International Journal of Medical Informatics最新文献_第10页

Efficient medical NER with limited data: Enhancing LLM performance through annotation guidelines 有限数据的高效医疗NER：通过注释指南增强LLM性能。

IF 4.1 2区医学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

International Journal of Medical Informatics

Pub Date : 2025-12-18 DOI: 10.1016/j.ijmedinf.2025.106230

Emiko Shinohara, Yoshimasa Kawazoe

Background

Named entity recognition (NER) is critical in natural language processing (NLP), particularly in the medical field, where accurate identification of entities, such as patient information and clinical events, is essential. Traditional NER approaches rely heavily on large, annotated corpora, which are resource intensive. Large language models (LLMs) offer new NER approaches, particularly through in-context and few-shot learning.

Objective

This study investigates the effects of incorporating annotation guidelines into prompts for NER via LLMs, with a specific focus on their impact on few-shot learning performance across various medical corpora.

Methods

We designed eight different prompt patterns, combining few-shot examples with annotation guidelines of varying complexity, and evaluated their performance via three prominent LLMs: GPT-4o, Claude 3.5 Sonnet, and gpt-oss-120b. Additionally, we employed three diverse medical corpora: i2b2-2014, i2b2-2012, and MedTxt-CR. Accuracy was assessed via precision, recall, and the F1 score, with evaluation methods aligned with those used in relevant shared tasks to ensure the comparability of the results.

Results

Our findings indicate that adding detailed annotation guidelines to few-shot prompts improves the recall and F1 score in most cases.

Conclusion

Including annotation guidelines in prompts enhances the performance of LLMs in NER tasks, making this a practical approach for developing accurate NLP systems in resource-constrained environments. Although annotation guidelines are essential for evaluation and example creation, their integration into LLM prompts can further optimize few-shot learning, especially within specialized domains such as medical NLP.

背景：命名实体识别（NER）在自然语言处理（NLP）中至关重要，特别是在医学领域，准确识别实体（如患者信息和临床事件）至关重要。传统的NER方法严重依赖于大型的、带注释的语料库，这是资源密集型的。大型语言模型（llm）提供了新的NER方法，特别是通过上下文学习和少镜头学习。目的：本研究探讨了通过llm将注释指南纳入NER提示的效果，并特别关注了它们对跨各种医学语料库的少射学习性能的影响。方法：我们设计了8种不同的提示模式，结合了不同复杂性的注释指南，并通过三个著名的llm: gpt- 40、Claude 3.5 Sonnet和gpt- ss-120b来评估它们的性能。此外，我们还采用了三种不同的医疗资料库：i2b2-2014、i2b2-2012和MedTxt-CR。准确性通过精密度、召回率和F1分数来评估，评估方法与相关共享任务中使用的方法一致，以确保结果的可比性。结果：我们的研究结果表明，在大多数情况下，为少量提示添加详细的注释指南可以提高召回率和F1分数。结论：在提示中包含注释指南可以增强llm在NER任务中的性能，使其成为在资源受限环境中开发准确的NLP系统的实用方法。尽管注释指南对于评估和示例创建至关重要，但将它们集成到LLM提示中可以进一步优化少量学习，特别是在医疗NLP等专业领域中。

{"title":"Efficient medical NER with limited data: Enhancing LLM performance through annotation guidelines","authors":"Emiko Shinohara, Yoshimasa Kawazoe","doi":"10.1016/j.ijmedinf.2025.106230","DOIUrl":"10.1016/j.ijmedinf.2025.106230","url":null,"abstract":"<div><h3>Background</h3><div>Named entity recognition (NER) is critical in natural language processing (NLP), particularly in the medical field, where accurate identification of entities, such as patient information and clinical events, is essential. Traditional NER approaches rely heavily on large, annotated corpora, which are resource intensive. Large language models (LLMs) offer new NER approaches, particularly through in-context and few-shot learning.</div></div><div><h3>Objective</h3><div>This study investigates the effects of incorporating annotation guidelines into prompts for NER via LLMs, with a specific focus on their impact on few-shot learning performance across various medical corpora.</div></div><div><h3>Methods</h3><div>We designed eight different prompt patterns, combining few-shot examples with annotation guidelines of varying complexity, and evaluated their performance via three prominent LLMs: GPT-4o, Claude 3.5 Sonnet, and gpt-oss-120b. Additionally, we employed three diverse medical corpora: i2b2-2014, i2b2-2012, and MedTxt-CR. Accuracy was assessed via precision, recall, and the F1 score, with evaluation methods aligned with those used in relevant shared tasks to ensure the comparability of the results.</div></div><div><h3>Results</h3><div>Our findings indicate that adding detailed annotation guidelines to few-shot prompts improves the recall and F1 score in most cases.</div></div><div><h3>Conclusion</h3><div>Including annotation guidelines in prompts enhances the performance of LLMs in NER tasks, making this a practical approach for developing accurate NLP systems in resource-constrained environments. Although annotation guidelines are essential for evaluation and example creation, their integration into LLM prompts can further optimize few-shot learning, especially within specialized domains such as medical NLP.</div></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"208 ","pages":"Article 106230"},"PeriodicalIF":4.1,"publicationDate":"2025-12-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145821980","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}

引用次数: 0

Development and validation of a bleeding risk model for off-pump coronary artery bypass grafting: a multi-center retrospective cohort study 非体外循环冠状动脉旁路移植术出血风险模型的建立和验证：一项多中心回顾性队列研究

IF 4.1 2区医学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

International Journal of Medical Informatics

Pub Date : 2025-12-16 DOI: 10.1016/j.ijmedinf.2025.106226

Zi Wang , Runhua Ma , Qiming Wang , Fan Yang , Xiaotong Xia , Xiaoyu Li , Qing Xu , Yao yao , Hongyi Wu , Chunsheng Wang , Qianzhou Lv

Background

Perioperative bleeding is a major challenge in coronary artery bypass grafting (CABG). Existing bleeding risk models often lack specificity for off-pump CABG (OPCABG) patients.

Objective

This study aims to develop and validate a novel perioperative bleeding prediction model tailored for OPCABG patients.

Methods

This retrospective, multi-center cohort study was conducted using both internal and external validation cohorts. Fourteen different models, including Binary Logistic Regression, Random Forest, Decision Tree, Extra Trees, Adaptive Boosting, Extreme Gradient Boosting, Categorical Boosting, Gradient Boosting, Naive Bayes, Artificial Neural Network, Light Gradient Boosting Machine, K-nearest Neighbors, Support Vector Machine, and LogitBoost, were applied for model development. SHapley Additive exPlanations (SHAP) were used to interpret feature importance and the model’s outputs.

Results

The final model, CABG Bleeding Risk of 10 Variables (CABG-BR10), was built using Random Rorest. This model identified 10 key variables: antiplatelet drug discontinuation, N-terminal pro B-type natriuretic peptide, activated partial thromboplastin time, hemoglobin, urea, cardiac troponin T, estimated glomerular filtration rate, total bilirubin, fibrinogen, and international normalized ratio. In the internal and external validation cohorts, the model demonstrated solid performance with Receiver Operating Characteristic − Area Under the Curve values of 0.90 and 0.87, and Precision-Recall − Area Under the Curve values of 0.70 and 0.67, respectively. SHAP analysis identified key predictors of bleeding risk, and an online tool was developed to facilitate bleeding risk assessment.

Conclusion

The CABG-BR10 model accurately predicts perioperative bleeding risk in OPCABG patients, outperforming traditional scoring systems and providing interpretable, clinically relevant insights into bleeding risk factors.

背景：围手术期出血是冠状动脉旁路移植术（CABG）的主要挑战。现有的出血风险模型对非体外循环CABG （OPCABG）患者往往缺乏特异性。目的建立并验证一种适合OPCABG患者的围手术期出血预测模型。方法采用内部和外部验证队列进行回顾性、多中心队列研究。采用了二元逻辑回归、随机森林、决策树、额外树、自适应增强、极端梯度增强、分类增强、梯度增强、朴素贝叶斯、人工神经网络、轻梯度增强机、k近邻、支持向量机和LogitBoost等14种不同的模型进行模型开发。SHapley加性解释（SHAP）用于解释特征的重要性和模型的输出。结果采用随机抽样方法建立CABG- br10 （CABG- br10）模型。该模型确定了10个关键变量：抗血小板药物停药、n端前b型利钠肽、活化部分凝血活蛋白时间、血红蛋白、尿素、心肌肌钙蛋白T、估计肾小球滤过率、总胆红素、纤维蛋白原和国际标准化比率。在内部和外部验证队列中，该模型表现出良好的性能，接收者工作特征-曲线下面积值分别为0.90和0.87，精确召回率-曲线下面积值分别为0.70和0.67。SHAP分析确定了出血风险的关键预测因素，并开发了一个在线工具来促进出血风险评估。结论CABG-BR10模型可准确预测OPCABG患者围手术期出血风险，优于传统评分系统，为出血危险因素提供可解释的、临床相关的见解。

{"title":"Development and validation of a bleeding risk model for off-pump coronary artery bypass grafting: a multi-center retrospective cohort study","authors":"Zi Wang , Runhua Ma , Qiming Wang , Fan Yang , Xiaotong Xia , Xiaoyu Li , Qing Xu , Yao yao , Hongyi Wu , Chunsheng Wang , Qianzhou Lv","doi":"10.1016/j.ijmedinf.2025.106226","DOIUrl":"10.1016/j.ijmedinf.2025.106226","url":null,"abstract":"<div><h3>Background</h3><div>Perioperative bleeding is a major challenge in coronary artery bypass grafting (CABG). Existing bleeding risk models often lack specificity for off-pump CABG (OPCABG) patients.</div></div><div><h3>Objective</h3><div>This study aims to develop and validate a novel perioperative bleeding prediction model tailored for OPCABG patients.</div></div><div><h3>Methods</h3><div>This retrospective, multi-center cohort study was conducted using both internal and external validation cohorts. Fourteen different models, including Binary Logistic Regression, Random Forest, Decision Tree, Extra Trees, Adaptive Boosting, Extreme Gradient Boosting, Categorical Boosting, Gradient Boosting, Naive Bayes, Artificial Neural Network, Light Gradient Boosting Machine, K-nearest Neighbors, Support Vector Machine, and LogitBoost, were applied for model development. SHapley Additive exPlanations (SHAP) were used to interpret feature importance and the model’s outputs.</div></div><div><h3>Results</h3><div>The final model, CABG Bleeding Risk of 10 Variables (CABG-BR10), was built using Random Rorest. This model identified 10 key variables: antiplatelet drug discontinuation, N-terminal pro B-type natriuretic peptide, activated partial thromboplastin time, hemoglobin, urea, cardiac troponin T, estimated glomerular filtration rate, total bilirubin, fibrinogen, and international normalized ratio. In the internal and external validation cohorts, the model demonstrated solid performance with Receiver Operating Characteristic − Area Under the Curve values of 0.90 and 0.87, and Precision-Recall − Area Under the Curve values of 0.70 and 0.67, respectively. SHAP analysis identified key predictors of bleeding risk, and an online tool was developed to facilitate bleeding risk assessment.</div></div><div><h3>Conclusion</h3><div>The CABG-BR10 model accurately predicts perioperative bleeding risk in OPCABG patients, outperforming traditional scoring systems and providing interpretable, clinically relevant insights into bleeding risk factors.</div></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"208 ","pages":"Article 106226"},"PeriodicalIF":4.1,"publicationDate":"2025-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145799858","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}

引用次数: 0

Predictive modeling of hospital emergency department demand using artificial intelligence: A systematic review 基于人工智能的医院急诊科需求预测建模：系统综述。

IF 4.1 2区医学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

International Journal of Medical Informatics

Pub Date : 2025-12-13 DOI: 10.1016/j.ijmedinf.2025.106215

Jorge Blanco , Marina Ferreras , Oscar Cosido

Background

Accurately forecasting patient arrivals in hospital emergency departments (EDs) is critical for hospital capacity and planning and clinical decision-making. Artificial intelligence (AI), particularly machine learning (ML) and deep learning (DL), has shown promising performance over traditional time series approaches. However, the extent to which these models are validated and generalizable remains uncertain.

Objective

To systematically review the literature on predictive models for hospital ED demand forecasting, focusing on algorithms used, internal and external variables, validation strategies and limitations pre- and post-pandemic developments.

Methods

A systematic literature review (SLR) was conducted following PRISMA guidelines. Five databases (PubMed, IEEE, Springer, ScienceDirect, ACM) were searched for peer-reviewed articles published between January 2019 and July 2025. Eligible studies applied predictive algorithms – excluding those focused on COVID-19 – to forecast ED visits. Extracted data included modeling approaches, feature types, evaluation metrics, and validation methods.

Results

Eleven studies met the inclusion criteria. Classical models such as ARIMA and SARIMA remain in use, but ML (e.g., XGBoost, Random Forest) and DL (e.g., LSTM, CNN) showed higher predictive accuracy, especially with high-dimensional, nonlinear data. Incorporating external variables—such as weather (temperature, humidity, wind), air quality, and calendar events—consistently improved performance. Common metrics included Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and Mean Absolute Percentage Error (MAPE), with MAPE ranging from 3 % to 18 %. Few studies performed external validation, and only a minority employed explainable AI methods (e.g., SHAP) to address interpretability.

Conclusions

AI-based models offer strong potential for ED demand forecasting, particularly when integrating environmental and temporal features. However, limited external validation and lack of interpretability remain significant barriers to clinical adoption. Future research should prioritize multicenter validation, standardized evaluation, and explainable AI to support reliable, transparent, and scalable use in hospital emergency departments.

背景：准确预测医院急诊科（EDs）的患者到达对医院容量、规划和临床决策至关重要。人工智能（AI），特别是机器学习（ML）和深度学习（DL），已经比传统的时间序列方法表现出了很好的表现。然而，这些模型被验证和推广的程度仍然不确定。目的：系统回顾有关医院急诊科需求预测预测模型的文献，重点关注所使用的算法、内部和外部变量、验证策略以及大流行前后发展的局限性。方法：根据PRISMA指南进行系统文献回顾（SLR）。五个数据库（PubMed， IEEE， b施普林格，ScienceDirect， ACM）检索了2019年1月至2025年7月之间发表的同行评议文章。符合条件的研究应用了预测算法（不包括那些关注COVID-19的研究）来预测急诊科就诊。提取的数据包括建模方法、特征类型、评估指标和验证方法。结果：11项研究符合纳入标准。经典模型如ARIMA和SARIMA仍在使用，但ML（如XGBoost、Random Forest）和DL（如LSTM、CNN）显示出更高的预测精度，特别是在高维、非线性数据下。结合外部变量—例如天气（温度、湿度、风）、空气质量和日历事件—可以持续提高性能。常用指标包括平均绝对误差（MAE）、均方根误差（RMSE）和平均绝对百分比误差（MAPE）， MAPE的范围从3%到18%。很少有研究进行外部验证，只有少数研究采用可解释的AI方法（例如，SHAP）来解决可解释性问题。结论：基于人工智能的模型为ED需求预测提供了强大的潜力，特别是在整合环境和时间特征时。然而，有限的外部验证和缺乏可解释性仍然是临床采用的重大障碍。未来的研究应优先考虑多中心验证、标准化评估和可解释的人工智能，以支持在医院急诊科可靠、透明和可扩展的使用。

{"title":"Predictive modeling of hospital emergency department demand using artificial intelligence: A systematic review","authors":"Jorge Blanco , Marina Ferreras , Oscar Cosido","doi":"10.1016/j.ijmedinf.2025.106215","DOIUrl":"10.1016/j.ijmedinf.2025.106215","url":null,"abstract":"<div><h3>Background</h3><div>Accurately forecasting patient arrivals in hospital emergency departments (EDs) is critical for hospital capacity and planning and clinical decision-making. Artificial intelligence (AI), particularly machine learning (ML) and deep learning (DL), has shown promising performance over traditional time series approaches. However, the extent to which these models are validated and generalizable remains uncertain.</div></div><div><h3>Objective</h3><div>To systematically review the literature on predictive models for hospital ED demand forecasting, focusing on algorithms used, internal and external variables, validation strategies and limitations pre- and post-pandemic developments.</div></div><div><h3>Methods</h3><div>A systematic literature review (SLR) was conducted following PRISMA guidelines. Five databases (PubMed, IEEE, Springer, ScienceDirect, ACM) were searched for peer-reviewed articles published between January 2019 and July 2025. Eligible studies applied predictive algorithms – excluding those focused on COVID-19 – to forecast ED visits. Extracted data included modeling approaches, feature types, evaluation metrics, and validation methods.</div></div><div><h3>Results</h3><div>Eleven studies met the inclusion criteria. Classical models such as ARIMA and SARIMA remain in use, but ML (e.g., XGBoost, Random Forest) and DL (e.g., LSTM, CNN) showed higher predictive accuracy, especially with high-dimensional, nonlinear data. Incorporating external variables—such as weather (temperature, humidity, wind), air quality, and calendar events—consistently improved performance. Common metrics included Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and Mean Absolute Percentage Error (MAPE), with MAPE ranging from 3 % to 18 %. Few studies performed external validation, and only a minority employed explainable AI methods (e.g., SHAP) to address interpretability.</div></div><div><h3>Conclusions</h3><div>AI-based models offer strong potential for ED demand forecasting, particularly when integrating environmental and temporal features. However, limited external validation and lack of interpretability remain significant barriers to clinical adoption. Future research should prioritize multicenter validation, standardized evaluation, and explainable AI to support reliable, transparent, and scalable use in hospital emergency departments.</div></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"207 ","pages":"Article 106215"},"PeriodicalIF":4.1,"publicationDate":"2025-12-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145769787","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}

引用次数: 0

Impact of AI recommendation correctness on diagnostic accuracy in clinical decision-making 人工智能推荐正确性对临床决策诊断准确性的影响

IF 4.1 2区医学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

International Journal of Medical Informatics

Pub Date : 2025-12-13 DOI: 10.1016/j.ijmedinf.2025.106223

Florian Kücking , Dorothee A. Busch , Mareike Przysucha , Jan-Oliver Kutza , Niels Hannemann , Jens Hüsers , Birgit Babitsch , Ursula Hübner

Background

Clinical decision-making is shaped by healthcare provider-related factors such as experience, qualification and cognitive skills. AI-based Clinical Decision Support Systems (CDSS) promise to enhance diagnostic accuracy but may also introduce risks, particularly through automation bias. The relative impact of correct and incorrect AI recommendations compared to human factors remains poorly understood.

Methods

A simulated diagnostic intervention study was conducted with 223 physicians and nurses, who generated 1,338 decisions when assessing wound maceration from images combined with AI recommendations. Participants first completed a baseline assessment of diagnostic performance without AI support, followed by a second phase including AI recommendations (correct or incorrect, based on a CNN). Diagnostic decisions were analysed using a generalised linear mixed model (GLMM) to examine the influence of AI recommendation correctness and healthcare provider-related factors (diagnostic performance, qualification, experience, trust in AI, gender, profession, age, healthcare sector) on decision accuracy.

Results

AI recommendations had a strong and bidirectional influence on diagnostic accuracy. Participants were ten times more likely to make correct decisions when receiving a correct AI recommendation (OR = 10.0, p < 0.001), but their accuracy decreased reciprocally when the AI recommendation was incorrect. Among provider-related factors, high baseline diagnostic performance (OR = 2.44, p = 0.019), pertinent formal qualifications (OR = 1.40, p = 0.049), longer work experience (OR = 1.89, p = 0.018), and female gender (OR = 1.55, p = 0.008) were associated with higher diagnostic accuracy. Trust in AI, age, profession, and healthcare sector showed no significant effects in the multivariate model. The overall effect of introducing AI was equivocal compared to baseline, however, there was a differential effect whether the recommendation was correct or wrong.

Conclusions

AI recommendations can exert a stronger influence on diagnostic decisions than healthcare provider-related factors. While AI support improved accuracy when correct, it reduced accuracy when incorrect, indicating overreliance on the system and posing a substantial safety risk. These findings highlight the dual nature of AI in clinical decision support and underscore the imperative for systems with consistently high quality in clinical practice. Equally important, clinicians must receive training and support to critically assess AI recommendations when making clinical decisions.

临床决策受到医疗保健提供者相关因素的影响，如经验、资格和认知技能。基于人工智能的临床决策支持系统（CDSS）有望提高诊断准确性，但也可能引入风险，特别是通过自动化偏见。与人为因素相比，正确和不正确的人工智能建议的相对影响仍然知之甚少。方法对223名医生和护士进行模拟诊断干预研究，在结合人工智能推荐的图像评估伤口浸渍时，他们产生了1338个决策。参与者首先在没有人工智能支持的情况下完成诊断性能的基线评估，然后是第二阶段，包括人工智能建议（基于CNN的正确或错误）。使用广义线性混合模型（GLMM）分析诊断决策，以检查人工智能推荐正确性和医疗保健提供者相关因素（诊断性能、资格、经验、对人工智能的信任、性别、职业、年龄、医疗保健部门）对决策准确性的影响。结果推荐对诊断准确性有很强的双向影响。当收到正确的人工智能推荐时，参与者做出正确决策的可能性增加了10倍（OR = 10.0, p < 0.001），但当人工智能推荐不正确时，他们的准确率会相应下降。在医护人员相关因素中，较高的基线诊断表现（OR = 2.44, p = 0.019）、相关的正式资格（OR = 1.40, p = 0.049）、较长的工作经验（OR = 1.89, p = 0.018）和女性（OR = 1.55, p = 0.008）与较高的诊断准确性相关。对人工智能的信任、年龄、职业和医疗保健部门在多变量模型中没有显着影响。与基线相比，引入人工智能的总体效果是模棱两可的，然而，无论建议是正确还是错误，都会产生不同的影响。结论推荐对诊断决策的影响大于医疗服务提供者相关因素。虽然人工智能支持在正确时提高准确性，但在错误时却降低了准确性，这表明过度依赖系统并带来了巨大的安全风险。这些发现突出了人工智能在临床决策支持中的双重性质，并强调了在临床实践中始终保持高质量的系统的必要性。同样重要的是，临床医生必须接受培训和支持，以便在做出临床决策时批判性地评估人工智能建议。

{"title":"Impact of AI recommendation correctness on diagnostic accuracy in clinical decision-making","authors":"Florian Kücking , Dorothee A. Busch , Mareike Przysucha , Jan-Oliver Kutza , Niels Hannemann , Jens Hüsers , Birgit Babitsch , Ursula Hübner","doi":"10.1016/j.ijmedinf.2025.106223","DOIUrl":"10.1016/j.ijmedinf.2025.106223","url":null,"abstract":"<div><h3>Background</h3><div>Clinical decision-making is shaped by healthcare provider-related factors such as experience, qualification and cognitive skills. AI-based Clinical Decision Support Systems (CDSS) promise to enhance diagnostic accuracy but may also introduce risks, particularly through automation bias. The relative impact of correct and incorrect AI recommendations compared to human factors remains poorly understood.</div></div><div><h3>Methods</h3><div>A simulated diagnostic intervention study was conducted with 223 physicians and nurses, who generated 1,338 decisions when assessing wound maceration from images combined with AI recommendations. Participants first completed a baseline assessment of diagnostic performance without AI support, followed by a second phase including AI recommendations (correct or incorrect, based on a CNN). Diagnostic decisions were analysed using a generalised linear mixed model (GLMM) to examine the influence of AI recommendation correctness and healthcare provider-related factors (diagnostic performance, qualification, experience, trust in AI, gender, profession, age, healthcare sector) on decision accuracy.</div></div><div><h3>Results</h3><div>AI recommendations had a strong and bidirectional influence on diagnostic accuracy. Participants were ten times more likely to make correct decisions when receiving a correct AI recommendation (OR = 10.0, p < 0.001), but their accuracy decreased reciprocally when the AI recommendation was incorrect. Among provider-related factors, high baseline diagnostic performance (OR = 2.44, p = 0.019), pertinent formal qualifications (OR = 1.40, p = 0.049), longer work experience (OR = 1.89, p = 0.018), and female gender (OR = 1.55, p = 0.008) were associated with higher diagnostic accuracy. Trust in AI, age, profession, and healthcare sector showed no significant effects in the multivariate model. The overall effect of introducing AI was equivocal compared to baseline, however, there was a differential effect whether the recommendation was correct or wrong.</div></div><div><h3>Conclusions</h3><div>AI recommendations can exert a stronger influence on diagnostic decisions than healthcare provider-related factors. While AI support improved accuracy when correct, it reduced accuracy when incorrect, indicating overreliance on the system and posing a substantial safety risk. These findings highlight the dual nature of AI in clinical decision support and underscore the imperative for systems with consistently high quality in clinical practice. Equally important, clinicians must receive training and support to critically assess AI recommendations when making clinical decisions.</div></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"207 ","pages":"Article 106223"},"PeriodicalIF":4.1,"publicationDate":"2025-12-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145736658","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}

引用次数: 0

Generative artificial intelligence as a source of advice on resuscitation and first aid for laypeople: A scoping review 生成人工智能作为外行人复苏和急救建议的来源：范围审查

IF 4.1 2区医学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

International Journal of Medical Informatics

Pub Date : 2025-12-13 DOI: 10.1016/j.ijmedinf.2025.106224

Alexei A. Birkun

Introduction

The performance of cutting-edge generative artificial intelligence (GenAI) in guiding laypeople on how to give help in health emergencies is attracting growing attention. This study aimed to map and summarise original research evidence on the quality of GenAI-synthesised advice on resuscitation and first aid.

Methods

The review encompassed journal publications that reported original quantitative data on the quality (accuracy, correctness, completeness, appropriateness) of GenAI-synthesised advice on how laypeople should perform cardiopulmonary resuscitation or provide first aid. Relevant papers were identified through PubMed, Scopus, and Google Scholar. Studies were included if they were published in English as an article, short report, letter, or note during the period 2017–2025. The review was conducted following the recommendations of the PRISMA extension for Scoping Reviews.

Results

Among the 19 eligible studies, 17 evaluated the performance of text-generating GenAI tools, one tested user-to-GenAI voice interaction and another one investigated text-to-video generation capabilities. The studies exhibited substantial heterogeneity in research design, methods, and reporting. Most of them (89.5 %) presented evidence of flaws in the generation of advice on resuscitation or first aid, including a failure to synthesise requested content (reported by 15.8 % of the studies), the creation of incomplete instructions (57.9 %), inaccurate instructions (57.9 %), or superfluous guidance (36.8 %), irrelevant or potentially harmful. The prevalence of misinformation varied from study to study, at times encompassing the whole sample of evaluated GenAI responses. Some authors did not accentuate the issue of misinformation despite the reported data indicating quality defects.

Conclusions

Current evidence indicates risks associated with the unsupervised generation of resuscitation and first aid guidance by publicly available GenAI, as the synthesised content often contains misinformation that may mislead users and induce harmful actions. There is a growing need for international collaboration to develop coordinated strategies to limit GenAI-driven misinformation and mitigate potential health risks.

尖端的生成式人工智能（GenAI）在指导非专业人员如何在突发卫生事件中提供帮助方面的表现越来越受到关注。本研究旨在绘制和总结genai合成的复苏和急救建议质量的原始研究证据。方法：本综述纳入了报道genai综合建议的质量（准确性、正确性、完整性、适当性）的原始定量数据的期刊出版物，这些建议是关于外行人应该如何进行心肺复苏或提供急救的。相关论文通过PubMed、Scopus和b谷歌Scholar检索。在2017-2025年期间以英文发表的文章、简短报告、信件或笔记被纳入研究。该审查是根据PRISMA扩展范围审查的建议进行的。结果在19项符合条件的研究中，17项评估了文本生成GenAI工具的性能，一项测试了用户对GenAI的语音交互，另一项调查了文本到视频的生成能力。这些研究在研究设计、方法和报告方面表现出实质性的异质性。其中大多数（89.5%）提出了在生成复苏或急救建议方面存在缺陷的证据，包括未能综合所要求的内容（15.8%的研究报告），创建不完整的说明（57.9%），不准确的说明（57.9%）或多余的指导（36.8%），不相关或潜在有害。错误信息的流行程度因研究而异，有时包括被评估的GenAI反应的整个样本。尽管报告的数据表明存在质量缺陷，但一些作者并未强调错误信息的问题。结论：目前的证据表明，在没有监督的情况下，通过公开的GenAI生成复苏和急救指南存在风险，因为合成内容通常包含可能误导用户并诱发有害行为的错误信息。越来越需要进行国际合作，制定协调一致的战略，以限制基因技术驱动的错误信息并减轻潜在的健康风险。

{"title":"Generative artificial intelligence as a source of advice on resuscitation and first aid for laypeople: A scoping review","authors":"Alexei A. Birkun","doi":"10.1016/j.ijmedinf.2025.106224","DOIUrl":"10.1016/j.ijmedinf.2025.106224","url":null,"abstract":"<div><h3>Introduction</h3><div>The performance of cutting-edge generative artificial intelligence (GenAI) in guiding laypeople on how to give help in health emergencies is attracting growing attention. This study aimed to map and summarise original research evidence on the quality of GenAI-synthesised advice on resuscitation and first aid.</div></div><div><h3>Methods</h3><div>The review encompassed journal publications that reported original quantitative data on the quality (accuracy, correctness, completeness, appropriateness) of GenAI-synthesised advice on how laypeople should perform cardiopulmonary resuscitation or provide first aid. Relevant papers were identified through PubMed, Scopus, and Google Scholar. Studies were included if they were published in English as an article, short report, letter, or note during the period 2017–2025. The review was conducted following the recommendations of the PRISMA extension for Scoping Reviews.</div></div><div><h3>Results</h3><div>Among the 19 eligible studies, 17 evaluated the performance of text-generating GenAI tools, one tested user-to-GenAI voice interaction and another one investigated text-to-video generation capabilities. The studies exhibited substantial heterogeneity in research design, methods, and reporting. Most of them (89.5 %) presented evidence of flaws in the generation of advice on resuscitation or first aid, including a failure to synthesise requested content (reported by 15.8 % of the studies), the creation of incomplete instructions (57.9 %), inaccurate instructions (57.9 %), or superfluous guidance (36.8 %), irrelevant or potentially harmful. The prevalence of misinformation varied from study to study, at times encompassing the whole sample of evaluated GenAI responses. Some authors did not accentuate the issue of misinformation despite the reported data indicating quality defects.</div></div><div><h3>Conclusions</h3><div>Current evidence indicates risks associated with the unsupervised generation of resuscitation and first aid guidance by publicly available GenAI, as the synthesised content often contains misinformation that may mislead users and induce harmful actions. There is a growing need for international collaboration to develop coordinated strategies to limit GenAI-driven misinformation and mitigate potential health risks.</div></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"207 ","pages":"Article 106224"},"PeriodicalIF":4.1,"publicationDate":"2025-12-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145737403","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}

引用次数: 0

Exploring hospital staff experiences and requirements for safer and more equitable EHR-integrated medication management system: A qualitative study 探讨医院员工对更安全、更公平的ehr整合药物管理系统的经验和需求：一项定性研究

IF 4.1 2区医学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

International Journal of Medical Informatics

Pub Date : 2025-12-11 DOI: 10.1016/j.ijmedinf.2025.106222

Sreyon Murthi , Shane Scahill , Nazanin Ghahreman-Falconer , Nataly Martini

Background

Medication-related harm remains a global challenge, causing preventable illness, mortality, and cost. Electronic health records (EHRs) and electronic medication management systems (eMMS) aim to improve safety through standardisation and decision support, yet persistent usability issues, fragmented workflows, and limited consideration of equity continue to compromise outcomes, particularly for Indigenous, older, and disabled populations.

Aim

To explore how doctors, nurses, pharmacists, digital health professionals, and organisational leaders experience EHR-integrated medication systems with particular attention to medication safety, workflow, and inequities.

Methods

An exploratory descriptive qualitative study was conducted at a large tertiary hospital in New Zealand (January – February 2025). Data were analysed using Braun and Clarke’s reflexive thematic analysis, guided by User-Centred Design, the Sociotechnical Model, the Digital Health Equity Framework, and the Māori wellbeing model Te Whare Tapa Whā. Rigour was supported through reflexive journaling, supervisory review, and COREQ adherence.

Results

Thirteen semi-structured interviews and two focus groups were held with 22 participants. Three overlapping themes were identified. System design and organisational factors influencing medication safety showed how complex interfaces, rigid automation, and unreliable infrastructure increased cognitive load and required vigilance. Limited recognition of cultural and social needs in eMMS, revealed that missing fields for language, disability, and family context constrained culturally safe communication and equitable care. Design priorities for safer, more inclusive systems captured participants’ vision of adaptive decision support, integrated data views, and embedded equity features such as interpreter prompts, disability and family engagement fields, and co-design to ensure systems evolve with clinical and cultural practice.

Conclusions

Improving medication safety requires more than digital transformation of medication processes. Integrating user-centred design, sociotechnical awareness, and equity principles can create intelligent, context-aware systems that support, rather than replace, clinical judgement, empathy, and human connection in patient care.

与药物相关的伤害仍然是一个全球性的挑战，造成可预防的疾病、死亡率和成本。电子健康记录（EHRs）和电子药物管理系统（eMMS）旨在通过标准化和决策支持来提高安全性，但持续存在的可用性问题、碎片化的工作流程和对公平的有限考虑继续损害结果，特别是对土著、老年人和残疾人群体。目的探讨医生、护士、药剂师、数字卫生专业人员和组织领导者如何体验电子病历集成的药物系统，特别关注药物安全、工作流程和不公平现象。方法于2025年1 - 2月在新西兰某大型三级医院进行探索性描述性定性研究。在以用户为中心的设计、社会技术模型、数字健康公平框架和Māori健康模型Te Whare Tapa Whā的指导下，使用Braun和Clarke的反思性主题分析对数据进行了分析。严谨性通过反思性日志记录、监督审查和COREQ依从性得到支持。结果共进行了13次半结构化访谈和2次焦点小组访谈，共22人。确定了三个重叠的主题。影响用药安全的系统设计和组织因素表明，复杂的界面、严格的自动化和不可靠的基础设施如何增加认知负荷和需要警惕。eMMS对文化和社会需求的认识有限，表明语言、残疾和家庭背景等领域的缺失限制了文化安全沟通和公平护理。设计更安全、更具包容性的系统的优先事项体现了参与者对自适应决策支持、集成数据视图和嵌入式公平功能（如口译提示、残疾和家庭参与领域）的愿景，并共同设计以确保系统随着临床和文化实践而发展。结论提高用药安全需要的不仅仅是用药流程的数字化改造。将以用户为中心的设计、社会技术意识和公平原则相结合，可以创建智能的情境感知系统，支持而不是取代临床判断、移情和患者护理中的人际联系。

{"title":"Exploring hospital staff experiences and requirements for safer and more equitable EHR-integrated medication management system: A qualitative study","authors":"Sreyon Murthi , Shane Scahill , Nazanin Ghahreman-Falconer , Nataly Martini","doi":"10.1016/j.ijmedinf.2025.106222","DOIUrl":"10.1016/j.ijmedinf.2025.106222","url":null,"abstract":"<div><h3>Background</h3><div>Medication-related harm remains a global challenge, causing preventable illness, mortality, and cost. Electronic health records (EHRs) and electronic medication management systems (eMMS) aim to improve safety through standardisation and decision support, yet persistent usability issues, fragmented workflows, and limited consideration of equity continue to compromise outcomes, particularly for Indigenous, older, and disabled populations.</div></div><div><h3>Aim</h3><div>To explore how doctors, nurses, pharmacists, digital health professionals, and organisational leaders experience EHR-integrated medication systems with particular attention to medication safety, workflow, and inequities.</div></div><div><h3>Methods</h3><div>An exploratory descriptive qualitative study was conducted at a large tertiary hospital in New Zealand (January – February 2025). Data were analysed using Braun and Clarke’s reflexive thematic analysis, guided by User-Centred Design, the Sociotechnical Model, the Digital Health Equity Framework, and the Māori wellbeing model Te Whare Tapa Whā. Rigour was supported through reflexive journaling, supervisory review, and COREQ adherence.</div></div><div><h3>Results</h3><div>Thirteen semi-structured interviews and two focus groups were held with 22 participants. Three overlapping themes were identified. <em>System design and organisational factors influencing medication safety</em> showed how complex interfaces, rigid automation, and unreliable infrastructure increased cognitive load and required vigilance. <em>Limited recognition of cultural and social needs in eMMS</em>, revealed that missing fields for language, disability, and family context constrained culturally safe communication and equitable care. <em>Design priorities for safer, more inclusive systems</em> captured participants’ vision of adaptive decision support, integrated data views, and embedded equity features such as interpreter prompts, disability and family engagement fields, and co-design to ensure systems evolve with clinical and cultural practice.</div></div><div><h3>Conclusions</h3><div>Improving medication safety requires more than digital transformation of medication processes. Integrating user-centred design, sociotechnical awareness, and equity principles can create intelligent, context-aware systems that support, rather than replace, clinical judgement, empathy, and human connection in patient care.</div></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"207 ","pages":"Article 106222"},"PeriodicalIF":4.1,"publicationDate":"2025-12-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145736657","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}

引用次数: 0

Implementation of digital health interventions in thyroid cancer care: A scoping review 数字健康干预在甲状腺癌护理中的实施：范围综述

IF 4.1 2区医学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

International Journal of Medical Informatics

Pub Date : 2025-12-10 DOI: 10.1016/j.ijmedinf.2025.106221

Mingxia Zhu, Xinlu Wang, Lei Mei

Background

Thyroid cancer is one of the fastest-growing malignancies worldwide, with an increasing number of long-term survivors requiring ongoing care. Digital health interventions offer scalable and patient-centered approaches to support this population, yet it remains unclear how these interventions have been designed, delivered, evaluated, and adopted in practice. This scoping review aims to synthesize current evidence on the characteristics, effectiveness, and uptake of digital health interventions in thyroid cancer care.

Methods

A systematic search was performed in PubMed, Web of Science, Embase, CINAHL, CENTRAL, PsycINFO, China National Knowledge Infrastructure (CNKI), Wanfang, and the Chinese Biomedical Literature Database (CBM) from inception to April 6, 2025. Reference lists of included articles were also screened manually. Information on study characteristics, participant details, intervention types and settings, outcomes, and reported barriers and facilitators was extracted.

Results

A total of 23 studies were included, most of which were conducted since 2018 and focused on postoperative patients. Interventions primarily targeted follow-up care, postoperative rehabilitation, and psychological support, using technologies such as smartphone apps, AI-driven systems, and telehealth. Preliminary evidence suggests improvement in quality of life, psychological well-being, and treatment-related adherence. However, few studies evaluated implementation outcomes or addressed barriers like digital literacy and privacy concerns. Theoretical frameworks and stakeholder co-design were rarely used.

Conclusions

Digital health interventions in thyroid cancer care show promise but remain underutilized in clinical practice. Future research should enhance uptake and implementation by expanding target populations, integrating emerging technologies, and adopting more rigorous and grounded approaches. Addressing these gaps may guide continued innovation and support broader integration of digital health interventions in this population.

背景：甲状腺癌是世界范围内增长最快的恶性肿瘤之一，越来越多的长期幸存者需要持续治疗。数字卫生干预措施提供了可扩展和以患者为中心的方法来支持这一人群，但目前尚不清楚这些干预措施是如何设计、交付、评估和在实践中采用的。本综述旨在综合目前关于甲状腺癌护理中数字健康干预的特征、有效性和吸收的证据。方法系统检索PubMed、Web of Science、Embase、CINAHL、CENTRAL、PsycINFO、中国知网（CNKI）、万方、中国生物医学文献数据库（CBM）自建站至2025年4月6日的相关文献。纳入文章的参考文献列表也进行了人工筛选。提取了有关研究特征、参与者详细信息、干预类型和设置、结果以及报告的障碍和促进因素的信息。结果共纳入23项研究，其中大部分研究于2018年以后开展，主要针对术后患者。干预措施主要针对随访护理、术后康复和心理支持，使用智能手机应用程序、人工智能驱动系统和远程医疗等技术。初步证据表明生活质量、心理健康和治疗相关依从性得到改善。然而，很少有研究评估实施结果或解决数字素养和隐私问题等障碍。理论框架和利益相关者协同设计很少使用。结论数字化健康干预在甲状腺癌护理中的应用前景广阔，但在临床实践中仍未得到充分利用。未来的研究应通过扩大目标人群、整合新兴技术和采用更严格和更有根据的方法来加强吸收和实施。解决这些差距可以指导持续创新，并支持在这一人群中更广泛地整合数字卫生干预措施。

{"title":"Implementation of digital health interventions in thyroid cancer care: A scoping review","authors":"Mingxia Zhu, Xinlu Wang, Lei Mei","doi":"10.1016/j.ijmedinf.2025.106221","DOIUrl":"10.1016/j.ijmedinf.2025.106221","url":null,"abstract":"<div><h3>Background</h3><div>Thyroid cancer is one of the fastest-growing malignancies worldwide, with an increasing number of long-term survivors requiring ongoing care. Digital health interventions offer scalable and patient-centered approaches to support this population, yet it remains unclear how these interventions have been designed, delivered, evaluated, and adopted in practice. This scoping review aims to synthesize current evidence on the characteristics, effectiveness, and uptake of digital health interventions in thyroid cancer care.</div></div><div><h3>Methods</h3><div>A systematic search was performed in PubMed, Web of Science, Embase, CINAHL, CENTRAL, PsycINFO, China National Knowledge Infrastructure (CNKI), Wanfang, and the Chinese Biomedical Literature Database (CBM) from inception to April 6, 2025. Reference lists of included articles were also screened manually. Information on study characteristics, participant details, intervention types and settings, outcomes, and reported barriers and facilitators was extracted.</div></div><div><h3>Results</h3><div>A total of 23 studies were included, most of which were conducted since 2018 and focused on postoperative patients. Interventions primarily targeted follow-up care, postoperative rehabilitation, and psychological support, using technologies such as smartphone apps, AI-driven systems, and telehealth. Preliminary evidence suggests improvement in quality of life, psychological well-being, and treatment-related adherence. However, few studies evaluated implementation outcomes or addressed barriers like digital literacy and privacy concerns. Theoretical frameworks and stakeholder co-design were rarely used.</div></div><div><h3>Conclusions</h3><div>Digital health interventions in thyroid cancer care show promise but remain underutilized in clinical practice. Future research should enhance uptake and implementation by expanding target populations, integrating emerging technologies, and adopting more rigorous and grounded approaches. Addressing these gaps may guide continued innovation and support broader integration of digital health interventions in this population.</div></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"207 ","pages":"Article 106221"},"PeriodicalIF":4.1,"publicationDate":"2025-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145736656","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}

引用次数: 0

Virtual care in residential aged care and primary care settings: a systematic literature review using the SEIPS framework 虚拟护理在住宅养老和初级保健设置：使用SEIPS框架的系统文献综述。

IF 4.1 2区医学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

International Journal of Medical Informatics

Pub Date : 2025-12-10 DOI: 10.1016/j.ijmedinf.2025.106218

Meredith A.B. Makeham , Tamasha Jayawardena , Samiha Elkheir , Ann Carrigan , Soumya , Heather Russell , Mirela Prgomet , Vaibhav Tyagi , Amy Von Huben , Melissa Baysari , Adeola Bamgboje-Ayodele

Background

The global need to improve access to primary care in residential aged care homes (RACHs) has driven interest in virtual care models. Despite rapid telehealth adoption, little is known about the sociotechnical factors influencing use in aged care settings, particularly from the perspectives of primary care providers, aged care staff, and residents. This review applied the Systems Engineering Initiative for Patient Safety (SEIPS) framework to synthesise evidence on barriers, enablers, processes, and outcomes of virtual care delivery in RACHs and primary care.

Methods

We conducted a systematic review in accordance with PRISMA guidelines and registered with PROSPERO (CRD42024562423). Databases searched included MEDLINE, Embase, CINAHL, and Scopus (January 2016-March 2025). Eligible studies reported qualitative, quantitative, or mixed-methods findings on virtual care involving RACHs and primary care. Data were extracted using a SEIPS-informed template and synthesised deductively across sociotechnical domains.

Findings

Thirteen studies met the inclusion criteria. Common barriers included limited digital literacy, sensory and cognitive impairments, poor audio-visual quality, lack of staff training, and workflow disruption. System-level challenges included poor technology interoperability, inadequate digital infrastructure, and insufficient organisational and policy support. Enablers included strong clinician–resident relationships, access to remote monitoring tools, and peer support. Reported outcomes were mixed: improved access, communication, and reduced emergency transfers were noted, alongside concerns about increased workload, reduced relational care, and diagnostic limitations. Studies reporting resident perspectives are lacking.

Interpretation

Virtual care has the potential to improve aged care access and outcomes, but effective implementation requires more than technology alone. Hybrid models integrating virtual with in-person care require supportive policies, funding models, and organisational workflows. Addressing interoperability gaps, infrastructure needs, and increasing co-design with residents are essential to create virtual care models that are sustainable, person-centred, and scalable in primary care and aged care contexts.

背景：全球需要改善居家老年护理之家（RACHs）获得初级保健的机会，这推动了人们对虚拟护理模式的兴趣。尽管远程医疗采用迅速，但人们对影响老年护理环境中使用远程医疗的社会技术因素知之甚少，特别是从初级保健提供者、老年护理人员和居民的角度来看。本综述应用了患者安全系统工程倡议（SEIPS）框架来综合关于乡村医院和初级保健中虚拟护理提供的障碍、推动因素、过程和结果的证据。方法：我们按照PRISMA指南进行了系统评价，并在PROSPERO注册（CRD42024562423）。检索数据库包括MEDLINE、Embase、CINAHL和Scopus（2016年1月- 2025年3月）。合格的研究报告了定性、定量或混合方法对虚拟护理的发现，包括RACHs和初级保健。使用seips知情模板提取数据，并在社会技术领域进行演绎合成。结果：13项研究符合纳入标准。常见的障碍包括数字素养有限、感觉和认知障碍、视听质量差、缺乏工作人员培训以及工作流程中断。系统级的挑战包括技术互操作性差、数字基础设施不足以及组织和政策支持不足。促成因素包括牢固的临床医师关系、远程监控工具的使用以及同伴支持。报告的结果好坏参半：注意到改善了获取、沟通和减少了紧急转移，同时担心工作量增加、关系护理减少和诊断局限性。缺乏报告居民观点的研究。解读：虚拟护理有可能改善老年护理的可及性和结果，但有效实施需要的不仅仅是技术。将虚拟护理与面对面护理相结合的混合模式需要支持性政策、资助模式和组织工作流程。解决互操作性差距、基础设施需求以及加强与居民的共同设计，对于在初级保健和老年保健环境中创建可持续、以人为本、可扩展的虚拟护理模式至关重要。

{"title":"Virtual care in residential aged care and primary care settings: a systematic literature review using the SEIPS framework","authors":"Meredith A.B. Makeham , Tamasha Jayawardena , Samiha Elkheir , Ann Carrigan , Soumya , Heather Russell , Mirela Prgomet , Vaibhav Tyagi , Amy Von Huben , Melissa Baysari , Adeola Bamgboje-Ayodele","doi":"10.1016/j.ijmedinf.2025.106218","DOIUrl":"10.1016/j.ijmedinf.2025.106218","url":null,"abstract":"<div><h3>Background</h3><div>The global need to improve access to primary care in residential aged care homes (RACHs) has driven interest in virtual care models. Despite rapid telehealth adoption, little is known about the sociotechnical factors influencing use in aged care settings, particularly from the perspectives of primary care providers, aged care staff, and residents. This review applied the Systems Engineering Initiative for Patient Safety (SEIPS) framework to synthesise evidence on barriers, enablers, processes, and outcomes of virtual care delivery in RACHs and primary care.</div></div><div><h3>Methods</h3><div>We conducted a systematic review in accordance with PRISMA guidelines and registered with PROSPERO (CRD42024562423). Databases searched included MEDLINE, Embase, CINAHL, and Scopus (January 2016-March 2025). Eligible studies reported qualitative, quantitative, or mixed-methods findings on virtual care involving RACHs and primary care. Data were extracted using a SEIPS-informed template and synthesised deductively across sociotechnical domains.</div></div><div><h3>Findings</h3><div>Thirteen studies met the inclusion criteria. Common barriers included limited digital literacy, sensory and cognitive impairments, poor audio-visual quality, lack of staff training, and workflow disruption. System-level challenges included poor technology interoperability, inadequate digital infrastructure, and insufficient organisational and policy support. Enablers included strong clinician–resident relationships, access to remote monitoring tools, and peer support. Reported outcomes were mixed: improved access, communication, and reduced emergency transfers were noted, alongside concerns about increased workload, reduced relational care, and diagnostic limitations. Studies reporting resident perspectives are lacking.</div></div><div><h3>Interpretation</h3><div>Virtual care has the potential to improve aged care access and outcomes, but effective implementation requires more than technology alone. Hybrid models integrating virtual with in-person care require supportive policies, funding models, and organisational workflows. Addressing interoperability gaps, infrastructure needs, and increasing co-design with residents are essential to create virtual care models that are sustainable, person-centred, and scalable in primary care and aged care contexts.</div></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"207 ","pages":"Article 106218"},"PeriodicalIF":4.1,"publicationDate":"2025-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145783589","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}

引用次数: 0

Effectiveness of computerized decision support systems linked to electronic health records: An updated systematic review with meta-analysis 与电子健康记录相关的计算机化决策支持系统的有效性：一项包含meta分析的最新系统综述

IF 4.1 2区医学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

International Journal of Medical Informatics

Pub Date : 2025-12-08 DOI: 10.1016/j.ijmedinf.2025.106220

Annalisa Biffi , Greta Castellini , Gabriele del Castillo , Francesca De Nard , Camilla Vismara , Federico Cabitza , Giovanni Corrao , Silvia Gianola

Background

Computerized decision support systems (CDSSs) integrated into electronic health records are intended to support continuous use of evidence in clinical decision-making, tailored to individual patients. We aimed to update a previous systematic review on the effectiveness of CDSSs linked with patient-data via electronic health records (EHRs) published in 2014.

Methods

We updated the searches on MEDLINE, Embase, and Cochrane Central Register of Controlled Trials databases from 2013 up to January 2023. We included randomized controlled trials (RCTs) that evaluated as intervention CDSSs featuring rule- or algorithm-based software integrated with EHRs and evidence-based knowledge compared with usual care, CDSSs without advice, or non-evidence-based CDSSs in any professional healthcare setting. Two independent reviewers extracted relevant data from the included RCTs and assessed the certainty of evidence using the Grading of Recommendations, Assessment, Development, and Evaluations approach. Meta-analyses with fixed- and random-effects models were performed for two primary outcomes: mortality and morbidity.

Results

We included 47 RCTs, incorporating data from 29 new RCTs in this update. Compared with controls, CDSS use may result in little to no reduction in mortality (38 trials, 127,623 patients; fixed-effects model risk ratio [RR] = 0.98; 95 % confidence interval [CI] 0.93 to 1.02; I² = 0 %; moderate certainty). The meta-analysis on morbidity reached nominal statistical significance: CDSS use may have trivial or small benefits with respect to morbidity (34 RCTs; 133,504 patients; fixed-effects model RR = 0.92, 95 % CI 0.90–0.97; random-effects model RR = 0.93, 95 % CI 0.87–0.99; I² = 48 %; high certainty). Our meta-analysis did not highlight substantial effects on mortality while tiny reductions in morbidity are possible. In specific therapeutic areas, such as cardiovascular, a small effect may be present. Nevertheless, CDSSs could improve care processes and clinician behavior, potentially influencing long-term health outcome.

Registration CRD42014007177.

计算机决策支持系统（cdss）集成到电子健康记录中，旨在支持临床决策中证据的持续使用，为个体患者量身定制。我们的目的是更新先前2014年发表的通过电子健康记录（EHRs）与患者数据相关的cdss有效性的系统综述。方法更新2013年至2023年1月在MEDLINE、Embase和Cochrane Central Register of Controlled Trials数据库中的检索结果。我们纳入了随机对照试验（RCTs），这些试验评估了干预cdss的特点，包括基于规则或算法的软件与电子病历和循证知识相结合，与常规护理相比较，无建议的cdss，或任何专业医疗机构中的非循证cdss。两名独立审稿人从纳入的随机对照试验中提取相关数据，并使用推荐、评估、发展和评价分级方法评估证据的确定性。采用固定效应和随机效应模型对两个主要结局进行meta分析：死亡率和发病率。结果我们纳入了47项随机对照试验，纳入了本次更新中29项新随机对照试验的数据。与对照组相比，使用CDSS可能导致死亡率几乎没有降低（38项试验，127,623例患者；固定效应模型风险比[RR] = 0.98； 95%可信区间[CI] 0.93至1.02;I2 = 0%；中等确定性）。发病率的荟萃分析达到名义统计学意义：CDSS的使用可能对发病率有微不足道或很小的益处（34项rct； 133,504例患者；固定效应模型RR = 0.92, 95% CI 0.90-0.97；随机效应模型RR = 0.93, 95% CI 0.87-0.99; I2 = 48%；高确定性）。我们的荟萃分析没有强调对死亡率的实质性影响，而发病率可能有微小的降低。在特定的治疗领域，如心血管，可能会出现小的影响。然而，cdss可以改善护理过程和临床医生的行为，潜在地影响长期健康结果。登记CRD42014007177。

{"title":"Effectiveness of computerized decision support systems linked to electronic health records: An updated systematic review with meta-analysis","authors":"Annalisa Biffi , Greta Castellini , Gabriele del Castillo , Francesca De Nard , Camilla Vismara , Federico Cabitza , Giovanni Corrao , Silvia Gianola","doi":"10.1016/j.ijmedinf.2025.106220","DOIUrl":"10.1016/j.ijmedinf.2025.106220","url":null,"abstract":"<div><h3>Background</h3><div>Computerized decision support systems (CDSSs) integrated into electronic health records are intended to support continuous use of evidence in clinical decision-making, tailored to individual patients. We aimed to update a previous systematic review on the effectiveness of CDSSs linked with patient-data via electronic health records (EHRs) published in 2014.</div></div><div><h3>Methods</h3><div>We updated the searches on MEDLINE, Embase, and Cochrane Central Register of Controlled Trials databases from 2013 up to January 2023. We included randomized controlled trials (RCTs) that evaluated as intervention CDSSs featuring rule- or algorithm-based software integrated with EHRs and evidence-based knowledge compared with usual care, CDSSs without advice, or non-evidence-based CDSSs in any professional healthcare setting. Two independent reviewers extracted relevant data from the included RCTs and assessed the certainty of evidence using the Grading of Recommendations, Assessment, Development, and Evaluations approach. Meta-analyses with fixed- and random-effects models were performed for two primary outcomes: mortality and morbidity.</div></div><div><h3>Results</h3><div>We included 47 RCTs, incorporating data from 29 new RCTs in this update. Compared with controls, CDSS use may result in little to no reduction in mortality (38 trials, 127,623 patients; fixed-effects model risk ratio [RR] = 0.98; 95 % confidence interval [CI] 0.93 to 1.02; I<sup>2</sup> = 0 %; moderate certainty). The <em>meta</em>-analysis on morbidity reached nominal statistical significance: CDSS use may have trivial or small benefits with respect to morbidity (34 RCTs; 133,504 patients; fixed-effects model RR = 0.92, 95 % CI 0.90–0.97; random-effects model RR = 0.93, 95 % CI 0.87–0.99; I<sup>2</sup> = 48 %; high certainty). Our <em>meta</em>-analysis did not highlight substantial effects on mortality while tiny reductions in morbidity are possible. In specific therapeutic areas, such as cardiovascular, a small effect may be present. Nevertheless, CDSSs could improve care processes and clinician behavior, potentially influencing long-term health outcome.</div><div><strong>Registration</strong> CRD42014007177.</div></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"207 ","pages":"Article 106220"},"PeriodicalIF":4.1,"publicationDate":"2025-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145736655","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}

引用次数: 0

The impact of digital self-management programmes on stroke survivors: a systematic review of randomised controlled trials 数字自我管理程序对中风幸存者的影响：随机对照试验的系统回顾。

IF 4.1 2区医学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

International Journal of Medical Informatics

Pub Date : 2025-12-07 DOI: 10.1016/j.ijmedinf.2025.106210

Weiwei Guo , Kim Lam Soh , Kim Geok Soh , Hasni Idayu Saidi

Purpose

To synthesise evidence on the effectiveness of digital self-management programmes for stroke survivors’ health outcomes, self-efficacy, and quality of life.

Methods

Relevant English-language studies published between 2015 and 2025 were retrieved from PubMed, Embase, the Cochrane Library, Web of Science and Scopus databases. This systematic review was conducted in accordance with the PRISMA guidelines and registered with PROSPERO (CRD420251059348). Studies were included if they investigated digital self-management interventions for adult stroke survivors, with outcome measures including secondary prevention, self-efficacy, self-management ability, and quality of life. Study quality was assessed using the Cochrane Risk of Bias tool. Owing to substantial heterogeneity across digital platforms, intervention duration, and outcome measurement tools, a descriptive synthesis approach was adopted.

Results

A total of 12 randomised controlled trials (RCTs) involving 3,049 participants were included. Among these, all three studies assessing self-efficacy reported significant improvements in stroke survivors (p < 0.05), two out of three studies demonstrated enhanced self-management ability (p < 0.05), all six studies evaluating quality of life showed significant positive effects (p < 0.05), and all six studies assessing medication adherence reported improvement. However, effects on secondary prevention behaviours such as smoking, alcohol use, physical activity, and blood pressure control were inconsistent. Few studies assessed motor function or long-term outcomes. Intervention content, delivery platforms, and intensity varied widely.

Conclusion

Digital self-management via technology shows positive impacts on self-efficacy, medication adherence, and quality of life in stroke survivors. The impact on motor rehabilitation remains unclear, indicating a need for further research. Digital self-management can enhance stroke survivors’ self-efficacy and self-management abilities, promoting active rehabilitation. This intervention effectively improves medication adherence and quality of life but has limited impact on behaviour changes such as smoking cessation and alcohol reduction. It is important to consider integrating digital tools with conventional care while addressing patients’ digital literacy and accessibility challenges. Further development and research are needed to evaluate the effects of digital self-management on stroke functional recovery and activity capacity.

目的：综合有关数字自我管理方案对中风幸存者健康结局、自我效能和生活质量有效性的证据。方法：从PubMed、Embase、Cochrane Library、Web of Science和Scopus数据库中检索2015 - 2025年间发表的相关英文研究。本系统评价按照PRISMA指南进行，并在普洛斯彼罗注册（CRD420251059348）。如果研究调查了成年中风幸存者的数字化自我管理干预措施，结果测量包括二级预防、自我效能、自我管理能力和生活质量，则纳入研究。使用Cochrane偏倚风险工具评估研究质量。由于数字平台、干预持续时间和结果测量工具之间存在实质性异质性，因此采用了描述性综合方法。结果：共纳入12项随机对照试验（RCTs），涉及3049名受试者。其中，所有三项评估自我效能的研究都报告了卒中幸存者的显著改善(p结论：通过技术进行数字化自我管理对卒中幸存者的自我效能、药物依从性和生活质量有积极影响。对运动康复的影响尚不清楚，表明需要进一步研究。数字化自我管理可以提高脑卒中幸存者的自我效能感和自我管理能力，促进积极康复。这种干预有效地改善了药物依从性和生活质量，但对戒烟和减少酒精等行为改变的影响有限。重要的是要考虑将数字工具与传统护理相结合，同时解决患者的数字素养和可及性挑战。评估数字化自我管理对脑卒中功能恢复和活动能力的影响需要进一步的开发和研究。

{"title":"The impact of digital self-management programmes on stroke survivors: a systematic review of randomised controlled trials","authors":"Weiwei Guo , Kim Lam Soh , Kim Geok Soh , Hasni Idayu Saidi","doi":"10.1016/j.ijmedinf.2025.106210","DOIUrl":"10.1016/j.ijmedinf.2025.106210","url":null,"abstract":"<div><h3>Purpose</h3><div>To synthesise evidence on the effectiveness of digital self-management programmes for stroke survivors’ health outcomes, self-efficacy, and quality of life.</div></div><div><h3>Methods</h3><div>Relevant English-language studies published between 2015 and 2025 were retrieved from PubMed, Embase, the Cochrane Library, Web of Science and Scopus databases. This systematic review was conducted in accordance with the PRISMA guidelines and registered with PROSPERO (CRD420251059348).Studies were included if they investigated digital self-management interventions for adult stroke survivors, with outcome measures including secondary prevention, self-efficacy, self-management ability, and quality of life. Study quality was assessed using the Cochrane Risk of Bias tool. Owing to substantial heterogeneity across digital platforms, intervention duration, and outcome measurement tools, a descriptive synthesis approach was adopted.</div></div><div><h3>Results</h3><div>A total of 12 randomised controlled trials (RCTs) involving 3,049 participants were included. Among these, all three studies assessing self-efficacy reported significant improvements in stroke survivors (<em>p</em> < 0.05), two out of three studies demonstrated enhanced self-management ability (<em>p</em> < 0.05), all six studies evaluating quality of life showed significant positive effects (<em>p</em> < 0.05), and all six studies assessing medication adherence reported improvement. However, effects on secondary prevention behaviours such as smoking, alcohol use, physical activity, and blood pressure control were inconsistent. Few studies assessed motor function or long-term outcomes. Intervention content, delivery platforms, and intensity varied widely.</div></div><div><h3>Conclusion</h3><div>Digital self-management via technology shows positive impacts on self-efficacy, medication adherence, and quality of life in stroke survivors. The impact on motor rehabilitation remains unclear, indicating a need for further research. Digital self-management can enhance stroke survivors’ self-efficacy and self-management abilities, promoting active rehabilitation. This intervention effectively improves medication adherence and quality of life but has limited impact on behaviour changes such as smoking cessation and alcohol reduction. It is important to consider integrating digital tools with conventional care while addressing patients’ digital literacy and accessibility challenges. Further development and research are needed to evaluate the effects of digital self-management on stroke functional recovery and activity capacity.</div></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"207 ","pages":"Article 106210"},"PeriodicalIF":4.1,"publicationDate":"2025-12-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145696318","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}

引用次数: 0