Training Rural Doctors: The KU Salina Program

.

Among the many forms of disparity in access to health care, the imbalance between rural and urban/suburban areas in terms of physician and other healthcare provider workforce is one of the most obstinate. The problem has been well-documented and been discussed by myself and others on a number of blog posts (e.g., Primary Care, Medical School Debt, and US Health Needs: Analysis from the Graham Center, May 31, 2001, Training rural family doctors, November 5, 2010). While the latest data from the Census Bureau show that the percentage of Americans living in rural areas has dropped from the 20% I have often cited to 16% in the 2010 census (as reported by Hope Yen for the Associated Press in Rural US disappearing? Population share hits low), 16% is not yet “disappeared”. In addition, the percent of doctors working in these areas (which I have previously cited at 9%) is almost certainly lower as well, given both retirements of older doctors and the low level of interest in entering even primary care (requisite for rural practice), not to mention rural practice itself, among graduating medical students and residents.

The problem of lack of availability of health care in many rural areas is further complicated by the aging of the rural population, which is associated with increased need for health care. There are wider reasons to fear the loss of rural doctors. Rural medical practices not only enhance health directly, they provide income and economic development for a community (jobs), and may indeed be one of the key determinants (along with schools) of why some rural communities will survive while others slowly disappear. We need strategies to increase the number of students matriculating in medical schools who are interested in rural practice, and strategies for encouraging and supporting them to enter family medicine (which is the main medical field that works in rural areas, although there are also roles for other primary care doctors such as general internists and general pediatricians, and some specialists, particularly general surgeons).

This is why the fact that the University of Kansas Medical School (KUSoM) is opening a 4-year medical school campus in the small central Kansas city of Salina is big news, not only in the state (as in reports from the Kansas Health Institute [KHI]) but nationally (as demonstrated by front-page coverage in the New York Times and on NPR). Like many state medical schools, KUSoM has long had a regional campus, in Wichita, where a portion of medical students have done their third and fourth (“clinical”) years of training. That campus will be expanded to have 8 new first-year students this year, and 28 beginning next year, who will spend 4 years in Wichita. But the bigger news is in Salina, a north-central Kansas city of just under 50,000 whose previous opportunity for fame as the place where Bobby McGee “slipped away” in Kris Kristoffersen’s song itself slipped away when Janice Joplin, in the most widely-known version, sang it incorrectly as “Salinas” (well, she was from Texas and lived in California!) [OK. I HAVE IT ON GOOD AUTHORITY THAT I AM WRONG AND IT WAS ALWAYS SALINAS. OH WELL, IT WOULD HAVE BEEN A GOOD STORY.]. Eight students per year will spend their entire 4 years in Salina, making it the smallest city to have a 4-year campus and making the “New Salina med school campus unique in US” according to Dave Ranney of the KHI. Or almost unique; Indiana University School of Medicine has done something similar, expanding a first-two-years campus in Terre Haute to 4 years. The KUSoM Wichita campus has done a good job of producing primary care doctors who practice in rural Kansas compared to most places, but Wichita is the largest city in Kansas, not in itself at all rural.

A.G. Sulzberger notes in his piece, “Small town doctors made in a small Kansas town” (NY Times, July 23, 2011) that “when one visitor from the Liaison Committee on Medical Education, an accrediting body whose approval was considered a major hurdle, [s/he]remarked with surprise that the area was not just cornfields.”  Of course not! Kansas is the Wheat State! That aside, Salina, on the banks of the Saline River in the Smoky Hills, does sit in the middle of an agricultural region. Indeed, it is the home of the Land Institute, a wonderful organization dedicated to developing a perennial prairie grass that will produce an economically-usable grain, that I take every opportunity to direct people to. 

The perspective of New Yorkers aside, Salina is not really a small town. A 4-year curriculum equivalent to that in KC or Wichita would be impossible in too small a town because the community must have at least the “core” specialists of internal medicine, pediatrics, surgery, obstetrics-gynecology, and psychiatry as well as family medicine. Salina is a prosperous town with a large enough medical community to support all the “core” clerkships that medical students need to take in their 3^rd year, as well as faculty to lead the small groups for Problem-Based Learning (PBL) sessions, and to support the basic biomedical science education. All of the materials for the modular curriculum for the first two years of medical school is available on-line, and the fact that all lectures are podcast means that Salina students will have the option of emulating many Kansas City-based students and not attend them all. But if they do, they will be directly tied in via high-quality, high-resolution interactive TV, and will even be able to ask questions of the KC-based lecturer just as if they were in the room.

The CEO of Salina Regional Medical Center, Micheal Terry, is quoted by Sulzberger as saying “When they go off to the ritz and the glitz and pick up a spouse from the big city, it’s always hard to get them back to small-town America.” Even if you think that the ritz and glitz of Kansas City and Wichita do not make them New York or Paris (but you should visit these cities before judging), it is not just “how you gonna keep ‘em down on the farm after they seen Paree”. Medical students are usually in their 20s, at the age where they often meet their spouses, and they meet them where they live. If that is in Kansas City or Wichita, it is more likely that the spouse will be from that area and unwilling to move to a rural area, or have a job that precludes them from doing so. If the Salina program is successful, it could be a model for decentralizing even more of the KUSoM curriculum to other Kansas cities, and of course for other states.

But how will we know if it is successful? In the very short term, we will see how the students and their teachers assess the experience, as well as student performance on exams. In the relatively short term (4 years), we will see whether they mostly enter primary care programs, and some years after that, whether the indeed enter rural practice. We don’t know, of course, whether, even if they do, it will actually increase the number of students entering rural practice, because some already do. Since the students were admitted first and then those who wanted to go to a rural site chose Salina, there is at least a possibility that 8 students who would have entered rural primary care practice anyway are matriculating in Salina, and we just took 8 additional suburban students in Kansas City who will not. We hope this is not the case, and look forward to expanding the program.

In any case, it will be extremely important if it demonstrates that medical education can be decentralized and effectively taught in smaller communities, and that it can be done with high quality and in a setting that does not demand relocation to major metropolitan areas. With this new setting, we have to make sure that our admission process favors students who are most likely to become rural primary care physicians. If we do that well, we may really have something here!

.

Training rural family doctors

.
In a recent report from the University of Washington’s WWAMI Rural Research Center, “Family Medicine Residency Training in Rural Locations”,[1] Chen et. al. repeat their 2000 study of rural training in the US. They note that this is very important given the health needs of the American people, 20% of whom live in rural areas, most of which are underserved given the less than 9% of doctors who practice there (Primary Care and Rural Areas, April 28, 2010; Medicare Costs in Rural America: A case of reaping what we haven't sown?, March 26, 2009). Most rural training is of family doctors, but a majority of it comes from urban programs.

How is this so? This is because over 90% of the family medicine residency programs in the nation are located in urban areas, and only 7% in “large rural areas”; it is very hard to have a program entirely in a “small rural area” as the requirements for family medicine training include a great deal of time spent in the hospital and working with other specialists. Nonetheless, the authors surveys discovered that nearly half of the family medicine programs felt that training rural doctors was an important part of their mission.

From the Executive Summary: “Rural residency programs provided a higher proportion of rural training than urban programs, but because more programs were in urban settings, there was more rural training in urban than rural programs. Overall, 15% of respondents had a formal RTT [rural training track]; of these, 61% were in rural and 10% in urban programs. Rural residency programs conducted 83% of all rural residency training in RUCA[2]-defined rural areas. Comparing the 2000 and 2007 surveys, rural training increased from 372 to 408 FTEs, but rural training in urban programs declined from 186 to 79 FTEs, resulting in an overall decrease in the amount of rural-focused family medicine training.” See figure for example of a “RUCA map” in a state (Kansas) with a high percentage of “isolated rural” areas; Other states (say, New Jersey) look different!

The authors found that while virtually all (99.9%) of the training FTEs reported as rural in rural programs in fact occurred in rural (RUCA-defined) areas, only 21.7% of those reported by urban programs were in such areas. The net result was that only 7.3% of family medicine residency training FTEs are in rural areas, with 83% of these conducted by rural programs. Interestingly, to me, a higher percentage of urban programs listed rural training as “very important” (45.3%) than listed urban (underserved) training (40.4%).

The authors discuss the policy implications of this situation, including the possible impact of several components of the ACA health reform law that encourage training in rural areas. An important one has to do with Medicare funding of GME (residency) positions, the arcane but major source of funding for such training (Funding Graduate Medical Education, May 25, 2009; Public Law 111-148, Sect 5506, p 661). The change would make it easier for non-hospital sponsors of training (such as community health centers, an area the authors have also written about[3]) to receive this funding (Public Law 111-148, Sect 5508, p.668). The law – and policy – also favors redistribution of unfilled GME spots to rural hospitals. However, there are counterpressures: many big urban hospital training sites are “over their cap”, meaning they have more residency positions than are funded by Medicare. In these cases, the hospital, or another entity, funds the positions; these are virtually always in those specialties that make more money for the hospital, not in primary care, and certainly not in rural tracks (Primary Care and Residency Expansion, January 7, 2010 ). As they expand, in response to the demand for more physicians and medical school class increases, they will exert great political pressure to have new GME slots assigned to them to help cover their costs.

So here is the situation:
· There is a tremendous shortage of doctors in rural areas.

· Rural family medicine training programs, particularly those located in rural areas, are very effective in producing physicians who will practice in those rural underserved areas.

· Only 7.3% of family medicine (FM) residency positions are in such rural programs. This is way better than any other specialty, and may in fact be higher because while the authors did not count many of the trainees claimed as rural by urban FM residencies because they are not actually occurring in RUCA-defined rural areas, “Training locations may be defined as 'urban' using RUCA definitions but may still be sparsely populated and serve a predominantly rural patient populations.”

· Even if all FM positions claimed by their sponors as rural are in fact so, FM residents are only about 10% of all residency positions (2,630 of 25,500 in 2010), and virtually no other residents are trained for, not to mention likely to enter, rural practice…despite the need for other specialists, especially general surgeons, in these communities.

· ACA has a number of components that target an increase in the production of rural physicians (particularly family doctors and general surgeons, Public Law 111-148, Sect 5501, p. 534), but even if these are fully implemented, and taken advantage of by students, it will be many years before they have a significant impact.

· Despite these ACA changes, there will be counterpressures to use many of the new training slots for training non-primary care residents in urban sites, and particularly in academic health centers; these will come from both the academic health centers and their teaching hospitals themselves, the subspecialty physician medical societies, and the medical students themselves who overwhelmingly prefer to live in urban areas, as well as to earn the much higher incomes of subspecialists.

How might this be changed or modified? The provisions in the ACA law funding of GME slots to non-hospital settings such as community health centers, preferentially training primary care doctors (and general surgeons), and supporting the growth of community health centers and health extension services are a start, but they are not sufficient. The key is going to be greater incomes for doctors practicing primary care and general surgery in rural areas compared to those practicing subspecialties in urban areas. Supply and demand is insufficient; while there is demand for such doctors in rural areas, there is demand for more subspecialists by urban hospitals that see them as cash cows. Medicare sets reimbursement rates, as almost all payers model their rates as percents (usually higher) of Medicare’s. The reimbursement has to change so dramatically as to make primary care doctors in rural areas make, if not more, at least not nearly so much less as subspecialists in cities.

This financial change will begin to level the playing field. Medical students who have no interest in rural practice can still stay in urban areas, and even in subspecialties, but they should have to pay a financial price, making less than they would if they were to practice in what (to them) might be considered a less desirable area. As long as we make some careers pay a lot more, often for no more or even less work, as well as be located more popular (urban metropolitan) areas, we cannot expect any different outcome from the one we currently have.



[1] Chen FM, Andrilla CHA, Doescher MP, Morris C, “Family Medicine Residency Training in Rural Locations”, Final Report #126, WWWAMI Rural Health Research Center, University of Washington School of Medicine Department of Family Medicine, July 2010.
[2] RUCA = Rural-Urban Commuting Areas; a measure of population density that accounts for nearness to an urban area; obviously two counties may have similarly low population densities, but if one is surrounded by similar counties while the other is adjacent to a county with a large – or moderate – urban area, the first is “more” rural.
[3] Morris CG, Chen FM, Training Residents in Community Health Centers: Facilitators and Barriers, Annals of Family Medicine, Nov2009;7(6):488-94

Medicare Costs in Rural America: A case of reaping what we haven't sown?

This guest column is by Donald Frey, MD. Dr. Frey is Professor and Chair of the Department of Family Medicine at Creighton University in Omaha, Nebraska, and an expert in rural primary care. He is also the accomplished author of the fiction book "Medicine: a Novel"

The recent New England Journal of Medicine article by Elliot Fisher describing the latest data release from the highly respected Dartmouth Atlas Project[1] underscored what many have known for some time—Medicare costs are soaring.

The study compared state and regional growth in Medicare costs between 1992 and 2006. But in addition to the overall rise in costs, the continued marked variance among costs across all states/regions was particularly disturbing. These cost discrepancies, as in previous studies, were independent of patient severity and disease status. In some instances, states that had previously been low on the cost index had jumped significantly. Others showed relatively moderate increases. None showed cost reductions.

The figures for Nebraska are particularly interesting. Sitting squarely in the center of the country, Nebraska has two metropolitan areas (Omaha and Lincoln) and a large region of rural and frontier counties. In 1992, Nebraska was at the top of the heap when it came to efficient use of Medicare dollars, ranking 51st out of 51 states and territories in Medicare costs per enrollee.

But something funny happened on the way to the 2006 survey. Costs per enrollee skyrocketed 5.3 % and Nebraska soared to 39th in Medicare spending. Interestingly, the greatest cost increases were in the most rural portions of the state.

What happened? The authors offered very little in the way of specificity. Undoubtedly, further analysis of these data may result in significant insights into why costs are rising more rapidly in specific areas as opposed to others. But at present, it is important to note what the extensive earlier work of the Dartmouth Atlas project has already shown.

Repeatedly, the Dartmouth researchers have shown the positive correlation between the percent of state physician workforce comprised of primary care physicians and lower costs and improved outcomes. States with higher concentrations of specialty physicians tended to have higher costs and poorer outcomes. Such findings are right in line with the international data generated by Dr. Barbara Starfield at Johns Hopkins.

All of this would be consistent with Nebraska's status in 1992. At that time, Family Medicine was the backbone of health care delivery in the state, particularly in rural areas. Even in the metro areas of Omaha and Lincoln, primary care was readily available and widely utilized. Four residency programs, all with a strong rural component, produced 32 residency graduates per year, many of whom remained in state to practice following graduation.

But in the ensuing years, Nebraska's urban centers began investing heavily in developing overlapping and competing subspecialty services. The number of medical students entering Family Medicine plummeted, and urban hospital systems found that a cost effective way to cover their primary care needs was to loot rural areas of their physicians. Rural communities began to experience increasing difficulty meeting their workforce needs. Some towns saw workforce reduced by more than 50%, with some 4-5 doctor communities reduced to only 2 Family Physicians.

Physicians in these communities suddenly found themselves overwhelmed. Local systems that previously were highly organized with respect to call coverage, division of labor, procedural practice, etc., were now finding it impossible to continue providing the same level of service in the face of such huge demand simply for basic care. Thresholds for referral began to lower. In order to manage the most basic community needs, cases that previously would have been readily handled locally by community Family Physicians had to be referred out to regional centers. These centers, specializing in more intensive—and more costly—care, began to manage these patients instead.

As the demand for more intensive (and more lucrative) care increased, urban health system recruitment of sub-specialists intensified, with an all-too-willing crop of medical graduates flooding local sub-specialty training programs, compromising even further the availability of Family Medicine graduates to "backfill" rural areas.

The result? Care that was more costly, further removed from the community, with no discernable improvement in quality.

Whether changes in workforce alone can fully explain Nebraska's explosion in Medicare costs is certainly open to question at this time. Hopefully, further analysis of state-wide Medicare costs, along with comparative data from other states will more fully illuminate this issue. In the meantime, given the exhaustive data regarding primary care and Medicare costs already produced by the Dartmouth Atlas project, the notion that Nebraska's significant rise in Medicare costs could be best explained by a statewide shift away from primary care remains a disturbingly plausible possibility.

If this is the case, rural states like Nebraska, where Family Medicine has traditionally provided a greater percent of overall care, may turn out to be harbingers of even greater increases in overall healthcare costs for the nation, as even those states that traditionally have relied less on primary care for health care delivery shift an even greater percent of their workforce to sub-specialties.

Regrettably, the influential Association of American Medical Colleges (AAMC) remains steadfast in its opposition to restructuring residency training positions to reflect the growing national need for Family Physicians, instead advocating for an increase only in total medical school student enrollment, with the seemingly blind-faith position that “the market” will drive graduates to where they are needed most.

An unregulated market that’s supposed to make everything work out fine? Certainly sounds familiar in 2009.

If the planned expansion of medical school enrollment by 30% occurs as planned, along with a continued insistence that market forces will auto-correct physician workforce, the work by Fisher, Starfield, Wennberg, Shi, and so many others[2] would predict that health care outcomes in this country will actually worsen because of the AAMC’s actions. If this occurs, it will reflect a perverse twist on one of the oldest principles of agriculture known to everyone here in Nebraska—rather than reaping what we sow, the outcomes we inherit will instead be due to what we haven’t sown.

[1] http://content.nejm.org/cgi/content/full/360/9/849

[2] Many of these articles are previously referenced on this blog. See especially December 11, 2008.