{"title":"Can Real-World Data From Hospital Information Systems Be Directly Utilized for Efficacy Studies: An Analysis of Data Time Points.","authors":"Jiaying Luo, Aomeng Zhang, Wulin Gao, Xiangwei Bu, Runming Li, Zehui Ye, Xinyi Zhang, Lili Ren, Yin Jiang, Zhiyue Guan, Hui Guan, Liyuan Tao, Guohua Dai, Chen Zhao, Hongcai Shang","doi":"10.1111/jebm.70079","DOIUrl":"https://doi.org/10.1111/jebm.70079","url":null,"abstract":"<p><strong>Background: </strong>Data from Hospital Information Systems are crucial components of real-world data, but concerns arise regarding the transparency of measurement time-points when directly utilizing for efficacy studies. The objective of the study was to analyze the distribution characteristics of time-points for laboratory test records from HIS.</p><p><strong>Method: </strong>Medical records before December 31, 2019 from Affiliated Hospital of Shandong University of Traditional Chinese Medicine, for patients primarily diagnosed with coronary heart disease (CHD), or heart failure combined with secondary diagnosis of CHD were retrieved from HIS. Fifteen test groups were extracted. The number of records, average of test times, and distribution characteristics of time points were analyzed.</p><p><strong>Results: </strong>The renal function tests have the most records, and the blood glucose have the most times. Tests time-points distribution showed concentration in the early stage of hospitalization, with the majority occurring within the first 0-10%. For those measured ≥2 times, their first tests are also centrally distributed in the early stage, while the lasts in all period of hospitalization. Besides, substantial difference is showed in the time span of the first and last test. Abnormal value may be a trigger that promotes more intensive examination, and normal or abnormal status of the first examination is significantly weak to moderate correlated with the number of examinations and time span.</p><p><strong>Discussion: </strong>The disclosure of time-points of clinical studies based on HIS should be encouraged and the pre-survey of data measuring time point is essential to the design of trial.</p>","PeriodicalId":16090,"journal":{"name":"Journal of Evidence‐Based Medicine","volume":" ","pages":"e70079"},"PeriodicalIF":3.5,"publicationDate":"2025-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145301526","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"A Hybrid Approach for Exploring Real-World Linear Causality Under Multicollinearity Based on Ischemic Post-Stroke Case Series Treated With Integrated Traditional Chinese and Modern Medicine Therapies.","authors":"Zixin Han, Jianxin Chen, Cheng Yu, Chunyu Wang, Xinlin Li, Weici Zheng, Ziyan Gu, Juanjuan Sun, Shuangshuang Hou, Wentao Zhu","doi":"10.1111/jebm.70070","DOIUrl":"https://doi.org/10.1111/jebm.70070","url":null,"abstract":"<p><strong>Objective: </strong>Integration of traditional Chinese and modern medicine (TCM-MM) aids rehabilitation of muscle strength among ischemic stroke (IS) survivors. However, it faces statistical challenges (e.g., multicollinearity, small sample) in the real-world setting. This study tried to provide an analytical framework for investigating linear causality with a retrospective small-sample case series.</p><p><strong>Methods: </strong>Original data was sourced from hospital information system and processed by many means. Wilcoxon signed-rank test was utilized to execute a self-controlled before-and-after comparison, before multiple linear regression (MLR) models were established for exploring prognostic factors of muscle strength improvement. Afterward, Bayesian networks (BN), mediation analysis and between-subjects effects tests were undertaken the detection of underlying multicollinearity sources progressively. Both clinical interpretability and model performance, containing R<sup>2</sup> and mean squared error (MSE), served as the indices for modelling comparison.</p><p><strong>Results: </strong>Muscle strength was significantly improved among 112 post-IS patients after accepting TCM-MM therapies (p < 0.01). Initially, MLR analysis with 11 explanatory variables (EVs) (MLR_1) revealed a probable multicollinearity-driven bias, resulting in reduced interpretability. Consequently, we traced collinearity among EVs using a BN structure that provided clues to mediating and mutual effects for establishing MLR with interactions embracing 11 EVs (MLR_2). Eventually, MLR_2 demonstrated superior model performance (ΔR<sup>2</sup> = 0.097, ΔMSE = -0.004), and better clinical interpretability. Whereas, we cannot deny a 1/3 probability of diminished statistical efficacy due to the small sample size.</p><p><strong>Conclusion: </strong>Our study proposed a practically hybrid approach for exploring linear causality under multicollinearity using real-world small-sample data, which suggested that balancing model performance with clinical interpretability can resolve statistical trade-offs in modelling optimization.</p>","PeriodicalId":16090,"journal":{"name":"Journal of Evidence‐Based Medicine","volume":" ","pages":"e70070"},"PeriodicalIF":3.5,"publicationDate":"2025-10-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145280374","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Min Chen, RuiRui Wu, Rui Chen, Qiong Guo, Ya Deng, Yuan Wang, Youlin Long
{"title":"Acupuncture for Treatment of Obesity: An Umbrella Review.","authors":"Min Chen, RuiRui Wu, Rui Chen, Qiong Guo, Ya Deng, Yuan Wang, Youlin Long","doi":"10.1111/jebm.70083","DOIUrl":"https://doi.org/10.1111/jebm.70083","url":null,"abstract":"<p><strong>Background: </strong>Systematic reviews on acupuncture for obesity report conflicting findings, and the certainty of this evidence remains unclear. This umbrella review appraises the evidence to identify which effects on body mass index (BMI) and body weight (BW) are supported by high-quality findings.</p><p><strong>Methods: </strong>We conducted an umbrella review by systematically searching PubMed, Embase, and the Cochrane Library through April 28, 2025 to obtain systematic reviews of randomized controlled trials evaluating acupuncture interventions for obesity. Two reviewers independently assessed methodological quality using a measurement tool to assess systematic reviews (AMSTAR) and evidence quality using the Grading of Recommendations Assessment, Development, and Evaluation (GRADE). We identified best evidence from high-quality systematic reviews with high or moderate GRADE ratings.</p><p><strong>Results: </strong>Our analysis included 22 systematic reviews encompassing 60 meta-analyses. Seventeen reviews (77.3%) achieved high methodological quality ratings according to AMSTAR criteria. Five meta-analyses provided high-quality evidence consistently demonstrating significant benefits of acupuncture interventions. The strongest evidence supported acupuncture combined with lifestyle interventions compared to lifestyle interventions alone for both BMI and BW outcomes. Additional high-quality evidence demonstrated significant benefits for acupuncture versus no treatment and versus sham acupuncture. Fourteen meta-analyses provided moderate-quality evidence confirming acupuncture effectiveness, with no significant differences between acupuncture and pharmaceutical treatments.</p><p><strong>Conclusions: </strong>Based on high-quality evidence, clinicians can recommend acupuncture to patients with obesity, particularly as adjunctive therapy to lifestyle interventions. For patients unable to tolerate pharmacological treatments, acupuncture represents a reasonable alternative. However, optimal benefits require integration with comprehensive lifestyle modifications rather than standalone use.</p>","PeriodicalId":16090,"journal":{"name":"Journal of Evidence‐Based Medicine","volume":" ","pages":"e70083"},"PeriodicalIF":3.5,"publicationDate":"2025-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145274710","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Xinyu Cui, Kexin Liu, Yuanyi Ji, Su Han, Yongzhong Cheng
{"title":"Two-Way Referrals in Chinese Medical Consortia Under the Framework of Hierarchical Diagnosis and Treatment: Progress and Challenges.","authors":"Xinyu Cui, Kexin Liu, Yuanyi Ji, Su Han, Yongzhong Cheng","doi":"10.1111/jebm.70069","DOIUrl":"https://doi.org/10.1111/jebm.70069","url":null,"abstract":"<p><p>The hierarchical diagnosis and treatment system's two-way referral mechanism is crucial for optimizing medical resource allocation, with medical consortia significantly enhancing this process. It evaluates the implementation dynamics of two-way referral systems within China's medical consortia from 2019 to 2024, focusing on policy evolution, regional disparities, and stakeholder engagement. Despite a positive overall trend, referral rates remain low, with notable regional and institutional disparities. The implementation outcomes span various dimensions, including referral metrics, specific services, timing, and costs; however, these effects are inconsistent and warrant further investigation. The current evaluation index system is diverse but tends to prioritize quantity over quality. Additionally, awareness and satisfaction levels among medical personnel and patients regarding the two-way referral system are uneven and influenced by multiple factors. Currently, China's two-way referral system faces challenges, including low referral volumes, an inadequate evaluation framework, limited research perspectives, and insufficient motivation for stakeholder participation. Future efforts should focus on strengthening primary care infrastructure, enhancing resource collaboration, advancing health insurance reforms, refining the evaluation system, and fostering synergy between hierarchical diagnosis and treatment, the Healthy China strategy, and referral model innovation to advance the hierarchical diagnosis and treatment system. Recommendations emphasize strengthening primary care capacity, reforming insurance payment models, and leveraging digital health technologies to align with the \"Healthy China 2030\" strategic goals.</p>","PeriodicalId":16090,"journal":{"name":"Journal of Evidence‐Based Medicine","volume":" ","pages":"e70069"},"PeriodicalIF":3.5,"publicationDate":"2025-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145274789","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Intervention at Circulating Homocysteine Levels >10 µmol/L for Primary Prevention of Cardiovascular Diseases: A Systematic Review and Dose-Response Meta-Analysis.","authors":"Zihui Huang, Jiupeng Zou, Zhen Li, Mengchu Li, Ruikun He, Jing Zeng, Zhijun Yang, Huabin Li, Xiaoting Lu, Huilian Zhu","doi":"10.1111/jebm.70080","DOIUrl":"https://doi.org/10.1111/jebm.70080","url":null,"abstract":"<p><strong>Objective: </strong>Several studies suggested that the risk of cardiovascular disease (CVD) had increased before individuals developed hyperhomocysteinemia. This study aimed to investigate the dose-response relationship between circulating homocysteine (Hcy) levels and the risk of CVD.</p><p><strong>Methods: </strong>Observational studies examining the relationship between circulating Hcy levels and CVD risks were included. Searches were conducted on English databases (PubMed, Embase, and Web of Science) and Chinese databases (CNKI, WanFang, VIP, and SinoMed) in May 2025. Combined relative risks (RRs) or odds ratios (ORs) were calculated using random-effects models. The dose-response relationship between circulating Hcy levels and CVD risks was assessed using restricted cubic spline analysis. The risk of bias was assessed using the Newcastle-Ottawa Scale for cohort and case-control studies, and the Agency for Healthcare Research and Quality criteria for cross-sectional studies. Publication bias was assessed using funnel plots and the trim-and-fill method.</p><p><strong>Results: </strong>A total of 117 original studies (35 cohort studies, 60 case-control studies, and 22 cross-sectional studies) were included, involving 504,469 participants with 43,089 CVD cases. Elevated circulating Hcy levels were significantly associated with increased CVD risks in all study types (RRs/ORs: 1.61-2.14). A non-linear dose-response relationship was observed between circulating Hcy levels and CVD risks (all p < 0.001), with thresholds at 10.4 µmol/L for cohort studies, and 10.0 µmol/L for both case-control studies and cross-sectional studies.</p><p><strong>Conclusions: </strong>In this meta-analysis, increased Hcy levels were linked to higher CVD risks. Circulating Hcy level >10 µmol/L may implement nutritional intervention in primary prevention of CVD.</p>","PeriodicalId":16090,"journal":{"name":"Journal of Evidence‐Based Medicine","volume":" ","pages":"e70080"},"PeriodicalIF":3.5,"publicationDate":"2025-09-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145199639","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Fen-Fen Li, Ke Han, Zi-Yue Fu, Bing-Yu Liang, Yan-Xun Han, Yu-Chen Liu, Ye-Hai Liu, Bu-Sheng Tong, Hai-Feng Pan
{"title":"Ecological and Regional Disparities in the Impact of Autoimmune Diseases on Global Labor Market Dynamics: A Cross-Sectional Study","authors":"Fen-Fen Li, Ke Han, Zi-Yue Fu, Bing-Yu Liang, Yan-Xun Han, Yu-Chen Liu, Ye-Hai Liu, Bu-Sheng Tong, Hai-Feng Pan","doi":"10.1111/jebm.70073","DOIUrl":"10.1111/jebm.70073","url":null,"abstract":"<div>\u0000 \u0000 \u0000 <section>\u0000 \u0000 <h3> Aim</h3>\u0000 \u0000 <p>Climate change has intensified the prevalence of chronic diseases, particularly autoimmune diseases (ADs), which severely affect the health and labor market participation of the working-age population. While ADs are not typically fatal, their chronic nature and high disability rates lead to significant labor force attrition. This study explores the impact of ADs on the labor market, particularly in regions affected by climate change.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Methods</h3>\u0000 \u0000 <p>This study integrates labor market data with re-estimated ADs burden data from 1990 to 2021. Using time series analysis, multivariate regression, and geographic variation analysis, the research examines the relationship between ADs burden and labor force participation, with a focus on the exacerbating effects of climate change. Data was sourced from the International Labour Organization (ILO) and the Global Burden of Disease (GBD) database.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Results</h3>\u0000 \u0000 <p>In 2021, the global labor force with ADs was 86,295,350, with a prevalence rate of 1644.55 per 100,000. Women had a significantly higher prevalence (1841.96 per 100,000) compared to men (1448.6 per 100,000). The total disability-adjusted life Years (DALYs) for ADs was 18,513,645 person-years, with women experiencing higher DALYs (386.3 per 100,000). Regions severely affected by climate change showed increased ADs prevalence and a decline in labor force participation.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Conclusion</h3>\u0000 \u0000 <p>ADs significantly contribute to global labor market decline, with climate change amplifying the health burden. Gender disparities are notable, and ADs' impact on labor force participation highlights the need for comprehensive public health policies and labor market interventions.</p>\u0000 </section>\u0000 </div>","PeriodicalId":16090,"journal":{"name":"Journal of Evidence‐Based Medicine","volume":"18 3","pages":""},"PeriodicalIF":3.5,"publicationDate":"2025-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145176155","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Hu Zhenyu, Haizhou Xiang, Zeng Ziran, Wu Jiali, Liu Li, Tang Jianwen, Long Menghong, Wang Maohua
{"title":"Efficacy and Safety of Different Preoperative Sedative Regimens in Alleviating Pediatric Preoperative Anxiety: A Systematic Review and Network Meta-Analysis","authors":"Hu Zhenyu, Haizhou Xiang, Zeng Ziran, Wu Jiali, Liu Li, Tang Jianwen, Long Menghong, Wang Maohua","doi":"10.1111/jebm.70068","DOIUrl":"10.1111/jebm.70068","url":null,"abstract":"<div>\u0000 \u0000 \u0000 <section>\u0000 \u0000 <h3> Background</h3>\u0000 \u0000 <p>Pediatric preoperative anxiety (PPA) is a prevalent condition that exhibits significant effects on the psychological and physiological status of children both preoperatively and postoperatively.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Methods</h3>\u0000 \u0000 <p>We conducted systematic review and network meta-analysis. PubMed, Embase, Web of Science Core Collection, and The Cochrane Library were searched up to December 1, 2024. RCTs of pediatric patients (0–14 years) receiving preoperative sedatives were included. Primary outcome was Parental Separation Anxiety Scale (PSAS); secondary outcomes were Mask Acceptance Scale (MAS), postoperative nausea/vomiting (PONV), and delirium/agitation (PODA).</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Results</h3>\u0000 \u0000 <p>Seventy studies (16,626 participants) were included. Five sedatives including midazolam, dexmedetomidine, ketamine (oral, intranasal, nebulized), clonidine (oral, intranasal), and melatonin (oral) were compared with placebo. Data from 20 interventions (5581 patients) assessed PPA. Intranasal dexmedetomidine (ID) showed highest single-drug efficacy (SUCRA: PSAS 68.1%, MAS 48.8%, PONV 65.7%, PODA 67.8%). Oral ketamine (OK) and midazolam (OM/IM) were effective alternatives. Combined regimens were promising but inconclusive.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Conclusions</h3>\u0000 \u0000 <p>ID significantly alleviated PPA with minimal adverse effects in single-drug regimens (optimal dose: 1–2 µg/kg). OK, OM or IM served as potential alternative options for clinical application. While combination regimens (notably OM+OK) demonstrated superior efficacy across outcomes, small sample sizes necessitate cautious interpretation, underscoring the need for future comparative studies.</p>\u0000 </section>\u0000 </div>","PeriodicalId":16090,"journal":{"name":"Journal of Evidence‐Based Medicine","volume":"18 3","pages":""},"PeriodicalIF":3.5,"publicationDate":"2025-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145137814","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Systematic Review and Network Meta-Analysis of the Comparative Effectiveness of Adherence Enhancement Strategies in Chronic Kidney Disease","authors":"Bin Ma, Yuanmin Jia, Haixia Wang, Ou Chen","doi":"10.1111/jebm.70078","DOIUrl":"10.1111/jebm.70078","url":null,"abstract":"<div>\u0000 \u0000 \u0000 <section>\u0000 \u0000 <h3> Objective</h3>\u0000 \u0000 <p>To evaluate and compare the effectiveness of adherence-enhancement strategies in patients with chronic kidney disease (CKD).</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Methods</h3>\u0000 \u0000 <p>Nine databases (PubMed, Embase, The Cochrane Library, Web of Science, Scopus, CNKI, VIP, WanFang, and CBM) were searched for randomized controlled trials (RCTs) to April 1, 2025. Two reviewers independently screened, extracted data, and assessed risk of bias with the Cochrane Risk of Bias 2.0 tool. Certainty of evidence was appraised using the Confidence in Network Meta-Analysis (CINeMA) tool. Network meta-analysis was performed, and surface under the cumulative ranking curve (SUCRA) was calculated to rank interventions. The review was registered in PROSPERO (CRD42024604771).</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Results</h3>\u0000 \u0000 <p>Thirty-five RCTs with 5084 patients were included. Evidence quality was limited by high risk of bias and low certainty. For medication adherence, education plus phone follow-up with short message service education showed the greatest effect (standardized mean differences [SMD] = 2.15, 95% confidence interval [CI] 1.09–3.21; SUCRA = 98.7), followed by empowerment (SMD = 1.19, 95% CI 0.20–2.18; SUCRA = 82.1). For diet adherence, education with phone follow-up was most effective (SMD = 6.68, 95% CI 5.64–7.71; SUCRA = 100), with empowerment also beneficial (SMD = 1.83, 95% CI 1.23–2.43; SUCRA = 87.9). For fluid adherence, education with medication management and pharmacist follow-up was most effective on scale-based outcomes (SMD = 3.06, 95% CI 2.18–3.94; SUCRA = 99.8), while cognitive behavioral therapy reduced interdialytic weight gain (SMD = −0.76, 95% CI −1.30 to −0.22; SUCRA = 71.1).</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Conclusions</h3>\u0000 \u0000 <p>Adherence-enhancement strategies improve medication, diet, and fluid adherence in CKD. High-quality RCTs are needed to confirm these findings.</p>\u0000 </section>\u0000 </div>","PeriodicalId":16090,"journal":{"name":"Journal of Evidence‐Based Medicine","volume":"18 3","pages":""},"PeriodicalIF":3.5,"publicationDate":"2025-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145137823","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Fengxian Chen, Yan Li, Yaolong Chen, Zhaoxiang Bian, La Duo, Qingguo Zhou, Lu Zhang, ADVANCED Working Group
{"title":"Strategies for the Analysis and Elimination of Hallucinations in Artificial Intelligence Generated Medical Knowledge","authors":"Fengxian Chen, Yan Li, Yaolong Chen, Zhaoxiang Bian, La Duo, Qingguo Zhou, Lu Zhang, ADVANCED Working Group","doi":"10.1111/jebm.70075","DOIUrl":"10.1111/jebm.70075","url":null,"abstract":"<div>\u0000 \u0000 <p>The application of artificial intelligence (AI) in healthcare has become increasingly widespread, showing significant potential in assisting with diagnosis and treatment. However, generative AI (GAI) models often produce “hallucinations”—plausible but factually incorrect or unsubstantiated outputs—that threaten clinical decision-making and patient safety. This article systematically analyzes the causes of hallucinations across data, training, and inference dimensions and proposes multi-dimensional strategies to mitigate them. Our findings reveal three critical conclusions: The technical optimization through knowledge graphs and multi-stage training significantly reduces hallucinations, while clinical integration through expert feedback loops and multidisciplinary workflows enhances output reliability. Additionally, implementing robust evaluation systems that combine adversarial testing and real-world validation substantially improves factual accuracy in clinical settings. These integrated strategies underscore the importance of harmonizing technical advancements with clinical governance to develop trustworthy, patient-centric AI systems.</p>\u0000 </div>","PeriodicalId":16090,"journal":{"name":"Journal of Evidence‐Based Medicine","volume":"18 3","pages":""},"PeriodicalIF":3.5,"publicationDate":"2025-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145124918","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}