Examination of nursing data elements from evidence-based recommendations for clinical decision support

The purpose of this descriptive, exploratory study was to examine the current use, metadata, and availability of delirium data elements in an electronic health record and clinical data repository. The investigation explored risk for ICU delirium by comparing the delirium data elements representing nursing practice at one clinical agency with a Synthesis of evidence-based practice recommendations developed for the Knowledge-Based Nursing Initiative. The analysis provided a description of the representation of data elements and issues related to the availability of data for use in the future development of clinical decision support system that was intended to prevent ICU delirium. Content analysis and descriptive statistics were used to categorize and analyze the instance-level data from the convenience sample of 1714 patients. Forty-one data element categories were derived from the synthesis based on nursing process components and were matched to 160 data elements identified in the clinical agency's electronic health record. The matched data elements were primarily text based, entered by RNs using flow sheets and care planning documentation. Even though there was a high number of potential data element matches, there was considerable variable data availability related to clinical, conceptual, and technical factors. The further development of valid and reliable data that accurately capture the interaction between nurse, patient, and family is necessary before embarking on electronic clinical decision support.