During the last decade, neonatal intensive care programs were developed in Alabama as elsewhere. Regionalization was associated with burgeoning transport demands and clinical activity.1 The influx of patients initially managed in the community led to the recognition that there were significant differences between the medical practice of primary and tertiary care physicians. Our unit, like others, sought to narrow these differences through continuing medical education (CME). Conventional CME techniques, such as monthly professional updates and semiannual lectures, were supplemented with care protocols or clinical algorithms as they are sometimes termed. The hypothesis was that care protocols would be a more effective and efficient means to change medical practice. A controlled-trial technique was used to evaluate the impact of these interventions on both neonatal care process and outcomes. The results showed that clinical algorithms can affect practice performance. We recommend their consideration as a supplementary CME technique to those seeking. © 1986, American Medical Association. All rights reserved.