Research Appraisals
Evidence-based critical appraisals of the latest medical research, systematically evaluated using Oxford CEBM methodology.
Showing 90 appraisals
Clinical gastroenterology and hepatology : the official clinical practice journal of the American Gastroenterological Association
Artificial Intelligence Tools for Gastrointestinal Research: A Practical Guide
Artificial intelligence (AI) tools, including large language models (LLMs) such as ChatGPT, Claude, and Gemini, are increasingly used in biomedical research. However, practical guidance for integrating these tools into the manuscript writing workflow remains limited. This narrative review provides a structured, step-by-step guide for gastrointestinal (GI) researchers and clinicians seeking to use AI responsibly across the stages of manuscript preparation.
2 Aug 2026
Read appraisal →Clinical spine surgery
An Introduction to Machine Learning for the Practicing Spine Surgeon
The applications of new and emerging technologies in spine surgery are constantly expanding. Specifically, machine learning algorithms have seen a rise in utilization in clinical research, allowing for interpretation of large datasets that have the capability of experiential learning. The goal of this work is to present a guide for surgeons to better understand model design, key takeaways, and common pitfalls related to machine learning to ensure accurate and appropriate interpretation of analytical findings in their practice.
2 Aug 2026
Read appraisal →European journal of pediatrics
Health-related quality of life and associated factors in children with bronchiectasis: A systematic review
UNLABELLED: This systematic review aimed to summarize the current evidence on health-related quality of life (HRQoL) and its associated factors in children with bronchiectasis. Following the PRISMA guidelines, we systematically searched PubMed, Embase, Web of Science, ScienceDirect, CINAHL, CENTRAL, CNKI, WanFang, VIP, and CBM published from inception to October 2025. Two reviewers independently screened studies, extracted data, and assessed quality using standardized tools. Descriptive synthesis was performed. Nine studies were included, with overall moderate to high quality. HRQoL scores in children with bronchiectasis were significantly lower than those in healthy controls across most scales, though findings varied across physical and psychosocial domains. Parent-proxy scores were generally lower than child self-reports. Lower HRQoL was associated with impaired lung function, frequent exacerbations, higher symptom burden, and anxiety/depression. CONCLUSION: HRQoL is substantially impaired in children with bronchiectasis. Routine assessment of HRQoL and integrated interventions addressing both physical and psychosocial aspects are warranted in clinical practice. WHAT IS KNOWN: • Bronchiectasis is a chronic respiratory disease that causes coughing and breathing difficulties in children. • Previous studies have reported inconsistent findings on quality of life in children with bronchiectasis. WHAT IS NEW: • This systematic review consolidates evidence from nine studies, confirming that children with bronchiectasis have significantly lower quality of life than healthy children. • Poor lung function, frequent exacerbations, and anxiety/depression are key determinants; parents tend to rate their child's quality of life lower than the children themselves, underscoring the need for integrated care including psychological support.
31 July 2026
Read appraisal →Journal of robotic surgery
Robot-assisted versus open kidney transplantation: an umbrella review of systematic reviews and meta-analyses
Robot-assisted kidney transplantation (RAKT) has emerged as a minimally invasive alternative to open kidney transplantation (OKT). However, the comparative benefits and limitations of both approaches remain unclear. To compare the outcomes of RAKT and OKT through an umbrella review of published systematic reviews and meta-analyses. This umbrella review was written following PRISMA guidelines. Only meta-analysis and systematic review were included. Many outcomes were studied such as warm (WIT) and cold ischemia times (CIT), rewarming time, total ischemia time, blood loss, blood transfusion, delayed graft function, surgical site infections (SSI), hospital stay, graft rejection, graft failure, all-cause mortality, operation time, incision length, hospital readmission, vascular and ureteral anastomosis time, and other complications. We assessed methodological quality via the AMSTAR-2 tool. Consistency and strength of evidence were also evaluated. 6 studies were included: four meta-analyses and two systematic reviews. Studies were of low to moderate quality. Robotic kidney transplant surgery yields better outcomes in terms of blood loss (MDs ranging from - 16 to - 55 mL across studies), post-operative pain, incision length, SSI and ureteral leak. On the other hand, CIT, rewarming time, total ischemia time, operative time and ileus rate were favored in open surgery. Same rejection rates, graft failure, mortality and hospital readmission rate were found between both techniques. RAKT appears to improve several perioperative outcomes while maintaining comparable graft and patient survival outcomes to OKT. However, the available evidence remains limited by the low-to-moderate quality of existing reviews, highlighting the need for high-quality prospective studies.
27 July 2026
Read appraisal →Medicine
Role of antazoline in cardioversion of atrial fibrillation: A PRISMA-compliant systematic review and meta-analysis
BACKGROUND: Atrial fibrillation (AF) is the most common sustained cardiac arrhythmia. Antazoline, a first-generation antihistamine with antiarrhythmic properties, has shown efficacy in the rapid conversion of recent-onset AF to sinus rhythm; however, direct comparative evidence against standard antiarrhythmic agents remains limited. This study aims to systematically evaluate the efficacy and safety of antazoline for pharmacological cardioversion of AF, in comparison with standard antiarrhythmic agents such as amiodarone and propafenone. METHODS: The methodology of this systematic review and meta-analysis adhered to the preferred reporting items for systematic reviews and meta-analyses guidelines. A systematic search was conducted in ScienceDirect, Embase, PubMed, and Cochrane Library from inception to April 2025, including citation and reference screening. Statistical analyses were performed using R version 2025.05.0 + 496. RESULTS: From 172 records, 5 studies comparing antazoline with propafenone (n = 5) or amiodarone (n = 3) for AF were included. For antazoline vs propafenone, conversion to sinus rhythm favored antazoline (risk ratio [RR] 1.13 [95% confidence interval (CI): 1.04-1.23], P = .017). No significant difference was observed in time to conversion (mean difference -10.98 [95% CI: -124.93-102.98], P = .436), bradycardia (RR 0.84 [95% CI: 0.35-2.03], P = .478), hypotension (RR 0.40 [95% CI: 0.14-1.12], P = .056), or other adverse events (RR 1.24 [95% CI: 0.01-247.13], P = .694). For antazoline vs amiodarone, conversion to sinus rhythm favored antazoline (RR 1.22 [95% CI: 1.08-1.36], P = .018, I2 = 0%), with no significant differences in bradycardia (RR 1.36 [95% CI: 0-1072357.09], P = .823) or hypotension (RR 1.05 [95% CI: 0.04-26.16], P = .877). CONCLUSION: Antazoline might be more effective than propafenone and amiodarone in achieving sinus rhythm in AF, with no significant differences in safety outcomes.
26 July 2026
Read appraisal →BMJ open
Potential benefits of music-based therapies on rehabilitation outcomes in people with multiple sclerosis: a systematic review
OBJECTIVES: To assess the evidence for a benefit of music-based therapies (MBTs) on rehabilitation outcomes in adults with multiple sclerosis (MS). SETTING: Scholarly literature (published in English) from any country. This systematic review received no specific funding and was registered on PROSPERO as CRD420251083656. DATA SOURCES: Six bibliographic databases: AMED (Ebsco), APA PsycINFO (Ovid), CINAHL (Ebsco), Cochrane CENTRAL, EMBASE (Ovid) and MEDLINE (Ovid), plus supplementary searches and citation chasing (search date July 2025). PRIMARY AND SECONDARY OUTCOME MEASURES: Eligible studies used a clinical trial or controlled quantitative observational study design to assess the benefit of MBTs on rehabilitation outcomes in adults with MS. Risk of bias was assessed using Critical Appraisal Skills Programme checklists. Data were synthesised using narrative synthesis. RESULTS: 23 studies were included (955 participants, of whom 880 people had MS and 691 received MBTs). The most frequently assessed outcome was motor function (16 studies), followed by quality of life (12 studies). Evidence suggests potential benefits for MBTs, particularly rhythmic auditory stimulation and music-cued motor imagery, on gait and mobility outcomes, including walking speed, walking distance, cadence, stride length and gait symmetry. Findings for other outcomes were less clear. CONCLUSIONS: MBTs may benefit people with MS, with the most consistent evidence observed for gait and mobility outcomes. However, heterogeneity in intervention type, study design and outcome measurement limits the strength of conclusions. Higher-quality, adequately powered trials with standardised outcomes and direct comparisons between MBTs are needed to clarify effectiveness and inform clinical implementation. PROSPERO REGISTRATION NUMBER: CRD420251083656.
24 July 2026
Read appraisal →Journal of robotic surgery
Reliability and readability of AI chatbot responses to patient questions about robot-assisted radical cystectomy
Robot-assisted radical cystectomy (RARC) is a complex procedure that requires patients to understand surgical indications, urinary diversion, perioperative treatment, complications, recovery, and long-term functional outcomes. Although artificial intelligence (AI) chatbots are increasingly used to obtain medical information, their suitability for RARC patient education remains unclear. We conducted a cross-sectional comparative evaluation of four contemporary AI chatbots: ChatGPT-5, DeepSeek-V4, Claude Sonnet 4.6, and Gemini 3.5 Pro. A set of 20 core patient-education questions on RARC was developed by three senior urologic experts. Chatbot responses were assessed using DISCERN, the Ensuring Quality Information for Patients tool, the Global Quality Scale, and JAMA benchmark criteria. Readability was evaluated using the Automated Readability Index, Coleman-Liau Index, Flesch-Kincaid Grade Level, Flesch Reading Ease, Gunning Fog Index, and SMOG. Reliability scores differed significantly across models for DISCERN, EQIP, and GQS, while JAMA benchmark criteria were summarized descriptively as transparency signals. DeepSeek-V4 achieved the highest mean scores for DISCERN, EQIP, and GQS, while ChatGPT-5 and DeepSeek-V4 showed the strongest JAMA benchmark performance. Gemini 3.5 Pro generally had the lowest reliability and transparency scores. Readability also varied across models. DeepSeek-V4 produced the most readable responses overall, whereas Gemini 3.5 Pro generated the most complex text. However, all models exceeded the recommended sixth-grade reading level, and FRES scores remained below the recommended threshold. Contemporary AI chatbots generated responses with variable presentation quality, transparency, and readability for common RARC patient-education questions. Because factual accuracy was not directly assessed, these tools should not be interpreted as validated sources of clinical guidance and should not replace individualized counseling by urologists.
23 July 2026
Read appraisal →European journal of pediatrics
Probiotics in pediatric functional abdominal pain disorders: a systematic review and meta-analysis
UNLABELLED: Functional abdominal pain disorders (FAPDs), currently referred to as abdominal pain-related disorders of gut-brain interaction (AP-DGBIs), are characterized by recurrent abdominal discomfort in children that cannot be fully explained by an identifiable organic disorder. Probiotics are increasingly used as microbiota-targeted treatments, although their true clinical benefit remains uncertain. The objective of this study is to evaluate the efficacy and safety of probiotic supplementation compared with placebo in children with AP-DGBIs. PubMed, Embase, Web of Science and Scopus were searched from the start of each database to April 2026; the review followed PRISMA guidelines. Randomized controlled trials in patients < 18 years with FAPDs (IBS, FAP-NOS, functional dyspepsia, abdominal migraine) diagnosed by Rome II-IV, in which probiotic supplementation was evaluated. Two reviewers independently extracted data and assessed risk of bias using RoB2 tool. Meta-analyses and network meta-analyses were used to synthesize the data; the certainty of the evidence was evaluated using GRADE and CINeMA. This review covers 21 trials (2005-2025) with 1899 patients treatments completed. Probiotics modestly reduced pain intensity and weekly pain episodes, and modestly increased the rate of complete symptom resolution compared with placebo. Treatment with probiotics was associated with complete symptom resolution in 35.3% versus 22.8% of placebo-treated patients, corresponding to a number needed to treat (NNT) of 7 (95%CI = 3.97-29.04). However, the certainty of evidence was rated as low to moderate across outcomes. Moreover, from a clinical perspective, when success was defined as an improvement in symptoms according to an author-defined threshold, probiotic supplementation was not associated with different response rates compared with placebo. CONCLUSION: Probiotics provide modest symptom relief in pediatric FAPDs and may be considered a safe adjunctive therapy rather than a stand-alone treatment. While these findings are in line with the ESPGHAN/NASPGHAN 2025 guidelines, larger standardized trials are required to define optimal strain selection, dosage and treatment duration. TRIAL REGISTRATION: https://www.crd.york.ac.uk/PROSPERO/view/CRD420251149947 , identifier CRD420251149947. WHAT IS KNOWN: • AP-DGBIs (formerly FAPDs) are common childhood disorders characterized by recurrent abdominal pain without identifiable organic disease. • Probiotics are increasingly used to restore gut microbial balance, but their clinical benefit remains uncertain. WHAT IS NEW: • Our meta-analysis showed that probiotic supplementation reduced pain intensity and weekly pain episodes and increased symptom resolution versus placebo. • Benefi ts were modest, with substantial heterogeneity across studies and low-to-moderate certainty of evidence.
21 July 2026
Read appraisal →Journal of robotic surgery
Technical advancements in robot-assisted radical prostatectomy: systematic review and bibliometric analysis
Robot-assisted radical prostatectomy has transformed the surgical management of localized prostate cancer, yet its research landscape remains incompletely mapped. We conducted a systematic review and bibliometric analysis of 578 articles and conference papers from Scopus and Web of Science (2005-2026), following PRISMA guidelines, and analyzed publication trends, geographic and institutional distribution, author networks, journal output, keyword co-occurrence, and thematic evolution. The results reveal exponential growth in research, with an inflection around 2015-2016 and accelerated momentum after 2020, reflecting the widespread adoption and maturation of robotic programs. The United States leads in publication volume, followed by Italy and the rapidly growing contribution of China, while The University of Turin (Italy) emerges as the most productive institution due to its pioneering Retzius-sparing approach; the Journal of Robotic Surgery remains the leading publication venue. Thematic analysis demonstrates a progression from early feasibility studies to the standardization of nerve-sparing and continence-preserving techniques, advancing toward single-port surgery, augmented reality navigation, and AI-assisted intraoperative decision-making. Emerging technologies such as augmented reality, the NeuroSAFE protocol, and deep learning models for outcome prediction are particularly transformative, and geographic disparities are highlighted as critical priorities for future research.
21 July 2026
Read appraisal →PloS one
A novel dynamic Optuna hybrid Harris Hawks Optimization approach for classification of CAD
Coronary Artery Disease (CAD) is a leading cause of mortality worldwide and is primarily associated with atherosclerotic plaque formation, resulting in coronary artery stenosis. The accurate prediction of CAD using deep learning models is often constrained by the limitations of conventional optimization techniques, including premature convergence and limited adaptability of the models. To address these challenges, this study proposes a Dynamic Optuna Hybrid Harris Hawks Optimization (Dynamic Optuna H-HHO) framework to enhance the performance of deep learning-based CAD prediction models. The proposed approach integrates dynamic parameter adjustment, adaptive escape energy mechanisms, and Optuna-based hyperparameter tuning. The framework was applied to optimize several deep-learning classifiers, including ResNet-50, VGG-16, InceptionV3, and MobileNet, using a coronary artery stenosis dataset. The performance was evaluated through a comparative analysis with models optimized using the conventional Hybrid Harris Hawks Optimization (H-HHO) algorithm. The experimental results indicate that the proposed Dynamic Optuna H-HHO framework consistently improves the predictive accuracy across all evaluated models. InceptionV3 achieved the highest accuracy of 97.9%, followed by MobileNet with 97.6%, compared with the maximum accuracy of 82.46% obtained using traditional HHO-based optimization. By combining adaptive optimization strategies with automated hyperparameter tuning, the proposed framework provides a robust and scalable solution for improving the accuracy of coronary artery disease prediction.
20 July 2026
Read appraisal →Renal failure
Effects on mortality of different blood purification techniques in sepsis patients: an umbrella review of systematic reviews and meta-analyses
Sepsis remains a leading cause of mortality worldwide, and extracorporeal blood purification has been widely implemented despite ongoing uncertainty regarding its survival benefit. We conducted an umbrella review of 42 systematic reviews and meta-analyses evaluating mortality across major extracorporeal blood purification modalities. Mortality estimates generally favored extracorporeal blood purification over conventional care, but effect sizes and consistency varied substantially by modality, and the overall certainty of evidence for mortality was low. At the review level, polymyxin B hemoperfusion showed a relatively consistent direction toward lower mortality, whereas renal replacement therapy-based strategies were typically closer to the null. However, most included reviews were of low or critically low methodological quality, with downgrading driven by risk of bias, inconsistency, and potential publication bias. Therefore, these findings should be interpreted cautiously and should not be considered definitive evidence of survival benefit. They highlight persistent evidentiary fragility and underscore the need for rigorously designed, phenotype-informed trials to clarify modality-specific effects in sepsis. What is known: Extracorporeal blood purification has been widely used as an adjunctive therapy for sepsis, but its mortality benefit remains uncertain across different modalities.What this study adds: This umbrella review synthesizes evidence from 42 systematic reviews and meta-analyses and shows that mortality estimates generally favor extracorporeal blood purification, although effects vary substantially by modality and the overall certainty of evidence remains low.Potential impact: These findings support cautious, modality- and phenotype-informed interpretation of extracorporeal blood purification in sepsis and highlight the need for rigorously designed trials with standardized protocols.
19 July 2026
Read appraisal →Neurosurgical review
Stereotactic radiosurgery for treatment-refractory obsessive-compulsive disorder: a systematic review of targets, targeting methods, lesion geometry, and clinical outcomes.
BACKGROUND: Stereotactic radiosurgery (SRS) is a noninvasive option for carefully selected patients with treatment-refractory obsessive-compulsive disorder (OCD), but the literature is heterogeneous in targets, targeting methods, lesion geometry, and outcomes. METHODS: PRISMA-guided systematic review of PubMed, Scopus, and Web of Science. Full-text human studies of radiosurgical capsulotomy for treatment-refractory OCD were included. Narrative synthesis was performed across technical and clinical domains. When ≥3 comparable studies reported the same endpoint, exploratory random-effects single-arm meta-analyses were performed. RESULTS: Twenty-one OCD-specific studies were included (1999-2024). The anterior limb of the internal capsule (ALIC), particularly the ventral ALIC, was the dominant target. Technique and dose reporting varied across cohorts, with commonly reported maximum doses spanning approximately 120-180 Gy, often using 4-mm collimation and single- or double-shot paradigms. Earlier reports primarily used anatomical MRI/coordinate-based targeting, whereas newer studies increasingly incorporated lesion mapping, tractography, and network-based analyses. Eleven cohorts were eligible for quantitative pooling. The pooled clinical response rate was 57.1% (95% CI, 49.1%-64.9%; I²=21.7%), and pooled remission was 36.8% (95% CI, 26.9%-47.9%; I²=0%). Severe delayed radiation-related injury was uncommon but present (6.4%, 95% CI, 3.1%-12.8%; I²=0%). Mechanistic studies suggested that lesion topography and pathway engagement may be more important than lesion size alone. CONCLUSIONS: SRS is a meaningful option for treatment-refractory OCD. The ventral ALIC remains the principal target, but outcomes likely depend on precise target identification and lesion placement. Standardized reporting and prospective evaluation of pathway-informed targeting are needed.
19 July 2026
Read appraisal →Biomedical physics & engineering express
NurtureNest: an IoT-wearable predictive analytics framework for real-time maternal risk assessment
Pregnancy-related complications are increasing globally, necessitating timely and accurate risk prediction for effective clinical intervention. This paper presents NurtureNest, an internet of things-enabled machine learning framework for automated pregnancy risk assessment. The system integrates wearable sensor data from smartwatches with user-provided clinical parameters via a mobile application, enabling continuous remote monitoring. Historical clinical data are used to train multiple machine learning models for multi-class classification of pregnancy risk into low-, medium-, and high-risk categories. Ensemble learning techniques are independently trained and evaluated alongside conventional machine learning classifiers to assess their effectiveness in pregnancy risk prediction. Experimental results show that LightGBM achieves the highest performance with 92.86% test accuracy and 95.04% cross-validation accuracy. Model performance is validated using ROC analysis and feature importance evaluation. The proposed framework enables early risk detection and supports timely clinical decision-making, improving maternal healthcare outcomes.
17 July 2026
Read appraisal →Pediatric surgery international
Single-stage versus staged reconstruction for high and intermediate anorectal malformations in neonates: a systematic review and meta-analysis
The surgical approach to high and intermediate anorectal malformations remains controversial. This systematic review compared continence and infection outcomes between the two approaches. MEDLINE, Embase, and Cochrane CENTRAL were searched per a registered protocol (PROSPERO CRD420261291319). Comparative studies of single-stage versus staged repair in neonates with high or intermediate anorectal malformations were included. Risk of bias was assessed using ROBINS-I. Random-effects meta-analysis with prediction intervals was performed. Four comparative studies (746 patients: 399 single-stage, 347 staged) were included. Pooled analysis showed no significant difference in incontinence outcomes (RR 0.93, 95% CI: 0.52-1.66, p = 0.80; 95% prediction interval: 0.28-3.08) with substantial heterogeneity (I²=83%). Infection data could not be pooled due to methodological concerns in outcome reporting. All included studies demonstrated moderate to serious risk of bias. GRADE certainty was LOW for incontinence and VERY LOW for infection. Current evidence does not demonstrate superiority of either approach for bowel continence (LOW certainty). Infection evidence was insufficient for synthesis (VERY LOW). Wide prediction intervals indicate substantial uncertainty; available data cannot confirm or refute inferior outcomes after single-stage repair. Prospective studies with standardized outcomes are needed. Protocol registration: PROSPERO CRD420261291319.
17 July 2026
Read appraisal →Biology of reproduction
Noninvasive pre-transfer identification of women with recurrent implantation failure using a plasma microRNA-based machine learning model
Recurrent implantation failure (RIF) refers to patients who have undergone at least three implantation failures with several high-quality embryos. Identifying patients with RIF in advance is crucial for effective clinical managements. However, uterine biopsy, the primary research approach, is not only invasive but also poses a risk to the integrity of the endometrial microenvironment. Here, we provided a novel, safe, and immediate methodology for RIF populations identification through plasma microRNA (miRNA) profiles. Peripheral blood samples of 163 women were collected and sequenced to quantify the relative expression levels of plasma miRNAs. Through a comparative analysis of the temporal expression between patients with RIF and those with successful embryo implantation during the peri-implantation period, we determined that the RIF-associated miRNA signature was established at the early stage of the window of implantation. By global screening, we identified 217 miRNAs that were differentially expressed between groups on the ovulation day (D0), using a threshold of |fold change| ≥ 1.5. Through refining selection criteria, 10 eligible miRNAs were ultimately selected and further validated through real-time quantitative polymerase chain reaction. These miRNAs were employed to construct a logistic regression-based RIF model. The model demonstrated robust performance in the validation and test cohorts, with area under the curve of 0.947 and 0.912, respectively. We also demonstrated that miRNAs maintain relative stability within a specific phase, making them suitable as clinical biomarkers. This study provides a noninvasive methodology for identifying women with RIF and suggests a potential role in predicting RIF risk before embryo transfer.
16 July 2026
Read appraisal →International ophthalmology
Safety and efficacy of silicone intubation compared to standard techniques in endoscopic dacryocystorhinostomy: a systematic review and meta-analysis
BACKGROUND: Primary Acquired nasolacrimal duct obstruction (PANDO) is a lacrimal drainage disorder, commonly seen in adults. Endoscopic Dacryocystorhinostomy (EN-DCR) is a widely adopted surgical procedure for treatment of nasolacrimal duct obstruction. Silicon intubation is a technique, which involves inserting a soft silicone stent to keep the ostium patent post operatively. METHODS: This meta-analysis followed Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) 2020 guidelines, with searches across 4 databases. Primary outcomes were overall success rate, and overall failure rate. Secondary outcomes included granulations/intracanalicular obstruction (ICO) threats, hemorrhage, anatomical failure, functional failure, recurrent symptoms/epiphora, and cicatricial stenosis/post-procedural restenosis. Data were analyzed using R software version 4.3. Heterogeneity was assessed using the I2 statistic. RESULTS: From 3,974 studies initially, 20 studies were selected in the final review, in which 17 were RCTs and 3 observational studies. A statistically significant improvement in surgical success rates, while reducing failure rates were observed in the silicone intubation (stented) group compared to the unstented group. No significant difference was observed in the incidence of restenosis, hemorrhage, functional failure, epiphora or recurrent symptoms. Reduction of anatomical failure was observed. Increase in adverse effects of granulation tissue formation and internal canalicular obstruction were observed. Heterogeneity was low and no publication bias was detected in outcomes. CONCLUSIONS: Silicone intubation during mechanical endoscopic dacryocystorhinostomy (EN-DCR) was associated with improved success and patency rates, as well as lower failure rate. While more resource-intensive, it provides a slightly higher advantage, particularly in recurrent or revision instances, recommending its implementation in routine EN-DCR where possible.
16 July 2026
Read appraisal →Journal of ethnopharmacology
Efficacy of Wendan Decoction combined with conventional medicine for depression: a systematic review and meta-analysis
ETHNOPHARMACOLOGICAL RELEVANCE: Wendan Decoction is well-known traditional Chinese medicine formula commonly used in China for the treatment of gastrointestinal disorders and mood disorders related to phlegm-dampness and spleen deficiency. In clinical practice, depending on different symptoms, additional herbs may be included, resulting in a modified formula known as modified Wendan Decoction. AIM OF THE STUDY: To evaluate the efficacy of modified Wendan Decoction plus conventional medicine in the treatment of depression. MATERIALS AND METHODS: We independently conducted a meta-analysis of studies on the efficacy of modified Wendan Decoction for depression from inception to February 28, 2025, using electronic databases, including PubMed, Embase, Cochrane, Web of Science, Wanfang, VIP and SinoMed. Data from randomized controlled trials were extracted to evaluate treatment outcomes based on depression scales. RESULTS: A total of 402 articles were identified through database searches, and 13 studies, encompassing 1117 patients, were included. The initial pooled analysis revealed a significant treatment effect of the experimental group, yielding a standardized mean difference (SMD) of -4.84 (95% CI: -7.54 to -2.14), with high heterogeneity (I2 = 99%). Sensitivity analysis, excluding studies with high risk of bias, demonstrated an SMD of -4.18 (95% CI: -5.26 to -3.10) with moderate heterogeneity (I2 = 83%). Notable regional differences were found, with the study in Southern region exhibiting greater efficacy (SMD = -4.81, 95% CI: -6.38 to -3.23) compared to the study in Northern region (SMD = -3.03, 95% CI: -3.73 to -2.32). CONCLUSIONS: Our meta-analysis supports the efficacy of combining modified Wendan Decoction with conventional medicine for treating depression. Exploratory subgroup results suggest slightly greater improvements in Southern China, though this finding requires cautious interpretation and confirmation in future high-quality trials. PROSPERO Registration ID: CRD420251015690.
15 July 2026
Read appraisal →Depression and anxiety
The Impact of Psoriasis Treatment on Depression and Suicide Risk: A Systematic Review
BACKGROUND: Psoriasis is a long-lasting inflammatory condition of the skin associated with various comorbidities, including depression and suicidal ideation. Management strategies for psoriasis include symptom alleviation, quality-of-life enhancement, and prevention of disease progression. Psoriasis treatments include topical therapies, phototherapy, oral systemic medications, and biologics. OBJECTIVES: In this review, we evaluate the impact of psoriasis treatments on depression and suicidal ideation in affected patients. METHODS: We systematically searched multiple databases, including PubMed, Scopus, and Web of Science, until September 30, 2024, to identify relevant articles. Studies that examined the effects of psoriasis therapies on depression and suicidal thoughts were included. Data on treatment modalities, psychological health outcomes, and psoriasis severity was obtained. Quality evaluation instruments, such as JBI, consolidated standards of reporting trials (CONSORTs), Center for Evidence-Based Management (CEMBa), and appraisal tool for cross-sectional studies (AXIS), were used to appraise study quality. RESULTS: Ten studies met inclusion criteria, predominantly focusing on biologic therapies. Biologics like guselkumab, brodalumab, and bimekizumab have shown notable reductions in depression symptoms, probably through the alleviation of psoriasis severity and the enhancement of quality-of-life. Suicidal ideation occurred in some cases, especially with brodalumab; however, a definitive causal relationship was not established. CONCLUSIONS: Treatments for psoriasis, especially biologics, have shown some advantages in reducing depression symptoms as well as relieving skin symptoms. However, cautious monitoring is required due to the possible hazards of suicidal thoughts with certain therapeutic options such as biologics. Addressing the complex issues of psoriasis requires comprehensive therapy that includes dermatologists and mental health specialists.
13 July 2026
Read appraisal →European journal of pediatrics
Efficacy and safety of procedural sedation-analgesia during laser photocoagulation for retinopathy of prematurity: a systematic review
UNLABELLED: Laser photocoagulation for retinopathy of prematurity (ROP) in preterm infants causes significant procedural pain. Although general anaesthesia (GA) is the standard of care, many settings frequently rely on inadequate sedation-analgesia or topical anaesthesia alone, exposing neonates to considerable pain and clinical instability. This systematic review aimed to evaluate the efficacy and safety of procedural sedation-analgesia during laser photocoagulation for ROP. We searched MEDLINE, Embase, and Cochrane CENTRAL for trials investigating sedation-analgesic agents during laser photocoagulation. Primary outcome was analgesic efficacy (pain score reduction, rescue analgesics required). Additional outcomes included adverse effects (e.g., apnea, oxygen requirement). Two reviewers extracted relevant information and assessed risk-of-bias using ROB2 and evidence certainty using GRADE methodology. Of 540 articles screened, four relevant RCTs were identified (n = 265). These RCTs compared varying combinations of agents (oral dextrose vs. no dextrose, fentanyl vs. oral sucrose, fentanyl vs. ketamine, ketamine-propofol vs. GA). Three RCTs evaluating pain-related outcomes (PIPP-R score and intra-procedural crying duration) suggested that the investigated non-general anaesthetic regimens may provide inadequate procedural analgesia (low to very low-certainty evidence). The evidence is very uncertain about the effect of ketamine-propofol on post-procedure ventilation requirement compared with general anaesthesia. Apnea was the commonest adverse effect (three studies), especially with high-dose fentanyl (30%). Three studies had a low risk-of-bias, while one had some concerns (randomisation, outcome measurement). CONCLUSION: Overall, the included studies were small RCTs evaluating varying combinations of interventions, precluding concrete conclusions about the superiority of any analgesic regimen during laser photocoagulation. Further high-quality studies are needed to identify safe and effective analgesia for settings where GA is unavailable. REVIEW REGISTRATION: Prospectively registered on PROSPERO (CRD420251070146). WHAT IS KNOWN: • Laser photocoagulation procedure for retinopathy of prematurity is immensely painful. • Despite general anaesthesia being the standard, most neonates undergo the procedure under inadequate analgesia, particularly in high-burden settings. Optimal analgesic regimen for the laser procedure, in settings where general anaesthesia is unavailable, is unknown. WHAT IS NEW: • This systematic review of trials investigating analgesic agents for laser procedure for ROP found four RCTs with significant heterogeneity in intervention agents. • Low- to very low-certainty evidence suggests that none of the evaluated sedo-analgesic regimens consistently achieved optimal analgesia, and most infants continued to experience moderate-to-severe procedural pain.
13 July 2026
Read appraisal →Medicine
Risk factors for episiotomy during vaginal birth: A systematic review and meta-analysis
BACKGROUND: Despite the numerous harms associated with episiotomy during vaginal birth, the risk factors for this procedure remain a topic of controversy. This study aimed to identify the potential risk factors for episiotomy during vaginal birth. METHODS: A systematic search was conducted across 7 databases, including PubMed/MedLine, Embase, Cochrane, Web of Science, Chinese Knowledge Infrastructure (CNKI), and Wanfang, up to January 2025. Statistical analyses, including heterogeneity tests, sensitivity analysis, and publication bias assessment, were conducted using Stata 18.0 and Review Manager 5.4. RESULTS: A total of 16 articles were included in this meta-analysis, comprising a sample size of 17,502 individuals. Newborn weight (odds ratio [OR] = 2.393; 95% confidence interval [CI] = 1.627-3.520; P < .001) and oxytocin use (OR = 2.341; 95% CI = 1.652-3.318; P < .001) were identified as significant risk factors for episiotomy. In contrast, maternal age (OR = 0.978; 95% CI = 0.922-1.038; P = .466), the type of childbirth attendant (OR = 0.920; 95% CI = 0.569-1.488; P = .733), and the duration of the second stage of labor (OR = 1.299; 95% CI = 0.781-2.162; P = .314) were not statistically significant risk factors for episiotomy. CONCLUSION: High newborn weight and oxytocin use are associated with higher rates of episiotomy. However, other potential risk factors are still controversial due to a lack of sufficient evidence, highlighting the need for more rigorous studies to strengthen the evidence base.
13 July 2026
Read appraisal →JMIR research protocols
Applications of Large Language Models in Ovarian Cancer Management: Protocol for a Systematic Review and Meta-Analysis
BACKGROUND: Ovarian cancer (OC) is a highly fatal gynecologic malignancy with complex management challenges and limited long-term survival for advanced stages. Large language models (LLMs)-including systems such as GPT-4, Claude, Google Gemini, and others-are emerging artificial intelligence (AI) tools capable of performing health care-related tasks such as diagnostic support, treatment planning, report generation, and patient communication. However, their applications in OC care have not yet been comprehensively assessed. OBJECTIVE: This protocol outlines a systematic review and meta-analysis aimed at evaluating the use, performance, and clinical impact of LLMs in OC management. We will examine how LLMs have been applied across various domains (eg, diagnosis, prognosis, treatment planning, and patient engagement), the metrics used to assess their performance (eg, accuracy, sensitivity, and area under the curve), and their strengths and limitations. METHODS: This review will be conducted in accordance with PRISMA-P (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Protocols) guidelines. A comprehensive search strategy will be implemented across biomedical, technical, and Chinese-language databases (eg, PubMed, Embase, Web of Science, IEEE Xplore, and China National Knowledge Infrastructure) from inception to December 31, 2025. Eligible studies include clinical evaluations, validation studies, and real-world implementation reports involving LLMs in OC care. Two independent reviewers will perform screening, data extraction, and quality appraisal using validated tools (eg, version 2 of the Cochrane risk-of-bias tool for randomized trials, Risk of Bias in Nonrandomized Studies of Interventions, Quality Assessment of Diagnostic Accuracy Studies 2, and Prediction Model Study Risk of Bias Assessment Tool+AI). Outcomes of interest include model performance metrics, clinical process impacts, safety concerns, and usability. Meta-analyses will be conducted where feasible using random-effects models in R (meta, metafor, and mada packages), including bivariate models for sensitivity and specificity. RESULTS: The review is currently in progress. The PROSPERO registration has been completed, and the literature search and selection process is underway. Study selection, data extraction, and quality assessment are expected to be completed by mid-2026. Final results will include pooled performance metrics (eg, accuracy, F1-score, and area under the curve), qualitative insights into clinical integration, and identification of limitations such as reporting bias or insufficient external validation. CONCLUSIONS: This systematic review will provide the first comprehensive synthesis of evidence on the application of LLMs in OC care. It will identify promising use cases, highlight safety and reporting challenges, and inform future research directions. The findings are expected to support evidence-based integration of LLMs into gynecologic oncology workflows while promoting transparency and methodological rigor in AI evaluation.
12 July 2026
Read appraisal →Journal of medical Internet research
Magnetic Resonance Imaging-Based Artificial Intelligence in Predicting Prostate Cancer Biochemical Recurrence: Systematic Review and Meta-Analysis
BACKGROUND: Artificial intelligence (AI) has emerged as a promising tool for prostate cancer (PCa) risk stratification and outcome prediction. However, current studies often lack multicenter external validation, have limited sample sizes, present significant intermodel variability, and face overfitting concerns. OBJECTIVE: This study aimed to comprehensively evaluate the diagnostic performance of magnetic resonance imaging (MRI)-based AI models in predicting biochemical recurrence (BCR) of PCa. METHODS: Systematic searches were conducted in the PubMed, Embase, Web of Science, and Cochrane Library databases up to January 13, 2026. Studies were included that involved participants diagnosed with PCa, used MRI-based AI for predicting BCR, and had clearly defined reference standards. The quality of the included studies was assessed using the PROBAST+AI tool. A bivariate random effects model was used to pool sensitivity, specificity, and area under the curve (AUC) statistics. RESULTS: A total of 28 studies were included, with 2623 patients in internal validation and 1134 patients in external validation. Diagnostic contingency tables were reconstructed from published performance metrics for most studies, while others were extracted from receiver operating characteristic curves due to the lack of direct reporting. In the internal validation set, pooled sensitivity was 0.80 (95% CI 0.73-0.86; prediction interval [PI] 0.48-0.99), specificity was 0.83 (95% CI 0.77-0.89; PI 0.49-1.00), and AUC was 0.86 (95% CI 0.83-0.89; PI 0.74-0.99). In the external validation set, pooled sensitivity was 0.82 (95% CI 0.72-0.91; PI 0.54-0.99), specificity was 0.83 (95% CI 0.71-0.92; PI 0.49-1.00), and AUC was 0.84 (95% CI 0.79-0.90; PI 0.70-0.98). No statistically significant differences were observed between internal and external validation in sensitivity (P=.73), specificity (P>.99), AUC (P=.53), or diagnostic odds ratio (P=.98). Medical Net and Extreme Gradient Boosting achieved the highest sensitivity and AUC, whereas multiple kernel learning and support vector machine had the highest specificity. Subgroup and meta-regression analyses suggested that AI method, model type, timing of MRI acquisition, and treatment modality may contribute to heterogeneity. CONCLUSIONS: This meta-analysis innovatively realizes the quantitative direct comparison of MRI-based AI model performance between internal and external validation cohorts for PCa BCR prediction. It comprehensively evaluates AI performance across diverse PCa treatment modalities and integrates machine learning and deep learning approaches. For the field, it identifies key sources of performance heterogeneity (eg, MRI acquisition timing and treatment modality) and quantifies the sensitivity-specificity trade-off in integrated radiomic-clinical models, advancing the systematic understanding of MRI-based AI for BCR prediction. In real-world practice, it provides actionable guidance to prioritize pretreatment MRI for AI model development and clinical BCR assessment and underscores the urgent need for standardized imaging protocols and prospective multicenter studies, laying a foundation for the safe clinical translation of these AI tools as adjunctive decision support instruments.
8 July 2026
Read appraisal →Spectrochimica acta. Part A, Molecular and biomolecular spectroscopy
AI-enhanced non-invasive diagnosis of chronic kidney disease using LIBS of fingernail biomarkers
Chronic kidney disease (CKD) is a systemic condition that leads to progressive renal failure and metabolic imbalances that may be detected in the keratinized bio-tissues of the body such as fingernails. Nevertheless, still its early detection is difficult due to the invasive nature of current clinical screening approaches. This research paper provides a precise, non-invasive, and AI-enhanced diagnostic model of CKD screening through the combination of Laser-Induced Breakdown Spectroscopy (LIBS) and the cutting-edge ensemble machine learning to determine the elemental patterns of fingernail biomarkers. In this study, 55 participants with chronic kidney disease and 45 healthy controls were taken as the subjects and their emission spectra were measured with a nanosecond LIBS spectrometer using a laser of a fundamental wavelength @ 1064 nm. To handle the high-dimensional spectral data, Principal Component Analysis (PCA) was used as a feature extraction approach, forming the basis for novel machine-learning models. A sophisticated ensemble learning algorithm specifically Extreme Gradient Boosting (XGBoost) was applied to examine the biochemical change between CKD and control samples and the predictive quality was tested by a relative comparison with a Support Vector Machine (SVM) algorithm. Using 10-fold cross-validation, the XGBoost model outperformed SVM, attaining 97% accuracy, 98% sensitivity, 96% precision and specificity, and a 97% F1-Score. Moreover, external validation on a separate dataset showed that the model is robust and generalizable with 95% accuracy, 97% sensitivity, and an F1-score of 95%. The results conclude that fingernail-based LIBS integrated with ensemble machine learning is a potential and non-invasive instrument for CKD classification.
6 July 2026
Read appraisal →Journal of pediatric health care : official publication of National Association of Pediatric Nurse Associates & Practitioners
Systematic Review of Studies Investigating Infant Feeding Difficulties: Focus on Machine Learning Applications
INTRODUCTION: This study aims to systematically review studies using machine learning for infant feeding difficulties and identify the clinical applicability and limitations of such approaches. METHODS: A literature search was conducted in accordance with the PRISMA guidelines. Studies on the use of machine learning for infants with feeding difficulties were analyzed. RESULTS: We screened 1,104 studies, and 10 were eligible for inclusion. These studies applied machine learning to measure physiological signals during feeding, such as the number of sucking, swallowing and breathing events, to classify or predict feeding difficulties. The most common algorithms in the analyzed studies included support vector machines (SVMs), k-nearest neighbors (KNNs), and decision trees (DTs), which generally achieved high classification accuracy. DISCUSSION: Although all included studies used sensor-based data, only four directly applied machine learning models. Future research should expand the direct applications of machine learning, standardize measurements, and validate algorithms to improve clinical utility.
2 July 2026
Read appraisal →BMJ supportive & palliative care
Time toxicity of patients with cancer enrolled in clinical trials: a systematic review and meta-analysis
PURPOSE: Time toxicity-the time patients spend receiving treatment, managing side effects, attending follow-ups and undergoing rehabilitation-is increasingly recognised as an important burden of cancer care. While time toxicity has been studied in routine clinical settings, it has rarely been quantified among patients with cancer enrolled in pharmacotherapy clinical trials. This systematic review aims to quantify time toxicity in this population. DESIGN: We searched Ovid MEDLINE, Embase and Cochrane CENTRAL from inception to 12 June 2025. Studies were included if they reported time toxicity among patients with cancer enrolled in clinical trials. Time toxicity was defined as calendar days with healthcare contact. Primary outcomes were (1) total days, (2) planned days and (3) unplanned days per month. Study quality was assessed using the Joanna Briggs Institute's (JBI) Checklist for Cohort Studies. RESULTS: Four retrospective studies reporting on an aggregate of 1809 patients were included. Two studies had low risk for bias and two had unclear risk of bias. Patients experienced time toxicity a mean of 7.6 days per month, with the highest burden in the first month of trial participation. Planned and unplanned days averaged 2.2 and 2.0 days per month, respectively. The ratio of planned to unplanned days remained consistent over time. CONCLUSIONS: Time toxicity is a substantial, under-recognised burden in cancer pharmacotherapy trials-approximately two times that experienced by non-participants. Its measurement and mitigation should be prioritised to improve patient-centred care, support shared decision-making and inform trial design, policy and consent processes. REGISTRATION: Study protocol preregistration: https://doi.org/10.17605/OSF.IO/BVY7C.
1 July 2026
Read appraisal →Diving and hyperbaric medicine
Evidence-informed decision aid for fitness-to-dive assessment after otologic surgery
INTRODUCTION: Fitness-to-dive after otologic surgery is often approached conservatively, with some procedures historically labelled as absolute contraindications despite limited empirical evidence. The available literature is heterogeneous and includes clinical reports, experimental pressure studies, guidance documents, and manufacturer specifications, leading to uncertainty in clinical counseling. We aimed to characterise the available evidence regarding fitness-to-dive after otologic surgery and to develop an evidence-informed clinical decision aid. METHODS: A scoping review was conducted in accordance with PRISMA-ScR guidance. PubMed/MEDLINE, Embase, Scopus, and relevant non-indexed sources were searched. Eligible sources included clinical reports and series, experimental or hyperbaric chamber studies, guidance or consensus documents, and manufacturer statements providing explicit pressure- or depth-related information. Data were charted descriptively by procedure type and evidence stream. RESULTS: The search identified 324 records; after removal of duplicates and screening, 40 sources were included. The evidence base was predominantly non-comparative. Across procedures, recommendations emphasised postoperative stability and reliable pressure equalisation rather than surgical history alone. Canal wall down mastoidectomy was consistently portrayed as incompatible with diving, whereas selected middle ear reconstructions and stapes surgery were commonly described as potentially compatible in appropriately selected individuals. For cochlear implantation, guidance was mainly conditional and based on hyperbaric testing, limited clinical diving reports, and manufacturer-specified pressure or depth limits. Communication emerged as an additional practical consideration in cases of significant hearing loss. CONCLUSIONS: Relevant evidence is limited and heterogeneous, and does not consistently support blanket prohibitions for all otologic procedures. A function-based, individualised approach is supported, while specific higher-risk scenarios warrant restriction. Prospective registries and standardised outcome reporting are needed to refine procedure-specific recommendations.
30 June 2026
Read appraisal →BMJ open
Comparative diagnostic accuracy of image-enhanced endoscopy modalities and artificial intelligence-assisted endoscopy for gastric intestinal metaplasia and gastric dysplasia: protocol for a systematic review and diagnostic test network meta-analysis.
INTRODUCTION: Gastric intestinal metaplasia (GIM) and gastric dysplasia are critical precursors in the Correa cascade of gastric carcinogenesis. Although multiple image-enhanced endoscopy (IEE) modalities and artificial intelligence (AI)-assisted systems have been developed to improve their detection, the comparative diagnostic accuracy of these technologies remains unclear. Direct comparative studies are limited and conventional pairwise meta-analyses do not provide a coherent framework for comparing multiple diagnostic technologies. METHODS AND ANALYSIS: This protocol follows the Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols (PRISMA-P) and PRISMA-Network Meta-Analysis guidelines. We will search MEDLINE (Ovid), Embase (Ovid), the Cochrane Central Register of Controlled Trials (CENTRAL; Cochrane Library/Wiley), Web of Science Core Collection (Clarivate), Scopus (Elsevier), ClinicalTrials.gov and the WHO International Clinical Trials Registry Platform from inception, without date or language restrictions. Eligible studies will be diagnostic accuracy studies evaluating IEE modalities (including narrow-band imaging with or without magnification, blue laser imaging, linked colour imaging and iScan) or AI-assisted endoscopy against histology. Comparative studies will form the primary evidence base for the diagnostic test network meta-analysis; single-arm studies will be summarised separately and will not contribute to the primary network estimates. The coprimary diagnostic parameters will be sensitivity and specificity for GIM and dysplasia. These will be analysed jointly using a hierarchical bivariate diagnostic test network meta-analysis with binomial likelihoods and correlated random effects. Diagnostic OR and area under the receiver operating characteristic curve will be treated as secondary summary measures. Two reviewers will independently screen, extract data and apply Quality Assessment of Diagnostic Accuracy Studies 2. Heterogeneity, transitivity and inconsistency will be assessed using prespecified clinical and methodological effect modifiers, model-based diagnostics and sensitivity analyses. The certainty of evidence will be assessed using Confidence in Network Meta-Analysis, with interpretation adapted to diagnostic accuracy outcomes. ETHICS AND DISSEMINATION: This study does not require ethical approval as it relies on secondary analysis of published literature. Findings will be disseminated through peer-reviewed publication and conference presentations. PROSPERO REGISTRATION NUMBER: CRD420261363078.
29 June 2026
Read appraisal →Biomedical physics & engineering express
Transformer imputation in CTG: a length-dependent evaluation of reconstruction methods
Objective.Computer and artificial intelligence (AI) analyses are being increasingly used in intrapartum cardiotocography (CTG). However, fetal heart rate (FHR) signal loss, which frequently occurs in clinical practice, hinders visual interpretation and reduces accuracy. Although the impact is well recognized, there is no consensus on the maximum continuous gap length that can be reliably reconstructed under clinical conditions. Therefore, we aim to identify suitable imputation methods and clarify the clinical limits of valid missing-segment lengths.Methods.Using an open FHR dataset (CTU-UHB), we extracted continuous segments and artificially introduced Removed data of varying lengths. Using performance metrics such as difference and similarity, we compared the performance among a Transformer-based model and linear and spline interpolation. Additionally, we quantified the similarity between the Pre-impute and removed data to assess task difficulty.Results.We analyzed 2727 segments from 552 cases across multiple gap lengths. In terms of numerical accuracy root mean square error (RMSE), spline consistently performed significantly worse than others. The Transformer generally maintained a better mean accuracy than linear interpolation, although significant differences were observed only under specific conditions. Conversely, for waveform preservation (correlation), the Transformer consistently outperformed linear interpolation. Notably, in highly complex imputation tasks, the Transformer proved most robust, yielding the lowest RMSE and highest correlation. However, performance systematically degraded for all methods as gap lengths increased.Conclusion.The Transformer provides an effective baseline for FHR imputation under clinical conditions, achieving a favorable balance between waveform and numerical accuracy. By clarifying the clinical limits of valid missing-segment lengths-specifically the decline in reliability beyond 30 s-this study provides guidance for standardizing preprocessing in future CTG AI research and clinical implementation.Significance.For imputing intrapartum FHR data, the Transformer generally improves waveform reproducibility over linear interpolation for short-to-moderate gaps. Defining the reliability limit provides a crucial baseline for future CTG AI.
29 June 2026
Read appraisal →European respiratory review : an official journal of the European Respiratory Society
Mucoactive agents in bronchiectasis: a systematic review and meta-analysis
BACKGROUND: Mucoactive agents aim to improve mucociliary clearance in bronchiectasis, disrupting the cycle of impaired clearance, infection and inflammation. Registry data show 28% of patients use these agents, but prescribing varies due to limited evidence. We aimed to update evidence on mucoactive agents' effects in adults with bronchiectasis. METHODS: We conducted a systematic review and meta-analysis, searching Medline, Embase, CENTRAL and trial registries to 1 October 2025. Eligible studies included adults with bronchiectasis, evaluating any mucoactive agent versus placebo or control. Cystic fibrosis and paediatric studies were excluded. The primary outcome was exacerbation frequency; secondary outcomes included lung function, quality of life and adverse events. FINDINGS: From 1512 records, 24 studies (20 randomised, four observational: n=7051) evaluated eight mucoactive agents (ambroxol, bromhexine, carbocisteine, erdosteine, hypertonic saline, mannitol, N-acetylcysteine, rhDNase). Most trials had high or some risk of bias. Seven randomised studies (n=1259) reported exacerbations; pooled analysis did not demonstrate a statistically significant difference in annualised exacerbation incidence (mean difference -0.40 per patient per year, 95% CI -1.04-0.24; p=0.22; I2=91%; very low certainty). Nine studies (n=767) reported percent predicted forced expiratory volume in 1 s (FEV1 % pred), showing a small mean increase of 3.23% (95% CI 0.31-6.15; p=0.03; I2=69.9%; very low certainty). Pooled analyses did not demonstrate significant differences for other spirometry measures, quality of life or adverse events. INTERPRETATION: Pooled evidence did not show a reduction in exacerbations and FEV1 % pred improvements were small. Evidence certainty was low/very low, meaning overall clinical benefit remains uncertain, highlighting the need for targeted trials of specific agents and subgroups.
25 June 2026
Read appraisal →NEJM evidence
A Randomized Trial of Interval Likelihood Ratios for Clinical Decision-Making
BACKGROUND: Dichotomized reporting of diagnostic test results may oversimplify test characteristics. Interval likelihood ratios (ILRs), test characteristics across a range of values, offer a nuanced alternative, but their impact on decision-making is uncertain. We assessed whether ILRs improve clinical decisions compared to two traditional formats: single-threshold and binary test characteristics. METHODS: We conducted a multicenter randomized trial of internal medicine residents, attending physicians, and advanced practice providers. Participants were presented with two fictional vignettes (cases involving pulmonary embolism diagnosis and heart failure treatment), each with a clearly correct response centered on a key diagnostic test result. The vignettes were identical except for participants' random assignment to one of three test characteristic presentations: (1) single threshold, (2) binary test characteristics, or (3) ILRs. We compared the correct management decision rates between the ILR and threshold groups and between the ILR and binary groups. Secondary objectives included comparing threshold and binary groups and subgroup analyses by case vignette. RESULTS: The trial enrolled 611 participants. The correct decision rate was 93.4% in the ILR group compared with 82.2% in the threshold group (difference of 11.2 percentage points; 95% confidence interval [CI], 6.9-15.6, P<0.001) and compared with 75.1% in the binary group (difference of 18.3 percentage points; 95% CI, 13.5-23.1, P<0.001). In the pulmonary embolism case, the correct response rate was 91.9% in the ILR group, 70.9% in the threshold group, and 64.7% in the binary group. In the heart failure case, the correct response rate was 94.9% in the ILR group, 93.4% in the threshold group, and 85.6% in the binary group. CONCLUSIONS: Presenting diagnostic test information using ILRs improved vignette-based clinical decisions. The magnitude of improvement appeared to vary by case vignette.
24 June 2026
Read appraisal →Acta orthopaedica
Registration and reporting characteristics of trials investigating exercise therapy following total knee arthroplasty: a systematic review.
BACKGROUND AND PURPOSE: Prospectively registering the primary trial outcome is important to reduce selective outcome reporting and increase the trustworthiness of findings, which guide clinical practice. The objectives of our systematic review were to explore and compare the reporting characteristics of prospectively and non-prospectively registered trials investigating exercise therapy following total knee arthroplasty. METHODS: Randomized trials comparing effects of exercise therapy after total knee arthroplasty for osteoarthritis were identified in 4 databases from 2000 to August 12, 2024. One primary outcome per trial was extracted, using a pre-specified hierarchical algorithm, irrespective of outcome domain. Pooled standardized mean differences (SMDs) were calculated on pre-specified outcome domains, and risk-of-bias assessed using the Cochrane Risk-of-Bias tool v2. RESULTS: 94 trials, comprising 9,396 participants, were included, of which 13 were prospectively registered, 33 retrospectively registered, and 48 unregistered. A single primary outcome was defined in 44% of the 94 trials, and 4 trials reported a primary outcome consistent with a prospective registration. The pooled SMD of primary outcomes was 0.06 (95% confidence interval [CI] -0.03 to 0.16) for prospectively registered trials, 0.67 (CI 0.22-1.11) for retrospectively registered trials, and 0.59 (CI 0.32-0.86) for unregistered trials. Lower risk-of-bias ratings and higher proportions of intention-to-treat adherence, dropout reporting, and adverse event reporting were observed among prospectively registered trials. CONCLUSION: Among prospectively registered trials we showed smaller effect size estimates between interventions with lower risk-of-bias ratings, and higher proportions of intention-to-treat adherence, dropout reporting, and adverse event reporting in contrast to trials without prospective registration; furthermore, clear specification of a single primary outcome was uncommon among trials evaluating exercise therapy after total knee arthroplasty.
23 June 2026
Read appraisal →PloS one
Inhaled nitric oxide in fibrotic and advanced interstitial lung disease: A systematic review and meta-analysis of randomized controlled trials.
PURPOSE: Interstitial lung disease (ILD) can lead to pulmonary hypertension (PH), contributing to reduced exercise capacity in patients with ILD. Inhaled nitric oxide (iNO) reduces pulmonary artery pressure and pulmonary vascular resistance and may improve exercise capacity in this population; however, the available evidence remains limited. Therefore, we conducted a systematic review and meta-analysis to compare the efficacy and safety of iNO versus placebo in patients with ILD. METHODS: We systematically searched PubMed, Embase, Cochrane databases, and ClinicalTrials.gov for randomized controlled trials (RCTs) comparing iNO with placebo in patients with ILD. The primary endpoint was 6-minute walk distance (6MWD). Secondary endpoint was moderate-to-vigorous physical activity (MVPA). Safety endpoint was any adverse events. Continuous endpoints were reported using mean differences (MDs) or standardized mean differences (SMDs), and binary endpoints were reported using risk ratios (RRs), all with 95% confidence intervals (CIs). RESULTS: Four RCTs, including one crossover trial, enrolling 274 patients were analyzed. Of these, 172 received iNO. The median follow-up period was 12 weeks. In the pooled analyses, iNO likely results in little to no difference in 6MWD compared with placebo (MD 1.83m; 95% CI -12.98 to 16.64; p = 0.81), and may result in little to no difference in any adverse events (RR 1.12; 95% CI 0.97 to 1.30; p = 0.12) and MVPA (SMD 0.32; 95% CI -0.93 to 1.56; p = 0.39). CONCLUSION: INO did not demonstrate improvements in 6MWD or MVPA. These findings do not support routine use of iNO in this population.
23 June 2026
Read appraisal →Journal of robotic surgery
A focused systematic review and meta-analysis of robot-assisted inguinal lymphadenectomy versus open inguinal lymph node dissection in penile cancer
This focused systematic review and meta-analysis evaluated robot-assisted inguinal lymphadenectomy (RAIL) versus open inguinal lymph node dissection (OILND) in penile cancer, aiming to provide robot-specific comparative estimates rather than a broad minimally invasive synthesis. PubMed, EMBASE, Web of Science, and the Cochrane Library were searched through December 2025 for comparative studies. All meta-analyses were conducted using STATA 18, employing random-effects modeling. Binary outcomes were summarized using odds ratios (ORs), and continuous outcomes were presented as weighted mean differences (WMDs), with all estimates accompanied by 95% confidence intervals (CIs). Risk of bias in the included non-randomized comparative studies was assessed using the ROBINS-I tool. Four retrospective comparative studies involving 276 patients were included, and no randomized controlled trials were identified. A lower odds of total postoperative complications was observed with RAIL (OR = 0.52, 95% CI: 0.27, 0.97; P = 0.047), but this borderline finding should be interpreted cautiously given the small number of retrospective studies and potential residual confounding. Across individual studies, operative time generally tended to be longer with RAIL, whereas findings for estimated blood loss and drainage-related outcomes were inconsistent. Lymph node yield, skin-related complications, minor complications, and groins with positive nodes did not differ significantly between approaches. According to the GRADE assessment, the certainty of evidence was low for most pooled outcomes and very low for skin-related complications and lower-limb edema/lymphedema. RAIL may be associated with lower overall postoperative complications than open surgery, but the evidence remains low certainty and insufficient to establish a definitive morbidity-reduction benefit. In addition, several continuous perioperative variables showed extreme between-study heterogeneity and were not clinically interpretable as reliable pooled effects.
22 June 2026
Read appraisal →Oral health & preventive dentistry
Natural Agents for the Improvement of Gingival Health: Systematic Review and Meta-Analysis of Randomized Clinical Trials
PURPOSE: This systematic review and meta-analysis evaluated the efficacy of natural ingredients used in toothpastes and gels in improving gingival health. MATERIALS AND METHODS: A comprehensive search strategy using different databases was conducted in accordance with PRISMA guidelines to identify randomized clinical trials (RCTs) on the use of natural ingredients for improving gingival health. Included studies focussed on outcomes measuring gingival health parameters, such as gingival index (GI), or bleeding on probing (BoP). Out of 249 identified studies, 10 RCTs met the inclusion criteria. Formulations included natural ingredients from plant extracts such as Aloe vera or Pudilan, and hydroxyapatite (as tooth-like calcium phosphate). RESULTS: The meta-analysis demonstrated a statistically significant improvement in gingival health when natural ingredients were used compared to control toothpastes and gels (p=0.0051). These findings suggest that natural ingredients may offer comparable or superior efficacy to conventional oral care products in terms of gingival health. Notably, Aloe vera and hydroxyapatite consistently demonstrated clinical benefits. CONCLUSIONS: Natural ingredients used in oral care products represent a promising strategy in gingivitis prevention and management. Their use in toothpastes and gels may therefore provide clinicians and patients with evidence-based, well-tolerated alternatives or adjuncts to conventional formulations for maintaining and improving gingival health.
20 June 2026
Read appraisal →World journal of urology
Non-operative management of early onset small urethrocutaneous fistula post hypospadias repair: a systematic review
BACKGROUND: Urethrocutaneous fistulas (UCF) after hypospadias repair are typically managed surgically, but conservative approaches may offer a less invasive, cost-effective alternative. While surgical repair remains the standard, non-operative strategies may eliminate anesthesia exposure, shorten recovery time, and lower healthcare costs. However, systematic evidence regarding these approaches remains limited. METHODS: This narrative systematic review evaluates non-operative management for small (typically ≤ 2 mm) UCFs. A comprehensive search was conducted across PubMed, Cochrane, ScienceDirect, Wiley, and Google Scholar up to April 2025. The study followed PRISMA guidelines and is registered in PROSPERO (CRD420251069870). We included randomized controlled trials (RCTs), comparative trials, case series, and anecdotal reports involving children (< 18 years). Risk of bias was assessed using Cochrane RoB-2, ROBINS-I, and JBI Critical Appraisal tools. RESULTS: Five studies involving 73 patients were included for analysis. Due to significant clinical heterogeneity in fistula characteristics and interventions, a pooled closure rate was not calculated. Reported individual study closure rates ranged from 53.8% to 62.5% in the primary series, with one anecdotal report (n = 1) showing 100% success. Success was more frequently observed in early-onset UCFs (diagnosed within 14 days post-catheter removal) and those with a diameter ≤ 2 mm. Minor adverse events included localized heat (42.8%) and transient dysuria (4.7%). Cost-analysis from one RCT indicated that non-operative adhesive application cost approximately 9.3% compared to total operative costs. Most included studies were classified as having a high risk of bias. CONCLUSION: Non-operative approaches using tissue adhesives, fibrin glue, or recatheterization may be viable for small (≤ 2 mm), early-onset UCFs. However, the current evidence is limited, heterogeneous, and carries a high risk of bias. These findings should be considered hypothesis-generating, and further high-quality comparative studies are required to establish definitive clinical protocols. TRIAL REGISTRATION: PROSPERO CRD420251069870.
19 June 2026
Read appraisal →The American journal of cardiology
Limus- Versus Paclitaxel-Coated Balloons for In-Stent Restenosis Treatment: A Systematic Review and Study-Level Meta-Analysis of Randomized Controlled Trials
Drug-coated balloons are the preferred treatment for in-stent restenosis (ISR), avoiding the need for a second metallic layer. While both paclitaxel-coated balloons (PCB) and limus-coated balloons (LCB, typically sirolimus or biolimus) are used, direct comparative evidence is limited, particularly regarding their long-term angiographic efficacy and clinical safety. We conducted a systematic review and study-level meta-analyzis of studies comparing LCB with PCB for coronary ISR. A comprehensive search was performed across PubMed, Embase, Cochrane, Scopus, Web of science for randomized controlled trials. Outcomes were compared using risk ratios (RR) for categorical data and mean differences for continuous data in a random-effects model. Heterogeneity was assessed using the I2 statistic. Six randomized controlled trials involving 1,038 patients were included (552 in the LCB group and 481 in the PCB group). Compared with PCB, LCB were associated with a significantly higher risk of clinically driven target lesion revascularization (RR: 1.48; 95% CI: 1.02 to 2.14; p = 0.04; I2 = 0%) and target lesion failure (RR: 1.19; 95% CI: 0.87 to 1.62; p = 0.27; I2 = 0%). In contrast, both platforms demonstrated no differences in all-cause mortality (RR: 0.98; p = 0.96), myocardial infarction (RR: 0.73; p = 0.46), stent thrombosis (RR: 0.66; p = 0.57), or MACE (RR: 1.09; p = 0.58). Angiographic outcomes were comparable, including late lumen loss (mean differences: -0.00 mm; p = 0.65; I2 = 74.7%) and minimal lumen diameter (mean differences: -0.12 mm; p = 0.07). In patients with ISR, PCB reduced repeat revascularization compared with LCB. These data suggest a modest advantage for PCB in preventing restenosis following drug-coated balloon angioplasty for ISR.
16 June 2026
Read appraisal →PloS one
Development and validation of search hedges for Transgender and Gender Diverse (TGD) populations in Ovid MEDLINE and Ovid APA PsycInfo
INTRODUCTION: This paper describes the development and validation of highly sensitive search hedges for Ovid MEDLINE and Ovid APA PsycInfo that effectively identify literature on transgender and gender diverse (TGD) populations. METHODS: Two librarians developed the search hedges using relevant keywords and controlled vocabulary terms, building on previous work on identifying transgender populations in evidence synthesis. The hedges were tested and refined to capture diverse and expansive gender identities across cultures and disciplines. The hedges were validated for sensitivity using a gold standard set of 144 articles from the Knowsy portal of evidence syntheses tagged as Two-Spirit, transgender, or gender non-binary. To assess precision an international research team of subject experts independently screened a randomized sample of search results in a two-stage screening process with an additional screener resolving disputes. RESULTS: The final search hedges demonstrated 100% sensitivity in both MEDLINE and APA PsycInfo, identifying all 144 relevant articles from the Knowsy gold standard set. The MEDLINE search hedge achieved a 71% precision, and the APA PsycInfo hedge achieved a 67% precision. These results balance comprehensive retrieval while minimizing non-relevant articles for an efficient screening process. CONCLUSIONS: These search hedges in MEDLINE and APA PsycInfo are valuable tools for researchers and librarians to more effectively identify literature on TGD populations. These tools will be crucial for ongoing work in addressing gaps in research and health disparities faced by TGD populations and will be particularly valuable for researchers conducting evidence synthesis projects related to this population.
14 June 2026
Read appraisal →Systematic reviews
Using full agreement across multiple large language models for title-and-abstract screening in systematic reviews: a proof-of-concept
BACKGROUND: The exponential growth of scientific literature poses significant challenges for conducting systematic reviews, particularly in the labor-intensive title-and-abstract screening phase. This study examines the feasibility of using multiple large language models (LLMs) for title-and-abstract screening in systematic reviews. METHODS: We propose and evaluate a full-agreement approach using three commercially available LLMs from different model families (ChatGPT, Gemini, and Claude), in which automated classification decisions are only accepted when all three models assign the same label. This approach was examined across six datasets to assess its effectiveness. A structured workflow was developed to support implementation without requiring specialized technical expertise. Additionally, a stop criterion was introduced to ensure that LLM-based classification is only applied when predefined performance thresholds are met. RESULTS: Across the six datasets, approximately four in five abstracts received full agreement across models and could therefore be classified automatically. For this subset of abstracts, classification performance was consistently higher than that of previous automated approaches using LLMs, with statistically significant improvements in all prespecified performance metrics. CONCLUSIONS: A full agreement approach across three LLMs may offer a promising and conservative strategy for title-and-abstract screening in systematic reviews by automating concordant decisions while reserving human review for discordant cases. As a proof-of-concept, the present findings support this approach as a possible workflow for reducing screening burden in the face of continued growth in the scientific literature, although its broader generalizability remains to be established.
14 June 2026
Read appraisal →JMIR research protocols
Effectiveness of Guideline-Based Clinical Decision Support Systems: Protocol for a Systematic Review
BACKGROUND: Clinical guidelines (CGs) standardize care through evidence-based recommendations, while clinical decision support systems (CDSS) can assist in applying these guidelines to individual patients. The scientific basis for the decisions offered by decision support systems is often not explicitly stated or not clearly specified in the literature on CDSS. Therefore, a systematic examination of the literature is needed to map the current state of CDSS, with a particular focus on the integration of CGs. OBJECTIVE: This study aims to systematically collect, describe, and synthesize evidence of randomized controlled studies of interventions using CDSS with a well-defined integration of evidence-based CGs and evaluating direct medical outcomes. METHODS: This systematic review adheres to the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) and PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews) checklists. The eligibility criteria for this review are defined using the patient, intervention, control, outcome, and study design framework, including studies involving patients with any medical condition or disease. Study interventions need to include guideline-based CDSS, encompassing all types of interventions used for treatment. Each intervention must provide a sufficiently accessible technical description, including the types of data and algorithms used for decision-support procedures. The guidelines used within these CDSS interventions must be derived from a clearly defined, evidence-based guideline development process published by a discernible guideline-producing body. Studies must use a randomized controlled study design. Only studies evaluating the effectiveness of the interventions on direct medical outcomes are included. Web of Science, including MEDLINE, and Scopus will be searched with search expressions aligned with the eligibility criteria. RESULTS: On August 11, 2022, the initial search was conducted on Web of Science and Scopus. From a total of 6203 records, 1347 were removed prior to screening as duplicates, 2506 records were excluded during the first screening step, and 2291 were excluded during the second step. Next, 41 papers were excluded based on full-text review, and 18 papers were finally included in the review following this initial search. This review explores whether CDSS based on CGs can improve clinical outcomes, although their effectiveness may vary depending on various factors. Potential limitations, such as high study heterogeneity, have already been identified. An update of the review has been started in April 2025. CONCLUSIONS: To our knowledge, this is the first rigorous systematic review on the effectiveness of guideline-based decision support systems in which the technical integration and algorithmic embedding of CGs have been described or can be inferred from secondary literature. With this review, we aim to address this gap by providing a detailed analysis of existing research and identifying best practices, challenges, and areas for future investigation.
14 June 2026
Read appraisal →BMJ open gastroenterology
Safety and efficacy of advanced combination therapies for treating inflammatory bowel disease in adults: a systematic review and meta-analysis
OBJECTIVES: To systematically synthesise available evidence on the safety of advanced combination therapy (ACT) in adults with inflammatory bowel disease (IBD), with secondary exploratory evaluation of efficacy outcomes. DESIGN: Systematic review and meta-analysis. DATA SOURCES: Searches of Embase (via Ovid), MEDLINE and PubMed were initially undertaken on 9 May 2025 and updated on 8 April 2026. Additional studies were identified through hand-searching and reference list screening. Eligible publications were restricted to those published between 1 January 2015 and 31 March 2026. ELIGIBILITY CRITERIA: We included randomised controlled trials as well as observational and descriptive studies. Conference abstracts, reviews, editorials and commentaries were excluded. Inclusion was restricted to publications in English. DATA EXTRACTION AND SYNTHESIS: Two reviewers independently screened studies, extracted data and assessed risk of bias using Joanna Briggs Institute tools and Cochrane RoB-1. Pooled proportions of total adverse events (TAEs), serious adverse events (SAEs) and treatment discontinuations were calculated using random-effects meta-analysis. Subgroup analyses were conducted by drug class. Where meta-analysis was not appropriate, findings were narratively synthesised. RESULTS: Fifty-two studies (n=2022 participants) were included. The most frequent combinations were anti-TNFa plus integrin inhibitors and interleukin (IL) 23 plus integrin inhibitors. Pooled analyses demonstrated low rates of SAEs (eg, anti-TNFa plus integrin inhibitors: 2.7%, 95% CI 0.22% to 6.86%) and discontinuations (6.38%, 95% CI 2.36% to 11.58%), although heterogeneity was substantial and certainty of evidence was very low, with some analyses based on small sample sizes. CONCLUSION: ACT may be associated with low rates of SAEs in selected patients with refractory IBD; however, the evidence is limited by small sample sizes, heterogeneity and predominantly observational designs. Robust conclusions regarding safety and efficacy cannot be made due to very low GRADE certainty. PROSPERO REGISTRATION NUMBER: CRD420251025883.
13 June 2026
Read appraisal →The New England journal of medicine
IVUS-Guided versus Angiography-Guided PCI in Unprotected Left Main Coronary Disease
BACKGROUND: Percutaneous coronary intervention (PCI) is increasingly used for revascularization of unprotected left main coronary artery disease. Whether intravascular ultrasonographic (IVUS) guidance during PCI results in better clinical outcomes than conventional angiographic guidance alone is uncertain. METHODS: In an international, multicenter, open-label trial, we randomly assigned patients with unprotected left main coronary artery disease in a 1:1 ratio to undergo either IVUS-guided PCI or angiography-guided PCI. The primary end point was a patient-oriented composite of any stroke, any myocardial infarction, any revascularization, or death from any cause at the longest follow-up. RESULTS: A total of 806 patients underwent randomization; 401 were assigned to undergo IVUS-guided PCI and 405 to undergo angiography-guided PCI. The mean (±SD) age of the patients was 71.4±10.7 years, 78.4% of the patients were men, and 34.7% had diabetes. At a median follow-up of 2.9 years, a primary end-point event had occurred in 135 patients (33.7%) in the IVUS-guided PCI group and in 125 patients (30.9%) in the angiography-guided PCI group (hazard ratio, 1.11; 95% confidence interval, 0.87 to 1.42; P = 0.40). The incidence of death, myocardial infarction, or revascularization appeared to be similar in the two groups. The percentages of patients with procedure-related and overall safety events also appeared to be similar in the two groups. CONCLUSIONS: Among patients with unprotected left main coronary artery disease, IVUS-guided PCI showed no additional benefit over angiography-guided PCI with respect to the incidence of stroke, myocardial infarction, any revascularization, or death from any cause at a median follow-up of 2.9 years. (Funded by Philips Image Guided Therapy Devices and Boston Scientific; OPTIMAL ClinicalTrials.gov number, NCT04111770.).
12 June 2026
Read appraisal →Medicine
Triple oral therapy combining metformin, SGLT-2 and DPP-4 inhibitors versus dual therapy in type 2 diabetes mellitus: A systematic review and meta-analysis
BACKGROUND: Triple oral therapy combining metformin, sodium-glucose cotransporter 2 inhibitor, and a dipeptidyl peptidase-4 inhibitor has been proposed as a synergistic approach to intensify glycemic control in patients with type 2 diabetes mellitus. We conducted a systematic review and meta-analysis to evaluate the efficacy and safety of triple therapy compared to dual therapy (metformin plus either sodium-glucose cotransporter 2 or dipeptidyl peptidase-4 inhibitor). METHODS: Following preferred reporting items for systematic review and meta-analysis guidelines, we searched PubMed, Embase, Scopus, and Web of Science through January 2026. Studies included randomized controlled trials comparing triple versus dual therapy in adults with type 2 diabetes mellitus. Outcomes included hemoglobin A1c (HbA1c), fasting plasma glucose, body weight, achievement of HbA1c < 7%, and adverse events (AEs). Pooled standardized mean differences (SMDs) and risk ratios (RRs) were calculated using random-effects models. RESULTS: Eight studies encompassing 2606 participants were included. Findings indicate triple therapy significantly reduced HbA1c levels compared to dual therapy, with a SMD of - 0.54 (95% confidence interval [CI]: -0.92 to -0.16; P = .005). Triple therapy resulted in greater reduction in fasting plasma glucose, with an SMD of -0.30 (95% CI: -0.62 to 0.01; P = .06). Patients on triple therapy were more likely to achieve HbA1c levels below 7% (RR: 2.02; 95% CI: 1.55-2.63; P < .0001). Weight reduction was modest, with an SMD: -0.14 (95% CI: -0.22 to -0.07; P = .0002). No significant differences were found in total AEs (RR = 0.97; P = .69) or hypoglycemia (RR = 1.32; P = .32), although there was higher discontinuation due to AEs (RR = 2.62; P = .03). CONCLUSION: Triple therapy offers superior glycemic control over dual therapy without major safety trade-offs, though tolerability may affect long-term adherence.
31 May 2026
Read appraisal →Medicine
The clinical efficacy of Astragalus-containing Chinese patent medicines in the effective treatment of heart failure: A systematic review and network meta-analysis
BACKGROUND: The incidence of cardiovascular diseases, particularly heart failure, has been rising year by year. Traditional Chinese medicines containing Astragalus have been widely used in clinical research. However, there is no definitive research on the efficacy of proprietary Chinese medicines containing Astragalus. A network meta-analysis was conducted to investigate the efficacy and safety of oral traditional Chinese medicines containing Astragalus for the treatment of heart failure. METHODS: This study searched 6 databases for clinical randomized controlled trials involving the use of traditional Chinese herbal medicines containing Astragalus alone or in combination with other drugs, with a search period from the establishment of the databases to May 27, 2025. Using the Cochrane Quality Assessment Manual, we conducted a bias analysis of the included studies, and the extracted data were analyzed using network meta-analysis in Stata 18. RESULTS: A total of 18 studies were included, involving 1584 patients with heart failure, with 16 types of traditional Chinese medicines containing Astragalus. The network meta-analysis via Surface Under the Cumulative Ranking Curve analysis indicated that: Compared to using Western medicine (WM) alone, Astragalus Granule + WM ranked best for improving overall clinical efficacy rate; Qiangxin Capsule + WM was optimal for improving six-minute walk test; Astragalus Granule + WM showed the best efficacy for reducing B-type natriuretic peptide levels; Yanxing Decoction + WM ranked first for lowering N-terminal pro-B-type natriuretic peptide levels; Xinlishen Compound + WM was the optimal combination for improving left ventricular ejection fraction; Qili Qiangxin Capsule + WM performed best for reducing left ventricular end-diastolic diameter; Yangxin Tongmai II Prescription + WM exhibited the highest efficacy for decreasing Minnesota Heart Failure Quality of Life Questionnaire scores. CONCLUSIONS: Adding oral traditional Chinese medicine containing Astragalus to WM treatment can further improve the clinical efficacy of heart failure. However, due to the limited number and quality of the included studies in this research, the above conclusions still require further validation through well-designed randomized double-blind controlled trials.
31 May 2026
Read appraisal →Philosophical transactions. Series A, Mathematical, physical, and engineering sciences
Predictability-Computability-Stability workflow for veridical data science in the age of artificial intelligence
Data science is a pillar of artificial intelligence (AI), which is transforming nearly every domain of human activity, from the social and physical sciences to engineering and medicine. While data-driven findings in AI offer unprecedented power to extract insights and guide decision-making, many are difficult or impossible to replicate. A key reason for this challenge is the uncertainty introduced by the many choices made throughout the data science life cycle (DSLC). Traditional statistical frameworks often fail to account for this uncertainty. The Predictability-Computability-Stability (PCS) framework for veridical (truthful) data science (VDS) offers a principled approach to addressing this challenge throughout the DSLC. This paper presents an updated and streamlined PCS workflow, tailored for practitioners and enhanced with guided use of generative AI (GenAI). We include a running example to display the PCS framework in action and conduct a related case study that showcases the uncertainty in downstream predictions caused by judgement calls in the data cleaning stage. This article is part of the theme issue 'Statistical workflow'.
30 May 2026
Read appraisal →Open heart
Impact of three-dimensional heart models on surgical planning and outcomes in congenital heart disease: a systematic review and meta-analysis
BACKGROUND: Congenital heart disease surgery includes intricate structures that are perhaps hard to completely understand by traditional two-dimensional shapes. Individualised, centre-based three-dimensional (3D) heart models may enhance structural imaging and surgical preparation, while data on their medical effect continue to be unpredictable and uneven. This systematic review and meta-analysis assessed the impact of 3D heart simulations on surgical preparation and outcomes in congenital cardiac disease. OBJECTIVE: To assess the successful outcomes of patient-centred 3D cardiac simulations in medical preparation, intraoperative execution and surgical results in patients experiencing operation for congenital cardiac disease. OBJECTIVE: To evaluate the impact of patient-specific 3D heart models on surgical planning precision and preoperative outcomes in congenital heart disease surgery. METHODS: A systematic review and meta-analysis were performed in accordance with Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 standards and prospectively registered in PROSPERO (CRD42024211985). PubMed, Embase, Scopus and Cochrane CENTRAL were searched through March 2025 for comparative studies evaluating 3D-printed or virtual heart models versus conventional imaging. Outcomes were harmonised and pooled as risk ratios (RRs), mean differences (MDs) or standardised MDs (SMDs) with 95% CIs using random-effects models. RESULTS: 32 studies (n=1842) met inclusion criteria. The use of 3D models was associated with more frequent surgical plan modification (RR 1.42; 95% CI 1.21 to 1.67), reduced operative time (MD -18.4 min; 95% CI -27.6 to -9.2), lower postoperative complications (RR 0.74; 95% CI 0.56 to 0.98) and reduced re-operation (RR 0.52; 95% CI 0.31 to 0.86). Hospital stay was shortened (MD: 1.8 days; 95% CI -3.0 to -0.6), and surgeon confidence improved (SMD 0.88; 95% CI 0.60 to 1.15). The confidence of indication extended from modest to small because of non-experimental research and variability. CONCLUSION: Individualised-centred 3D heart stimulation improves structural imaging and medical preparation in congenital cardiac disease, chiefly for intricate lacerations and can increase preoperative competence, while statistics for conclusive diagnosis results continue to be restricted and mainly experiential. Upcoming potential, large-scale investigator-led research with consistent result events are required to ratify their medical success.
26 May 2026
Read appraisal →European journal of clinical pharmacology
Ondansetron augmentation for obsessive-compulsive disorder: A systematic review and meta-analysis of randomized placebo-controlled trials
BACKGROUND: A substantial proportion of adults with obsessive-compulsive disorder (OCD) experience persistent symptoms despite optimized first-line treatment with exposure and response prevention and/or serotonin reuptake inhibitors (SRIs). Ondansetron, a selective 5-hydroxytryptamine-3 (5-HT3) receptor antagonist with plausible circuit-level relevance to cortico-striato-thalamo-cortical dysfunction, has been evaluated as an augmentation strategy. Still, evidence is dispersed across small randomized trials. METHODS: We conducted a PRISMA 2020-adherent systematic review and random-effects meta-analysis (PROSPERO: CRD420261302132). Embase (Ovid and Embase.com), MEDLINE, PubMed, and Scopus were searched from inception to 12 February 2026. Eligible studies were randomized, double-blind, placebo-controlled trials of oral ondansetron augmentation in adults (≥ 18 years) with OCD on stable background treatment. The primary outcome was change in Yale-Brown Obsessive-Compulsive Scale (Y-BOCS) total score, pooled at prespecified acute windows (week 4, ~week 8 primary, week 12). Secondary outcomes included Y-BOCS obsession and compulsion subscales. Risk of bias was assessed using RoB 2. Random-effects models (DerSimonian-Laird) were used; Wan's method was applied when required. RESULTS: Five trials (total N = 289) were included. Ondansetron augmentation was associated with greater improvement in Y-BOCS total score versus placebo at week 4 (MD 2.63; 95% CI 0.87-4.39; I²=62%); heterogeneity resolved after excluding one influential study (MD 1.94; 0.37-3.51; I²=0%). At week 8 (four trials; N = 244), benefit was larger (MD 5.53; 3.45-7.60; I²=48%) and remained robust in sensitivity analysis (MD 6.76; 5.81-7.71; I²=0%). At week 12 (three trials; N = 200), effects persisted (MD 6.41; 4.28-8.53; I²=55%) with reduced heterogeneity after excluding one study (MD 5.06; 2.82-7.30; I²=0%). In two trials reporting subscales (N = 214), obsessions improved at weeks 4 and 8 (overall MD 2.67; 1.66-3.67; I²=0%), while compulsions showed a smaller, later signal (overall MD 1.02; 0.19-1.86; I²=0%). Safety reporting was variably detailed across trials; known QT-prolongation considerations remain relevant for off-label use. CONCLUSIONS: In randomized, placebo-controlled trials, ondansetron augmentation was associated with greater improvement in clinician-rated OCD severity than placebo over short-term follow-up, with signals emerging by week 4 and persisting through 8-12 weeks. However, these findings should be interpreted cautiously because the evidence base is small, geographically concentrated, and partly influenced by individual studies. The current evidence should therefore be considered preliminary and hypothesis-generating rather than definitive. Larger multicenter trials with standardized background SRI optimization, preregistered outcomes, and rigorous safety monitoring are required before routine clinical adoption can be considered.
25 May 2026
Read appraisal →Journal of gastrointestinal cancer
Optimal First-line Immune-related Therapy Selection for Advanced Gastric Cancer: A Systematic Review and Network Meta-analysis
PURPOSE: Although immune checkpoint inhibitors combined with chemotherapy improve survival in advanced gastric cancer (AGC), comparative evidence across first-line regimens remains insufficient. This study aimed to systematically evaluate their relative efficacy and safety. METHODS: PubMed, Embase, the Cochrane Library, Web of Science, and major conferences were searched up to February 28, 2025. Eligible RCTs comparing ICI-related regimens as first-line AGC treatment were included. Primary outcomes were overall survival (OS) and progression-free survival (PFS). RESULTS: Eleven RCTs involving 8,999 patients, six integrated strategies, and twelve specific regimens were included. Dual-target ICI-chemo and single ICI-chemo showed significant survival benefits over chemotherapy, whereas dual ICIs±chemo and single ICI alone did not. Dual-target ICI-chemo was associated with significantly better survival than both dual ICIs±chemo and single ICI±chemo. Among the specific regimens, cadon-chemo achieved the most favorable OS and PFS. No significant OS differences were observed among the single ICI-chemo regimens. In PD-L1-positive subgroups, both dual-target ICI-chemo and single ICI-chemo significantly improved survival compared with chemotherapy, with cadon-chemo being most effective in CPS ≥ 10 and CPS ≥ 1, and SHR1701-chemo in CPS ≥ 5. In the CPS < 5 subgroup, dual-target ICI-chemo significantly improved OS, while single ICI-chemo showed a trend toward benefit. However, no ICI-based strategy demonstrated benefit in CPS < 1. CONCLUSIONS: Dual-target ICI-chemo appears to be the most effective first-line strategy for AGC based on current indirect evidence, surpassing single ICI-chemo and dual ICI-based regimens. These findings support personalized therapy selection but await confirmation by head‑to‑head trials. Future studies should address CPS < 1 and explore optimized combinations to enhance outcomes.
25 May 2026
Read appraisal →Behavior research methods
Automating data extraction in meta-research: A multi-model benchmark in network psychometrics papers
Manual data extraction in meta-research is often tedious, time-consuming, and error-prone. In this paper, we investigate whether the current generation of large language models (LLMs) can be used to extract accurate information from scientific papers. Across the meta-research literature, these tasks usually range from extracting verbatim information (e.g., the number of participants in a study, effect sizes, or whether a study is preregistered) to making subjective inferences. Using a publicly available dataset containing a wide range of metascientific variables from 43 network psychometrics papers, we tested five LLMs (Claude 4.6 Opus, Claude 4.5 Sonnet, Claude 4.5 Haiku, GPT-5.2, and GPT-5 mini). We used an automated API-based pipeline to extract variables from the documents. This approach allows batch processing of research papers. As such, it represents a more efficient and scalable way to extract metascientific data than the default chat interface. The extraction accuracy ranged from 79.6% to 91.3% across the models. The extraction performance was generally higher for more explicit, verbatim information and worse for variables that required more complicated inference. Furthermore, most models were able to convey uncertainty in more contentious cases. We provide a comparison of the accuracy and cost-effectiveness of the individual models and discuss the characteristics of variables that are and are not suitable for automatic coding. Furthermore, we describe some of the common pitfalls and best practices of automated LLM data extraction. The proposed procedure can substantially reduce the time and costs associated with conducting meta-research.
25 May 2026
Read appraisal →Journal of ethnopharmacology
Comparative efficacy and safety of Chinese patent medicines in coronary heart disease patients after percutaneous coronary intervention: A systematic review and network meta-analysis
ETHNOPHARMACOLOGICAL RELEVANCE: Patients with coronary heart disease (CHD) remain at high risk of major adverse cardiovascular events (MACEs) after percutaneous coronary intervention (PCI). Chinese Patent Medicines (CPMs) have demonstrated therapeutic potential as adjunctive therapies, however, direct comparative evidence regarding their effectiveness is lacking. AIM OF THE STUDY: This network meta-analysis (NMA) aims to systematically evaluate and compare the clinical efficacy and safety of CPMs combined with conventional therapy (CT) in post-PCI management. MATERIALS AND METHODS: A systematic literature search was performed in PubMed, Web of Science, Embase, Cochrane Library, CNKI, CBM, VIP, and Wanfang databases up to December 24, 2024. Randomized controlled trials investigating CPMs in CHD patients following PCI were included. Methodological quality was assessed using the Cochrane Risk of Bias tool. Outcome measures included major adverse cardiovascular events (MACEs), left ventricular ejection fraction (LVEF), left ventricular end-diastolic diameter (LVEDD), high-sensitivity C-reactive protein (hs-CRP), N-terminal pro-brain natriuretic peptide (NT-proBNP), and nitric oxide (NO) levels. NMA was implemented using STATA 16.0 and R 4.1.2. Odds ratios and standardized mean differences with 95% confidence intervals were calculated for dichotomous and continuous outcomes, respectively. Model selection was guided by heterogeneity assessed using the I2 statistic. Intervention rankings were estimated using the surface under the cumulative ranking curve (SUCRA), with robustness and publication bias evaluated through sensitivity analyses and funnel plots. RESULTS: A total of 84 RCTs involving 10,524 patients and eight CPMs were included. For the primary outcome (MACEs), the SUCRA ranking indicated that Tongxinluo capsules combined with conventional therapy (TXLJN + CT) was the most effective intervention, followed by Shensong Yangxin capsules plus CT (SSYXJN + CT), Shexiang Baoxin pills plus CT (SXBXW + CT), Qili Qiangxin capsules plus CT (QLQXJN + CT), Qishen Yiqi dripping pills plus CT (QSYQDW + CT), Xinyu capsules plus CT (XYJN + CT), Fufang Danshen Dripping Pills plus CT (FFDSDW + CT), and Guanxin Shutong capsules plus CT (GXSTJN + CT), while CT alone ranked last. Correspondingly, the NMA demonstrated that TXLJN + CT was associated with the greatest reduction in LVEDD (SMD -1.01, 95% CI -1.35 to -0.66) and hs-CRP levels (SMD -2.50, 95% CI -3.33 to -1.68). The most significant improvement in LVEF (SMD 1.25, 95% CI 0.85 to 1.64) and NO levels (SMD 1.02, 95% CI 0.73 to 1.29) was observed with QSYQDW + CT. Meanwhile, SXBXW + CT showed the greatest efficacy in reducing NT-proBNP levels (SMD -3.43, 95% CI -4.68 to -2.18). CONCLUSION: CPMs combined with CT may provide clinical benefits in post-PCI management. According to SUCRA rankings, TXLJN + CT ranked first in reducing MACE and in lowering LVEDD and hs-CRP levels. QSYQDW + CT ranked first in improving LVEF and NO levels, while SXBXW + CT showed the highest ranking for reducing NT-proBNP levels. However, larger-scale, high-quality RCTs are needed to validate these findings.
25 May 2026
Read appraisal →The Cochrane database of systematic reviews
Early erythropoiesis-stimulating agents in preterm or low-birthweight infants
RATIONALE: Preterm and low-birthweight infants are at risk of anemia due to a variety of factors, including low levels of erythropoietin. They may therefore benefit from erythropoiesis-stimulating agents (ESAs). However, controversy remains about their effectiveness and safety, not only in reducing blood transfusions but also in improving clinical outcomes. OBJECTIVES: To assess the benefits and harms of early administration of ESAs in preterm or low-birthweight infants. SEARCH METHODS: We searched CENTRAL, MEDLINE, Embase, Scopus, DOAJ, and trial registries to 31 March 2025, and checked the reference lists of studies and systematic reviews for potentially relevant studies. ELIGIBILITY CRITERIA: Randomized controlled trials (RCTs) comparing early (initiated before eight days of life) ESAs (erythropoietin or darbepoetin) to placebo or no intervention among preterm (< 37 weeks' gestation) or low-birthweight (< 2500 g) infants. OUTCOMES: Outcomes included mortality during initial hospital stay (all-cause mortality), moderate to severe neurodevelopmental impairment (NDI) at 18 to 26 months' corrected age, proportion of infants exposed to one or more red blood cell (RBC) transfusions, retinopathy of prematurity (ROP), necrotizing enterocolitis (NEC), and severe intraventricular hemorrhage (sIVH). RISK OF BIAS: We used the Cochrane RoB 1 tool to assess risk of bias. SYNTHESIS METHODS: We performed data collection and analyses according to the methods of Cochrane Neonatal (fixed-effect meta-analysis using the inverse-variance method). We analyzed categorical data using risk ratio (RR) and continuous data using mean difference (MD), and reported the 95% confidence interval (CI) on all estimates. We used GRADE to assess the certainty of evidence. INCLUDED STUDIES: We included 37 studies (n = 6724), five (n = 3052) of which are new. Two studies (n = 752) investigated darbepoetin, and 35 studies (n = 5972) investigated erythropoietin. SYNTHESIS OF RESULTS: ESAs compared to placebo or no intervention Mortality: ESAs probably result in little to no difference in mortality during initial hospital stay (RR 0.93, 95% CI 0.80 to 1.09; I² = 0%; 24 studies, 5217 participants; moderate-certainty evidence). Subgroup analysis suggests that early use of erythropoietin probably results in little to no difference in mortality (RR 0.91, 95% CI 0.77 to 1.09; I² = 0%; 23 studies, 4505 participants), and early use of darbepoetin probably results in little to no difference in mortality (RR 1.00, 95% CI 0.71 to 1.43; I² = 0%; 2 studies, 712 participants). NDI: ESAs may reduce moderate to severe NDI at 18 to 26 months' corrected age (RR 0.81, 95% CI 0.68 to 0.95; I² = 86%; 3 studies, 1707 participants; low-certainty evidence). Subgroup analysis suggests that early use of erythropoietin may reduce moderate to severe NDI at 18 to 26 months' corrected age (RR 0.79, 95% CI 0.66 to 0.95; I² = 93%; 2 studies, 1239 participants), while early use of darbepoetin may result in little to no difference in moderate to severe NDI at 18 to 26 months' corrected age (RR 0.87, 95% CI 0.59 to 1.28; I² not applicable; 1 study, 468 participants). Transfusions: ESAs may reduce the proportion of infants exposed to one or more RBC transfusions (RR 0.79, 95% CI 0.76 to 0.83; I² = 56%; 21 studies, 3507 participants; low-certainty evidence). Subgroup analysis suggests that erythropoietin may reduce the proportion of infants exposed to one or more RBC transfusions (RR 0.80, 95% CI 0.76 to 0.84; I² = 56%; 20 studies, 2795 participants); darbepoetin may reduce the proportion of infants exposed to one or more RBC transfusions (RR 0.75, 95% CI 0.68 to 0.84; I² = 0%; 2 studies, 712 participants). ROP (any stage): ESAs result in little to no difference in incidence of ROP (RR 0.91, 95% CI 0.83 to 1.00; I² = 0%; 14 studies, 3844 participants; high-certainty evidence). Subgroup analysis suggests the use of either erythropoietin (RR 0.92, 95% CI 0.84 to 1.02; I² = 0%; 13 studies, 3230 participants) or darbepoetin (RR 0.83, 95% CI 0.58 to 1.17; I² = 0%; 2 studies, 614 participants) results in little to no difference in incidence of ROP. NEC (any stage): ESAs probably reduce the incidence of NEC (RR 0.74, 95% CI 0.62 to 0.87; I² = 0%; 20 studies, 5749 participants; moderate-certainty evidence). Subgroup analysis suggests that early use of erythropoietin probably reduces the incidence of NEC (RR 0.72, 95% CI 0.61 to 0.85; I² = 0%; 19 studies, 5041 participants), while early use of darbepoetin probably results in little to no difference in incidence of NEC (RR 1.08, 95% CI 0.54 to 2.15; I² = 0%; 2 studies, 708 participants). sIVH (grade ≥ 3): ESAs probably reduce the incidence of sIVH (RR 0.70, 95% CI 0.56 to 0.89; I² = 43%; 9 studies, 2458 participants; moderate-certainty evidence). Subgroup analysis suggests that early use of erythropoietin probably reduces the incidence of sIVH (RR 0.72, 95% CI 0.57 to 0.91; I² = 47%; 9 studies, 2396 participants), while early use of darbepoetin probably results in little to no difference in incidence of sIVH (RR 0.40, 95% CI 0.11 to 1.41; I² not applicable; 1 study, 62 participants). Of note, the test for subgroup differences between erythropoietin and darbepoetin revealed no evidence of a difference for the outcomes of mortality, moderate to severe NDI, NEC, sIVH, ROP, and the proportion of infants exposed to RBC transfusion. We downgraded the certainty of evidence for inconsistency, imprecision, and high risk of bias in the included studies. AUTHORS' CONCLUSIONS: Early use of ESAs probably results in little to no difference in mortality (moderate-certainty evidence); may reduce moderate to severe NDI at 18 to 26 months' corrected age (low-certainty evidence); has little to no effect on ROP (high-certainty evidence); may reduce the proportion of infants exposed to one or more RBC transfusions (low-certainty evidence); and probably reduces both NEC and sIVH (moderate-certainty evidence). Given the benefits of ESAs, future research should focus on cost-effectiveness, equity, feasibility of implementation, and acceptability to different stakeholders of the routine use of erythropoietin or darbepoetin among preterm or low-birthweight infants. FUNDING: No funding was received for this review. REGISTRATION: Protocol (and previous versions) available via 10.1002/14651858.CD004863; 10.1002/14651858.CD004863.pub2; 10.1002/14651858.CD004863.pub3; 10.1002/14651858.CD004863.pub4; 10.1002/14651858.cd004863.pub5; 10.1002/14651858.CD004863.pub6.
25 May 2026
Read appraisal →Postgraduate medical journal
Peer review in the age of artificial intelligence: a comparative study of human and AI-generated review reports
BACKGROUND: Peer review is central to maintaining scientific quality and helps editors make decisions. However, the volume of scientific publications continues to rise, placing pressure on the peer review system. With the rise of generative AI, its role in supporting peer review is gaining attention. This study aims to compare human-written and AI-generated peer review reports. METHODS: We analysed 398 peer review reports linked to 119 research articles published in BMJ Open in 2024. Publicly available reports and manuscripts were included. Editorials, corrections, and protocols were excluded. AI reports were generated using ChatGPT. All reports were anonymized and assessed by two independent reviewers. We conducted a hybrid thematic analysis. Frequencies of themes were calculated and compared by reviewer type. For quantitative comparison, we used the Mann-Whitney U test to assess differences in review quality scores and Fisher's exact test to compare the distribution of themes. All analyses were conducted using R software. RESULTS: Human reviewers gave more detailed and diverse comments. They addressed deeper issues like interpretation, originality, and applicability. AI reviews covered more sections but focused on routine or structural elements. AI outperformed slightly in format-related domains. Co-occurrence analysis showed human reviews linked diverse themes, while AI comments were structurally clustered. Shannon Index confirmed that human reviews were more thematically diverse. CONCLUSIONS: AI can support peer review by screening for basic errors. However, it lacks insight, critical judgment, and contextual awareness. Human input remains essential for meaningful review. Review-specific AI tools that preserve confidentiality are needed for future integration.
22 May 2026
Read appraisal →BMJ (Clinical research ed.)
Concordance between target trial emulation and randomised controlled trials: systematic review and meta-analysis
OBJECTIVES: To assess the concordance of results between paired target trial emulations and their benchmarking randomised controlled trials and to explore determinants of emulation success. DESIGN: Systematic review and meta-analysis. DATA SOURCES: Medline, EMBASE, Scopus, PsycINFO, and Web of Science (2010 to 2025). STUDY SELECTION: Observational studies that explicitly aimed to emulate target randomised controlled trials of a health intervention were included. DATA EXTRACTION: Data were extracted on study domain, target population, intervention, comparator, and main outcomes for each emulation and corresponding randomised controlled trial. MAIN OUTCOME MEASURES: For 21 predefined emulation design features, concordance was assessed using Pearson correlation coefficients, standardised difference agreement, and ratio of ratios, with subgroup and regression analyses of study characteristics and concordance. RESULTS: Among 107 target trial emulation-randomised controlled trial pairs, the Pearson correlation coefficient was 0.59 (95% confidence interval (CI) 0.45 to 0.70), standardised difference agreement was 79% (85/107), and summary ratio of ratios was 0.96 (95% CI 0.92 to 1.01; I2=36%). In 63 pairs with closer emulation of trial design, a higher level of agreement was observed, with a Pearson correlation coefficient of 0.83 (0.73 to 0.89) and standardised difference agreement of 87% (55/63). In subgroup analysis, trial emulations tended to systematically underestimate treatment effects from randomised controlled trials for specific outcomes related to venous thromboembolism and major adverse cardiovascular events, while overestimating those for respiratory outcomes, with pooled ratio of ratios of 0.76 (0.58 to 1.00; I2=40%), 0.91 (0.86 to 0.96; I2=32%), and 1.20 (1.03 to 1.40; I2=40%), respectively. Moreover, emulations based on claims data tended to underestimate the treatment effects (ratio of ratios 0.90, 0.82 to 0.99; I2=38%). In univariate regression analyses, poorer target trial emulation-randomised controlled trial concordance was associated with emulation designs with imbalanced baseline population characteristics, treatment started in hospital, and lower outcome emulation quality. CONCLUSIONS: Current target trial emulations achieve moderate concordance when replicating their corresponding randomised controlled trials. Concordance could be improved by better emulation design, including better baseline and outcome emulation and leveraging multisource linked databases. SYSTEMATIC REVIEW REGISTRATION: PROSPERO CRD42024619811.
20 May 2026
Read appraisal →The British journal of general practice : the journal of the Royal College of General Practitioners
Does recommending healthy diets in general practice benefit secondary prevention of cardiovascular disease? A systematic review and meta-analysis.
BACKGROUND: Suboptimal diet is the main preventable risk factor globally for cardiovascular disease (CVD), posing major burden for primary care. However, healthy dietary recommendations' evidence in patients with established CVD is unclear. AIM: Systematically review and meta-analyse dietary patterns' effect on CVD patients' all-cause and CVD mortality, and CVD recurrence. METHOD: Publication databases were searched for randomised controlled trials (RCTs) (≥6 months follow-up) and prospective cohort studies. Random-effects meta-analysis calculated relative-risks (RR) and 95% confidence-intervals (95% CI). Cochrane bias risk tools assessed study quality. RESULTS: Screening 28 291 abstracts identified 41 articles meeting inclusion criteria. Nineteen articles from 14 unique RCTs conducted 1956-2020, enrolling 14 065 CVD patients, compared fat-replacement, Mediterranean, low-fat, low-sodium, and high-fibre diets with minimal/no dietary advice. Pooled results showed no diet intervention significantly impacted mortality/CVD outcomes. However, stratified meta-analysis revealed a Mediterranean-diet significantly reduced CVD mortality (RR: 0.40; 95% CI: 0.19, 0.83). Twenty-two cohort studies, conducted 1976-2017 for 104 386 CVD patients, predominantly assessed a Mediterranean diet (60%), showing greatest healthy dietary pattern adherence significantly reduced all-cause mortality by 29% (0.71; 0.65, 0.78), CVD mortality by 25% (0.75; 0.66, 0.84), and CVD recurrence by 17% (0.83; 0.75, 0.92). Overall, there was considerable between-study heterogeneity (RCTs: I2 = 34-60%, cohorts: I2 = 49-76%) partly explained by disparate geographical locations, and studies were mainly classified as of some concerns (65% RCTs) or high bias risk (52% cohorts). CONCLUSION: GPs providing healthy dietary advice, particularly the Mediterranean diet, likely enhances CVD secondary prevention. However, cogent studies with clearly defined diets are required to bolster the evidence.
16 May 2026
Read appraisal →BMJ open
Regular human insulins versus rapid-acting insulin analogues in children and adolescents with type 1 diabetes: a systematic review with meta-analysis and trial sequential analysis
OBJECTIVES: To assess the beneficial and harmful effects of regular human insulins versus rapid-acting insulin analogues in children and adolescents with type 1 diabetes. DESIGN: Systematic review of randomised clinical trials with meta-analysis and trial sequential analysis. DATA SOURCES: CENTRAL, MEDLINE, Embase, LILACS and other sources from inception to 30 January 2026. STUDY SELECTION: Randomised clinical trials comparing regular human insulins versus rapid-acting insulin analogues (insulin aspart, lispro, glulisine) in children and adolescents with type 1 diabetes. ANALYSES: Data were analysed using meta-analysis and trial sequential analysis. Risk of bias was assessed using the Cochrane Risk of Bias tool, V.2, and the certainty of the evidence was assessed using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach. PRIMARY OUTCOMES: Severe hypoglycaemia, ketoacidosis and serious adverse events. RESULTS: 10 trials randomising 1107 participants were included. The certainty of evidence was very low mainly due to high risk of bias and small sample sizes. Meta-analysis showed no evidence of a difference between regular human insulins and rapid-acting insulin analogues on severe hypoglycaemia (risk ratio (RR) 1.28, 95% CI 0.81 to 2.03; I2=0.0%; p=0.2851; nine trials), ketoacidosis (RR 0.88, 95% CI 0.26 to 2.93; I2=0.0%; p=0.8593; two trials) and serious adverse events (RR 1.00, 95% CI 0.44 to 2.25; I2=0.0%; p=0.9958; two trials). Trial sequential analysis showed that all meta-analyses of primary outcomes were underpowered. CONCLUSIONS: Current research shows no differential effects between regular human insulins and rapid-acting insulin analogues for children and adolescents with type 1 diabetes, but the evidence is very uncertain. PROSPERO REGISTRATION NUMBER: CRD42024508625.
16 May 2026
Read appraisal →JMIR research protocols
Evaluating the Methodological Quality of Artificial Intelligence-Assisted Systematic Reviews: Protocol for a Mixed Methods Meta-Research Study
BACKGROUND: Artificial intelligence (AI), including large language models (LLMs), is increasingly integrated into systematic review (SR) workflows. AI tools may accelerate searching, screening, data extraction, and reporting, but their effects on methodological quality, reporting completeness, transparency, and reproducibility remain uncertain. Existing evaluations largely examine isolated tasks, and inconsistent disclosure of AI use limits reproducibility and oversight. OBJECTIVE: This 4-phase mixed methods meta-research study will (1) compare the methodological quality of AI-assisted versus traditional SRs; (2) refine, finalize, and apply a preliminary AI Transparency and Disclosure Index (AITDI); (3) evaluate reproducibility by comparing outputs across repeated runs of the same AI model, across different AI models, and between AI models and human reviewers at multiple SR stages; and (4) explore knowledge user perspectives on rigor, transparency, and trust in AI-assisted SRs. METHODS: We will conduct a matched cohort analysis of SRs published from 2023 to 2025 in biomedical journals. Each AI-assisted SR will be matched 1:2 with traditional SRs by publication year, clinical domain, review type, and meta-analysis status. Two independent reviewers will apply A Measurement Tool to Assess Systematic Reviews, version 2 (AMSTAR 2; methodological quality), PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) 2020 (reporting completeness), and, when applicable, Risk of Bias in SRs (ROBIS; risk-of-bias rigor). A preliminary AITDI will be refined and then applied to all AI-assisted SRs. Reproducibility will be assessed using SR-derived task sets to compare outputs across repeated runs of the same model, across different models, and between AI and human reviewers at key SR stages. Semistructured interviews with authors, editors, clinicians, policymakers, and patient partners will be analyzed using reflexive thematic analysis. RESULTS: As of December 2025, the study has been preregistered on the Open Science Framework (OSF; DOI: 10.17605/OSF.IO/Q5JRW), the search strategy has been finalized, and title/abstract screening has begun. Data extraction is planned for March-May 2026, followed by AITDI refinement and reproducibility testing from May 2026 to October 2026. Qualitative interviews are anticipated from October 2026 to February 2027, with final analyses by April 2027 and dissemination planned for mid-2027. CONCLUSIONS: This study will provide one of the first empirical comparisons of methodological quality, transparency, and reproducibility of AI-assisted versus traditional SRs in the LLM era. Findings will inform expectations for responsible AI integration and support refinement of reporting and methodological best practices, including future development of AI-specific reporting and appraisal extensions (eg, PRISMA-LLM [Preferred Reporting Items for Systematic Reviews and Meta-Analyses-large language model] and AMSTAR-LLM [A Measurement Tool to Assess Systematic Reviews-large language model]).
16 May 2026
Read appraisal →European journal of heart failure
Surrogate markers of clinical endpoints in clinical trials of heart failure.
AIMS: We aimed to determine whether treatment related changes in surrogate markers predict longer-term therapeutic effects on all-cause mortality and heart failure (HF) hospitalization in randomized trials of HF. METHODS: Systematic literature search included randomized trials of interventions in patients with HF until October 2024. Random-effects meta-regression models with inverse variance weighting were calculated between fifteen surrogate markers and the clinical endpoints of (i) all-cause mortality and (ii) HF hospitalization. The degree of heterogeneity of each model explained by the surrogate marker was determined with R2. Surrogate threshold effects (STEs) were also calculated. RESULTS: Ninety-six randomized trials with median follow-up 12 months were included, enrolling a total of 120 304 patients with HF. Treatment related changes in natriuretic peptides (NPs) were weakly correlated with all-cause mortality (64 comparisons, P = .002, R2 = 12%, and STE = -34%) and HF hospitalization (41 comparisons, P < .001, R2 = 32%, and STE = -30%). Changes in LVEF were moderately correlated with mortality (69 comparisons, P < .001, R2 = 59%, and STE = +6%) and strongly correlated with HF hospitalization (45 comparisons, P < .001, R2 = 90%, and STE = +2%), although this finding was only observed in trials of HFrEF. LV end-diastolic diameter was correlated with HF hospitalization but not mortality. Other echocardiographic markers were not predictive of clinical endpoints. Treatment related changes in patient reported outcomes and exercise capacity were either weakly correlated with clinical outcomes or derived from small clinical trials. CONCLUSION: In HF trials, most surrogate markers, including NPs and echocardiographic measures were only weakly or moderately correlated with treatment effects on all-cause mortality. There is insufficient evidence to support the use of a single biomarker or echocardiographic measure in regulatory trials of HF.
14 May 2026
Read appraisal →BMC medical research methodology
When to stop reviewing: validation of stop criteria in ASReview.
BACKGROUND: Systematic reviews are essential for evidence-based research, but they often require a great deal of time and effort. Although title and abstract (T&A) screening is just one part of the review process, it can be very time-consuming when search strategies retrieve large numbers of records. Given the exponential growth of scientific publications in recent decades, tools such as ASReview, which use machine learning (ML) for active learning-based screening, aim to reduce the workload. However, since ASReview helps users prioritise potentially relevant studies rather than supporting them in screening all records, a key question arises: at what point can the screening process safely stop without risking the omission of relevant studies when the entire dataset is not being reviewed? METHODS: This simulation study tested three proposed stop criteria for terminating screening in ASReview without loss of relevant data: (1) stopping after a calculated number of relevant studies based on an initial sample; (2) stopping after a fixed number of consecutively studies deemed irrelevant; (3) stopping after a predefined percentage of the dataset has been screened. A total of 35,000 automated title and abstract screenings were conducted using five datasets from the SYNERGY repository. Key outcomes included the percentage of studies screened until the last relevant study was found and the number of relevant studies missed under each stop criterion. RESULTS: The proportion of the dataset that needed to be screened to identify all relevant studies (as pre-classified in the SYNERGY dataset) varied greatly across datasets, ranging from 2.9% to 76.9% on average. None of the tested stop criteria could consistently identify all relevant studies across all datasets. Stop criterion 1 was reliable in only 2% of simulations. Stop criterion 2 showed high variability, with thresholds ranging from 2% to 61%, depending on the dataset. Stop criterion 3 failed to define a universal percentage applicable across datasets. CONCLUSIONS: ASReview can reduce screening workload by prioritizing potentially relevant studies through ML-based ranking, thereby allowing researchers to identify relevant studies earlier in the screening process. However, no stop criterion reliably ensures that all relevant studies are identified. Early stopping may result in missed studies, depending on dataset characteristics. Current stop criteria should be applied cautiously and potentially combined with quality assurance measures. Further research is needed to develop more robust and generalizable stopping rules. TRIAL REGISTRATION: Not applicable - this is a simulation study, not a registered systematic review.
13 May 2026
Read appraisal →European journal of pediatrics
Prognosis of neonates receiving invasive mechanical ventilation in low-resource settings: a systematic review and prognostic meta-analysis
Provision of invasive mechanical ventilation (IMV) in the neonatal intensive care has seen a steady rise in low-esource settings (LRS). However, outcomes among those exposed to IMV remain under-reported, with the current evidence base being restricted to single-centre observational studies, thus limiting comparative analyses and effective healthcare planning. This study aims to estimate the pooled proportion of mortality and morbidity among neonates exposed to IMV in low-resource settings. Medline, Embase, and CENTRAL were searched until 22 August 2025. Randomised and non-randomised studies were included. Two reviewers, blinded to each other, extracted data independently. Proportion-based meta-analyses using random-effects model were performed. Risk of bias was assessed using ROBINS-E, and evidence-certainty was evaluated using the GRADE approach. One hundred of 117 studies were included, with most conducted in South Asia. In-hospital mortality was reported in 68 studies (7193 neonates), with a pooled estimate of 45% (39%-50%), evidence-certainty being very low. Among the secondary outcomes, the pooled rates were as follows: bronchopulmonary dysplasia, 10% (5%-18%); intraventricular haemorrhage (any grade), 10% (5%-19%); necrotising enterocolitis (any stage), 14% (6%-31%); retinopathy of prematurity (any stage), 33% (22%-46%); ventilator-associated pneumonia, 21% (14%-29%); sepsis, 32% (25%-40%) and pulmonary haemorrhage, 9% (6%-14%). Evidence-certainty for all the secondary outcomes was also very low. Subgroup analysis comparing two distinct time epochs revealed a significant difference in mortality, 43% (36%-50%) (I2 = 93.5%) in the post-2010 epoch compared to 55% (48%-63%) (I2 = 82.9%) in the pre-2010 epoch (p = 0.004). Heterogeneous outcome definitions and predominance of unadjusted analyses across studies limit the existing evidence.Conclusions: In LRS, the mortality and morbidity rates among neonates receiving IMV remain substantially high with a modest improvement in survival in the past decade. Improving outcomes mandates moving beyond access to the provision of IMV to investing in comprehensive training and scaling up critical auxiliary resources. Future research must adopt standardised outcome definitions and adjusted analyses to precisely quantify the impact of IMV in LRS.
11 May 2026
Read appraisal →Graefe's archive for clinical and experimental ophthalmology = Albrecht von Graefes Archiv fur klinische und experimentelle Ophthalmologie
Comparing the performance of four mainstream large language models on medical literature review generation: a human expert evaluation in SMILE surgery
PURPOSE: To systematically evaluate and compare the performance of four leading large language models (LLMs) in generating medical literature reviews across topics of varying research maturity, thereby providing insights for their effective and responsible application in academic writing. METHODS: In this comparative study, using standardized prompts, we instructed four leading LLMs (GPT-4, Gemini 2.5 Pro, Grok-3, and DeepSeek R1) to generate literature reviews on nine topics related to small incision lenticule extraction (SMILE) surgery. These topics were categorized into three groups by research maturity: well-researched, controversial, and open. Seven ophthalmology experts evaluated the generated content across four dimensions: quality, accuracy, bias, and relevance, while all references were verified for authenticity. Performance differences among models were evaluated using group comparison tests followed by post-hoc analysis. RESULTS: Significant performance variations were identified across all four models and dimensions (p < 0.001). Specifically, Gemini ranked highest in content quality, accuracy, and bias control. In contrast, DeepSeek, despite its high-quality score, received the lowest relevance score. Grok-3 demonstrated the highest reference authenticity (p < 0.001), whereas GPT-4's was the lowest (p < 0.001). All models showed diminished performance on open topics and exhibited severe reference fabrication ("hallucinations"). CONCLUSION: Rather than excelling universally, LLMs exhibit distinct and task-specific strengths that mandate a task-driven, hybrid strategy in tool selection. Reference fabrication was found to be a pervasive issue across all models, regardless of the task topic, elevating human verification from a best practice to an essential safeguard for academic integrity.
9 May 2026
Read appraisal →Research synthesis methods
Transforming evidence synthesis: A systematic review of the evolution of automated meta-analysis in the age of AI
Exponential growth in scientific literature has heightened the demand for efficient evidence-based synthesis, driving the rise of the field of automated meta-analysis (AMA) powered by natural language processing and machine learning. This PRISMA systematic review introduces a structured framework for assessing the current state of AMA, based on screening 13,216 papers (2006-2024) and analyzing 61 studies across diverse domains. Findings reveal a predominant focus on automating data processing (52.5%), such as extraction and statistical modeling, while only 16.4% address advanced synthesis stages. Just one study (approximately 2%) explored preliminary full-process automation, highlighting a critical gap that limits AMA's capacity for comprehensive synthesis. Despite recent breakthroughs in large language models and advanced AI, their integration into statistical modeling and higher-order synthesis, such as heterogeneity assessment and bias evaluation, remains underdeveloped. This has constrained AMA's potential for fully autonomous meta-analysis (MA). From our dataset spanning medical (67.2%) and non-medical (32.8%) applications, we found that AMA has exhibited distinct implementation patterns and varying degrees of effectiveness in actually improving efficiency, scalability, and reproducibility. While automation has enhanced specific meta-analytic tasks, achieving seamless, end-to-end automation remains an open challenge. As AI systems advance in reasoning and contextual understanding, addressing these gaps is now imperative. Future efforts must focus on bridging automation across all MA stages, refining interpretability, and ensuring methodological robustness to fully realize AMA's potential for scalable, domain-agnostic synthesis.
6 May 2026
Read appraisal →Journal of clinical ultrasound : JCU
Illuminating Research Dynamics: Medical Ultrasound and Deep Learning
BACKGROUND: Deep learning (DL) has enabled advances in ultrasound imaging, but challenges like limited datasets and device variability hinder progress. This study provides the first bibliometric overview of DL research in medical ultrasound. METHOD: We retrieved related publications (2004-Apr 2025) on medical ultrasound and DL from the Web of Science Core Collection. Bibliometric tools (Bibliometrix, VOSviewer, CiteSpace) were used to quantify publication growth, country/institution contributions, and thematic evolution. RESULTS: A total of 3386 publications were included, with the annual output increasing steadily at an average growth rate of 31.02%. While China led all countries in publication volume (1355), the United Kingdom achieved the highest citation rate per article (55.5 citations). Fudan University was the most prolific institution (110), and key authors such as Luca Saba and Jasjit S. Suri formed the principal collaboration hubs. In terms of venues, Diagnostics published the most papers, whereas IEEE Transactions on Medical Imaging exerted the greatest citation impact (5646 citations; IF 8.9). Keyword analysis revealed core themes in classification, segmentation, and CNN-based diagnosis. From 2020 onward, ultrasound-specific topics (e.g., diagnosis, images) dominate; advanced architectures (e.g., U-Net) and emerging clinical concepts such as noninvasive assessment. CONCLUSION: This mapping highlights a rapidly growing, internationally collaborative field. Emerging trends include advanced artificial intelligence (AI) architectures (U-Net, attention-based networks, vision Transformers) and a shift toward surgical and therapeutic applications. These insights underscore the need for multicenter validation, standardized data protocols, and clinically interpretable AI models to accelerate safe translation into practice.
5 May 2026
Read appraisal →Clinical pharmacology and therapeutics
Off-the-Shelf Large Language Models for Guiding Pharmacoepidemiological Study Design
This study aimed to assess the ability of two off-the-shelf large language models, ChatGPT and Gemini, to support the design of pharmacoepidemiological studies. We assessed 48 study protocols of pharmacoepidemiological studies published between 2018 and 2024, covering various study types, including disease epidemiology, drug utilization, safety, and effectiveness. The coherence (i.e., "Is the response coherent and well-formed, or is it difficult to understand?") and relevance (i.e., "Is the response relevant and informative, or is it lacking in substance?") of the large language models' responses were evaluated by human experts across seven key study design components. Coding accuracy was assessed. Both large language models demonstrated high coherence, with over 90% of study components rated as "Strongly agree" by experts for most categories. ChatGPT achieved the highest coherence for "Index date" (97.9%) and "Study design" (95.8%). Gemini excelled in "Study outcome" (93.9%) and "Study exposure" (95.9%). Relevance, however, was more variable, with ChatGPT aligning with expert recommendations in over 90% of cases for "Index date" and "Study design" but showing lower agreement for covariates (65%) and follow-up (70%). Coding agreement percentages reveal varying levels of concordance, with the Anatomical Therapeutic Chemical classification system coding system demonstrating the highest agreement at 50% with experts. In contrast, the Current Procedural Terminology and International Classification of Diseases systems showed agreements of 22.2% and 20%, respectively. While ChatGPT and Gemini show promise in certain tasks supporting pharmacoepidemiological study design, their limitations in relevance and coding accuracy highlight the need for critical oversight by domain experts.
4 May 2026
Read appraisal →BMJ supportive & palliative care
Hand-foot syndrome prevention: pharmacological interventions-systematic review and network meta-analysis
INTRODUCTION: Hand-foot syndrome (HFS) is a dermatological side effect of chemotherapies such as capecitabine and pegylated liposomal doxorubicin. Over recent years, numerous randomised controlled trials (RCTs) have investigated various pharmacological strategies to prevent HFS. Although urea cream and celecoxib have shown promising results, the efficacy of topical diclofenac remains unclear. To provide a comprehensive comparison of pharmacological interventions for HFS prevention in cancer patients undergoing chemotherapy, we conducted a network meta-analysis (NMA). METHODS: We systematically searched PubMed, Cochrane and Scopus for relevant RCTs published up to November 2024. The primary outcome was the incidence of moderate to severe HFS (grade 2-3), while secondary outcomes included the occurrence of all-grade HFS (grade≥1) and chemotherapy modifications. Risk ratios (RR) with 95% CIs were estimated for all outcomes using NMA. RESULTS: Thirteen RCTs comprising 1983 patients, most of whom received capecitabine (n=1849) were included, evaluating five pharmacological interventions. Both topical diclofenac (RR 0.26, 95% CI 0.10 to 0.66) and celecoxib (RR 0.46, 95% CI 0.26 to 0.84) significantly reduced the risk of moderate to severe HFS compared with placebo. Notably, topical diclofenac was the only intervention that significantly reduced the incidence of all-grade HFS and the need for chemotherapy modifications. CONCLUSION: Our NMA demonstrates that topical diclofenac provides the most robust protective effect among available pharmacological interventions for HFS prevention and is also associated with fewer chemotherapy modifications. These findings support its consideration as a preferred preventative strategy for patients at risk of HFS during chemotherapy. PROSPERO REGISTRATION NUMBER: CRD42025643635.
3 May 2026
Read appraisal →BMJ health & care informatics
Omission and hallucination prevalence of clinical guidelines in diagnostic large language model outputs
OBJECTIVE: Meaningful assessments of how large language models (LLMs) incorporate clinical guidelines require large-scale testing over many queries. Here, we evaluate the prevalence of clinical guideline omissions and hallucinations in a large sample of diagnostic LLM outputs. METHODS: We used simulated case vignettes and zero-shot prompting to generate diagnostic outputs and rationales from GPT-4.1 and DeepSeek-V3. English case vignettes were created for hypercholesterolaemia and type-2 diabetes mellitus. Each vignette contained identical medical information, while sociodemographic characteristics varied in terms of sex, ethnicity and location. We calculated the prevalence of existing and hallucinated clinical guidelines in LLM outputs across disease, LLM and sociodemographic characteristics. RESULTS: We analysed a total of 12 197 LLM outputs, which quantifies three hazard areas: omissions (up to 97% for DeepSeek-V3 and 46% for GPT-4.1), hallucinations (up to 9%) and inconsistencies (guideline citation rate ranging from 0% to 78.39% across sociodemographic vignettes). Omission and hallucination rates were generally similar across vignettes with different sex or ethnicity data, yet were particularly sensitive to patient location. DISCUSSION: This study highlights significant variability in clinical guideline prediction across two different diseases, three different sociodemographic variables and two LLMs, even when the LLMs were instructed by identical prompts, establishing clinical guideline prediction in LLM outputs as a stochastic event. CONCLUSION: The stochastic nature of LLMs creates a unique challenge for evidence generation and clinical deployment. Being able to measure and capture this stochasticity within high-quality research designs will be a prerequisite to advancing the responsible deployment of LLMs in healthcare.
27 Apr 2026
Read appraisal →BMJ open
Do studies of interventions to improve laypeople's critical thinking about health choices assess potential harms? A systematic review.
OBJECTIVES: To make informed health choices, and avoid waste and unnecessary suffering, people need critical thinking skills. However, like health interventions, educational interventions can have adverse effects. In this systematic review, the objective was to assess the extent to which researchers have included potential adverse effects in studies of interventions intended to improve the critical thinking of laypeople about health choices. DESIGN: This study was a systematic review, in which we updated the search for an earlier systematic review of intended effects of relevant interventions. The earlier review did not address potential adverse effects. We did not update the analysis of intended effects. DATA SOURCES: We searched Cochrane Central Register of Controlled Trials (CENTRAL), Cumulative Index to Nursing and Allied Health Literature (CINAHL), Embase, Epistemonikos, Medical Literature Analysis and Retrieval System Online (MEDLINE), Education Resources Information Center (ERIC) and Web of Science up to March 2025. In addition to studies from the original review and updated search, we included any additional studies included in a similar, even earlier review. Our unit of analysis was study report (eg, journal article). ELIGIBILITY CRITERIA: We included all studies from the original review. We applied the same inclusion criteria to the results of the updated search: the study included a comparison, the population was laypeople and the intervention was intended to improve understanding of ≥1 key concept for informed health choices. DATA EXTRACTION AND SYNTHESIS: We extracted data about study design (randomised trial or other), participants (children, adolescents or adults), study setting (countries), main intervention (resources delivered to participants) and comparator (usual/no intervention or other). For the analysis, we extracted verbatim text describing any assessment of a potential adverse effect of the intervention. We conducted a narrative synthesis of the extracted data. RESULTS: We included 35 reports of quantitative studies (including multi-method and mixed-methods studies). Most often, the study was a randomised trial, the setting was a high-income country, the population included adults (including university students) and the intervention was school-based (including university). In one of the 35 reports, authors described assessing a potential adverse effect. CONCLUSION: To our knowledge, this is the first systematic review assessing the extent to which researchers have assessed adverse effects of any category of educational interventions. Our review shows that researchers generally have not assessed potential adverse effects of interventions to improve critical thinking about health choices. Researchers should pay more attention to such effects, while policymakers and educators making decisions about implementing relevant interventions should consider the lack of evidence. The findings of this study suggest a need for research that facilitates assessing potential adverse effects of interventions to improve critical thinking about health choices.
27 Apr 2026
Read appraisal →BMJ health & care informatics
Using a large language model artificial intelligence agent to improve the efficiency of clinical quality measure evidence evaluation: a case study.
OBJECTIVES: To evaluate the feasibility and performance of a large language model (LLM)-based artificial intelligence (AI) agent, implemented within a structured Claim-Argument-Evidence System (CAES), for supporting the review of clinical quality measure (CQM) evidence in the Centers for Medicare & Medicaid Services Consensus-Based Entity (CBE) endorsement process. METHODS: The CBE conducted a pilot study using a previously endorsed measure. CAES extracted claims and citations from a submitted diagnostic performance measure for pneumonia, automatically retrieved additional relevant evidence from PubMed abstracts and assessed the quality, confidence and agreement of evidence supporting each claim. The system's assessments were compared with the judgement of a subject matter expert (SME). RESULTS: CAES completed the assessment in approximately 5 hours. The SME agreed with the CAES-assigned claim statuses for 69% of claims, was neutral for 11% and disagreed for 14%. Disagreements primarily stemmed from the need for contextual interpretation beyond abstracts. DISCUSSION: Manual evaluation of CQM evidence requires significant time and resources, estimated at over 2400 labour hours per review cycle, limiting efficiency and transparency.The AI agent evaluated 64 claims and 355 claim-evidence pairs related to the pneumonia diagnosis measure. It assigned claim statuses based on evidence strength and generated justifications. CONCLUSION: This pilot demonstrated the feasibility and potential of LLM-based AI agents to improve the efficiency and transparency of evidence review for CQMs. Further development is needed to incorporate additional data sources and extend applicability across the measure development lifecycle.
27 Apr 2026
Read appraisal →Journal of the American Academy of Orthopaedic Surgeons. Global research & reviews
Academic Influence and Industry Funding in Platelet-Rich Plasma Injections: Coauthorship Network Analysis
BACKGROUND: Platelet-rich plasma (PRP) injections have gained popularity in the fields of orthopaedics and sports medicine despite inconsistent evidence regarding clinical efficacy. Concerns have placed a role on industry funding in shaping the development of PRP injection literature. Therefore, the aim of this study is to investigate the relationship between industry funding and academic influence in PRP injection research. METHODS: A systematic review was done of PRP injection literature published between 2017 and 2024, all from Q1 (top quartile) peer-reviewed journals based on Scopus journal ranking metrics. Overall, 81 studies were included in final analysis. Authors were linked to the Open Payments Database (OPD) and were identified through their national provider identifier (NPI) number. A coauthorship network analysis was used to map patterns of collaboration among authors in the field. RESULTS: The majority of studies identified were randomized control trials (38.3%). From 2017 to 2024, total industry funding to PRP authors totaled $36.1 million. Markedly more general payments than research payments were there (P < 0.05). Funding did not markedly correlate with publication count (r = 0.368, P = 0.40). No notable association was found between funding type and study outcome for either general (P = 0.482) or research (P = 0.481) payments. A centralized structure of 621 unique authors revealed that the top 10% of contributors accounted for more than 40% of all betweenness activity among these authors. CONCLUSION: The majority of industry funding in PRP injection research is directed as general payments. A small cohort of authors hold disproportionate influence over the community.
26 Apr 2026
Read appraisal →Behavior research methods
A review and evaluation of doubly robust approaches for estimating average treatment effects
In nonexperimental studies, obtaining an unbiased estimate of the average treatment effect (ATE) typically requires two key assumptions: that all relevant covariates are measured (i.e., no unmeasured confounding) and that the statistical model used for covariate adjustment is correctly specified. Two common approaches for adjustment are specifying an outcome model and propensity score weighting. To mitigate bias from model misspecification, doubly robust methods combine both approaches, ensuring unbiased ATE estimates if either the outcome model or the propensity score model is correctly specified. In this study, we review four doubly robust methods that have received considerable attention in the methodological literature but remain underutilized in psychological research: augmented inverse probability weighting, regression weighted by the inverse propensity score, regression incorporating the inverse propensity score as a covariate, and calibrated propensity score weighting. Using two simulation studies, we compare these methods with regression estimation and inverse probability weighting estimators. Our results suggest that doubly robust methods-particularly regression weighted by the inverse propensity score-offer greater protection against bias from model misspecification across various data-generating scenarios. We also discuss practical considerations for implementing doubly robust methods, including weight normalization, propensity score truncation, and potential efficiency losses due to overfitting. The different methods for estimating the ATE are illustrated in a data example.
26 Apr 2026
Read appraisal →The British journal of ophthalmology
Use of patient-reported outcomes in ophthalmology clinical trials between 2014 and 2023.
BACKGROUND/AIMS: Patient-reported outcomes (PROs) evaluate health and functional status, and PRO measures (PROMs) are standardised tools for measuring PROs. Together, they provide valuable insights into treatment efficacy, safety and practicality not captured by traditional clinical endpoints. This cross-sectional analysis with a systematic search component aims to investigate the use, interpretation and reporting of PROs and PROMs in ophthalmic randomised controlled trials (RCTs). METHODS: Ophthalmic RCTs published in the top 10 highest impact factor ophthalmic and medical journals between 2014 and 2023 were systematically reviewed. The frequency of PRO inclusion and adherence to Consolidated Standards of Reporting Trials (CONSORT) PRO guidelines was assessed. The relationship between PRO utilisation and study-level and journal-level characteristics was explored with multivariable regression. RESULTS: Among 9436 records screened, 333 RCTs met eligibility criteria. Of these, 87 (26.1%) included PROs, and 28 (8.4%) used them as primary outcomes. Most studies (83/87, 95.4%) leveraged PROMs, with ophthalmology-specific tools predominating (73.5%). Minimal important differences (MIDs) were rarely used (2.3%) for PRO interpretation. At least 8/13 CONSORT PRO Extension items were reported in 33.3% studies, and trials with primary PRO endpoints had better adherence (p<0.001). PRO utilisation was less likely in trials with lower 5-year journal impact factor (adjusted OR (aOR) 0.99, 95% CI 0.98 to 1.00, p=0.037) and pharmaceutical compared with health service interventions (aOR 0.15, 95% CI 0.03 to 0.67, p=0.013). CONCLUSIONS: PRO integration and interpretation remain limited in high-impact ophthalmic RCTs, despite offering a meaningful adjunct to objective endpoints. Future trials should adopt validated and condition-specific PROMs and establish MIDs to enhance interpretation.
26 Apr 2026
Read appraisal →The American journal of nursing
Using Artificial Intelligence to Enhance Evidence Search and Synthesis
Accelerating the path from research to practice.
25 Apr 2026
Read appraisal →Journal of neurology
Efficacy and safety of monoclonal antibodies in the treatment of relapsing remitting multiple sclerosis: a systematic review
INTRODUCTION: In relapsing-remitting multiple sclerosis (RRMS), conventional immunomodulatory and immunosuppressive therapies are widely used. However, in many cases, optimal control of inflammatory activity and disease progression is not achieved, which has led to the use of biological drugs such as monoclonal antibodies that act specifically on key components of the immune system. The aim was to evaluate the efficacy and safety of monoclonal antibodies compared to other drugs or placebo in adult patients with RRMS. METHODS: A systematic review was performed based on randomized, double-blind, phase III controlled clinical trials published between 2012 and 2025 in the PubMed, Cochrane Library, and Web of Science databases, assessing efficacy and safety in adult patients with RRMS. The review was carried out following the PICO methodology and PRISMA guidelines. RESULTS: A total of 11 studies were included, evaluating 5 monoclonal antibodies: alemtuzumab, daclizumab, ocrelizumab, ofatumumab, and ublituximab. These therapies showed superior efficacy compared to conventional treatments in reducing the annual relapse rate, MRI inflammatory activity and MRI activity, particularly in patients with highly active disease. However, effects on disability progression were heterogeneous across trials and not consistently significant. In addition, decreases in biomarkers of axonal damage were observed. Nevertheless, relevant adverse effects were identified, including infections, autoimmune reactions, hepatic and cutaneous toxicity, whose incidence varies depending on the drug, requiring close clinical monitoring. CONCLUSIONS: Monoclonal antibodies are an effective option in RRMS, with clinical and radiological benefits superior to those of conventional treatments. Their use requires individualized assessment and close follow-up due to the risk of adverse effects, especially in high-risk patients.
25 Apr 2026
Read appraisal →Journal of the American College of Cardiology
Extension Reports in Clinical Trials: A Systematic Review and Consequent Guidance
BACKGROUND: Following the primary publication of a clinical trial, questions of longer-term efficacy and safety remain unanswered. Investigators often have sequel publications based on extended follow-up; yet, no guidelines exist for such extension reports. OBJECTIVES: The purpose of this analysis was to assess the characteristics of extension reports of cardiovascular trials published in 7 high-impact medical journals. METHODS: We searched MEDLINE to identify cardiovascular trial extension papers published between January 2019 and October 2023 in 7 high-impact medical journals. Extension reports were defined as those presenting additional follow-up data after publication of the primary results. Papers were screened by 2 reviewers before extracting data using a standard proforma. RESULTS: After screening 1,388 papers, 69 extension reports across 64 trials were identified. A variety of cardiovascular topics were covered, coronary heart disease (n = 20) being the most common. Interventions were medical devices, drugs, surgery, or treatment strategies. Duration of follow-up varied considerably, most common being an original 1-year report (n = 20) with extension at 2, 3, 5, or 10 years. Nearly all extension reports had a superiority hypothesis (n = 68). Most extensions (n = 53) examined the same primary outcome as the original publication. There were also 7 nonrandomized open-label extensions, which suffered from the lack of a concurrent control arm. In 18 reports, treatment crossovers were permitted during the extension, and in 16 cases, alternatives to analysis by intention to treat were utilized. One-half of the extension studies were prespecified (n = 38), and only 10 had dedicated protocols or statistical analysis plans. No study used multiplicity corrections to control type I error across the totality of evidence. Landmark analyses were common (n = 39) which, while descriptively useful, were not based on the original randomized groups. Considerable loss to follow-up (>10%) occurred in 10 of 50 extension reports with clear reporting of this issue. For some trials, the conclusions regarding treatment effects were altered by the extended follow-up. In a previous survey of cardiovascular trials published in the New England Journal of Medicine, Lancet, and Journal of the American Medical Association in 2019, we identified a total of 23 of 84 trials involving extensions in some form. CONCLUSIONS: This first systematic investigation into the reporting of extension studies found a great diversity in practice, noting several issues that need more adequate consideration. Considerations are provided on what trialists should consider when planning, conducting, and reporting extension studies.
25 Apr 2026
Read appraisal →Journal of the American College of Cardiology
African Representation in Randomized Controlled Trials Published in Leading Medical and Cardiovascular Journals, 2019-2024
BACKGROUND: Despite accounting for a substantial proportion of the global population and disease burden, African countries are underrepresented in randomized controlled trials (RCTs), including those informing cardiovascular (CV) care. OBJECTIVES: In this study, we sought to quantify African representation in RCTs published from 2019 to 2024 in: 1) 5 leading general medical journals; and 2) 3 leading CV journals. METHODS: We conducted a systematic review of RCTs published from 2019 to 2024 in the British Medical Journal, the Journal of the American Medical Association, The Lancet, Nature Medicine, and the New England Journal of Medicine, and in Circulation, the European Heart Journal, and the Journal of the American College of Cardiology. Eligible studies included traditional, pragmatic, cluster, and stepped-wedge RCTs. African representation was assessed by trial scope (Africa-only vs multicontinental), country and regional participation, disease category, and African authorship. RESULTS: Among 2,138 RCTs published in leading general medical journals, only 83 (3.9%) were conducted exclusively in Africa, and 195 (9.1%) were multicontinental studies including at least 1 African site. In the CV journals, 2 out of 334 RCTs (0.6%) were conducted exclusively in Africa, and African sites were included in only 9 multicontinental trials (2.7%). South Africa accounted for the majority of Africa-based RCTs across both journal categories. Regionally, southern Africa predominated and central Africa was minimally represented. Trials published in general medical journals and conducted exclusively in Africa largely focused on infectious diseases (n = 63; 75.9%), with only 3 addressing cardiovascular disease (CVD). In contrast, Africa-including multicontinental trials more frequently investigated noncommunicable diseases, including CVD. African leadership was common in Africa-only trials but rare in multicontinental studies. CONCLUSIONS: African countries are profoundly underrepresented in RCTs published in the world's most influential medical and CV journals. Addressing this imbalance requires expanding African participation in global trials, investing in local research capacity, and promoting equitable leadership to strengthen the relevance and validity of clinical evidence. (Underrepresentation of African Countries in Randomized Controlled Trials: A Systematic Review of Leading General Medical and Cardiovascular Journals; CRD42024603157).
24 Apr 2026
Read appraisal →Journal of the American College of Cardiology
Using Artificial Intelligence to Adjudicate Major Adverse Cardiovascular Events in Clinical Trials
BACKGROUND: Major adverse cardiovascular events (MACE)-cardiovascular (CV) death, nonfatal myocardial infarction (MI), and nonfatal stroke-are highly relevant clinical outcomes. In global randomized trials, medical records review by a physician clinical events committee (CEC) is the conventional standard for adjudicating MACE but is labor intensive. Automated adjudication with the use of artificial intelligence (AI) could reduce cost and improve reproducibility. OBJECTIVES: In this study, the authors sought to develop and validate an AI-based adjudication system for MACE and compare its performance with CEC adjudication in a large global randomized trial. METHODS: We developed an AI-based system ("Auto-MACE") that uses an iteratively refined prompt of the OpenAI o1-mini language model to adjudicate MACE events, and a Clinical Longformer model trained on adjudicated events to assign a confidence level. We validated Auto-MACE against CEC adjudication in the PARADISE-MI global clinical trial comparing sacubitril/valsartan and ramipril in 5,661 patients with MI complicated by systolic dysfunction or pulmonary congestion. RESULTS: Auto-MACE provided a confident adjudication in 315/455 deaths (69%), 301/659 potential MIs (46%), and 136/167 potential strokes (81%). Auto-MACE agreed with the CEC adjudication in 97%, 89%, and 88% of confident events, respectively. Among all events, Auto-MACE agreed with CEC adjudications in 86% of deaths, 76% of potential MIs, and 84% of potential strokes. The estimated effect of sacubitril/valsartan vs ramipril on composite MACE was similar with Auto-MACE adjudication (HR: 0.91; 95% CI: 0.78-1.07) and CEC adjudication (HR: 0.90; 95% CI: 0.77-1.05). CONCLUSIONS: AI-based adjudication of MACE showed high agreement with human CEC adjudication, especially for CV death and stroke, and where the model was confident. Initial AI-based adjudication with CEC review of uncertain events may reduce workload while maintaining accuracy.
24 Apr 2026
Read appraisal →Sleep & breathing = Schlaf & Atmung
Health-related quality of life in patients with obstructive sleep apnea - a systematic review update assessing the quality of measurement properties of patient-reported outcome measures
PURPOSE: The purpose of this systematic review update was to identify existing health-related quality of life (HRQOL) patient-reported outcome measures (PROMs) for obstructive sleep apnea (OSA). Building upon a 2016 systematic review, which recommended four PROMs with the highest evidence, this investigation was warranted to synthesize new evidence and evaluate new PROMs that have emerged since then via the Consensus based Standards for the Selection of Health Measurement Instruments (COSMIN). METHODS: We searched Medline via PubMed, Embase and the Cochrane Library for HRQOL PROM development and validation studies published in English or German. Studies focusing on cross-cultural validation, those in which PROMs were assessed as secondary outcomes, and studies limited to specific demographic subgroups or single HRQOL domains were excluded. For risk of bias and quality of measurement properties (MP) assessment, we used the COSMIN methodology to facilitate comparability with the previous review, as well as the GRADE approach for evidence level evaluation. RESULTS: We identified two new PROMs, the Patient-Reported Apnea Questionnaire (PRAQ) and the Symptoms, Tiredness, Alertness, Mood and Psychosocial instrument (STAMP), which were validated in two studies and a single study, respectively. Overall, the PRAQ displays an moderate risk of bias with excellent quality of MP and high evidence for content validity in particular, whereas the STAMP shows a higher risk of bias and low quality of MP. CONCLUSION: The PRAQ has demonstrated the strongest evidence among PROMs since 2016. In total, including the previous review, five PROMs are currently recommended for assessing HRQOL in individuals with OSA. REVIEW REGISTRATION: The review was registered with PROSPERO under the registration number CRD42024602020.
24 Apr 2026
Read appraisal →BMJ quality & safety
Sustainability of the de-implementation of low-value care in infants with bronchiolitis: 2-year follow-up of a cluster randomised controlled trial
BACKGROUND: In 2017, the PREDICT (Paediatric Research in Emergency Departments International Collaborative) network conducted a cluster randomised controlled trial (cRCT) at 26 Australian and New Zealand hospitals to improve bronchiolitis care. Findings demonstrated that targeted interventions significantly improved adherence with five evidence-based low-value bronchiolitis practices (no chest radiography, salbutamol, glucocorticoids, antibiotics and epinephrine) in the first 24 hours of hospitalisation (adjusted risk difference, 14.1%; 95% CI: 6.5% to 21.7%; p<0.001). During the intervention year (2017), intervention hospital (n=13) compliance was 85.1% (95% CI: 82.6% to 89.7%). This study aimed to determine if improvements in bronchiolitis management were sustained at intervention hospitals 2 years post-trial completion. METHODS: International, multicentre follow-up study of hospitals in Australia and New Zealand that participated in a cRCT of de-implementation of low-value bronchiolitis practices, 1 year (2018) and 2 years (2019) post-trial completion, obtained retrospectively from medical audits. Sustainability was defined a priori as no more than a <7% decrease to any level of improvement in adherence for all five low-value practices (composite outcome) from the cRCT intervention year. RESULTS: Of the 26 hospitals, 11 intervention and 10 control hospitals agreed to participate in the follow-up study. Data were collected on 3299 infants with bronchiolitis 1 year (intervention and control hospitals) and 1689 infants 2 years post-trial (intervention hospitals). Adherence with no use of the five low-value practices 2 years post-trial completion was 80.9% (adjusted predicted adherence, 80.8%, 95% CI: 77.4% to 84.2%; estimated risk difference from cRCT outcome -3.9%, 95% CI: -8.6% to 0.8%) at intervention hospitals, fulfilling the a priori definition of sustainability. DISCUSSION: Targeted interventions, delivered over one bronchiolitis season, resulted in sustained improvements in bronchiolitis management in infants 2 years later. This follow-up study provides evidence for sustainability in de-implementing low-value care in bronchiolitis management. TRIAL REGISTRATION DETAILS: Australian and New Zealand Clinical Trials Registry No: ACTRN12621001287820.
23 Apr 2026
Read appraisal →Trials
The wood and the trees: estimands in cluster randomised trials.
BACKGROUND: The estimand framework was introduced into guidance on good clinical practice to address a variety of shortcomings and ambiguities in the reporting of trials, including the use of terms such as "intention to treat" and the handling of non-adherence to treatment. The framework was primarily grounded in individually randomised trials, and some thorny issues still cloud understanding of its application to cluster randomised trials. ESTIMANDS IN CLUSTER RANDOMISED TRIALS: This commentary addresses some of the challenges in thinking about the estimands behind cluster randomised trials. These challenges include informative cluster size-the possibility that a treatment effect may be modified by the size of the cluster. In particular, we consider the perspectives of different actors-a population-level decision maker or politician, a cluster manager, and a patient-and examine possible estimands for each, and how they differ. CONCLUSIONS: In the cluster randomised trial context, the estimand framework can be complex to navigate. Different perspectives lead to different estimands. We caution against abandoning careful statistical modelling. This is particularly true in the presence of informative cluster size, where modelling any interaction between cluster size and treatment effect could be useful from a number of perspectives.
22 Apr 2026
Read appraisal →International journal of epidemiology
Structured tools for assessing quality and risk of bias in Mendelian randomization studies: an updated systematic review
BACKGROUND: The growing use of Mendelian randomization (MR) has heightened the need for rigorous quality and bias assessment tools. A previous systematic review included studies published up to July 2021 identified 14 structured instruments for conducting, evaluating, and reporting MR studies. However, methodological developments have accelerated in the years since. METHODS: We updated the previous systematic review to include tools published between July 2021 and January 2025, applying the same search strategy and eligibility criteria. Two reviewers independently screened articles, extracted data, and mapped tool content to bias domains. RESULTS: We identified 15 additional articles, bringing the total to 29 tools. Of these 29, 19 provided structured evaluation tools. Twelve of the 19 evaluation tools were newly added in the present review, which addressed broader methodological domains beyond core instrumental variable assumptions, including genetic instrument selection, population stratification, sensitivity analyses, and dataset considerations. However, substantial variation in bias domains, structure, and scoring methods across tools persists. Key gaps remain in the assessment of linkage disequilibrium, missing data, and dynastic effects. CONCLUSIONS: While the number of structured tools has increased in recent years, the lack of standardization across tools makes it difficult to compare across systematic reviews of MR studies. Developing more complete and standardized evaluation frameworks and properly testing these tools in practice are important next steps to improve the overall quality of MR research.
20 Apr 2026
Read appraisal →Journal of medical Internet research
Artificial Intelligence for Identifying Patient-Reported Outcome and Experience Measures in Oncology: Retrospective Cross-Sectional Study Using ClinicalTrials.gov
BACKGROUND: Traditional health care systems have evolved to increasingly recognize patients' perspectives as key to improving the quality of care, especially in oncology. Hence, patient-reported outcome measures (PROMs) and patient-reported experience measures (PREMs) can enhance patient-health care provider communication while facilitating individualized care. This tailored approach improves patient outcomes and highlights the need for research methods to account for variability in patient experiences across diverse sociodemographic and clinical backgrounds while proposing artificial intelligence (AI) to automate and accelerate the identification of PROMs and PREMs. OBJECTIVE: This study aimed to estimate the proportion of clinical studies including PROMs or PREMs, using either a traditional expert-based identification method or an AI-enriched approach. METHODS: In a retrospective cross-sectional study using the ClinicalTrial.gov database, we focused on oncology studies between 2012 and 2022. Two methods were used to identify PROMs and PREMs: (1) a traditional expert-based method, where an algorithm identified PROMs and PREMs from a list of 346 oncology-specific PROMs and PREMs (extracted from the PROQOLID database, Mapi Research Trust) and/or 11 PROM- and PREM-specific terms, and (2) an AI-enriched method using a bidirectional encoder representations from transformers model, trained on 2399 outcomes labeled by experts. To evaluate algorithm performance, results were compared with expert decisions. Studies were classified by PROM/PREM use, analyzed using logistic regression to identify drivers, and a named entity recognition model identified frequently used measures. RESULTS: A total of 24,491 studies were included. According to the traditional expert-based algorithm, 7549 (31%) studies used at least one PROM or PREM, as compared to 8029 (33%) studies identified by the AI-enriched algorithm, increasing from 2012 to 2022 (P<.001). With 90% (95% CI 88%-92%) accuracy, the AI-enriched algorithm outperformed (P<.001) the expert-based algorithm (84%, 95% CI 82%-86%) in identifying PROMs and PREMs. Breast and digestive cancers accounted for nearly 50% of all oncology studies using PROMs and PREMs, with the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire-Core 30 being the most frequently used. As expected, trials in phases 2 to 4 more frequently included PROMs and PREMs than preclinical or early phase 1 studies (odds ratio [OR] 1.8, 95% CI 1.1-2.8 for phase 2; OR 3.6, 95% CI 2.3-5.8 for phase 3; and OR 2.6, 95% CI 1.6-4.4 for phase 4). In observational studies, cross-sectional and prospective studies incorporated more PROMs and PREMs than retrospective studies (OR 4.6, 95% CI 3.3-6.4 and OR 3.2, 95% CI 2.5-4.1, respectively). CONCLUSIONS: Our study shows that an AI-enriched algorithm outperforms traditional expert methods in identifying PROMs and PREMs in oncology. Combining expert-labeled data with AI enables scalable, automated trial monitoring, supports efficient research, informs stakeholders, enhances patient-centered decisions, and can be extended to other diseases and databases.
17 Apr 2026
Read appraisal →Journal of medical Internet research
Beyond Missingness: Systematizing Methods for Comprehensive Data Fitness Assessment in Clinical Research
Secondary use of clinical data offers unprecedented opportunities to rapidly conduct large-scale research and improve patient care. However, incomplete understanding of data quality requirements for a study often causes significant delays in executing analyses and validating results. Current practice has largely followed 2 paths. First, multi-institutional networks have developed general data quality programs, but these are typically tied to unique network characteristics and do not address study-specific requirements well. Second, models have been proposed to formalize the requirements for data fitness analyses without extending to the methods needed to meet these requirements. More recently, tools have been developed to conduct cohort-centric screening, focusing on generally applicable structural checks such as missingness or facial implausibility. These provide a first level of information but incompletely capture the fitness requirements of an analysis. In turn, investigators conduct per-study exploratory analyses, but these efforts are typically ad hoc and partially reported, which can hinder reproducible science and delay advances in patient care. Analogously to advances over the past decade in data modeling and reproducible analytics, there is a need for a more systematic, capable approach to study-specific data quality assessment (SSDQA). We discuss such a model, which guides improved SSDQA design and implementation, including metadata for consistent annotation and reporting of data quality assessment results. The model integrates theoretical principles of data quality testing with pragmatic considerations of application to clinical data, providing a consistent approach to specifying data quality assessment checks. Additionally, it proposes to regularize check application through a standard set of options. The SSDQA model builds on current practice, providing a path toward more complete, sound, and reproducible assessments. These characteristics foster multidisciplinary collaboration to identify data quality issues that, in turn, inform decisions about study design and provide important context that has a bearing on adoption of results.
17 Apr 2026
Read appraisal →European journal of pediatrics
The impact of growth hormone therapy on ocular structural parameters in idiopathic short stature: a systematic review and meta-analysis
This study aimed to conduct a systematic review and meta-analysis of the effects of growth hormone (GH) therapy on ocular structures and biometric parameters in children with idiopathic short stature (ISS), synthesizing existing evidence for future research. We systematically searched PubMed, EMBASE, the Cochrane Library, CNKI, Wanfang, and VIP databases from inception to September 2025. Randomized controlled trials and observational studies comparing the effects of GH therapy versus control on ocular biometric and refractive outcomes in children with ISS were included. Study quality was assessed using the Cochrane Risk of Bias Tool (RoB 2) and the NOS. Meta-analysis was performed using R software version 4.3.2, with effect sizes expressed as MD or SMD and 95% CI. Nine studies were included (4 RCTs, 5 cohort studies). Meta-analysis showed no significant association of GH therapy with corneal curvature (SMD = 0.13, 95% CI: -0.34 to 0.61). For axial length, the random-effects model showed no significant difference (MD = 0.72, 95% CI: -0.36 to 1.81), with high heterogeneity (I2 = 91.7%). Similarly, no significant association was found for intraocular pressure (MD = 1.38, 95% CI: -0.66 to 3.43; I2 = 95.9%). However, a significant association was observed for refractive status (MD = 0.53, 95% CI: 0.20 to 0.86; I2 = 0.0%). Sensitivity analysis showed robust results for corneal curvature and refractive outcomes, while axial length and intraocular pressure results were influenced by individual studies. Funnel plots and Egger's test indicated no significant publication bias.Conclusion: The associations between GH therapy and ocular biometric parameters in children with ISS remain uncertain. Current evidence suggests a possible link with refractive changes but shows no significant associations with axial length or intraocular pressure when accounting for high heterogeneity. The findings are inconsistent and should be regarded as hypothesis-generating, insufficient to recommend changes in clinical monitoring practice. Future large-sample, long-term studies are required.
17 Apr 2026
Read appraisal →JMIR medical informatics
Classification of Cochrane Plain Language Summaries by Conclusiveness Using Transformer-Based Models and ChatGPT: Retrospective Observational Study
BACKGROUND: Cochrane plain language summaries (PLSs) aim to make systematic review findings more accessible to the general public. However, inconsistencies in how conclusions are presented may impact comprehension and decision-making. Classifying PLSs based on conclusiveness can improve clarity and facilitate informed health decisions. OBJECTIVE: This study aimed to develop and evaluate deep learning language models for the classification of PLSs according to 3 levels of conclusiveness (conclusive, inconclusive, and unclear) and to compare their performance with a general-purpose large language model (GPT-4o). METHODS: We used a publicly available dataset containing 4405 Cochrane PLSs of systematic reviews published until 2019, already classified by humans according to 9 categories of conclusiveness regarding the intervention's effectiveness or safety. We merged these categories into 3 classes based on the strength of conclusiveness: conclusive, inconclusive, and unclear. For the fine-tuning, we used Scientific Bidirectional Encoder Representations from Transformers (SciBERT), a pretrained language model trained on 1.14 million papers primarily from the health sciences, and Longformer, a transformer model designed specifically to process long documents. The script was developed using the Python programming language and the PyTorch framework. We computed evaluation metrics using the scikit-learn machine learning library and determined the area under the curve of the receiver operating characteristic (AUCROC) to measure the model performance in balancing sensitivity and specificity. We also analyzed a separate set of 213 PLSs and compared the predictions of our pretrained models with both manual verification and outputs generated by ChatGPT. RESULTS: The model based on SciBERT achieved a balanced accuracy of 56.6%. The AUCROC was 0.91 for "conclusive," 0.67 for "inconclusive," and 0.75 for "unclear" conclusiveness classes. The Longformer-based model had a balanced accuracy of 60.9%, with AUCROCs of 0.86 for "conclusive," 0.67 for "inconclusive," and 0.72 for "unclear" conclusiveness classes. Both models underperformed compared with ChatGPT, which demonstrated higher accuracy (74.2%), better precision and recall, and a higher Cohen κ (0.57). CONCLUSIONS: Fine-tuning 2 transformer-based language models showed mixed results in classifying Cochrane PLSs by conclusiveness, likely due to semantic overlap and subtle linguistic differences. Despite satisfactory internal test metrics, the fine-tuned models failed to generalize to newly published PLSs, where performance dropped to near-chance levels. These findings suggest that general-purpose large language models like GPT-4o may currently offer more reliable results for practical classification tasks in biomedical applications.
16 Apr 2026
Read appraisal →European journal of pediatrics
Clinical response and risk factors of fecal microbiota transplantation in children: a systematic review and meta-analysis
UNLABELLED: The objective of this study is to investigate the clinical response and incidence of adverse events (AEs) following fecal microbiota transplantation (FMT) in children, across various diseases, populations, and treatment protocols. A systematic search was conducted across eight major Chinese and English databases, identifying 47 studies up to August 28, 2025, for inclusion. Study quality was assessed using the Quality Assessment with Diverse Studies (QuADS) tool. Single-arm rates were pooled via meta-analysis employing the Freeman-Tukey double arcsine transformation, followed by extensive subgroup comparisons to identify influencing factors. FMT demonstrated efficacy in pediatric recurrent Clostridium difficile infection (rCDI), inflammatory bowel disease (IBD), and autism spectrum disorder (ASD), although a higher incidence of AEs was observed in children with IBD. Subgroup analyses revealed that the use of donor feces from relatives or friends was associated with a higher clinical response rate in rCDI. The presence of comorbidities such as IBD diminished the response rate in rCDI patients. Younger age in rCDI and IBD patients showed a trend towards higher clinical response rates, though this did not reach statistical significance. No statistically or clinically significant differences were found in other subgroup comparisons. Meta-regression suggested IBD to be a risk factor for FMT-related AEs. CONCLUSION: This study innovatively delineates the efficacy-safety profile of pediatric FMT and outlines a pathway for optimizing individualized treatment regimens, providing crucial evidence-based guidance for clinical practice. TRIAL REGISTRATION: This study has been registered on the PROSPERO database (CRD42024614196). WHAT IS KNOWN: • Fecal Microbiota Transplantation (FMT) demonstrates preliminary therapeutic potential in several pediatric diseases. • Existing evidence remains fragmented, with limited systematic data on factors modifying efficacy and safety in children. WHAT IS NEW: • The study revealed FMT's high efficacy across rCDI, IBD, and ASD, and identified IBD as a risk factor for elevated FMT-related adverse events in pediatric patients. • Notably, related/friend donors improved rCDI response rates, while comorbidities like IBD reduced rCDI treatment efficacy.
15 Apr 2026
Read appraisal →Journal of medical Internet research
The Alberta Quality Assessment Tool: Risk of Bias (AQAT:RoB) for the Evaluation of Medical Large Language Model Question-Answer Studies: Development and Pilot Validation
BACKGROUND: Despite the transformative potential of large language models (LLMs) in health care, the rapid development of these tools has outpaced their rigorous evaluation. While artificial intelligence-specific reporting guidelines have been developed to address standardized reporting of artificial intelligence studies, there is currently no specific tool available for risk of bias assessment of LLM question-answer (QA) studies. Existing risk-of-bias tools for medical research are not well suited to the unique challenges of evaluating LLM-QA studies, which creates a critical gap in assessing their safety and effectiveness. OBJECTIVE: This study aims to develop the Alberta Quality Assessment Tool: Risk of Bias (AQAT:RoB) for LLM-QA studies to systematically evaluate the validity and risk of bias in LLM-QA studies. METHODS: We conducted 2 literature reviews. The first was on quality assessment tools for LLM-QA studies, and the second was on LLM-QA studies, which informed the first draft of the AQAT:RoB. The draft AQAT:ROB was further refined through a prespecified iterative process of modified Delphi, consensus meeting, and validation. The first Delphi process occurred between May 1 and May 20, 2025, and the first consensus meeting was held on May 22. The first round of validation was completed by 4 evaluators, who were not part of the consensus meeting, on 16 randomly selected studies. As this first round of validation surpassed our a priori threshold of ≥80% agreement and a Cohen κ of ≥0.61 between evaluators, no further rounds of development and validation were undertaken. A second Delphi process occurred between February 20 and February 23, 2026, to vote on postpilot changes in response to peer review. RESULTS: The AQAT:RoB consists of 5 high-level domains (Questions, Reference Answers, LLM Answers, Evaluators, Outcomes). These domains are subdivided into 9 subdomains. Each subdomain includes at least one "Support for Judgment" and at least one "Type of Bias" and is to be rated "low," "high," or "unclear" for risk of bias. A pilot evaluation was completed by internal validators who were not part of the consensus discussion and were asked to complete the AQAT:RoB form for each assigned study. Each of the 16 studies was evaluated by 2 evaluators independently. Pilot validation showed a percent agreement of 86.1% and a Cohen κ of 0.70 between assessors. CONCLUSIONS: The AQAT:RoB demonstrates promising initial reliability for assessing the validity or risk of bias in LLM-QA studies. The tool will benefit from future refinements, external validation, and periodic updates to keep pace with evolving technology.
10 Apr 2026
Read appraisal →Philosophical transactions. Series A, Mathematical, physical, and engineering sciences
The need for verification in artificial intelligence-driven scientific discovery
Artificial intelligence (AI) is transforming the practice of science. Machine learning (ML) and large language models (LLMs) can generate hypotheses at a scale and speed far exceeding traditional methods, offering the potential to accelerate discovery across diverse fields. However, the abundance of hypotheses introduces a critical challenge; without scalable and reliable mechanisms for verification, scientific progress risks being hindered rather than advanced. In this article, we trace the historical development of scientific discovery, examine how AI is reshaping established practices for discovery and review the principal approaches, ranging from data-driven methods and knowledge-aware neural architectures to symbolic reasoning frameworks and LLM agents. While these systems can uncover patterns and propose candidate laws, their scientific value ultimately depends on rigorous and transparent verification, which we argue must be the cornerstone of AI-assisted discovery. This article is part of the discussion meeting issue 'Symbolic regression in the physical sciences'.
10 Apr 2026
Read appraisal →PloS one
The impact of physical adjunctive interventions on outcomes of clear aligner treatment: A systematic review of randomized controlled trials
INTRODUCTION: Physical adjunctive interventions (PAIs), including vibration devices and low-level laser therapy, are promoted to accelerate tooth movement, improve aligner tracking, and reduce discomfort in clear aligner treatment (CAT), but randomized evidence remains inconsistent. This systematic review aimed to assess whether PAIs improve CAT outcomes in terms of objective tooth-movement metrics, aligner tracking, overall treatment efficiency, and patient-centered outcomes, and to appraise the risk of bias and the certainty of the evidence at the outcome level. METHODS: Comprehensive electronic searches of PubMed, Embase, Scopus, Web of Science, and CENTRAL were conducted from database inception (earliest available indexing date in each database) to 30 June 2025 (coverage cut-off). Grey literature searching and hand-searching were also performed, with no restrictions on language or publication status. Eligible studies were human randomized controlled trials (RCTs) comparing CAT combined with PAIs versus CAT alone or sham interventions. Two independent reviewers performed study selection and data extraction, with risk of bias assessed using the Cochrane RoB 2 tool and certainty of evidence appraised with GRADE. Random-effects meta-analyses were conducted when feasible; otherwise, results were narratively summarized. The protocol was prospectively registered (PROSPERO CRD420251132229). RESULTS: Seven RCTs involving 266 participants were included. No significant improvement in Little's Irregularity Index (maxilla MD = 0.08, p = 0.869; mandible MD = 0.44, p = 0.487). Vibration improved aligner tracking under a 7-day change schedule (p = 0.003) but not case completion (p = 0.999). Overall compliance was approximately 77%, and no between-group difference was observed (p = 0.390). Pain was slightly lower on days 1-3 only (p < 0.05); no quality-of-life benefits were observed; periodontal indices remained unchanged; biomarkers showed inconsistent results. The certainty of evidence was low to very low, suggesting that further well-designed RCTs are likely to change the effect estimates and may alter the conclusions. CONCLUSIONS: Across objective tooth-movement metrics, aligner tracking, treatment efficiency, and patient-centered outcomes, current randomized evidence does not demonstrate a consistent benefit of physical adjunctive interventions in clear aligner treatment. Interpretation is limited by outcome-level risk of bias concerns and low to very low certainty of evidence.
9 Apr 2026
Read appraisal →Journal of the American Heart Association
Comparison of Intravascular Imaging-, Physiology-, or Angiography-Guided Approaches for Percutaneous Coronary Intervention: A Systematic Review and Network Meta-Analysis
BACKGROUND: Despite current guidelines recommending physiology- and intravascular imaging-guided percutaneous coronary intervention (PCI) in specific lesion subsets, angiography-guided PCI remains common in practice. The comparative effectiveness of these strategies remains uncertain. We aimed to compare clinical outcomes of PCI guided by intravascular imaging or physiological assessment versus conventional angiography. METHODS: We conducted a systematic review and network meta-analysis of randomized controlled trials, searching PubMed and EMBASE up to May 31, 2025. Eligible studies compared at least 2 of the following 6 guidance modalities in PCI: angiography, intravascular ultrasound (IVUS), optical coherence tomography/optical frequency domain imaging, fractional flow reserve, angiography-derived fractional flow reserve, and instantaneous wave-free ratio. The primary outcome was trial-defined major adverse cardiovascular events (MACEs). Hazard ratios (HRs) with 95% CIs were pooled using a frequentist random-effects network meta-analysis. Subgroup analyses assessed clinical presentation and guidance objectives such as decision making and procedural optimization. RESULTS: We identified 43 randomized controlled trials involving 39 291 patients. IVUS-guided PCI (HR, 0.69 [95% CI, 0.60-0.79]), optical coherence tomography/optical frequency domain imaging-guided PCI (HR, 0.75 [95% CI, 0.63-0.90]), and fractional flow reserve-guided PCI (HR, 0.81 [95% CI, 0.70-0.95]) were associated with a lower risk of MACEs compared with angiography-guided PCI. Furthermore, IVUS-guided PCI was associated with a lower risk of MACEs compared with instantaneous wave-free ratio-guided PCI (HR, 0.74 [95% CI, 0.55-1.00]). IVUS-guided PCI reduced the risk of MACE in both acute coronary syndrome and non-acute coronary syndrome patients. CONCLUSIONS: IVUS- and optical coherence tomography/optical frequency domain imaging-guided PCI were superior to angiography-guided PCI in reducing MACEs. Among the physiology-based approaches, only fractional flow reserve showed a clear benefit.
9 Apr 2026
Read appraisal →Journal of ethnopharmacology
Efficacy and safety of Qiliqiangxin capsule as adjunctive therapy in heart failure with reduced ejection fraction: A systematic review and meta-analysis of randomized controlled trials
ETHNOPHARMACOLOGICAL RELEVANCE: The Qiliqiangxin (QLQX) capsule is a prominent traditional Chinese medicine formulation developed based on the collateral disease theory, specifically for the treatment of heart failure (HF). Its composition includes 11 medicinal botanicals, such as Huang Qi, Ren Shen, and Dan Shen, and is designed to tonify qi, activate blood circulation, and promote diuresis. Supported by over 20 years of extensive clinical application, modern pharmacological research, and stringent quality control, QLQX has been incorporated into the national heart failure guidelines in China. AIM OF THE STUDY: This study evaluates the efficacy and safety of QLQX as adjunctive therapy for heart failure with reduced ejection fraction (HFrEF). METHODS: We systematically searched seven databases until February 24, 2025, for randomized controlled trials (RCTs) evaluating QLQX in patients with HFrEF. Two independent researchers screened trials, extracted data, and assessed the risk of bias using the Cochrane RoB 2.0 tool. Meta-analyses were conducted for efficacy and safety outcomes, with evidence quality evaluated through the GRADE framework. RESULTS: Twenty-four RCTs (n = 5580) were included. Compared with conventional treatments (CTs) alone, QLQX adjunctive therapy significantly reduced the risk of cardiovascular death (RR = 0.82, 95 %CI [0.69, 0.96], P = 0.02), hospitalization for HF (RR = 0.78, 95 %CI [0.67, 0.91], P = 0.001), and composite cardiovascular events by 18 % (RR = 0.82, 95 %CI [0.73, 0.92], P = 0.0005), with evidence quality rated as moderate. QLQX significantly improved cardiac function parameters compared to controls: increased left ventricular ejection fraction (MD = 7.14, 95 % CI [6.75, 7.54], P < 0.05), reduced left ventricular end-diastolic diameter (MD = -5.25, 95 % CI [-5.75, -4.75], P < 0.05), improved NYHA class (RR = 1.19, 95 %CI [1.14, 1.25]), decreased NT-proBNP (SMD = -2.01, 95 %CI [-2.26, -1.95], P < 0.05), and increased 6-min walking distance (MD = 44.80, 95 %CI [42.37, 47.22], P < 0.05). CONCLUSION: QLQX adjunct therapy effectively reduces cardiovascular death and HF hospitalization while improving cardiac function in HFrEF, with favorable safety. Large-scale multicenter RCTs with extended follow-up are required to confirm long-term benefits and address existing study limitations.
6 Apr 2026
Read appraisal →Regional anesthesia and pain medicine
Human versus artificial intelligence: evaluating ChatGPT's performance in conducting published systematic reviews with meta-analysis in chronic pain research
INTRODUCTION: Artificial intelligence (AI), particularly large-language models like Chat Generative Pre-Trained Transformer (ChatGPT), has demonstrated potential in streamlining research methodologies. Systematic reviews and meta-analyses, often considered the pinnacle of evidence-based medicine, are inherently time-intensive and demand meticulous planning, rigorous data extraction, thorough analysis, and careful synthesis. Despite promising applications of AI, its utility in conducting systematic reviews with meta-analysis remains unclear. This study evaluated ChatGPT's accuracy in conducting key tasks of a systematic review with meta-analysis. METHODS: This validation study used data from a published meta-analysis on emotional functioning after spinal cord stimulation. ChatGPT-4o performed title/abstract screening, full-text study selection, and data pooling for this systematic review with meta-analysis. Comparisons were made against human-executed steps, which were considered the gold standard. Outcomes of interest included accuracy, sensitivity, specificity, positive predictive value, and negative predictive value for screening and full-text review tasks. We also assessed for discrepancies in pooled effect estimates and forest plot generation. RESULTS: For title and abstract screening, ChatGPT achieved an accuracy of 70.4%, sensitivity of 54.9%, and specificity of 80.1%. In the full-text screening phase, accuracy was 68.4%, sensitivity 75.6%, and specificity 66.8%. ChatGPT successfully pooled data for five forest plots, achieving 100% accuracy in calculating pooled mean differences, 95% CIs, and heterogeneity estimates (I2 score and tau-squared values) for most outcomes, with minor discrepancies in tau-squared values (range 0.01-0.05). Forest plots showed no significant discrepancies. CONCLUSION: ChatGPT demonstrates modest to moderate accuracy in screening and study selection tasks, but performs well in data pooling and meta-analytic calculations. These findings underscore the potential of AI to augment systematic review methodologies, while also emphasizing the need for human oversight to ensure accuracy and integrity in research workflows.
4 Apr 2026
Read appraisal →Am J Perinatol
Application of Generative AI to Enhance Obstetrics and Gynecology Research
The rapid evolution of large-language models such as ChatGPT, Claude, and Gemini is reshaping the methodological landscape of obstetrics and gynecology (OBGYN) research. This narrative review provides a comprehensive account of generative AI capabilities, key use cases, and recommended safeguards for investigators. First, generative AI expedites hypothesis generation, enabling researchers to interrogate vast corpora and surface plausible, overlooked questions. Second, it streamlines systematic reviews by composing optimized search strings, screening titles and abstracts, and identifying full-text discrepancies. Third, AI assistants can draft reproducible analytic code, perform preliminary descriptive or inferential analyses, and create publication-ready tables and figures. Fourth, the models support scholarly writing by suggesting journal-specific headings, refining prose, harmonizing references, and translating technical content for multidisciplinary audiences. Fifth, they augment peer-review and editorial workflows by delivering evidence-focused critiques. In educational settings, these models can create adaptive curricula and interactive simulations for trainees, fostering digital literacy and evidence-based practice early in professional development among clinicians. Integration into clinical decision-support pipelines is also foreseeable, warranting proactive governance. Notwithstanding these opportunities, responsible use demands vigilant oversight. Large-language models occasionally fabricate citations or misinterpret domain-specific data ("hallucinations"), potentially propagating misinformation. Outputs are highly prompt-dependent, creating a reliance on informed prompt engineering that may disadvantage less technical clinicians. Moreover, uploading protected health information or copyrighted text raises privacy, security, and intellectual property concerns. We outline best-practice recommendations: maintain human verification of all AI-generated content; cross-validate references with primary databases; employ privacy-preserving, on-premises deployments for sensitive data; document prompts for reproducibility; and disclose AI involvement transparently. In summary, generative AI offers a powerful adjunct for OBGYN scientists by accelerating topic formulation, evidence synthesis, data analysis, manuscript preparation, and peer review. When coupled with rigorous oversight and ethical safeguards, these tools can enhance productivity without compromising scientific integrity. Future studies should quantify accuracy, bias, and downstream patient impact. · Generative AI supports various research stages in OBGYN, such as hypothesis generation, systematic review assistance, data analysis, and scientific writing, demonstrating its potential to streamline research workflows and improve research efficiency.. · Generative AI has notable limitations, including the risk of generating inaccurate references ("hallucinations") and the need for careful supervision.. · Effective usage requires skill in prompt engineering, posing a challenge for those without technical expertise.. · Utilizing generative AI in sensitive fields like OBGYN raises privacy, security, and ethical concerns..
1 Dec 2025
Read appraisal →