Using Evidence-Based Nursing Practices and Electronic Health Record Decision Support to Reduce Fall-related Patient Injuries in Acute Care

This project focused on developing and implementing clinical decision support (CDS) tools to support nurses in the development of care plans and involvement in quality improvement activities in the area of fall prevention in acute care. The project was a joint effort between researchers at the Aurora Health Care System Nursing Research, the University of Wisconsin-Milwaukee Colleges of Nursing and Health Sciences, and Cerner Corporation.

The main objectives of the project were to:

• Design, build, and implement CDS tools that were populated with data extracted from an electronic health record (EHR).
• Assess whether the CDS tools could help nurses improve care planning and quality-improvement activities, and patient and family education related to fall prevention in acute care.

A mixed-methods study design included pre-post queries of data, direct observation, focus groups, surveys, and usability testing. The team examined the "As-Is" state in order to inform the development of the CDS tools. Three tools were built: (1) a care planning CDS tool (CP-CDS), (2) a quality improvement CDS tool (QI-CDS), and (3) a fall prevention educational tool for patients and family members. The team also developed training materials for each of the tools.

Once the CDS tools were developed, they were populated with EHR-based data. Despite the early involvement of nurses in the project, they were slow to adopt the tools. Feedback indicated that the CP-CDS tool was outside the nurses' normal workflow, that the tool was slow to load, and that it had insufficient assessment and intervention details. Feedback on the QI-CDS tool from leaders indicated that they were able to access the tool, believed it brought disparate data together, and could save time and improve data quality. However, leaders were unable to find the time to actually use the tool. Staff-nurse feedback on the Fall Prevention Tool was positive, but again the tool was not widely adopted post-implementation. The team theorized that the reasons for the poor adoption of the tools included competing EHR implementations and resource reduction during training, go-live, and adoption periods. Despite these limitations, this study illustrates the complexities of nursing workflow and the need to understand the sociotechnical context and how CDS can support nurses' decisionmaking.