Rural Pipelines Versus Long Term Obligations

Generic expansions nationwide can be expected to result in no change from the 9% of physicians found in rural areas. To few remain instate, choose needed locations, and remain instate in needed locations.

Voluntary pipelines are still limited by lack of specific focus. Voluntary pipelines are based on the theory that improvements in admission and training and family medicine choice will multiply the result for higher percentages instate in the desired primary care or rural or underserved outcome. Multipliers do work, but represent small relative changes.

Rural pipelines applied nationally can deliver about 20% instate rural practice location. This is better than 3% instate most needed rural for generic expansion, but is still not optimal. Lack of specific instate practice retention and lack of career choice for specialties suitable for rural location limit desired outcomes. Voluntary choice allows career choices outside of family medicine (FM), the only specialty choice that multiplies rural location significantly (triple). Even when choosing IM, surgery, ortho, and ob-gyn the US MD graduates have been subspecializing rather than remaining in the general careers.  Instate rural proportions of graduates are single digit for the average and for most state medical schools in the US.

Family medicine tracks linking medical school to residency training can deliver 30 – 40 percentage points of instate rural at least for the first half of a career (the limits of data on such tracks). After 7 years instate with additional time likely prior to admission, the impact on practice location instate is strong. Drift out of state is likely to reduce this proportion over the second half. The rural contribution indication overall in areas such as accelerated tracks has been 40% rural. The family medicine model also addresses small and isolated rural locations. For example the FP to nonFP ratio for U of KS grads indicates a 16 times multiplier for small and isolated rural). Those choosing KS and family practice have characteristics that facilitate optimal health access placements.

Family medicine tracks do not address the other half of rural workforce needs such as general surgeons, ob-gyn physicians, orthopedists, anesthetists, urologists, and general types of cardiology, GI, etc.

Long Term Obligations

The contributions of voluntary models pale compared to long term obligation models. Long term obligation models as illustrated by Jichi in Japan (Matsumoto) can deliver 70% instate and at least 50% instate rural workforce over a career. Over half of this contribution is up front by the 8 year design of the obligation. Such a design also addresses the specialties needed beyond just family medicine.

The challenges are not small for a long term obligation. Iron-clad contracts, informed consent, and interstate or federal compacts are required. States such as Montana are making specific contracts to assure return to Montana after training - by design. Also protections are needed for those under obligation - of course much of primary care workforce needs protection from marginal treatment under the current primary care design.

Medical schools such as Jichi have a cohort effect with all graduates heading to the same types of practices. Japan has implemented commitment tracks in public schools in all 47 prefectures. There will be similar obligations but the cohort effect is likely to be less. Of course Jichi graduates have been leading many of these efforts.

Commitment tracks themselves are a useful admissions tool. Rather than candidates for medical school claiming all types of service, rural, small town, or family practice interests - they would be known as willing to sign the obligation contract or not. This might help the admission committee choose other candidates more likely to remain instate and in careers most needed by the state.

Summary Estimates

The estimates using rural workforce years after graduation indicate 20% of a career instate and rural or about 7 rural instate years per graduate for 28 per year in a pipeline design with voluntary choice

This compares to compared to 50% instate rural or 17 years spend instate rural in a career using the long term obligation.

With 8 long term committed rural graduates for 8 years and decreasing down to 3 after age 54, a state can count on at least 38 rural physicians at any given time 30, 50, or 70 years later - by design. If the output seemed to be falling short, the obligation length could be increased - again resulting in no increase in cost with a specific increase in rural outcomes. Ten year obligations would likely increase yield to about 44 rural physicians.

Voluntary methods are regressive under US policy designs. Failures of voluntary designs require more generic expansions resulting in substantially more costs for no better primary care result (or less result). Pipelines require increasing maintenance costs across each of 5 or so segments not counting support of rural training sites if included. Funding must be injected to improve the yield across each segment - yields that have been declining already. Long term obligations are selected by those that have every incentive to shape their training for their first 8 years of practice - rural instate practice.

For most of my career I have promoted pipelines. Objective analysis indicates that pipelines only worked prior to 1980 when just about anything worked - by designs that increased spending in primary care, rural, and underserved areas.  Recovery of rural workforce, underserved workforce, and primary care workforce will require specific efforts. Assumptions have not worked for 30 years with more to come.

Generic Expansions Are Not Smart

Medical and medical education associations want generic expansions of graduate medical education positions. Billions taken from health care diverted to medical education will not address primary care deficits SMARTly. 

Generic expansion fails for the purpose of primary care. Generic is not SMART - specific, measurable, achievable, realistic, or timely. Obviously the added GME positions outside of primary care will not result in primary care. IM and PD graduate increases have been countered by decreases in the percentages remaining in primary care for no gain. Both are mired at about 1400 - 1600 entering primary care and this is likely to remain so. Both have primary care graduates most likely to crowd their practice location in top concentrations of primary care already.

Specific family medicine expansions could help, but gains of positions could result in conversions of family medicine training to non-primary care positions. What is most likely is that family medicine will remain at 3000 annual graduates as it has for the last 30 years. After 30 years family medicine has reached a peak with 100,000 as a primary care workforce. Internal medicine and pediatric primary care with both be about 50,000 or together about 100,000. HRSA 2008 projections of substantially more are in error as the projections assume retention of primary care graduates within primary care, something lacking in US primary care for decades with more to come.

Only expansions specific to primary care with permanent primary care result for a career are SMART.

Nurse practitioner and physician assistant annual graduates have doubled in number each 6 - 12 years since 1980 but the primary care results have been limited due to departures form primary care and family practice during training, at graduation, and after graduation. The last doubling of physician assistants from 3000 to 6000 resulted in a 30% increase in those entering primary care in 2008 but a 200% increase in non-primary care. Conversions of primary care to non-primary care will result in no gain in primary care and over a tripling of non-primary care.

Generic expansions of osteopathic schools have not increased family medicine numbers. Before 1970 the family practice component was 70%, in the 1990 decade it averaged 35%, and recently the AOA study noted 17%. Each increase in annual graduates has been countered by a decrease in FM proportion resulting in the same 500 - 600 family medicine DO graduates.

Over time the international medical graduate numbers have increased, but the preference has been internal medicine resulting in the least primary care yield. Caribbean annual graduate increases have resulted in forced primary care choice to return to the US via residency programs, but generic expansions of GME will allow more to bypass primary care - negating the expansion. Ross is the number one US source of primary care with by far the most annual graduates and with 26% family medicine and with over 50% primary care training program entry, but more GME positions will allow a bypass of primary care choice.

US MD expansions also fail to increase primary care. The 30% increase in annual graduates seems impressive. Unfortunately the proportion remaining in primary care continues to decline with 7% for family medicine, 4 - 5% for internal medicine, and 6% for pediatrics with 20% or less as a total. Departures from primary care and conversions of family medicine to other positions negate generic GME expansion. Also with generic expansion, graduates choosing FM or IM as a backup plan are more likely to get their higher priority subspecialty choices.

For a SMART primary care result, generic should be avoided and the focus should be specific, measurable, achievable, realistic, and timely. Even with a change to a totally SMART primary care design, it will take 30 class years of sufficient permanent primary care graduates (about 14,000 a year). About 2050 or 2060 the US could have sufficient primary care just as it chose in 1980 to have too little for 2010 to 2030 - by design.

SMART Primary Care : Family Practice Contributions

Specific in primary care is a source that remains in primary care closest to 100% for a career contribution.

Not Specific and therefore not SMART is a source that is flexible with graduates that serve less than one-third of their careers in primary care delivery.

Without specific retention within primary care, it is difficult to measure primary care delivery. Flexible primary care training sources do not remain in primary care and result 75% in non-primary care workforce. Internal medicine, nurse practitioner, and physician assistant graduates steadily depart primary care during training, at graduation, and each year after graduation. Flexible primary care sources are difficult to assess with regard to career primary care contributions, especially in the past 15 years of changes. During periods of declining primary care retention, all methods overestimate primary care delivery including the Standard Primary Care Year method.

Most permanent sources such as family medicine result in the most specific translation of annual graduates into reliable and measurable and achievable primary care delivery.

Primary care recovery requires SMART in at least 4 dimensions - retention in primary care for a career at closest to 100%, activity in practice at closest to 100%, years in a career closest to 40 years (age 25 to 65),and volume at or greater than the top volume sources such as pediatrics or family medicine.

Measurable involves estimates based on a tool such as the Standard Primary Care Year. The SPCYr is the product of career estimates of years in a career, % in primary care, % active in US workforce (or any specific location type), and volume set relative to 100% for FM and PD. Sources of primary care least specific  in years, PC retention, activity, and volume are least specific, difficult to measure, and deliver the least priamry care during a career.

SMARTest for primary care is family medicine with 85% retention, 34 years, 85% active, and 100% volume. Caribbean, US MD, US DO, and non-citizen international graduate FM all have 20 - 26 Standard Primary Care Years per graduate. This is a measure that is relative or comparative to other sources.

Nurse practitioners with 25 years, 70% active as direct clinicians, 35% of direct clinicians in primary care, and lowest volume at 70% (wide range) represent least primary care delivery per generic nurse practitioner graduate for only about 4 Standard Primary Care Years per graduate. By specific type of training, the family nurse practitioner can be expected to deliver about 6 to 7 SPCYrs over a career due to higher retention. All other nurse practitioner graduates combined deliver less than 2 Standard Primary Care Years per graduate or less than graduates of Yale or Harvard or other prestigious medical schools. Family medicine residency graduates in their careers will deliver 4 to 8 times more primary care than nurse practitioner graduates. Medical schools missing in family medicine career choice also miss out on primary care delivery for the United States.

Physician assistants have declined to 28% entering primary care, 75% active, 33 years, and 75% of the volume for about 5 Standard Primary Care Years per graduate. In the 1990s graduates with over 50% primary care and lower costs of training, the PA was a bargain in training cost and primary care delivery.

The last doubling of annual PA graduates resulted in a 200% increase in non-primary care entry and a 30% increase in primary care entry numbers - a level that will be negated by departures after graduation. PA annual graduates increased 100% but will have about zero increase in primary care delivery. PA leaders are not at fault. Physician assistants mirror physician designs and policies that shape only 25% of physicians entering primary care shape PAs the same way. Flexible designs cannot fight policy constructs. Only permanent designs can result specifically in primary care during periods of poorly supportive of primary care as in all recent decades other than 1970 - 1980.

SMART is possible for non-physician clinicians but only with required retention within family practice. Distribution is optimal as well as noted in family practice DO, MD, NP, and PA sources that distribute at maximum levels to those in need of care. If NP and PA matched the 95% retained in family practice of family physicians, NP and PA graduates would contribute 3 to 4 times more primary care per graduate and the combination of FM, NP, and PA would be optimal for primary care recovery. Regardless of the source, permanent family practice is required for recovery to have sufficient primary care and to distribute primary care where needed.

Only family practice solutions work for nearly all Americans

for basic health access needs for nearly all of the years of their lives

delivered in nearly all locations.

Internal medicine is not SMART with lack of specificity the major SMART failure. Only 20% of graduates indicate primary care entry with departures in the years after graduation. Highest cost of training defeats medicine and medicine pediatrics as sources. Internal medicine also concentrates in top concentrations of people, income, and health professionals. Non-citizen international graduates do tend to chose internal medicine (about 35% of total IM). Delays in entry, limited distribution, lack of retention in primary care, and departures from the United States.

Pediatric residency expansions are contraindicated for the purpose of increasing US primary care delivery. Saturation of pediatric primary care workforce has been known for a decade. Expansions of annual graduates have bounced away from primary care and toward academic and part time positions. Pediatric graduates crowd into academic and highest workforce concentrations (similar to origins and training) where 70% are already found. Fewer have been locating in rural and underserved locations (Randolph, Committee on Pediatric Workforce).

In summary, primary care that stays in family practice is SMART. Family practice provides the bulk of primary care and even greater shares of primary care where it is most needed.

Non-family practice sources are more likely to depart primary care and are more likely to crowd into top workforce concentration locations with the most specialized and the most concentrated. Even the most general still are found at 75% - 80% inside of existing workforce concentrations.

Family practice is found about 50 - 55% outside of concentrations where 65% of the population is found in 30,000 zip codes. All other sources are found 70 - 92% inside of concentrations - 3400 zip codes clustered together in 4% of the land area with 35% of the population.

Sources that are more specific than current family medicine are required for efficient and effective primary care recovery. Generic is not measurable and is actually not achievable with regard to primary care recovery. Permanent primary care that is predominantly family practice is required - by design.

 

Thanks to all 12,000 who have visited Basic Health Access in 2011.

Robert C. Bowman, M.D.        Basic Health Access Web    Basic Health Access Blog

SMART Basic Health Access     World of Rural Medical Education

Basic Health Access Blogspot 2011 - Summaries and Links for 2011

Dr. Bowman is the North American Co-Editor of Rural and Remote Health and a Professor in Family Medicine at A T Still University School of Osteopathic Medicine. He was the founding chair of the Rural Medical Educators Group of the National Rural Health Association, he was the long term chair of the STFM Group on Rural Health, he is the founding director of Priority Infrastructure at http://www.infrastructureamerica.org/ and he is the author of the World of Rural Medical Education, and Physician Workforce Studies

Meeting Primary Care Needs in the Last Half of the 21st Century - Really!. 

Clinician Specific Medical Education

SMART – Specific, Measurable, Achievable, Realistic, Timely

To Be SMART or Not to Have Health Access

Many new proposals attempt to be seen as solutions for basic health access. Even more fail to be solutions. Few will remain objective in the next 20 years of desperation in primary care workforce. Those eager to meet their own agendas will continue to wave the primary care banner even with less than one-third of graduates delivering primary care workforce. Readers should understand  that what is being promoted as a primary care solution is typically generic, not specific.

This blog will focus on solutions that have worked and that will work. This blog will help eliminate solutions that will not work or cannot work.

What works in primary care is SMART focus upon primary care - Specific, Measurable, Achievable, Realistic, and Timely. What does not work is less spending upon primary care, less spending in locations in need of primary care, flexible sources of primary care that fail to remain in primary care, and claims of primary care that are not measurable or achievable or common sense.

Designs that work:

• spend more upon primary care, 
• spend more in locations in need of primary care, 
• spend most specifically on primary care service delivery, and 
• spend training funds on types of workforce that are most likely to serve in health access for 100% of their career delivering the most primary care per graduate. 

These are designs that work for an entire nation when applied to an entire nation. When these designs are not compromised by spending elsewhere not primary care, nations can count on primary care staff, nurses, and professionals that remain in primary care, that continue to gain primary care experience, and that deliver the best primary care for the least spending.

SMART works when the focus is specific for primary care. SMART principles also work for rural health with specific rural focus.