Analyzing the effect of data quality on the accuracy of clinical decision support systems: a computer simulation approach.

Clinical decision support systems (CDSS) use data from a variety sources to provide guidance to physicians at the point of care. However, several studies have shown that data from these registries often can not be trusted to be accurate or complete. For instance, studies show that accuracy and completeness in medical registries may be as low as 67% and 30.7%, respectively. Consequently, since CDSS rely on this data for generating guidance, the possibility that the medical decisions facilitated by the system may result in negative patient outcomes still exists. To analyze the extent of this problem, we present a two-pronged approach using simulation, followed by regression in order to quantify the relative impact of poor data quality on overall CDSS accuracy. The results from this analysis can be beneficial to developers and hospitals who can use the results to inform the development of procedures for minimizing incorrect medical decisions facilitated by these systems.