Office Based Proportions in IM and FM and Pediatrics

Robert C. Bowman, M.D.


With the collapse of generalist primary care choices in internal medicine, the nation has lost a top source of primary care. The changes in internal medicine, choice of family medicine, and midlevel providers all suggest health policy influences. The nation's health policy continues to remain unfavorable to primary care and any location outside of major medical centers. Five Periods of Health Policy and Physician Career Choice define two periods of favorable health policy impacting the class years from 1965 - 1978 and from 1992 - 1997. The nation has twice been successful in primary care and distribution with increased investment in the primary care patients in most need of care. See also Changes in Specialty Choice 1987 - 1999


Health policy changes are unlikely in the next few years. Both prior health policy periods favorable to distribution have required years of deliberation and debate. It is important to consider other means of improving primary care delivery. There are other factors that impact the proportion remaining in office based generalist practice. The most obvious loss involves subspecialization but there are factors that could be influenced by admissions, GME policy, training, and other areas.


The following data involves the AMA Masterfile for the 1987 – 1999 US allopathic graduates, the most recent 40% of the workforce who have distributed as of 2005 careers and locations. This is data that involves equilibrium conditions and physicians who are more likely to have transitioned to their final choices involving specialty choice and primary practice activity. This is much different than senior medical student match data where claims of primary care tend to be exaggerated. When considering only those listed as generalists in 2005 about 89 - 90% of all 3 specialties are office based. However when comparing the office based generalists to the total residency graduates in a specialty, there is great variation. See also Primary Care Retention



Residency Grads

Office Based

Not Office Based


1987-1999 Grads


All Office

Office Generalist

Hospital Based

Listed As Resident

Not Class-ified

Major Medical Center

Rural Office

All Rural

Internal Medicine



20979  45.2%





2151 10.3%



Family Medicine



22952  90.2%





5810 25.3%






13582  70.1%





1292 9.5%



Medicine Pediatrics



1582  87.1%





271 17.1%



All Other











The above data does not include 5866 graduates of the military school, military residencies, those listed in military locations, or those designating military activities. About 0.7% of this group is inactive.


Changes in Specialty Choice 1987 - 1999


Family medicine residency graduates average 90.2% listed in office based primary care. Pediatric residency graduates are 70.1% office based. Internal medicine residency graduates were office based in 2005 at 45.2% levels.

The actual losses outside of office based care may be conservative. Internal medicine, pediatrics, and medicine pediatrics specialties lose residency graduates to other specialties outside of medicine or pediatrics that are only picked up by more detailed studies involving residencies attended over many decades of tracking. Primary Care Retention Family medicine residency graduates from 1997 – 2003 remained in family medicine at 98% with very few moving to geriatrics, primary care, women’s health, and psychiatry careers that would not be considered “losses” in terms of the most needed workforce.


The losses due to subspecialty choice involve the factors related to choice of family medicine. Distributional or humble origin types choose family medicine and distributional careers at the highest levels. Elite students make subspecialty and major medical center choices. (Birth Origins and Distribution Tables) Medical schools with higher MCAT scores, exclusive origins, eastern locations, and those created before 1966 graduate the most subspecialists and the fewest family physicians and office based primary care physicians. (Choice of Family Medicine Regression)


Health policy impacts on office based choices and family medicine choice are greatest followed by education and admission and training factors. Finally there are variations within the primary care specialties themselves.


The greatest variation involves office internal medicine ranging from

·         34% remaining in office based care for internal medicine residency graduates choosing to locate in medical schools and in military bases to

·         38% for graduates of medical schools with MCAT higher than 10.6 (top 16%) to

·         40% for those up to age 25 or the youngest at graduation from medical school and those born in the most densely populated counties in the US to

·         53% for those over age 29 at graduation or those born in the most isolated rural counties to

·         55% for graduates of the Historically Black medical schools and the West Coast Distributional Schools (both with more older and diverse graduates) to

·         58% for physicians in rural locations to

·         64% for graduates of Duluth and Mercer with top distribution levels and the most humble origin students.


Internal medicine residency graduates located outside of major medical centers for in urban served, urban underserved, or rural locations had 50 – 58% office based choice .


Pediatrics choice varied from 62 - 68% for the core metro, foreign born, urban, highest income, out of state, and top MCAT/research medical school types to 72 – 76% for the distributional schools and distributional student types: older, lower income, rural born, and instate born. Pediatricians located in areas of high poverty, low income, and underserved locations had 84% office based designation. Pediatricians in rural areas had 87% office based choice compared to 86% in urban served locations 78% in major medical center and 49% for those in medical school locations. Those retained in the same state as their medical school for practice had 76% office choice compared to 66% for those who moved to a different state.


Family medicine office choice rarely varied much outside of 88 – 90% with age, birth county income, instate birth, population density, or medical school type. The range of graduates in office family medicine was 50% in military family physicians (mostly hospital based) to 74% for family physicians located in Washington DC to 84% for family physicians in medical schools to 92% for the West Coast distributional medical schools (UCLA, UC Davis, UC Irvine, U of Washington) to 95.4% for the graduates of Duluth and Mercer. Family physicians in major medical centers, served urban areas, underserved urban areas, and rural locations have 89 – 90% office designations.


Physicians from distributional student types and from Distributional Medical Schools remained in primary care at higher levels. Physicians from elite schools and likely those with the elite scores and most urban origins and highest income levels were the most likely to subspecialize.


Older age and diverse students have more office based choice in internal medicine. Over 54% of internal medicine graduates were once office based as noted the literature and in the Masterfile data, but the levels have continued to decline. New studies indicate a decline to 19% or less. (Garibaldi, Collapsing Choice of General IM)


Health policy is clearly the major culprit in studies that consider changes in choice by class year. (Distributional Choices and Health Policy)


Funding for clinical services, graduate medical education, research funding, public health funding, and supplemental hospital funding all favor major medical center and medical school location. Primary care outside of major medical centers gets declining Medicaid and Medicare and insurance, the lowest levels of reimbursement, crippling regulation, inability to negotiate based on low numbers of patients and providers, reduced patient care revenues (co-pays, urgent care, ER use, barriers of transportation and day care), and increasing costs in liability, salaries, utilities, rent, and numerous areas.


Family medicine maintains a consistently high choice even inside medical schools and major medical centers. The office based proportion in family medicine and pediatrics appear to be more resistant to change, even with health policy change.  Medical school locations have the lowest office based choices for internal medicine and pediatrics. The most distributional schools, students, and locations have the greatest office based choice in the primary care specialties.


Medicine pediatrics resembles the office based patterns of family medicine but does not share all of the scope and distribution. Medicine pediatrics also has low numbers so far and remains a transitional career similar to pediatrics and internal medicine with losses to subspecialty choice and academic careers and major medical center locations over time. (Medicine Pediatrics References)


The focus on gaining more for office based primary care involves the same factors as the choices that focus on physician distribution and the focus on choice of family medicine. These include improvements in education in lower and middle income populations, admissions of more humble origin students, preferential choice of family medicine within medical schools or graduate medical education positions, and health policy supportive of lower and middle income populations and locations.

·         The most outstanding gains in office based primary care involve a preferential choice of family medicine. The differential between 89% and 65% or 45% are substantial.

·         Health policy shifts that influenced hospital based specialties to change to family medicine and primary care were an outstanding success in retaining office based proportions. Increased support for lower and middle income patients and limitations of graduate medical education positions outside of primary care were the key factors. When health policy continues to favor major medical center location, the graduates in medical schools and major centers have the lowest choices of office based care.

·         Replications of branch campuses and family medicine specific schools such as Duluth and Mercer have the highest probability of improving much needed areas. Replications of accelerated family medicine residency programs with 100% greater rural choice and 50% greater underserved choice and top office based choice would also greatly improve distribution and primary care. Accelerated programs were terminated, however. Medical schools have also had the data from the Jefferson Physician Shortage Area Program for over 20 years and have failed to embrace these changes even during the last expansion. The clear focus of the current expansion so far has been subspecialists. Expansion involving different students will distribute physicians where needed. Expansion involving elite schools and elite students will not improve distribution or office based proportions. The typical 70% of medical students who are the youngest, who are the most urban in origin, who have the highest MCAT scores, and have parents with the highest levels of income, education, and professional degree will be choosing the most elite medical schools. They will be choosing subspecialities, hospital based care, and major medical center practice at the highest levels.

·         An admissions focus on different medical students results in more office based physicians. Different origins in terms of socioeconomics, age, geography, culture, test scores, and colleges are important. Broader admissions and focus on rural careers and family medicine can result in high levels of distribution and office based care. Admissions can even move beyond student origins to the highest levels of distribution and office based care, but only a few schools have accomplished this. 

·         Medical school leadership has yet to sound a consistent call for the much needed improvements in education that would improve health care quality from both ends: better prepared physicians and a population that could better make health care decisions.

·         New specialty choices such as emergency medicine and medicine-pediatrics may move medical students away from family medicine. This progressively results in more in major medical centers and less away from major medical centers and primary care over time. (Family Medicine Contributes Much More) If the distributional types of students are lost to family medicine, the nation is losing a 50% probability as found in a rural or underserved location years later. There is some evidence that some rural born, lower income, and diverse students are impacted by these choices. Family physicians have provided many if not most of the emergency physicians for decades and medicine and pediatrics and women’s health and geriatrics and mental health and public health….


More elite students, expansions of elite medical schools, decreasing choice of family medicine, nonspecific increases of graduate medical education positions, and poorly supportive state and federal health policy are not going to improve physician distribution, increase the proportion of office based care, retain more physicians instate for practice, or graduate the family physicians that fill in the gaps in underserved, low income, shortage, and rural areas with fewer people, facilities, and physicians.


Physician Workforce Studies


Primary Care Retention


Changes in Specialty Choice 1987 - 1999


louis vuitton outlet nike kd vii louis vuitton outlet louis vuitton outlet jordan 6 sport blue lebron 12 louis vuitton outlet wolf grey 3s sport blue 3s Wolf Grey 3s sport blue 3s cheap jordans sport blue 6s louis vuitton outlet louis vuitton outlet louis vuitton outlet coach factory outlet new jordans kate spade outlet lebron 12