DOES RESIDENCY SIZE MATTER? THE IMPACT OF PROGRAM SIZE AND FREESTANDING CHILDREN?S HOSPITAL STATUS ON PEDIATRIC RESIDENCY TRAINING Scott A. Shipman, William L. Cull, Sarah E. Brotherton, Richard J. Pan Pediatrics & Health Policy and Management, Johns Hopkins, Baltimore, MD; AAP, Elk Grove Village, IL; AMA, Chicago, IL; Pediatrics, UC Davis, Davis, CA

BACKGROUND: Pediatric residencies differ greatly in size, ranging from 12 to 120 residents. Some programs are based in freestanding children?s hospitals, and others are in general medical centers. These organizational differences may impact residency training.

OBJECTIVE: To study the influence of program size and children's hospital status on certain residency experiences and career preparation.

DESIGN/METHODS: Data from the 1997-99 AAP National Third Year Resident Survey, the AMA FREIDA Online Database, and the American Board of Pediatrics were utilized. Programs were grouped by their total number of categorical pediatric residents in 1997 as follows: small (<30 residents), medium (30-59) and large (60+). Residents' training experiences and career decisions were compared across program sizes, and multivariate regression was conducted, adjusted for the nested study design.

RESULTS: Residents in small programs were more likely to be married, have children, and to be foreign medical graduates. Small programs were less likely to be affiliated with academic centers and freestanding children?s hospitals, and graduates of smaller programs were less likely to pass the Board certifying examination (pass rates: small 76%, medium 82%, large 85%). Residents from small programs reported less time in emergency departments and inpatient wards, and more time in outpatient clinics and community sites. Child advocacy and community needs assessment training were rated higher in small programs. In multivariable regression adjusted for resident and program characteristics, residents' career plans were similar across programs of all sizes. However, primary care training in small programs was rated very good or excellent more frequently than in medium (OR=2.2) and large (OR=2.9) programs. Conversely, preparation for fellowship training was rated highest in large programs. Freestanding children's hospitals had no independent effect on Board pass rates or the reported readiness for primary care or fellowship.

CONCLUSIONS: Residency size has a strong influence on the pediatric training experience. Training in a freestanding children?s hospital has a negligible effect on similar measures. Program size is an easily quantifiable measure that may be a proxy for the emphasis placed on primary care during training.