Presenter Information

Preslie BretonFollow

Presentation Type

Poster Presentation

Abstract

Interpreting free-text medication orders in Electronic Health Record (EHR) systems is challenging due to variability, increasing the risk of errors and non-compliance with Joint Commission (TJC) standards. This research develops Natural Language Processing (NLP) tools to enhance medication order accuracy, particularly for Duplicate PRN orders. Our open-source, vendor-neutral approach ensures broad applicability across healthcare institutions while maintaining data security.

Using Python and the Natural Language Toolkit (NLTK), we implemented a four-phase methodology. First, the NLP framework was developed on Google Colab for collaborative accessibility. Next, models were trained on 10,000 anonymized synthetic free-text medication orders to detect non-compliant entries. Model validation was conducted using precision, recall, and F1 score metrics, achieving high accuracy. The precision score of 1.000 demonstrated no false positives, while a recall of 0.989 indicated minimal missed duplicates, resulting in an F1 score of 0.995. These findings highlight the model’s strong reliability in identifying non-compliant PRN orders.

Our NLP tools flagged 12.7% of PRN orders as potential duplicates, underscoring the prevalence of therapeutic duplication in free-text entries. These results validate the potential of NLP-driven automation for improving TJC compliance and medication safety. Future work will focus on real-world validation and integration into educational resources to enhance healthcare professionals’ understanding of compliance standards.

Faculty Mentor

Dr. Jay Dorris- Lipscomb University College of Pharmacy

Share

COinS
 

Developing NLP-Based Open-Source Tools to Improve Medication Order Interpretation in EHR Systems

Interpreting free-text medication orders in Electronic Health Record (EHR) systems is challenging due to variability, increasing the risk of errors and non-compliance with Joint Commission (TJC) standards. This research develops Natural Language Processing (NLP) tools to enhance medication order accuracy, particularly for Duplicate PRN orders. Our open-source, vendor-neutral approach ensures broad applicability across healthcare institutions while maintaining data security.

Using Python and the Natural Language Toolkit (NLTK), we implemented a four-phase methodology. First, the NLP framework was developed on Google Colab for collaborative accessibility. Next, models were trained on 10,000 anonymized synthetic free-text medication orders to detect non-compliant entries. Model validation was conducted using precision, recall, and F1 score metrics, achieving high accuracy. The precision score of 1.000 demonstrated no false positives, while a recall of 0.989 indicated minimal missed duplicates, resulting in an F1 score of 0.995. These findings highlight the model’s strong reliability in identifying non-compliant PRN orders.

Our NLP tools flagged 12.7% of PRN orders as potential duplicates, underscoring the prevalence of therapeutic duplication in free-text entries. These results validate the potential of NLP-driven automation for improving TJC compliance and medication safety. Future work will focus on real-world validation and integration into educational resources to enhance healthcare professionals’ understanding of compliance standards.

 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.