Monitoring quality of health care with relational patterns.

This paper describes the extraction of relational patterns from hospital discharge data for the purpose of monitoring the quality of health care. We discuss what relational patterns are and how they support the extraction of meaningful patterns from data that are expressive, comprehensive, and easy to interpret. We demonstrate how relational patterns can be applied to identify poor practices embedded in hospitalization processes, for instance, help trigger subsequent inquiries and support decision-making processes.