This study projected the incidence, prevalence, death and disability-adjusted life-years (DALYs) attributable to total strokes and pathological types in people aged ≥ 15 years for 204 countries and territories to 2050. Age- and sex-specific trends in rates were developed using XGBoost models incorporating national human development index, gross domestic product per capita, and demographic data from the Global Burden of Disease study, the World Bank, and the United Nations. Uncertainty intervals were calculated as the 2.5th and 97.5th percentiles of the distribution using a bootstrap-like method. From 2021 to 2050, the global absolute number of incident strokes is anticipated to increase by 31.64

BACKGROUND:Oxaliplatin-induced peripheral neuropathy (OIPN) affects up to 80% of patients receiving oxaliplatin-based chemotherapy, and effective preventive strategies remain limited. Huangqi Guizhi Wuwu Decoction (HQGZWWD), a traditional Chinese herbal formula included in China's National Classic Famous Formulas, is widely used for OIPN in clinical practice, but high-quality clinical evidence remains lacking. PURPOSE:This study evaluated the efficacy and safety of HQGZWWD for OIPN prevention and explored its neuroprotective mechanisms. METHODS:We conducted a multicenter, randomized, double-blind, placebo-controlled trial at 12 tertiary hospitals in China. Adults with colorectal cancer scheduled to receive XELOX chemotherapy were randomly assigned (1:1) to oral HQGZWWD granules or placebo (47.5 g per sachet, twice daily) throughout chemotherapy. Participants, outcome assessors, and statisticians were blinded. The primary outcome was the incidence of grade ≥2 chronic OIPN in the modified intention-to-treat population, analyzed using logistic regression with multiple imputation for missing data. RESULTS:A total of 360 participants were enrolled, and 354 were included in the modified intention-to-treat analysis. Grade ≥2 chronic OIPN occurred in 21.3% of patients receiving HQGZWWD group and 38.6% in the placebo group (OR, 0.43; 95% CI, 0.27-0.69; P<0.001). Adverse events were comparable between groups (OR, 1.27; 95% CI, 0.68-2.39; P=0.524). In animal models, HQGZWWD increased pain thresholds, preserved dorsal root ganglia, reduced IL-1β, MCP-1, and MDA levels, and enhanced SOD activity. Consistently, clinical serum enzyme-linked immunosorbent assay (ELISA) analyses further validated these mechanisms, demonstrating that HQGZWWD significantly attenuated chemotherapy-induced elevations of pro-inflammatory cytokines (IL-1β, MCP-1) and oxidative stress marker (MDA), while preserving antioxidant enzyme (SOD) activity in CRC patients receiving XELOX chemotherapy. CONCLUSION:HQGZWWD significantly reduced the incidence of OIPN without increasing toxicity. Preclinical findings suggest that these effects may involve reduced oxidative stress and improved mitophagy. These findings support HQGZWWD as a promising preventive strategy for OIPN. TRIAL REGISTRATION:The trial is registered with ClinicalTrials.gov (NCT04913376).

OBJECTIVE:Identifying stroke patients at different disease stages is a prerequisite for clinical research using electronic medical records (EMRs), whereas an artificial intelligence-based model that can be directly applied remains lacking. We therefore develop a large language model (LLM) pipeline for stroke staging model (StrokeSM) in retrospective clinical research. METHODS:StrokeSM was developed using a Chinese national stroke database comprising EMRs from 33,637 patients. A total of 2000 patients were randomly selected from the Tianjin regional stroke database for external validation. StrokeSM comprised three phases: stroke hospitalization identification based on BERT and a bidirectional cross-attention network to fuse present illness history and discharge diagnosis, symptom-time extraction based on chief complaint through a UIE-base LLM, and stroke staging classification according to the predefined rules. RESULTS:On the test set, StrokeSM achieved accuracy, F1 score, precision, and recall of 0.90, 0.91, 0.91, and 0.90, respectively. The F1 score, precision, and recall of StrokeSM for acute phase was 0.91, 0.89, and 0.93, respectively. On the external validation set, StrokeSM had an accuracy, F1 score, precision, and recall of 0.92, 0.93, 0.93, and 0.92, respectively. Moreover, StrokeSM performed remarkably well in acute phase, with F1 score, precision, and recall of 0.97, 0.98, and 0.96, respectively. CONCLUSIONS:StrokeSM had achieved state-of-the-art performance, providing an accurate method of classifying stroke populations with different disease stages in EMRs, especially in the acute phase. StrokeSM heralds automatic and accurate identification of disease stage phenotypes based on LLM in EMRs, laying the foundation for drawing reliable conclusions in clinical research.

Objective:Lower extremity varicose veins (LEVV) are common chronic venous disorders. Adherence to perioperative self-care plays a vital role in managing postoperative pain and enhancing long-term quality of life (QoL). Traditional health education models often struggle with issues such as incomplete information and high cognitive load. Methods:This study utilized a single-blind, randomized controlled trial design. LEVV patients were randomly assigned to either an intervention group [checklist-enhanced multimedia interactive education (CE-MIE), n = 96] or a control group [multimedia interactive education (MIE), n = 97]. Both groups received standard perioperative care and multimedia educational resources. The intervention group also used a structured checklist for a comprehensive, bidirectional verification of educational content (including ankle pump exercises, limb elevation and discharge instructions) and key skills, with patients required to score >80 points on elastic stocking wearing skills. Primary outcomes included QoL scores, pain levels, and complication rates. Results:The CE-MIE group showed significantly better QoL scores at 1 month postoperatively compared to the MIE group (32.74 ± 4.72 vs. 35.49 ± 4.01, p < 0.001). Additionally, the CE-MIE group reported lower pain scores on the 3rd and 7th postoperative days. However, there were no significant differences in QoL scores between the two groups at the 1-year follow-up, and long-term pain assessment at 1 year was not included in the study design as the primary focus was on acute recovery. Conclusion:CE-MIE is an effective short-term intervention for improving QoL and pain management in LEVV patients. To address the challenge of long-term decay in intervention effectiveness, future studies should explore ways to extend the "in-hospital standardization" model to promote "out-of-hospital sustainability."

Non-randomized studies of interventions (NRSIs) provide important evidence on harms, especially for rare adverse events that randomized controlled trials (RCTs) are often underpowered to detect. Evidence synthesis is therefore needed to integrate findings across study designs and to inform a comprehensive assessment of harms. However, synthesizing evidence from RCTs and NRSIs remains methodologically challenging. We examined how evidence from RCTs and NRSIs is synthesized in practice and how conclusions were drawn when findings conflict. The meta-epidemiological study included systematic reviews indexed in PubMed between 1 January 2017 and 31 December 2024 that synthesized evidence from both RCTs and NRSIs for the same outcome. We evaluated methodological practices across four synthesis scenarios. For reviews that combined RCTs and NRSIs in a meta-analysis, we assessed key methodological components of the review process. For reviews that meta-analyzed RCTs and NRSIs separately, we assessed qualitative agreement between RCTs and NRSIs based on the magnitude, direction, and statistical significance of the estimates. When qualitative disagreement was observed, we further evaluated whether the review conclusions were reasonable, taking into account the certainty of evidence and the heterogeneity of the estimates. Of 42,341 records screened, 195 systematic reviews were included. 49 (25.1

Introduction: The identification of nondiabetic kidney disease (NDKD) in diabetic patients is critically important. Unlike diabetic nephropathy, NDKD often requires additional therapeutic interventions beyond standard diabetes care. There is a need to develop computational methods using electronic medical record data to identify NDKD in diabetic patients for whom kidney biopsy is not an option. Methods: The study included 1,136 diabetic patients who underwent kidney biopsy at a tertiary teaching hospital. We collected 103 parameters from electronic medical records, including demographic characteristics, physical examination results, laboratory tests, and the status of diabetic retinopathy. We developed seven models to detect NDKD, including k nearest neighbors, random forest, extreme gradient boosting (XGB), LASSO logistic regression, support vector machine, naïve Bayes, and multilayer perceptron (MLP), in the training set (n = 908) and compared their performances in the testing set (n = 228). The SHapley Additive exPlanations (SHAP) approach was used to analyze the importance of features. Results: Biopsy-confirmed NDKD was present in 53% of the 1,136 participants. In the testing set, the area under the receiver operating characteristic curve (AUC) for NDKD detection using XGB, LASSO regression, and MLP reached 0.8, with performances that were stable regardless of whether variable normalization was performed. Among them, XGB revealed the highest AUC (0.833; 95% CI: 0.800–0.864) without feature normalization, which was statistically superior to the other models according to DeLong’s tests. After feature normalization, SVM achieved the highest AUC of 0.841 (95% CI: 0.817–0.861) among all models. In addition to established predictive factors for NDKD (e.g., hematuria and absence of diabetic retinopathy), SHAP analysis identified several features, such as low IgG levels, that contributed significantly to the differentiation models. Conclusion: Despite performance variations in different modeling techniques, machine learning models may have the potential to facilitate the detection of NDKD for patients with contraindications for kidney biopsy. Further efforts are warranted to improve accuracy and facilitate their translation into clinical practice.

Importance Randomized clinical trials (RCTs) provide the optimal design for evaluating the effects of Chinese herbal medicine (CHM) on patient outcomes. However, how trialists have designed, conducted, and analyzed CHM RCTs remains largely unknown. Objective To investigate the design, conduct, and analysis of CHM RCTs; to examine whether there are differences between RCTs published in English and Chinese and between higher-impact and lower-impact English journals; and to determine whether CHM RCTs have improved over time. Design, Setting, and Participants In this cross-sectional study, PubMed, EMBASE, Cochrane Central Register of Controlled Trials, and 4 Chinese databases were searched from inception to April 2024. Parallel CHM RCTs published in journals covered in the Journal Citation Reports or Chinese core journals were included. Main Outcomes and Measures The primary outcomes were the general and methodological characteristics of included RCTs published in English vs Chinese, publication year up to vs after 2015, and higher-impact vs lower-impact English journals. To compare characteristics of included RCTs published in different journals, χ 2 or Fisher exact test was use for dichotomous variables, and t test was used for continuous variables when the distribution proved normal or Mann-Whitney U test when it did not. Results The 400 CHM RCTs (200 from Chinese language journals and 200 from English language journals) enrolled 100 to 4870 patients. Most RCTs (370 RCTs [92.5%]) did not specify the study hypothesis; approximately one-third (102 RCTs [30.6%]) were registered. The protocols were available for 15 RCTs (3.8%), and statistical analysis plans were available for 4 RCTs (1.0%). Approximately two-fifths (159 RCTs [39.8%]) reported inadequate methods of allocation sequence generation, and three-fifths (242 RCTs [60.2%]) described inadequate methods of allocation concealment. More than one-third (138 RCTs [34.5%]) explicitly specified a primary outcome, and 115 RCTs (28.8%) reported sample size estimation. Ony 10 RCTs (2.5%) had an independent data monitoring committee. More than two-thirds (254 RCTs [73.5%]) stated reasons for prescribing CHM, most commonly the limited or no effect of Western medicine (215 RCTs [53.8%]) and adverse effects of Western medicine (80 RCTs [20.0%]). Most RCTs did not mention whether there was prior clinical (279 RCTs [69.8%]), pharmacological (201 RCTs [50.2%]), or toxicological (388 RCTs [97.0%]) evidence to support the trial hypotheses. A minority (146 RCTs [36.5%]) specified the prescription of CHM according to traditional Chinese medicine syndrome diagnosis. Most RCTs with missing data conducted only a complete case analysis (70 RCTs [77.8%] for dichotomous outcomes and 79 RCTs [84.0%] for continuous outcomes). A small proportion of RCTs (62 RCTs [15.5%]) used an intention-to-treat analysis, and trialists rarely performed sensitivity analysis (29 RCTs [7.2%]) and subgroup analysis (30 RCTs [7.5%]). The design, conduct, and analysis of CHM RCTs improved over time, and were superior in English-language journals, especially higher-impact English-language journals. Conclusions and Relevance These findings suggest that the conduct and analysis of CHM RCTs are generally suboptimal, highlighting areas that urgently need improvement, including statement of study hypothesis and provision of a protocol; registration of the trial; implementation of allocation concealment; specification of primary outcome and sample size estimation; mention of prior clinical, pharmacological, and toxicological support for the trial hypotheses; and satisfactory conduct of sensitivity analysis or subgroup analysis. Although improvements occurred over time, further enhancing the fundamental research capabilities and developing methodological guidelines remains necessary.

OBJECTIVE:To explore the predictive value of the first modified Nutrition Risk in the Critically Ill Score (mNUTRIC) assessed within 24 hours of admission to the intensive care unit (ICU) for the risk of ICU mortality in patients with sepsis. METHODS:A single-center prospective cohort study was conducted, enrolling septic patients admitted to the department of critical care medicine, the First Affiliated Hospital of Guangxi Medical University, from November 2024 to June 2025. Basic patient information, past medical history, complications, critical care-related scores and initial laboratory parameters obtained within 24 hours of ICU admission, as well as the duration of continuous renal replacement therapy (CRRT), mechanical ventilation and physical restraint during ICU stay were collected. The mNUTRIC score was calculated for each patient. Patients were followed up to record ICU mortality and length of ICU stay. According to the mNUTRIC score, patients were divided into four groups: Q1 group (mNUTRIC score<4 points), Q2 group (4 points≤mNUTRIC score<5 points), Q3 group (5 points≤mNUTRIC score<6 points) and Q4 group (mNUTRIC score≥6 points). The above-mentioned indicators were compared among the four groups. Multivariate Cox regression analysis was used to evaluate the association between the first mNUTRIC score within 24 hours of ICU admission and ICU mortality in septic patients. Restricted cubic spline (RCS) analysis was performed to test the dose-response relationship between mNUTRIC score and ICU mortality. Receiver operator characteristic curve (ROC curve) was plotted, and the area under the curve (AUC) was calculated to assess the predictive efficacy of mNUTRIC score for ICU mortality. Patients were further stratified according to the optimal cut-off value determined by the Youden index. Kaplan-Meier survival curves were drawn to estimate the cumulative ICU survival probability of patients in the two subgroups, and the Log-rank test was used to compare the survival difference between the two groups. Subgroup analysis and interaction test were conducted to evaluate the consistency of the association between mNUTRIC score and ICU mortality across different subgroups. RESULTS:A total of 335 septic patients were enrolled in this study, including 81 cases in Q1 group, 76 cases in Q2 group, 76 cases in Q3 group and 102 cases in Q4 group. The overall ICU mortality rate was 17.6% (59/335). Significant differences were observed among the four groups in terms of age, past medical history (hypertension, diabetes mellitus, coronary heart disease), complications [septic shock, acute kidney injury (AKI), multiple organ dysfunction syndrome (MODS)], Acute Physiology and Chronic Health Evaluation II (APACHE II), Sequential Organ Failure Assessment (SOFA), Charlson Comorbidity Index (CCI), albumin (Alb), aspartate aminotransferase (AST), blood lactic acid (Lac), duration of CRRT and mechanical ventilation, and ICU mortality (all P<0.05). In the multivariate Cox regression model with full adjustment for confounding factors, mNUTRIC score was significantly and non-linearly positively correlated with ICU mortality. Each 1-point increase in mNUTRIC score was associated with an 89% increase in the risk of ICU mortality [hazard ratio (HR)=1.89, 95% confidence interval (95%CI) was 1.52-2.36, P<0.001]. The ROC curve analysis showed that the AUC of mNUTRIC score for predicting ICU mortality was 0.769 (95%CI was 0.711-0.828), with a Youden index of 0.424, an optimal cut-off value of 4.5 points, a sensitivity of 88.1% and a specificity of 54.3%. Based on the optimal cut-off value, the cohort was further divided into the mNUTRIC score<4.5 subgroup and the mNUTRIC score≥4.5 subgroup. The Kaplan-Meier survival curves demonstrated that the cumulative survival probability of patients in the mNUTRIC score≥4.5 subgroup was significantly lower than that in the mNUTRIC score<4.5 subgroup (Log-rank test: χ2=28.540, P<0.000 1), suggesting that mNUTRIC score≥4.5 points was associated with a higher risk of ICU mortality. In all subgroups, mNUTRIC score was robustly and positively correlated with ICU mortality (all HR>1), with no significant interaction observed (all P>0.05), indicating that the positive association between mNUTRIC score and ICU mortality risk in septic patients was consistent across different subgroups. CONCLUSIONS:The mNUTRIC score has good predictive efficacy for ICU mortality risk in patients with sepsis. Timely implementation of nutritional or supportive interventions based on the mNUTRIC score may improve the adverse ICU outcomes of these patients.

BACKGROUND: The increasing adoption of Bayesian methods in clinical trials necessitates robust sensitivity analyses to validate their assumptions, particularly the prior distribution. However, the design, implementation, and interpretation of these analyses are not well characterized. The purpose of this study is to examine the design, conduct and interpret characteristics of sensitivity analyses in Bayesian clinical trials. METHODS: We systematically searched PubMed, Embase, and the Cochrane Central Register of Controlled Trials to identify Bayesian clinical trials published from database inception through December 2024. Two reviewers independently performed study selection and full-text screening. Data regarding sensitivity analyses were extracted in duplicate using standardized, pilot-tested data collection forms. Multivariable logistic regression was employed to examine the association between five variables with more likely reporting sensitivity analyses for prior information. RESULTS: Among the 171 included trials, 103 had an accessible study protocol or statistical analysis plan (SAP), of which 36 (35.0%) trials with available protocols or SAPs failed to pre-specify any sensitivity analyses. Sensitivity analyses were reported in 91 (53.2%) trials, with four reporting at least one sensitivity analysis that contradicted the primary analyses. 38 (22.2%) interpreted the results of sensitivity analysis in the discussion section. Post-hoc sensitivity analyses were explicitly described in 16 trials, only five provided a rationale for these analyses. 24 (23.3%) reported sensitivity analyses not pre-specified in the protocol, 47 (45.6%) omitted pre-specified sensitivity analyses from their results, none of these provided a justification. Of the 122 trials that reported using prior distributions, 33 (27.0%) conducted sensitivity analyses on the priors. Among the 58 trials that used informative priors, 22 (37.9%) performed sensitivity analyses; correspondingly, 11 out of 64 trials (17.2%) with non-informative priors did so. Multivariable analysis indicated that both the use of informative priors and the availability of a protocol or SAP were significantly associated with an increased likelihood of conducting sensitivity analyses for priors. CONCLUSIONS: This study reveals gaps in the design, conduct, and reporting of sensitivity analyses for Bayesian trials, especially concerning priors. There is a pressing need to improve and standardize the use of sensitivity analyses in Bayesian clinical trials.

Importance:Externally controlled trials (ECTs) can serve as an alternative in settings where randomized clinical trials (RCTs) are unfeasible. However, the methodological rigor of ECTs, particularly with regard to bias control, is often inadequately assessed, which can compromise the validity of studies and lead to incorrect decisions. Objective:To examine the design, conduct, and analysis characteristics of current ECTs and to assess whether appropriate methods were used to control bias. Design, Setting, and Participants:This cross-sectional study searched PubMed for ECTs published between January 1, 2010, and December 31, 2023. Eligible ECTs included single-arm trials with an external control or that used a treatment group from an RCT compared with an external control and evaluated the efficacy and/or safety of a drug or medical device. Data analysis was conducted from March 5 to 20, 2025. Main Outcomes and Measures:Extracted information included design characteristics, external control data sources, transparency in covariate selection, statistical methods, and the use of sensitivity and quantitative bias analyses. The characteristics of included ECTs were compared between journals in the top 25% in their Journal Citation Reports category (Q1) and non-Q1. Results:This study included 180 ECTs, of which 85 (47.2%) focused on oncology. Only 64 (35.6%) provided reasons for using external controls, and 29 (16.1%) were prespecified to use external controls. The main sources of external controls were clinical (also termed real-world) data (98 [54.4%]) and trial-derived controls (67 [37.2%]), while concurrent data collection with the treatment arm was relatively infrequent (18 [10.0%]). Only 14 studies (7.8%) conducted feasibility assessments to evaluate the adequacy of data sources, and 13 (7.2%) specified how to handle missing data in external control datasets. Covariate selection procedures were described in 37 of the 164 studies (22.6%) that reported important covariates. Sixty studies (33.3%) used statistical methods to adjust for important covariates when generating the external control, with the propensity score method being the most common (35 of 60 [58.3%]). Among 120 ECTs that generated external controls without statistical methods, 91 (75.8%) used univariate analysis to estimate treatment effects, and only 18 (15.0%) used multivariable regression analysis. Sensitivity analyses for primary outcomes were performed in 32 studies (17.8%), and quantitative bias analyses (2 [1.1%]) were nearly absent. ECTs in Q1 journals were more likely to prespecify the use of external controls (χ21 = 9.86; P = .002) and provided rationales for using external controls (χ21 =  4.33; P = .04). Thirteen recommendations for the careful practice of ECTs are proposed. Conclusions and Relevance:In this cross-sectional study of ECTs, current practices in the design, conduct, and analysis were suboptimal, limiting their reliability and credibility. The study identified several critical methodological issues, such as the lack of justification for using external controls, failure to prespecify external controls in the protocol, insufficient use of confounding adjustment techniques, inadequate sensitivity analyses, and almost complete absence of quantitative bias analyses. Therefore, actionable suggestions for future ECT practices are proposed.

BACKGROUND:Standard random-effects meta-analysis models for rare events exhibit significant limitations, particularly when synthesizing studies with double-zero events. While methodological advances in both frequentist and Bayesian frameworks now offer robust alternatives that bypass continuity corrections, the comparative performance of these approaches-especially between Bayesian and frequentist paradigms-remains understudied. METHODS:This study evaluates the performance of ten widely used meta-analysis models for binary outcomes, using the odds ratio as the effect measure. The evaluated models comprise seven frequentist and three Bayesian approaches. Simulations systematically varied key parameters, including control event rates, treatment effects, study numbers, and heterogeneity levels, to compare model performance across four metrics: percentage bias, 95% confidence/credible interval width, root mean square error, and coverage. The methods were further illustrated through applications to two published rare events meta-analyses. RESULTS:The results show that the beta-binomial model proposed by Kuss generally performed well, while the generalised estimating equations did not. In cases where heterogeneity is not large, all models tended to have a good performance except for the generalised estimating equations. When the heterogeneity is large, none of the compared models produced good performance. The Bayesian model incorporating the Beta-Hyperprior proposed by Hong et al. performed well, followed by the binomial-normal hierarchical model proposed by Bhaumik. CONCLUSIONS:In summary, the beta-binomial model proposed by Kuss is recommended for rare events meta-analyses, and the Bayesian model is a promising method for pooling rare events data.

The iMeta Conference 2025, part of the iMeta Conference series, themed "Creating High-Impact International Journals," held at the Huangjiahu Campus of Hubei University of Chinese Medicine from August 23rd to 25th, 2025, and focused on frontier topics such as microbiology, medicine, traditional Chinese medicine, botany, and research career development. The event aimed to support the development of researchers and strengthen the impact of academic journals. Through invited reports, thematic seminars, and poster presentations, the conference highlighted hot topics including multi-omics technologies, microbe-host interactions, AI-assisted research, live biotherapeutic products, and the modernization of traditional Chinese medicine. The event demonstrated the innovative momentum of interdisciplinary integration and technological convergence, providing an international platform for academic exchange and laying a foundation for building an innovative scientific research ecosystem and enhancing the global influence of Chinese academic journals.

BACKGROUND AND PURPOSE:Despite stroke center advancements in China, real-world adherence to acute care protocols of ischemic stroke remains understudied. We aimed to systematically investigate the clinical characteristics and in-hospital treatment of acute ischemic stroke (AIS) patients, and explore their association with prognosis. METHODS:We developed a nationwide cohort of AIS using data from the China National Electronic Disease Surveillance System. Patients were identified from the first discharge diagnosis. Comorbidities and prescription names were standardized by natural language processing and manual verification. Stepwise Cox regression models with fixed and time-dependent covariates explored the possible association between treatments and in-hospital mortality. RESULTS:This cohort included 14,046 patients with AIS from 111 hospitals between 2015 and 2020. Only a small proportion of patients received intravenous thrombolysis (2.76%) or endovascular interventional therapy (3.23%). Neuroprotective agents were used by 59.90% of patients, and dual antiplatelet therapy by 45.77%. Most patients (80.79%) received traditional Chinese medicine, including Chinese patent medicines (79.04%), Chinese herbal medicine slices (10.95%), and acupuncture (7.35%). Rehabilitation services were provided to 7.48% of patients. Cox regression analysis showed that neuroprotective agents (hazard ratio (HR) = 0.73, 95% confidence interval (CI) = 0.61-0.88), Chinese patent medicine (circulate blood and transform stasis: 0.49, 0.41-0.59; clear heat and remove toxins: 0.71, 0.52-0.98), Chinese herbal medicine slices (0.28, 0.17-0.44), acupuncture (0.58, 0.41-0.84), and rehabilitation therapies (0.95, 0.93-0.97) were potentially associated with reduced in-hospital mortality risk. CONCLUSIONS:Our findings showed relatively low utilization rates of thrombolytic (2.76%) and interventional therapies (3.23%) in China, highlighting the urgent need to improve access to these evidence-based reperfusion strategies. The use of neuroprotective agents, Chinese herbal medicine, acupuncture, and rehabilitation might be associated with reduced in-hospital mortality in AIS patients; however, future high-quality prospective studies are still warranted to confirm the clinical effects of these treatments.

Promoting the international acceptance of clinical studies about traditional Chinese medicine (TCM) interventions is a key strategy for internationalization of TCM. However, the complexities of TCM interventions—in terms of the theories, practice patterns, and components—pose challenges to the design and implementation of clinical studies that are well accepted by the international community. This article summarized the current status of clinical studies about TCM interventions that were published in international journals, explored underlying barriers hindering the international acceptance, and discussed potential strategies for future development.

OBJECTIVES:The demand for high-quality clinical evidence supporting acupuncture remains urgent, necessitating the establishment of a suitable methodological framework to promote its generation. METHODS:Following internal deliberations and extensive online discussions with experts in the IDEAL Collaboration, we proposed the IDEAL-Acu framework specifically for acupuncture, based on the surgery-focused IDEAL model with necessary modifications to accommodate the characteristics of acupuncture. To ensure consensus on recommendations, a panel of external experts and internal research team members was convened, and any disagreements were iteratively resolved through expert review. RESULTS:This article introduces an IDEAL-Acu framework with five stages for evaluating acupuncture outcome and improving practice to optimize treatment. The framework includes Idea (proposal of an acupuncture regime), Development (optimization or standardization of the acupuncture regime), Exploration (feasibility assessment for conducting a definitive RCT), Assessment (evaluation of effects through comparison with standard therapy or sham acupuncture), and Long-term monitoring (examination of long-term efficacy and safety) stages. We provide clear recommendations for each stage along with specific examples. CONCLUSION:The framework highlights the importance of conducting studies at each stage in acupuncture evaluation process and can serve as a helpful guide for assessing its effects and promoting evidence-based practice in acupuncture.

BACKGROUND:The use of inserted sham acupuncture as a placebo in randomized controlled trials (RCTs) is controversial, because it may produce specific effects that cause an underestimation of the effect of acupuncture treatment. OBJECTIVE:This systematic survey investigates the magnitude of insert-specific effects of sham acupuncture and whether they affect the estimation of acupuncture treatment effects. SEARCH STRATEGY:PubMed, Embase and Cochrane Central Register of Controlled Trials were searched to identify acupuncture RCTs from their inception until December 2022. INCLUSION CRITERIA:RCTs that evaluated the effects of acupuncture compared to sham acupuncture and no treatment. DATA EXTRACTION AND ANALYSIS:The total effect measured for an acupuncture treatment group in RCTs were divided into three components, including the natural history and/or regression to the mean effect (controlled for no-treatment group), the placebo effect, and the specific effect of acupuncture. The first two constituted the contextual effect of acupuncture, which is mimicked by a sham acupuncture treatment group. The proportion of acupuncture total effect size was considered to be 1. The proportion of natural history and/or regression to the mean effect (PNE) and proportional contextual effect (PCE) of included RCTs were pooled using meta-analyses with a random-effect model. The proportion of acupuncture placebo effect was the difference between PCE and PNE in RCTs with non-inserted sham acupuncture. The proportion of insert-specific effect of sham acupuncture (PIES) was obtained by subtracting the proportion of acupuncture placebo effect and PNE from PCE in RCTs with inserted sham acupuncture. The impact of PIES on the estimation of acupuncture's treatment effect was evaluated by quantifying the percentage of RCTs that the effect of outcome changed from no statistical difference to statistical difference after removing PIES in the included studies, and the impact of PIES was externally validated in other acupuncture RCTs with an inserted sham acupuncture group that were not used to calculate PIES. RESULTS:This analysis included 32 studies with 5492 patients. The overall PNE was 0.335 (95% confidence interval [CI], 0.255-0.415) and the PCE of acupuncture was 0.639 (95% CI, 0.567-0.710) of acupuncture's total effect. The proportional contribution of the placebo effect to acupuncture's total effect was 0.191, and the PIES was 0.189. When we modeled the exclusion of the insert-specific effect of sham acupuncture, the acupuncture treatment effect changed from no difference to a significant difference in 45.45% of the included RCTs, and in 40.91% of the external validated RCTs. CONCLUSION:The insert-specific effect of sham acupuncture in RCTs represents 18.90% of acupuncture's total effect and significantly affects the evaluation of the acupuncture treatment effect. More than 40% of RCTs that used inserted sham acupuncture would draw different conclusions if the PIES had been controlled for. Considering the impact of the insert-specific effect of sham acupuncture, caution should be taken when using inserted sham acupuncture placebos in RCTs. Please cite this article as: Luo XC, Liu JL, Yao MH, Chen YM, Fan AY, Liang FR, Zhao JP, Zhao L, Zhou X, Zhong XY, Yang JH, Li B, Zhang Y, Sun X, Li L. Specific effect of inserted sham acupuncture and its impact on the estimation of acupuncture treatment effect in randomized controlled trials: A systematic survey. J Integr Med. 2025; 23(6):630-640.

Machine learning (ML) models have been developed to identify randomised controlled trials (RCTs) to accelerate systematic reviews (SRs). However, their use has been limited due to concerns about their performance and practical benefits. We developed a high-recall ensemble learning model using Cochrane RCT data to enhance the identification of RCTs for rapid title and abstract screening in SRs and evaluated the model externally with our annotated RCT datasets. Additionally, we assessed the practical impact in terms of labour time savings and recall improvement under two scenarios: ML-assisted double screening (where ML and one reviewer screened all citations in parallel) and ML-assisted stepwise screening (where ML flagged all potential RCTs, and at least two reviewers subsequently filtered the flagged citations). Our model achieved twice the precision compared to the existing SVM model while maintaining a recall of 0.99 in both internal and external tests. In a practical evaluation with ML-assisted double screening, our model led to significant labour time savings (average 45.4%) and improved recall (average 0.998 compared to 0.919 for a single reviewer). In ML-assisted stepwise screening, the model performed similarly to standard manual screening but with average labour time savings of 74.4%. In conclusion, compared with existing methods, the proposed model can reduce workload while maintaining comparable recall when identifying RCTs during the title and abstract screening stages, thereby accelerating SRs. We propose practical recommendations to effectively apply ML-assisted manual screening when conducting SRs, depending on reviewer availability (ML-assisted double screening) or time constraints (ML-assisted stepwise screening).

IntroductionAlthough various sham acupuncture techniques have been employed to ensure blinding in randomised controlled trials (RCTs) of acupuncture, the effectiveness of blinding in these trials and its influence on trial effect size estimates remain unclear. The objectives of this study are the following: (1) to investigate the proportion and study characteristics of sham-controlled trials reporting on blinding assessment, (2) to assess the blinding effectiveness of different types of sham acupuncture, (3) to investigate the relationship between blinding effectiveness and effect sizes in acupuncture RCTs.Methods and analysisWe will search PubMed and EMBASE from inception to 1 January 2025 to identify RCTs that compared acupuncture with sham acupuncture in humans with any disease or symptom, with no restrictions on language. Paired investigators will independently determine eligibility and use pilot-tested standardised forms for data extraction. We will calculate the proportion of sham-controlled trials that assessed and reported blinding success and conduct descriptive analyses of general study characteristics, acupuncture treatment details, sham acupuncture details and blinding assessments for included trials. We will assess the effectiveness of blinding success using the James blinding index (BI) and Bang BI, and pool data from included trials using random-effects models. We will use Hedges’ g, a standardised mean difference, with its 95% CI, to calculate treatment effects. We will use Pearson’s r correlation coefficient to assess the relationship between blinding effectiveness and trial effect sizes when variable distributions meet the assumptions of normality and linearity; otherwise, we will consider employing non-parametric tests. When sufficient data are available, we will also use random-effects meta-regression to explore the relationship.Ethics and disseminationEthical approval is not required. The findings of this study will be disseminated through peer-reviewed publications, conference presentations and condensed summaries for clinicians, health policymakers and guideline developers regarding the design, conduct, analysis and interpretation of blinded assessment of sham acupuncture RCTs.Study registrationOpen Science Framework (https://doi.org/10.17605/OSF.IO/B3U7K).

PurposeThis study aimed to explore the association between the Nutrition Risk in Critical Illness (NUTRIC) score and the risk of ICU mortality in patients with sepsis.MethodsThis was a single-center, prospective cohort study that enrolled septic patients admitted between November 2024 and May 2025 to Wards 1 and 2 of the Department of Critical Care Medicine at the First Affiliated Hospital of Guangxi Medical University. A multivariable logistic regression model was applied to evaluate the association between the NUTRIC score assessed within 24 h of ICU admission and ICU mortality. Restricted cubic spline (RCS) analysis was conducted to model this relationship, and robustness was verified via subgroup analysis. Kaplan–Meier survival curve analysis was used to compare cumulative ICU survival rates among different NUTRIC score groups, with differences between groups tested using the log-rank test.ResultsA total of 245 patients with sepsis were included in the study, and the ICU mortality rate was 17.1% (42/245). Multivariable logistic regression showed a statistically significant association between the NUTRIC score and ICU mortality, with each 1-point increase in the score associated with a 92% increase in risk (OR = 1.92, 95% CI: 1.32–2.80, p = 0.002). RCS analysis indicated a significant linear relationship between the NUTRIC score and ICU mortality risk (P non-linearity = 0.704). Subgroup analysis further demonstrated a positive association across all subgroups (ORs > 1), and did not identify any significant interactions. Kaplan–Meier survival curves showed that patients with high nutritional risk (NUTRICb group) had significantly poorer ICU survival than those with low nutritional risk (NUTRICa group) (log-rank test, p = 0.00024).ConclusionThe NUTRIC score is significantly associated with ICU mortality in patients with sepsis, highlighting its potential utility in early ICU risk stratification. Incorporating the NUTRIC score into ICU assessment protocols may help identify high-risk patients early and guide early nutrition or supportive care, potentially improving clinical outcomes.