ARTICLE |
a Ambulatory Pediatric Association, McLean, Virginia
b Department of Practice and Research, Center for Child Health Research, American Academy of Pediatrics, Elk Grove Village, Illinois
| ABSTRACT |
|---|
|
|
|---|
METHODS. A cross-sectional study comparing data for Continuity Research Network practice visits during a 1-week period in 2002 with data from the 2000 National Ambulatory Medical Care Survey was performed. Continuity Research Network and National Ambulatory Medical Care Survey data were derived from 30 patient visits per practice site for patients <22 years of age, with the primary care providers being residents and practicing pediatricians, respectively.
RESULTS. Eighteen Continuity Research Network practices reported on 540 visits, compared with 32 National Ambulatory Medical Care Survey physicians reporting on 792 visits. Continuity Research Network patients were more likely to be black non-Hispanic or Hispanic/Latino and to have public insurance. The top 5 reasons for visits were the same for Continuity Research Network and National Ambulatory Medical Care Survey visits, although the orders varied slightly. These 5 reasons accounted for 58% of Continuity Research Network visits and 49% of National Ambulatory Medical Care Survey visits. Continuity Research Network visits were more likely to result in patient instructions to return at a specific time (78% vs 52%).
CONCLUSIONS. Residents in Continuity Research Network practices provide care to more underserved patients but evaluate problems that are similar to those observed in office practices; the Continuity Research Network practices thus provide important training experiences for residents who will serve both minority and nonminority children.
Key Words: resident continuity clinic resident education National Ambulatory Medical Care Survey Continuity Research Network
Abbreviations: CORNETContinuity Research Network NAMCSNational Ambulatory Medical Care Survey PROSPediatric Research in Office Settings
The Accreditation Council for Graduate Medical Education has mandated the pediatric residency continuity experience since 1989.1 The goals are to provide residents with experience as the primary care provider for children of all ages, with a variety of diseases, in an environment that promotes continuity of care and emulates a practice model. The Ambulatory Pediatric Association has developed extensive primary care guidelines for pediatric residency education.2 What residents learn during their training is critical because they are likely to conform to the assumptions, beliefs, values, and practice norms of physician role models and opinion leaders and to emulate what they learn during residency in their future practice behavior.3,4 Because >70% of residents choose to practice community-based primary care,5 it seems prudent, in our opinion, to evaluate the residency educational experiences to determine whether the continuity experience is training our residents to meet the needs of our nation's children and to evaluate the worth of this longitudinal time investment.
To date, studies examining the pediatric residency continuity experience have included primarily single residency programs.611 Investigators have studied resident-patient continuity,6,7,9 site of practice,7,9 preparation for future practice,1,8 and resident satisfaction with the continuity experience.12,13 Although these studies have provided some valuable information about the continuity experience, most are limited to single residency programs611 and are based on resident self-reporting.7,8,12,13 Results have been conflicting at times, which perhaps reflects institutional differences. Although the continuity experience has been mandated, few objective data are available regarding the population of patients cared for, the content and disposition of visits, or the care patients receive when residents serve as the primary care providers. To our knowledge, no studies have compared patient visits at continuity practices with national normative findings for office settings where pediatric visits with practicing pediatricians occur.
With recognition of the need for more-generalizable data and a commitment to studying the continuity experience, the Continuity Research Network (CORNET) was established in 2002 as a national pediatric practice-based research network of resident continuity practices and was endorsed as a core function of the Ambulatory Pediatric Association. The overall CORNET research goals are to study the health care of minority and underserved children, to examine health care disparities, and to study resident education, with comparisons of physician behaviors between pediatric residents and pediatricians in practice. This comparison of CORNET visits with visits from the National Ambulatory Medical Care Survey (NAMCS), a national survey of visits to office-based physicians, is important to determine the demographic features of patients who receive their primary care from continuity practices, to determine the types and content of visits in which pediatric residents participate, and to ascertain the patient characteristics of this new practice-based research network, to determine how representative the network is and how CORNET practices compare with national practices. Patient visit details provide information that a research network can use to understand the distinctive aspects of its patients, which can guide additional research.14
Using data about pediatric visits to NAMCS practices and to CORNET practices, we hypothesized that visits to CORNET practices would be more likely to be made by minority and underserved children but would be similar with respect to the patients' chief complaints and the clinicians' leading diagnoses, compared with visits described in the NAMCS. The study objectives were to compare patient visits to pediatric resident continuity practices with patient visits to office practices, with respect to the variables of patient demographic features, parents'/patients' chief complaints, clinicians' leading diagnoses, longitudinal care, types of visits, referrals made, and visit dispositions.
| METHODS |
|---|
|
|
|---|
Included in our analysis were visits to NAMCS pediatric practitioners by patients <22 years of age. The CORNET study used the same methods, patient age group, and time frame; the only difference was that pediatric residents served as the primary care providers. Types of contacts not included for either sample were those made by telephone, those made outside the physician's office (for example, house calls), and those made for administrative purposes only (ie, to leave a specimen or to pick up a prescription).
Each CORNET practice collected data on a random sample of 30 patient visits during a 1-week period in May through August 2002 (summer months similar to the time frame for NAMCS data collection). The CORNET practices replicated the NAMCS systematic sampling methods, which were based on the expected number of visits for the practice in 1 week. During this period, data for a systematic random sample were recorded by the physician or office staff members, on an encounter form provided for that purpose. Practices were asked to enter the data at a secure, Internet-based, entry site maintained by the American Academy of Pediatrics. This study was performed in conjunction with the Pediatric Research in Office Settings (PROS) network, a practice-based research network of the American Academy of Pediatrics, although the data presented represent CORNET data only.
The CORNET study used a data collection instrument almost identical to that for the NAMCS. The patient demographic variables included race, ethnicity, age, gender, insurance status, type of visit, patient's/parent's chief complaint, clinician's leading diagnosis (defined by International Classification of Diseases, 9th Revision, code), whether the patient was a new or established patient, whether the patient was seen by his or her regular primary care provider, whether referrals were made, and visit disposition. The only departure from NAMCS methods was that CORNET practices had Internet access for data entry and included a question to document the level of training of the pediatric resident.
Pediatric practitioners and their designated office personnel gained access to the Internet-based NAMCS data collection instrument through the PROS home Web page, by entering a member identification number. Site principal investigators could gain access only to their own survey records. Surveys were completed on a restricted-access basis by using the secure, encrypted, Internet site and were deidentified by the practices to protect the rights of human subjects. Data instruments completed online were downloaded directly into study-specific Microsoft Access (Microsoft, Redmond, WA) tables and imported to SPSS software (SPSS, Chicago, IL) for analyses. Institutional review board approval was obtained at each participating institution.
Thirty visits from each practice were selected randomly, consistent with the NAMCS methods. Data analysis was performed by using visit weighting to reflect the sampling probability. With the relative weights being retained, the visit numbers were adjusted so that the total N value was the number of observations in each data set. SAS software (version 9.1; SAS Institute, Cary, NC) was used for weighting and analyses. Data collected from each specific NAMCS patient record were compared with findings for CORNET as a whole, with respect to variables such as age, gender, practice arrangement, and specialty work. Data analyses for the study included descriptive and univariate procedures, Student's t test, and
2 analysis.
| RESULTS |
|---|
|
|
|---|
Table 1, which compares the 18 participating CORNET sites with 10 nonparticipating CORNET sites, demonstrates no statistically significant differences. Figure 1 shows the national distribution of the 18 participating CORNET practices. In a comparison of geographic locations for CORNET versus NAMCS practices, 43% vs 25% were located in the South, 33% vs 22% in the Northeast, 17% vs 34% in the West, and 7% vs 19% in the Midwest.
|
|
|
|
|
|
|
| DISCUSSION |
|---|
|
|
|---|
The adequacy of residency training in continuity clinics as a preparation for practice is an important and complex issue. The chief complaints and provider diagnoses were very similar between CORNET and NAMCS practices. This concordance suggests that the clinical content in the residency continuity experience at large academic center-based programs is similar to that for pediatricians in practice, which provides some reassurance that the continuity experience is helping to prepare residents for their futures as pediatricians. The major reasons for CORNET visits were for well-child care or physical examinations, which is another important training aspect because these skills of physical examination, anticipatory guidance, screening, developmental assessment, and nutritional counseling over a time period are not covered in other portions of residency training.2 However, there are other identified factors that influence residents' preparation for practice and satisfaction with the continuity experience; clinic site (private practice versus other settings), number of patients seen per session, and variety of chronic medical problems have all been identified as important determinants of the quality of the continuity experience.7,8,12,13 These factors are all subject to large institutional variations, which explains some of the differences in previous research findings but reinforces the importance of studies from multiple practice settings. Future CORNET studies can examine these and other questions that require larger sample sizes and other practice-based research network attributes.
It is important to educate pediatric residents about the unique health care needs of underserved patients and to introduce concepts such as community health, the medical home, advocacy, and other ideas from the Future of Pediatric Education II.2,23,24 Residents from academic sites gain exposure in working with underserved populations. In a multisite national study, >60% of residents who had their continuity experience in hospital-based clinics and provided care to underserved patients stated that they served as advocates for their primary care patients.12 It is important to address the needs of the underserved population and to influence residents' interest in caring for this population. Weitzman et al25 demonstrated that pediatric residents who had an interest in caring for underserved patients when they entered residency tended to maintain that interest; however, those who did not enter with an interest did not develop an interest during residency training. This finding suggests that, despite likely exposure to an underserved population during the continuity experience, the training experience does not encourage residents to care for these patients. Clearly, there is a need to explore the reasons for this finding and to modify experiences to encourage this interest. CORNET practices have an opportunity to study health care issues affecting underserved patients, compared with national probability samples and the PROS network.
Another important element of residency training is continuity with patients. Although the majority of patients from CORNET visits saw their resident primary care provider, this continuity was less than that noted for pediatricians in practice, whereas a large proportion of CORNET patients had been seen in the practice previously. Other continuity-based studies illustrated an overall modified degree of patient continuity.6,7,9 Although a previous single-program study showed that residents in private practices had higher measures of continuity of care, those practices also were more likely to have patients not seen for health maintenance visits, compared with a resident continuity clinic practice.7 A previous multisite study demonstrated superior continuity at hospital-based clinics, compared with continuity experiences in private practitioners' offices.12 The 58% continuity identified in this study was identical to the 58% determined through resident self-report in a large multisite study assessing the continuity at hospital-based clinic sites.12 This finding reinforces the importance of having multisite data, as opposed to data from single training programs or practices. In contrast to full-time private pediatricians such as those studied in NAMCS, residents have limited clinic time. Given these time constraints, we think that a continuity rate of 58% is satisfactory and allows exposure to longitudinal experiences for residents and patients.
CORNET practices were more likely to be in the South and Northeast than in the West and Midwest, compared with the location of NAMCS practices. This finding is likely because continuity clinics in academically based training programs such as those enrolled in CORNET tend to be located in urban population centers. Also, participants in CORNET are more likely to be academicians interested in education and the majority of pediatric residency programs in the western United States are based in university medical schools and not in freestanding children's hospitals. In addition, urban hospital-based clinics serve a large proportion of the nation's underserved children, which may account for the demographic characteristics of CORNET patients.26 CORNET patients were also more likely to be between 1 and 4 years of age, compared with NAMCS patients. This finding may reflect enrollment by residents of patients from inpatient services such as the NICU or newborn service into their continuity practices, which has been substantiated in other studies.6,7,10,12 Although the longer duration of visits in CORNET practices is most likely associated with residency training, Zuckerman et al27 noted that parents were more likely to report having received excellent care with visits averaging 20 minutes, compared with 12.4 minutes for children who received poor care. Although there are delays, perhaps parents appreciate having more time spent with them and receiving input from >1 physician (both resident and supervising preceptor).
Although this study did not address health care disparities specifically, the findings indicate some interesting avenues for future research. CORNET and NAMCS patients visit their doctors for largely the same reasons but have markedly different insurance and socioeconomic status. Therefore, CORNET represents an opportunity to study potential health care disparities. Patients at CORNET visits were more likely to have a specific time for a follow-up visit documented in their medical records. However, we do not know whether NAMCS practices educated their patients to schedule follow-up visits automatically or whether they had fewer concerns regarding compliance and found that specific follow-up instructions were unnecessary. CORNET practices enhanced the documentation to ensure continued involvement and longitudinal care within the practice. This may have important implications for future health care use. Flores et al28 noted that pediatric hospitalizations could be avoided if better follow-up visits were scheduled. CORNET patient disposition was also more likely to reflect that a referral was given. Future studies need to assess inequities between the sites with respect to chronic illnesses, receipt of prescriptions, and use of durable medical equipment. It would also be of interest to look at practice variables for the resident physicians in continuity clinics and the pediatricians in NAMCS practices, to determine what role experience has with respect to the chronic illness rates of the patient population.
There are several limitations that must be considered. NAMCS practices were selected randomly, whereas CORNET practices volunteered to participate and may not be representative of all continuity practices. However, in a comparison of participating and nonparticipating CORNET practices, there were no significant differences in the type of practice, location, or number of residents. A poststudy analysis that compared participating CORNET practices with 56 practices that joined after the study showed no differences except for the number of residents per practice; participating practices had significantly more residents per practice (mean: 35.2 vs 25.6 residents per practice; P = .04). Because of institutional review board delays, the CORNET data collection for 2 of the 18 practices extended beyond the May to June time frame for collection of NAMCS data into July and August. This might have led to some differences in the types of visits, perhaps with more well-child care visits, because more school physicals for the upcoming school year tend to occur during the summer. Finally, most residents participate in their continuity experiences only 1 half-day per week, which surely would affect the continuity variable of whether the patients were seen by their primary care providers.
| CONCLUSIONS |
|---|
|
|
|---|
| ACKNOWLEDGMENTS |
|---|
Study sites and coinvestigators were as follows: Baylor College of Medicine, Jan Drutz, MD; Brody School of Medicine/East Carolina University, John Olsson, MD; Children's National Medical Center, Rachel Moon, MD; Johns Hopkins Children's Center, Janet Serwint, MD; Medical College of Georgia, Jack Benjamin, MD; Naval Medical Center-Portsmouth, Timothy Shope, MD, MPH; New York Medical College, Theresa Hetzler, MD; Oregon Health and Science University, Cynthia Ferrell, MD; Rush Children's Hospital, Beth Volin, MD; University of California, Irvine, Lynn Hunt, MD; University of Maryland, Susan Feigelman, MD; University of Miami, Lee Sanders, MD, MPH; University of Pittsburgh, Evelyn Reis, MD; University of South Florida, Sharon Dabrow, MD; University of Texas at Houston, Michelle S. Barratt, MD, MPH; University of Utah, Sarah Croskell, MD; University of Vermont, Wendy Davis, MD; Virginia Commonwealth University, Helen Ragazzi, MD.
We acknowledge our coinvestigators at the continuity sites who participated in this study, the Continuity Special Interest Group Task Force that helped to form CORNET, the Ambulatory Pediatric Association for endorsing this network, and the PROS network of the American Academy of Pediatrics for assistance, guidance, and use of the Internet-based data entry system.
| FOOTNOTES |
|---|
Address correspondence to Janet R. Serwint, MD, Johns Hopkins Hospital, Park 389, 200 N Wolfe St, Room 2076, Baltimore, MD 21287. E-mail: jserwint{at}jhmi.edu
This work was presented in part at the Pediatric Academic Society Meeting, Ambulatory Pediatric Association Presidential Plenary Session; May 3, 2004; San Francisco, CA.
The authors have indicated they have no financial relationships relevant to this article to disclose.
| REFERENCES |
|---|
|
|
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||