Central Journal of Family Medicine & Community Health

Cite this article: Aysola J, Groves DS, Hicks LS (2015) Health Center Professional Programs and Primary Care Workforce. J Family Med Community Health 2(8): 1063.

Abstract

Background: Current policy promotes health center professional training and pipeline programs as solutions to bolster primary care workforce in shortage areas, despite the paucity of evidence.

Methods: We analyzed data from US health centers we surveyed from March to June 2010, merged with federal health center data, to estimate associations between health center training and pipeline programs and provider recruitment and retention.

Results: Of the 976 surveyed, 391 health centers responded. Health centers with career ladder programs compared to those without had higher adjusted rates of no/minimal difficulty in recruitment of primary care providers. (17.6% vs. 10.6%; p=.01) and close to double the adjusted rates of reporting no/minimal difficulty in retention of primary care providers (39.4% vs. 21.2%; p=.0001).

Discussion: There remains a need for further evaluation of health professional programs in order to expand models, such as career ladder programs, that demonstrate effectiveness in improving the primary care workforce in shortage areas.

*Corresponding authorJaya Aysola, Division of General Internal Medicine, Perelman School of Medicine, University of Pennsylvania, 1111 Blockley Hall, 423 Guardian Drive, Philadelphia, PA, 19104-6201, Department of Health Care Policy, Harvard Medical School, Division of General Medicine, Brigham and Women’s Hospital, USA, Tel: 215-573-9475; Fax: 215-573-8778; Email: jaysola@upenn.edu

Submitted: 03 November 2015

Accepted: 03 December 2015

Published: 05 December 2015

ISSN: 2379-0547

Copyright© 2015 Aysola et al.

OPEN ACCESS

Keywords•Access to care•Primary care•Workforce•Health care reform•Safety net

Research Article

Health Center Professional Programs and Primary Care WorkforceJaya Aysola1,2,3*, DaShawn Groves4 and LeRoi S Hicks2,5,6

1Division of General Internal Medicine, University of Pennsylvania, USA2Department of Health Care Policy, Harvard Medical School, USA3Division of General Medicine, Brigham and Women’s Hospital, USA4Department of Health Care, National Association of Community Health Centers, USA5Division of Hospital Medicine, University of Massachusetts Memorial Medical Center, USA6Department of Medicine, Christiana Care Health Care System, USA

INTRODUCTIONOver the past decade the number of medical students going

into primary care has steadily declined, even as the need for primary care physicians continues to increase [1]. Accounting for population growth, aging, and the current rate of production of primary care physicians, the projected national shortage of primary care physicians by 2025 is over 35,000 [2]. Further, it is likely that expansion of health insurance coverage under the Affordable Care Act will require a greater number of primary care physicians and likely exacerbate this projected shortage [3,4].

Primary care providers serve as the first and sometimes only contact with our health care system for many individuals [5,6]. Evidence supports the importance of both access to primary care providers, and receipt of primary care services for the prevention of early morbidity and mortality, especially for vulnerable populations [7-10]. Therefore, it is imperative to identify programs that increase the likelihood of trainees going into primary care and practicing in underserved communities.

Health centers play a crucial role in delivering comprehensive

primary care services in designated health professional shortage areas, yet they continue to face challenges recruiting primary care providers [11-15, 16]. Many participate in health professionals training programs, such as physician residency programs, or pipeline programs, such as Area Health Education Centers (AHEC), designed to encourage individuals to pursue careers in health care and work in underserved areas. Prior studies suggest that health center affiliation with such programs is a potential solution for the mal-distribution of primary care providers [17-21]. For example, Morris et.al surveyed a regional network of family medicine residents and found family physicians that trained at health centers were twice more likely to work in underserved settings than those who did not [17]. Nottingham et.al demonstrated positive effects of West Virginia AHEC, on provider recruitment and retention [21]. However, studies to date are limited to either providing theoretical frameworks in support of the workforce benefits of health center training and pipeline programs or to findings from local studies that may lack generalizability.

The Affordable Care Act includes several provisions for

Central

Aysola et al. (2015)Email:

J Family Med Community Health 2(8): 1063 (2015) 2/6

expanding the role of such health professional programs in an effort to address future primary care workforce needs [22]. However, there lacks sufficient knowledge of the prevalence and impact of these programs. Therefore, we conducted a national survey to identify health centers with professional training and pipeline programs, and evaluated their effects on primary care provider recruitment and retention.

MATERIALS AND METHODS

Survey sample and administration

In partnership with the National Association of Community Health Centers (NACHC), we developed a self-administered survey to evaluate the effects of health center professional training and pipeline activities. Survey questions elicited closed-ended responses on health center professional programs: 1) descriptive characteristics 2) funding sources 3) experiences with participation and 4) provider recruitment and retention. We adapted questions from a previously operationalized 2007 NACHC survey of health centers and created additional questions informed from key informant interviews and literature review. The survey was pilot tested for clarity and face validity. We fielded the online survey from March to June 2010, to executive directors/designee of 976 of 1084 federally funded health centers with active email addresses in 2009 Uniform Data System (UDS), excluding only 108 health centers that previously opted out of survey solicitations. We sent out reminders on average twice a week to non-responders and to encourage participation employed a lottery-based incentive of 500 dollars for ten selected health centers that completed the survey. We linked our data to the 2010 Uniform Data System(UDS), which provides a range of information on the operations and performance on all federally-funded health centers (n=1124). The Human Studies Committee of Harvard Medical School approved this study.

Survey questionnaire

Our primary predictors were the presence or absence of 1) health professional training programs 2) Area Health Education Center programs (AHEC) and 3) career ladder programs. Health professional training programs include any training program for health professional students or providers affiliated with health centers. Examples include medical residency training programs, clinical electives for medical students, or nurse practitioner training programs. Level of health center involvement in training could range from hosting a one month elective to hosting/managing the primary care residency program, the latter termed a ‘teaching health center’ (THC) [18,23]. AHECs implement community-based training programs in health careers for students and continuing education for healthcare providers in effort to improve supply and distribution of healthcare professionals, with a statewide or regional focus [24-26]. Career ladder programs offer further training or education to any employee with the opportunity of promotion and career advancement within the health center upon completion. Training and educational opportunities with career ladder programs include, but are not limited to advanced degrees in administration, finance, dentistry, or nursing. For example, a registered nurse (RN) is becoming a Nurse Practitioner (NP) or a physician acquiring a Masters in Business Administration (MBA) [27]. Additional covariates

include hospital/medical school affiliation and receipt of Title VII funding for primary care training, previously shown to be associated with primary care providers working in health centers or other National Health Service Corps (NHSC) sites [28,29].

Our primary outcomes were difficulty in primary care provider recruitment and retention in the past five years, measured with a scaled response (no/minimal, some, great). We defined ‘provider’ to encompass primary care nurse practitioners, physician assistants, and physicians (MD/DO). We defined provider retention as health care providers who have stayed with the health center beyond three years of hire. Federal loan repayment and J-1 visa waiver programs require a minimum of two to three years of service in designated provider shortage areas and many providers fulfill this obligation at health centers [30]. A cited goal for health centers is to retain providers at least one year beyond this obligatory period; therefore, we defined retention as longer than three years [31].

Uniform Data System (UDS)

We obtained the following health center characteristics from the 2010 UDS: 1) urban versus rural location; 2) census region; 3) the percent of health center patients/users by categories of income level, race/ethnicity, and insurance status; 4) the number of clinical delivery sites;5) patients seen at least once during the 12 month reporting period; and 6) medical providers (total number of full time equivalent (FTE) primary care providers per health center) [32] (Table 1).

Statistical analysis

We compared survey respondent characteristics to all health centers in the 2010 UDS, using t-tests or chi-square tests as appropriate to determine if our sample was nationally representative. In ordinal logistic regression models, we evaluated the association between health center professional programs and degree of difficulty in recruitment and retention of primary care providers, adjusting for urban versus rural location, census region, hospital affiliation, receipt of Title VII funding, total number of sites, patients seen annually, number of providers’ full time equivalents (FTE), percent minority patients, percent uninsured, and percent of patients with incomes 100% below the federal poverty line (FPL). We added interaction terms to our models to determine any significant combined effects of our predictors on workforce outcomes. We also estimated associations between the level and type of involvement of health centers in professional training programs (e.g. hosting one-month primary care resident elective vs. managing a residency program) and workforce outcomes, adjusting for the same health center characteristics detailed above. When the proportional odds assumption was not met, we ran partial proportional odds models [33].

We restricted our analysis to US health centers and excluded responses from health centers in US territories (n=14). We utilized STATA 12 for analyses. We report two-tailed p values and 95 percent confidence intervals for all statistical tests; p<0.05 was considered statistically significant.

RESULTS AND DISCUSSIONWe received completed surveys from 391 (40%) of the

Central

Aysola et al. (2015)Email:

J Family Med Community Health 2(8): 1063 (2015) 3/6

976 centers surveyed of which 371 (95%) were CHCs. When compared to health centers nationally, more study centers were located in the South and less in the West. There were no other significant differences between study and national health center characteristics (Table 1).

Of the 391 health center respondents, 341 (89.5%) reported having health professional training programs, 160 (47.6%) reported having AHEC programs, and 189 (54.8%) reported having career ladder programs. Of those that trained medical residents, 62 (18.2%), reported having an ongoing continuity care clinical experience for residents, and 19 (5.6%) reported managing a primary care residency-training program. Among centers reporting career ladder programs, 128 (67.7%) specified having advancement programs in administration, 70 (37.0%) in finance, 141 (74.6%) in nursing, and 56 (29.6%) in dentistry.

Provider recruitment

Of the 391 health centers in our sample, only 52 (15.0%)

reported no/minimal difficulty in recruiting primary care providers, while 114 (32.8%) reported great difficulty in provider recruitment. In unadjusted analysis, 3% more centers with career ladder programs reported minimal difficulty in recruiting primary care providers than those without. (Table 2) Conversely, a smaller number of centers with AHEC programs reported having minimal difficulty in primary care provider recruitment. (Table 2) In addition, health centers managing their own residency programs had a greater odds of reporting minimal difficulty in recruitment of primary care providers [OR (95% CI): 3.6 (1.2- 10.9)] then non-‘teaching health centers’ in unadjusted models.

Adjusted analyses revealed that 7% more centers with career ladder programs reported minimal difficulty in recruitment of primary care providers compared to those without (p=..01) (Table 3). After adjustment, we also found urban centers reporting significantly less difficulty in provider recruitment than rural centers. (Table 3) We did not find any significant interactions between our key predictors for provider recruitment.

PROVIDER RETENTIONHealth centers reported less difficulty with provider retention

as compared to recruitment, with 113 (33.5%) reporting no/ minimal difficulty in retaining providers beyond a 3 year period, compared with 54 (16%) reporting great difficulty. Unadjusted analysis demonstrated about 15% more centers with career ladder programs reported no/minimal difficulty in provider retention, compared to those without such programs (Table 2). After adjustment, about twice as many centers with career ladder programs versus those without, reported no/minimal difficulty in retention of primary care providers (p<.001) (Table 3). Other significant covariates include health center location and percentage minority patients served. Urban health centers were more likely to report minimal difficulty with provider retention than rural centers (Table 3). Health centers with a greater percentage of minority patients were less likely to report no/minimal difficulty in retaining providers. [Adjusted OR (95% CI): 0.99 (0.98, 0.99); p=.02] We did not find any significant interactions between our key predictors for provider retention.

CONCLUSIONIn this national assessment of health center training and

pipeline programs, we found 54% of health centers had career ladder programs for their employees that significantly enhanced their ability to both recruit and retain providers.

We did not detect differences in workforce outcomes between centers with and without AHEC affiliations; however, we found a positive association between AHEC affiliation and pediatric primary vaccination rate. Previous evaluations of AHEC, conducted within a decade of its inception, demonstrated benefit on the supply and distribution of providers in shortage areas [25]. Three decades later, our findings underscore the importance of continued assessment of these programs.

While 90% of health centers engage in some form of training health professionals, only 6% at the time of this survey would be characterized as ‘teaching health centers’ that host/manage their own primary care residency programs. These health centers

Table 1: Comparison of Study Health Centers to National Sample, 2010

CharacteristicsStudy Health

Centersn=391

National Health Centersn=1124

Mean number of sites 7.6 6.7Mean number of providers

(FTE)* 41.0 39.0

Mean number of users 18,434 17,322Census Region (%) †

South 44.1 36Northeast 18.1 17.5

West 14.9 27.3Midwest 22.9 19.2

Location (%)Urban 46.5 47.4Rural 53.5 52.6

Users by Race/Ethnicity (%)Non-Hispanic White 44.3 44.7Non-Hispanic Black 18.5 20.4

Hispanic 28.6 26.2Multi/Other 8.5 8.7

Users by Income (% FPL)Below 100 68.1 71.8

100-150 16.4 14.4150-200 7.2 6.5Over 200 8.3 7.2

Users by Insurance (%)Uninsured 38.4 37.5Medicaid 34 38.5Medicare 8.3 7.5

Private 17.1 13.9Other public 2.1 2.5

SOURCE: Bureau of Primary Health Care Uniform Data System, 2010.NOTES: Table compares the characteristics of study health centers to health centers nationally, using chi-square tests for categorical variables and t-tests for mean users and number of sites.*Number of providers’ full time equivalents (FTE)†p = .04 between study health centers and health centers nationally; comparisons of all other characteristics were not significant.

Central

Aysola et al. (2015)Email:

J Family Med Community Health 2(8): 1063 (2015) 4/6

Table 2: Number and Percent Health Centers Reporting Degree of Difficulty in Provider Recruitment and Retention by Key PredictorsDegree of Difficulty in Provider Recruitment

Degree of Difficulty in Provider Retention

Health Centers with Key Predictors Minimal Some Great p Value* Minimal Some Great p Value*

Health Professional Training n (%) .27 .18

Presence 48 (15.7) 160 (52.5) 97 (31.8) 98 (33.2) 153(51.9) 44 (14.9)

Absence 4 (11.4) 15 (42.9) 16 (45.1) 12 (34.3) 13 (37.1) 10 (28.6)

AHEC n (%) .01 .97

Presence 16 (10.4) 96 (62.3) 42 (27.2) 49 (33.8) 74 (51.0) 22 (15.2)

Absence 32 (19.2) 73 (43.7) 62 (37.1) 56 (33.5) 84 (50.3 27 (16.1)

Career Ladder n (%) .008 .02

Presence 31 (16.9) 105 (57.1) 48 (26.1) 73 (40.1) 87 (47.8) 22 (12.1)

Absence 21 (13.8) 67 (44.1) 64 (42.1) 36 (24.8) 78 (53.8) 31 (21.4)SOURCE: All information derived from Authors’ computations using survey response data.Notes: Health center responses regarding key predictors do not include missing or ‘other’ and therefore, percent health center with and without each key predictor may not add up to 100.*Using Chi-sq. exact tests and univariate ordinal logistic regression models to compare reported difficulty in provider recruitment between health centers with and without each key predictor

Table 3: Adjusted* Predicted Probabilities† for Reported Degree of Difficulty in Provider Recruitment and Retention among health centers by Significant Binary Predictors

Predicted Probabilities (%)Degree of Difficulty in Provider

Recruitment§

Predicted Probabilities (%)Degree of Difficulty in Provider Retention§

Significant Predictors Minimal Some Great p Value Minimal Some Great p Value

Career Ladder Programs .01 <.001

Presence 17.6 56.0 26.5 39.4 50.8 3.6

Absence 10.6 50.1 39.2 21.2 57.9 8.2

Location .04 <.001

Urban 18.6 56.2 25.2 40.7 50.0 3.4

Rural 10.6 50.0 39.4 22.0 57.9 7.9SOURCE: All information derived from Authors’ computations using survey response data.*Adjusted for key predictors and census region, location (urban vs. rural), hospital affiliation, title 7 funding, number of sites, patients seen annually, and providers (FTE), percent minority patients, percent uninsured, and percent of patients with incomes 100% below FPL †Predicted probabilities (marginal effects) for reported degree of difficulty in provider recruitment and retention for health centers with significant binary predictors when holding remaining covariates at their means ‡Partial proportional odds model applied with listed significant predictors meeting the proportional odds assumption for ordinal regression model §Proportional odds assumption met for this ordinal regression model

informed the 2010 ACA legislation that created the teaching health center graduate medical education (THCGME) program [22,34]. We found a significant unadjusted benefit for provider recruitment amongst the 6% of ‘teaching health centers’. Evidence demonstrating physicians from rural health residency training tracks practice in rural areas [35], as well as local studies of the workforce benefits of health center residency programs [19], lends support to our findings. Continued investment in the THCGME program [36], will allow for further evaluation of the effects of teaching health centers on primary care provider supply and distribution.

Factors driving primary care provider supply and distribution are complex with a lack of consensus on how best to determine

future workforce requirements [37-42]. However, agreement exists on the mal-distribution of primary care providers and the reliance on health centers in medically underserved areas [43]. Therefore, policymakers have continued to create and sustain programs to bolster the health center workforce, but there lacks coordination, evaluation, or publication of these efforts [44]. Two separate studies found benefit with Title VII programs on provider recruitment, with the Title VII funded physicians having 1.2 to 1.3 times the odds of working in CHCs or other medical underserved communities, compared to those without [28,29]. We revealed health centers with career ladder programs have 1.8 times the odds of reporting minimal difficulty in provider recruitment compared to those without these programs.

Central

Aysola et al. (2015)Email:

J Family Med Community Health 2(8): 1063 (2015) 5/6

Few studies have established promising provider retention enhancing strategies. Evaluations of the National Health Service Corps (NHSC), a federal program that provides scholarships and loan repayment incentives in exchange for service in shortage areas, reveal about 1 in 4 of every new primary care physician entering a health professional shortage area in the late 1980s [6], and that 1 in 5 physicians practicing in rural communities were brought there by NHSC [45]. However, in a 9 year cohort study Pathman et al determined that significantly fewer NHSC than non-NHSC physicians remained in their index practices, index communities, or rural practice settings after three years and every year thereafter [46]. In contrast, we found that health centers with career ladder programs compared to those without had 2.4 times the odds of reporting minimal difficulty retaining providers beyond three years.

We conducted a nationally representative survey identifying previously unknown effects and characteristics of health center professional training and pipeline programs. Despite these strengths, our study had some limitations. In this cross-sectional study, we could not make any causal inferences. Both key predictors and primary outcome variables were assessed in the same survey and may have been subject to selection bias and with our outcomes potentially recall bias; however, these variables could only be determined by health center reporting.

Our study reveals that career ladder programs, in particular, represent a promising model for developing the primary care workforce within health centers. These findings call for ongoing research in this area to advance, integrate, and expand programs that prove effective in addressing the mal-distribution of primary care providers nationwide.

ACKNOWLEDGEMENTSWe would like to thank Leonard Davis Institute of Health

Economics, University of Pennsylvania for support and Jianing Yang, BS for her assistance with statistical programming.

Conflict of Interest

Financial Disclosure: Jaya Aysola was funded by a grant (T32HP10251) from the Health Resources and Services Administration of the Department of Health and Human Services for the data collection portion of this study. Jaya Aysola was funded by the grant AHRQ’s PCOR Institutional Award (K12 HS021706-01) during the final data analysis and manuscript preparation for this study.

Conflicts of Interest Disclosure: The funders had no role in the design and conduct of the study; the collection, management, analysis, and interpretation of the data; or the preparation, review, or approval of the manuscript. Grant/Cooperative Agreement Number U30CS16089 from the Health Resources and Services Administration, Bureau of Primary Health Care (HRSA/BPHC), supported this publication. Its contents are solely the responsibility of the authors and do not necessarily represent the official view of HRSA/BPHC.

REFERENCES1. Bodenheimer T, Grumbach K, Berenson RA. A lifeline for primary care.

N Engl J Med. 2009; 360: 2693-2696.

2. Petterson SM, Liaw WR, Phillips RL Jr, Rabin DL, Meyers DS, Bazemore AW. Projecting US primary care physician workforce needs: 2010-2025. Ann Fam Med. 2012; 10: 503-509.

3. Schwartz MD. Health care reform and the primary care workforce bottleneck. J Gen Intern Med. 2012; 27: 469-472.

4. Dall TM, Gallo PD, Chakrabarti R, West T, Semilla AP, Storm MV. An aging population and growing disease burden will require a large and specialized health care workforce by 2025. Health Aff (Millwood). 2013; 32: 2013-2020.

5. Blumenthal D, Rizzo JA. Who cares for uninsured persons? A study of physicians and their patients who lack health insurance. Medical care. 1991; 29: 502-520.

6. Rosenblatt RA, Hart LG. Physicians and rural America. West J Med. 2000; 173: 348-351.

7. Starfield B, Shi L, Macinko J. Contribution of primary care to health systems and health. Milbank Q. 2005; 83: 457-502.

8. Macinko J, Starfield B, Shi L. Quantifying the health benefits of primary care physician supply in the United States. International journal of health services: planning, administration, evaluation. 2007; 37:111-126.

9. Chang CH, Stukel TA, Flood AB, Goodman DC. Primary care physician workforce and Medicare beneficiaries’ health outcomes. JAMA. 2011; 305: 2096-2104.

10. Baicker K, Chandra A. Medicare spending, the physician workforce, and beneficiaries’ quality of care. Health Aff (Millwood). 2004; 4: 184-97.

11. O’Malley AS, Forrest CB, Politzer RM, Wulu JT, Shi L. Health center trends, 1994-2001: what do they portend for the federal growth initiative? Health Aff (Millwood). 2005; 24: 465-472.

12. DeVries A, Li CH, Sridhar G, Hummel JR, Breidbart S, Barron JJ. Impact of medical homes on quality, healthcare utilization, and costs. Am J Manag Care. 2012; 18: 534-544.

13. Forrest CB, Whelan EM. Primary care safety-net delivery sites in the United States: A comparison of community health centers, hospital outpatient departments, and physicians’ offices. JAMA. 2000; 284: 2077-2083.

14. Blumenthal DS, Lukomnik JE, Hawkins DR Jr. A proposal to provide care to the uninsured through a network of community health centers. J Health Care Poor Underserved. 1993; 4: 272-279.

15. Rosenblatt RA, Andrilla CH, Curtin T, Hart LG. Shortages of medical personnel at community health centers: implications for planned expansion. JAMA. 2006; 295: 1042-1049.

16. Iglehart JK. Spreading the safety net--obstacles to the expansion of community health centers. N Engl J Med. 2008; 358: 1321-1323.

17. Morris CG, Johnson B, Kim S, Chen F. Training family physicians in community health centers: a health workforce solution. Fam Med. 2008; 40: 271-276.

18. Rieselbach RE, Crouse BJ, Frohna JG. Teaching primary care in community health centers: addressing the workforce crisis for the underserved. Ann Intern Med. 2010; 152: 118-122.

19. Savageau JA, Ferguson WJ, Bohlke JL, Cragin LJ, O’Connell E. Recruitment and retention of primary care physicians at community health centers: a survey of Massachusetts physicians. Journal of health care for the poor and underserved. 2011; 22: 817-835.

20. Gordon MS. Health education centers. Their role in medical education. JAMA. 1971; 218: 1192-1194.

21. Nottingham LD, Lewis MJ. AHEC in West Virginia: a case study. Area

Central

Aysola et al. (2015)Email:

J Family Med Community Health 2(8): 1063 (2015) 6/6

Aysola J, Groves DS, Hicks LS (2015) Health Center Professional Programs and Primary Care Workforce. J Family Med Community Health 2(8): 1063.

Cite this article

health education centers. The Journal of rural health: official journal of the American Rural Health Association and the National Rural. Health Care Association. 2003; 19: 42-46.

22. Rachel Morgan SHPS. Summary of the health workforce provisions in the patient protection and affordable care act: H.R. 3590 444 North Capitol Street, N.W., Suite 515 Washington, D.C. 20001 National Conference of State Legislatures. 2010.

23. Chen FM, Lesko SA, Gray B. Teaching health centers and the primary care workforce crisis for the underserved. Ann Intern Med. 2010; 152: 616.

24. Bacon TJ, Baden DJ, Coccodrilli LD. The national Area Health Education Center program and primary care residency training. J Rural Health. 2000; 16: 288-294

25. Gessert CE, Smith DR. The national AHEC program: review of its progress and considerations for the 1980s. Public Health Rep. 1981; 96: 116-120.

26. Gessert C, Jones C. Urban AHECs: a comparison with rural AHECs. Public Health Rep. 1986; 101: 637-643.

27. Trude L. Community Health Center Workforce Initiatives North Dakota: Health Workforce Information Center. 2010.

28. Rittenhouse DR, Fryer GE Jr, Phillips RL Jr, Miyoshi T, Nielsen C, Goodman DC, et al. Impact of Title VII training programs on community health center staffing and national health service corps participation. Ann Fam Med. 2008; 6: 397-405.

29. Krist AH, Johnson RE, Callahan D, Woolf SH, Marsland D. Title VII funding and physician practice in rural or low-income areas. J Rural Health. 2005; 21: 3-11.

30. Brooks RG, Mardon R, Clawson A. The rural physician workforce in Florida: a survey of US- and foreign-born primary care physicians. J Rural Health. 2003; 19: 484-491.

31. NHSC Clinician Retention Policy Brief. 2012.

32. Bureau of Primary Health Care H, U.S. Department of Health and Human Services. 2010 Uniform Data System Reporting Manual. 2010.

33. Williams R. Generalized ordered logit/partial proportional odds

models for ordinal dependent variables. Stata Journal. 2007; 6: 58-82.

34. Chen C, Chen F, Mullan F. Teaching health centers: a new paradigm in graduate medical education. Acad Med. 2012; 87: 1752-1756.

35. Rosenthal TC. Outcomes of rural training tracks: a review. J Rural Health. 2000; 16: 213-216.

36. Rich EC. Commentary: teaching health centers and the path to graduate medical education reform. Acad Med. 2012; 87: 1651-653.

37. Tarlov AR. Estimating physician workforce requirements. The devil is in the assumptions. JAMA. 1995; 274: 1558-1560.

38. Feil EC, Welch HG, Fisher ES. Why estimates of physician supply and requirements disagree. JAMA. 1993; 269: 2659-2663.

39. Cooper RA. Perspectives on the physician workforce to the year 2020. JAMA. 1995; 274: 1534-1543.

40. Politzer RM, Gamliel S, Cultice J, Sekscenski ES. Physician workforce projections: too many or just right? JAMA. 1996; 275: 685-686.

41. Chen PG, Mehrotra A, Auerbach DI. Do we really need more physicians? Responses to predicted primary care physician shortages. Med Care. 2014; 52: 95-96.

42. Phillips RL Jr, Bazemore AM, Peterson LE. Effectiveness over efficiency: underestimating the primary care physician shortage. Med Care. 2014; 52: 97-98.

43. Adashi EY, Geiger HJ, Fine MD. Health care reform and primary care--the growing importance of the community health center. N Engl J Med. 2010; 362: 2047-2050.

44. Pathman DE, Taylor DH Jr, Konrad TR, King TS, Harris T, Henderson TM, Bernstein JD. State scholarship, loan forgiveness, and related programs: the unheralded safety net. JAMA. 2000; 284: 2084-2092.

45. Cullen TJ, Hart LG, Whitcomb ME, Rosenblatt RA. The National Health Service Corps: rural physician service and retention. J Am Board Fam Pract. 1997; 10: 272-279.

46. Pathman DE, Konrad TR, Ricketts TC, 3rd. The comparative retention of National Health Service Corps and other rural physicians. Results of a 9-year follow-up study. JAMA. 1992; 268: 1552-1558.