Performance of large language models in data extraction for evidence synthesis: A systematic review

Data extraction is among the most resource-intensive and error-prone stages of systematic review production. Large language models (LLMs) offer potential for automating or semi-automating this process, yet their performance characteristics remain incompletely characterised. This systematic review aimed to comprehensively evaluate LLM accuracy, reliability, and efficiency for data extraction in evidence synthesis, and to identify optimal implementation strategies.

We searched PubMed, Embase, Web of Science, and preprint servers (medRxiv, arXiv) through December 2025. Studies were eligible if they evaluated one or more LLMs for data extraction against a human reference standard and reported quantitative performance metrics. Two reviewers independently extracted data and assessed methodological quality using PROBAST + AI and reporting completeness using TRIPOD-LLM. Narrative synthesis was performed due to substantial heterogeneity precluding meta-analysis.

Twenty-seven studies met inclusion criteria, evaluating models including GPT-4/4o (n = 15), Claude versions 2-3.5 (n = 10), Gemini (n = 3), and open-source alternatives including Llama, Mistral, Qwen, and DeepSeek. Overall accuracy ranged from 47% to 99.9%, with substantial heterogeneity by task type and data granularity. Categorical and string variables were extracted more reliably (74-96%) than numerical data (47-88%). Claude 3.5 Sonnet achieved high accuracy in an assistive workflow (91.0%; 95% CI: 90.4-91.6%), exceeding human-only extraction (89.0%). Claude models outperformed GPT in head-to-head comparisons (OR 1.70 for event counts). Omissions were the dominant error type (60-74%), with hallucination rates of only 0.08-6%, challenging widespread fabrication concerns. Time savings of 33% to 87% were reported, although most included studies did not quantitatively assess efficiency. Methodological quality was generally robust, with 74.1% of studies rated low risk of bias under PROBAST + AI. Mean TRIPOD-LLM compliance was 88.5%, though gaps in inference settings and model version documentation were common.

LLMs demonstrate promising but variable performance for data extraction in evidence synthesis. Current evidence supports their integration as assistive tools within dual-extraction workflows requiring human verification, rather than as autonomous extractors. Categorical data is extracted more reliably than numerical outcomes, and few-shot prompting with structured output formats consistently improves performance. Standardised benchmarks and prospective comparative studies remain priorities for future research.