High-precision information retrieval for rapid clinical guideline updates

Area of Science:

• Medical Informatics
• Clinical Practice Guidelines
• Biomedical Research

Background:

• Translating medical evidence into clinical practice is often delayed, impacting patient access to optimal treatments.
• Current clinical guideline updates involve slow, manual evidence synthesis, contributing to significant delays.
• An average of nine years is observed from research initiation to guideline adoption, with 1.7-3.0 years lost post-publication.

Purpose of the Study:

• To address delays in clinical guideline updates by automating evidence synthesis.
• To introduce the Next Generation Evidence (NGE) system for data-driven analysis of evidence translation timelines.
• To develop precision-focused literature search filters for efficient guideline maintenance.

Main Methods:

• Utilizing state-of-the-art biomedical Natural Language Processing (NLP) methods.
• Integrating diverse evidence sources, including clinical trial reports and digital guidelines.
• Developing and evaluating precision-focused literature search filters for guideline updates.

Main Results:

• The Next Generation Evidence (NGE) system enables automated, data-driven analysis of evidence translation.
• Precision-focused filters demonstrated exceptional accuracy in identifying pivotal publications for guideline updates.
• Benchmarking against German oncology guidelines confirmed the filters' effectiveness.

Conclusions:

• The NGE system offers a solution to accelerate the integration of new medical evidence into clinical practice.
• Automated processes and precision filters can significantly reduce the time lag in guideline updates.
• This approach has the potential to improve patient access to the latest evidence-based treatments.