A chart review process aided by natural language processing and multi-wave adaptive sampling to expedite validation of code-based algorithms for large database studies

TL;DR

This paper presents a faster validation method for code-based algorithms in large databases, combining NLP and adaptive sampling to reduce review time and resource use while maintaining accuracy.

Contribution

It introduces a novel validation process integrating NLP-assisted chart review and multi-wave adaptive sampling to improve efficiency in large database studies.

Findings

NLP reduced chart review time by 40%.

Adaptive sampling prevented 77% of chart reviews with limited accuracy loss.

The method enhances routine validation of database algorithms.

Abstract

Background: One of the ways to enhance analyses conducted with large claims databases is by validating the measurement characteristics of code-based algorithms used to identify health outcomes or other key study parameters of interest. These metrics can be used in quantitative bias analyses to assess the robustness of results for an inferential study given potential bias from outcome misclassification. However, extensive time and resource allocation are typically re-quired to create reference-standard labels through manual chart review of free-text notes from linked electronic health records. Methods: We describe an expedited process that introduces efficiency in a validation study us-ing two distinct mechanisms: 1) use of natural language processing (NLP) to reduce time spent by human reviewers to review each chart, and 2) a multi-wave adaptive sampling approach with pre-defined criteria to stop the validation study once performance characteristics are identified with sufficient precision. We illustrate this process in a case study that validates the performance of a claims-based outcome algorithm for intentional self-harm in patients with obesity. Results: We empirically demonstrate that the NLP-assisted annotation process reduced the time spent on review per chart by 40% and use of the pre-defined stopping rule with multi-wave samples would have prevented review of 77% of patient charts with limited compromise to precision in derived measurement characteristics. Conclusion: This approach could facilitate more routine validation of code-based algorithms used to define key study parameters, ultimately enhancing understanding of the reliability of find-ings derived from database studies.