Epidemiologic Method Review at Scale: Assessing Charlson Comorbidity Versioning Using a LLM
Fuchs, J.; Johnson, C.; Foster, N.; Leese, P. J.
Show abstract
The Charlson Comorbidity Index (CCI) is widely used in epidemiologic studies. However, many versions of the CCI have been developed since the original method was published in 1987, and it is unclear which version is used most frequently and how version utilization in research has changed over time. We present an approach using a large language model to extract data from articles by detecting which specific CCI version is employed. We analyzed 31,767 articles published since 2012 to evaluate the landscape of CCI implementation. We show that 63% of articles that cite a single method version cite only the original CCI publication, which cannot be applied in the modern real-world data era, leading to ambiguity in actual CCI implementation. More generally, this paper introduces a generalizable approach to scaling methods literature review beyond typical human-review, which we validate and demonstrate through application to the CCI.
Matching journals
The top 5 journals account for 50% of the predicted probability mass.
Similar papers in this journal
Similar papers in this journal
- Use of unstructured text in prognostic clinical prediction models: a systematic review 94%
- Incorporating Preprints in Systematic Reviews: A Preliminary Study of a Novel Method for Rapid Evidence Synthesis 94%
- Automating literature screening and curation with applications to computational neuroscience 93%
Similar papers in this journal
- COVID-19-related research data availability and quality according to the FAIR principles: A meta-research study 96%
- Introducing the EMPIRE Index: A novel, value-based metric framework to measure the impact of medical publications 96%
- Knowledge and motivations of training in peer review: an international cross-sectional survey 95%
Similar papers in this journal
- Large language models for conducting systematic reviews: on the rise, but not yet ready for use – a scoping review 95%
- The use of the Registered Reports format for publication of randomized clinical trials: a cross-sectional study 94%
- The impact of retracted randomised controlled trials on systematic reviews and clinical practice guidelines: a meta-epidemiological study 94%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.