Instate Retention of Family Physicians: Dependable Primary Care Workforce Retained within States

Robert C. Bowman, M.D.

Family medicine is a critical part of the lubrication and insulation of a nation that includes efficient health and efficient education. Without the personal physicians that connect people to systems, particularly those with the least, there is no warning light. The engine just fails not long after the lubrication fails. You never know losses in serving professionals until it gets cold outside and the cold begins to penetrate the inner workings and the whole machine grinds to a halt.

Family medicine is clearly the sustained and retained primary care workforce, losing few over time and distributing most where they are most needed. The characteristics that help distribute family physicians are connections outside of major medical centers. These connections involve instate origins, rural origins, lower and middle income levels, and older age at admissions. Each of these categories (especially instate birth) and also family medicine all add to increased percentages retained within a state for medical practice. These studies involve long term sustained retention of 1987 - 1999 graduates as of 2005 locations, not first time locations.

It is clear that many states have lost track of the importance of basic necessities such as education, primary care, and public health. Although most states have been forced to gain a grip on economics, there seems to be a poor grasp of how investments in education and health care influence economics. Basically states that invest in education and health care efforts are more efficient and effective in economics and a number of areas. Health care costs, unemployment, insurance premiums, legal costs, prison costs, and social costs are only a few measures of the inefficiencies resulting when states, cities, or nations fail to invest in education, primary care, and public health.

The current focus of many states is one of desperation, with officials searching for any way to decrease costs. Federal cuts in spending coupled with taxpayer demands for no new taxes have created great challenge. The only new taxes that have been implemented are those that also may end up increasing costs in the long run, such as investing in lotteries and co-pays that prevent those in most need of health care and preventive care from getting health care at all. There is a great danger that short term budgetary improvements may only pile up increased costs in the long run. There is a disconnect between spreadsheet savings and the realities of life faced by underserved populations. At the center of care of underserved populations is family medicine.

Early Lessons for a New Specialty

Family medicine funding at the state level has long been considered an important investment. In the early years of family medicine, there was little medical school support. Also the funding that was sent by legislatures to medical schools for family medicine tended to be diverted to other uses. Hospitals still do the same thing with Graduate Medical Education funding and filter it through to the training programs that brought them the funding. This resulted in direct funding lines for family medicine for training of residents and support of programs. These have been effective in graduating family physicians, although it has been a constant challenge for family medicine to get state and federal governments to recognize the importance of infrastructure and health policy.

In more recent years it has become apparent that education policy is also important. States that fail to invest in lower income children in rural and inner city areas are also failing to invest in those most likely to become family physicians. Birth Origins and FP Choice The losses may also include school teachers, nurses, public servants, and others that begin as the children of first, second, or third generation college families.

Many states have begun to examine the concept of human capital retention. When states make the education investments that result in more jobs, services, and economics in their state, the future looks bright. This human capital perspective is a major advantage family medicine since FPs have 36% higher instate retention in the state of their medical school location compared to other physicians. This increases to 44% higher for the family physicians graduating from the 81 public allopathic schools. The size and number of these schools still dominate physician workforce in the United States. The very few medical schools that do not have this level of retention are allopathic private schools and the following public schools located in some of the least favorable areas for family medicine: U Mass, U Miami, SUNY Stony Brook, SUNY Brooklyn. Massachusetts, the District of Columbia, and the nation’s most densely populated areas have the lowest concentration of family physicians, for a number of reasons.

To understand retention of family physicians there are a number of areas to consider. Perhaps one of the primary considerations is the period of history evaluated. Evaluation of choice by medical school class can assist in this analysis. The graduates of 1995 – 1997 in the United States had the best choice of family medicine and the best rural and poverty distribution in the nation’s history. Managed Care Comparison Table This was an era influenced by managed care, increasing primary care reimbursement, increasing investment in Medicaid (from 4% to 7% of physician income), and consistent decreasing reimbursement of subspecialists. Panic on the part of medical students regarding lower probability of hospital-based careers resulted in an increase in GME positions for primary care and a decrease in positions in hospital-based careers. Many of these chose family medicine. The 50% of medical students born in the most urban areas of the nation were 50% more likely to choose family medicine during this time period. These students born in areas of over 1 million are also the most likely to be practicing in 2005 within 60 miles of their birth location with 40% of the 1987 – 2000 graduates in such locations. During managed care the urban born and instate born were more likely to choose family medicine and this resulted in even greater retention within a state. In the past few years the choices have returned to pre-health reform era career and practice location choices, in other words back to market forces and less efficient use of state education and medical education investments.

The contrast between rural born and urban born is important, beyond greater choice of family medicine, primary care, and poverty practices. Rural born students have a 40% retention level as well, but this involves retention in a county with the same population density. This may or may not result in a location within state borders, but it does result in an even higher location in the same state plus adjacent states. Essentially adjacent states are trading students. This often involves those who were born in another state or who attended medical school in another state who are returning. Also those who married spouses from other states are likely to be moving to an adjacent state where the spouse has family, job, or other contacts. It is not uncommon in the Midwest or South (or any state) to meet and marry a spouse from an adjacent state that was first encountered in college, medical school, or residency training.

Given the current day challenges of recruiting family medicine residents, residency programs have very little control over the origins and interests of the residents that they select. In particular the programs that have set call and other workload demands must be able to recruit enough just to maintain the care of patients and to maintain the some number of residency positions.

• Family practice residencies in western states are more likely to have the ability to accept and reject candidates.
• While academic programs have some advantages in recruiting residents who desire to stay in the same location for the entire 7 years of training instead of 4, community-based programs must be willing to take all who have acceptable qualifications, especially in the East, Midwest, and South where there is less interest.

There is 62% retention of family medicine graduates within state locations for family medicine programs in the Midwest on average, but 76% stay in the same state or an adjacent state. Certain residency programs are more likely to have adjacent state retention.

Ranked By Greatest Margins Between State Retention Compared with Adding Adjacent States

 There can more than 1 residency program for some sponsors. The Masterfile is weak on residency program names and linking with specific programs over time.

If a state takes a narrow view of who locates in their state alone and cuts back on family medicine funding, then each state individually and collectively is likely to lose out on the benefits of family physicians. The situation with University of Nebraska programs can be instructive. When UNMC retooled residency programs to emphasize rural training tracks and accelerated rural programming, the UNMC medical students most interested in rural practice tended to stay with rural residency programs. The programs losing out on UNMC students were programs in adjacent states such as in Wichita KS as well as the program in Lincoln Nebraska. The impacts to name just a few:

• Retention rates decreased for Lincoln with the change in medical student origins. However there was an advantage. Lincoln does a much better job of attracting students from other states and once they are attracted to the state, they are more likely to choose state locations for practice.
• Academic programs were filled with Nebraska born students. This meant that they did not have to recruit as many from distant states or nations. Distant origin students prefer large urban practice settings and are much less likely to serve rural areas and to stay instate for practice.
• There were more positions filled in Nebraska FP residency programs statewide.
• There are also important workforce receptor considerations. With more positions filled and more graduates, there will be more choosing urban locations.
• This makes it more likely that other FP graduates in the state will choose rural locations, just as in the managed care era when the nation “saturated the urban receptors” and distributed physicians at the best level ever. In times of low choice of FP, the urban and urgent care positions get far too many who might otherwise choose rural and poverty locations.

States must do all that they can to graduate, train, recruit, and retain family physicians. There are major benefits of this effort.

These benefits for the classes graduating 1987 – 2000, about 40% of the total US physician workforce, include

• Primary care access – family medicine averages 44% of the primary care and 65% in rural areas of the nation (not adjusting for the additional numbers of encounters since FP docs see more and work longer). The range varies greatly from 19% of primary care in states like Massachusetts or the District of Columbia to 70% for the most rural states.
• Family medicine provides 60% of the primary care for those in the military zip codes in Europe and the Pacific.  
• Family medicine is by far the most efficient form of primary care with 87 – 92% of graduates in office-based primary care compared to 76% for pediatrics and 50% for internal medicine residency graduates. When students choose internal medicine or pediatric residency training, they are more likely to subspecialize and to choose hospital-based, research, teaching, or non-office-based practice. Those choosing family medicine are interested in delivering direct patient care, especially for rural areas and poverty populations.
• The medical schools that graduate more family physicians also graduate more office-based internal medicine physicians, indicating a close connection between supportive policies in a state, admissions of students more likely to choose primary care, medical school environments, and career choice.  
• Mental health access – family physicians are the mental health capacity in rural areas, according to studies (according to the Minnesota Office of Rural Health report and experts such as Jack Geller in North Dakota)
• Family medicine physicians have lower income origins (Bowman, Birth Origins) and are more likely to return to populations that are similar, including those who are poor, income, less educated, elderly, non-white (Mold, Graham Center studies)
• The similar origins also mean that FPs share important common ground with their patients. This may well have important quality and cost saving considerations. With 60% of US physicians from the top 20% in income level, it may help to have more and more physicians who are known to come from lower income origins.
• As a group, family physicians have a 36% increased retention within the state of the medical school compared to other physicians, 44% for public medical school FPs.
• Family physicians are the only physicians that distribute in the same pattern as the US population.
• Family physician economic impacts are estimated at $500,000 to $1 million per family physician. By the numbers: Rural Doctors and Rural Economies  The impacts tend to be higher in rural areas where there are more procedures, hospital impacts, and nursing home impacts. This is critical for preserving facilities and health care jobs, for keeping all jobs and population intact, and for recruiting new jobs and businesses. Declines in family physicians will accelerate declines in rural and underserved areas, the ones that devastate education, population, and quality of life.
• Studies involving McLennan County in Texas where the family medicine residency program was the primary source of primary care, public health, and women’s health demonstrated savings in health care costs for the county of 30 – 50%. Cooperative efforts between the medical association, the city, the county, the hospitals, and state have made this possible. McLennan County FP Residency Impacts
• Family medicine residency programs are an important source of care for uninsured, underinsured, and working poor patients. Community and Migrant Centers get the funding and attention. Family medicine programs do the same work with the same results and actually train the physicians needed for Community Health and all other underserved locations in primary care, mental health, geriatrics, and Medicare. The best concentrations of primary care services in poverty locations are provided by CHCs, by FP residency programs, and especially in zip codes where there are both CHCs and FP residency programs. (poverty) Poverty Locations and Physicians Family medicine residency programs donate millions in faculty, resident, and gratis services to Community and Migrant Health Centers each year, and receive much less in return. However, CHCs treat family medicine training just like neglectful deans and other medical education leaders in the past. They often appear to consider their own patient care needs much more important than graduating the physicians who will serve in their clinics now and in the future.
• Family medicine residency programs are more and more a last resort for uninsured patients, Medicaid patients, Medicare patients, and TriCare military-related patients. Other clinics who are supposedly “not for profit” have long ago deserted such populations. The efforts of some family medicine leaders have prevented medical schools, clinics, and other entities from leaving these populations behind. Further cuts in these areas will leave family medicine no choice but to stop caring for such populations just to be able to survive.  
• Family physicians are important leaders at the local, state, and medical school level. They provide leadership in state health areas, including outreach, medical school admissions, and planning.

Family medicine is an integral part of efficient education, government, and health care. Inefficient states with inefficient systems cannot graduate, recruit, or retain family physicians.

• They do not have the education investment that allows lower income students to be educated well enough.
• They do not invest in enough medical school positions.
• They do not prioritize family medicine, mental health, public health.
• They do not cover health for poor children or other low income populations.
• They do not support training of family physicians.

Basically inefficient states focus on urban areas and higher income peoples. States that are efficient and supportive of education and primary care and rural care are likely to graduate, recruit, and retain family physicians.

The tendency of states to experience a “domino effect” and replicate changes such as Medicaid co-pays, pharmacy restrictions, treatment of dual eligibles, and other areas that make primary care more difficult make this a matter of great concern for family medicine, particularly at a critical time.

Factors in Retention of Family Medicine

• Size of the state and geography
• Birth origins of the family physicians
• Perceptions of the State
• Population Distribution in a State
• Status of state education and health policy
• Health policy in neighboring states and the nation
• Current workforce needs for family physicians and state market impacts
• Training of family physicians

Size of the State and Geography

Small states and those with many bordering states and states who have training locations in close proximity to other states are more likely to lose family physicians to other states. The data on family medicine has always shown an impact on FP location within 60 miles of a residency training program. If 60 miles includes other states, there will be losses. Geographic barriers can also act to help retain physicians. In the largest states in the nation where

• training is distant from borders and
• there are few adjacent states because of significant coastal location
• geographically isolated by mountain ranges
• much higher population levels centrally located

the retention levels are the highest.

Birth Origins of the Family Physicians

Family physicians are more likely to be retained in their birth state or the state of their medical school, but this has to do with who chooses family medicine. Those choosing family medicine in greater percentage include those born in a state, those who are from rural areas in the state, and those who are from lower income origins.

Perceptions of the State

Family physicians do tend to locate in states that have high growth rates as with any physician or worker. There are also states such as Wisconsin and North Carolina that clearly invest much more in recruitment efforts and do have patterns of workforce that reflect success in these efforts. Midwest and southern states do tend to donate physicians to residencies and practice locations in other states. There is a consistent westward migration of family physicians throughout training and practice location.

Some of this is preference for the most urban locations in the nation that involve birth origins, ethnicity, and culture. For the 1997 – 2003 family physicians, there were 100 Asian family physicians born in California, 200 attended medical school in the state, 400 did a family medicine residency in California, and 700 Asians from all medical schools in the world, particularly the US and India, located practice in California by 2005.

Population Distribution in a State

States that have urban areas or significant population in large rural areas are attractive. Small but not isolated rural locations of 6000 – 12000 people have a higher percentage of family physicians compared to the population distribution in the nation. (RUCA) Large rural locations of 20,000 to 50,000 represent a clear advantage for family physicians with enough population for urban living and health resources, but not too much to lose rural lifestyle and respect as a family physician. States with stacked population in one corner with mostly small or isolated rural towns represent a challenge for recruitment and retention. States such as Kansas, North Dakota, South Dakota, and Nebraska face difficult issues of geography, frontier locations, and population distributions near adjacent states. Iowa has a nice distribution of large rural or urban locations accessible from just about any portion of the state represent a maximal environment for family medicine.

Status of State Education and Health Policy

Ranking states on health policy and education reveals a 0.7 correlation between the percentage of family physicians in a state and these rankings. Basically states who do not invest in the future and those that do not allow efficient operation of primary care practices do not educate, graduate, recruit, or retain family physicians.

• When states do not educate rural or inner city students well, those most likely to choose family medicine as well as those most likely to stay in a state to practice are prevented from accessing college and medical school and family medicine residency programs.
• Not having enough medical school positions creates much the same situation. States such as West Virginia have the highest rates of family physicians and the highest rural physician rate in the nation. States with few medical school positions select elite students that do not tend to choose family medicine or rural locations.
• Scholarships and obligations
• Recruitment and retention
• Practice structure support, hospital viability
• There are clearly states that do a much better job in salaries for family physicians, reimbursement for primary care, child health coverage especially for poor or rural children, liability, and support infrastructure for primary care. Family physicians are more likely to care for the poor and rural populations and poor support of these populations in health care can be problematic and lead to poor recruitment and retention.

Health Policy in Neighboring States and the Nation

Sometimes it is difficult to retain physicians when neighboring states do much better in health and education. This is much the same with school teachers, nurses, counselors, and other professionals. Nebraska, North Dakota, South Dakota, Iowa, and Kansas donate thousands of school teachers, counselors, nurses, and health professionals to other states. Minnesota absorbs these professionals. Illinois donates all these categories except physicians and professionals where there is in migration. States near Minnesota and major urban areas are going to lose more young professionals.

States do have choices. They can invest in students likely to stay in a state or those likely to migrate. States may invest in such students for years or decades, and then lose them to other states. When states focus on students with elite scores for merit scholarships or college funding or professional school, they actually increase the probability of losing their investments to states more distant. The mantra should be “High SAT, high ACT, high MCAT, will travel.” This includes rural born and inner city students as well. That is why lower income students, rural born students, and long term connections with a state are important in retention measurements. Studies within states can reveal the students that are the best investment for a state. In Nebraska, those most likely to gain admission to medical school are from suburbs of Omaha, Lincoln, and the major rural towns along the interstate. These physicians are the most likely to be located outside of Nebraska in 2005. These are the students with the highest probability of higher income, higher parent education, and higher scores.

There are important examples of retention at the medical school level that are the result of state investments and interventions. Few schools have the dedicated effort that Duluth has to graduate 50% into family medicine and 30% into rural careers. Duluth also graduates consistently more Native physicians than just about any other medical school and with only 60 in each class. The retention for Duluth for practice in MN is over 60%. The only schools higher include osteopathic schools (see training below), schools in California or Texas (geography) or Mercer (which has had similar admissions focus on FP and gets older, rural, instate). Duluth represents an $8 million dollar investment a year for the medical education of 120 total students, 60 first year and 60 second year, but the economic, health access, and other impacts alone are outstanding. Duluth represents an entire confluence of college, admissions, and training impacts, not just a 2 year focus.

Current Workforce Needs for Family Physicians and State Market Impacts, Differentials In Pay

States do vary in pay rates, usually based on procedures and demand for family physicians. Highly competitive and often overserved urban markets have actually acted to suppress pay for all in primary care. Medical leaders, state leaders, and national leaders have been slow to realize important workforce changes in recent years.

Emergent, locums, and urgent care demands have reduced the numbers available for true primary care. Changes in health policy and reimbursement make it tempting to give up the high overhead, liability, paperwork, and other hassles of a full service continuity primary care setting to do a temp job. This has impacts on retention as well since going this direction is a focus on salary and fewer hours and can involve out of state location or services.

This also means double the work and triple the costs. Temp positions usually involve care that was not needed or will need to be repeated at the primary care physician’s office. The care costs much more and involves too much testing. The patients receive little explanation and the primary focus of urgent care is to avoid responsibility. Urgent situations are unwilling to be responsible for patients, liability, future care, or follow-up. States that make primary care difficult are hurting only themselves, their businesses, and their people.

Training of Family Physicians

This is a huge category. The major impact involves medical school type.

• Osteopathic schools have some of the highest retention rates. Again they tend to have lower income origin students who are older and more mature, more likely to be rural born, and less likely to have the top MCAT scores that mean distant travel. Osteopathic schools tend to train not only for 4 years but for 5, 6, or 7 years in the same location through their state/regional OPTI system. They also graduate more family physicians. All of these categories involve more connections with a state and a much higher degree of retention.
• Much the same is true of allopathic US medical schools that focus on rural health or primary care. These schools also have more older, rural born, lower income, and lower scoring students. Such schools also take more resources because they actually must invest in teaching and make this a primary focus. Elite schools who choose non-distributional students rarely need to do as much with teaching, accreditation, travel, outreach, or primary care investment. Their focus is on research and practice revenues.

Another factor is intensity of training, particularly involving procedures. If the training programs in the state have little procedural training influenced by

• Liability problems
• Private hospitals that refuse to train residents intensively
• Underfunded primary care programs that cannot support procedural efforts and faculty
• Primary care leadership not focused on the needs of a state or procedures

Then the state is likely to have less attractiveness and retention potential. In the current national reimbursements schemes coupled with liability issues and hospital privilege challenges, it takes a dedicated effort to train for procedures but this can help improve the finances and satisfaction of family practitioners, especially in rural areas.

Training can also be innovative in ways favorable to retention. There are many transitions involved in a medical career: high school, move to college, admissions to medical school, matching into a family medicine residency, and choosing a practice location. Coordination of these transitions is a matter of cooperation between all of the parties. When states work to retain high school graduates, when college health professional advisors are linked to medical schools by admissions or special admissions programs (RHOP, tracks, BA/MD programs), when medical schools are linked to FP residencies (accelerated programs), when residency programs are linked to the state (state recruitment efforts, resident salary based on choice of FP with added salary for matching to a community in the state as in Oklahoma) or when linkages are formed across all transitions by loan repayment and scholarship efforts, then retention can be enhanced.

In studies of 150 accelerated family medicine graduates (3 years med school, 3 years FP residency at same med school) nationwide in 12 programs, basically the only accelerated family medicine residents not staying in the same state as their medical school were those at private allopathic schools that appeared to have federal obligations that drove them out of state. Over 50% chose rural, another 23% choose office based primary care in poverty locations, and 10% chose faculty locations, also serving the state and poverty populations. Over 80% choosing the most needed physician workforce careers and locations with 71% instate retention from medical school to practice location is about as pure workforce efficiency as you can get. Rural training tracks have even better rates of rural practice choice and similar instate retention.

South Dakota’s brief rural training track program (RTT) program had 5 graduates and 100% rural location and 100% instate practice location when I checked 2 years ago. The kind of programming that would retain family physicians has not always been supported by the state.

South Dakota only graduates 47% of the family physicians that serve in the state. This is the lowest number of instate family physicians in the Midwest. South Dakota has the most to lose if it takes measures that decrease overall family physician supply in the Midwest.

Medical students born in South Dakota chose family medicine at 16.5% and the physicians practicing in the state chose FP at 20.8%, a 26% improvement. Just about every other state around South Dakota has greater choice of family medicine. This is another measure of how much South Dakota could lose if it is not able to obtain family physicians from other states.

In summary

Family medicine is a critical part of the lubrication and insulation of a nation that includes efficient health and efficient education. Without the personal physicians that connect people to systems, particularly those with the least, there is no warning light. The engine just fails when the lubrication fails. You never know losses in serving professionals until it gets cold outside and the cold begins to penetrate the inner workings and the whole machine grinds to a halt.

Cuts in family medicine mean fewer services for those most in need of physicians, now and in the future. Cuts in family medicine funding mean cuts in faculty, particularly in the outreach positions that connect rural and academic communities. Less family medicine funding means that academic family medicine has to focus on revenue generation instead of the needs of the state. This means more patient care and more emphasis on grants and research. The means that family physicians are unable to serve on leadership positions at the medical school, particularly areas such as admissions that is so time consuming. This means even less ability to influence medical school admissions that would distribute physicians where needed, and other areas critical to medical education and health care in a state. The end result is poorer health access and even more problems with education and health care and underserved populations.

Without family physicians it is more and more likely that medical schools, medical school leadership, and governments will not be able to perform their functions in a way that maintains trust in authority and better opportunities for citizens and residents.

For the long term there is one area that has a higher priority. If government reinvests away from  family medicine and toward education such that more inner city, rural, and low income children receive better education, there is a good potential (after some years of problems) that long term the state will be more efficient and more likely to graduate more family physicians and have lower costs in many areas. Short of this investment in children and education in the earliest ages, there are few investments as important as family medicine.

Physician Workforce Studies

Poverty Locations and Physicians

Bright Future vs Dependent States

The efficient states clearly have education and health policy that result in better graduation, recruitment, and retention of family physicians. These policies also result in more teachers and other young professionals. Such states have lower costs for health care, for insurance of all types, for prison, for welfare, for legal costs, and for unemployment. When ranking states by these factors, there are important differences.

• The top ranking states with Bright Futures include the Dakotas, Nebraska, Kansas, Minnesota, Wisconsin, Iowa, Vermont, Montana, Utah, Idaho, Hawaii, Oregon, West Virginia, and Arkansas. They have prioritized investments in children and education and health care.
• At the bottom are the dependent states of Florida, New York, Louisiana, California, Texas, Alabama, New Jersey, Illinois, Georgia, Connecticut, and Arizona. These states have not made the correct investments even though many have great wealth. They have become dependent upon other states and nations for their future. One of the most inefficient places in the nation is Washington DC, with legendary failures in so many systems and more to come. When so many systems have failed, it becomes necessary to coordinate all recovery operations at the same time.

The states that increase in family physicians from birth origins out to practice are Bright Future states. The states that retain family physicians also have the highest ranking Bright Future scores. The states that graduate the most family physicians have the highest ranking scores. The states with the least standards for education are the ones that do educate their children and they also graduate more family physicians.

States have yet to inform their graduates about so many of the problems of some of these dependent states. States have also failed to help graduates discover so many of the great benefits of states that really do run efficiently, even if they do not have the highest salaries. Nor do states do a good job of recovering their graduates when they begin to learn the truth. Iowa’s Roger Tracy has done a great job of tracking its family physicians with Iowa connections no matter where they end up to bring them back to Iowa. Wisconsin and North Carolina have invested in offices that not only recruit from other states, they do an excellent job of retaining graduates in the state. Much more can be done.

Physician Distribution in the United States

Retention Within FP

Physician Workforce Studies

Family Medicine Central: National Comparisons of Workforce

www.ruralmedicaleducation.org