Collaboration on Improving the Efficiency of Data Abstraction - 2020 Quality Summit presentation (Non-CE)-Collaboration on Improving the Efficiency of Data Abstraction

Video Transcription

Video Summary

The speaker begins by expressing gratitude for the opportunity to present their experience using automated data abstraction with natural language processing (NLP) at Lucio Packard Children's Hospital at Stanford. They emphasize the importance of clinical registries and the need to improve their efficiency. The speaker explains that the current manual data abstraction process is inefficient, error-prone, and limits the amount of data that can be collected. They highlight that NLP, a subset of artificial intelligence, can automate chart abstraction and curate both structured and unstructured data from electronic health records. The speaker presents their findings from implementing NLP in cardiac registries, demonstrating significant time savings and improved efficiency. They discuss the benefits and limitations of using NLP in registry support, highlighting its effectiveness in variables that require computation, logic, and standardized definitions. However, variables requiring adjudication or flexible interpretation still pose challenges. The speaker concludes that NLP improves the user interface for abstractors and allows for comprehensive data capture, enhancing research and quality improvement opportunities. The presentation acknowledges the partnership with Carta Healthcare in developing the NLP software and the cost-effectiveness of improving data abstraction efficiency. Contact information is provided for further inquiries.

Keywords

automated data abstraction

natural language processing

clinical registries

electronic health records

NLP software