{"doi":"10.1002/gin2.70029","title":"Opportunities and Challenges in Using Artificial Intelligence in Guideline Development and Implementation","abstract":"Artificial intelligence (AI) within clinical care is a burgeoning area of interest. AI is rapidly transforming the landscape of clinical care, revolutionising how healthcare is delivered, managed and optimised [1]. With its ability to analyse vast amounts of data, identify patterns and make real-time predictions, AI holds immense promise in improving patient outcomes, enhancing efficiency and reducing healthcare costs when the technology is used responsibly and its potential risks are mitigated. AI tools are currently being explored for training clinicians, enhancing the rigour, breadth and precision of environmental and biomedical health data, and reporting, evaluating and synthesising data in biomedical research and clinical practice [2, 3]. In response, biomedical research and public health agencies are developing frameworks and guidelines for the use of AI in clinical research and direct patient care [1, 4, 5]. Similarly, funding agencies, scholarly editors and shared oversight bodies are releasing advisories and guidelines for investigators and authors engaged in peer review and writing [6, 7]. Clinical Practice Guidelines (CPGs) are cornerstones of evidence-based medicine, providing standardised recommendations for patient care [8]. As part of ‘Bringing Guidelines to the Digital Age’, an effort by the Guidelines International Network North America (GIN-NA) regional community that applied human-centred design (HCD) to solve pain points in guideline development and implementation, a multidisciplinary team of systematic reviewers, guideline developers, informaticists, and others worked collaboratively to gain insights, conduct research, and identify possible solutions for three pain points in guideline development and implementation [9, 10]. The pain points included (1) insufficient informatics resources and expertise to translate ambiguous or complex language in written guidelines into computable formats, (2) incomplete information in existing guidelines that necessitates adapting guidelines to cater for the needs of diverse patient populations, and (3) unclear language, which creates barriers for patients and clinicians to understand and use guidelines [9]. Given the widespread availability of AI tools, the team emphasised the need to help guideline developers understand how approaches that include AI tools could address these challenges and facilitate guideline development and implementation. These pain points serve as a framework for exploring potential AI applications within each stage of guideline development. Through weekly meetings, the team looked at potential areas for automation with AI tools in the life cycle of guideline development, adapted from the GIN-McMaster Guideline Development Checklist and the 2011 Institute of Medicine Report, Clinical Practice Guidelines We Can Trust [8, 10]. These automation approaches require a shift in behaviours and mindsets among guideline developers to effectively integrate AI tools into guideline development workflows. Amid the rapid evolution of AI technology, we anticipate that enhanced AI tools with accuracy, validation and adherence to governance frameworks will facilitate the effective development of guidelines without significantly compromising quality. These tools could eventually surpass human performance in certain steps in the near future, but their use must remain grounded in principles of transparency, equity and accountability to fully realise their potential while upholding scientific integrity and public trust [11, 12]. The primary objective of this manuscript is to enlighten guideline developers to re-evaluate current methods and leverage AI's power effectively within the CPG development and implementation continuum. By structuring our discussion to first explore AI's transformative potential, we aim to provide a comprehensive and practical perspective on its role in shaping the future of clinical guidelines, with a later section dedicated to addressing its ris","journal":"Clinical and Public Health Guidelines","year":2025,"id":532835,"datarank":0.0,"base_score":0.0,"endowment":0.0,"self_citation_contribution":0.0,"citation_network_contribution":0.0,"self_endowment_contribution":0.0,"citer_contribution":0.0,"corpus_percentile":null,"corpus_rank":null,"citation_count":3,"citer_count":0,"citers_with_citation_signal":0,"citers_with_endowment":0,"datacite_reuse_total":0,"is_dataset":false,"is_dataset_confidence":0.9575,"is_data_producer":false,"deposit_databanks":null,"is_oa":true,"file_count":0,"downloads":0,"has_version_chain":false,"published_date":"2025-01-01","fair_score":null,"fair_percentile":null,"algorithm_id":"datarank_citation_only_1hop_v6","ranking_scope":"data_only","authors":[{"id":1414347,"name":"Xiaomei Yao","orcid":"0000-0002-1210-3758","position":1,"is_corresponding":false},{"id":1414840,"name":"Jonathan L. Heald","orcid":null,"position":2,"is_corresponding":false},{"id":107201,"name":"Stacy Lathrop","orcid":null,"position":3,"is_corresponding":false},{"id":642651,"name":"Heba Hussein","orcid":"0000-0001-6042-361X","position":4,"is_corresponding":false},{"id":1414348,"name":"Maria Michaels","orcid":"0000-0002-8268-5714","position":5,"is_corresponding":false},{"id":1414346,"name":"Chirine Chehab","orcid":"0000-0003-0261-6020","position":0,"is_corresponding":true}],"reference_count":31,"raw_metadata":null,"created_at":"2026-07-19T02:51:27.893975Z","pmid":null,"pmcid":null,"fwci":null,"citation_percentile":null,"influential_citations":0,"oa_status":null,"license":null,"views":0,"total_file_size_bytes":0,"version_count":0,"fair_f":null,"fair_a":null,"fair_i":null,"fair_r":null,"fair_zscore":null,"fair_rationale":null,"fair_model":null,"fair_agent_version":null,"fair_fulltext_source":null,"fair_has_llm":null,"fair_computed_at":null,"clinical_trials":[],"software_tools":[],"db_accessions":[],"linked_datasets":[],"topics":[]}