of the greatest challenges in attempting to match up with Black males who are less likely to be physicians or professionals, especially outside of the most urban areas. Academic careers may be a nice match for those attempting to balance family, personal, and professional life.

The most difficult matchups involve the needs of the 6 million rural peoples in predominantly Black, predominantly Hispanic, and Native reservations. This is the 10% of rural America with the greatest challenges regarding income, economics, education, and health care access. Outside of these areas rural America is 85 – 90% white. Over 60 years of declines of whites, males, and rural born US MD Grads are a problem for most of rural America regarding all types of rural physicians. Asian, foreign born, Black, Hispanic, and other groups are less likely long term matches for such populations. Those choosing internal medicine are even less likely to be found in rural America. Internal medicine physicians and all other primary care types other than family medicine have been returning to major medical centers and urban locations and subspecialty training for over 50 years. Generalist pediatricians do not tend to subspecialize over time, but 75% are consistently found in major medical center locations. Rural born pediatricians and other specialists may maintain a different equilibrium since about 14 – 20% are found in rural areas consistently over time. The only consistent sources of rural physicians are rural born physicians and family physicians.

Family physicians do not have the same major medical center location or focus as other specialties. This is a likely result of the advantages and disadvantages of family medicine. In some ways family physicians are excluded from major medical centers. In other ways family physicians prefer locations outside of major medical centers. When tracking family physicians over the period from birth to medical school to residency to practice, family physicians leave states with high concentrations of major medical center physicians and move toward states with the fewest major medical center physicians. FP docs are found in the lowest concentrations at 7 – 9% in urban major medical center locations nationwide. In urban served and underserved locations FPs are 16% of physicians, in rural areas FPs are 30%, and in the rural low income and CHC settings (Rosenblatt, Jama) FPs are 50 – 70% or more.

Exclusion may also apply regarding other settings. Black and Hispanic family physicians may be more comfortable in urban underserved locations with predominant Black or Hispanic populations. Retention levels within 60 miles of FP residency training are highest for Mexican American family physicians. Whites may not be retained for prolonged years in Native reservation locations or inner city locations without significant white populations. Rural locations, isolated rural locations, and rural underserved locations are dominated by whites and Natives, populations with the most rural born physicians.

Underserved urban zip codes have a CHC, an NHSC physician, a whole county primary care designation, or serve a population of 20% or more in poverty. Urban underserved location is 3.2% for all physicians, begins at 3.8% for white family physicians, increases to 6% for Asian FPs, then 10% for Other Hispanic and Native FPs, then 14% for Black FPs, 17% for Mexican American FPs, and over 20% for Puerto Rican FPs. Some needs of urban underserved populations are met by major medical center and urban served physicians, but these are locations that decide the types of physicians and the types of patients accepted. Urban underserved locations have higher concentrations of poverty, are more dependent upon federal and state efforts, have decided upon increased concentrations of primary care. Urban underserved locations are the best indication of specific underserved care in urban areas.

Foreign Born Considerations

Foreign born US MD Grads are increasing. Asian US MD Grads share the characteristics of Asians in census reports and are 90% foreign born or have a parent who is. Asian and foreign born components of US MD Grads are over 30% of current graduates. In general Asian and foreign born US MD Grads have the lowest choice of family medicine, primary care, rural locations, and underserved areas. They tend to have parents who were most urban, most educated, highest status, and professionals.

The Historically Black medical schools have increased admissions of foreign born medical students of recent decades, but overall at rates less than other private medical schools. Howard has increased from 20 to 40 in each class or from 15% to 30% from 1987 to recent graduating classes. Meharry has remained about 15% foreign born. Morehouse has admitted 3 or less in 12 of 17 class years since 1987. Since 1970 the Black medical students and foreign born Black medical students have a much wider range of medical school locations. In recent years the Black medical students have had fewer opportunities outside of a select group of "distributional" medical schools that focus more on admissions of individuals rather than their scores. Admissions changes in the United States have been most apparent by income level with 3000 more highest income admissions replacing 1500 lowest income and 1500 middle income medical students now compared to 1997 matriculants.

Admissions Ratios, Changing Admissions, and Physician Distribution

Age is a key factor in elite or humble origins. Those less than 26 at graduation tend to be higher income, most urban, and foreign born medical students. Age 26 is the most common graduation age with 30% and a normal distribution of various types and origins of students. Age 30 and above represents students delayed in admission by obstacles of income and education related to parents and origins and those with second careers in medicine. Historically black medical schools have some of the older allopathic graduates and in recent years the average age has been increasing. This may represent a shift of Black students away from schools that once admitted more Black students back to the historically black schools and the few schools that clearly admit the widest range of students of all origins, ages, and income levels.

One other factor does seem to impact choice of family medicine. Different types of medical schools and different regions of the nation have greater choice of family medicine. The Midwest has greater FP choice followed closely by the west. The south and the Northeast have lower choice of family medicine. Higher MCAT schools have lower choice of family medicine. Populations with eastern, northeastern, southeastern, or east central distributions tend to have lower choice of family medicine. Because black, Puerto Rican, and Other Hispanic populations are eastern in location, FP choice would be expected to be 11 or 12% rather than an average of 14%. Vietnamese and Mexican American populations are located in areas and medical schools with greater family medicine choice. Asian populations are more likely in the most metro areas and Asian populations concentrate on the east coast and west coast in these locations.

An unknown factor is local and state economics. Declines in inner cities are present in most major metropolitan Midwestern and Northeastern states. These are counties that have been losing 5 - 15% per decade. The impacts on education, education policies, reversals of affirmative action, and local economics are significant. Opportunities for Black students may be impacted by these changes.

Finally although family medicine choice is stable across class years dating back for decades, the earliest years after the graduation of any physicians are subject to obligations for military or underserved location. In family medicine this tends to balance out, but this is also a group with the most atypical career and location choices in the history of US workforce. The 1997 – 2003 family medicine residency graduates are generally the same population as the 1994 – 2000 US MD Grads. This group had maximal family medicine choice for the 1995 – 1997 class years and rapidly declining choice until the 2000 US MD Grads. The impacts on workforce are more difficult to predict in this situation. After 2000 the declines in health policy make predictions even more difficult. Below an unknown baseline of health policy support, declines in family medicine choice may make predictions chaotic. This was seen prior to Medicare and Medicaid.

Given the background on choice of family medicine, Black physicians, and family physicians, it is time to return to the questions:

Do foreign born Black Medical Students choose family medicine?

Do foreign born Black family physicians choose underserved locations?

Methods

The following fields were accessed in the AMA Masterfile 2005 Office Max version: birth city, state, and nation of birth; self-designated career choice; ethnicity, gender, and race for family physician graduates of 1997 - 2003 only; age at graduation; and type of medical school. The Masterfile birth origins version involves cleaning and coding of origins by geography and birth county income levels and coding of zip code practice locations. The nation is divided into military, major medical center, served, and underserved locations and rural and urban coding using RUCA 2.0. In addition Historically Black medical schools and the top 20 MCAT medical schools were coded for comparison.

Results

Foreign born Black family physicians are a balance of male and female. US born Black family physicians are 60 - 70% female.

Ethnicity and race data was not provided to the author except for family physicians. By assuming that nations with the highest Black populations contribute Black physicians and checking this assumption with ethnicity and race in known family physicians, birth country estimates can be made.  

The average for FPGP choice for this time period is 14% for US MD Grads. The 15.5% choice is similar to the average and similar to FPGP choice in US born Black Physicians. The average FPGP choice for foreign born US MD Grads increases to 17.3% without Ethiopia and the 3 lower choice African nations. This is similar to FPGP choice in older and lower income US MD Grads. Only rural origin grads have greater FPGP choice and are more of a combination of rural, lower income, and older US MD Grads.

In studies of family physicians, the most exclusive populations have lower family medicine choice. This includes higher income or more elite types of students by US origins, medical school type, or birth in other nations. With increases in the percentages of physicians born in cities or counties with medical schools in the United States or in other nations, there is a decreased choice of family medicine and physician distribution. The most likely explanation is that children of physicians or professionals are a primary factor in career and location choice. The Masterfile with secondary data does not capture scores, parent income, or parent occupation. However comparisons by ethnicity and race also are consistent. Asian US MD Grads do have the highest levels of professional parents followed by White and by other groups. Within Whites the % choice of family medicine goes down as levels of professionals in counties go up.

Actually the total FP admission rate is a constant 1 per 100,000 per year in normally distributed populations, but the total medical school admissions for all specialties increases from 3 to 20 with higher levels of people, income, education, and professionals. This means a lower percentage choosing family medicine in areas with a higher rate of admission as in 1 in 4 or 25% for 100,000 rural or lowest income people compared to 1 in 20 or 5% for 100,000 people in top income or most urban areas. One consideration is that in lower income groups, individual student characteristics make more impact. In the higher income groups, the admission and the chance for admission is about parents and how they structure environments for their children and others in the area. Shaping a Nation: Physicians Who Serve

So far there is no difference regarding foreign born and US Born Black physicians.

Income Origins

Comparisons by birth income level do not reveal differences in distribution for underserved or urban underserved locations. Rural location for US born top income birth origin black family physicians is 7% compared to 11 or 12% for other origins and foreign origins. Those raised in the highest income and most urban environments have few rural experiences and their training tends to be in locations surrounded by the least rural workforce.

Age at Graduation

Comparisons of younger US MD Grads can indicate those of more exclusive origins, regardless of US or foreign birth. Younger US born family physicians have only 7% rural location and concentrate in major medical centers at 70%. These are low levels of distribution common in general internal medicine graduates. Underserved urban location remained at 6 – 7% which is twice the average of all physicians, but half of the usual urban underserved distribution of Black family physicians. Younger graduates and elite medical schools both have about 20% of the younger graduates found in teaching positions.

Perhaps the best measure of physician distribution overall is a measure of the total physicians minus those found in major medical centers. A group concentrated in major medical centers is not distributed to rural, urban underserved, or military careers. Historically Black US MD Grad family physicians are found outside of major medical centers at greater levels. Foreign born Black US MD Grad FPs from the three Historically Black medical schools give up some rural distribution but maximize underserved distribution at 26.8% while US Born Grads have greater rural and rural underserved distribution and maintain superior overall underserved distribution at 21.5%. Overall the small numbers of younger Black US MD Grads limits much analysis.

Lower underserved and rural numbers in elite schools have compensation in greater teaching positions and some military FP. In a few more years the study population can be repeated to see if academic, military, and other locations change from 2005 locations and careers.

The West Coast Distributional medical schools are UC Davis, Irvine, UCLA, and the University of Washington. The totals include the early admission schools of UMKC and NEOUCOM (not shown) that had too few Black US MD Grads to include in this table.

Black family physicians from each of the different types of medical schools make significant contributions involving underserved urban or rural distribution although the US MD Grads of schools with the highest MCAT scores seem to be the most limited. The elite schools do make other contributions, however. The Black family physicians in these schools have 20% of graduates in academic family medicine instead of a more normal 10 – 12% for other Black family physicians or an average of about 4 – 6% for all US MD Grad family physicians. The other contributions made by Black US MD Grads involve males in military family medicine. Military components increase to 8 – 12% for the Black FP graduates of the most elite medical schools, which also cost the most for tuition and cost of living.  

Limitations -

1. about 20% of ethnicity or race data is missing

2. the data on Black graduates and grads of Historically Black medical schools is delayed or missing with higher levels of unclassified, similar to Puerto Rican schools or locations and osteopathic, the Masterfile is not a completely uniform source and this may impact the most recent graduates to a greater degree

3. NHSC obligations are greater for Black and Hispanic medical students, especially in family medicine. It is difficult to know how much this shapes career and location choice and future location choice

Other Discussion Points

Point 1 We have admitted a broader range of all students in the past
Point 2 We could choose a broader range and support this choice and get better outcomes
Point 3 Few differences in important career outcomes, as long as the career choice is FP
Point 4 FPs are the distributional specialty
Point 5 Progressive losses in temporary forms of primary care make FP a most important choice

The nation has admitted up to 1400 US MD Grad black medical students in the past. The nation managed this level in just a few years from a baseline of 1000. This was a 40% increase and it was followed abruptly by a decline back to 1000 after interventions. The interventions included individual medical school efforts, state and federal efforts, and admissions training in 3000 by 2000 efforts backed by national affirmative action efforts. That was 12 years ago before significant population increases and the capacity should have increased. The osteopathic 2006 report notes that osteopathic schools admit about 150 Black medical students. Older Black medical students admitted at distributional and Historically black allopathic schools are getting older in a pattern that suggests delays in admissions. Declines in admission by income level have been dramatic in the past decade. Increases in MCAT scores have also been dramatic, but neither of these can be linked to better medical students or physicians.

The various changes indicate that more diverse students in a number of areas have been admitted and could be admitted, but American medical schools are making other choices. Overall this tells Black and other populations that the route to medical school admission is more difficult, restricted, and may not be worth pursuing. - a bad signal. This is a likely signal given to all socially distant by color, culture, or class and especially combinations of these 3 as noted by medical educators in recent years such as Whitcomb. Also huge tracts of populations in middle and lower income areas do not even know about admissions tracks, osteopathic medicine, or Caribbean routes. Medical schools without a rural track record or a record for Black, or Hispanic admissions are not likely to attract rural, Black, or Hispanic applicants. Medical schools have been able to boost admissions and medical school performance with significant medical school efforts (Crump at Louisville, U TN, 3000 by 2000) The schools made these efforts most often with state and federal support, but this support is lacking. Waiting for support to do what needs to be done is a common problem in higher education. Long term progress is about child development and a better start in life. Gross estimates are that we could increase admissions steadily to about 1800 a year and then watch applicant pools, performance, and age at entry to see if we can continue to increase more. An admissions focus on the individual rather than their scores is the most important requirement.

Black medical school applicants have a wide range of MCAT scores, as do accepted Black medical students. Regarding all medical school admissions of all types of students, the MCAT can identify 90% of those likely to fail, but 80% of those identified will become physicians and some will clearly be the best physicians from a number of perspectives. The MCAT is a less valid test of future medical school performance in those most different than standard students. The standard student is getting more and more narrow as the nation grows more diverse in a number of dimensions. Regardless of ethnicity, race, income or other divisions, the process of admissions is a bad process with great sensitivity but poor specificity. It is clear that we are selecting for ultimate attrition prevention rather than selecting physicians for the best qualities for health care now or in the 3 decade life span of future physicians. Most specifically, the current admissions process is selecting away from those most likely to choose family medicine, primary care, rural, and underserved locations. The current process, especially the preparation and testing process, is biased toward those of higher income, more urban, and professional parent origins.

Workforce is more than selections however. Those who choose certain careers and locations are influenced by other factors. Being rural and comfortable with rural people, patients, and locations is important, so is being Black or being comfortable with Black people, patients, and locations. Foreign born Blacks do face many of the same issues as US born. There are great divisions among Blacks as Stephen Wilson (Family L) noted and as Colin Powell notes in his autobiography right up front. However other factors apply, such as exclusion. This may shape career and location choice. FP choice also shapes location choice.
 

What is apparent with these and other divisions of family physicians by geographic, income, or other origins:

The major difference is choice of family medicine or not - not the subgroups of various family physicians.

Family physicians all make contributions to important pools of physician workforce

1. Total Physicians - one of the top 2 or 3 if not the most numerous physician type, especially after a few years when internal medicine physicians subspecialize

2. Primary Care Workforce - a consistent and persisting top contributor to primary care, in numbers, in volume, there are no major leaks in the primary care pool of family physicians, we stay in primary care

3. Distributional physician workforce outside of major medical centers - FPs are 50% inside and 50% outside, there is some local variation between 40 - 60%, but unless researchers go to great lengths to cherry pick FPs, there are near half and half. This leaves 45,000 FPs to distribute to areas with less people and fewer other physicians. Because there are no leaks and because there is an equilibrium between inside and outside, and because FPs do not collapse "inside" to concentrate back in major medical centers over time     -   we stay

4. Women's Health - family medicine stays in women's health, FPs also contribute more in more distant locations

5. Mental Health - family medicine contributes, and increases in more distant locations

FPs do have minor leaks such as about 1% for sports medicine and geriatrics and 1% for other careers, but even these contribute to primary care pools. Bigger leaks at 2 - 4% each are administrative, hospital based, research, and teaching, but each of these areas have specific roles to play.

Contrast medicine pediatrics with leakage of 80% of those who match, or similar levels of current internal medicine residency graduates with only 20% intending primary care careers at the completion of residency (garibaldi) or the 100,000 physician assistants who decline 1 percentage point or 1000 fewer each year in rural location, 1000 fewer working with FPGP, and 1500 - 2000 or more fewer each year working in primary care. NP data is missing nationwide, but is likely to be declining at the same or a more rapid rate as in physician assistants.Primary Care Retention

You cannot supply a primary care workforce pool by using a bucket that has a hole that grows bigger with each trip to the well. When the primary care supply also leaks out with each passing month, the job is made more difficult.

Supplying the primary care pool with a bucket that does not leak, with a specialty that does not seep out, and with health care policy that resists all leakage and recruits workforce also, is the best way to supply a top quality, stable, dependable, foundation for health care.

Physician Workforce Studies

[email protected] for questions

www.ruralmedicaleducation.org