Clinical registries and quality measurement in surgery: a systematic review

Surgical clinical registries provide clinical information with the intent of measuring and improving quality. This study aimed to describe how surgical clinical registries have been used to measure surgical quality, the reported findings, and the limitations of registry measurements.

Medline, CINAHL, and Cochrane were queried for English articles with the terms: "registry AND surgery AND quality." Eligibility criteria were studies explicitly assessing quality measurement with registries as the primary data source. Studies were abstracted to identify registries, define registry structure, uses for quality measurement, and limitations of the measurements used.

A total of 111 studies of 18 registries were identified for data abstraction. Two registries were financed privately, and 5 registries were financed by a governmental organization. Across registries, the most common uses of process measures were for monitoring providers and as platforms for quality improvement initiatives. The most common uses of outcome measures were to improve quality modeling and to identify preoperative risk factors for poor outcomes. Eight studies noted improvements in risk-adjusted mortality with registry participation; one found no change. A major limitation is bias from context and means of data collection threatening internal validity of registry quality measurement. Conversely, the other major limitation is the cost of participation, which threatens the external validity of registry quality measurement.

Clinical registries have advanced surgical quality definition, measurement, and modeling as well as having served as platforms for local initiatives for quality improvement. The implication of this finding is that subsidizing registry participation may improve data validity as well as engage providers in quality improvement.