Helen Pearce

Background: Financial incentives are used in National Health Service (NHS) primary care to influence practice, yet their effects on General Practitioner (GP) behaviour and unintended consequences remain unclear. Recent UK GP contract reforms have changed the incentive landscape, highlighting the need for a contemporary, GP-focused synthesis. Aim: To synthesise evidence on the impact of financial incentives on GP behaviour in high-income countries with predominantly publicly funded health systems, including incentives at GP, organisational or national level and with an effect on GP behaviour. We will assess whether effects differ across population and practice contexts, including implications for health inequalities. Design & Setting: The review focuses on GPs in high-income, predominantly publicly funded healthcare systems where primary care is the first point of contact and gatekeeper to secondary care, and on evidence from 2010 onwards to reflect incentive structures. Method: Seven databases will be searched from 2010 onwards. Records will be managed in EPPI-Reviewer using de-duplication, priority screening of titles and abstracts , and dual independent full-text review. Eligible studies include GPs (or outcomes attributable to GP behaviour) and financial incentives at individual, organisational or system level. Data will be extracted and combined by structural narrative synthesis, with meta-analysis if studies are comparable. Risk of bias will be assessed, qualitative evidence synthesised by meta-aggregation; and certainty graded using GRADE where feasible. Conclusion: This review will synthesise the evidence on how financial incentives affect GP behaviour and health inequalities, informing clinicians and policy makers locally and nationally as incentive models evolve.

Background: Rapidly and accurately synthesizing large volumes of evidence is a time- and resource-intensive process. Once published, reviews often risk becoming outdated, limiting their usefulness for decision makers. Recent advancements in artificial intelligence (AI) have enabled researchers to automate stages of the evidence synthesis process, from literature searching and screening to data extraction and analysis. As previous reviews on this topic have been published, a significant number of tools have been further developed and evaluated. Furthermore, as generative AI increasingly automates evidence synthesis, understanding how it is studied and applied is crucial, given both its benefits and risks. Objective: This review aimed to map the current landscape of evaluated AI tools used to automate evidence synthesis. Methods: Following the Joanna Briggs Institute methodology for scoping reviews, we searched Ovid MEDLINE, Ovid Embase, Scopus, and Web of Science in February 2025 and conducted a gray literature search in April 2025. We included articles published in any language from January 2021 onward. Two reviewers independently screened citations using Rayyan, and data were extracted based on study design and key AI-related technical features. Results: We identified 7841 unique citations through database searches and 19 records through gray literature searching. A total of 222 articles were included in the review. We identified 65 AI tools and 25 open-source models or machine learning (ML) algorithms that automate parts of or the whole evidence synthesis pathway. A total of 54.1% (n=120) of the studies were published in 2024, reflecting a trend toward researching general-purpose large language models (LLMs) for evidence synthesis automation. The most popular tool studied was generative pretrained transformer models, including its conversational interface ChatGPT (n=70, 31.5%). Moreover, 31.1% (n=69) studied tools automated by traditional ML algorithms. No studies compared traditional ML tools to LLM-based tools. In addition, 61.7% (n=137) and 26.1% (n=58) studied AI-assisted automation of title and abstract screening and data extraction, respectively, the 2 most intensive stages and, therefore, amenable to automation. Technical performance outcomes were the most frequently reported, with only 4.1% (n=9) of studies reporting time- or workload-specific outcomes. Few studies pragmatically evaluated AI tools in real-world evidence synthesis settings. Conclusions: This review comprehensively captures the broad, evolving suite of AI automation tools available to support evidence synthesis, leveraged by increasingly complex AI approaches that range from traditional ML to LLMs. The notable shift toward studying general-purpose generative AI tools reflects how these technologies are actively transforming evidence synthesis practice. The lack of studies in our review comparing different AI approaches for specific automation stages or evaluating their effectiveness pragmatically represents a significant research gap. Optimal tool selection will likely depend on the review topic and methodology and researcher priorities. While they offer potential for reducing workload, ongoing evaluation to mitigate AI bias and to ensure the integrity of reviews is essential for safeguarding evidence-based decision-making.

BACKGROUND Rapidly and accurately synthesizing large volumes of evidence is a time and resource-intensive process. Once published, reviews often risk becoming outdated, limiting their usefulness for decision-makers. Recent advancements in artificial intelligence (AI) have enabled researchers to automate various stages of the evidence synthesis process, from literature searching and screening to data extraction. OBJECTIVE We aimed to map the current landscape of AI tools used to automate evidence synthesis. METHODS Following the JBI methodology for scoping reviews, we searched Ovid MEDLINE, Ovid Embase, Scopus, and Web of Science in February 2025, and conducted a grey literature search in April 2025. We included articles published in any language from January 2021 onwards. Two reviewers independently screened citations using Rayyan, and we extracted data based on study design and key AI-related technical features. RESULTS We identified 7,841 unique citations through database searches and 19 additional records through a grey literature search. A total of 222 articles were included in the review. We identified 65 AI tools that automate either specific tasks or the entire evidence synthesis process. More than half of the included studies were published in 2024, reflecting a trend in the use of general-purpose large language models (LLMs) for evidence synthesis. Title and abstract screening, as well as data extraction, were the most studied tasks for automation. CONCLUSIONS A broad, evolving suite of AI tools is available to support automation in evidence synthesis, leveraged by increasingly complex AI methods. Optimal tool selection likely will depend on the review topic, researcher priorities, and specific tasks. While these tools offer potential for reducing manual workload, ongoing evaluation to mitigate AI bias, and ensure quality and integrity of reviews, is essential for safeguarding evidence-based decision-making.