To Follow the Money, Follow the Workforce

Dramatic headlines now highlight profiteering in hospice services.This should not be a surprise to those who track workforce. Hospice has consumed steadily more nurses in recent HRSA nursing reports. When the nation sends more spending to a service, more workforce follows because workforce is required to be able to deliver the services.

Nurses have moved steadily toward higher pay and toward the most lines of revenue with the highest revenue in each line just like physicians, physician assistants, and nurse practitioners. All also move toward higher pay for less hours as well. About 85% of nurses are forced though hospital entry positions after nursing school - by designs that send the most spending to hospitals. Witness the current nursing problems as hospitals have already ratcheted back.

With 30% more spending compared to the costs of delivering care, health services expand rapidly - as in just about any other health care business. More profits draw more investors and result in more services - including those that might not be as needed. This is the case for health services other than primary care or public health.

Primary care physicians, nurse practitioners, physician assistants, and registered nurses get paid less - unless they convert away from primary care. Conversion is easy for these flexible sources of workforce. Conversion away from primary care is not simple for family physicians. Not surprisingly 70 - 85% of NP, PA, and IM primary care sources are found outside of primary care.

Only school nurses get paid less that primary care nurses for a variety of reasons including schools having to cut all types of personnel including teachers, teaching assistants, and nurses. Education and health systems are personnel intensive and therefore health care cost intensive and with increasing health care costs, the cuts must involve those most important to the delivery of education in the United States. School district budget officers have been telling this to us for years. Why don't we listen?

Why do we allow profiteering that results in such rapid health care cost increases? This is guaranteed to defeat our businesses, our economy, our schools, our health care delivery, and government budgets at all levels.Why did we forget the lessons of the 1980s when health care drove us to our knees and reining in health care costs for a few years allowed the economy to kick off the longest recent run of economic progress in the nation's history? We thought it crazy to pay thousands more for a car due to health care costs. Now we pay thousands more for education, for health care, for government, and more.

One way to cut back on health care costs is just to say no - that is to more workforce in certain areas. The US has doubled non-primary care workforce each 15 years by design changes resulting in ever more non-primary care workforce (See Three Dimensions of Non-Primary Care Increase) A design that rewards higher pay for non-primary care results in more growth and also more workforce moving to non-primary care, even those trained in primary care that are now 70% moving to non-primary care careers.

Generic expansions must stop and the US must have specific training in primary care that is forced to remain in primary care. This is partly for restoration of primary care and mostly about stopping the incredible cost increases for non-primary care.

Cutting back on non-primary care is a very good idea as no one has found a good way to rein in non-primary care spending. Primary care is already cut back. The cut back is due to too little primary care spending that has limited the primary care personnel and the services that can be provided. Primary care is personnel intensive and has relatively few codes that have not had major improvements in reimbursement.

Non-primary care profits can be preserved by more services, by fewer personnel, by more technology, by billing code creep, and by new services not yet facing cost controls. Also people desperate for very specific non-primary care services have thousands of advocacy groups clamoring for more when we have ever less available for health spending - if we hope to have a nation at all.

Even the highest income only give 14% for taxes and we passed 14% years ago. We are at 17.5% of our Gross Domestic Product spend on health care and will not stop until we reach at least 20% according to government sources. By that time we will have doubled non-primary care once again and incredible pressures will push health spending higher.

Primary care and basic health access and public health lose out in the competition - as in the past 50 years except for 1965 - 1980. But even then when we doubled primary care, we increased non-primary care slightly higher. And since 1980 we have had stagnant primary care because we have diverted more and more resources to non-primary care - which has doubled each 15 years as we have gone along with those haphazardly shaping the design over the past 100 years.

Three Dimensions of Non-Primary Care vs Zero Growth in Primary Care

Promotors of generic expansion have commonly mentioned primary care workforce increase. At best primary care workforce remains a side effect of expansion. Training and spending and accreditation in primary care remain subservient to non-primary care interests and influences. Even primary care associations are compromised by members and leaders influenced primarily by non-primary care influences. The US should not expect an increase in primary care because this is what has been designed. Non-primary care is quite another result.

Primary care projections are actually quite easy to make. The best estimates are guided by recent decades. Recent decades of stagnation in primary care indicate no growth. Primary care spending is stagnant, permanent primary car such as family medicine is stagnant, and flexible primary care sources have been departing primary care. The US will remain at zero growth even with expansions as fewer remain in primary care during training, at graduation, and after graduation. Predictions of primary care as no growth should be easy due to stagnation in primary care spending and increases in the cost of delivering primary care - major determinants that insure declines in primary care delivery capacity. This is a design that sends primary care steadily away from primary care.

Addressing Disbelief and Assumption

For those still not convinced, the opposite approach may help. Why not calculate non-primary care workforce increase? Predictions of non-primary care workforce are most difficult with three dimensions of increase in 4 sources of health professional workforce.

• Generic expansions of nurse practitioner, physician assistant, osteopathic (DO), and allopathic (MD) annual graduates
• Increasing proportions of NP, PA, DO, and MD entering non-primary care (from 60% past 70%)
• Primary care graduates steadily converting to non-primary care over their careers

People and politicians and even academics like to think in short term solutions and cures. Short term solutions are not possible in workforce. Workforce designs result in an annual graduate number and it takes  30 - 35 years or the average career length to change the overall workforce design (25 years for NP with shortest careers).

For example it took over 30 years of graduates from 1980 to 2010 in family medicine to translate 3000 annual graduates into 100,000 for the current workforce. Steady increases from 30,000 to 100,000 were seen over this time. The Health Resources and Services Administration projected family practice to increase to 144,000. This is not possible as the family medicine annual graduates remain fixed at 3000. Even with a doubling of family medicine to 6000 in 2011, family medicine would fall short. Internal medicine primary care projections are similarly in error. Projection methods are exposed as problematic for primary care, particularly when primary care does not remain in primary care. Poor understanding of the basics goes all the way to the top experts.

Other sources not bound by primary care limitations have increased in annual graduates although the result is far more non-primary care than primary care. Sources such as NP and PA have doubled in annual graduates each 6 - 12 years since 1980 and have moved from 50% to over 65% not in primary care. Steady departures from NP and PA primary care continue with teaching hospital, emergency care, and subspecialty diversions leading the way. The last PA doubling (100% increase from 3000 to 6000 annual graduates) resulted in over a 200% increase in numbers entering non-primary care and just a 30% increase in primary care entry (AAPA). This small increase of about 3% a year will be negated due to departures in the years after graduation as Larson and Hart noted even during the 1990s - a much better primary care time period.

Osteopathic increases in annual graduates could have resulted in more primary care under a different national policy design, but the expansions have been countered with declining family practice choice from 65% (prior to 1970) to 35% in the 1990s to 17% for no gain in annual family practice - the predominant DO primary care vehicle. Family medicine entry remains fixed at about 500 - 600 per year. Only the fact that family medicine is the most permanent primary care source results in any stability of osteopathic primary care output - but there is no gain either.

And US population growth continues and the elderly that use 2 to 3 times more primary care are doubling by 2030 and we might have increased primary care demand from expansions of health care coverage (might).

MD declines in primary care involve all sources. There is a decrease in family medicine from 14% to 7% of US MD graduates, half as many internal medicine graduates (20 - 25%) enter primary care as compared to 1990s levels (over 50%), and pediatric graduates have decreased from 70% to 40% remaining in primary care. A 30% US MD expansion will not cover the losses for an overall decrease in primary care delivery per graduate from by far the major source.

What If We Stopped at Current Annual Graduate Levels and at Current Proportions Entering Primary Care?

Even if the US stopped all expansions and held at the current levels without further departures from primary care, this would be 6,500 annual physician assistant graduates times 33 years per graduate times 75% non-primary care for 161,000 in non-primary care and 56,000 in primary care (predominantly employed in family practice). About as many physician assistants will be inactive as will be found in primary care.

The nurse practitioner maximum would be 200,000 given about 8000 annual graduates. These have most recently been 70% found in the direct clinician component with 35% found in primary care (HRSA Nursing 2008). This results over 25 class years in 90,000 for a non-primary care workforce and 60,000 in nursing (especially nurse staff) positions, and 50,000 in primary care (predominantly family nurse practitioners remaining in family practice employment). Primary care comes in third in priority for nurse practitioner workforce.

The physician contribution will remain 100,000 for primary care from FM, about 42,000 from primary care IM and 48,000 for pediatrics. The non-primary care result is 192,000 for IM graduates and 58,000 for PD graduates. Family medicine and medicine pediatrics both contribute about 10,000 each to result in 270,000 for non-primary care with a total of 196,000 for physician primary care.

The total would result in 280,000 in primary care and 525,000 for non-primary care and 60,000 for nursing - and this is just the result from primary care graduates.

The total non-primary care is already set for 1.4 million as a workforce and will be higher with annual graduate expansions and stable or declining primary care proportions.

What resulted in increased primary care delivery in the 1990s with substantial NP and PA primary care efforts working together with MD and DO is quite different. The boost in workforce effort from working together will be found predominantly in non-primary care.

The 1990s design was 29% for primary care for the 2020s result. The 2010 design is set for 16% primary care result or less. The non-primary care workforce result is important to examine.

In 10 or 20 years, we will once again revisit the continual major blunder in US health care workforce reports - the lack of a SMART design steady for 30 - 50 years in the future instead of oscillating wildly.

The consequences of non-primary care excesses are substantial. Those that have promoted generic expansions and non-primary care excesses will have once again visited more problems upon our children and grandchildren as fantasy does not translate into reality.

The financial and economic repercussions of health care design are still poorly recognized. The US has clearly found a way to limit primary care workforce and primary care spending. Low priority assures little growth in this spending relative to increasing non-primary care demands.

Non-primary care workforce is quite another matter. Non-primary care workforce has always found a way to escape cost limitations (increased volume, more testing). Non-primary care expansions have been a major reason for continued health care cost increases and will be a major reason why 2020 will bring 20% of the GDP spent upon health care. This leaves less and less for all other areas other than health care. Even worse is increasing costs for people intensive endeavors such state government budgets, local and school district budgets, federal costs, and American employers. Greater deficits and lesser productivity are programmed in place by failure of health care design.

Only Specific, Measurable, Achievable, Realistic, and Timely designs work - with permanent primary care the specific focus along with sufficient primary care spending to deliver primary care to an entire nation rather than just half.

A best guess is that about 10 - 15 years from now (sooner if health care costs are reigned in more) there will be too much non-primary care workforce. Of course those claiming economic benefits from expansion will still be claiming these benefits as health care consumes an ever larger proportion.

By the time we figure out the consequences of too much, it will be too late to stop the momentum as there will be another 15 to 30 years of increase even if annual graduate levels no longer increase. Once again it is easy to forget that an annual graduate level takes 30 years for full realization. Increases in non-primary care (or any workforce) that are too rapid inevitably overshoot the mark 15 - 20 years later. There are other consequences to consider. If nurse practitioner annual graduates double as in each 6 to 12 years since 1980 the consequences will be dire for basic registered nursing workforce depleted of more and more experienced RNs.

Another decade of expansion of non-primary care workforce will result in many other national financial and economic concerns. The US rapidly forgot the lessons of economic decline as the result of rapidly rising health care costs. Few now remember $1200 of the cost of a car required for health care. The cost is now higher. Not surprisingly manufacturing has been substantially removed from the United States due to past, present, and future health care costs.

The lessons of the 1990s remain valid - costs reigned in for even a short time set the nation on course for one of the longest recent runs in economic progress in recent US history - even with a brief design change.

Spending more on health care is not going to contribute to economics in any area other than economic ruin. Designs that result in too much non-primary care must also be reigned in if there is to be hope for economic recovery.

Meeting Primary Care Needs in the Last Half of the 21st Century

Achievable primary care is one of the weakest areas in the workforce literature. Enough annual graduate expansion of any primary care source will increase primary care, but steady declines in the proportion remaining in primary care in the years after graduation make this a less efficient and less effective process for the purpose of addressing primary care workforce.

With primary care graduates departing primary care
it is not possible to recover primary care. 

Despite projections of primary care increases by government and by major associations, the US is actually losing ground in basic health access primary care. Increases in cost of training, more graduates required for less primary care result, and increases in the costs of locums, recruitment, and retention insure failure in primary care delivery with the current voluntary and flexible design that facilitates departures from primary care.

Studies and reports that project primary care workforce place too much emphasis on annual graduate numbers. The US has created 4 new primary care sources and has doubled annual primary care graduates from 14,000 in 1980 to 28,000 in 2010 but this matters less than the decline from 18 Standard Primary Care Years per graduate to 7 SPCYrs. With less than one-third remaining in primary care, the United States must find ways not to lose primary care rather than throwing more dollars at more graduates that deliver less primary care per graduate.

More important than annual graduates is what each graduate accomplishes. This is more about years in a career, activity in US health care, volume, and primary care retention. The Standard Primary Care Year uses estimates of these four factors to generate a relative measure of primary care delivery for each source specific for each class year.

The figure of 400 million people represents a good estimate for a steady state US  population in the latter half of the 21^st Century as population growth slows. Health Resources and Services Administration recommendations of 95 primary care physicians per 100,000 can be adjusted to 110 primary care physicians per 100,000 as the result of aging changes effective for 2030 and beyond. This results in a need for 440,000 primary care physicians to reach a sufficient 110 per 100,000. Already it should be apparent that moving down in the last decade instead of up is not the path to sufficient primary care.


These levels can be converted to Standard Primary Care Year estimates by source to indicate how many annual graduates of each source will be required per class year for 30 class years to reach sufficient primary care.

Setting Primary Care Sufficiency: Total Workforce, Annual Graduates, and Primary Care per Graduate

Family medicine is most consistent in primary care retention and has reached a steady state. Family medicine can be used to convert total workforce to annual graduates to Standard Primary Care Years per graduate. This figure can be used to convert other sources to annual graduate requirements for sufficient primary care.

Family medicine at 3000 annual graduates for 30 years has resulted in 100,000 active family physicians for a workforce. The 100,000 for 310 million people is a workforce of 32.4 family physicians per 100,000.

In 30 years the US will have about 400 million people. For 400 million people at 110 primary care physicians per 100,000 this would require 440,000 family physicians.

For 440,000 family physicians rather than the current 100,000 this would require 4.4 times more annual graduates. The steady state 3000 annual graduates times 4.4 is 13,200 annual graduates for sufficient primary care defined as 110 per 100,000 for 400 million people.

These beginning annual graduate points and ending family physician primary care workforce numbers can be converted to Standard Primary Care Years. The SPCYr estimate of 24 SPCYrs for a family physician can be multiplied by 13,200 annual graduates for about 320,000 Standard Primary Care Years per class year. This is significantly above the recent annual yield of 200,000 SPCYrs per class year from six sources from 28,000 annual graduates (NP, PA, IM, FM, PD, MPD). Once again the SPCYr allows career years, activity, volume, and primary care retention to be considered such that all primary care training sources can be compared to each other and across the class years.  

One more adjustment is needed due to declines in family medicine of about 10% less per graduate in SPCYrs over decades of time. This results in about 350,000 Standard Primary Care Years annually required of 30 class years of graduates to result in steady state 110 primary care physician equivalents per 100,000 people.Note that this figure does not include a 10% - 20% fudge factor that should be included in all primary care estimates. This is because new workforce creations have consistently stolen primary care to result in substantally less tha predicted as noted with the creations of emergency medicine, geriatric, hospitalist, and urgent care workforce. Unanticipated losses also include tens of thousands converted from primary care to teaching hospital workforce due to resident work hours restrictions. Hospitalist changes also thrust more workload from hospital to primary care with primary care nurses most impacted. Increasing fragmentation and regulation by government and insurance also result in less efficient and effective primary care delivery.

The 350,000 SPCYr figure is not likely enough, but can be used to determine how many annual graduates are needed in each source to reach a level equivalent to 440,000 primary care physicians for 110 primary care physicians per 100,000. This can be used to estimate annual graduates for 80 and for 60 primary care physicians per 100,000 as well.


Annual Graduates Required to Reach Sufficient Primary Care for 2050 and Beyond

	FM	PD	NP	PA	IM
Higher SPCYr Estimate (requires least annual graduates)	12,709	25,995	66,667	63,738	51,237
Lower SPCYr Estimate (requires most annual graduates)	17,090	39,063	115,056	108,025	114,379
Using Highest or Best Case Estimate in Each Factor
Primary Care Retention	90%	44%	40%	26%	25%
Active for Career	90%	90%	70%	80%	90%
Years in Career	34	34	25	33	33
Volume Adjustment	100%	100%	75%	80%	92%
Higher SPCYr Estimate	27.54	13.46	5.25	5.49	6.83
Using Lowest in Each Factor
Primary Care Retention	80%	35%	30%	18%	15%
Active for Career	80%	80%	65%	75%	80%
Years in Career	32	32	24	32	30
Volume Adjustment	100%	100%	65%	75%	85%
Lower SPCYrEstimate	20.48	8.96	3.04	3.24	3.06
Using the Average
110 per 100,000	14,899	32,529	90,861	85,882	82,808
80 per 100,000	10,836	23,657	66,081	62,459	60,224
60 per 100,000	8,127	17,743	49,561	46,844	45,168

Incredibly high levels of annual gradutes are required to reach sufficient primary care when the sole source of primary care is not permanent.


Permanent primary care sources most active for the most years at highest volume require the fewest annual graduates to reach sufficient primary care. Flexible sources less likely to remain in primary care with fewer years, lesser activity and lower volume  require more graduates than the United States can supply or afford.

The Impossibility of Reaching Sufficient Primary Care without SMART

Currently the US only graduates about 100,000 Registered Nurses, 30,000 physicians, 9000 NPs, 7000 Internists, 6500 physician assistants, 3000 pediatricians, 3000 family physicians, and 500 medicine pediatric physicians each year.


RN workforce demand will increase substantially in the next 20 years for the same elderly and health care coverage increase reasons as primary care. Too few enter and remain as RNs already. Increasing nurse practitioners by tens of thousands of annual graduates to reach sufficient primary care would devastate RN workforce. 

The last physician assistant expansion doubling actually increased primary care numbers entering the workforce by only 30% - a level likely to be negated in future departures from primary care. Physician assistants and nurse practitioners are widely sought by a wide range of employers with new specialties created with each passing year. Win-Win-Win non-primary care benefits to the practitioner, employer, and specialist physician are innate in the US policy design and insure departures from primary care.

Low primary care yield defeats generic expansions as a primary care workforce intervention. Generic physician expansions of medical students or generic expansions of graduate medical education positions are just not specific to primary care. Even expansion limited to primary care GME positions fail as pediatric and internal medicine expansions fail to yield much primary care workforce increase. About half of family nurse practitioners deliver the predominant primary care of all nurse practitioners. Expansions without retention are mostly about non-primary care workforce result. Three dimensions insure too much non-primary care. Meanwhile primary care remains stagnant by design.

Only SMART expansion works – specific, measurable, achievable, realistic, and timely. A best approach is a primary care source that delivers more primary care than family medicine for about 12,000 annual graduates needed at a cost of 12 billion per year for sufficient primary care. This compares to the current 16 billion for 28,000 annual graduates with half enough primary care delivery capacity result (all costs of higher education and training).

Specific is the least costly because it is most specific. Primary care sources that are flexible have become 55 – 80% not primary care in yield in the United States. Sources claiming to be primary care solutions are not good solutions. Specific language is required. Advanced nurses and generic nurse practitioners are not primary care solutions.Only the nurse practitioner that remains in family practice employment (25%) is a good primary care solution. Requiring permanent family practice is the only way to reach this solution. Only the physician assistant that remains in family practice is a good solution as demonstrated by consistent primary care delivery and consistent distribution where needed but again this is only 25% of physician assistants and only 20% of new graduate PAs. Schools such as Duluth manage 50% family medicine despite the policies that drive medical students away from a permanent primary care source, but 100% entry is possible in an even better design.

What is clear is that there has been little thought and even less planning

with regard to sufficient primary care workforce.

Rapid increases in the cost of training make these even more challenging. Also primary care delivery capacity per graduate is likely to decrease with quality focus or collaborative care emphasis – models that result in lower volume per primary care graduate. Departures of primary care workforce to urgent, emergent, hospital, and subspecialty workforce will also result in fewer Standard Primary Care Years per graduate. New workforce is created each few years that steals from primary care. Historically this has included emergency medicine, geriatrics, sports medicine, and hospitalists. Resident work hours restrictions alone converted tens of thousands of NPs and PAs away from primary care.

The United States has actually avoided expanding the most permanent primary care source for over 30 years. It has chosen to expand the sources least likely to remain in primary care. These have not been SMART choices specific to primary care and indicate choices made for other reasons – including non-primary care workforce, academic interests, and teaching hospitals.

Longer training also defeats primary care workforce. Each additional year of training results in 3 - 4% less workforce yield - resulting in even more graduates required to reach sufficient workforce. Two more years required for nurse practitioner doctorates in all nurse practitioner graduates in 2015 will result in an 8 - 10% loss of NP workforce and will substantially raise the cost of training.

The consequence of increased cost of training with less workforce result
has not even been recognized

as yet another move in the opposite direction from sufficient primary care.

Family medicine and other sources remaining in family practice have demonstrated consistent distribution to the elderly, poor, near poor, CHC, rural, and underserved populations as well as 53% found in 30,000 zip codes with 65% of the US population most left behind in health access. Expansions involving IM, PD, most nurse practitioners and most physician assistants are expansions of graduate types known to fail in distribution. This is also why hybrids or combinations of primary care are a bad idea.

Better than family medicine is required for efficient and effective primary care recovery using SMART principles, but family medicine is a starting point.

HRSA Nursing Reports for 2004 and 2008 were used for NP data as well as Advance for NP and PA. AAPA data was used for PAs. HRSA 2008 Physician Projections were used for sufficient primary care. Other references came from the Standard Primary Care Year literature, Ferrer, Mold, and Rosenblatt.

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

SMART basic health access - Specific, Measurable, Achievable, Realistic, Timely

Summaries and Links to Fifty Basic Health Access 2011 Blogs

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