Understanding Delivery of Computer-based Intensive Insulin Therapy

Abstract

Intensive insulin therapy (IIT), a nurse-driven protocol combining frequent blood glucose testing and insulin administration to tightly control blood glucose, became the standard of critical care following a 2001 study. Many institutions subsequently implemented computer-based clinical decision support systems (CDSSs) for IIT. However, recent studies question IIT’s benefit and safety. Whereas previous research investigated effects of patient characteristics on IIT performance, this dissertation evaluated IIT CDSS with respect to the interaction of people, process, and technology. An organizational analysis using institutional theory explored the influence of peers, regulators, and professions in IIT adoption. A literature review and case study demonstrated the underreported role of social, organizational, and contextual factors affecting IIT CDSS. A quantitative analysis of system records established the frequency and effect of blood glucose data mismatches as well as characteristics and effects of nurse dosing overrides on IIT CDSS performance. An ethnographic study of nurse workflow yielded understanding of how IIT CDSS functions with respect to other clinical information systems and care processes. Using a mixed quantitative and qualitative approach informed by social theory, this research demonstrates how sociotechnical interactions affect IIT CDSS and may be leveraged to improve care delivery.