International Journal of Medical Informatics最新文献_第2页

Patient engagement and performance expectancy towards epilepsy digital health interventions: systematic literature review and meta-analysis 患者对癫痫数字健康干预的参与和表现预期：系统文献综述和荟萃分析

IF 4.1 2区医学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

International Journal of Medical Informatics

Pub Date : 2026-01-30 DOI: 10.1016/j.ijmedinf.2026.106306

Tolesa Fanta Jilcha , Peter Richard Christopher Leeson , Khin Than Win

Background

Digital Health is currently showing promising results in reducing patient and caregiver suffering that arise from misconceptions.

Objective

To synthesize existing evidence on Perceived Usefulness, interest in use and willingness to use towards Epilepsy Digital Health Interventions.

Method

Databases were searched for studies reporting on the outcomes of interest by using a comprehensive search strategy. Studies published in English from January 2015 to September 2025 were included. The Newcastle-Ottawa Quality Assessment Scale was employed to evaluate the quality of included studies. Stata version 19 was used to compute a pooled proportion using a random-effects model. Heterogeneity was assessed using the Cochrane chi-square and the index of heterogeneity test. Sensitivity tests and subgroup analyses were performed. Publication bias was examined by funnel plots and Egger’s test.

Result

Overall, 6041 studies were found from databases. After a step-by-step screening, 23 studies were included in this review. The total number of participants was 6703 with a sample size ranges from 12 to 1168. The pooled proportions of Perceived Usefulness, interest to use, and willingness to use Digital Health were 0.66 (0.58, 0.75), 0.69 (0.50, 0.88), and 0.75 (0.66, 0.83), respectively. In this review, Sensitivity tests indicated that none of the included studies exerted extreme influence on the pooled prevalence; and Funnel plots and Egger’s test (p ≤ 0.772) showed no evidence of publication bias.

Conclusion

In this review, 66% of respondents perceive Digital Health as useful; 69% were interested in using Digital Health, and 75% were willing to engage with Digital Health. Most of the studies were from high-income countries, with no studies found from developing countries. This review emphasizes the importance of focusing on the user’s perceptions, their interest and willingness to use Digital Health Interventions. It also stresses the need for further studies in low-income countries.

背景数字健康目前在减少因误解引起的患者和护理人员痛苦方面显示出有希望的结果。目的综合现有的癫痫数字健康干预措施的感知有用性、使用兴趣和使用意愿的证据。方法采用综合检索策略在数据库中检索有关相关结果的研究报告。纳入了2015年1月至2025年9月以英文发表的研究。采用纽卡斯尔-渥太华质量评定量表评价纳入研究的质量。使用Stata version 19使用随机效应模型计算合并比例。采用Cochrane卡方检验和异质性指数检验评估异质性。进行敏感性试验和亚组分析。发表偏倚采用漏斗图和Egger检验。结果共从数据库中检索到6041项研究。经过逐步筛选，本综述纳入了23项研究。参与者总数为6703人，样本量从12到1168人不等。感知有用性、使用兴趣和使用数字健康意愿的总比例分别为0.66（0.58,0.75）、0.69（0.50,0.88）和0.75（0.66,0.83）。在本综述中，敏感性试验表明，纳入的研究均未对总患病率产生极端影响；漏斗图和Egger检验（p≤0.772）均未发现发表偏倚的证据。在本次审查中，66%的受访者认为数字医疗是有用的；69%的人对使用数字医疗感兴趣，75%的人愿意参与数字医疗。大多数研究来自高收入国家，没有发现来自发展中国家的研究。这篇综述强调了关注用户的认知、他们使用数字健康干预措施的兴趣和意愿的重要性。报告还强调需要在低收入国家进行进一步研究。

{"title":"Patient engagement and performance expectancy towards epilepsy digital health interventions: systematic literature review and meta-analysis","authors":"Tolesa Fanta Jilcha , Peter Richard Christopher Leeson , Khin Than Win","doi":"10.1016/j.ijmedinf.2026.106306","DOIUrl":"10.1016/j.ijmedinf.2026.106306","url":null,"abstract":"<div><h3>Background</h3><div>Digital Health is currently showing promising results in reducing patient and caregiver suffering that arise from misconceptions.</div></div><div><h3>Objective</h3><div>To synthesize existing evidence on Perceived Usefulness, interest in use and willingness to use towards Epilepsy Digital Health Interventions.</div></div><div><h3>Method</h3><div>Databases were searched for studies reporting on the outcomes of interest by using a comprehensive search strategy. Studies published in English from January 2015 to September 2025 were included. The Newcastle-Ottawa Quality Assessment Scale was employed to evaluate the quality of included studies. Stata version 19 was used to compute a pooled proportion using a random-effects model. Heterogeneity was assessed using the Cochrane chi-square and the index of heterogeneity test. Sensitivity tests and subgroup analyses were performed. Publication bias was examined by funnel plots and Egger’s test.</div></div><div><h3>Result</h3><div>Overall, 6041 studies were found from databases. After a step-by-step screening, 23 studies were included in this review. The total number of participants was 6703 with a sample size ranges from 12 to 1168. The pooled proportions of Perceived Usefulness, interest to use, and willingness to use Digital Health were 0.66 (0.58, 0.75), 0.69 (0.50, 0.88), and 0.75 (0.66, 0.83), respectively. In this review, Sensitivity tests indicated that none of the included studies exerted extreme influence on the pooled prevalence; and Funnel plots and Egger’s test (p ≤ 0.772) showed no evidence of publication bias.</div></div><div><h3>Conclusion</h3><div>In this review<strong>,</strong> 66% of respondents perceive Digital Health as useful; 69% were interested in using Digital Health, and 75% were willing to engage with Digital Health. Most of the studies were from high-income countries, with no studies found from developing countries. This review emphasizes the importance of focusing on the user’s perceptions, their interest and willingness to use Digital Health Interventions. It also stresses the need for further studies in low-income countries.</div></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"211 ","pages":"Article 106306"},"PeriodicalIF":4.1,"publicationDate":"2026-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146081744","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

Bridging performance and uncertainty: Cautionary notes on machine learning and large language models in TBI prognostication 桥接性能和不确定性：机器学习和大型语言模型在TBI预测中的警示

IF 4.1 2区医学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

International Journal of Medical Informatics

Pub Date : 2026-01-28 DOI: 10.1016/j.ijmedinf.2026.106315

Hasan Nawaz Tahir , Anfal Khan , Muhammad Yousaf , Shahnila Javed , Mursala Tahir , Yousaf Ali

引用次数: 0

Smart insurance analytics: A novel ensemble feature selection approach to unlock health insurance coverage predictions in Sierra Leone. 智能保险分析：一种新颖的集成特征选择方法来解锁塞拉利昂的健康保险覆盖预测。

IF 4.1 2区医学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

International Journal of Medical Informatics

Pub Date : 2026-01-27 DOI: 10.1016/j.ijmedinf.2026.106313

David B Olawade, Augustus Osborne, Afeez A Soladoye, Olaitan E Oluwadare, Emmanuel O Awogbindin, Ojima Z Wada

Background: Predicting health insurance uptake remains a critical challenge for policymakers and insurance providers seeking to optimise coverage strategies and resource allocation. In Sierra Leone, health insurance uptake remains extremely low, and understanding determinants is vital for universal health coverage goals.

Objective: To develop and evaluate an innovative ensemble feature selection methodology for health insurance uptake prediction, establishing new performance benchmarks through systematic comparison of multiple machine learning algorithms using comprehensive validation strategies.

Methods: This study employed supervised machine learning to predict health insurance uptake among 15,574 women using data from the 2019 Sierra Leone Demographic and Health Survey (SLDHS). We implemented an ensemble feature selection approach that requires consensus across Adaptive Ant Colony Optimisation, Recursive Feature Elimination, and Backwards Elimination techniques. Seven algorithms were systematically compared: Logistic Regression, Support Vector Machines, K-Nearest Neighbors, Random Forest, Gradient Boosting, XGBoost, and LightGBM. SMOTE addressed class imbalance, whilst validation employed nested 5-fold cross-validation, 10-fold cross-validation, and hold-out testing to prevent information leakage.

Results: Random Forest achieved exceptional performance with 0.9973 accuracy, 0.9973 precision, 0.9973 recall, 0.9973 F1-score, and perfect 1.0000 ROC AUC on hold-out testing. XGBoost delivered comparable results with 0.9914 across all metrics and 0.9998 ROC AUC. Backward Feature Elimination consistently yielded superior results across ensemble methods. However, the near-perfect performance warrants cautious interpretation and requires external validation to confirm generalizability.

Conclusions: This research establishes new performance benchmarks for health insurance prediction, significantly exceeding existing literature, which has direct implications for health insurance policy and practice in Sierra Leone. The innovative ensemble feature selection methodology provides a robust framework for enhancing prediction accuracy across healthcare applications, offering immediate practical value for stakeholders. Future work should prioritize external validation, explainability analysis, and temporal stability assessment to ensure practical deployment readiness.

背景：预测健康保险的吸收仍然是政策制定者和保险提供者寻求优化覆盖策略和资源分配的关键挑战。在塞拉利昂，健康保险的接受程度仍然极低，了解决定因素对于实现全民健康覆盖目标至关重要。目的：开发和评估一种用于健康保险摄取预测的创新集成特征选择方法，通过使用综合验证策略对多种机器学习算法进行系统比较，建立新的性能基准。方法：本研究利用2019年塞拉利昂人口与健康调查（SLDHS）的数据，采用监督式机器学习来预测15574名女性的医疗保险吸收情况。我们实现了一种集成特征选择方法，该方法需要在自适应蚁群优化、递归特征消除和向后消除技术之间达成共识。系统地比较了七种算法：逻辑回归、支持向量机、k近邻、随机森林、梯度增强、XGBoost和LightGBM。SMOTE解决了类不平衡问题，而验证采用嵌套的5次交叉验证、10次交叉验证和保留测试来防止信息泄漏。结果：Random Forest在hold-out测试中取得了0.9973的准确率、0.9973的精密度、0.9973的召回率、0.9973的f1得分和完美的1.000 ROC AUC的优异表现。XGBoost在所有指标上提供了0.9914和0.9998 ROC AUC的可比结果。在集成方法中，向后特征消除始终产生优越的结果。然而，近乎完美的性能需要谨慎的解释，并需要外部验证来确认普遍性。结论：本研究为健康保险预测建立了新的绩效基准，显著超过现有文献，这对塞拉利昂的健康保险政策和实践具有直接影响。创新的集成特征选择方法为提高医疗保健应用程序的预测准确性提供了一个强大的框架，为利益相关者提供了直接的实用价值。未来的工作应该优先考虑外部验证、可解释性分析和时间稳定性评估，以确保实际部署就绪。

{"title":"Smart insurance analytics: A novel ensemble feature selection approach to unlock health insurance coverage predictions in Sierra Leone.","authors":"David B Olawade, Augustus Osborne, Afeez A Soladoye, Olaitan E Oluwadare, Emmanuel O Awogbindin, Ojima Z Wada","doi":"10.1016/j.ijmedinf.2026.106313","DOIUrl":"https://doi.org/10.1016/j.ijmedinf.2026.106313","url":null,"abstract":"<p><strong>Background: </strong>Predicting health insurance uptake remains a critical challenge for policymakers and insurance providers seeking to optimise coverage strategies and resource allocation. In Sierra Leone, health insurance uptake remains extremely low, and understanding determinants is vital for universal health coverage goals.</p><p><strong>Objective: </strong>To develop and evaluate an innovative ensemble feature selection methodology for health insurance uptake prediction, establishing new performance benchmarks through systematic comparison of multiple machine learning algorithms using comprehensive validation strategies.</p><p><strong>Methods: </strong>This study employed supervised machine learning to predict health insurance uptake among 15,574 women using data from the 2019 Sierra Leone Demographic and Health Survey (SLDHS). We implemented an ensemble feature selection approach that requires consensus across Adaptive Ant Colony Optimisation, Recursive Feature Elimination, and Backwards Elimination techniques. Seven algorithms were systematically compared: Logistic Regression, Support Vector Machines, K-Nearest Neighbors, Random Forest, Gradient Boosting, XGBoost, and LightGBM. SMOTE addressed class imbalance, whilst validation employed nested 5-fold cross-validation, 10-fold cross-validation, and hold-out testing to prevent information leakage.</p><p><strong>Results: </strong>Random Forest achieved exceptional performance with 0.9973 accuracy, 0.9973 precision, 0.9973 recall, 0.9973 F1-score, and perfect 1.0000 ROC AUC on hold-out testing. XGBoost delivered comparable results with 0.9914 across all metrics and 0.9998 ROC AUC. Backward Feature Elimination consistently yielded superior results across ensemble methods. However, the near-perfect performance warrants cautious interpretation and requires external validation to confirm generalizability.</p><p><strong>Conclusions: </strong>This research establishes new performance benchmarks for health insurance prediction, significantly exceeding existing literature, which has direct implications for health insurance policy and practice in Sierra Leone. The innovative ensemble feature selection methodology provides a robust framework for enhancing prediction accuracy across healthcare applications, offering immediate practical value for stakeholders. Future work should prioritize external validation, explainability analysis, and temporal stability assessment to ensure practical deployment readiness.</p>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"211 ","pages":"106313"},"PeriodicalIF":4.1,"publicationDate":"2026-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146115071","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

Machine learning-based prediction of three-year mortality in elderly inpatients with coronary artery disease combined with heart failure 基于机器学习的老年冠心病合并心力衰竭住院患者三年死亡率预测

IF 4.1 2区医学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

International Journal of Medical Informatics

Pub Date : 2026-01-25 DOI: 10.1016/j.ijmedinf.2026.106307

Shihui Fu , Zilei Zhao , Xuhui Liu , Jianfeng Guo , Kai Wang , Yali Zhao , Shan Nan , Qianshuo Liu , Yan Nie , Jinwen Tian

Objective

Accurate prediction of survival outcome is essential for early intervention and treatment optimization. This study aimed to develop a model utilizing machine learning techniques for predicting three-year mortality in elderly inpatients with coronary artery disease (CAD) combined with heart failure (HF).

Methods

This study enrolled 987 elderly inpatients with CAD. This cohort was randomly divided into the training and validation datasets in a 7:3 ratio. Five machine learning methods, including Logistic Regression, Random Forest, Support Vector Machine, eXtreme Gradient Boosting, and Gradient Boosting Decision Trees, were implemented to construct predictive models.

Results

Overall, the median age of this cohort was 85 [81,89] years. Three-year mortality in elderly inpatients with CAD combined with HF was 56.46%. The least absolute shrinkage and selection operator method and five-fold cross-validation identified that ten features were significantly associated with three-year mortality. Logistic Regression showed better performance than other models in the Brier Score, Area Under The Curve, Accuracy, Precision, Recall, and F1 Score of 0.1105, 0.9014, 0.8764, 0.9167, 0.8627, and 0.8889, respectively. The Shapley Additive exPlanations method revealed that age, interventricular septum thickness, gamma gap, serum creatinine, N-terminal pro-B-type natriuretic peptide (NT.proBNP), and neutrophil-to-lymphocyte ratio were identified as risk factors, and mean systolic blood pressure, hemoglobin, albumin, and sodium were protective factors. Age, albumin, and NT.proBNP were three features most associated with three-year mortality. The network application could be available at https://cad-hf-predict.tracebook.org.cn.

Conclusion

Logistic Regression exhibits excellent predictive performance for predicting three-year mortality in elderly inpatients with CAD combined with HF.

目的准确预测生存预后对早期干预和优化治疗至关重要。本研究旨在开发一个利用机器学习技术预测老年住院冠心病（CAD）合并心力衰竭（HF）患者三年死亡率的模型。方法本研究纳入987例老年冠心病住院患者。该队列按7:3的比例随机分为训练数据集和验证数据集。采用逻辑回归、随机森林、支持向量机、极端梯度增强和梯度增强决策树五种机器学习方法构建预测模型。结果总体而言，该队列的中位年龄为85岁[81,89]。老年冠心病合并心衰住院患者3年死亡率为56.46%。最小绝对收缩和选择算子方法以及五倍交叉验证确定了10个特征与三年死亡率显着相关。Logistic回归在Brier Score、Area Under the Curve、Accuracy、Precision、Recall和F1 Score分别为0.1105、0.9014、0.8764、0.9167、0.8627和0.8889方面均优于其他模型。Shapley加性解释法显示，年龄、室间隔厚度、γ间隙、血清肌酐、n -末端前b型利钠肽（NT.proBNP）和中性粒细胞与淋巴细胞比值是危险因素，平均收缩压、血红蛋白、白蛋白和钠是保护因素。年龄、白蛋白和NT.proBNP是与三年死亡率最相关的三个特征。该网络应用程序可在https://cad-hf-predict.tracebook.org.cn.ConclusionLogistic上获得，回归在预测老年CAD合并心衰住院患者的三年死亡率方面表现出出色的预测性能。

{"title":"Machine learning-based prediction of three-year mortality in elderly inpatients with coronary artery disease combined with heart failure","authors":"Shihui Fu , Zilei Zhao , Xuhui Liu , Jianfeng Guo , Kai Wang , Yali Zhao , Shan Nan , Qianshuo Liu , Yan Nie , Jinwen Tian","doi":"10.1016/j.ijmedinf.2026.106307","DOIUrl":"10.1016/j.ijmedinf.2026.106307","url":null,"abstract":"<div><h3>Objective</h3><div>Accurate prediction of survival outcome is essential for early intervention and treatment optimization. This study aimed to develop a model utilizing machine learning techniques for predicting three-year mortality in elderly inpatients with coronary artery disease (CAD) combined with heart failure (HF).</div></div><div><h3>Methods</h3><div>This study enrolled 987 elderly inpatients with CAD. This cohort was randomly divided into the training and validation datasets in a 7:3 ratio. Five machine learning methods, including Logistic Regression, Random Forest, Support Vector Machine, eXtreme Gradient Boosting, and Gradient Boosting Decision Trees, were implemented to construct predictive models.</div></div><div><h3>Results</h3><div>Overall, the median age of this cohort was 85 [81,89] years. Three-year mortality in elderly inpatients with CAD combined with HF was 56.46%. The least absolute shrinkage and selection operator method and five-fold cross-validation identified that ten features were significantly associated with three-year mortality. Logistic Regression showed better performance than other models in the Brier Score, Area Under The Curve, Accuracy, Precision, Recall, and F1 Score of 0.1105, 0.9014, 0.8764, 0.9167, 0.8627, and 0.8889, respectively. The Shapley Additive exPlanations method revealed that age, interventricular septum thickness, gamma gap, serum creatinine, N-terminal pro-B-type natriuretic peptide (NT.proBNP), and neutrophil-to-lymphocyte ratio were identified as risk factors, and mean systolic blood pressure, hemoglobin, albumin, and sodium were protective factors. Age, albumin, and NT.proBNP were three features most associated with three-year mortality. The network application could be available at <span><span>https://cad-hf-predict.tracebook.org.cn</span><svg><path></path></svg></span>.</div></div><div><h3>Conclusion</h3><div>Logistic Regression exhibits excellent predictive performance for predicting three-year mortality in elderly inpatients with CAD combined with HF.</div></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"210 ","pages":"Article 106307"},"PeriodicalIF":4.1,"publicationDate":"2026-01-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146080697","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

Rural and urban patient perceptions of electronic health record data use in research and clinical care: A cross-sectional survey research study 农村和城市患者对研究和临床护理中电子病历数据使用的看法：一项横断面调查研究

IF 4.1 2区医学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

International Journal of Medical Informatics

Pub Date : 2026-01-24 DOI: 10.1016/j.ijmedinf.2026.106310

Melissa L. Harry , Morgan L. Brenholdt , Anthony W. Olson , Claudia C. Ramjattan , Megan A. Schwalbe

Purpose

Understand patient perceptions of sharing and use of electronic health record (EHR) data in research and clinical care, including whether differences exist between rural and urban patients.

Methods

We adapted survey items from existing surveys and developed some new items, making revisions following cognitive interviews with eight patients in a rural-serving health system. We then invited 18,251 health system patients to take the electronic survey (7/11/2023–10/13/2023, 10/31/2023–01/02/2024). Analyses included bivariate statistics and multivariable ordered and binary logistic regression examining associations between participant responses and rurality using respondent zip code and associated 2020 Rural-Urban Commuting Area code (rural: 4–10; urban: 1–3). We analyzed open-ended survey questions with qualitative content analysis.

Findings

Of 1,929 participants who started the survey (10.6% response rate), 1,912 completed questions beyond demographics and were included in the analytical sample. Most respondents were female (66.9%), White (93.4%), employed for wages (45.1%) or retired (37.2%), had at least some college (88.3%), and lived in urban locales (55.0%). Rural respondents had significantly lower medical mistrust levels than urban. Comfort with sharing data for research was high amongst respondents, particularly when de-identified. Some differences were seen between rural and urban respondents in adjusted models, foremost being rural respondents having higher adjusted odds (aOR = 1.43, 95% CI = 1.16–1.77, p = 0.001) of being more comfortable sharing data if their zip code was removed. Rural respondents had significantly higher odds of being comfortable with some demographic data being in the EHR and accessible to health system providers and researchers compared to urban respondents.

Conclusions

Respondents generally supported sharing health data for research and care purposes. Although zip code is frequently used to demarcate rurality in U.S.-based research, rural respondents may be more comfortable sharing data when zip code is removed. Opportunities to assuage concerns regarding data use are also described.

目的了解患者对在研究和临床护理中共享和使用电子健康记录（EHR）数据的看法，包括农村和城市患者之间是否存在差异。方法根据对8名农村卫生系统患者的认知访谈，对现有调查项目进行改编，并开发了一些新的调查项目。然后，我们邀请了18251名卫生系统患者进行电子调查（2023年7月11日- 2023年10月13日，2023年10月31日- 2024年2月1日）。分析包括双变量统计和多变量有序和二元逻辑回归，使用受访者的邮政编码和相关的2020年城乡通勤区代码（农村：4-10；城市：1-3）来检验参与者的回答与乡村性之间的关联。我们对开放式调查问题进行定性内容分析。在开始调查的1,929名参与者（10.6%的回复率）中，1,912名完成了人口统计学以外的问题，并被纳入分析样本。大多数受访者是女性（66.9%），白人（93.4%），有工资工作（45.1%）或退休（37.2%），至少有一些大学（88.3%），居住在城市（55.0%）。农村受访者对医疗的不信任程度明显低于城市受访者。受访者对分享研究数据的满意度很高，尤其是在去识别的情况下。在调整后的模型中，农村和城市受访者之间存在一些差异，最重要的是农村受访者在删除其邮政编码后更愿意分享数据的调整几率更高（aOR = 1.43, 95% CI = 1.16-1.77, p = 0.001）。与城市受访者相比，农村受访者对电子病历中的一些人口统计数据感到满意，并且卫生系统提供者和研究人员可以获得这些数据的可能性要高得多。结论受访者普遍支持出于研究和护理目的共享健康数据。虽然在美国的研究中，邮政编码经常被用来划分农村地区，但当邮政编码被删除时，农村受访者可能更愿意分享数据。还描述了缓解对数据使用的担忧的机会。

{"title":"Rural and urban patient perceptions of electronic health record data use in research and clinical care: A cross-sectional survey research study","authors":"Melissa L. Harry , Morgan L. Brenholdt , Anthony W. Olson , Claudia C. Ramjattan , Megan A. Schwalbe","doi":"10.1016/j.ijmedinf.2026.106310","DOIUrl":"10.1016/j.ijmedinf.2026.106310","url":null,"abstract":"<div><h3>Purpose</h3><div>Understand patient perceptions of sharing and use of electronic health record (EHR) data in research and clinical care, including whether differences exist between rural and urban patients.</div></div><div><h3>Methods</h3><div>We adapted survey items from existing surveys and developed some new items, making revisions following cognitive interviews with eight patients in a rural-serving health system. We then invited 18,251 health system patients to take the electronic survey (7/11/2023–10/13/2023, 10/31/2023–01/02/2024). Analyses included bivariate statistics and multivariable ordered and binary logistic regression examining associations between participant responses and rurality using respondent zip code and associated 2020 Rural-Urban Commuting Area code (rural: 4–10; urban: 1–3). We analyzed open-ended survey questions with qualitative content analysis.</div></div><div><h3>Findings</h3><div>Of 1,929 participants who started the survey (10.6% response rate), 1,912 completed questions beyond demographics and were included in the analytical sample. Most respondents were female (66.9%), White (93.4%), employed for wages (45.1%) or retired (37.2%), had at least some college (88.3%), and lived in urban locales (55.0%). Rural respondents had significantly lower medical mistrust levels than urban. Comfort with sharing data for research was high amongst respondents, particularly when de-identified. Some differences were seen between rural and urban respondents in adjusted models, foremost being rural respondents having higher adjusted odds (aOR = 1.43, 95% CI = 1.16–1.77, <em>p</em> = 0.001) of being more comfortable sharing data if their zip code was removed. Rural respondents had significantly higher odds of being comfortable with some demographic data being in the EHR and accessible to health system providers and researchers compared to urban respondents.</div></div><div><h3>Conclusions</h3><div>Respondents generally supported sharing health data for research and care purposes. Although zip code is frequently used to demarcate rurality in U.S.-based research, rural respondents may be more comfortable sharing data when zip code is removed. Opportunities to assuage concerns regarding data use are also described.</div></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"210 ","pages":"Article 106310"},"PeriodicalIF":4.1,"publicationDate":"2026-01-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146081153","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

Reply to comment on “medication-based mortality prediction in COPD using machine learning and conventional statistical methods” 回复关于“利用机器学习和传统统计方法预测COPD药物死亡率”的评论

IF 4.1 2区医学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

International Journal of Medical Informatics

Pub Date : 2026-01-24 DOI: 10.1016/j.ijmedinf.2026.106304

Ana Paula Bruno Pena-Gralle , Amélie Forget , Yohann Moanahere Chiu , Marc-André Legault , Marie-France Beauchesne , Lucie Blais

引用次数: 0

Development and temporal validation of machine learning models for predicting clinically relevant medication reconciliation discrepancies at the emergency department: A single-center retrospective study. 用于预测急诊科临床相关药物调和差异的机器学习模型的开发和时间验证：一项单中心回顾性研究

IF 4.1 2区医学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

International Journal of Medical Informatics

Pub Date : 2026-01-24 DOI: 10.1016/j.ijmedinf.2026.106309

Greet Van De Sijpe, Tuur Schrooten, Sabrina De Winter, Lorenz Van der Linden, Peter Vanbrabant, Annabel Dompas, Bo Bertels, Maarten De Vos, Isabel Spriet

Objective: Medication discrepancies at hospital admission are common and can cause preventable patient harm. Predictive models can help prioritize medication reconciliation for high-risk patients. This study aimed to develop and validate machine learning (ML) models for predicting clinically relevant medication reconciliation discrepancies in emergency department (ED) patients, and to compare their performance with logistic regression.

Methods: We conducted a single-center, retrospective study at UZ Leuven. The dataset included patients admitted to the ED between 2017 and 2019 (development set) and 2021-2022 (temporal validation set). The outcome variable was the presence of at least one clinically relevant medication discrepancy, defined by expert panel adjudication. Variables were extracted from the electronic health record, with care to avoid data leakage. Three models - logistic regression, random forest, and eXtreme Gradient Boosting - were developed using tailored variable selection strategies, and validated temporally. Model performance was assessed via discrimination, calibration, and classification metrics. Clinical utility was assessed using decision curve analysis.

Results: The development and validation cohorts included 817 and 349 patients, respectively. LR and RF models demonstrated moderate discrimination on temporal validation (AUROC 0.67-0.68). The XGBoost model showed lower discrimination (AUROC 0.63). Calibration was comparable across models. Decision curve analysis showed only small differences in net benefit between models across clinically relevant threshold probabilities.

Conclusion: ML models provided no clear improvement over logistic regression, which achieved similar predictive performance and greater interpretability. These findings highlight both the potential and the limitations of ML for supporting targeted medication reconciliation in ED workflows. Future research should explore the added value of richer data sources, such as unstructured clinical narratives.

目的：住院用药不一致是常见的，可造成可预防的患者伤害。预测模型可以帮助对高危患者的药物调节进行优先排序。本研究旨在开发和验证机器学习（ML）模型，用于预测急诊科（ED）患者的临床相关药物调和差异，并将其性能与逻辑回归进行比较。方法：我们在鲁汶大学进行了一项单中心回顾性研究。该数据集包括2017年至2019年（发展集）和2021年至2022年（时间验证集）期间入住急诊科的患者。结果变量是存在至少一种临床相关的药物差异，由专家小组裁决确定。从电子健康记录中提取变量，小心避免数据泄露。使用量身定制的变量选择策略开发了三个模型-逻辑回归，随机森林和极端梯度增强，并进行了时间验证。通过判别、校准和分类指标评估模型性能。采用决策曲线分析评估临床效用。结果：开发和验证队列分别包括817例和349例患者。LR和RF模型在时间验证上表现出中度差异（AUROC为0.67-0.68）。XGBoost模型具有较低的判别性（AUROC为0.63）。各模型的校准具有可比性。决策曲线分析显示，在临床相关阈值概率的模型之间，净收益只有很小的差异。结论：与逻辑回归相比，ML模型没有明显的改进，预测性能相似，可解释性更强。这些发现强调了ML在支持ED工作流程中靶向药物调节方面的潜力和局限性。未来的研究应探索更丰富的数据来源的附加价值，如非结构化的临床叙述。

{"title":"Development and temporal validation of machine learning models for predicting clinically relevant medication reconciliation discrepancies at the emergency department: A single-center retrospective study.","authors":"Greet Van De Sijpe, Tuur Schrooten, Sabrina De Winter, Lorenz Van der Linden, Peter Vanbrabant, Annabel Dompas, Bo Bertels, Maarten De Vos, Isabel Spriet","doi":"10.1016/j.ijmedinf.2026.106309","DOIUrl":"https://doi.org/10.1016/j.ijmedinf.2026.106309","url":null,"abstract":"<p><strong>Objective: </strong>Medication discrepancies at hospital admission are common and can cause preventable patient harm. Predictive models can help prioritize medication reconciliation for high-risk patients. This study aimed to develop and validate machine learning (ML) models for predicting clinically relevant medication reconciliation discrepancies in emergency department (ED) patients, and to compare their performance with logistic regression.</p><p><strong>Methods: </strong>We conducted a single-center, retrospective study at UZ Leuven. The dataset included patients admitted to the ED between 2017 and 2019 (development set) and 2021-2022 (temporal validation set). The outcome variable was the presence of at least one clinically relevant medication discrepancy, defined by expert panel adjudication. Variables were extracted from the electronic health record, with care to avoid data leakage. Three models - logistic regression, random forest, and eXtreme Gradient Boosting - were developed using tailored variable selection strategies, and validated temporally. Model performance was assessed via discrimination, calibration, and classification metrics. Clinical utility was assessed using decision curve analysis.</p><p><strong>Results: </strong>The development and validation cohorts included 817 and 349 patients, respectively. LR and RF models demonstrated moderate discrimination on temporal validation (AUROC 0.67-0.68). The XGBoost model showed lower discrimination (AUROC 0.63). Calibration was comparable across models. Decision curve analysis showed only small differences in net benefit between models across clinically relevant threshold probabilities.</p><p><strong>Conclusion: </strong>ML models provided no clear improvement over logistic regression, which achieved similar predictive performance and greater interpretability. These findings highlight both the potential and the limitations of ML for supporting targeted medication reconciliation in ED workflows. Future research should explore the added value of richer data sources, such as unstructured clinical narratives.</p>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"211 ","pages":"106309"},"PeriodicalIF":4.1,"publicationDate":"2026-01-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146108314","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

Characteristics of online medication consultation from home-based patients on a tertiary hospital WeChat platform: a cross-sectional study 三级医院微信平台家庭患者在线用药咨询的特点：一项横断面研究

IF 4.1 2区医学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

International Journal of Medical Informatics

Pub Date : 2026-01-24 DOI: 10.1016/j.ijmedinf.2026.106308

Chen Wang , Liu-Cheng Li , Jie Ying , Qin Chen , Kai-Li Mao , Lian-Di Kan

Background: Amidst the rapid digital transformation of healthcare, internet hospitals have catalyzed substantial growth in pharmacist-led online medication consultation services. Nevertheless, there remains a notable paucity of empirical analysis and evaluation regarding pharmacists’ provision of online pharmaceutical care.

Objective: The present study was to analyze the characteristics of online medication consultation from home-based patients on a tertiary hospital WeChat platform.

Methods: A retrospective analysis was performed on 5,746 consultation records from April 2022 to March 2025. Consultation categories and frequencies, response rates, and patient demographic characteristics (gender distribution and age profiles) were systematically analyzed to elucidate specific patient needs within the online consultation paradigm. Statistical modeling was employed to examine associations between consultation efficacy and variables including patient gender, consultation year, and temporal patterns. Furthermore, frequently inquired medications were quantified to discern prevailing consultation trends.

Results: A total of 5,746 consultations were analyzed. Physical examination result inquiries were most frequent but least response rate (16.30 %). Medication timing and administration methods ranked as the second and third most frequent consultation categories, respectively, with higher response rates of 74.38 % and 68.94 %. The patient population was predominantly female (p = 0.017) with a median age of 30 years. Among the three annual periods, April 2023-March 2024 yielded the highest consultation volume but lowest response rate. Across four daily time intervals, consultation volume peaked during afternoon hours and was lowest in the late-night period, with comparable response rates among periods. Antibiotics and gastrointestinal medications represented the most frequent consultation topics.

Conclusion:Online consultation provides patients with convenient access to professional guidance on medication administration, timing and selection. However, the user demographic is predominantly younger, necessitating strategies to enhance accessibility for elderly populations. Increased pharmacist staffing during afternoon hours is warranted to accommodate peak consultation volumes, particularly specialists in antimicrobial therapy and Helicobacter pylori regimens. These findings inform targeted quality improvements at dispensing windows, emphasizing proactive counseling on high-frequency consultation topics identified through online interactions.

背景：在医疗保健的快速数字化转型中，互联网医院促进了药剂师主导的在线药物咨询服务的大幅增长。尽管如此，关于药剂师提供在线药学服务的实证分析和评估仍然显着缺乏。目的：分析某三级医院微信平台居家患者在线用药咨询的特点。方法：回顾性分析2022年4月~ 2025年3月5746例会诊记录。系统地分析了咨询类别和频率、应答率和患者人口统计学特征（性别分布和年龄概况），以阐明在线咨询范式中的特定患者需求。采用统计模型来检验会诊效果与患者性别、会诊年份和时间模式等变量之间的关系。此外，经常询问的药物被量化，以辨别普遍的咨询趋势。结果：共分析了5746例咨询。体检结果查询频率最高，但应答率最低（16.30%）。用药时间和给药方法分别为第二和第三高的咨询类别，有效率分别为74.38%和68.94%。患者以女性为主（p = 0.017），中位年龄为30岁。在三个年度期间，2023年4月至2024年3月的咨询量最高，但回应率最低。在每天的四个时间间隔中，咨询量在下午达到高峰，在深夜达到最低，各时间段的回应率相当。抗生素和胃肠道药物是最常见的咨询主题。结论：网上咨询为患者在给药、用药时机和选择等方面提供了方便的专业指导。然而，用户主要是年轻人，因此需要采取战略来提高老年人的可及性。在下午增加药剂师的工作人员是必要的，以适应高峰咨询量，特别是在抗菌治疗和幽门螺杆菌方案的专家。这些发现为分配窗口提供了有针对性的质量改进，强调了通过在线互动确定的高频咨询主题的主动咨询。

{"title":"Characteristics of online medication consultation from home-based patients on a tertiary hospital WeChat platform: a cross-sectional study","authors":"Chen Wang , Liu-Cheng Li , Jie Ying , Qin Chen , Kai-Li Mao , Lian-Di Kan","doi":"10.1016/j.ijmedinf.2026.106308","DOIUrl":"10.1016/j.ijmedinf.2026.106308","url":null,"abstract":"<div><div><strong>Background</strong>: Amidst the rapid digital transformation of healthcare, internet hospitals have catalyzed substantial growth in pharmacist-led online medication consultation services. Nevertheless, there remains a notable paucity of empirical analysis and evaluation regarding pharmacists’ provision of online pharmaceutical care.</div><div><strong>Objective</strong>: The present study was to analyze the characteristics of online medication consultation from home-based patients on a tertiary hospital WeChat platform.</div><div><strong>Methods</strong>: A retrospective analysis was performed on 5,746 consultation records from April 2022 to March 2025. Consultation categories and frequencies, response rates, and patient demographic characteristics (gender distribution and age profiles) were systematically analyzed to elucidate specific patient needs within the online consultation paradigm. Statistical modeling was employed to examine associations between consultation efficacy and variables including patient gender, consultation year, and temporal patterns. Furthermore, frequently inquired medications were quantified to discern prevailing consultation trends.</div><div><strong>Results</strong>: A total of 5,746 consultations were analyzed. Physical examination result inquiries were most frequent but least response rate (16.30 %). Medication timing and administration methods ranked as the second and third most frequent consultation categories, respectively, with higher response rates of 74.38 % and 68.94 %. The patient population was predominantly female (<em>p</em> = 0.017) with a median age of 30 years. Among the three annual periods, April 2023-March 2024 yielded the highest consultation volume but lowest response rate. Across four daily time intervals, consultation volume peaked during afternoon hours and was lowest in the late-night period, with comparable response rates among periods. Antibiotics and gastrointestinal medications represented the most frequent consultation topics.</div><div><strong>Conclusion:</strong>Online consultation provides patients with convenient access to professional guidance on medication administration, timing and selection. However, the user demographic is predominantly younger, necessitating strategies to enhance accessibility for elderly populations. Increased pharmacist staffing during afternoon hours is warranted to accommodate peak consultation volumes, particularly specialists in antimicrobial therapy and <em>Helicobacter pylori</em> regimens. These findings inform targeted quality improvements at dispensing windows, emphasizing proactive counseling on high-frequency consultation topics identified through online interactions.</div></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"210 ","pages":"Article 106308"},"PeriodicalIF":4.1,"publicationDate":"2026-01-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146080698","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

Prioritising digital health technologies in Australian community pharmacy: a delphi study identifying barriers, enablers, and policy implications for implementation 优先考虑澳大利亚社区药房的数字卫生技术：一项德尔菲研究，确定了实施的障碍、推动因素和政策影响。

IF 4.1 2区医学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

International Journal of Medical Informatics

Pub Date : 2026-01-23 DOI: 10.1016/j.ijmedinf.2026.106305

Amina Hareem , Julie E. Stevens , Joon Soo Park , Ieva Stupans , Elton Lobo , Jae Pyun , Kate Wang

Objective

To determine priority digital health technologies for Australian community pharmacies and identify the key barriers, enablers, and policy/funding factors that can inform future implementation planning through expert consensus.

Methods

A two-round Delphi study was conducted with 31 experts representing pharmacy, academia, policy, and digital health. In Round 1, participants identified priority technologies, barriers, and enablers. In Round 2, 27 participants ranked five technologies, nine policy options, and six financial models. Consensus was assessed using descriptive statistics and interquartile ranges (IQRs).

Results

E-prescriptions and My Health Record (MyHR) were ranked as top priorities (mean = 1.70 and 2.22; IQR ≤ 1.0). Key barriers included financial constraints, interoperability issues, and digital literacy gaps. Telehealth incentives received the strongest agreement among participants, while reimbursement-based funding and government support were rated as the most supportive financial models for implementation. Broader enablers, such as a national medicine repository and stronger cross-disciplinary collaboration, were also endorsed.

Conclusion

Digital health adoption in community pharmacy requires prioritisation of core technologies, improved system integration, workforce training, and practical funding mechanisms. These findings offer guidance for policymakers, pharmacy leaders, and digital health stakeholders aiming to embed digital tools more consistently and effectively into pharmacy practice.

目的：确定澳大利亚社区药房的优先数字卫生技术，并确定通过专家共识可以为未来实施规划提供信息的关键障碍、推动因素和政策/资金因素。方法：采用两轮德尔菲法对31名来自药学、学术界、政策和数字健康领域的专家进行调查。在第一轮中，参与者确定了优先技术、障碍和促成因素。在第二轮中，27名参与者对5项技术、9项政策选择和6种金融模型进行了排名。使用描述性统计和四分位数范围（IQRs）评估共识。结果：电子处方和我的健康记录（MyHR）排名最高（平均值分别为1.70和2.22,IQR≤1.0）。主要障碍包括财政限制、互操作性问题和数字素养差距。远程保健奖励措施得到与会者最强烈的赞同，而基于补偿的供资和政府支助被评为最有利于实施的财政模式。会议还批准了更广泛的推动因素，如建立国家药物储存库和加强跨学科合作。结论：社区药房采用数字健康需要优先考虑核心技术，改进系统集成，进行劳动力培训，并建立切实可行的资助机制。这些发现为决策者、药房领导和数字健康利益相关者提供了指导，旨在将数字工具更一致、更有效地嵌入药房实践。

{"title":"Prioritising digital health technologies in Australian community pharmacy: a delphi study identifying barriers, enablers, and policy implications for implementation","authors":"Amina Hareem , Julie E. Stevens , Joon Soo Park , Ieva Stupans , Elton Lobo , Jae Pyun , Kate Wang","doi":"10.1016/j.ijmedinf.2026.106305","DOIUrl":"10.1016/j.ijmedinf.2026.106305","url":null,"abstract":"<div><h3>Objective</h3><div>To determine priority digital health technologies for Australian community pharmacies and identify the key barriers, enablers, and policy/funding factors that can inform future implementation planning through expert consensus.</div></div><div><h3>Methods</h3><div>A two-round Delphi study was conducted with 31 experts representing pharmacy, academia, policy, and digital health. In Round 1, participants identified priority technologies, barriers, and enablers. In Round 2, 27 participants ranked five technologies, nine policy options, and six financial models. Consensus was assessed using descriptive statistics and interquartile ranges (IQRs).</div></div><div><h3>Results</h3><div>E-prescriptions and My Health Record (MyHR) were ranked as top priorities (mean = 1.70 and 2.22; IQR ≤ 1.0). Key barriers included financial constraints, interoperability issues, and digital literacy gaps. Telehealth incentives received the strongest agreement among participants, while reimbursement-based funding and government support were rated as the most supportive financial models for implementation. Broader enablers, such as a national medicine repository and stronger cross-disciplinary collaboration, were also endorsed.</div></div><div><h3>Conclusion</h3><div>Digital health adoption in community pharmacy requires prioritisation of core technologies, improved system integration, workforce training, and practical funding mechanisms. These findings offer guidance for policymakers, pharmacy leaders, and digital health stakeholders aiming to embed digital tools more consistently and effectively into pharmacy practice.</div></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"210 ","pages":"Article 106305"},"PeriodicalIF":4.1,"publicationDate":"2026-01-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146055263","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

Reproducing real-world clinical prediction models using the DIVE platform: A comparative validation study across three chronic diseases 使用DIVE平台再现真实世界的临床预测模型：一项跨三种慢性疾病的比较验证研究

IF 4.1 2区医学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

International Journal of Medical Informatics

Pub Date : 2026-01-22 DOI: 10.1016/j.ijmedinf.2026.106303

Francesco Lapi , Ettore Marconi , Marco Gorini , Lorenzo Nuti , Gerardo Medea , Iacopo Cricelli

Objectives

The aim of this analysis is to evaluate the performance and reproducibility of the Python-based Data Insight Validation Engine (DIVE), a modular analytics interface implemented in Python to facilitate real-world evidence (RWE) generation from clinical (e.g. primary care) data. The platform was used to replicate three previously published studies focused on chronic kidney disease (CKD), chronic obstructive pulmonary disease (COPD), and severe asthma, each originally developed using conventional statistical environments.

Methods

Using a primary care data source, DIVE was employed to replicate three studies on development and validation of prediction scores using machine learning (ML) and traditional inferential analyses. Namely, a ML-based Generalized Additive² Model (GA²M) predicting CKD, and two Cox-based regression models for COPD exacerbations (CEX-HScore) and severe asthma (AS-HScore). Data referred to over one million patients under the care of approximately 800 general practitioners (GPs) in Italy. Although the initial studies were carried out between 2013 and 2021, the DIVE-based investigations extended from 2013 to 2022, thereby also demonstrating “external” temporal validation. Results obtained via DIVE were compared to the “original” prior findings.

Results

DIVE demonstrated high fidelity in replicating published results. The CKD model achieved largely consistent discrimination (AUC: 89.2% vs. 89.3%) and average precision (22.1% vs. 22.4%) using GA²M. The COPD model showed AUC of 65.5%, pseudo-R² of 12.7%, and calibration slope of 1.01 (p = 0.317) which were consistent with original CEX-HScore (AUC: 66%; pseudo-R²: 13%; calibration slope: 1.03 (p = 0.345)). For severe asthma, the prediction model exhibited an AUC equals to 71.9%, pseudo-R² of 17.6%, and calibration slope of 1.09 (p = 0.211), still aligned with the original AS-HScore (AUC: 72.5%; pseudo-R²: 18%; calibration slope: 1.12 (p = 0.182)).

Conclusion

DIVE represents a reliable, scalable, and interoperable solution for RWE analytics, demonstrating equivalence with traditional analytic methods and aligning with best practices in data reproducibility. Continued development toward integrating federated (multi-database) analyses protocols and broader interoperability might expand its utility across several clinical domains.

目的：本分析的目的是评估基于Python的数据洞察验证引擎（DIVE）的性能和可重复性，这是一个用Python实现的模块化分析接口，用于促进临床（例如初级保健）数据的真实世界证据（RWE）生成。该平台用于重复先前发表的三项研究，重点是慢性肾脏疾病（CKD）、慢性阻塞性肺疾病（COPD）和严重哮喘，每项研究最初都是使用传统的统计环境开发的。方法：使用初级保健数据源，采用DIVE重复三项研究，利用机器学习（ML）和传统的推理分析来开发和验证预测分数。即基于ml的广义加法模型（GA2M）预测CKD，以及两个基于cox的COPD恶化（CEX-HScore）和重度哮喘（AS-HScore）回归模型。数据涉及意大利约800名全科医生（gp）护理下的100多万患者。虽然最初的研究是在2013年至2021年之间进行的，但基于dive的调查从2013年延长到2022年，从而也证明了“外部”时间验证。通过DIVE获得的结果与“原始”先前的发现进行比较。结果：DIVE在复制已发表的结果方面表现出高保真度。使用GA2M， CKD模型获得了基本一致的识别率（AUC: 89.2% vs 89.3%）和平均精度（22.1% vs 22.4%）。COPD模型AUC为65.5%，拟合r2为12.7%，校正斜率为1.01 (p = 0.317)，与原始CEX-HScore （AUC: 66%，拟合r2: 13%，校正斜率：1.03 (p = 0.345)）一致。对于重度哮喘，预测模型AUC为71.9%，拟合r2为17.6%，校正斜率为1.09 (p = 0.211)，与原始AS-HScore （AUC: 72.5%，拟合r2: 18%，校正斜率：1.12 (p = 0.182)）保持一致。结论：DIVE是一种可靠的、可扩展的、可互操作的RWE分析解决方案，与传统分析方法相当，在数据再现性方面符合最佳实践。继续朝着集成联邦（多数据库）分析协议和更广泛的互操作性的方向发展，可能会扩展其在多个临床领域的实用性。

{"title":"Reproducing real-world clinical prediction models using the DIVE platform: A comparative validation study across three chronic diseases","authors":"Francesco Lapi , Ettore Marconi , Marco Gorini , Lorenzo Nuti , Gerardo Medea , Iacopo Cricelli","doi":"10.1016/j.ijmedinf.2026.106303","DOIUrl":"10.1016/j.ijmedinf.2026.106303","url":null,"abstract":"<div><h3>Objectives</h3><div>The aim of this analysis is to evaluate the performance and reproducibility of the Python-based Data Insight Validation Engine (DIVE), a modular analytics interface implemented in Python to facilitate real-world evidence (RWE) generation from clinical (e.g. primary care) data. The platform was used to replicate three previously published studies focused on chronic kidney disease (CKD), chronic obstructive pulmonary disease (COPD), and severe asthma, each originally developed using conventional statistical environments.</div></div><div><h3>Methods</h3><div>Using a primary care data source, DIVE was employed to replicate three studies on development and validation of prediction scores using machine learning (ML) and traditional inferential analyses. Namely, a ML-based Generalized Additive<sup>2</sup> Model (GA<sup>2</sup>M) predicting CKD, and two Cox-based regression models for COPD exacerbations (CEX-HScore) and severe asthma (AS-HScore). Data referred to over one million patients under the care of approximately 800 general practitioners (GPs) in Italy. Although the initial studies were carried out between 2013 and 2021, the DIVE-based investigations extended from 2013 to 2022, thereby also demonstrating “external” temporal validation. Results obtained via DIVE were compared to the “original” prior findings.</div></div><div><h3>Results</h3><div>DIVE demonstrated high fidelity in replicating published results. The CKD model achieved largely consistent discrimination (AUC: 89.2% vs. 89.3%) and average precision (22.1% vs. 22.4%) using GA<sup>2</sup>M. The COPD model showed AUC of 65.5%, pseudo-R<sup>2</sup> of 12.7%, and calibration slope of 1.01 (p = 0.317) which were consistent with original CEX-HScore (AUC: 66%; pseudo-R<sup>2</sup>: 13%; calibration slope: 1.03 (p = 0.345)). For severe asthma, the prediction model exhibited an AUC equals to 71.9%, pseudo-R<sup>2</sup> of 17.6%, and calibration slope of 1.09 (p = 0.211), still aligned with the original AS-HScore (AUC: 72.5%; pseudo-R<sup>2</sup>: 18%; calibration slope: 1.12 (p = 0.182)).</div></div><div><h3>Conclusion</h3><div>DIVE represents a reliable, scalable, and interoperable solution for RWE analytics, demonstrating equivalence with traditional analytic methods and aligning with best practices in data reproducibility. Continued development toward integrating federated (multi-database) analyses protocols and broader interoperability might expand its utility across several clinical domains.</div></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"210 ","pages":"Article 106303"},"PeriodicalIF":4.1,"publicationDate":"2026-01-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146068320","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