Collaborative large language models for automated data extraction in living systematic reviews
Open accessPublished version of a preprint
The DOI matches its Crossref record; 16 of the 16 references it cites were found in a registry or an index.
Abstract
Objective: Data extraction from the published literature is the most laborious step in conducting living systematic reviews (LSRs). We aim to build a generalizable, automated data extraction workflow leveraging large language models (LLMs) that mimics the real-world 2-reviewer process. Materials and Methods: A dataset of 10 trials (22 publications) from a published LSR was used, focusing on 23 variables related to trial, population, and outcomes data. The dataset was split into prompt development (n = 5) and held-out test sets (n = 17). GPT-4-turbo and Claude-3-Opus were used for data extraction. Responses from the 2 LLMs were considered concordant if they were the same for a given variable. The discordant responses from each LLM were provided to the other LLM for cross-critique. Accuracy, ie, the total number of correct responses divided by the total number of responses, was computed to assess performance. Results: In the prompt development set, 110 (96%) responses were concordant, achieving an accuracy of 0.99 against the gold standard. In the test set, 342 (87%) responses were concordant. The accuracy of the concordant responses was 0.94. The accuracy of the discordant responses was 0.41 for GPT-4-turbo and 0.50 for Claude-3-Opus. Of the 49 discordant responses, 25 (51%) became concordant after cross-critique, increasing accuracy to 0.76. Discussion: Concordant responses by the LLMs are likely to be accurate. In instances of discordant responses, cross-critique can further increase the accuracy. Conclusion: Large language models, when simulated in a collaborative, 2-reviewer workflow, can extract data with reasonable performance, enabling truly “living” systematic reviews.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Language Models are Few-Shot Learners
Brown, T. B., Mann, B., Ryder, N., Subbiah, M., Kaplan, J., Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G., Askell, A., et al. (2020) · arXiv · doi:10.48550/arxiv.2005.14165 · Cited by 23,372
Large Language Models Encode Clinical Knowledge
Singhal, K., Azizi, S., Tu, T., Mahdavi, S. S., Wei, J., Chung, H. W., Scales, N., Tanwani, A., Cole-Lewis, H., Pfohl, S., et al. (2022) · arXiv · doi:10.48550/arxiv.2212.13138 · Cited by 4,083
Large language models encode clinical knowledge
Singhal, K., Azizi, S., Tu, T., Mahdavi, S. S., Wei, J., Chung, H. W., Scales, N., Tanwani, A., Cole-Lewis, H., Pfohl, S., et al. (2023) · Nature · doi:10.1038/s41586-023-06291-2 · Cited by 4,083
Prompt Programming for Large Language Models: Beyond the Few-Shot Paradigm
Reynolds, L., & McDonell, K. (2021) · Extended Abstracts of the 2021 CHI Conference on Human Factors in Computing Systems · doi:10.1145/3411763.3451760 · Cited by 952
Drug treatments for covid-19: living systematic review and network meta-analysis
Siemieniuk, R. A., Bartoszko, J. J., Zeraatkar, D., Kum, E., Qasim, A., Díaz Martinez, J. P., Izcovich, A., Rochwerg, B., Lamontagne, F., Han, M. A., et al. (2020) · BMJ · doi:10.1136/bmj.m2980 · Cited by 846
PaLM-E: An Embodied Multimodal Language Model
Driess, D., Xia, F., Sajjadi, M. S. M., Lynch, C., Chowdhery, A., Ichter, B., Wahid, A., Tompson, J., Vuong, Q., Yu, T., et al. (2023) · arXiv · doi:10.48550/arxiv.2303.03378 · Cited by 591
The Wilcoxon signed rank test for paired comparisons of clustered data
Rosner, B., Glynn, R. J., & Lee, M. T. (2006) · Biometrics · doi:10.1111/j.1541-0420.2005.00389.x · Cited by 557
Large language models are few-shot clinical information extractors
Agrawal, M., Hegselmann, S., Lang, H., Kim, Y., & Sontag, D. (2022) · Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/2022.emnlp-main.130 · Cited by 358
Large Language Models Are Human-Level Prompt Engineers
Zhou, Y., Muresanu, A. I., Han, Z., Paster, K., Pitis, S., Chan, H., & Ba, J. (2022) · arXiv · doi:10.48550/arxiv.2211.01910 · Cited by 330
Large language models for generative information extraction: a survey
Xu, D., Chen, W., Peng, W., Zhang, C., Xu, T., Zhao, X., Wu, X., Zheng, Y., Wang, Y., & Chen, E. (2024) · Frontiers of Computer Science · doi:10.1007/s11704-024-40555-y · Cited by 326
Research works printed but not linked: 18
- Jaiyesimi, 2024, Therapy for stage IV non-small cell lung cancer without driver alterations: ASCO living guideline, version 2023.3, J Clin Oncol, 42, e23, 10.1200/JCO.23.02746 · DOI
- Michelen, 2021, Characterising long COVID: a living systematic review, BMJ Glob Health, 6, e005427, 10.1136/bmjgh-2021-005427 · DOI · Paper page
- Sutton, 2012, An introduction to conditional random fields, FNT Mach Learn, 4, 267, 10.1561/2200000013 · DOI · Paper page
- Hasan, 10.1136/bmjebm-2023-112597 · DOI · Paper page
- Kartchner, 2023, 10.18653/v1/2023.bionlp-1.37 · DOI
- Ma, 2023, Understanding and mitigating overfitting in prompt tuning for vision-language models, IEEE Trans Circuits Syst Video Technol, 33, 4616, 10.1109/TCSVT.2023.3245584 · DOI
- Riaz, 2018, A meta-analysis of randomized controlled trials for efficacy and safety of vascular endothelial growth factor tyrosine kinase inhibitors (VEGF-TKIs) adjuvant therapy in high-risk renal cell cancer (RCC), J Clin Oncol, 36, 4567, 10.1200/JCO.2018.36.15_suppl.4567 · DOI
- Siddiqi, 2020, Adjuvant therapy in high-risk renal cell cancer: a systematic review and cumulative meta-analysis, J Clin Oncol, 38, 708, 10.1200/JCO.2020.38.6_suppl.708 · DOI
- Naqvi, 2023, The role of volume of disease for treatment selection in patients with metastatic castration sensitive prostate cancer (mCSPC): a living meta-analysis, J Clin Oncol, 41, 5088, 10.1200/JCO.2023.41.16_suppl.5088 · DOI
- Naqvi, 2023, Deep prostate-specific antigen response and overall survival in patients with metastatic castration-sensitive prostate cancer: a systematic review and meta-analysis, J Clin Oncol, 41, 195, 10.1200/JCO.2023.41.6_suppl.195 · DOI
- Riaz, 2023, A living interactive systematic review and network meta-analysis evaluating systemic therapies in metastatic castration-sensitive prostate cancer (mCSPC), J Clin Oncol, 41, 193, 10.1200/JCO.2023.41.6_suppl.193 · DOI
- Naqvi, 2022, Indirect comparisons of triplet therapy as compared to novel hormonal therapy doublets in patients with metastatic castration sensitive prostate cancer, J Clin Oncol, 40, 5083, 10.1200/JCO.2022.40.16_suppl.5083 · DOI
- Riaz, 2022, First-line treatment options in metastatic castration-sensitive prostate cancer: a systematic review and network meta-analysis, J Clin Oncol, 40, 132, 10.1200/JCO.2022.40.6_suppl.132 · DOI
- Naqvi, 2022, Triplet therapy in metastatic castration-sensitive prostate cancer: a systematic review and meta-analysis, J Clin Oncol, 40, 136, 10.1200/JCO.2022.40.6_suppl.136 · DOI
- Khan, 2024, First-line PARP inhibitors (PARPi) with androgen receptor pathway inhibitors (ARPi) in metastatic castration-resistant prostate cancer (mCRPC): a systematic review and meta-analysis, J Clin Oncol, 42, 173, 10.1200/JCO.2024.42.4_suppl.173 · DOI
- Bibi, 2024, Efficacy of first line (1L) poly ADP-ribose polymerase inhibitors (PARPi) in combination with androgen pathway inhibitors (API) by homologous recombination repair (HRR) genes in metastatic castration-resistant prostate cancer (mCRPC): a living meta-analysis, J Clin Oncol, 42, e17059, 10.1200/JCO.2024.42.16_suppl.e17059 · DOI
- Khan, 2023, Toxicity profile of Janus kinase inhibitors (JAKi) in patients with myelofibrosis: a report from a living systematic review and meta-analysis, Blood, 142, 6433, 10.1182/blood-2023-188107 · DOI
- Valmeekam, 2023, On the planning abilities of large language models—a critical investigation, Adv Neural Inf Process Syst, 36, 75993
Cited by
Artificial intelligence across the cancer care continuum
Riaz, I. B., Khan, M. A., & Osterman, T. J. (2025) · Cancer · doi:10.1002/cncr.70050 · Cited by 40
Self-citationNovel AI applications in systematic review: gPT-4 assisted data extraction, analysis, review of bias
Kim, J. K., Chua, M. E., Li, T. G., Rickard, M., & Lorenzo, A. J. (2025) · BMJ Evidence-Based Medicine · doi:10.1136/bmjebm-2024-113066 · Cited by 11
Diagnosing the limits of large language models for structured scientific evidence extraction
Tan, Z., D’Souza, J., & Auer, S. (2026) · International Journal on Digital Libraries · doi:10.1007/s00799-026-00449-9
Structured Schemas for Provenance‐Rich, LLM ‐Assisted QSP Model Calibration
Eliason, J., & Popel, A. S. (2026) · CPT: Pharmacometrics & Systems Pharmacology · doi:10.1002/psp4.70306
Metadata
- Authors
- Khan, M. A., Ayub, U., Naqvi, S. A. A., Khakwani, K. Z. R., Sipra, Z. B. R., Raina, A., Zhou, S., He, H., Saeidi, A., Hasan, B., Rumble, R. B., Bitterman, D. S., Warner, J. L., Zou, J., Tevaarwerk, A. J., Leventakos, K., Kehl, K. L., Palmer, J. M., Murad, M. H., Baral, C., & Riaz, I. B.
- Venue
- Journal of the American Medical Informatics Association
- Publisher
- Oxford University Press (OUP)
- Published
- 2025-01-21
- Volume
- 32
- Issue
- 4
- Pages
- 638-647
- ISSN
- 1067-5027 (ISSN-L)
- 1527-974X
- OpenAlex
- W4406663495
- Cited by
- 50
- References
- 16
- FWCI
- 13.24
- Type
- journal article
- Language
- en
- Open access
- green
- Funders
- National Institutes of Health (U24 CA265879)
- AstraZeneca
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Collaborative large language models for automated data extraction in living systematic reviews. https://app.cochat.ai/papers/10.1093/jamia/ocae325
- The paper
- Khan, M. A., Ayub, U., Naqvi, S. A. A., Khakwani, K. Z. R., Sipra, Z. B. R., Raina, A., Zhou, S., He, H., Saeidi, A., Hasan, B., Rumble, R. B., Bitterman, D. S., Warner, J. L., Zou, J., Tevaarwerk, A. J., Leventakos, K., Kehl, K. L., Palmer, J. M., Murad, M. H., . . . Riaz, I. B. (2025). Collaborative large language models for automated data extraction in living systematic reviews. Journal of the American Medical Informatics Association. https://doi.org/10.1093/jamia/ocae325
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.