AIM:The aim of this study is to evaluate the efficacy and safety of blood-activating and stasis-removing Chinese patent medicines (BASR-CPMs) during the perioperative period of percutaneous coronary intervention (PCI) for myocardial infarction (MI). METHODS:We searched eight databases (PubMed, Embase, the Cochrane Library, Web of Science, CNKI, WanFang Data, SinoMed, and VIP) from database inception to February 15, 2025, for randomized controlled trials (RCTs) comparing standard care plus BASR-CPMs versus standard care alone (or with other BASR-CPMs) in MI patients undergoing PCI. A Bayesian network meta-analysis was conducted to estimate relative effects. Primary outcomes included major adverse cardiovascular events (MACE) and major adverse cardiac and cerebrovascular events (MACCE), while key secondary outcomes included thrombolysis in MI (TIMI) grade 3 flow, angina, and bleeding events. Risk of bias was assessed using the Cochrane Risk of Bias 2 (RoB 2) tool. The certainty of evidence was evaluated using the GRADE framework. A protocol of the systematic review and network meta-analysis was registered with PROSPERO (CRD420251048208). RESULTS:We included 160 RCTs (21,147 participants) evaluating 25 BASR-CPMs. Regarding hard clinical endpoints, no BASR-CPMs differed significantly from standard care for MACE or MACCE at any time point. For example, Tongxinluo capsule showed no significant reduction in MACE at 1 month (relative risk [RR] 0.62, 95% credible interval [CrI] 0.11 to 1.70; low certainty). Tongxinluo capsule may improve TIMI grade 3 flow immediately after PCI (RR = 1.12, 95% CrI: 1.04-1.26; low certainty) and at 3 months (RR = 1.47, 95% CrI: 1.10-1.98; moderate certainty). Danhong injection (RR = 1.15, 95% CrI: 1.02-1.31) and Shexiang Baoxin pill (RR = 1.15, 95% CrI: 1.02-1.30) may also improve TIMI grade 3 flow immediately after PCI (moderate certainty). For reducing angina incidence at 1 and 6 months, Tongxinluo, Danhong, Salvianolate injection, Shexiang Baoxin, Guanxin Shutong, and Shexiang Tongxin dripping pill showed potential benefits (low to high certainty). Most interventions did not increase bleeding risk, and Tongxinluo possibly reduced adverse effects (low certainty). CONCLUSIONS:Exploratory findings suggest that BASR-CPMs, notably Tongxinluo, Danhong, and Shexiang Baoxin, improve immediate reperfusion and reduce angina recurrence post-PCI in MI patients. These intermediate benefits do not translate into reductions in hard clinical outcomes. Further rigorous RCTs are needed to confirm their long-term impact on MACE and MACCE.

OBJECTIVE:To evaluate harms reporting practices in placebo-controlled randomized clinical trials (RCTs) of Chinese herbal medicine (CHM) formulas published in Quartile 1 (Q1) English-language and Tier-1 Chinese journals. METHODS:This systematic survey evaluated harms reporting in CHM formula RCTs. We systematically identified eligible RCTs published in English-language journals Q1 (2024 Journal Citation Reports) and Chinese Tier-1 journals (2023 Traditional Chinese Medicine ranking). Two reviewers independently evaluated harms reporting using items derived from the CONSORT Extension for Harms (CONSORT Harms) recommendations and CHM formula-specific reporting elements. RESULTS:Among 96 eligible RCTs (49 English Q1; 47 Chinese Tier-1), only 19.8% (n = 19) of trials had a published protocol, with a significantly higher proportion in English Q1 journals (p < 0.001). Only one trial explicitly reported adherence to the CONSORT Harms extension (2004 or 2022). Syndrome differentiation was significantly more frequent in Chinese Tier-1 journals than in English Q1 journals (76.6% vs. 38.8%, p < 0.001). Sixty-seven percent of studies (n = 64) did not report methods for assessing the relationship between harms and interventions, and 15.6% (n = 15) relied solely on clinical judgment. Eighty percent of studies (n = 77) did not report statistical methods for harms analysis. In the results section, 62% (n = 59) did not present harms data in tables. CONCLUSIONS:Harms reporting in CHM formula RCTs remains inadequate. Routine adoption of the CONSORT Harms framework, together with CHM formula-specific harms reporting elements, is needed to improve transparency and facilitate the interpretation of harms data in CHM formula research.

Traditional Chinese medicine (TCM) faces persistent gaps between evidence generation and clinical use. Building on Qian Xuesen's theory of open complex giant systems (OCGSs) and its qualitative-to-quantitative metasynthesis, we advance a complex-systems evidence framework tailored to TCM's hallmarks: a holistic perspective and pattern-based diagnosis and therapy. The framework integrates systems science, artificial intelligence, and allied disciplines to coordinate qualitative and quantitative approaches and to align macro-level effectiveness evaluation with micro-level mechanistic inquiry. It organizes multi-source evidence into a four-phase loop-production, differentiation, application, and validation: (1) standardizes evidence production; (2) conducts integrated evaluation along the disease-pattern-formula axis; (3) supports individualized effectiveness evaluation and decision-making; and (4) uses real-world feedback to verify and refine evidence. By linking clinical phenotypes, pathways, and outcomes into an evidence chain, the framework is intended to improve clinical decision quality and accelerate translational research. Beyond TCM, it offers a generalizable model for complex interventions acting on complex human systems, positioning TCM research for international scientific dialogue and modernization.

Medical guideline texts are characterized by strong domain specificity, complex conditional logic, and implicit semantics, which make their automatic transformation into executable process models highly challenging.To address this problem, we reformulate the task as a structure-constrained, multi-stage semantic reasoning problem and propose a dependency-aware BPMN automatic modeling method based on large language models. The proposed approach explicitly decomposes the modeling process into three progressively structured semantic stages: activity and event extraction, dependency relation identification, and BPMN structure generation, thereby improving logical consistency while effectively reducing model hallucinations. Experimental results demonstrate that the proposed method achieves F1 scores of 0.88 and 0.87 in activity recognition and dependency extraction, respectively, and yields significant improvements in gateway identification and overall structural matching accuracy. The results indicate that the proposed approach exhibits clear advantages in modeling complex conditional branches, enhancing process structure interpretability, and controlling modeling errors, representing a novel methodological paradigm for medical guideline process modeling.

Objective To summarize the key evidence in the development of drug treatment strategies for type 2 diabetes mellitus (T2DM), and to provide an evidence-based roadmap for tailoring pharmacological strategies in patients with T2DM. Methods A systematic search of domestic and foreign databases, relevant guidelines websites, and T2DM guidelines and expert consensus organized by the society was conducted. The search time limit is from January 1, 2010, to April 1, 2024, and the update search is until April 1, 2025. A series of guidelines with early publication, stable and timely update, clear use process of antihyperglycemic drugs and high clinical applicability of the guidelines were screened. Results Two guideline series compiled by the Chinese Diabetes Society and the American Diabetes Association were selected to summarize the development of drug treatment strategies for T2DM, and to summarize the key evidence. With the version updates, the overall development trend of the two guideline series is consistent: both of them pay more attention to individualized treatment and comprehensive management, from blood glucose oriented to outcome oriented, and the dual benefits of GLP-1RA, SGLT2i and other drugs hypoglycemic+cardiorenal protection are the key to promoting change. However, the two series differ in the timing of addressing cardiac and renal complications and weight management. Differences also exist in the details of first-line agent selection and the prioritization of cardiac and renal complications. Conclusion This study summarizes the key evidence in the development of drug treatment strategies for T2DM, and to promote the rational planning of drug treatment strategies for patients with T2DM by medical staff.

Version, regional, and conclusion differences in clinical practice guidelines (CPGs) cause inconsistent recommendations. A structured modeling method maps recommendations to multi-field representation. Logistic regression-based weighted scoring selects candidate pairs to reduce data scale and annotation cost; structured-text semantic fusion fine-tunes LLMs for fine-grained recommendation relationship identification. Experiments show the method outperforms baselines in screening efficiency and classification performance, supporting clinical guideline knowledge integration and intelligent decision support.

BackgroundMyocardial infarction (MI) poses a formidable health challenge, frequently necessitating management through percutaneous coronary intervention (PCI). However, PCI comes with potential complications that can impact patient outcomes. Traditional Chinese medicine (TCM), particularly the utilization of Chinese patent medicines with blood-activating and stasis-resolving properties, offers another approach to enhance PCI efficacy and improve patient quality of life. The aim of this study is to assess the comparative efficacy and safety of blood activating and stasis removing Chinese patent medicines for the perioperative period of PCI for MI.MethodsThis systematic review and network meta-analysis will be reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. A comprehensive search strategy will be implemented across seven electronic databases to identify relevant studies. Eligible studies will be limited to randomized controlled trials that compare any Chinese patent medicine (added to standard care) with standard care or another treatment in patients in the perioperative period of PCI after MI. Two independent reviewers will screen all retrieved citations, extract pertinent data, and assess the risk of bias. We will conduct Bayesian random-effects network meta-analysis and network meta-regression. To elucidate whether the intervention has an important impact on certain outcomes within the perioperative period of PCI for MI, we will conduct a patient values and preferences survey to determine the minimum important difference for outcomes. We will assess the certainty of evidence using the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) framework.DiscussionThis study will provide new insights into the efficacy and safety of blood activating and stasis removing Chinese patent medicines for the perioperative period of PCI for MI patients, providing help for future clinical practice and research.Ethics and disseminationEthical approval is not required for this review. The findings will be disseminated through publication in a peer-reviewed academic journal, presentations at scientific conferences, and outreach via various media platforms.

OBJECTIVES:Many clinical practice guidelines (CPGs) claim to follow the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) approach, yet their actual adherence remains uncertain. We aimed to identify and evaluate existing criteria for evaluating adherence to GRADE in CPGs and describe the extent of adherence revealed by their application. STUDY DESIGN AND SETTING:We systematically searched PubMed, Embase, Web of Science, and the Cochrane Library from January 2004 to November 2024, supplemented by manual reviews of the GRADE handbook and related websites. Eligible documents were those that either proposed or applied criteria for evaluating GRADE adherence. We summarized the criteria offered, evaluated the extent to which they meet standards for a structured instrument, and analyzed adherence findings. RESULTS:We identified 11 eligible documents: eight proposed evaluation criteria (four applying their own to evaluate CPGs) and three applied existing criteria. From these criteria, we derived 10 clearly discrete items that can be used to define good adherence to GRADE. The most common items addressed the criteria for assessing certainty of evidence (CoE) and criteria for making recommendations (seven of eight documents), while question framing and outcome selection and importance rating were less frequent (3/8). Among the eight documents proposing criteria, four failed to report their development methods and response options; furthermore, none conducted user testing or provided guidance manuals. The number of CPGs assessed in the included studies ranged from 4 to 86. Reported adherence to GRADE varied considerably across these studies; nevertheless, the median adherence rates were suboptimal for five of the 10 items. CONCLUSION:Existing criteria for assessing GRADE adherence are varied and suboptimal in usability. Evaluations based on these criteria reveal significant application gaps in key areas among CPGs that claim to use GRADE. These findings highlight the need to develop a structured, user-tested evaluation instrument and improve the implementation of GRADE in CPGs. PLAIN LANGUAGE SUMMARY:Many medical guidelines say that they use a trusted scientific method called GRADE to make their advice reliable. But we neither know if they actually follow it correctly nor were there good tools to check. We wanted to find out: 1) what checklists exist to audit GRADE use? and 2) what do those checklists tell us about how well guidelines really apply GRADE? We searched scientific databases and official GRADE sources for studies that either created or used checklists to evaluate GRADE in guidelines. We reviewed these checklists, checked their quality, and summarized what they found when applied. We found 11 relevant studies. The checklists they proposed were inconsistent and poorly designed; none were tested with real users or came with instructions. When used, they showed that many guidelines fail to properly apply key steps of the GRADE method, even though they claim to use it. For half of the important steps, less than 50% of guidelines followed them correctly. This means that current tools for checking GRADE use are unreliable, and many guidelines may not be as trustworthy as they claim. To protect patients and improve medical advice, we urgently need a better, user-friendly audit tool and more support to help guideline developers apply the GRADE method correctly.

OBJECTIVE:To develop and validate a rapid, evidence-based benefit-risk assessment index for off-label drug use in pediatrics, thereby promoting standardized clinical decision-making. METHODS:Under the auspices of the Guangdong Pharmaceutical Association, a multidisciplinary expert panel from pediatrics, pharmacy, methodology, and ethics was convened. Systematic literature searches informed a preliminary index framework. Three Delphi rounds refined indicators. Weights were assigned using the analytic hierarchy process. Expert-based validation involved 20 hematology-oncology experts assessing 10 cases; intraclass correlation coefficient (ICC) and Spearman's correlations evaluated reliability and validity. RESULTS:The index consists of 2 primary dimensions (benefits: 56 points; risks: 44 points), 10 third-level, and 40 fourth-level indicators, with five prerequisites (e.g., no alternatives, informed consent). The ICC was 0.77 (good reliability); Spearman's correlations were rs = 0.97 (benefits) and rs = 0.68 (risks) (both p < 0.05), indicating strong alignment with expert consensus. Junior experts showed lower correlations (benefits rs = 0.89; risks rs = 0.56) than seniors (benefits rs = 0.97; risks rs = 0.78), highlighting the system's role in standardizing assessments. CONCLUSION:This quantitative index provides an evidence-based tool for pediatric off-label drug decisions, supporting clinicians, policymakers, and standardization efforts.

Evidence-based medicine (EBM), formalized in the 1990s, has redefined clinical practice by advocating the integration of research evidence, clinical expertise, and patient values. This paradigm has introduced methodological rigor through randomized controlled trials (RCTs) to establish causality, systematic reviews to synthesize findings, and the GRADE approach to evaluate evidence based on risk of bias, inconsistency, indirectness, imprecision, and publication bias. These advancements have shaped clinical guidelines, reduced practice variability, and influenced medical education toward evidence-based inquiry. Despite its contributions, EBM faces challenges in the evolving landscape of modern medicine. The lengthy process of evidence generation, often requiring years for trials and guideline updates, limits responsiveness to emerging health needs, as observed during the COVID-19 pandemic. The external validity of RCT results is constrained by strict inclusion criteria, posing difficulties in applying findings to diverse patient populations with comorbidities. Additionally, the siloed nature of evidence complicates comprehensive care for multifactorial conditions, while the annual influx of over one million medical publications overwhelms traditional synthesis methods. Artificial intelligence (AI) presents a promising avenue to address these issues, leveraging capabilities in processing heterogeneous data. Natural language processing may enhance literature analysis, machine learning could identify patterns in complex datasets, and causal inference might improve the reliability of observational data insights. These technologies hold potential to accelerate evidence development and tailor it to individual needs. This paper proposes digital intelligent evidence-based medicine (i-EBM) as a conceptual evolution of EBM, designed for the AI era. i-EBM envisions a three-layered framework. The data foundation layer aims to integrate structured evidence from RCTs, domain knowledge such as biomedical ontologies and traditional Chinese medicine principles, and multi- modal patient data, including electronic health records, genomics, and wearable device outputs. Knowledge graphs are proposed to link these elements into a unified, computable knowledge network. The intelligent processing layer seeks to apply AI for evidence retrieval, data extraction, quality assessment, and synthesis, potentially using large language models to assist these processes. The knowledge service layer intends to provide dynamic guidelines and individualized predictions, supported by ongoing human-machine collaboration to ensure clinical relevance and ethical considerations. i-EBM has the potential to mitigate EBM's limitations by facilitating real-time evidence updates, reducing knowledge fragmentation through integrated data, and offering personalized decision support. For instance, it may support precision medicine by connecting diverse data sources, with applications possibly extending to fields like oncology or traditional Chinese medicine. Future research could explore autonomous AI systems, optimized clinical workflows, and governance frameworks to address data privacy, bias, and global standardization. In conclusion, i-EBM offers a theoretical framework to extend EBM principles, harnessing AI's potential alongside human expertise to advance medical research and practice. Meanwhile, for issues such as the quantitative study of the complex intervention characteristics and syndrome differentiation patterns of traditional Chinese medicine, i-EBM can provide methodological support in data integration, pattern recognition, and causal inference, offering potential tools and insights for uncovering the intrinsic regularities of TCM evidence and optimizing its evaluative framework.

Background:Coronary artery disease (CAD) remains a leading cause of global mortality and disability. CAD patients face tradeoffs between antithrombotic therapy benefits and bleeding risks, underscoring the need to incorporate patient values and preferences into clinical guidelines. Establishing minimal important differences (MIDs) for patient-important outcomes supports clinical guideline development by determining the smallest change in outcomes that patients consider important. However, directly conducting patient surveys to establish MIDs presents several methodological challenges. Methods:We established a multidisciplinary working group to guide the MID investigation. Using a three-phase process, we identified key outcomes through literature review and discussion. We will develop draft health outcome descriptions by synthesizing evidence from clinical guidelines and qualitative studies, supplemented with patient interviews, and refine the drafts through iterative cognitive interviews. We then designed outcome-specific draft MID questionnaires and will employ cognitive interviews to assess clarity and comprehensibility. Discussion:This study will develop standard materials for surveying patient values and determining MIDs in Chinese CAD patients. The resulting methodology will support future investigations into patient-important outcomes and provide critical evidence for clinical guideline development.

Clinical practice guidelines contain substantial diagnostic and therapeutic process knowledge to guide clinical decisions. However, since this knowledge is often expressed in natural language, it is difficult for computers to directly understand and execute, which limits its intelligent application. To address this issue, this study proposes a systematic framework to achieve the computability of guideline process knowledge. The framework covers the full technical path from process concept standardization, ontology modeling, sentence recognition, to process visualization. The study first clarifies the standardized definition of diagnostic and therapeutic processes and their core elements, constructing a unified semantic framework. Next, based on ontology engineering methods and business process model and notation process models, a computable and inferable diagnostic process ontology was built. Then, combining manually annotated corpus with large language model prompt engineering strategies, high-precision automatic recognition and key element extraction of diagnostic process sentences were achieved. Finally, process modeling and visualization techniques were used to transform the extracted knowledge into structured graphical representations, providing intelligent knowledge graph support for clinical decision support systems. This study provides a reproducible technical foundation for the structured expression and intelligent application of clinical practice guideline diagnostic and therapeutic process knowledge.

With the acceleration of population aging, multimorbidity has become a critical challenge urgently requiring solutions in the global healthcare field. As a core tool for standardizing clinical practices and improving medical quality, clinical practice guidelines have demonstrated significant value in clinical settings. The number of current multimorbidity guidelines is relatively limited, and their development, evaluation, and implementation still face many challenges. These barriers include insufficient direct evidence, obstacles to multidisciplinary collaboration, and inadequate clinical applicability. The rapid development of digital health technologies provides a comprehensive solution for optimizing multimorbidity guidelines. this study explores how digital health technologies—such as big data, artificial intelligence, and digital platforms—can assist researchers in optimizing the entire process from the development to implementation of multimorbidity guidelines, providing new pathways to overcome bottlenecks in multimorbidity management and enhance the practical value of guidelines.

Objective To understand the perceptions and needs of target users of the public-facing diabetes knowledge platform, and to provide support and guidance for the construction and optimization of the platform. Methods Using convenience sampling and employing a descriptive qualitative research methodology, semi-structured individual interviews and focus group discussions were conducted with the platform's target users. Data analysis and thematic extraction were performed according to Colaizzi's seven-step analysis method. Results In-depth interviews were conducted with 27 participants, and four major themes were elicited: deficiencies in the diabetes platform, the design of platform functionalities, the frequency of platform updates, and the prospective role of the platform. Conclusion The study clarified that the platform should have the key functions of dynamic updating of recommendations and efficient and convenient retrieval. The construction of the platform will facilitate the dissemination and implementation of living guidelines, enhance the convenience of accessing recommendation, satisfy users’ information needs, promote the scientific and standardized approach to the diagnosis and treatment of diabetes. Moreover, it holds significant importance in improving the comprehensive level of diabetes management.

Objective To investigate the current status of evidence grading and/or recommendation classification in clinical practice guidelines for type 2 diabetes mellitus (T2DM) in China, and to provide references for improving the methodological quality of domestic diabetes guidelines.Methods The publicly released clinical practice guidelines/consensus(hereofter referred to as the guidelines) for adult T2DM in China published in domestic and international journals from January 1, 2010 to July 1, 2025 were systematically searched to screen the literature and extract data. Descriptive analysis was performed using Excel 2024, and visualized charts were generated with Origin 2024.Results A total of 118 Chinese T2DM guidelines were included, of which only 49 (41.5%) reported evidence grading and/or classes of recommendation, and 7 classification methods were summarized. The Grading of Recommendations Assessment, Development and Evaluation (GRADE) system was the most widely applied tool. Evidence grading was mostly presented by letters and four-level classification, while recommendation classification was mainly described in text with two-level classification. Ten guidelines (8.5%) adopted the GRADE system for both evidence and recommendation grading, covering 382 recommendations, of which 337 (88.2%) reported both evidence grading and recommendation grading.Conclusion The number of Chinese T2DM guidelines shows an increasing trend, and the application frequency of the GRADE system has gradually increased. Nevertheless, it is necessary to consolidate the evidence base and standardize classification criteria of guidelines, and to facilitate clinical decision-making and promote the standardized and scientific development of domestic guidelines.

Abstract Background Myocardial infarction (MI) remains a major global health burden, with percutaneous coronary intervention (PCI) as the primary revascularization strategy. However, complications such as no‐reflow, reperfusion injury and in‐stent restenosis persist, requiring adjunctive therapies. Traditional Chinese Medicine (TCM), especially blood‐activating and stasis‐resolving Chinese patent medicine (CPM), is increasingly used in MI management. Objective To develop a living clinical practice guideline for the perioperative use of CPMs in the perioperative period of PCI for MI, based on rigorous and transparent methodology. Methods The guideline followed the WHO Handbook for Guideline Development, methods for developing integrated Chinese–Western medicine guidelines, and the living guideline framework. It will be reported in accordance with the RIGHT‐TCM extension. Clinical questions and outcomes were framed, selected and rated through structured discussions with the Guideline Steering Committee and the Guideline Development Group. A systematic review and Bayesian network meta‐analysis will evaluate the efficacy and safety of CPM. The certainty of evidence will be assessed using the GRADE approach, and recommendations will be formulated via the Evidence‐to‐Decision framework, incorporating benefits, harms, evidence certainty, feasibility and costs. When new clinical studies are included and critical outcomes' combined effect demonstrates directional or magnitude changes, the consensus and guideline development groups will jointly decide on new updates. Conclusion By addressing key clinical uncertainties and leveraging real‐time evidence updates, the guideline aims to enhance patient outcomes and support clinicians in decision‐making. Trial Registration: Guideline Registration Number: PREPARE‐2024CN596

In recent years, artificial intelligence (AI) or machine learning (ML) methods have been increasingly used in the development and evaluation of clinical prediction models. Their algorithms differ from traditional regression modeling methods, resulting in significant limitations of the existing Prediction Model Risk of Bias Assessment Tool (PROBAST) for their evaluation. To address these limitations, the tool is updated to PROBAST+AI in 2025. PROBAST+AI extends the original framework to specifically address methodological challenges unique to developing or evaluating clinical prediction models based on AI/ML algorithms. PROBAST+AI assesses the quality of model development and the risk of bias in model evaluation across 4 domains: participants and data sources, predictors, outcome, and analysis, encompassing 16 and 18 signaling questions, respectively. Furthermore, the tool evaluates the applicability of the model across 3 domains: participants and data sources, predictors, and outcome. This article aims to compare the changes between the original and updated versions of the PROBAST tool, interpret the key content and items of PROBAST+AI, and apply the updated tool to evaluate an example clinical prediction model publication, to help domestic systematic review authors, clinicians, and policymakers critically appraise studies that develop or evaluate prediction models based on traditional or AI/ML methods.

OBJECTIVE:The purpose of this study is to validate the taxonomy and framework using Chinese guidelines and identify actionable statements. DESIGN AND SETTING:We searched five databases, to identify the health guidelines from 1 January 2020 to 1 May 2023. Five researchers categorised statements into six types: formal recommendations (Type I) with clear direction and strength, with explicit and direct evidence; good practice statements (GPS) (Type II), actionable in isolation with a significant benefit; remarks (Type III), an inseparable unit belonging to a formal recommendation or GPS that provides additional clarification; research only recommendations (Type IV) for specific populations; implementation considerations, tools and tips (Type V), that describe the how, who, where, what and when, in relation to implementing a recommendation and lacking a direct evidence link; and informal recommendations (Type VI), unrelated to evidence and not meeting GPS criteria. RESULTS:We included 116 guidelines, including 74 Western medicine guidelines, 12 traditional Chinese medicine guidelines and 30 integrated Chinese and Western medicine guidelines. 99 guidelines (85.3%) used the Grading of Recommendations Assessment, Development and Evaluation criteria. Medical specialty societies developed the highest number of guidelines (53.4%). Of all the statements, 4422 statements were extracted from the guidelines. Among them, 2154 (48.7%) were formal recommendations, 197 (4.4%) were GPS, 394 (8.9%) were remarks, 16 (0.4%) were research only recommendations, 1106 (25.0%) were implementation considerations, tools and tips, and 555 (12.6%) were informal recommendations. CONCLUSIONS:Up to date, the Chinese guideline developers tend to overestimate the number of formal recommendations and underestimate the number of GPS, remarks, research only recommendations, implementation considerations, tools and tips, and informal recommendations. Thus the current quality of actionable statements in Chinese health guidelines requires further enhancement.

OBJECTIVE:To provide up-to-date evidence on key benefits, harms, and uncertainties regarding medications for adults with type 2 diabetes. DESIGN:Living systematic review and network meta-analysis (NMA), using frequentist random effects and GRADE (grading of recommendations, assessment, development and evaluation) approaches. Updates are planned at least two times a year. DATA SOURCES:Medline and Embase, searched up to 31 July 2024 for the current iteration. STUDY SELECTION:Randomised controlled trials of at least 24 weeks comparing one or more medications with standard treatment, placebo, or each other. RESULTS:The systematic review and NMA includes 493 168 participants from 869 trials (adding 53 trials since October 2022) reporting data for 13 drug classes (63 drugs) and 26 outcomes of interest. Regarding benefits, moderate to high certainty evidence confirms the well established cardiovascular and kidney benefits of sodium-glucose cotransporter-2 (SGLT-2) inhibitors, glucagon-like peptide-1 receptor agonists (GLP-1RAs), and finerenone (the last for patients with established chronic kidney disease). The most effective drugs in reducing body weight were tirzepatide (mean difference (MD) -8.63 kg (95% confidence interval -9.34 to -7.93); moderate certainty) and orforglipron (MD -7.87 kg (-10.24 to -5.50); low certainty), followed by eight other GLP-1RAs (high to moderate certainty). Absolute benefits of medications vary substantially depending on the baseline risk of cardiovascular and kidney outcomes; risk-stratified absolute effects of medications are summarised using an interactive multiple comparisons tool (https://matchit.magicevidence.org/250709dist-diabetes/#!/). Regarding medication-specific harms, SGLT-2 inhibitors increase genital infections (odds ratio (OR) 3.29 (95% CI 2.88 to 3.77); high certainty) and ketoacidosis due to diabetes (OR 2.08 (1.45 to 2.99); high certainty), and probably increase amputations (OR 1.27 (1.01 to 1.61); moderate certainty); tirzepatide and GLP-1RAs probably increase severe gastrointestinal events (most increased risk with tirzepatide (OR 4.21 (1.87 to 9.49); moderate certainty)); finerenone increases severe hyperkalaemia (OR 5.92 (3.02 to 11.62); high certainty); and thiazolidinediones increase major osteoporotic fractures and probably increase hospitalisation for heart failure. Sulfonylureas, insulin, and dipeptidyl peptidase-4 inhibitors probably increase the risk of severe hypoglycaemia. There is low to very low certainty evidence for effects on other diabetes-related complications, including neuropathy and visual impairment. Despite interest in the issue, there is uncertainty about whether GLP-1RAs may reduce dementia (OR 0.92 (0.83 to 1.02); low certainty). CONCLUSIONS:This living systematic review provides a comprehensive summary of the cardiovascular, kidney, and weight loss benefits, as well as medication-specific harms of medications for adults with type 2 diabetes, including effects of SGLT-2 inhibitors, GLP-1RAs, finerenone and tirzepatide. SYSTEMATIC REVIEW REGISTRATION:PROSPERO number: CRD42022325948. A more detailed protocol is available at https://data.aliveevidence.org/records/q02rv-km486. READERS' NOTES:This article is the first version of a living systematic review. It is linked to a living BMJ Rapid Recommendation and other living clinical practice guidelines, presenting risk stratified recommendations for patients with type 2 diabetes at lower, moderate, and higher risk of cardiovascular and kidney complications. The latest evidence will be made available via the BMJ Rapid Recommendation and via an interactive GRADE evidence summary (MATCH-IT: https://matchit.magicevidence.org/250709dist-diabetes/#!/). Major updates will be published in The BMJ.