Improving Allergy Documentation and Clinical Decision Support in the Electronic Health Record

In the United States, allergies are one of the primary causes of chronic illness, impacting approximately 50 million individuals. Six percent of hospitalizations are attributable to medication allergies known to induce adverse drug reactions, resulting in longer hospital stays and higher healthcare expenses. However, due to the lack of specificity in the allergy entry, clinicians are less likely to recommend treatments that patients can tolerate. No internal infrastructure exists to reconcile allergy information throughout the entire electronic health record (EHR). Moreover, allergy records are rarely erased and reviewed, resulting in the buildup of erroneous or obsolete information.

In this study, the researchers created an allergy module that detects inconsistencies in allergy information within the EHR. In particular, the team proposed an innovative health information technology system redesign, including making a dynamic picklist that puts answers in order of how important they are based on the allergen input using natural language processing.

The specific aims of the research were as follows:

• Improve reaction documentation by developing a comprehensive and interactive value set.
• Develop an innovative allergy reconsolidation module within the EHR.
• Redesign drug allergy alerting mechanisms.
• Distribute study methods and tools so they are widely available to other researchers and healthcare institutions for noncommercial use.

The researchers used both quantitative and qualitative methods to analyze patient demographics and allergy histories within the EHR, understand usability and impact through clinician interviews, and compare existing allergy alert override behavior. They then implemented a real-time EHR clinical decision support tool that uses picklists to rank the most likely allergic reactions to a given allergen, comprehensive tools to support real-time allergy reconciliation, and alerting systems that reduce rather than increase cognitive burden, with the overall goal of improving allergy documentation.

The findings showed that the use of the tool decreased documentation time and decreased the number of free-text entries, suggesting that this alerting approach has the potential to significantly reduce the frequency of disruptive alerts and result in improved drug administration safety and a reduction of alert fatigue. The development of the allergy module in real time highlights the significance of comprehensive allergy clinical decision support tools. In terms of precision and practicality, the findings indicate that a dynamic response picklist has the potential to outperform the currently used static picklists. Overall, the results of the study give several important insights into the existing status and future development of allergy documentation and decision support infrastructure inside the EHR.