Printed from:

Scope of Practice and Family Medicine Match Rates: Results From a CERA Clerkship Directors’ Survey

David Anthony, MD, MSc; Jordan White, MD, MPH; Katherine Margo, MD; Derjung M. Tarn, MD, PhD

Background and Objectives: Family medicine clerkship directors depend on community preceptors to teach and attract medical students to family medicine. Many community preceptors do not provide the full range of family medicine services, and some are not family physicians. This study aimed to determine the types of practices in which family medicine clerkship students train and whether scope of practice is associated with family medicine Match rates.

Methods: Data were collected as part of the 2014 Council of Academic Family Medicine Educational Research Alliance (CERA) Family Medicine Clerkship Director Survey. Clerkship directors estimated the percentage of their preceptor sites providing each of nine clinical services and the percentage of students placed with internal medicine physicians for their family medicine rotations. We devised a Scope of Practice Index (SPI) to assess scope of practice and measured the association between a clerkship’s SPI and family medicine Match rate.

Results: Limited scopes of practice were common. SPI was lowest in the Northeast and highest in the West. In bivariate and multivariable analyses, a lower SPI was associated with lower family medicine Match rates. Preventive gynecological care was the service most highly associated with family medicine Match rates. Family medicine Match rates were lower when programs used internal medicine sites for their family medicine rotations.

Conclusions: Many clerkship students are exposed to practices with limited scopes of family medicine practice, and this is associated with lower family medicine Match rates. These findings have implications for the specialty as preceptor scope of practice declines.

(Fam Med 2017;49(3):177-82.)

To promote a broad scope of practice for family physicians, the 2004 Future of Family Medicine Report proposed a model of practice offering a full “basket of services,” in which a comprehensive approach to care is provided.1 Comprehensiveness is one of the pillars of primary care, particularly of family medicine. However, recent data indicate that in many family medicine practices the basket is missing a few items.

A 2011 study found that among family physicians taking the Maintenance of Certification exam, fewer than 40% reported providing more than half of the clinical services considered on the survey.2 In particular, between 1995 and 2004, the percentage of all prenatal visits provided by family physicians dropped from 11.6% to 6.1%,3 and between 2003 and 2009 the percentage of family physicians providing women’s gender-specific health care needs declined.4

Reductions in the scope of family medicine have raised concerns about primary care physician training opportunities.5,6 In one recent study the majority of family medicine program directors surveyed reported using hospitalists (frequently internal medicine physicians) to run their inpatient teams.7 In addition, 64% of programs reported that less than half of their graduates provided inpatient care. This suggests that there are fewer family physicians to model the provision of inpatient care for family medicine residents and medical students. Changes in the scope of family medicine may most profoundly influence medical students who are making career choices. Given these declines, many students likely will be trained in clerkship settings with limited scopes of practice. Family medicine clerkship directors have reported difficulty finding clinical training sites for their medical students due to increases in medical school class sizes and competition from other types of learners such as physician assistants.8 Under pressure, some clerkship directors place students at sites that have even more constrained scopes of practice (eg, internal medicine sites). Studies have not examined the influence of changes in preceptor scope of practice on students’ decisions to pursue careers in family medicine.

This study examines the influence of scopes of practice on medical student career decisions by determining: (1) family medicine clerkship directors’ assessments of the current scopes of practice at their clinical training sites, (2) how often family medicine clerkship directors place students at internal medicine sites, and (3) the relationship of medical students’ exposure to scopes of practice on family medicine Match rates.





This study analyzes data obtained as part of the 2014 Council of Academic Family Medicine Educational Research Alliance (CERA) Family Medicine Clerkship Director Survey. The study was approved by the American Academy of Family Physicians Institutional Review Board.

Survey Administration and Development

The CERA Family Medicine Clerkship Director Survey is distributed annually to the clerkship director (or his/her designate) at the main campus of qualifying medical schools. Qualifying medical schools are accredited by the Liaison Committee on Medical Education or the Committee on Accreditation of Canadian Medical Schools and are located within the United States, the Commonwealth of Puerto Rico, or Canada. Schools must have students who complete a family medicine clerkship or a primary care clerkship with a required family medicine component that is directed by a family medicine educator. CERA identified 121 US and 16 Canadian family medicine educators who met these criteria. As the US and Canadian medical education systems are quite different and have considerably different family medicine Match rates (9.5% and 42.8%, respectively) we restricted all analyses to US clerkship directors.

The survey was conducted during the fall of 2014 using SurveyMonkey.® Three email invitations were sent, each including a personalized greeting, a letter urging participation, and a survey. Nonrespondents were contacted to encourage participation.

Investigators submitted survey questions for inclusion in the CERA survey. Questions were pretested with family medicine educators and modified for flow, timing, and readability.

Survey Items

Respondents were asked about their school location (US Census region) and whether the school is public or private. Clerkship demographics included the clerkship design (block, longitudinal, or both), clerkship duration (in weeks), and payment for community preceptors (yes/no). Community preceptors were defined as “teachers who practice off campus and who do not have a primary appointment in your department or institution.”

Respondents estimated the percentage of their family medicine clerkship preceptor sites providing each of nine different clinical services (scope of practice assessment). These services were selected to match the services that family physicians are asked about when taking the American Board of Family Medicine Maintenance of Certification examination,2 though some items were removed or combined due to CERA survey question number limits. We asked for the percentage of sites with National Committee for Quality Assurance (NCQA) rating as patient-centered medical homes (PCMH) since this growing and promoted model of care may influence students’ decisions to pursue family medicine. They also were asked to estimate the percentage of their clerkship students who are placed in internal medicine practices and to report the percentage of their 2013 medical school graduating class who matched in family medicine.


All analyses were performed using SPSS version 21 and Stata/SE version 14.1. We performed descriptive statistics to examine characteristics of the US medical schools and clerkships and conducted bivariate analyses to assess the relationships between: (1) the percentage of sites providing each of nine clinical services and family medicine Match rates and (2) the percentage of students placed at internal medicine practices and family medicine Match rates.

Scope of Practice Index

To analyze the effects of exposing students to varying scopes of family medicine practice, we developed a measure called the Scope of Practice Index (SPI). The SPI is an ordinal value ranging from 0–10, with greater numbers indicating broader scopes of practice. Schools start with a score of 10, and points are subtracted: 1 point for each clinical service for which the percentage of sites providing that service was below the median value and 1 point if the school used internal medicine practices for their family medicine clerkship sites.

Relationship Between SPI and Family Medicine Match Rates

We used the Pearson correlation coefficient to examine the unadjusted relationship between the SPI and family medicine Match rates. We also used one-way ANOVAs to examine the SPI and family medicine Match rates stratified across the four US Census regions (Northeast, South, Midwest, and West). Multiple linear regression was used to investigate the association between SPI and family medicine Match rates, adjusting for the duration of the family medicine clerkship (in weeks) and the census region, both of which were associated with Match rates.

We ran two sensitivity analyses to determine whether different ways of calculating the SPI influenced our outcomes. First, the SPI was calculated without consideration of using internal medicine practices as training sites (so that the SPI ranged from 0–9). It was also calculated by subtracting 1 point for each clinical service for which fewer than 50% of its sites provided the service.



The overall survey response rate was 90.9%, with 110 out of 121 US clerkship directors responding.

Two thirds of the schools were public. Eighty percent had a block clerkship, with 5.5% having longitudinal clerkships and 14.5% having a combination of the two. The average duration of the family medicine clerkship, combining block and longitudinal clerkships, was 5.3 weeks (SD=1.5 weeks). About 20% of respondents reported paying the majority of their community preceptors to teach. Among those who reported paying, the average payment was $202.20 per week (SD=$147.80). The overall family medicine Match rate was 9.5% (SD=5.2%) (Table 1).



Respondents estimated that community preceptors fail to provide a large number of clinical services (Figure 1). The majority of community preceptors provided preventive gynecologic care (82.9%), care of children under 13 years of age (72.2%), and dermatology and orthopedic procedures (65.7%). They least commonly performed pediatric inpatient care (21.1%), vaginal deliveries (23.3%), and prenatal care (31.0%). In general, schools in the West provided the most services, and schools in the Northeast provided the least (service-specific data not shown). With the exception of adult inpatient care, the rate with which each of the clinical services was provided was weakly to moderately correlated with the family medicine Match rate. Preventive gynecologic care was the most strongly correlated (correlation coefficient 0.39, P value <.001). The percentage of students placed at internal medicine practices was negatively associated with the family medicine Match rate (correlation coefficient -0.34, P value=.006).




Bivariate analyses indicate that higher SPI (ie, broader scope of practice) is moderately correlated with Match rates (Pearson correlation coefficient=0.37, P value<.001) (Figure 2). Both the SPI and the family medicine Match rate varied by geographic region (Figure 3). The SPI was lowest in the Northeast and highest in the West. Similarly, the family medicine Match rate was lowest in the Northeast and highest in the West.



A multivariable model showed that the family medicine Match rate increases by approximately 0.5% with each 1-point increase in the SPI (95% CI=0.2 to 0.8). In addition, programs in the West had 5.9% higher Match rates than those in the Northeast (95% CI=2.7 to 9.2). Each 1-week increase in clerkship duration was associated with a 1.2% increase in the family medicine Match rate (95% CI=0.5 to 1.8). Overall, this model predicted 33% of the variability in Match rates. Sensitivity analyses using two different methods for calculating the SPI did not alter the overall results.





This study demonstrates that higher family medicine Match rates are associated with the use of family medicine clerkship preceptors with broader scopes of practice and with avoidance of the use of internal medicine sites for training. This relationship holds when adjusting for geographic region and clerkship duration, both of which have been shown in previous studies to influence family medicine Match rates.9,10 Many other factors (eg, personality type, the presence of a family medicine department, etc) have been shown to affect medical students’ choices to pursue careers in family medicine,11,12 but this is the first study to evaluate the relationship of family medicine preceptor scope of practice on family medicine Match rates.

We hypothesize that scope of practice influences medical students’ career choices in several ways. Exposure to the full scope of family medicine may have a powerful impact on a student’s view of the specialty. It may confirm the broad competencies of family physicians among students interested in primary care and help distinguish family medicine from other specialties.13,14 Comprehensiveness is clearly valued among people choosing careers in family medicine, as demonstrated by a recent study in which initial family medicine board certifiers report intending to practice a broader scope of practice than that reported by family physicians taking their recertification exam.15

Our findings are concerning because they indicate that many US family medicine clerkship students are exposed to limited scopes of family medicine practice, which may negatively influence students’ interest in the specialty. These data reflect national trends in which family physicians are increasingly limiting their scope of practice.2-4 Family physicians are needed to fill gaps in the US primary care physician workforce,16 and broader scope of practice among family physicians has been associated with beneficial patient outcomes.17 Exposure of clerkship students to the full scope of practice may therefore be difficult to implement given the overall decrease in family physicians’ scope of practice and increasing competition for community preceptors.8

Our findings suggest some possible solutions. Family medicine Match rates were lower when medical schools used internal medicine preceptors. This finding is consistent with a prior study showing that among schools with a primary care clerkship, those in which more than 30% of students were placed with family physicians had higher family medicine Match rates.18 Clerkship directors may place students with general internists because outside pressures have made it difficult to find family physicians to teach their students.8 Innovative solutions to recruit family medicine preceptors and decrease the number of internal medicine preceptors are needed.19 Paying preceptors for their time20 and more vigorous recruiting through state and national organizations may help. In our study, the provision of gynecologic procedures demonstrated the strongest correlation with family medicine Match rates. As such, family medicine clerkship directors may benefit from increasing medical students’ exposure to gynecologic procedures, either through didactic sessions or clinical exposures. Since internists are less likely than family physicians to comprehensively manage reproductive-aged women’s health care needs,21 this finding may partly explain the negative impact of placing students with general internists on family medicine Match rates. Lastly, in the present study the provision of adult inpatient medicine services did not correlate with Match rates, despite national concern about the decline in family physicians providing this service.22,23 This observation corroborates data showing declines in family medicine residency graduates’ intentions to provide inpatient services.7 More research is needed to determine if displaying inpatient medicine skills during family medicine clerkships is important in recruiting students to the specialty.

Our study has several limitations. Data on the services provided by community preceptors are based on clerkship director estimates and may be inaccurate. It is possible that awareness of the family medicine Match rate may alter a clerkship director’s assessment of the provision of these services. Since our study is based on cross-sectional data, we cannot show causality: other factors associated with scope of practice (eg, rural location) may account for these findings. Additionally, the SPI is an unvalidated measure. We equally weighted the clinical services used to determine the SPI, and four of the nine services related to obstetrics and gynecology. Our sensitivity analyses of alternate scope of practice calculations did not vary the results, leading to increased confidence in the utility of this novel measure, but further research could investigate the optimal components of the index.

In conclusion, US medical students, particularly in the Northeast, are doing their family medicine clerkships in sites with limited scopes of practice, when exposure to broader scopes of practice appears to promote family medicine as a career choice. Clerkship directors, faced with challenges in finding broad-spectrum family medicine sites, should consider strategies to help students appreciate the full scope of family medicine despite not experiencing this clinically. Similarly, methods to recruit and support community preceptors who provide a broad scope of practice are needed. Such strategies may be essential to help the United States meet its primary care demands.

Corresponding Author: Address correspondence to Dr Anthony, Brown University, Warren Alpert School of Medicine, 111 Brewster Street, Pawtucket, RI 02860. 401-729-2308. Fax: 401-729-2729.

Acknowledgments: This paper was presented at the 2015 Society of Teachers of Family Medicine Conference on Medical Student Education, Atlanta, GA.



  1. Martin JC, Avant RF, Bowman MA, et al. The Future of Family Medicine: a collaborative project of the family medicine community. Ann Fam Med 2004;2(Suppl 1):S3-S32.
  2. Bazemore AW, Petterson S, Johnson N, et al. What services do family physicians provide in a time of primary care transition? J Am Board Fam Med 2011;24(6):635-6.
  3. Cohen D, Coco A. Declining trends in the provision of prenatal care visits by family physicians. Ann Fam Med 2009;7(2):128-33.
  4. Xierali IM, Puffer JC, Tong STC. The percentage of family physicians attending to women’s gender-specific health needs is declining. J Am Board Fam Med 2012;25(4):406-7.
  5. Newton WP. Family physician scope of practice: what it is and why it matters. J Am Board Fam Med 2011;24(6):633-4.
  6. Green LA, Dodoo MS, Ruddy G, et al. The physician workforce in the United States: a family medicine perspective. 2004. Accessed March 31, 2016.
  7. Baldor R, Savageau JA, Shokar N, et al. Hospitalist involvement in family medicine residency training: a CERA study. Fam Med 2014;46(2):88-93.
  8. Anthony D, Jerpbak CM, Margo KL, et al. Do we pay our community preceptors? Results from a CERA clerkship directors’ survey. Fam Med 2014;46(3):167-72.
  9. Kozakowski SM, Crosley PW, Bentley A. Entry of US medical school graduates into family medicine residencies: 2012-2014. Fam Med 2014;46(9):696-700.
  10. Senf JH, Campos-Outcalt D, Kutob R. Factors related to the choice of family medicine: a reassessment and literature review. J Am Board Fam Pract 2003;16(6):502-12.
  11. Avery DM Jr, Wheat JR, McKnight JT, Leeper JD. Factors associated with choosing family medicine as a career specialty: what can we use? Am J Clin Med 2009; 6(4): 55-58.
  12. Avery DM Jr, Wheat JR, Leeper JD, et al. Admission factors predicting family medicine specialty choice: a literature review and exploratory study among students in the rural medical scholars program. J Rural Health 2012;28:128-36.
  13. Bennett KL, Phillips JP. Finding, recruiting, and sustaining the future primary care physician workforce: a new theoretical model of specialty choice process. Acad Med 2010 Supplment;85(10):S81-S88.
  14. Campos-Outcalt D, Senf J, Kutob R. A comparison of primary care graduates from schools with increasing production of family physicians to those from schools with decreasing production. Fam Med 2004;36(4):260-4.
  15. Coutinho AJ, Cochrane A, Stelter K, et al. Comparison of intended scope of practice for family medicine residents with reported scope of practice among practicing family physicians. JAMA 2015;314(22):2364-72.
  16. Petterson SM, Liaw WR, Phillips RL Jr, et al. Projecting US primary care physician workforce needs: 2010-2025. Ann Fam Med 2012;10:503-9.
  17. Bazemore A, Peterson S, Peterson SE, Phillips RL Jr. More comprehensive care among family physicians is associated with lower costs and fewer hospitalizations. Ann Fam Med 2015;13:206-13.
  18. Stine CC, Sheets KJ, Calonge BN. Association between clinical experiences in family practice or in primary care and the percentage of graduates entering family practice residencies. Acad Med 1992;67(7):475-7.
  19. Christner JG, Dallighan GB, Briscoe G, et al. The community preceptor crisis: recruiting and retaining community-based faculty to teach medical students—a shared perspective from the Alliance for Clinical Education. Teach Learn Med. Published online April 19, 2016. Accessed May 4, 2016.
  20. Peters AS, Schnaidt KN, Zivin K, et al. How important is money as a reward for teaching? Acad Med 2009;1:42-6.
  21. Petterson SM, Bazemore AW, Phillips RL, Rayburn WF. Trends in office-based care for reproductive-aged women according to physician specialty: a ten-year study. J Womens Health (Larchmt) 2014 Dec;23(12):1021-6.
  22. Association of Family Medicine Residency Directors. Is the family physician in or out of hospital medicine? A discussion of pertinent perspectives to consider as we address inpatient curricular review. Ann Fam Med 2009;7(5):471-2.
  23. Tiemstra J. Inpatient care is an important part of family medicine. J Am Board Fam Med 2007;20(6):612.

From the Warren Alpert Medical School, Brown University (Drs Anthony and White); the Perelman School of Medicine, University of Pennsylvania (Dr Margo); and David Geffen School of Medicine, University of California, Los Angeles (Dr Tarn).

Copyright 2018 by Society of Teachers of Family Medicine