Background and Objectives: This is the 16th report prepared by the American Academy of Family Physicians (AAFP) on the percentage of each US medical school's graduates entering family practice residency programs. Approximately 15.9% of the 16,029 graduates of US medical schools between July 1995 and June 1996 were first-year family practice residents in October 1996, compared with 14.6% in 1995 and 13.4% in 1994. This is the highest percentage since 1980-1981 (12.8%), when this series of studies began. Medical school graduates from publicly funded medical schools were almost twice as likely to be first-year family practice residents in October 1996 than were residents from privately funded schools, 19.1% compared with 11.2%. The Mountain region reported the highest percentage of medical school graduates who were first-year residents in family practice programs in October 1996 at 24%; the Middle Atlantic and New England regions reported the lowest percentages at 11.4% and 9.9%, respectively. Nearly half of the medical school graduates (48.4%) entering a family practice residency program as first-year residents in October 1996 entered a program in the same state where they graduated from medical school. The percentages for each medical school have varied substantially from year to year since the AAFP began reporting this information. This article reports the average percentage for each medical school for the last 3 years. Also reported are the number and percentage of graduates from colleges of osteopathic medicine who entered Accreditation Council for Graduate Medical Education-accredited family practice residency programs.

Results of the 1997 National Resident Matching Program: Family Practice
Norman B. Kahn, Jr, MD; Julea G. Garner, MD; Gordon T. Schmittling, MS; Daniel J. Ostergaard, MD; Robert Graham, MD

The 1997 National Resident Matching Program (NRMP) results reveal all-time highs for the number of positions filled in family practice residencies (2,905) and the number of positions filled with US seniors (2,340). This is the largest number of US seniors choosing family practice in the history of the NRMP. Of the 65 additional positions filled through the NRMP, compared with 1996 (2,905 versus 2,840), 64 were filled with additional US seniors (2,340 versus 2,276). Continuing a trend begun in 1992, the 1997 results showed more positions filled in family practice residencies on July 1 than in the previous year, with 3,570 in 1997, compared with 3,494 in 1996. Internal medicine residencies matched 128 more US seniors in 1997, and pediatric residencies matched 63 more US seniors. Of the US seniors matching through the NRMP, only 37.5% are predicted to practice as generalists. In the past 5 years, family practice has been the choice of 71.5% of additional US seniors choosing primary care. The continued record increased interest in family practice as a career, coupled with the nation's need for more family physicians, demands increased support for the nation's family practice residency programs.

Special Articles: 1997 Match Results
(Fam Med 1997;29(8):553-9.)

Alternative Medicine Instruction in Medical Schools and Family Practice Residency Programs
Michael Carlston, MD; Marian R. Stuart, PhD; Wayne Jonas, MD

Background and Objectives: The use of medical therapies outside of mainstream Western medicine, referred to as complementary medicine or alternative medicine (CAM), is rapidly increasing in the United States. Despite evidence of physician interest and willingness to refer to CAM providers, there is currently little information regarding medical education in these practices. This survey assessed the frequency and nature of alternative medicine instruction in US medical schools and family practice residency programs.

Methods: Society of Teachers of Family Medicine staff mailed a 16-question survey to all US medical school family medicine department chairs and non-university-based family practice residency program directors about existing instruction in alternative medicine, planned instruction, and educational programs under consideration. Results: The overall response rate was 77.9% (364/467), with 29.7% (108/364) of all respondents currently teaching, 6.0% (22/364) starting to teach, and 6.3% (23/364) considering teaching some form of alternative medicine. CAM instruction is most common in the Northeast and Rocky Mountain regions. The instruction is predominantly elective (72.2%). Instructional content and methodologies vary widely. Conclusions: Alternative medicine has begun to establish a presence in US medical schools and family practice residency programs. Offerings in this diverse subject vary widely in content and format.

Educational Research and Methods
(Fam Med 1997;29(8):559-62.)

Background and Objectives: This study identified associations between panic states and family 1) structure, 2) functioning, and 3) stress/support. Methods: Ninety-seven adults with panic disorder or infrequent panic attacks, based on the Structured Clinical Interview of the Diagnostic and Statistical Manual, Third Edition, Revised, were matched to 97 subjects without panic symptoms based on age, gender, and race/ethnicity. All subjects completed a structured interview concerning health care use by family members and family characteristics. Family functioning was assessed using the Family Adaptability and Cohesion Evaluation Scales, and family stress/support were assessed using the Duke Social Support and Stress Scales.

Results: Although groups did not differ in either perceived or ideal family cohesion or adaptability, the panic group perceived their families as more dysfunctional and reported higher levels of family stress and total stress but lower levels of support, including family support, nonfamily support, and total support. Conclusions: Subjects with panic symptoms have families with high levels of dysfunction and stress but low levels of support. Increased family dysfunction may be due to comorbid substance abuse.

Clinical Research and Methods
(Fam Med 1997;29(8):563-7.)

Background and Objectives: A variety of types of data ranging from administrative data sets to large-scale surveys have been collected for one purpose but can be accessed and used to address new questions of relevance to primary care. For many research questions, primary data collection isn't necessary because the data already exists. Although relatively few researchers consider analyzing existing data sets, there are a variety of advantages to secondary analysis. This paper introduces the practical aspects of performing secondary analysis. We review the types of questions that can be addressed, where and how to access data, and the steps involved in preparing data for analysis.

Research Series
(Fam Med 1997;29(8):568-71.)

Background and Objectives: Esophagogastroduodenoscopy (EGD) is a procedure that has been gaining interest among family physicians. To determine the extent of EGD training at family practice residencies, we surveyed all US family practice residency directors. Methods: In late 1994, we mailed a questionnaire to the residency directors of all ACGME-accredited family practice residency programs. Results: A total of 359 programs responded to the questionnaire (87.7%). Eighty-seven (24.2%) programs reported that their residents receive training to perform EGD. At the majority of these programs, residents perform 10 or less EGDs during their residency experience. Programs on the East and West coasts were less likely to provide this training than were programs in other regions of the country. Residents at programs in the South performed significantly more EGDs than did residents in programs in other regions. Southern residents also were more likely to perform these procedures in family practice centers rather than gastroenterology (GI) labs. Those programs whose residents perform more than 25 EGDs were more likely to have family physicians as trainers of this procedure and also were more likely to offer this training at family practice centers instead of GI labs. Conclusions: EGD training is now offered by a relatively large number of family practice residency programs. However, concern could be raised as to whether these programs provide adequate quantitative experience. Strong regional differences were noted in the availability of this training.

Special Series: More on Procedures in Family Medicine
(Fam Med 1997;29(8):572-4.)

Background and Objectives: Colonoscopy training is receiving greater emphasis in family practice residencies. However, no standards have been established to measure the adequacy of this training. This study assessed the colonoscopy experience of family practice residents at Louisiana State University Medical Center at Shreveport (LSUMC-S). Methods: We included all colonoscopies performed by the family practice service between August 1992 and December 1994 and matched them by gender and age with cases from the gastroenterology (GI) and general surgery (GS) services performed during the same time period. Family practice and GI were compared using 143 cases from each service; 166 cases were used to compare family practice to GS. Results: The cecum was intubated in 87% of patients on all services. The average time to complete the procedure was 35 minutes by the family practice service, 44 minutes by GI, and 25 minutes by GS. No significant differences were found between family practice and GI in the number of patients with polyp, normal colon, or biopsy performed. In comparison to GS, there were significantly fewer patients on the family practice service with normal colon and more with multiple polyps and biopsy performed. Significantly more cancers were found by the family practice service than by either GI or GS. There were no complications reported for any of the services. Results compared favorably with data in the current literature. Conclusions: The colonoscopy experience available to family practice residents at LSUMC-S is acceptable within the parameters studied.

Special Series: More on Procedures in Family Medicine
(Fam Med 1997;29(8):575-9.)

Background and Objectives: Fiberoptic flexible sigmoidoscopy (FFS) is widely used by family physicians to evaluate abdominal problems and screen for colorectal cancer. We evaluated data on exams performed by family practice residents to determine the number of supervised procedures needed for technical proficiency at FFS. Methods: We reviewed data recorded from all FFS procedures done at a family practice residency from October 1986-July 1994. Results: A total of 262 exams were performed by 55 residents at the Family Practice Center. There was a modest correlation between increasing numbers of exams and increased unassisted depth of insertion (UDI). Maximum UDI was achieved after 10-15 supervised exams. Factors such as patient gender, prior surgery, and preparation quality were also significantly correlated with UDI. Significant differences in training experience and patient selection were seen between male and female residents. Conclusions: Maximal UDI is reached after 10-15 procedures. Differences in training experiences and successful UDI based on resident gender should be studied further.

Special Series: More on Procedures in Family Medicine
(Fam Med 1997;29(8):580-3.)

Background and Objectives: A core curriculum for procedural training in family practice is desirable. However, opinions differ as to which of many emerging technologies should be taught. This lack of agreement is due in part to the political and financial burdens of securing hospital privileges, the scheduling burdens of expanding an overcrowded curriculum, and a generational barrier between physicians who feel that technology enhances the biopsychosocial model versus those who feel otherwise. Nevertheless, as emerging technologies are shown to have established value in primary care, the core procedural curriculum will continue to evolve.

Special Series: More on Procedures in Family Medicine
(Fam Med 1997;29(8):584-5.)

Background and Objectives: Teaching outpatient procedures is a major responsibility of family practice residencies. This requires equipment, faculty expertise, and curricula. This report documents one residency program's attempt to enhance the teaching of outpatient procedures. Methods: We tabulated all office procedures completed during the years 1993 through 1996 at a family practice residency program in the southeastern United States. We compared the number of procedures and dollar charges before and during an educational intervention designed to increase the number of procedures performed and to improve procedural training of residents. The intervention consisted of hiring a full-time procedural nurse and new procedural equipment. Results: The residency program spent $177,000 over 4 years to improve procedural training, but income from procedures was estimated to be only $127,000. From 1993 to 1996, the total number of annual procedures increased from 322 to 540. Both the number of procedures and procedural income declined in the final year of the project when the procedural nurse assumed other clinical and administrative duties. Conclusions: Volume and teaching of common procedures in a family practice residency program can be enhanced by a dedicated effort, but this effort requires considerable staffing and financial resources.

Special Series: More on Procedures in Family Medicine
(Fam Med 1997;29(8):586-9.)