AIM:Health system resilience (HSR) has gained prominence in response to acute shocks and chronic stressors, particularly following the COVID-19 pandemic. This study aimed to analyze the conceptualization, frameworks, and operationalization of HSR in empirical research. METHODS:We searched PubMed, Web of Science, and Global Health databases from inception to January 26, 2024 to identify empirical HSR studies. We screened studies independently, and systematically extracted data. Analyzed were conducted across study characteristics, shocks/stressors types, conceptualizations/definitions, framework traditions, and methodological approaches. RESULTS:A total of 125 empirical studies were included, with a marked increase from 2020 onward. Most studies examined acute shocks (84%), with the largest share in the Europe (23.3%) and Africa (16.8%). HSR was predominantly conceptualized as a system capacity to absorb and adapt to shocks, while learning and transformation were less frequently operationalized. Health system-specific frameworks (e.g., Kruk; Blanchet) were most commonly used, primarily as analytical lenses rather than measurement tools. Qualitative methods predominated, although mixed-methods approaches are emerging. Evidence on continuity of essential functions and everyday resilience remained limited. CONCLUSIONS:Despite growing conceptual sophistication, empirical HSR research remains constrained by fragmented frameworks use and limited operationalization. Advancing the field requires clearer conceptual boundaries, improved methodological integration, and greater attention to how resilience is enacted in routine system functioning.

Network meta-analysis (NMA) is a key tool for comparing multiple treatments. However, its reliability can be compromised by inconsistencies in methodology and reporting. We conducted this meta-research study to assess the characteristics and quality of NMAs published between 2019 and 2023 in BMJ, Lancet, JAMA, NEJM, and Annals of Internal Medicine, to identify limitations and propose improvement strategies. We sampled and evaluated NMAs from the target journals using a 36-item checklist. Reviewers independently extracted data and assessed methodological and reporting quality. We compared our sample’s compliance with NMAs from specific domains (Anesthesia, Cancer, Acupuncture, and Cochrane reviews) using odds ratios (ORs) with 95

Large language models (LLMs) are increasingly being explored for patient education, but evidence from patient-facing clinical studies involving real patients remains limited. We conducted a mixed-methods systematic review of LLM-supported patient education in clinical settings, focusing on patterns of use, patient perspectives, and patient-reported outcomes. Searches of PubMed, Embase, Scopus, Web of Science, and CINAHL identified 48 eligible studies from 17 countries. LLMs were mainly used to answer patient questions, generate educational materials, and simplify medical information. Patients generally valued LLM-supported education for clarity, accessibility, and usefulness, but raised concerns about accuracy, privacy, personalization, trust, and emotional support. Quantitative findings were most consistent for cognitive and information-related outcomes, whereas evidence for satisfaction, trust, empathy, preference, behavioral outcomes, and longer-term effects was mixed or limited. Current evidence supports the use of LLMs as adjuncts to patient education, particularly for improving access to and understanding of medical information, while clinician–patient communication remains essential for personalized explanation, trust building, and shared decision-making.

OBJECTIVES:The application of large language models (LLMs) to systematic review tasks is rapidly expanding, yet the transparency and methodological rigor of these evaluations remain unclear. We aimed to assess reporting transparency, methodological quality, and how authors frame claims and caveats in studies applying LLMs to systematic review tasks. STUDY DESIGN AND SETTING:We conducted a cross-sectional meta-epidemiological study by searching PubMed, Embase, Web of Science Core Collection, IEEE Xplore, and five other databases from inception to December 1, 2025, for peer-reviewed articles and preprints. We included empirical studies evaluating generative transformer-based LLMs (eg, Gemini) for core systematic review tasks (eg, screening, data extraction) against a reference standard. We assessed reporting transparency using an adapted Chatbot Assessment Reporting Tool and methodological quality using an adapted Quality Assessment of Diagnostic Accuracy Studies 2 tool. We also analyzed the frequency and strength of claims and caveats mentioned by the authors. The study is registered with the Open Science Framework (https://osf.io/8edhb). RESULTS:We identified and included 229 studies comprising 440 empirical tasks. Reporting transparency was moderate, with a mean item score of 0.52 (standard deviation 0.30) on a 0-1 scale, where higher values indicate more complete reporting. We observed substantial gaps in reproducibility-essential domains, including protocol information (mean score 0.12) and model details (0.30). Although 60.6% of assessments were rated as having a low risk, key safeguards against overfitting and data leakage were rarely reported; for example, locking the test set before prompt optimization, a basic protection against information leakage, was not reported in 99.8% of tasks. We identified 837 claims and 693 caveats. Authors framed claims weakly more often than strongly (66.8% vs 33.2%). Performance superiority over a comparator was the most common claim (64.8% of tasks). Readiness for practical use was claimed in 47.5% of tasks, almost always in qualified terms (93.3%). CONCLUSION:Studies applying LLMs to systematic review tasks are reported with moderate transparency but often omit reproducibility-critical details necessary to assess leakage and overfitting. Although authors frequently make claims about performance and practice readiness, these are typically expressed cautiously. Improved reporting standards and clearer safeguards are urgently needed before routine use of LLMs in evidence synthesis can be recommended. PLAIN LANGUAGE SUMMARY:LLMs, such as ChatGPT, are increasingly used to help carry out parts of systematic reviews, which summarize evidence to inform healthcare decisions. We examined 229 studies that tested LLMs on tasks such as screening articles, extracting data, and assessing study quality, covering 440 evaluations in total. On average, these studies reported their methods with moderate clarity, but often omitted information needed to repeat the work or judge whether the results were trustworthy. Notably, almost none described safeguards to ensure that the test data had not already influenced how the model was set up, a key step for avoiding overly optimistic results. Authors frequently described LLMs as performing well and nearly ready for practical use, though usually in cautious terms. Clearer reporting standards and stronger safeguards are needed before LLMs can be routinely relied upon in evidence synthesis.

Anti-infective agents are widely used during pregnancy to prevent, treat, and mitigate the transmission of infections, and their utilisation has significantly increased over the past decade. However, concerns persist regarding their potential embryotoxicity, compounded by inconsistencies in safety data that complicate medication decision-making. This protocol details an umbrella review that aims to provide a comprehensive and methodologically sound synthesis of evidence from systematic reviews (SRs) on the adverse effects of prenatal anti-infective exposure on offspring. The review will incorporate a standardised, three-stage data extraction strategy and an R-based algorithm to efficiently identify and manage overlapping evidence. We will conduct a comprehensive search across six databases (PubMed, Embase, Scopus, Web of Science, Cochrane Library, and Epistemonikos) from inception to June 22, 2024. The search will cover three key concepts—pregnant women, anti-infective agents, and systematic reviews—without restrictions on controls or outcomes. We will include SRs, with or without meta-analyses, that evaluate the adverse effects of prenatal anti-infective exposure and report either neonatal outcomes (e.g. birth defects) or long-term offspring outcomes (e.g. neurodevelopmental disorders). To address overlapping evidence, a structured three-stage data extraction process, supported by an R-based algorithm, will be implemented, sequentially handling PICO definition extraction, overlapping evidence management, and quantitative data extraction. At least two independent reviewers will screen studies, extract data, and assess methodological quality and evidence strength using AMSTAR-2 and GRADE, respectively. Evidence synthesis will be conducted narratively, drawing on meta-analytic associations from direct comparative analyses and organising findings by anti-infective agents, indications, outcomes, effect estimates, certainty of evidence, and methodological quality. Anti-infectives with safety concerns will be summarised in tables and evidence maps. By adopting a standardised process that incorporates an automated algorithm for managing overlapping evidence, the planned umbrella review will provide a comprehensive synthesis of existing evidence on the adverse effects of anti-infective use during pregnancy. The findings are expected to inform clinical practice, guide policy decisions, and support infection management in pregnant patients. Furthermore, the review will identify key knowledge gaps, help prioritise future research, and ultimately contribute to improving maternal and offspring health outcomes. PROSPERO CRD42024577013

AIMS:Pharmacologic treatment is widely used for managing gestational diabetes mellitus (GDM). However, evidence remains limited on the comparative effectiveness and safety across different drug classes for GDM. This study aims to compare the efficacy and safety of pharmacologic treatments for GDM using network meta-analysis of randomised controlled trials (RCTs). MATERIALS AND METHODS:We searched PubMed, Embase, Cochrane Central Register of Controlled Trials, and Web of Science up to May 16, 2025 to identify trials evaluating glucose levels and maternal or neonatal complications in GDM patients using any pharmacologic agents, e.g., insulin, biguanides, sulfonylureas, α-glycosidase inhibitors, sodium-glucose cotransporter-2(SGLT-2) inhibitors, dipeptidyl peptidase IV(DPP-IV) inhibitors or Glucagon-like peptide-1receptor (GLP-1 RAs) versus standard care. We used random-effects model for both pairwise meta-analyses and frequentist network meta-analyses and applied the Confidence in Network Meta-Analysis (CINeMA) framework to assess the certainty of evidence. RESULTS:Seventy-three articles were eligible, comprising 71 trials in which 14 877 participants were enrolled and seven drug classes assessed. All subsequent effects refer to comparisons with standard care. Because of their moderate to high evidence of certainty, sulfonylureas were established as the most effective drugs used for lowering fasting glucose (mean difference [MD]: -0.33 mmol/L; 95% confidence interval [CI]: -0.55 to -0.1), followed by insulin (MD: -0.3 mmol/L; 95% CI: -0.4 to -0.21) and biguanides (MD: -0.2 mmol/L; 95% CI: -0.28 to -0.12). Biguanides were the most effective drugs used to control haemoglobin A1c (MD: -0.1%; 95% CI: -0.16 to -0.03), but their use was associated with increased adverse events leading to low birth weight among infants (OR: 2.04; 95% CI: 1.04-4.01). Insulin (OR: 0.51; 95% CI: 0.34-0.75), biguanides (OR: 0.39; 95% CI: 0.26-0.59), and sulfonylureas (OR: 0.5; 95% CI: 0.31-0.79) use could decrease the risk of macrosomia. Sulfonylureas were found to be more easily get premature delivery and neonatal hypoglycaemia than biguanides; evidence regarding their impact on low birth weight and long-term safety remains lacking. CONCLUSION:Insulin use provides significant benefits in achieving glycaemic control and minimising maternal and foetal complications, and it has a favourable overall safety profile. Biguanide use may be associated with an increased risk of low birth weight, warranting careful consideration and thorough counselling for shared decision-making. Currently, insufficient evidence supporting the efficacy and safety of sulfonylureas, α-glycosidase, DPP-IV, and SGLT-2 inhibitors; and GLP-1 RAs in GDM management is available.

INTRODUCTION:Patients are increasingly recognized as key interest holders in health care decision‑making. Integrating patient perspective is crucial to patient‑centered, evidence‑based nutrition guideline recommendations. OBJECTIVES:Our aim was to examine individual willingness to decrease consumption of foods high in saturated fatty acids (SFAs) after being informed about the estimated absolute risk reduction (ARR) of myocardial infarction (MI), together with the certainty of the supporting evidence based on a Cochrane systematic review. PATIENTS AND METHODS:Respondents from 7 countries were presented with a conversation aid with the ARR of MI, together with the certainty of evidence, and asked about their willingness to reduce their intake of SFA‑rich foods. Using a multivariable logistic regression model, we explored 12 variables to identify factors underlying respondent willingness to introduce dietary fat changes. RESULTS:We analyzed 3663 respondents from Brazil, Canada, China, Croatia, Poland, Spain, and the United States. Overall, 50.2% were willing to reduce their SFA intake. Women (odds ratio [OR], 1.51; 95% CI, 1.29-1.77), nonomnivores (OR, 1.73; 95% CI, 1.37-2.19), and respondents from Spain, China, and Brazil (OR, 2.45; 95% CI, 1.83-3.27; OR, 4.36; 95% CI, 3.49-5.45; and OR, 1.31; 95% CI, 1.08-1.58, respectively) showed greater willingness, whereas those from Croatia or holding a university degree showed lower willingness (OR, 0.7; 95% CI, 0.51-0.95; OR, 0.73; 95% CI, 0.61-0.87, respectively). CONCLUSIONS:The significant variation in people willingness (depending, eg, on sex, education levels, or dietary pattern) to reduce their intake of SFA‑rich foods underscores potential importance of incorporating patient preference when developing dietary guideline recommendations, and of promoting individualized evidence‑based care using absolute effects and certainty of evidence.

ObjectiveTo explore the feasibility of using Large Language Models (LLMs) to generate evidence-based health science popularization materials by surveying public needs, designing specific prompts, and evaluating the quality of the generated content.MethodsAn online survey was conducted via SoJump platform to investigate public needs for healthy lifestyle information. The Kimi LLM was employed to generate health science popularization materials, with prompts optimized based on literature review and expert feedback. Two health educators independently evaluated the generated materials using the DISCERN instrument. Data were analyzed using SPSS 24.0 software.ResultsThe primary barriers hindering residents from accessing and using health information were distrust of publishing institutions (71.37%), excessive use of technical terminology (64.71%), and inaccessible channels (60.00%). The designed prompts, consisting of system and user prompts, comprised three sections: role definition, content generation, and structure/output formatting specifications. These prompts effectively guided the LLM to produce accessible health science materials with a standardized structure, including a title, introduction, specific interventions, and summary. Following prompt optimization, the DISCERN scores of the generated materials increased by (4.40±2.51) points, representing a statistically significant difference (P=0.041).ConclusionsHealth science popularization is a crucial factor in promoting the acquisition and utilization of health information among residents. Developing prompts to assist LLMs in creating health science materials holds significant potential; it facilitates health education work and enhances both the scientific accuracy and efficiency of content production.

Background: Artificial intelligence-based clinical decision support systems (AI-CDSS) have demonstrated high diagnostic and predictive performance, however, their impact in pediatric care remains uncertain. Therefore, this study aimed to evaluate whether AI-CDSS improve effectiveness and diagnostic or predictive performance in pediatric care and, to examine methodological development, and barriers and facilitators influencing implementation. Methods: PubMed, Cochrane Library, and Web of Science were searched from inception to October, 2025 for studies related to AI-CDSS targeting pediatric populations and reported on efficacy, development, or implementation factors. Given anticipated heterogeneity, a structured narrative synthesis was conducted. Findings: Sixty-one studies were included, with 80% published since 2020. 44 studies reported strong model performance, with AUC values typically ranging from 0·75-0·95 across diverse pediatric conditions, particularly in imaging-based diagnosis, acute care risk stratification, and chronic disease management. Studies evaluating genomic and screening applications also reported increased diagnostic yield and reductions in unnecessary diagnostic examinations. Seven prospective studies assessed clinical effectiveness and reported improvements in care processes, including reduced clinician workload, more appropriate diagnostic testing, enhanced antimicrobial stewardship, and greater standardization of care. Qualitative and mixed-methods evidence identified barriers including limited interpretability, workflow misalignment, alert fatigue, and variable clinician trust, while facilitators included usability, perceived benefit, and organizational support. Across development studies, internal validation predominated, whereas external validation and calibration assessment were infrequently reported. Interpretation: AI-CDSS in pediatric care demonstrate strong diagnostic and predictive performance, but evidence for consistent improvement in patient-centered outcomes remains limited. This performance-effectiveness paradox appears to be driven not only by insufficient external validation and vulnerability to dataset shift, but also by implementation challenges including limited interpretability, workflow misalignment, and variable clinician uptake. Future progress will depend on shifting from model-centric development toward implementation-ready systems that prioritize generalizability, human-centered design, and prospective evaluation to achieve meaningful improvements in pediatric health outcomes.

Background:Annual influenza vaccination is recommended for solid organ transplant (SOT) recipients, with various strategies explored to enhance efficacy. Given the ongoing uncertainty about the benefits and potential risks of different strategies in influenza vaccination in SOT recipients, comparing the indirect comparisons of the available evidence is necessary. Methods:We searched MEDLINE, Embase, CENTRAL, and CINAHL from inception to March 2025 to identify randomized controlled trials that compared different influenza vaccinations in SOT recipients. We conducted a network meta-analysis and used the GRADE approach to assess the certainty of evidence, categorize interventions, and present the findings. The protocol for this systematic review was registered in PROSPERO (CRD42024537277). Results:A total of 13 articles met our eligibility criteria (with 2,298 participants) and assessed 10 vaccination strategies. The high-dose (60IMSD) influenza vaccine was superior to the standard influenza vaccination regarding the seroconversion rate for H1N1 (RR 2.27, 95% CI 1.52 to 3.39; high certainty) and H3N2 (RR 1.61, 95% CI 1.34 to 1.94; high certainty). All of the included vaccination strategies might have no difference in graft rejection and other adverse events compared with the standard vaccination. Conclusions:We found that the 60IMSD vaccine demonstrated better immunogenicity outcomes compared with all other influenza vaccines studied in SOT recipients. Additionally, we observed little or no difference among various influenza vaccination strategies in terms of graft rejection and other serious adverse events (SAEs) or non-serious adverse events (non-SAEs).

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.

Network meta-analysis (NMA) plays an important role in comparative effectiveness research, particularly in fields such as traditional and complementary medicine, where multiple interventions often need to be assessed within a single evidence framework. However, conducting NMA remains labor-intensive, methodologically demanding, and difficult to complete efficiently using conventional workflows. Although recent advances in large language models have created new opportunities for supporting evidence synthesis, their routine use in NMA is still constrained by limited transparency, prompt dependency, and difficulty integrating with structured analytical procedures. In this study, we introduce SmartEBM, a web-based human-AI collaborative platform designed to support the full workflow of NMA. SmartEBM provides an integrated environment for title and abstract screening, full-text screening, data extraction, risk of bias assessment, statistical analysis, and certainty of evidence assessment. The platform is organized around a human-in-the-loop model, in which AI-assisted functions support repeated and labor-intensive tasks while researchers retain oversight of methodological judgement, verification, and final decision-making. Through its six functional modules, SmartEBM offers low-code interfaces, structured outputs, and verification-oriented workspaces that help connect major steps of evidence synthesis within one platform. Rather than functioning as a stand-alone automation tool, SmartEBM is intended as a practical platform for end-to-end NMA support. This platform-oriented approach may help make evidence synthesis more manageable, traceable, and accessible in routine research practice, especially in complex review settings.

Gram-negative bacteria (GNB) infection therapy poses a significant challenge in the field of infectious diseases due to their extensive resistance to clinically advanced antibiotics such as carbapenems and third-generation cephalosporins, and novel therapeutic options are urgently needed. Unlike single-target antibiotics, natural products derived from herbs exert therapeutic effects through synergistic mechanisms, including bacterial growth inhibition, modulation of inflammatory pathways, and regulation of microbial balance and host immune function. In this review, we systematically summarized natural products that have demonstrated promising efficacy against drug-resistant GNB. We highlighted several widely used natural product formulations with validated therapeutic outcomes through rigorous clinical trials, including Tanreqing injection (TRQ) and Xuebijing injection (XBJ). The active constituents of these natural products and their mechanisms of action are thoroughly discussed, providing valuable insights into the potential of natural products for developing combinational anti-drug-resistant GNB agents. Moreover, an overview of opportunities and challenges in achieving standardized quality control, multi-target/antibiotic synergy, exploration of molecular mechanisms of action, and clinical practice transformation. Finally, this review summarizes the key challenges in the clinical translation of natural products and emphasizes the importance of advancing multi-target/antibiotic synergistic therapies, establishing standardized quality control protocols, and elucidating molecular mechanisms of action.

Background: Integrated Chinese-Western medicine (ICWM) is a distinctive medical system that plays an important role in healthcare and has received increasing attention in recent years. To facilitate the dissemination of evidence in ICWM, we developed an Artificial Intelligence (AI)-empowered Clinical Evidence for Integrated Chinese-Western Medicine (ACE-iMed) platform. Methods: A multidisciplinary working group was established, including individuals with professional backgrounds in evidence-based medicine methodology, Chinese medicine (CM), Western medicine (WM), and ICWM clinical practice and research, and computer science. Through multiple rounds of discussions, the working group defined the framework and methodology of the platform, and then applied the platform to summarize evidence for eight diseases. Results: The ACE-iMed platform (website: www.aceimed.org) contains two interfaces. The first enables the developers to store and screen the literature, perform methodological quality assessments, and generate evidence summaries. The AI-empowered workflows showed good consistency and stability across multiple stages, including literature screening and assessment of risk of bias/methodological quality, and effectively support summarizing evidence for eight diseases. The second interface, intended for end users, provides synchronized access to the included literature and the generated summaries, enabling quick access to clinical question-oriented evidence resources. Conclusion: This study introduces an AI-empowered, clinical question-oriented ICWM evidence platform. Application across eight diseases demonstrated the platform's feasibility and practical utility. The platform not only supports the developers in summarizing evidence but also provides end users with a potential pathway to access evidence and its summaries.

Introduction Sepsis, a life-threatening organ dysfunction caused by a dysregulated host response to infection, may benefit from immunomodulatory drugs. Nevertheless, numerous clinical trials of these drugs have failed to demonstrate efficacy, partly due to substantial heterogeneous treatment responses. Subgroup analyses from these trials are frequently employed to investigate different treatment effects across subgroups. However, which drugs might have different effects across subgroups and how credible these findings are have not been well summarised and evaluated. Additionally, the differences in the characteristics and results of subgroup analyses based on whether the primary trial’s main effect is statistically significant remain unclear. We will conduct a systematic review to comprehensively address these questions.Methods and analysis We will include randomised controlled trials (RCTs) evaluating immunomodulatory drugs for adult sepsis and exclude quasi-randomised trials, single-arm studies, animal research, conference abstracts, study protocols and non-English publications. To comprehensively search for subgroup analyses, we will search both RCTs and their published secondary analyses across PubMed, Embase, Web of Science, ClinicalTrials.gov and the Cochrane Library from their inception. Four reviewers will independently screen eligible studies and only one subgroup analysis will be selected for data extraction using standardised forms. The credibility of subgroup effects will be assessed using the Instrument for assessing the Credibility of Effect Modification Analyses. We will analyse the proportion and characteristics of subgroup analyses reported in trials. We will qualitatively summarise the results of subgroup analyses, focusing on findings with a subgroup-specific p value<0.05 or an interaction p value<0.05. If two or more studies examined the same drug within a subgroup, we will perform data synthesis using a random-effects model to estimate pooled subgroup effects. We will compare the differences based on whether the primary trial’s main effect was statistically significant using t-tests, Mann-Whitney U tests, χ2 test or Fisher’s exact test, as appropriate.Ethics and dissemination No ethical approval is required because the data we will use do not include individual patient data. Findings will be disseminated through publication in a peer-reviewed journal.Trial registration number CRD420251089737.

Insomnia poses significant challenges to public health systems. Acupuncture, a widely used traditional therapy around the world, has shown promise in improving sleep outcomes, but clinical practice remains inconsistent. This guideline aims to provide evidence-based recommendations for acupuncture practice in insomnia management while addressing implementation across diverse healthcare settings. A systematic review and meta-analysis of randomized controlled trials (RCTs) was conducted, encompassing studies published before January 2025. Evidence certainty was assessed using the GRADE approach, and recommendations were formulated through a modified Delphi process involving a multidisciplinary panel of 26 experts from 11 countries that comprehensively considered patient preferences, cost-effectiveness, and regional accessibility. Based on moderate-to low-certainty evidence, it is conditionally suggested that electroacupuncture or manual acupuncture may be considered for patients with chronic insomnia particularly in institutions where it is available. Notably, electroacupuncture is conditionally suggested for patients with comorbid insomnia or for patients with poor sleep quality. To standardize clinical application, expert consensus established a core set of acupoints, including Baihui (DU20), Sanyinjiao (SP6), Sishencong (EX-HN1), and Anmian (EX-HN-54), and a treatment protocol involving needle retention for 30 min after obtaining "de qi", administered three times per week for at least four weeks.

Hypertension is a major risk factor for cardiovascular disease. Salt substitutes may reduce sodium intake while maintaining palatability, but comparative effects across formulations remain uncertain. We conducted a systematic review and frequentist random-effects network meta-analysis of randomised controlled trials in adults comparing salt substitutes with regular salt, other substitutes or no intervention. Databases (PubMed, Embase, CENTRAL, CNKI, Wanfang), WHO-ICTRP and ClinicalTrials.gov were searched from inception to Oct 3, 2025 (PROSPERO CRD42023451859). We assessed the risk of bias using a modified Cochrane tool and conducted a random-effects network meta-analysis, with evidence certainty evaluated through the GRADE approach. We included 34 randomised controlled trials involving 37,063 participants across 15 countries (17 from China, 17 from other countries; mean age 62.3 years). Our results indicate that moderate-potassium and low-sodium salt substitutes (25–40