Small-area variations in the treatment of prevalent medical conditions: a comparison of three cities in the Northeast

This article presents a small-area variation study that examines utilization differences for primary care physicians (PCPs) in treating a homogeneous set of prevalent medical conditions. The study used secondary data collected over a 24-month period from a large, Northeastern region independent practice association. The diagnostic cluster methodology was used to examine geographic differences for PCPs in treating prevalent medical conditions. This methodology groups International Classification of Diseases, 9th revision (ICD-9), codes into diagnostic clusters based on clinical homogeneity with respect to generating a similar clinical response from the physician. For each diagnostic cluster, diagnostic episode clusters (DECs) were formulated. Each DEC links all services incurred in treating a patient's medical condition within a specific period of time. Differences in use rates across small areas were tested using t tests. The data showed little variation in the physician office visit rate across small areas. However, services generated from these office visits exhibited large rate variations. The most significant small-area differences were for hospital inpatient days and surgical procedures. Pattern-of-treatment differences exist across small areas for the homogeneous set of prevalent medical conditions treated by PCPs.