A Maternal and Child Health Curriculum for Family Practice Residents: Results of an Intervention at the University of North Carolina

174 March 2003 Family Medicine A Maternal and Child Health Curriculum for Family Practice Residents: Results of an Intervention at the University of North Carolina Margaret Helton, MD; Bron Skinner, PhD; Clark Denniston, MD Background and Objectives: There has been a significant reduction in the number of family physicians who provide pregnancy care. This study examines the effects of an educational intervention at a university-based family practice residency. The intervention was designed to increase the number of graduates who include prenatal care or deliveries in their practices. Methods: The curriculum in pregnancy care was expanded to include a teaching service, better role modeling by family medicine faculty, more deliveries, stronger didactics, breast-feeding and child health services, and greater collaboration with other health care settings and professionals. Results: The intervention increased the average percentage of residents who included prenatal care or deliveries in their practices after graduation from 27.5% to 52%. Conclusions: This educational intervention increased the number of family practice residency graduates who plan to include prenatal care or deliveries in their practices. (Fam Med 2003;35(3):174-80.) The care of women and children is an integral component of family medicine, and family physicians play an important role in providing this care for the medically underserved, especially in rural areas. Nationwide, the percentage of family physicians who provide pregnancy care dropped from 43% in 1986 to 22.4% in 2000. 1 This trend is disturbing, since a lack of adequate pregnancy care is associated with poor pregnancy outcomes. 2 In North Carolina, the percentage of family physicians who deliver babies declined from 15.2% in 1994 to 11.8% in 1998. 3 This is particularly troubling because North Carolina s infant mortality rate is 9.2 deaths per 1,000 live births (US range=4.3 10.6), ranking among the highest in the United States. 4 Between 1988 and 1994, only 11 of 40 (27.5%) family practice residents from the University of North Carolina provided prenatal care or performed deliveries after graduation from the residency program. During that era, all obstetrical teaching in the formative first and second years of residency was by obstetricians, with From the Department of Family Medicine, University of North Carolina. third-year family practice residents (PGY-3) on call for 2 months for family practice center deliveries. Some family medicine faculty who supervised the PGY-3 s deliveries were not particularly interested in pregnancy care, so teaching was inconsistent, the number of deliveries in the practice was falling, and few graduates were including prenatal care or deliveries in their practices. Various factors such as malpractice insurance costs, lifestyle issues, and features of rural practices and communities also likely influenced residents choices on this matter. These complex issues are not always controllable, but residency training is a critical time in determining whether or not a resident will be favorably inclined to pregnancy care. There is evidence that family practice training programs may be able to increase the number of graduates who provide pregnancy care by changing the residency curriculum. 6,7 Taylor and Hansen identified several characteristics of residency training programs in which the pregnancy care component of training was perceived to have been successful. 8 Our paper describes changes we made in the pregnancy care training our residents received at a university-based program. Many of these changes echo the findings reported by Taylor and Hansen. We also report the outcome of efforts to

Residency Education increase family physician-led deliveries, residents evaluations of the curriculum, and percentage of residents visits that involved children, as well as the outcome of measuring our residents plans to provide pregnancy care after graduation. Methods Planning for a restructuring of our pregnancy care curriculum began in 1991. A planning group prepared a proposal and presented it for faculty approval. Several broad principles guided our efforts (Table 1). The specific interventions are summarized in Table 2, matched with the essential characteristics identified by Taylor and Hansen. 8 Table 1 Vol. 35, No. 3 Principles That Guided the Development of a Pregnancy Care Curriculum Guiding Principles Establish family practice models of pregnancy care. Implement the program sequentially in stages. Make content evidence based. Broaden the service base to increase volume. Cultivate support from the hospital and obstetrics department. Enhance the child health component of the curriculum. 175 Role Models An important characteristic of a successful pregnancy care training program is role modeling by the family medicine faculty. 8-10 Faculty who were no longer interested in deliveries were allowed to bow out, leaving several committed faculty to provide coverage. We created a new service, called the maternal and child health (MCH) service, which operated independently of our inpatient service. Though we chose to continue the PGY-1 rotation on the obstetrics department s service, the new MCH service ensured that, during the last 2 years of residency, residents would rotate on our own MCH service with family medicine faculty as supervisors. We called the new service the maternal and child health service, rather than the family practice obstetrics service, to better reflect the family medicine principle that the care of the family from pregnancy to child care is a natural continuum. We also felt it was important not to attempt to be mini-obstetricians. Family physicians performing deliveries are likely to enjoy it more when they operate in a paradigm of care they have established for themselves. 11 Essential Characteristics for Successful Obstetrics Training Interventions for the Maternal and Child Health Curriculum at the in Family Practice Residencies (Taylor and Hansen, 1997 8 ) University of North Carolina, 1995 2001 Family medicine faculty model behaviors by providing maternity care All PGY-2 and PGY-3 deliveries are supervised by family medicine faculty Family medicine faculty are competent at maternity care Only faculty who are committed to maternity care, and who practice it themselves, are supervising residents Positive obstetrical learning experiences Faculty display their support and enthusiasm for maternity care Family practice residents are given real responsibility Created our own MCH service, with only family practice residents serving as the caregivers Mutual respect between OB and family medicine faculty and residents PGY-1s continue to rotate on the OB service Adequate OB volume to ensure competence Increased volume of MCH service deliveries Emphasis on the longitudinal experience of maternity care Continuity care throughout PGY-2 and PGY-3 year, emphasis on child health care after delivery Curriculum encourages residents to do maternity care Curriculum exceeds RRC requirement, ample didactic teaching Hospital administration supports family physicians doing maternity care MCH service is busier, creating more of a presence on L&D Family medicine department credentials family physicians doing Our department credentials itself regarding maternity care, maternity care eliminated list of required consultations Family physicians readily accepted in the community as maternity care Positive local reputation due to support of midwifery and familyproviders centered care philosophy Community family physicians provide maternity care Presence of our MCH service allowed local physicians to provide prenatal care MCH maternal and child health service OB obstetrics RRC Residency Review Committee for Family Practice L&D labor and delivery Table 2 Comparison of Maternal and Child Health Interventions at the University of North Carolina With Essential Characteristics of Successful Obstetrics Training (Taylor and Hansen 8 )

176 March 2003 Family Medicine Sequential, Staged Implementation For the first year (1994 1995) of the MCH service, family medicine faculty provided coverage and performed all deliveries by themselves. The first residents were introduced in 1995 1996; these were PGY-3s who took call for a month at a time manageable because the volume was low. A second resident (PGY-2) on every-other-night call was added to the PGY-3 on the service in 1996 1997. In 2000, as volume increased to more than 300 deliveries per year, we added a third resident (PGY-2 or PGY-3) to reduce call requirements to every third night. Our new curriculum exceeds the minimum of 3 months of pregnancy care training mandated by the Residency Review Committee for Family Practice (RRC) by 6 weeks. 12 The continuity experience occurs during the PGY-2 and PGY-3 years, with patients assigned to residents who will be on the MCH service at the time of the estimated date of delivery. Evidenced-based Didactics Residents and faculty present interesting cases involving pregnancy or newborn issues, with complementary didactic teaching, at a conference held every other Friday. Our residents and faculty also take and pass the Advanced Life Support in Obstetrics (ALSO) course, taught yearly as part of our MCH program since 1995. Participants in ALSO report feeling greater comfort with delivery room emergencies and believe that by attending the course, they are more likely to perform deliveries in their future practices. 13 Based on a similar rationale, the Neonatal Resuscitation Program (NRP) has been in the curriculum since 1994. Collaboration With Other Practice Sites to Broaden the Experience and Increase Volume Residents need a high volume of deliveries to optimize the probability of exposure to problems that arise in the delivery room. Taylor and Hansen suggest that the ideal training volume is 100 deliveries during the course of a resident s training, of which 20 to 30 are continuity deliveries. 8 As noted, we continued the intern rotation on the obstetrics department s service, thus ensuring a high volume of deliveries for the PGY-1 year. Then we took steps to expand the volume on our own MCH service. To do this, we first made the family practice center more attractive to pregnant patients by hiring a prenatal clinic nurse coordinator. This individual ensures that prenatal patients are seen in a timely manner, provides patient education, organizes records, and assigns new patients to a continuity provider. The cost of the prenatal coordinator was covered by revenue from the increased number of deliveries. We also hired a maternity care social worker, who is stationed in the family practice center but whose salary is covered by the county health department. Second, we expanded our relationship with our county health department to deliver their prenatal patients. The MCH residents provide prenatal and pediatric care at the health department under the supervision of our faculty physicians. Our experience confirms reports that linking a family practice residency with a health department offers a unique educational setting, direct service to medically indigent patients, and shared financing. 14 This model has been successful enough that we recently added another health department clinic from a neighboring county to our MCH program. Third, we began providing deliveries for the patients of several local community family physicians, an arrangement that enabled these physicians to provide prenatal care. Several of these community family physicians also joined our MCH faculty call schedule. Last, we hired a midwife faculty member for the family practice center and also began providing consultation services and medical backup to a local freestanding birthing center that is run by a group of midwives. Such collaboration between physicians and midwives has established safety and acceptability 15 and can improve access to maternity care in rural areas. 16 In our case, it enhanced our position as an important, trustworthy provider in the local birthing scene, which is believed to be a key characteristic for a successful training program. 8 Developing Support From the Hospital and the Obstetrics Department Attitudes of obstetricians about the provision of maternity care by family physicians vary widely and can sometimes be negative. Family physicians who feel well supported by obstetricians during their training are more likely to develop positive attitudes toward obstetric practice. 17 We have tried to maintain good relationships with the obstetricians, and the establishment of our MCH service has given us more legitimacy in the labor and delivery suite. Our department is now represented on hospital committees, including committees involved with maternity care, newborn screening, and medical informatics. We chose, as mentioned, to c ontinue the PGY-1 r ota tion on the obste trics department s service. This is consistently a positive experience, and the family practice and obstetrical residents form collegial relationships that they carry with them throughout training. Family medicine departments that credential their own physicians for delivery room privileges are more likely to have strong training programs. 8 In 2001, we updated the credentialing documents, eliminating an addendum listing obstetrical consultation guidelines, a move supported by the hospital credentialing committee. We felt this was a relic from a time when the family medicine department was not empowered to practice independently. Further, the list was an ongoing

Residency Education Vol. 35, No. 3 177 source of contention between the family medicine and the obstetrics departments and was subject to nuances in interpretation about the need for consultation and other concerns. We believed instead that fostering collegial relationships with the obstetricians is the best way to ensure that timely and appropriate consultation occurs. Putting the Child in Maternal Child Health Our MCH service sought not just to strengthen training in delivery room skills but also in breast-feeding and child care. Enhancing this portion of the curriculum can overcome the low volume of children that plagues the practice demographics of many family practice training programs. To make our practice more child friendly, we created a maternal and child health same-day clinic for children with acute medical needs. The residents on the MCH rotation staff this clinic. To promote breast-feeding, we supported our prenatal nurse coordinator in becoming a certified lactation consultant. This not only provided support to new mothers but also developed the nurse into someone who taught the residents about breast-feeding. Finally, we added an adolescent health clinic experience in which residents spend a day and a half a week at a local teen clinic. Outcomes Assessment We counted the number of family physician-led deliveries in our program using a hospital perinatal database confirmed by billing data, both of which are collected independent of our department. We determined the number of residents who planned to include pregnancy care at their first practice after graduation based on their hospital credentialing requests. To assess the quality of the learning experience, we analyzed residents evaluations before 1995, when their teaching was primarily with the obstetricians and then after 1995, when their training was on our MCH service. A standard service evaluation form provided the feedback data from residents. These forms consist of an overall assessment scale and 22 items with five-point Likert scales to assess the quality of the experience from three perspectives: work experience (five items), educational experience (nine items), and affective experience (eight items). Items are clustered to produce summary indexes in each area of experience. Room is provided for comments. Data Analysis The significance of a difference between two independent proportions 18 was used to evaluate the percentage of residents, before and after the implementation of the new curriculum, who practiced prenatal and/or delivery care in the first practice situation they enter after graduation from the program. The same test was used to evaluate the percentage of children seen in the family practice center. A test of difference between two means for independent samples 18 was used to evaluate residents evaluations for the overall, educational, and affective indexes. Residents comments from service evaluations were subjected to a qualitative analysis, looking for themes and reports of incidents indicative of the quality of the residents experiences on the obstetrics and MCH rotations. Results The number of deliveries on the family physicianled MCH service has increased from a low of between 100 150 to more than 300 per year between 1988 and 2000 (Figure 1). The number of residency graduates including any pregnancy care in their first practice after graduation went from an average of 27.5% in the years 1988 1994 to an average of 52% in the years 1995 2001 (Table 3). This is a significant increase (P<.05) between the proportions in the two time intervals. Over the years of the study, residents completion rate for service evaluation forms was approximately 52% per year (range 32% 65%). The results listed in Table 4 provide the averages for summary indexes from residents evaluations: the overall rating, the index of the rotations as work experiences, the index of the experience educationally, and the affective index. For comparison, the average ratings on the PGY-1 obstetrics service, where our residents continued to train throughout the study period, have been reported separately. There has been little change in the PGY-1 ratings in the obstetrics department, except that, judging from the improved work and affective index, residents appeared more satisfied with the experience. All differences, except that of the overall rating, between the obstetrics services prior to 1995 and our new MCH service were significant (P<.01), with the MCH service more enjoyable and rated higher as an educational experience. MCH average overall assessment, however, is somewhat lower than that for the PGY-1 obstetrics experience. Residents exposure to children in the family practice center has increased. In 1997 1998, visits by patients under age 18 comprised 15.4% of total resident office visits. With our MCH curriculum intervention in full effect between 1997 and 2001, such visits comprised 19.7% of all resident office visits in 2000 2001, a significant difference of proportions (P<.001). Discussion We found that our interventions to strengthen our MCH program were successful in increasing the number of residents who include pregnancy care in their first practice after graduation. There are many factors that influence a resident s decision on providing pregnancy care, but we know of no significant confounding

178 March 2003 Family Medicine changes, such as malpractice rate increases or changes in reimbursement, during the time period we studied. Indeed, the percentage of our residents doing deliveries after graduation increased, even as the percentage of family physicians doing deliveries declined on a national level. Our program innovations have likely attracted tra inees with an interest in maternal and child health care, including an intention to perform deliveries after gradua tion, as training programs attract residents who are oriented to the same kinds of issues that turn on the faculty. 19 We have been delighted to observe that some residents who had no intention to perform deliveries change their minds and do so, apparently as a direct result of their training in our program. Analysis of the evaluation comments from residents reveals that the PGY-1 obstetrics rotation, which takes place at a community hospital that was used throughout the study period for training, was considered an excellent rotation overall. Residents worked hard, but learned a lot, and generally felt fairly treated. This remains a popular experience, especially since recent years have seen a reduction in the number of hours worked. Comments about the PGY-2 site in the University Hospital s obstetrics department prior to 1995 were less positive. While many felt they had learned useful material, the comments clearly point to the lack of fit because of the high-risk obstetrics and lack of integration into the obstetric team. Residents felt they learned how to refer to the obstetrician rather than how to provide pregnancy care themselves. Comments about our new MCH experience indicate that it was well liked because of the opportunity to learn continuity delivery care through one-on-one teaching by family practice attendings. Because our residents are delivering for patients they know through continuity prenatal visits, they gain a sense of responsibility and autonomy that they did not have on the obstetrics service. Greater autonomy has been shown to lead to higher levels of satisfaction, 20 Figure 1 Number of Deliveries Performed by the University of North Carolina Family Medicine Department 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 1992 1993 1994 1995 1996 1997 1998 1999 2000 Academic Years Data from 1988 1990 are by calendar year and thereafter by academic year. and that appears to be borne out in our residents evaluations in which the affective component of the experience is so highly rated. Table 3 Number of Residents Who Included Pregnancy Care in Practice Immediately After Graduation # Providing # Providing Total # Providing Year of # in Prenatal Prenatal and Prenatal Care and/or Graduation Class Care Only Delivery Care Delivery Services 1988 5 1 1 2 (40%) 1989 6 1 1 2 (33%) 1990 7 1 2 3 (43%) 1991 4 1 1 2 (50%) 1992 6 0 0 0 (0%) 1993 5 0 0 0 (0%) 1994 7 1 1 2 (29%) 1995 6 2 2 4 (67%) 1996 6 0 3 3 (50%) 1997 8 1 3 4 (50%) 1998 8 1 3 4 (50%) 1999 8 0 5 5 (62%) 2000 8 0 4 4 (50%) 2001 8 0 3 3 (38%)

Residency Education Vol. 35, No. 3 179 Table 4 Averages From Residents Service Evaluations* 1988 1994 1995 2000 PGY-2 PGY-1 PGY-2,3 PGY-1 OB OB MCH OB Averages of Evaluations Rotation Site Rotation** Site Overall rating 3.8 4.1 4.0 4.2 Work experience index 3.7 3.6 4.2 3.9 Educational index 3.5 3.7 4.0 3.7 Affective index 2.9 3.1 3.9 3.3 OB obstetrics MCH Maternal and Child Health service * Based on 5-point Likert scale (5=outstanding, 4=very good, 3=average, 2=below average, 1=unacceptable) **All differences between MCH and obstetrics service averages were significant at P<.01 except the ones with overall. We now have the best of both worlds: a strong PGY- 1 experience in obstetrics in which residents perform enough deliveries to gain fundamental technical skills and a family practice experience that provides residents with the opportunity to learn a model of pregnancy care that can work in a future community practice. Based on the increased number of residents who choose to perform deliveries in their practices, and judging by their evaluations, we conclude that it is the quality of the experience, rather than the quantity of the deliveries, that shapes their decisions regarding this aspect of their practice. Though the primary intent of our MCH program is to train family physicians to provide health care services that can improve outcomes in communities, there are new considerations that make it important to have a vigorous MCH program. The number of medical students choosing to enter family practice has persistently declined since 1997. 21 It may be that our discipline needs to reclaim clinical generalism, with teachers of family medicine modeling full-spectrum care. 22 Limitations of the Study This curriculum has worked well in our setting but its generalizability elsewhere will depend on many things. It takes committed attendings and the willingness to devote a significant proportion of the residency curriculum to it. In the face of many other demands for residency education, many may not wish to make the necessary tradeoffs. The politics of the institution must also be favorable to allow a program to establish an autonomous family practice pregnancy care service. Our outcome measures include the intent of graduates to include obstetrics in their practices after graduation from residency. Frequently, practice situations change during the years following graduation, so our estimates of the proportions of graduates doing obstetrics may not reflect the long term. Many will drop this aspect of their practice with time. We also know, however, that some of our graduates who first did not include pregnancy care when they graduated have since added it and reported comfort doing so because of the training they had here. Completion rates for the service evaluation forms were not as high as we might like. It is possible that those who chose not to complete forms somehow represent a population that would respond differently and affect the service evaluations. There is also the possibility that feedback by our residents would be positively biased in an effort to please the family practice faculty. However, we also have an inpatient service, which has been plagued with many difficulties over the years, and residents have never held back from giving negative critique about that service where they felt it was appropriate. Finally, none of our outcome measures has dealt with the question of competency of the residents. The usual methods for establishing resident competence have been applied during the implementation of the MCH curriculum, and we have no measurable differences that we are able to report from observation. Conclusions Family practice training programs can increase the number of graduates practicing pregnancy care by improving the curriculum. Any program seeking to improve this aspect of its residency training will need faculty who champion the effort. We believe that if we can make such a program thrive in our academic hospital environment, it can work in other settings as well. Acknowledgments: This curriculum intervention was supported in part by HRSA Grant #2 D15 PE 14206-18, Graduate Training in Family Medicine. A portion of the content of this manuscript was presented as a poster at the Society of Teachers of Family Medicine 2001 Annual Spring Conference in Denver. Corresponding Author: Address correspondence to Dr Helton, University of North Carolina, Department of Family Medicine, CB # 7595, Chapel Hill, NC 27599-7595. 919-966-3711. Fax: 919-966-6125. margaret_helton@ med.unc.edu. REFERENCES 1. American Academy of Family Physicians, Practice Profile I Survey, May 2000. Facts About Family Practice. www.aafp.org/facts/x829.xml. 2. Nesbitt TS, Larson EH, Rosenblatt RA, Hart LG. Access to maternity care in rural Washington: its effect on neonatal outcomes and resource use. Am J Pub Health 1997;87(1):85-90. 3. Cecil G. Sheps Center for Health Services Research, The University of North Carolina at Chapel Hill. Trends in the supply and distribution of obstetric care in North Carolina, 1998. Findings Brief 2001. 4. National Center for Health Statistics. Deaths: final data for 1997. Table 31. Natl Vital Stat Rep 1999;47(19),June 30.

180 March 2003 Family Medicine 5. Pathman D, Tropman S. Obstetrical practice among new rural family physicians. J Fam Pract 1995;40:457-64. 6. Nesbitt TS, Davidson RC, Paliescheskey MR, Fox-Garcia J, Arevalo JA. Trends in maternity care by graduates and the effect of an intervention. Fam Med 1994;26(3):149-53. 7. Peterson TC, Reiss PJ, Wadland WC. Restructuring a family practice obstetrics curriculum. J Fam Pract 1990;22:219-25. 8. Taylor HA, Hansen GH. Perceived characteristics of successful family practice residency maternity care training programs. Fam Med 1997;29(10):709-14. 9. Smith MA, Howard KP. Choosing to do obstetrics in practice: factors affecting the decisions of third-year family practice residents. Fam Med 1987;19(3):191-4. 10. Petry LJ, Bobula JA. Longitudinal teaching of obstetrics in family practice residency programs. Fam Med 1977;19(3):195-9. 11. Rosenblatt RA. The future of obstetrics in family practice: time for a new direction. J Fam Pract 1988;26:127-9. 12. American Academy of Family Physicians, American College of Obstetricians and Gynecologists, Joint Task Force on Core Educational Guidelines for Family Practice Residents. Am Fam Physician 1998;58: 275-8. 13. Beasley JW, Damos JR, Roberts RG, Nesbitt TS. The Advanced Life Support in Obstetrics course. Arch Fam Med 1994;3:1037-41. 14. Verdon ME. Establishing a linkage between a family practice residency and a local health department. Fam Med 1993;25(5):312-5. 15. Rook JP, Weatherby NL, Ernst EKM, Stapleton S, Rosen D, Rosenfield A. Outcomes of care in birth centers. N Engl J Med 1989;321:1804-11. 16. Nesbitt T. Rural maternity care: new models of access. Birth 1996;23:161-5. 17. Kruse J, Phillips DM, Wesley R. A comparison of the attitudes of obstetricians and family physicians toward obstetric practice, training, and hospital privileges of family physicians. Fam Med 1990;26:219-25. 18. Ferguson GA. Statistical analysis in psychology and education, fourth edition. New York: McGraw Hill, 1976. 19. Klein M. Obstetrics is too important to be left to the obstetricians. Fam Med 1987;19(3):167-9. 20. Weaver SP, Mills TL, Passmore C. Job satisfaction of family practice residents. Fam Med 2001;33:678-82. 21. Pugno PA, McPherson DS, Schmittling GT, Kahn NB Jr. Results of the 2001 National Resident Matching Program: family practice. Fam Med 2001;33:594-601. 22. Denniston CR. A program director s musings on the Match and the future of family practice residency education. Fam Med 2001;33:626-8.