A Systems Model of HIT-Induced Complexity

Abstract

Background While health information technology (HIT) is playing a key role in transforming the healthcare system into a collaborative patient centered system, it is common for unintended consequences (UICs) to emerge post-HIT implementation. Healthcare delivery is a complex adaptive system and UICs occur because of multiple interactions between technology, users, organizational policies, and other situational contexts. Understanding the nature of these interactions and the manner in which they occur is a necessary first step to managing UICs from HIT implementation.

Methodology We use a case study of a perioperative system to identify three categories of UICs. We then further analyzed the UICs using complex adaptive systems concepts to articulate the interactions that led to the UICs as well as the upstream and downstream implications of them.

Results From our analysis we developed a systems model of four dimensions of HIT induced complexity: temporal, policy, workflow, and connectivity complexity. It also emphasizes that we cannot think of HIT implementation as an in-the-moment event. Rather, tasks such as information entry or retrieval may have emerging properties and evolve in complexity as the tasks interact with other people, processes, and technologies.

Conclusion Implementing HIT in complex healthcare settings is a significant challenge. While the complexity of healthcare delivery prevents us from predicting the specific interactions that lead to UICs, our systems model of HIT complexity enables us to make inferences about how certain interactions occur and the contexts where they occur. Our model helps our understanding of the complexity of HIT implementation and improves our ability to proactively manage UICs.