Databases for assessing the outcomes of the treatment of patients with congenital and paediatric cardiac disease--a comparison of administrative and clinical data.

The introduction of the reporting of medical and surgical outcomes to the public and the potential implementation of initiatives involving pay-for-performance have invigorated debates about the relative benefits of administrative and clinical databases for comparing rates of mortality at the level of the hospital and surgeon. While general agreement exists that public performance report cards must use the highest quality data available, debate continues regarding whether administrative or clinical data should be utilized for this purpose. Clinical databases may contain information more relevant to risk-adjustment, but the currently available clinical databases are voluntary and suffer from validity concerns. Administrative data, however, suffer from inaccuracies of coding and a lack of potentially informative covariates. Particularly problematic to congenital heart surgery is the non-uniform application of coding algorithms to define complex reconstructive procedures for which there is no unique code assignment. The purposes of this manuscript are; therefore, to discuss the relative advantages and limitations of both clinical and administrative data, and to provide a brief introduction to currently available databases germane to the study of congenital cardiac disease.